diff --git "a/log_1724033313.txt" "b/log_1724033313.txt" new file mode 100644--- /dev/null +++ "b/log_1724033313.txt" @@ -0,0 +1,36305 @@ +Step 1 | loss:4.977115154266357 lr:1.25e-06 tokens_per_second_per_gpu:357.6114455483301 mfu:1.6869142034557307 +Step 2 | loss:5.802342891693115 lr:2.5e-06 tokens_per_second_per_gpu:3116.7879505486744 mfu:14.702420541597759 +Step 3 | loss:5.689752101898193 lr:3.75e-06 tokens_per_second_per_gpu:4595.4657867121805 mfu:21.67759618323503 +Step 4 | loss:4.5182342529296875 lr:5e-06 tokens_per_second_per_gpu:4752.505385003742 mfu:22.418378762094672 +Step 5 | loss:4.745183944702148 lr:6.25e-06 tokens_per_second_per_gpu:3284.8944310841425 mfu:15.495407492206729 +Step 6 | loss:4.297004699707031 lr:7.5e-06 tokens_per_second_per_gpu:5150.791942959616 mfu:24.297164410666586 +Step 7 | loss:3.370804786682129 lr:8.75e-06 tokens_per_second_per_gpu:5180.785945540063 mfu:24.438651238343095 +Step 8 | loss:3.1073412895202637 lr:1e-05 tokens_per_second_per_gpu:3217.520727673348 mfu:15.177594238078868 +Step 9 | loss:2.816117763519287 lr:1.125e-05 tokens_per_second_per_gpu:3811.8525727934402 mfu:17.981158954979513 +Step 10 | loss:2.4933037757873535 lr:1.25e-05 tokens_per_second_per_gpu:3654.7766945116487 mfu:17.24020523721595 +Step 11 | loss:2.582620620727539 lr:1.3750000000000002e-05 tokens_per_second_per_gpu:4474.363939388008 mfu:21.10633810730114 +Step 12 | loss:2.453038215637207 lr:1.5e-05 tokens_per_second_per_gpu:5014.676070854758 mfu:23.655082618185432 +Step 13 | loss:2.6531434059143066 lr:1.6250000000000002e-05 tokens_per_second_per_gpu:5333.14264591706 mfu:25.157343788754964 +Step 14 | loss:2.246553659439087 lr:1.75e-05 tokens_per_second_per_gpu:5359.8837005088335 mfu:25.28348590575874 +Step 15 | loss:2.1769909858703613 lr:1.8750000000000002e-05 tokens_per_second_per_gpu:4844.108608179268 mfu:22.850486794935126 +Step 16 | loss:2.3359761238098145 lr:2e-05 tokens_per_second_per_gpu:5316.3713038161495 mfu:25.07823050657894 +Step 17 | loss:2.296849250793457 lr:2.125e-05 tokens_per_second_per_gpu:4461.744968854108 mfu:21.046812270274355 +Step 18 | loss:1.922660231590271 lr:2.25e-05 tokens_per_second_per_gpu:5371.576557989068 mfu:25.33864310950021 +Step 19 | loss:2.0903518199920654 lr:2.375e-05 tokens_per_second_per_gpu:4437.414084146985 mfu:20.932039335833856 +Step 20 | loss:2.07597017288208 lr:2.5e-05 tokens_per_second_per_gpu:5325.087984438838 mfu:25.119348576295234 +Step 21 | loss:2.167999267578125 lr:2.625e-05 tokens_per_second_per_gpu:3886.2261433179633 mfu:18.331991776583315 +Step 22 | loss:1.956641435623169 lr:2.7500000000000004e-05 tokens_per_second_per_gpu:5366.496691336586 mfu:25.31468051178583 +Step 23 | loss:1.9976859092712402 lr:2.8749999999999997e-05 tokens_per_second_per_gpu:4943.181983537087 mfu:23.317832810158848 +Step 24 | loss:1.9248414039611816 lr:3e-05 tokens_per_second_per_gpu:4339.914388931594 mfu:20.472116638339486 +Step 25 | loss:1.920866847038269 lr:3.125e-05 tokens_per_second_per_gpu:4169.066719789499 mfu:19.66619903337806 +Step 26 | loss:1.9564201831817627 lr:3.2500000000000004e-05 tokens_per_second_per_gpu:4533.566601683585 mfu:21.38560716636515 +Step 27 | loss:1.8525471687316895 lr:3.375000000000001e-05 tokens_per_second_per_gpu:4983.198242069238 mfu:23.506596329132737 +Step 28 | loss:1.8630582094192505 lr:3.5e-05 tokens_per_second_per_gpu:4262.295371215247 mfu:20.105974488601465 +Step 29 | loss:1.691329836845398 lr:3.625e-05 tokens_per_second_per_gpu:4386.034689587668 mfu:20.689673965468902 +Step 30 | loss:1.6779667139053345 lr:3.7500000000000003e-05 tokens_per_second_per_gpu:5353.7765861956905 mfu:25.2546776055625 +Step 31 | loss:1.7953671216964722 lr:3.875e-05 tokens_per_second_per_gpu:5337.023588170976 mfu:25.175650855523713 +Step 32 | loss:1.8227676153182983 lr:4e-05 tokens_per_second_per_gpu:4372.157349367198 mfu:20.62421218393055 +Step 33 | loss:1.8691363334655762 lr:4.125e-05 tokens_per_second_per_gpu:3928.394464190658 mfu:18.530906940797244 +Step 34 | loss:1.9419254064559937 lr:4.25e-05 tokens_per_second_per_gpu:5362.875974400284 mfu:25.297600972239284 +Step 35 | loss:1.6252727508544922 lr:4.375e-05 tokens_per_second_per_gpu:4385.068892368229 mfu:20.68511813520281 +Step 36 | loss:2.120028495788574 lr:4.5e-05 tokens_per_second_per_gpu:2872.258828034838 mfu:13.548934949729563 +Step 37 | loss:1.6902426481246948 lr:4.6250000000000006e-05 tokens_per_second_per_gpu:3858.511153799265 mfu:18.201255442358796 +Step 38 | loss:1.8422335386276245 lr:4.75e-05 tokens_per_second_per_gpu:5352.250186256762 mfu:25.24747731661989 +Step 39 | loss:1.8644042015075684 lr:4.875e-05 tokens_per_second_per_gpu:3598.6478222242017 mfu:16.975435770062944 +Step 40 | loss:1.8438526391983032 lr:5e-05 tokens_per_second_per_gpu:4175.187038344371 mfu:19.695069620235415 +Step 41 | loss:1.9035049676895142 lr:5.125e-05 tokens_per_second_per_gpu:3520.310999929073 mfu:16.60590761365672 +Step 42 | loss:1.759334921836853 lr:5.25e-05 tokens_per_second_per_gpu:4468.468292046818 mfu:21.07852733289159 +Step 43 | loss:1.8935174942016602 lr:5.375e-05 tokens_per_second_per_gpu:4371.449114852385 mfu:20.62087132088691 +Step 44 | loss:1.7295974493026733 lr:5.500000000000001e-05 tokens_per_second_per_gpu:5116.020784328321 mfu:24.133142922831354 +Step 45 | loss:1.6646069288253784 lr:5.6250000000000005e-05 tokens_per_second_per_gpu:4219.226826028389 mfu:19.90281281270352 +Step 46 | loss:1.7425038814544678 lr:5.7499999999999995e-05 tokens_per_second_per_gpu:5332.733902757761 mfu:25.155415677533774 +Step 47 | loss:1.77764892578125 lr:5.8750000000000005e-05 tokens_per_second_per_gpu:3865.2193668523123 mfu:18.23289922786919 +Step 48 | loss:1.6717827320098877 lr:6e-05 tokens_per_second_per_gpu:5329.216963223615 mfu:25.13882567369945 +Step 49 | loss:1.6372060775756836 lr:6.125000000000001e-05 tokens_per_second_per_gpu:5323.284993701137 mfu:25.110843561363556 +Step 50 | loss:1.6749216318130493 lr:6.25e-05 tokens_per_second_per_gpu:5341.87651464233 mfu:25.19854293768373 +Step 51 | loss:1.6254881620407104 lr:6.375e-05 tokens_per_second_per_gpu:3928.9908752641923 mfu:18.53372031358933 +Step 52 | loss:1.5417120456695557 lr:6.500000000000001e-05 tokens_per_second_per_gpu:5376.466032447833 mfu:25.361707594752573 +Step 53 | loss:1.7423839569091797 lr:6.625e-05 tokens_per_second_per_gpu:3813.4688939043695 mfu:17.988783417432586 +Step 54 | loss:1.6658880710601807 lr:6.750000000000001e-05 tokens_per_second_per_gpu:4357.805630059024 mfu:20.556512675298084 +Step 55 | loss:1.6575756072998047 lr:6.875e-05 tokens_per_second_per_gpu:4152.37192668342 mfu:19.587446845871256 +Step 56 | loss:1.628978967666626 lr:7e-05 tokens_per_second_per_gpu:4192.002144603634 mfu:19.774389345413322 +Step 57 | loss:1.6093302965164185 lr:7.125000000000001e-05 tokens_per_second_per_gpu:5370.154025428626 mfu:25.33193278070352 +Step 58 | loss:1.5017375946044922 lr:7.25e-05 tokens_per_second_per_gpu:5321.652857712715 mfu:25.103144497435053 +Step 59 | loss:1.6383483409881592 lr:7.375e-05 tokens_per_second_per_gpu:3876.9208393924077 mfu:18.288097070317907 +Step 60 | loss:1.8423153162002563 lr:7.500000000000001e-05 tokens_per_second_per_gpu:4241.079605776108 mfu:20.005896103242204 +Step 61 | loss:1.5847941637039185 lr:7.625e-05 tokens_per_second_per_gpu:4532.4185529898705 mfu:21.380191624799597 +Step 62 | loss:1.7432739734649658 lr:7.75e-05 tokens_per_second_per_gpu:3651.695553278772 mfu:17.225670968323566 +Step 63 | loss:1.6111385822296143 lr:7.875e-05 tokens_per_second_per_gpu:3898.3766911834255 mfu:18.389307983961938 +Step 64 | loss:1.6226521730422974 lr:8e-05 tokens_per_second_per_gpu:3862.0591114568742 mfu:18.21799176397374 +Step 65 | loss:1.5231575965881348 lr:8.125000000000001e-05 tokens_per_second_per_gpu:4252.621181145402 mfu:20.06033968345498 +Step 66 | loss:1.6752289533615112 lr:8.25e-05 tokens_per_second_per_gpu:5292.63729708689 mfu:24.96627314739791 +Step 67 | loss:1.888590931892395 lr:8.375e-05 tokens_per_second_per_gpu:5353.634382147312 mfu:25.254006804803787 +Step 68 | loss:1.7112354040145874 lr:8.5e-05 tokens_per_second_per_gpu:4716.413171785805 mfu:22.248125634379104 +Step 69 | loss:1.8126922845840454 lr:8.625000000000001e-05 tokens_per_second_per_gpu:3805.686609871854 mfu:17.952073055856637 +Step 70 | loss:1.6470041275024414 lr:8.75e-05 tokens_per_second_per_gpu:5314.174672078094 mfu:25.06786861988651 +Step 71 | loss:1.5515538454055786 lr:8.875e-05 tokens_per_second_per_gpu:4588.34623233616 mfu:21.644012030522102 +Step 72 | loss:1.663364291191101 lr:9e-05 tokens_per_second_per_gpu:5283.954103656931 mfu:24.925313042483225 +Step 73 | loss:1.7292652130126953 lr:9.125e-05 tokens_per_second_per_gpu:4007.4758701111027 mfu:18.903947425203153 +Step 74 | loss:1.7287043333053589 lr:9.250000000000001e-05 tokens_per_second_per_gpu:4348.860850963916 mfu:20.514318626169775 +Step 75 | loss:1.4615694284439087 lr:9.375e-05 tokens_per_second_per_gpu:5364.508015103903 mfu:25.305299586693113 +Step 76 | loss:1.5816060304641724 lr:9.5e-05 tokens_per_second_per_gpu:5240.251841308045 mfu:24.7191620146087 +Step 77 | loss:1.6188013553619385 lr:9.625000000000001e-05 tokens_per_second_per_gpu:4500.494236063794 mfu:21.229599175009163 +Step 78 | loss:1.6986182928085327 lr:9.75e-05 tokens_per_second_per_gpu:4947.248724401873 mfu:23.33701632067558 +Step 79 | loss:1.6556849479675293 lr:9.875000000000002e-05 tokens_per_second_per_gpu:3797.736932291526 mfu:17.91457307035537 +Step 80 | loss:1.6750799417495728 lr:0.0001 tokens_per_second_per_gpu:4644.53582574287 mfu:21.909067929554762 +Step 81 | loss:1.594044804573059 lr:9.999999981197181e-05 tokens_per_second_per_gpu:4081.5728347294844 mfu:19.253475449553854 +Step 82 | loss:1.6763513088226318 lr:9.999999924788725e-05 tokens_per_second_per_gpu:4422.860003880772 mfu:20.863385256035976 +Step 83 | loss:1.6325174570083618 lr:9.99999983077463e-05 tokens_per_second_per_gpu:4527.060108458118 mfu:21.354914927698406 +Step 84 | loss:1.6169507503509521 lr:9.9999996991549e-05 tokens_per_second_per_gpu:5387.419860336825 mfu:25.413378669820705 +Step 85 | loss:1.6030534505844116 lr:9.999999529929532e-05 tokens_per_second_per_gpu:4728.525940882702 mfu:22.30526363286068 +Step 86 | loss:1.6149234771728516 lr:9.999999323098531e-05 tokens_per_second_per_gpu:4690.6026389489425 mfu:22.12637294725662 +Step 87 | loss:1.6076806783676147 lr:9.999999078661897e-05 tokens_per_second_per_gpu:4162.549143593606 mfu:19.635454514448863 +Step 88 | loss:1.6452643871307373 lr:9.999998796619632e-05 tokens_per_second_per_gpu:5367.7224471709715 mfu:25.320462611192138 +Step 89 | loss:1.6134248971939087 lr:9.999998476971737e-05 tokens_per_second_per_gpu:5331.120501835515 mfu:25.14780499010896 +Step 90 | loss:1.6035537719726562 lr:9.999998119718217e-05 tokens_per_second_per_gpu:4860.792788247754 mfu:22.92918891893283 +Step 91 | loss:1.5198533535003662 lr:9.999997724859073e-05 tokens_per_second_per_gpu:4484.911102369068 mfu:21.156090874614375 +Step 92 | loss:1.534193754196167 lr:9.999997292394307e-05 tokens_per_second_per_gpu:4303.393549014596 mfu:20.29984160535319 +Step 93 | loss:1.8662641048431396 lr:9.999996822323924e-05 tokens_per_second_per_gpu:5357.055914699305 mfu:25.270146757625184 +Step 94 | loss:1.7746814489364624 lr:9.999996314647925e-05 tokens_per_second_per_gpu:5344.986923610766 mfu:25.213215267478333 +Step 95 | loss:1.6424798965454102 lr:9.99999576936632e-05 tokens_per_second_per_gpu:5335.1058822360765 mfu:25.166604709434424 +Step 96 | loss:1.8308556079864502 lr:9.999995186479105e-05 tokens_per_second_per_gpu:3140.2528086679918 mfu:14.813108280864205 +Step 97 | loss:1.7396461963653564 lr:9.99999456598629e-05 tokens_per_second_per_gpu:4813.0741348582915 mfu:22.704091889252926 +Step 98 | loss:1.5237096548080444 lr:9.999993907887876e-05 tokens_per_second_per_gpu:4289.706910310291 mfu:20.235279395403623 +Step 99 | loss:1.7738392353057861 lr:9.999993212183873e-05 tokens_per_second_per_gpu:4727.378722572665 mfu:22.299852008356602 +Step 100 | loss:1.6969436407089233 lr:9.99999247887428e-05 tokens_per_second_per_gpu:4627.350283647534 mfu:21.828000795335228 +Step 101 | loss:1.4452462196350098 lr:9.999991707959106e-05 tokens_per_second_per_gpu:5299.901171721314 mfu:25.000538083393298 +Step 102 | loss:1.5933443307876587 lr:9.999990899438358e-05 tokens_per_second_per_gpu:3013.625880613429 mfu:14.215787456449092 +Step 103 | loss:1.7552030086517334 lr:9.999990053312041e-05 tokens_per_second_per_gpu:3588.4626988807695 mfu:16.92739080549074 +Step 104 | loss:1.7536749839782715 lr:9.999989169580159e-05 tokens_per_second_per_gpu:5338.247466815726 mfu:25.181424099906227 +Step 105 | loss:1.6716833114624023 lr:9.99998824824272e-05 tokens_per_second_per_gpu:3237.199204715337 mfu:15.270420971780327 +Step 106 | loss:1.5709445476531982 lr:9.999987289299732e-05 tokens_per_second_per_gpu:4344.333101549714 mfu:20.492960459669273 +Step 107 | loss:1.666176676750183 lr:9.999986292751203e-05 tokens_per_second_per_gpu:5324.053167496666 mfu:25.114467168221033 +Step 108 | loss:1.9560880661010742 lr:9.999985258597138e-05 tokens_per_second_per_gpu:2897.2304784031207 mfu:13.666730485119643 +Step 109 | loss:1.720683217048645 lr:9.999984186837546e-05 tokens_per_second_per_gpu:4138.1313296249755 mfu:19.520271519850176 +Step 110 | loss:1.4911872148513794 lr:9.999983077472434e-05 tokens_per_second_per_gpu:5207.756195442598 mfu:24.565874508732485 +Step 111 | loss:1.6964640617370605 lr:9.999981930501812e-05 tokens_per_second_per_gpu:4074.772678294537 mfu:19.221397951423878 +Step 112 | loss:1.9723231792449951 lr:9.999980745925688e-05 tokens_per_second_per_gpu:4102.857944971912 mfu:19.353880946181672 +Step 113 | loss:1.6166048049926758 lr:9.999979523744069e-05 tokens_per_second_per_gpu:4335.671253864479 mfu:20.452101046274414 +Step 114 | loss:1.5112602710723877 lr:9.999978263956968e-05 tokens_per_second_per_gpu:4788.770083344969 mfu:22.5894455315658 +Step 115 | loss:1.4996871948242188 lr:9.999976966564392e-05 tokens_per_second_per_gpu:4697.0560396277115 mfu:22.156814739322066 +Step 116 | loss:1.6079826354980469 lr:9.999975631566351e-05 tokens_per_second_per_gpu:4227.135628832462 mfu:19.940119984911234 +Step 117 | loss:1.6535558700561523 lr:9.999974258962856e-05 tokens_per_second_per_gpu:5102.664921461164 mfu:24.070141038941795 +Step 118 | loss:1.3643683195114136 lr:9.999972848753918e-05 tokens_per_second_per_gpu:5366.308672918664 mfu:25.313793596829246 +Step 119 | loss:1.8539003133773804 lr:9.999971400939543e-05 tokens_per_second_per_gpu:5029.921971802473 mfu:23.727000134174734 +Step 120 | loss:1.6671372652053833 lr:9.999969915519747e-05 tokens_per_second_per_gpu:4183.092257650961 mfu:19.73235988847409 +Step 121 | loss:1.5981459617614746 lr:9.999968392494538e-05 tokens_per_second_per_gpu:4606.760688466827 mfu:21.73087616192065 +Step 122 | loss:1.4918025732040405 lr:9.999966831863931e-05 tokens_per_second_per_gpu:5200.786704117041 mfu:24.532998229031996 +Step 123 | loss:1.6916927099227905 lr:9.999965233627934e-05 tokens_per_second_per_gpu:4881.953756969261 mfu:23.02900881882574 +Step 124 | loss:1.643632173538208 lr:9.99996359778656e-05 tokens_per_second_per_gpu:4443.618860633024 mfu:20.961308325162538 +Step 125 | loss:1.6451221704483032 lr:9.999961924339822e-05 tokens_per_second_per_gpu:5344.131187006038 mfu:25.20917861191011 +Step 126 | loss:1.5643641948699951 lr:9.999960213287733e-05 tokens_per_second_per_gpu:3802.0075153252606 mfu:17.934718139162097 +Step 127 | loss:1.896005630493164 lr:9.999958464630305e-05 tokens_per_second_per_gpu:5338.375142408649 mfu:25.18202636746175 +Step 128 | loss:1.464520812034607 lr:9.999956678367551e-05 tokens_per_second_per_gpu:4887.672095718366 mfu:23.055983198354465 +Step 129 | loss:1.8352677822113037 lr:9.999954854499484e-05 tokens_per_second_per_gpu:4220.895264201927 mfu:19.910683120233003 +Step 130 | loss:1.825022578239441 lr:9.99995299302612e-05 tokens_per_second_per_gpu:5279.541492126414 mfu:24.904498001024614 +Step 131 | loss:1.5183511972427368 lr:9.999951093947473e-05 tokens_per_second_per_gpu:4868.689040598239 mfu:22.96643688850964 +Step 132 | loss:1.6132062673568726 lr:9.999949157263553e-05 tokens_per_second_per_gpu:5001.348020117209 mfu:23.592211928856464 +Step 133 | loss:1.549281358718872 lr:9.999947182974379e-05 tokens_per_second_per_gpu:4893.529459617977 mfu:23.083613383238262 +Step 134 | loss:1.5950697660446167 lr:9.999945171079963e-05 tokens_per_second_per_gpu:5354.413973614564 mfu:25.257684270766834 +Step 135 | loss:2.0059516429901123 lr:9.999943121580322e-05 tokens_per_second_per_gpu:3873.575600902469 mfu:18.272317009604322 +Step 136 | loss:1.824257254600525 lr:9.99994103447547e-05 tokens_per_second_per_gpu:3626.2259619056917 mfu:17.10552656025636 +Step 137 | loss:1.5396379232406616 lr:9.999938909765425e-05 tokens_per_second_per_gpu:4904.557852934612 mfu:23.135636196150756 +Step 138 | loss:1.4718241691589355 lr:9.9999367474502e-05 tokens_per_second_per_gpu:4849.788657078466 mfu:22.877280554708918 +Step 139 | loss:1.4912599325180054 lr:9.999934547529814e-05 tokens_per_second_per_gpu:4066.2848258759814 mfu:19.1813593033886 +Step 140 | loss:1.4167594909667969 lr:9.999932310004283e-05 tokens_per_second_per_gpu:4787.820275182012 mfu:22.58496512440715 +Step 141 | loss:1.8283880949020386 lr:9.999930034873621e-05 tokens_per_second_per_gpu:3027.4288215850715 mfu:14.280898284036901 +Step 142 | loss:1.5872268676757812 lr:9.999927722137848e-05 tokens_per_second_per_gpu:4038.553617216313 mfu:19.050546460708855 +Step 143 | loss:1.4768949747085571 lr:9.999925371796981e-05 tokens_per_second_per_gpu:4957.487374381363 mfu:23.385313779522658 +Step 144 | loss:1.579728126525879 lr:9.999922983851037e-05 tokens_per_second_per_gpu:4730.215564666323 mfu:22.313233876527562 +Step 145 | loss:1.5528804063796997 lr:9.999920558300033e-05 tokens_per_second_per_gpu:4129.693071025225 mfu:19.480466814319776 +Step 146 | loss:1.562134861946106 lr:9.99991809514399e-05 tokens_per_second_per_gpu:3569.7964630643523 mfu:16.839338986356466 +Step 147 | loss:1.6979511976242065 lr:9.999915594382925e-05 tokens_per_second_per_gpu:5285.506081579421 mfu:24.93263398713835 +Step 148 | loss:1.7575689554214478 lr:9.999913056016857e-05 tokens_per_second_per_gpu:5310.985556435434 mfu:25.052825017280846 +Step 149 | loss:1.6058013439178467 lr:9.999910480045804e-05 tokens_per_second_per_gpu:5360.908796470733 mfu:25.28832145831043 +Step 150 | loss:1.3760288953781128 lr:9.999907866469786e-05 tokens_per_second_per_gpu:4793.4573588582025 mfu:22.61155620989339 +Step 151 | loss:1.6608830690383911 lr:9.999905215288825e-05 tokens_per_second_per_gpu:4821.5376553681 mfu:22.744015759523194 +Step 152 | loss:1.4542816877365112 lr:9.999902526502937e-05 tokens_per_second_per_gpu:4793.738922222587 mfu:22.612884392322854 +Step 153 | loss:1.6338752508163452 lr:9.999899800112144e-05 tokens_per_second_per_gpu:4825.9288566461055 mfu:22.76472980516774 +Step 154 | loss:1.4975342750549316 lr:9.999897036116468e-05 tokens_per_second_per_gpu:4290.292803817616 mfu:20.238043155041417 +Step 155 | loss:1.6309894323349 lr:9.999894234515928e-05 tokens_per_second_per_gpu:4067.805434546889 mfu:19.188532273931862 +Step 156 | loss:1.7686258554458618 lr:9.999891395310544e-05 tokens_per_second_per_gpu:3612.8900190099403 mfu:17.042618642270618 +Step 157 | loss:1.6812293529510498 lr:9.99988851850034e-05 tokens_per_second_per_gpu:3480.449507846522 mfu:16.417874154431473 +Step 158 | loss:1.724161982536316 lr:9.999885604085337e-05 tokens_per_second_per_gpu:5327.223321239401 mfu:25.129421324309515 +Step 159 | loss:1.5780503749847412 lr:9.999882652065555e-05 tokens_per_second_per_gpu:4602.180618802326 mfu:21.709271191872737 +Step 160 | loss:1.5159473419189453 lr:9.999879662441018e-05 tokens_per_second_per_gpu:5132.528276380541 mfu:24.211011579310245 +Step 161 | loss:1.474927544593811 lr:9.99987663521175e-05 tokens_per_second_per_gpu:4530.446172458659 mfu:21.370887569311048 +Step 162 | loss:1.6902960538864136 lr:9.999873570377768e-05 tokens_per_second_per_gpu:4647.8457700387025 mfu:21.924681501532937 +Step 163 | loss:1.7375913858413696 lr:9.999870467939103e-05 tokens_per_second_per_gpu:3652.8889148872368 mfu:17.23130025863918 +Step 164 | loss:1.6101139783859253 lr:9.999867327895772e-05 tokens_per_second_per_gpu:4859.523920349463 mfu:22.923203452565403 +Step 165 | loss:1.7314754724502563 lr:9.999864150247802e-05 tokens_per_second_per_gpu:4404.822067810878 mfu:20.778297234005045 +Step 166 | loss:1.6311572790145874 lr:9.999860934995214e-05 tokens_per_second_per_gpu:5351.522288409725 mfu:25.244043698283235 +Step 167 | loss:1.5634568929672241 lr:9.999857682138035e-05 tokens_per_second_per_gpu:5000.917113891789 mfu:23.590179270670976 +Step 168 | loss:1.6216775178909302 lr:9.99985439167629e-05 tokens_per_second_per_gpu:3968.569289047896 mfu:18.720418444180773 +Step 169 | loss:1.5599719285964966 lr:9.99985106361e-05 tokens_per_second_per_gpu:4418.479324956306 mfu:20.842720846128376 +Step 170 | loss:1.4251290559768677 lr:9.999847697939191e-05 tokens_per_second_per_gpu:5342.696630991388 mfu:25.20241156642851 +Step 171 | loss:1.4848564863204956 lr:9.999844294663891e-05 tokens_per_second_per_gpu:5344.630942696577 mfu:25.211536044769495 +Step 172 | loss:1.7887282371520996 lr:9.999840853784125e-05 tokens_per_second_per_gpu:5345.204616895118 mfu:25.21424216384293 +Step 173 | loss:1.5167880058288574 lr:9.999837375299916e-05 tokens_per_second_per_gpu:5324.377006432679 mfu:25.115994771734822 +Step 174 | loss:1.5607881546020508 lr:9.999833859211294e-05 tokens_per_second_per_gpu:5327.474662833434 mfu:25.130606945491866 +Step 175 | loss:1.8474515676498413 lr:9.999830305518282e-05 tokens_per_second_per_gpu:4197.777682068267 mfu:19.801633541042296 +Step 176 | loss:1.529289722442627 lr:9.99982671422091e-05 tokens_per_second_per_gpu:4571.883567657464 mfu:21.566354832412465 +Step 177 | loss:1.3915703296661377 lr:9.999823085319202e-05 tokens_per_second_per_gpu:5346.796225002286 mfu:25.22175005084078 +Step 178 | loss:1.5018274784088135 lr:9.999819418813188e-05 tokens_per_second_per_gpu:4337.505855326882 mfu:20.460755174387856 +Step 179 | loss:1.4370858669281006 lr:9.999815714702891e-05 tokens_per_second_per_gpu:4656.305823990227 mfu:21.964589019456323 +Step 180 | loss:1.3595609664916992 lr:9.999811972988345e-05 tokens_per_second_per_gpu:5346.83952163264 mfu:25.221954288433327 +Step 181 | loss:1.6727123260498047 lr:9.999808193669575e-05 tokens_per_second_per_gpu:4271.5407418565455 mfu:20.149586479339362 +Step 182 | loss:1.6479345560073853 lr:9.999804376746609e-05 tokens_per_second_per_gpu:3674.2491215826576 mfu:17.332059998048546 +Step 183 | loss:1.6015710830688477 lr:9.999800522219476e-05 tokens_per_second_per_gpu:5376.312592397596 mfu:25.360983791856032 +Step 184 | loss:1.5301748514175415 lr:9.999796630088205e-05 tokens_per_second_per_gpu:5372.9620486913045 mfu:25.345178705531755 +Step 185 | loss:1.5869094133377075 lr:9.999792700352826e-05 tokens_per_second_per_gpu:4570.447538781648 mfu:21.55958083919345 +Step 186 | loss:1.7009718418121338 lr:9.999788733013368e-05 tokens_per_second_per_gpu:3957.741426467113 mfu:18.669341568963365 +Step 187 | loss:1.5201878547668457 lr:9.999784728069862e-05 tokens_per_second_per_gpu:4234.654547381878 mfu:19.975588006568618 +Step 188 | loss:1.7807611227035522 lr:9.999780685522335e-05 tokens_per_second_per_gpu:4207.8398654229795 mfu:19.849098576711988 +Step 189 | loss:1.6269394159317017 lr:9.999776605370822e-05 tokens_per_second_per_gpu:4826.013217595213 mfu:22.765127750158143 +Step 190 | loss:1.5026233196258545 lr:9.99977248761535e-05 tokens_per_second_per_gpu:4462.28146229788 mfu:21.049343001383846 +Step 191 | loss:1.5280933380126953 lr:9.99976833225595e-05 tokens_per_second_per_gpu:3769.9488377855214 mfu:17.783491887432735 +Step 192 | loss:1.584761381149292 lr:9.999764139292656e-05 tokens_per_second_per_gpu:5346.320887340451 mfu:25.21950780198894 +Step 193 | loss:1.5381591320037842 lr:9.999759908725497e-05 tokens_per_second_per_gpu:4827.011008059293 mfu:22.769834498017815 +Step 194 | loss:1.575740933418274 lr:9.999755640554506e-05 tokens_per_second_per_gpu:4535.6296631910645 mfu:21.395338979490525 +Step 195 | loss:1.8287780284881592 lr:9.999751334779716e-05 tokens_per_second_per_gpu:4791.140979843801 mfu:22.600629454867445 +Step 196 | loss:1.5717638731002808 lr:9.999746991401158e-05 tokens_per_second_per_gpu:3421.061132375862 mfu:16.137729054634665 +Step 197 | loss:1.8906826972961426 lr:9.999742610418863e-05 tokens_per_second_per_gpu:4777.74590101351 mfu:22.53744258258885 +Step 198 | loss:1.4894214868545532 lr:9.999738191832867e-05 tokens_per_second_per_gpu:4160.75763695984 mfu:19.62700367199484 +Step 199 | loss:1.5609506368637085 lr:9.999733735643201e-05 tokens_per_second_per_gpu:5320.769784757978 mfu:25.09897889163978 +Step 200 | loss:1.434980869293213 lr:9.999729241849903e-05 tokens_per_second_per_gpu:3379.446025827288 mfu:15.941423496775705 +Step 201 | loss:1.6461706161499023 lr:9.999724710453e-05 tokens_per_second_per_gpu:4472.347688753219 mfu:21.096827109049013 +Step 202 | loss:1.6033382415771484 lr:9.999720141452532e-05 tokens_per_second_per_gpu:4992.897187824474 mfu:23.552347911070964 +Step 203 | loss:1.5628172159194946 lr:9.999715534848528e-05 tokens_per_second_per_gpu:3481.3756765213293 mfu:16.422243050091208 +Step 204 | loss:1.6909650564193726 lr:9.999710890641027e-05 tokens_per_second_per_gpu:5392.009935316893 mfu:25.43503083664923 +Step 205 | loss:1.6078797578811646 lr:9.999706208830062e-05 tokens_per_second_per_gpu:5320.312393686474 mfu:25.09682129991653 +Step 206 | loss:1.458487868309021 lr:9.99970148941567e-05 tokens_per_second_per_gpu:5310.738219560133 mfu:25.05165828704003 +Step 207 | loss:1.3689974546432495 lr:9.999696732397883e-05 tokens_per_second_per_gpu:4117.341296356398 mfu:19.422201385776848 +Step 208 | loss:1.645653247833252 lr:9.999691937776741e-05 tokens_per_second_per_gpu:3687.045390141797 mfu:17.392422180110554 +Step 209 | loss:1.5518417358398438 lr:9.999687105552275e-05 tokens_per_second_per_gpu:3396.317442385605 mfu:16.021008847241777 +Step 210 | loss:1.6789473295211792 lr:9.999682235724528e-05 tokens_per_second_per_gpu:3523.8263668796258 mfu:16.622490199345926 +Step 211 | loss:1.4223533868789673 lr:9.999677328293531e-05 tokens_per_second_per_gpu:5194.88851736789 mfu:24.50517547580182 +Step 212 | loss:1.6903934478759766 lr:9.999672383259322e-05 tokens_per_second_per_gpu:5322.528539309359 mfu:25.107275236933113 +Step 213 | loss:1.6120579242706299 lr:9.999667400621942e-05 tokens_per_second_per_gpu:3955.6955940121593 mfu:18.659691028218337 +Step 214 | loss:1.6739774942398071 lr:9.999662380381422e-05 tokens_per_second_per_gpu:3483.3741427182217 mfu:16.431670156115445 +Step 215 | loss:1.537605881690979 lr:9.999657322537805e-05 tokens_per_second_per_gpu:5262.707266052917 mfu:24.825088084420788 +Step 216 | loss:1.4700030088424683 lr:9.999652227091127e-05 tokens_per_second_per_gpu:5379.166348739343 mfu:25.374445447421095 +Step 217 | loss:1.6943227052688599 lr:9.999647094041428e-05 tokens_per_second_per_gpu:4834.948685379217 mfu:22.8072778762433 +Step 218 | loss:1.4100309610366821 lr:9.999641923388744e-05 tokens_per_second_per_gpu:4141.144975058617 mfu:19.534487399541593 +Step 219 | loss:1.4781394004821777 lr:9.999636715133114e-05 tokens_per_second_per_gpu:5318.058898430218 mfu:25.08619117830683 +Step 220 | loss:1.5388753414154053 lr:9.99963146927458e-05 tokens_per_second_per_gpu:4038.4636312701896 mfu:19.050121981647806 +Step 221 | loss:1.4801530838012695 lr:9.99962618581318e-05 tokens_per_second_per_gpu:3541.6126154549443 mfu:16.706390968522634 +Step 222 | loss:1.5958679914474487 lr:9.999620864748953e-05 tokens_per_second_per_gpu:3691.1562208758123 mfu:17.41181367006295 +Step 223 | loss:1.498438835144043 lr:9.999615506081939e-05 tokens_per_second_per_gpu:3724.5860087307706 mfu:17.569507683084748 +Step 224 | loss:1.4000016450881958 lr:9.999610109812179e-05 tokens_per_second_per_gpu:3600.408026037701 mfu:16.98373895177296 +Step 225 | loss:1.4556677341461182 lr:9.999604675939713e-05 tokens_per_second_per_gpu:5099.975202756073 mfu:24.057453176896544 +Step 226 | loss:1.5309364795684814 lr:9.999599204464583e-05 tokens_per_second_per_gpu:4610.2069037362435 mfu:21.747132547331447 +Step 227 | loss:1.3864469528198242 lr:9.999593695386831e-05 tokens_per_second_per_gpu:4260.499744725259 mfu:20.097504212082722 +Step 228 | loss:1.649256944656372 lr:9.999588148706493e-05 tokens_per_second_per_gpu:4419.905230332581 mfu:20.84944708507281 +Step 229 | loss:1.674020767211914 lr:9.999582564423617e-05 tokens_per_second_per_gpu:4693.793929001227 mfu:22.141426806923473 +Step 230 | loss:1.8304381370544434 lr:9.999576942538244e-05 tokens_per_second_per_gpu:4199.671075923492 mfu:19.810565002903697 +Step 231 | loss:1.7574007511138916 lr:9.999571283050413e-05 tokens_per_second_per_gpu:3989.0007826287288 mfu:18.81679728537383 +Step 232 | loss:1.4871970415115356 lr:9.999565585960169e-05 tokens_per_second_per_gpu:4157.717182596113 mfu:19.612661330006137 +Step 233 | loss:1.501407504081726 lr:9.999559851267553e-05 tokens_per_second_per_gpu:5349.748945589531 mfu:25.235678537636897 +Step 234 | loss:1.6097533702850342 lr:9.99955407897261e-05 tokens_per_second_per_gpu:5364.501190401237 mfu:25.305267393406226 +Step 235 | loss:1.602247714996338 lr:9.999548269075383e-05 tokens_per_second_per_gpu:5109.282408974295 mfu:24.101356856601033 +Step 236 | loss:1.4353002309799194 lr:9.999542421575915e-05 tokens_per_second_per_gpu:5327.280492842631 mfu:25.12969101251643 +Step 237 | loss:1.5901182889938354 lr:9.999536536474252e-05 tokens_per_second_per_gpu:4747.407211899475 mfu:22.39432980972382 +Step 238 | loss:1.4735808372497559 lr:9.999530613770436e-05 tokens_per_second_per_gpu:5328.489280894679 mfu:25.13539307199845 +Step 239 | loss:1.6683062314987183 lr:9.999524653464512e-05 tokens_per_second_per_gpu:4433.191219258177 mfu:20.912119361658302 +Step 240 | loss:1.4956086874008179 lr:9.999518655556526e-05 tokens_per_second_per_gpu:4111.737385402677 mfu:19.39576678168235 +Step 241 | loss:1.4202460050582886 lr:9.999512620046522e-05 tokens_per_second_per_gpu:5006.827648176215 mfu:23.618060269331753 +Step 242 | loss:1.7904207706451416 lr:9.999506546934545e-05 tokens_per_second_per_gpu:4768.515357593297 mfu:22.4939005343826 +Step 243 | loss:1.4392962455749512 lr:9.999500436220642e-05 tokens_per_second_per_gpu:5334.251866634643 mfu:25.162576172132372 +Step 244 | loss:1.5221153497695923 lr:9.999494287904857e-05 tokens_per_second_per_gpu:4581.004784496643 mfu:21.60938116848294 +Step 245 | loss:1.5613714456558228 lr:9.999488101987239e-05 tokens_per_second_per_gpu:4781.589342941606 mfu:22.55557275392249 +Step 246 | loss:1.548429012298584 lr:9.999481878467831e-05 tokens_per_second_per_gpu:4316.503498399964 mfu:20.361683473391984 +Step 247 | loss:1.4857447147369385 lr:9.999475617346684e-05 tokens_per_second_per_gpu:5330.363705478258 mfu:25.14423505256901 +Step 248 | loss:1.6085819005966187 lr:9.999469318623842e-05 tokens_per_second_per_gpu:5173.420895214461 mfu:24.403909039350065 +Step 249 | loss:1.463384747505188 lr:9.999462982299354e-05 tokens_per_second_per_gpu:4553.991328976555 mfu:21.481954089823915 +Step 250 | loss:1.5144425630569458 lr:9.999456608373267e-05 tokens_per_second_per_gpu:4843.5235833634115 mfu:22.847727133063206 +Step 251 | loss:1.6166424751281738 lr:9.999450196845628e-05 tokens_per_second_per_gpu:4084.07872323052 mfu:19.265296153161906 +Step 252 | loss:1.655717372894287 lr:9.999443747716486e-05 tokens_per_second_per_gpu:3767.2535940944795 mfu:17.770777963086097 +Step 253 | loss:1.7235912084579468 lr:9.999437260985891e-05 tokens_per_second_per_gpu:4674.109858366383 mfu:22.048573687290155 +Step 254 | loss:1.5652357339859009 lr:9.99943073665389e-05 tokens_per_second_per_gpu:3824.2958634445213 mfu:18.039856080024972 +Step 255 | loss:1.769311785697937 lr:9.999424174720531e-05 tokens_per_second_per_gpu:5349.974963378224 mfu:25.236744702108773 +Step 256 | loss:1.428949236869812 lr:9.999417575185867e-05 tokens_per_second_per_gpu:4633.176216871666 mfu:21.855482716360164 +Step 257 | loss:1.5178546905517578 lr:9.999410938049944e-05 tokens_per_second_per_gpu:4467.718894000947 mfu:21.074992294448727 +Step 258 | loss:1.4852381944656372 lr:9.999404263312814e-05 tokens_per_second_per_gpu:4229.992470449528 mfu:19.953596194246376 +Step 259 | loss:1.5199028253555298 lr:9.999397550974526e-05 tokens_per_second_per_gpu:5362.587849037525 mfu:25.296241835743896 +Step 260 | loss:1.5427311658859253 lr:9.999390801035132e-05 tokens_per_second_per_gpu:4598.108468518915 mfu:21.690062164205408 +Step 261 | loss:1.5531551837921143 lr:9.999384013494682e-05 tokens_per_second_per_gpu:4212.4338553073 mfu:19.870769210811826 +Step 262 | loss:1.4219180345535278 lr:9.999377188353226e-05 tokens_per_second_per_gpu:5341.075272160019 mfu:25.1947633401886 +Step 263 | loss:1.375824213027954 lr:9.999370325610816e-05 tokens_per_second_per_gpu:4979.726284607148 mfu:23.49021851340703 +Step 264 | loss:1.7403321266174316 lr:9.999363425267506e-05 tokens_per_second_per_gpu:5341.2241642052295 mfu:25.195465689369463 +Step 265 | loss:1.321394443511963 lr:9.999356487323344e-05 tokens_per_second_per_gpu:4308.26916952982 mfu:20.322840739190138 +Step 266 | loss:1.6402541399002075 lr:9.999349511778382e-05 tokens_per_second_per_gpu:5336.432351871213 mfu:25.172861893022414 +Step 267 | loss:1.480301856994629 lr:9.999342498632677e-05 tokens_per_second_per_gpu:4239.0803448966535 mfu:19.996465248564736 +Step 268 | loss:1.621330738067627 lr:9.99933544788628e-05 tokens_per_second_per_gpu:5337.903762663074 mfu:25.179802788776193 +Step 269 | loss:1.4551053047180176 lr:9.999328359539241e-05 tokens_per_second_per_gpu:4332.285484271898 mfu:20.43612979343435 +Step 270 | loss:1.492715835571289 lr:9.999321233591616e-05 tokens_per_second_per_gpu:4615.357036736561 mfu:21.77142660339724 +Step 271 | loss:1.5872430801391602 lr:9.99931407004346e-05 tokens_per_second_per_gpu:4262.040811459313 mfu:20.104773686800353 +Step 272 | loss:1.578521966934204 lr:9.999306868894822e-05 tokens_per_second_per_gpu:5159.477357566862 mfu:24.338134993253504 +Step 273 | loss:1.4488741159439087 lr:9.999299630145761e-05 tokens_per_second_per_gpu:3771.1911660604737 mfu:17.789352161868745 +Step 274 | loss:1.407361626625061 lr:9.99929235379633e-05 tokens_per_second_per_gpu:5356.401835420247 mfu:25.26706135406844 +Step 275 | loss:1.6321903467178345 lr:9.999285039846583e-05 tokens_per_second_per_gpu:4184.265833621116 mfu:19.737895846558022 +Step 276 | loss:1.2802722454071045 lr:9.999277688296573e-05 tokens_per_second_per_gpu:5317.345212314892 mfu:25.08282459913343 +Step 277 | loss:1.3447608947753906 lr:9.99927029914636e-05 tokens_per_second_per_gpu:5316.292987491727 mfu:25.077861075114495 +Step 278 | loss:1.3912814855575562 lr:9.999262872395995e-05 tokens_per_second_per_gpu:3791.9222453492494 mfu:17.88714419996076 +Step 279 | loss:1.4711099863052368 lr:9.999255408045539e-05 tokens_per_second_per_gpu:3965.9723923492306 mfu:18.708168439376905 +Step 280 | loss:1.5704023838043213 lr:9.999247906095044e-05 tokens_per_second_per_gpu:5352.1989126440185 mfu:25.247235450240396 +Step 281 | loss:1.6860631704330444 lr:9.999240366544566e-05 tokens_per_second_per_gpu:3660.030340099504 mfu:17.264987579818232 +Step 282 | loss:1.7025563716888428 lr:9.999232789394165e-05 tokens_per_second_per_gpu:5359.508697673474 mfu:25.28171695340227 +Step 283 | loss:1.7833980321884155 lr:9.999225174643896e-05 tokens_per_second_per_gpu:4277.337125422538 mfu:20.176929009585162 +Step 284 | loss:1.870025873184204 lr:9.999217522293816e-05 tokens_per_second_per_gpu:5353.14987329158 mfu:25.25172129386556 +Step 285 | loss:1.750087857246399 lr:9.999209832343983e-05 tokens_per_second_per_gpu:4246.719473194797 mfu:20.032500320116764 +Step 286 | loss:1.5927867889404297 lr:9.999202104794456e-05 tokens_per_second_per_gpu:5352.4336870116485 mfu:25.24834292098915 +Step 287 | loss:1.556875228881836 lr:9.999194339645292e-05 tokens_per_second_per_gpu:5234.291164931658 mfu:24.691044487145025 +Step 288 | loss:1.6802409887313843 lr:9.999186536896548e-05 tokens_per_second_per_gpu:4251.465760959157 mfu:20.05488936930145 +Step 289 | loss:1.528411626815796 lr:9.999178696548286e-05 tokens_per_second_per_gpu:3571.432950365525 mfu:16.847058576169623 +Step 290 | loss:1.5362863540649414 lr:9.999170818600562e-05 tokens_per_second_per_gpu:5333.910365174883 mfu:25.160965251480526 +Step 291 | loss:1.619497299194336 lr:9.999162903053436e-05 tokens_per_second_per_gpu:5064.217179931332 mfu:23.88877648228444 +Step 292 | loss:1.3742998838424683 lr:9.99915494990697e-05 tokens_per_second_per_gpu:5334.4806062187145 mfu:25.163655175777436 +Step 293 | loss:1.4336925745010376 lr:9.999146959161219e-05 tokens_per_second_per_gpu:5294.470772420504 mfu:24.974921963369717 +Step 294 | loss:1.4115139245986938 lr:9.999138930816247e-05 tokens_per_second_per_gpu:5352.710854228117 mfu:25.249650366038832 +Step 295 | loss:1.8543038368225098 lr:9.999130864872114e-05 tokens_per_second_per_gpu:3324.961900649773 mfu:15.684412582363041 +Step 296 | loss:1.461618185043335 lr:9.999122761328878e-05 tokens_per_second_per_gpu:4705.553086852714 mfu:22.196896760828153 +Step 297 | loss:1.3561164140701294 lr:9.999114620186604e-05 tokens_per_second_per_gpu:5349.518174132276 mfu:25.23458994929935 +Step 298 | loss:1.5196369886398315 lr:9.999106441445349e-05 tokens_per_second_per_gpu:3057.106311681033 mfu:14.420892068321493 +Step 299 | loss:1.3602254390716553 lr:9.999098225105179e-05 tokens_per_second_per_gpu:4050.399355901095 mfu:19.1064248311764 +Step 300 | loss:1.4917446374893188 lr:9.99908997116615e-05 tokens_per_second_per_gpu:4386.360765549366 mfu:20.691212121414527 +Step 301 | loss:1.4422775506973267 lr:9.99908167962833e-05 tokens_per_second_per_gpu:5337.304481025603 mfu:25.176975875044256 +Step 302 | loss:1.6644338369369507 lr:9.999073350491778e-05 tokens_per_second_per_gpu:3503.740991972621 mfu:16.527744058991253 +Step 303 | loss:1.5702590942382812 lr:9.999064983756559e-05 tokens_per_second_per_gpu:5365.399214917666 mfu:25.309503528268998 +Step 304 | loss:1.661752700805664 lr:9.999056579422732e-05 tokens_per_second_per_gpu:4459.730801367189 mfu:21.037311098586148 +Step 305 | loss:1.7117326259613037 lr:9.999048137490364e-05 tokens_per_second_per_gpu:4199.802262228812 mfu:19.81118383108858 +Step 306 | loss:1.3982068300247192 lr:9.999039657959517e-05 tokens_per_second_per_gpu:5332.621330957295 mfu:25.154884657144247 +Step 307 | loss:1.49591863155365 lr:9.999031140830256e-05 tokens_per_second_per_gpu:3002.463744873509 mfu:14.163133757707149 +Step 308 | loss:1.5977102518081665 lr:9.999022586102641e-05 tokens_per_second_per_gpu:5351.640284772418 mfu:25.244600307258317 +Step 309 | loss:1.724862813949585 lr:9.99901399377674e-05 tokens_per_second_per_gpu:3335.8272750879937 mfu:15.73566640741217 +Step 310 | loss:1.8515453338623047 lr:9.999005363852618e-05 tokens_per_second_per_gpu:3649.932246789677 mfu:17.217353150763394 +Step 311 | loss:1.7103312015533447 lr:9.998996696330338e-05 tokens_per_second_per_gpu:3672.183249440351 mfu:17.32231492667915 +Step 312 | loss:1.6907237768173218 lr:9.998987991209967e-05 tokens_per_second_per_gpu:3910.0097859188986 mfu:18.444183276639794 +Step 313 | loss:1.3911914825439453 lr:9.998979248491568e-05 tokens_per_second_per_gpu:4784.693187187046 mfu:22.570214116798667 +Step 314 | loss:1.491417407989502 lr:9.99897046817521e-05 tokens_per_second_per_gpu:5342.374607708956 mfu:25.200892527662443 +Step 315 | loss:1.718909740447998 lr:9.998961650260956e-05 tokens_per_second_per_gpu:4325.34405011893 mfu:20.403385864204008 +Step 316 | loss:1.6582530736923218 lr:9.998952794748873e-05 tokens_per_second_per_gpu:5382.364321844897 mfu:25.389530832190466 +Step 317 | loss:1.5627987384796143 lr:9.998943901639028e-05 tokens_per_second_per_gpu:4011.2711431579487 mfu:18.921850375705482 +Step 318 | loss:1.6187814474105835 lr:9.99893497093149e-05 tokens_per_second_per_gpu:5303.955985074192 mfu:25.019665329801317 +Step 319 | loss:1.4885013103485107 lr:9.998926002626324e-05 tokens_per_second_per_gpu:5349.170079707432 mfu:25.232947928506334 +Step 320 | loss:1.774170160293579 lr:9.998916996723596e-05 tokens_per_second_per_gpu:4431.79156583453 mfu:20.905516957653465 +Step 321 | loss:1.4962466955184937 lr:9.998907953223377e-05 tokens_per_second_per_gpu:3478.1995895058803 mfu:16.407260905743012 +Step 322 | loss:1.5922406911849976 lr:9.998898872125732e-05 tokens_per_second_per_gpu:5337.4602027984665 mfu:25.177710441215726 +Step 323 | loss:1.6073918342590332 lr:9.998889753430731e-05 tokens_per_second_per_gpu:5171.520605028922 mfu:24.394945046321936 +Step 324 | loss:1.498542070388794 lr:9.998880597138442e-05 tokens_per_second_per_gpu:4276.116613550831 mfu:20.171171646845526 +Step 325 | loss:1.536744236946106 lr:9.998871403248935e-05 tokens_per_second_per_gpu:5334.04713615162 mfu:25.16161042351333 +Step 326 | loss:1.3804266452789307 lr:9.998862171762278e-05 tokens_per_second_per_gpu:4677.135287408782 mfu:22.06284515227457 +Step 327 | loss:1.544562816619873 lr:9.998852902678541e-05 tokens_per_second_per_gpu:4208.037472775471 mfu:19.85003072430898 +Step 328 | loss:1.604543924331665 lr:9.998843595997795e-05 tokens_per_second_per_gpu:3981.6420218115954 mfu:18.782084755065352 +Step 329 | loss:1.5374857187271118 lr:9.998834251720106e-05 tokens_per_second_per_gpu:5340.227056153348 mfu:25.190762160564557 +Step 330 | loss:1.5223444700241089 lr:9.998824869845547e-05 tokens_per_second_per_gpu:4221.039047440292 mfu:19.911361370300323 +Step 331 | loss:1.6485155820846558 lr:9.998815450374189e-05 tokens_per_second_per_gpu:3415.651463387868 mfu:16.112210723033588 +Step 332 | loss:1.535568118095398 lr:9.998805993306102e-05 tokens_per_second_per_gpu:5343.187473412384 mfu:25.2047269538747 +Step 333 | loss:1.6076381206512451 lr:9.998796498641359e-05 tokens_per_second_per_gpu:5375.613372569677 mfu:25.357685452628537 +Step 334 | loss:1.801588773727417 lr:9.998786966380027e-05 tokens_per_second_per_gpu:4581.785859132898 mfu:21.613065630807228 +Step 335 | loss:1.479350209236145 lr:9.99877739652218e-05 tokens_per_second_per_gpu:5041.109677311654 mfu:23.77977444988093 +Step 336 | loss:1.456386923789978 lr:9.998767789067894e-05 tokens_per_second_per_gpu:4466.438982922138 mfu:21.068954735514637 +Step 337 | loss:1.3687913417816162 lr:9.998758144017234e-05 tokens_per_second_per_gpu:5318.782030569814 mfu:25.089602315988373 +Step 338 | loss:1.6961289644241333 lr:9.998748461370278e-05 tokens_per_second_per_gpu:4532.70857124238 mfu:21.381559690378896 +Step 339 | loss:1.4523617029190063 lr:9.998738741127095e-05 tokens_per_second_per_gpu:5358.09819859007 mfu:25.275063388569894 +Step 340 | loss:1.454153299331665 lr:9.998728983287762e-05 tokens_per_second_per_gpu:4871.562421960583 mfu:22.979991118645234 +Step 341 | loss:1.4549983739852905 lr:9.99871918785235e-05 tokens_per_second_per_gpu:5341.986466864772 mfu:25.19906160107677 +Step 342 | loss:1.454443335533142 lr:9.998709354820934e-05 tokens_per_second_per_gpu:5368.8994056049405 mfu:25.32601452493499 +Step 343 | loss:1.5429311990737915 lr:9.998699484193586e-05 tokens_per_second_per_gpu:3425.327524256879 mfu:16.157854353058934 +Step 344 | loss:1.459235429763794 lr:9.998689575970381e-05 tokens_per_second_per_gpu:5006.446366120333 mfu:23.61626169682073 +Step 345 | loss:1.5360223054885864 lr:9.998679630151395e-05 tokens_per_second_per_gpu:4298.7449636808 mfu:20.277913435203228 +Step 346 | loss:1.4415922164916992 lr:9.9986696467367e-05 tokens_per_second_per_gpu:4068.3277698928205 mfu:19.19099621887822 +Step 347 | loss:1.4344582557678223 lr:9.998659625726374e-05 tokens_per_second_per_gpu:3574.192685612652 mfu:16.860076718189706 +Step 348 | loss:1.9111155271530151 lr:9.998649567120491e-05 tokens_per_second_per_gpu:5356.9978444834305 mfu:25.269872830509108 +Step 349 | loss:1.617362141609192 lr:9.998639470919126e-05 tokens_per_second_per_gpu:5336.771904947096 mfu:25.174463622815086 +Step 350 | loss:1.5502482652664185 lr:9.998629337122358e-05 tokens_per_second_per_gpu:4832.1510573174055 mfu:22.794080987351172 +Step 351 | loss:1.4852972030639648 lr:9.998619165730259e-05 tokens_per_second_per_gpu:5046.569724812607 mfu:23.80553042552309 +Step 352 | loss:1.4100067615509033 lr:9.998608956742908e-05 tokens_per_second_per_gpu:5320.31424403036 mfu:25.09683002830434 +Step 353 | loss:1.2162272930145264 lr:9.998598710160381e-05 tokens_per_second_per_gpu:5309.969213162569 mfu:25.0480307526567 +Step 354 | loss:1.5554441213607788 lr:9.998588425982755e-05 tokens_per_second_per_gpu:3424.6985857572217 mfu:16.15488754284808 +Step 355 | loss:1.5856186151504517 lr:9.998578104210108e-05 tokens_per_second_per_gpu:3777.029636835136 mfu:17.81689322465866 +Step 356 | loss:1.5053528547286987 lr:9.998567744842517e-05 tokens_per_second_per_gpu:5369.264978531411 mfu:25.32773899480232 +Step 357 | loss:1.6122362613677979 lr:9.99855734788006e-05 tokens_per_second_per_gpu:4324.563776947912 mfu:20.399705182527985 +Step 358 | loss:1.5382310152053833 lr:9.998546913322817e-05 tokens_per_second_per_gpu:3440.342462891509 mfu:16.228682380411183 +Step 359 | loss:1.487890601158142 lr:9.998536441170863e-05 tokens_per_second_per_gpu:5349.520368561878 mfu:25.234600300798196 +Step 360 | loss:1.5868582725524902 lr:9.998525931424278e-05 tokens_per_second_per_gpu:5316.716985486763 mfu:25.0798611459986 +Step 361 | loss:1.6398998498916626 lr:9.998515384083144e-05 tokens_per_second_per_gpu:3986.8139428560667 mfu:18.806481589052844 +Step 362 | loss:1.672434687614441 lr:9.998504799147536e-05 tokens_per_second_per_gpu:4129.661903310745 mfu:19.480319790892818 +Step 363 | loss:1.4957809448242188 lr:9.998494176617536e-05 tokens_per_second_per_gpu:4913.367154828574 mfu:23.177191176206463 +Step 364 | loss:1.6395670175552368 lr:9.998483516493224e-05 tokens_per_second_per_gpu:5047.387748233388 mfu:23.8093891815681 +Step 365 | loss:1.5985544919967651 lr:9.99847281877468e-05 tokens_per_second_per_gpu:4476.776993741446 mfu:21.117720896395838 +Step 366 | loss:1.603684902191162 lr:9.998462083461985e-05 tokens_per_second_per_gpu:4798.654544449309 mfu:22.636072221057937 +Step 367 | loss:1.6394450664520264 lr:9.998451310555217e-05 tokens_per_second_per_gpu:5325.125573837509 mfu:25.119525891902178 +Step 368 | loss:1.7005245685577393 lr:9.99844050005446e-05 tokens_per_second_per_gpu:4145.895125674825 mfu:19.556894670457517 +Step 369 | loss:1.7839124202728271 lr:9.998429651959793e-05 tokens_per_second_per_gpu:2708.5456035158722 mfu:12.776671737317205 +Step 370 | loss:1.4841586351394653 lr:9.9984187662713e-05 tokens_per_second_per_gpu:4664.055494499361 mfu:22.001145537478042 +Step 371 | loss:1.6372931003570557 lr:9.99840784298906e-05 tokens_per_second_per_gpu:3770.5707828557192 mfu:17.786425708443627 +Step 372 | loss:1.5173344612121582 lr:9.998396882113159e-05 tokens_per_second_per_gpu:4094.117511231846 mfu:19.31265082895801 +Step 373 | loss:1.2926005125045776 lr:9.998385883643675e-05 tokens_per_second_per_gpu:5349.903889459635 mfu:25.236409434308328 +Step 374 | loss:1.4004876613616943 lr:9.998374847580694e-05 tokens_per_second_per_gpu:5340.270745753841 mfu:25.190968251864298 +Step 375 | loss:1.4465574026107788 lr:9.998363773924299e-05 tokens_per_second_per_gpu:5213.121478097641 mfu:24.59118346242794 +Step 376 | loss:1.5526392459869385 lr:9.99835266267457e-05 tokens_per_second_per_gpu:4708.227701584633 mfu:22.209513374854943 +Step 377 | loss:1.517961859703064 lr:9.998341513831594e-05 tokens_per_second_per_gpu:4093.135812069351 mfu:19.308019986513372 +Step 378 | loss:1.4117121696472168 lr:9.998330327395456e-05 tokens_per_second_per_gpu:5349.118975690793 mfu:25.232706862140603 +Step 379 | loss:1.7012068033218384 lr:9.998319103366234e-05 tokens_per_second_per_gpu:5342.591380123864 mfu:25.201915080128753 +Step 380 | loss:1.273396372795105 lr:9.998307841744019e-05 tokens_per_second_per_gpu:4460.969537851602 mfu:21.04315442993336 +Step 381 | loss:1.5992560386657715 lr:9.998296542528891e-05 tokens_per_second_per_gpu:4187.51162942466 mfu:19.75320681915321 +Step 382 | loss:1.609203815460205 lr:9.998285205720938e-05 tokens_per_second_per_gpu:5057.52126659496 mfu:23.857190716636154 +Step 383 | loss:1.3349337577819824 lr:9.998273831320244e-05 tokens_per_second_per_gpu:5350.646009860926 mfu:25.239910142860015 +Step 384 | loss:1.4451919794082642 lr:9.998262419326895e-05 tokens_per_second_per_gpu:4799.986819108628 mfu:22.642356787935444 +Step 385 | loss:1.7013812065124512 lr:9.998250969740976e-05 tokens_per_second_per_gpu:3606.3566189360563 mfu:17.011799479409106 +Step 386 | loss:1.505946397781372 lr:9.998239482562573e-05 tokens_per_second_per_gpu:4697.329512445416 mfu:22.158104757262453 +Step 387 | loss:1.6304787397384644 lr:9.998227957791775e-05 tokens_per_second_per_gpu:2991.2775107741936 mfu:14.110366382892336 +Step 388 | loss:1.5806316137313843 lr:9.998216395428665e-05 tokens_per_second_per_gpu:4916.983867298125 mfu:23.194251826000112 +Step 389 | loss:1.4268907308578491 lr:9.998204795473333e-05 tokens_per_second_per_gpu:4440.037348567397 mfu:20.944413721680004 +Step 390 | loss:1.3484035730361938 lr:9.998193157925863e-05 tokens_per_second_per_gpu:5345.310749734722 mfu:25.214742810555467 +Step 391 | loss:1.5137611627578735 lr:9.998181482786346e-05 tokens_per_second_per_gpu:5322.116153239062 mfu:25.1053299414792 +Step 392 | loss:1.517581820487976 lr:9.998169770054869e-05 tokens_per_second_per_gpu:4555.8132118063895 mfu:21.490548221977505 +Step 393 | loss:1.5128538608551025 lr:9.998158019731518e-05 tokens_per_second_per_gpu:4716.954022167924 mfu:22.25067691791015 +Step 394 | loss:1.4693433046340942 lr:9.998146231816383e-05 tokens_per_second_per_gpu:3810.6101217912205 mfu:17.975298101618044 +Step 395 | loss:1.4426344633102417 lr:9.998134406309554e-05 tokens_per_second_per_gpu:4656.457715866836 mfu:21.965305519353716 +Step 396 | loss:1.4179823398590088 lr:9.998122543211117e-05 tokens_per_second_per_gpu:3873.2183118098355 mfu:18.270631615994567 +Step 397 | loss:1.4720690250396729 lr:9.998110642521162e-05 tokens_per_second_per_gpu:5365.909907485869 mfu:25.31191255224635 +Step 398 | loss:1.4860426187515259 lr:9.99809870423978e-05 tokens_per_second_per_gpu:5375.456025990903 mfu:25.356943221970365 +Step 399 | loss:1.8121013641357422 lr:9.99808672836706e-05 tokens_per_second_per_gpu:3343.092412992816 mfu:15.769937302469563 +Step 400 | loss:1.6906031370162964 lr:9.998074714903092e-05 tokens_per_second_per_gpu:5239.656370546487 mfu:24.71635307742864 +Step 401 | loss:1.5776245594024658 lr:9.998062663847965e-05 tokens_per_second_per_gpu:3367.1638934468856 mfu:15.883486582789477 +Step 402 | loss:1.600898265838623 lr:9.998050575201771e-05 tokens_per_second_per_gpu:4308.9827285163965 mfu:20.32620671867554 +Step 403 | loss:1.5565658807754517 lr:9.998038448964603e-05 tokens_per_second_per_gpu:4308.63926774894 mfu:20.324586555634998 +Step 404 | loss:1.4171626567840576 lr:9.998026285136549e-05 tokens_per_second_per_gpu:5361.072703202222 mfu:25.289094634328233 +Step 405 | loss:1.5634620189666748 lr:9.998014083717702e-05 tokens_per_second_per_gpu:5320.8206623404385 mfu:25.099218889876852 +Step 406 | loss:1.5317381620407104 lr:9.998001844708151e-05 tokens_per_second_per_gpu:4800.74290415365 mfu:22.645923370094224 +Step 407 | loss:1.4980381727218628 lr:9.997989568107992e-05 tokens_per_second_per_gpu:5174.780191153393 mfu:24.41032107021381 +Step 408 | loss:1.6457068920135498 lr:9.997977253917316e-05 tokens_per_second_per_gpu:3755.1497744874832 mfu:17.713682180875324 +Step 409 | loss:1.4645628929138184 lr:9.997964902136217e-05 tokens_per_second_per_gpu:3979.7332052995257 mfu:18.77308054190024 +Step 410 | loss:1.638306975364685 lr:9.997952512764783e-05 tokens_per_second_per_gpu:5009.715988620411 mfu:23.63168506400872 +Step 411 | loss:1.4358224868774414 lr:9.997940085803112e-05 tokens_per_second_per_gpu:4811.107193326641 mfu:22.694813490453303 +Step 412 | loss:1.5889806747436523 lr:9.997927621251296e-05 tokens_per_second_per_gpu:4976.969351911483 mfu:23.47721359150035 +Step 413 | loss:1.6530141830444336 lr:9.997915119109428e-05 tokens_per_second_per_gpu:2987.9202932571357 mfu:14.094529815064 +Step 414 | loss:1.5827670097351074 lr:9.997902579377602e-05 tokens_per_second_per_gpu:5322.251073758018 mfu:25.10596638458693 +Step 415 | loss:1.4900614023208618 lr:9.997890002055914e-05 tokens_per_second_per_gpu:4674.501435565418 mfu:22.050420823747963 +Step 416 | loss:1.5056896209716797 lr:9.997877387144458e-05 tokens_per_second_per_gpu:4431.954955094408 mfu:20.90628769266939 +Step 417 | loss:1.688762903213501 lr:9.997864734643328e-05 tokens_per_second_per_gpu:3736.368688519738 mfu:17.625088593981776 +Step 418 | loss:1.5613564252853394 lr:9.997852044552619e-05 tokens_per_second_per_gpu:5056.97853753273 mfu:23.85463057105049 +Step 419 | loss:1.5224473476409912 lr:9.997839316872429e-05 tokens_per_second_per_gpu:4734.594563971358 mfu:22.333890363383738 +Step 420 | loss:1.3799349069595337 lr:9.997826551602849e-05 tokens_per_second_per_gpu:3776.0863029780016 mfu:17.812443357905206 +Step 421 | loss:1.4298081398010254 lr:9.997813748743979e-05 tokens_per_second_per_gpu:5379.413241225054 mfu:25.375610081400392 +Step 422 | loss:1.477967381477356 lr:9.997800908295914e-05 tokens_per_second_per_gpu:5356.225531489667 mfu:25.266229698347342 +Step 423 | loss:1.6183093786239624 lr:9.99778803025875e-05 tokens_per_second_per_gpu:4337.308720562963 mfu:20.459825256071788 +Step 424 | loss:1.3667815923690796 lr:9.997775114632587e-05 tokens_per_second_per_gpu:5109.275410825123 mfu:24.101323845137376 +Step 425 | loss:1.6811363697052002 lr:9.997762161417517e-05 tokens_per_second_per_gpu:5379.097048100785 mfu:25.374118544485107 +Step 426 | loss:1.640128254890442 lr:9.997749170613641e-05 tokens_per_second_per_gpu:3924.6734621638116 mfu:18.5133543393682 +Step 427 | loss:1.7752861976623535 lr:9.997736142221055e-05 tokens_per_second_per_gpu:3903.152227189426 mfu:18.411835002092705 +Step 428 | loss:1.4183013439178467 lr:9.99772307623986e-05 tokens_per_second_per_gpu:5158.5381473270545 mfu:24.33370457830654 +Step 429 | loss:1.465246319770813 lr:9.99770997267015e-05 tokens_per_second_per_gpu:3774.066283717323 mfu:17.80291458240176 +Step 430 | loss:1.7067395448684692 lr:9.997696831512027e-05 tokens_per_second_per_gpu:4317.131888059373 mfu:20.364647694629745 +Step 431 | loss:1.52360999584198 lr:9.997683652765585e-05 tokens_per_second_per_gpu:4607.284572473151 mfu:21.73334741216231 +Step 432 | loss:1.4720383882522583 lr:9.99767043643093e-05 tokens_per_second_per_gpu:4319.711118286205 mfu:20.376814363671773 +Step 433 | loss:1.549620509147644 lr:9.997657182508158e-05 tokens_per_second_per_gpu:3956.4104164580745 mfu:18.66306296765701 +Step 434 | loss:1.837656021118164 lr:9.997643890997368e-05 tokens_per_second_per_gpu:4017.3306307913863 mfu:18.950434012727605 +Step 435 | loss:1.5749653577804565 lr:9.99763056189866e-05 tokens_per_second_per_gpu:4577.104549535397 mfu:21.59098309472204 +Step 436 | loss:1.6261473894119263 lr:9.997617195212134e-05 tokens_per_second_per_gpu:3345.082675868198 mfu:15.779325712026601 +Step 437 | loss:1.4227317571640015 lr:9.997603790937893e-05 tokens_per_second_per_gpu:4793.873382178771 mfu:22.61351866287741 +Step 438 | loss:1.5919846296310425 lr:9.997590349076036e-05 tokens_per_second_per_gpu:4070.323377156925 mfu:19.20040983882871 +Step 439 | loss:1.5807573795318604 lr:9.997576869626664e-05 tokens_per_second_per_gpu:3478.4200922737214 mfu:16.40830105492057 +Step 440 | loss:1.5645266771316528 lr:9.997563352589878e-05 tokens_per_second_per_gpu:5366.227585284415 mfu:25.313411092623515 +Step 441 | loss:1.6435091495513916 lr:9.997549797965781e-05 tokens_per_second_per_gpu:3729.914599968917 mfu:17.594643557106497 +Step 442 | loss:1.6552213430404663 lr:9.997536205754474e-05 tokens_per_second_per_gpu:4287.912122322673 mfu:20.226813074243623 +Step 443 | loss:1.3904492855072021 lr:9.99752257595606e-05 tokens_per_second_per_gpu:5365.016443755813 mfu:25.307697931391047 +Step 444 | loss:1.535935640335083 lr:9.997508908570641e-05 tokens_per_second_per_gpu:4044.7459508545694 mfu:19.079756755001572 +Step 445 | loss:1.5388600826263428 lr:9.99749520359832e-05 tokens_per_second_per_gpu:4187.527984031693 mfu:19.753283966624966 +Step 446 | loss:1.4512848854064941 lr:9.997481461039199e-05 tokens_per_second_per_gpu:5266.10271323763 mfu:24.84110498811411 +Step 447 | loss:1.6353598833084106 lr:9.997467680893382e-05 tokens_per_second_per_gpu:4703.838504126251 mfu:22.18880878156945 +Step 448 | loss:1.3715705871582031 lr:9.997453863160975e-05 tokens_per_second_per_gpu:4711.060375673118 mfu:22.222875581834362 +Step 449 | loss:1.6105375289916992 lr:9.997440007842078e-05 tokens_per_second_per_gpu:5362.445342642508 mfu:25.295569608764502 +Step 450 | loss:1.4820888042449951 lr:9.997426114936796e-05 tokens_per_second_per_gpu:5280.755412692379 mfu:24.910224271450478 +Step 451 | loss:1.4465056657791138 lr:9.997412184445236e-05 tokens_per_second_per_gpu:5357.826734259829 mfu:25.27378284500756 +Step 452 | loss:1.7002277374267578 lr:9.997398216367502e-05 tokens_per_second_per_gpu:3403.0253920092086 mfu:16.05265139011054 +Step 453 | loss:1.4851562976837158 lr:9.997384210703697e-05 tokens_per_second_per_gpu:5362.246403763369 mfu:25.29463118012913 +Step 454 | loss:1.4933736324310303 lr:9.997370167453929e-05 tokens_per_second_per_gpu:4345.826770831628 mfu:20.500006352518323 +Step 455 | loss:1.5207353830337524 lr:9.997356086618302e-05 tokens_per_second_per_gpu:4951.776842616588 mfu:23.35837622687102 +Step 456 | loss:1.6840230226516724 lr:9.99734196819692e-05 tokens_per_second_per_gpu:5026.348838779153 mfu:23.710145056064373 +Step 457 | loss:1.5921177864074707 lr:9.997327812189894e-05 tokens_per_second_per_gpu:3476.132399326532 mfu:16.397509616968012 +Step 458 | loss:1.4898252487182617 lr:9.997313618597326e-05 tokens_per_second_per_gpu:4300.890831167911 mfu:20.288035858262678 +Step 459 | loss:1.5868380069732666 lr:9.997299387419328e-05 tokens_per_second_per_gpu:4317.700102123436 mfu:20.367328057317255 +Step 460 | loss:1.5142948627471924 lr:9.997285118656e-05 tokens_per_second_per_gpu:5125.966309621958 mfu:24.180057662523033 +Step 461 | loss:1.4498693943023682 lr:9.997270812307454e-05 tokens_per_second_per_gpu:4780.935177244697 mfu:22.552486942717994 +Step 462 | loss:1.5350189208984375 lr:9.997256468373797e-05 tokens_per_second_per_gpu:5362.680989020028 mfu:25.296681192932684 +Step 463 | loss:1.4388046264648438 lr:9.997242086855137e-05 tokens_per_second_per_gpu:4130.257828326702 mfu:19.483130870868376 +Step 464 | loss:1.5621970891952515 lr:9.99722766775158e-05 tokens_per_second_per_gpu:5271.886946741291 mfu:24.868390204442875 +Step 465 | loss:1.5380889177322388 lr:9.997213211063237e-05 tokens_per_second_per_gpu:3989.998419015929 mfu:18.82150330642657 +Step 466 | loss:1.4328320026397705 lr:9.997198716790216e-05 tokens_per_second_per_gpu:4530.027522862805 mfu:21.368912727738422 +Step 467 | loss:1.5229934453964233 lr:9.997184184932625e-05 tokens_per_second_per_gpu:4977.526680973077 mfu:23.47984260777404 +Step 468 | loss:1.5816993713378906 lr:9.997169615490576e-05 tokens_per_second_per_gpu:4946.88276840633 mfu:23.335290043806072 +Step 469 | loss:1.5867336988449097 lr:9.997155008464175e-05 tokens_per_second_per_gpu:3680.1967912836835 mfu:17.360116170805377 +Step 470 | loss:1.2895448207855225 lr:9.997140363853534e-05 tokens_per_second_per_gpu:5334.00498932614 mfu:25.16141160974655 +Step 471 | loss:1.4311707019805908 lr:9.99712568165876e-05 tokens_per_second_per_gpu:5327.5790666343655 mfu:25.131099436033985 +Step 472 | loss:1.6059939861297607 lr:9.99711096187997e-05 tokens_per_second_per_gpu:5180.453747478507 mfu:24.4370842034079 +Step 473 | loss:1.5034524202346802 lr:9.997096204517269e-05 tokens_per_second_per_gpu:5359.888385641051 mfu:25.28350800632678 +Step 474 | loss:1.7632604837417603 lr:9.997081409570771e-05 tokens_per_second_per_gpu:3788.0069524186447 mfu:17.86867509518924 +Step 475 | loss:1.737852692604065 lr:9.997066577040585e-05 tokens_per_second_per_gpu:5337.378915883964 mfu:25.177326996971036 +Step 476 | loss:1.4009217023849487 lr:9.997051706926824e-05 tokens_per_second_per_gpu:5367.186454378769 mfu:25.317934241741522 +Step 477 | loss:1.4400209188461304 lr:9.9970367992296e-05 tokens_per_second_per_gpu:5340.8006900416385 mfu:25.19346808948766 +Step 478 | loss:1.446315050125122 lr:9.997021853949025e-05 tokens_per_second_per_gpu:4157.5658779752675 mfu:19.611947600294375 +Step 479 | loss:1.555546760559082 lr:9.997006871085208e-05 tokens_per_second_per_gpu:5007.199330616099 mfu:23.61981355881625 +Step 480 | loss:1.5294324159622192 lr:9.996991850638268e-05 tokens_per_second_per_gpu:5331.936985701367 mfu:25.151656483810697 +Step 481 | loss:1.755823016166687 lr:9.996976792608312e-05 tokens_per_second_per_gpu:4106.828327025179 mfu:19.37260991574396 +Step 482 | loss:1.5623118877410889 lr:9.996961696995458e-05 tokens_per_second_per_gpu:5169.063969924843 mfu:24.383356679390833 +Step 483 | loss:1.5801191329956055 lr:9.996946563799818e-05 tokens_per_second_per_gpu:5243.514174798124 mfu:24.73455099829289 +Step 484 | loss:1.4915111064910889 lr:9.996931393021504e-05 tokens_per_second_per_gpu:5293.53874273743 mfu:24.970525420333995 +Step 485 | loss:1.5800511837005615 lr:9.996916184660631e-05 tokens_per_second_per_gpu:5331.748278518463 mfu:25.150766319832172 +Step 486 | loss:1.5662575960159302 lr:9.996900938717315e-05 tokens_per_second_per_gpu:4077.881067713951 mfu:19.236060754661178 +Step 487 | loss:1.3697993755340576 lr:9.996885655191668e-05 tokens_per_second_per_gpu:5196.724001979542 mfu:24.51383376988095 +Step 488 | loss:1.464963436126709 lr:9.996870334083807e-05 tokens_per_second_per_gpu:4469.907056496294 mfu:21.085314230278296 +Step 489 | loss:1.4855870008468628 lr:9.996854975393847e-05 tokens_per_second_per_gpu:5299.298213062936 mfu:24.997693824526372 +Step 490 | loss:1.34384024143219 lr:9.996839579121903e-05 tokens_per_second_per_gpu:5182.771312604057 mfu:24.448016553522503 +Step 491 | loss:1.16252863407135 lr:9.99682414526809e-05 tokens_per_second_per_gpu:4757.688271161797 mfu:22.442827320393782 +Step 492 | loss:1.3622504472732544 lr:9.996808673832526e-05 tokens_per_second_per_gpu:4284.547030030169 mfu:20.21093935975593 +Step 493 | loss:1.501899242401123 lr:9.996793164815325e-05 tokens_per_second_per_gpu:4387.62812050558 mfu:20.69719045097604 +Step 494 | loss:1.5833849906921387 lr:9.996777618216607e-05 tokens_per_second_per_gpu:4013.55199736619 mfu:18.93260955415961 +Step 495 | loss:1.4188214540481567 lr:9.996762034036486e-05 tokens_per_second_per_gpu:4134.6770564837025 mfu:19.503977124083022 +Step 496 | loss:1.5279145240783691 lr:9.99674641227508e-05 tokens_per_second_per_gpu:4011.967857908654 mfu:18.92513689805635 +Step 497 | loss:1.4954180717468262 lr:9.996730752932507e-05 tokens_per_second_per_gpu:4806.893163640124 mfu:22.674935193434393 +Step 498 | loss:1.3494651317596436 lr:9.996715056008885e-05 tokens_per_second_per_gpu:4566.084507937169 mfu:21.53899967829022 +Step 499 | loss:1.6805938482284546 lr:9.99669932150433e-05 tokens_per_second_per_gpu:4201.033218639089 mfu:19.81699046249868 +Step 500 | loss:1.3515243530273438 lr:9.996683549418963e-05 tokens_per_second_per_gpu:5368.308972264964 mfu:25.323229350132 +Step 501 | loss:1.4695074558258057 lr:9.9966677397529e-05 tokens_per_second_per_gpu:3477.5537421141153 mfu:16.40421433340313 +Step 502 | loss:1.475269079208374 lr:9.996651892506264e-05 tokens_per_second_per_gpu:3015.593013193078 mfu:14.225066756454721 +Step 503 | loss:1.7479525804519653 lr:9.99663600767917e-05 tokens_per_second_per_gpu:4117.697997406755 mfu:19.42388400549086 +Step 504 | loss:1.4868576526641846 lr:9.996620085271739e-05 tokens_per_second_per_gpu:5323.39738268452 mfu:25.111373719373717 +Step 505 | loss:1.6801795959472656 lr:9.99660412528409e-05 tokens_per_second_per_gpu:4116.556944758736 mfu:19.418501465466495 +Step 506 | loss:1.495944857597351 lr:9.996588127716344e-05 tokens_per_second_per_gpu:5066.375113593825 mfu:23.898955823551866 +Step 507 | loss:1.4623980522155762 lr:9.996572092568621e-05 tokens_per_second_per_gpu:5148.562216189717 mfu:24.286646409062353 +Step 508 | loss:1.7586923837661743 lr:9.996556019841043e-05 tokens_per_second_per_gpu:4625.639527794852 mfu:21.819930868093085 +Step 509 | loss:1.4998849630355835 lr:9.996539909533729e-05 tokens_per_second_per_gpu:4554.686401522182 mfu:21.485232865613238 +Step 510 | loss:1.7347557544708252 lr:9.996523761646802e-05 tokens_per_second_per_gpu:3670.4730954373053 mfu:17.314247838464425 +Step 511 | loss:1.6895431280136108 lr:9.996507576180381e-05 tokens_per_second_per_gpu:5283.483350028047 mfu:24.92309241729693 +Step 512 | loss:1.5797957181930542 lr:9.996491353134587e-05 tokens_per_second_per_gpu:3791.8937525646234 mfu:17.887009794634306 +Step 513 | loss:1.465261459350586 lr:9.996475092509547e-05 tokens_per_second_per_gpu:4136.94811030607 mfu:19.514690072444864 +Step 514 | loss:1.4339382648468018 lr:9.996458794305379e-05 tokens_per_second_per_gpu:4367.814047778853 mfu:20.603724089294662 +Step 515 | loss:1.6572500467300415 lr:9.996442458522208e-05 tokens_per_second_per_gpu:4100.674934229021 mfu:19.343583312047418 +Step 516 | loss:1.67189359664917 lr:9.996426085160153e-05 tokens_per_second_per_gpu:3726.5950608929616 mfu:17.578984724913447 +Step 517 | loss:1.476836919784546 lr:9.996409674219342e-05 tokens_per_second_per_gpu:5355.180122277427 mfu:25.261298324727996 +Step 518 | loss:1.594107747077942 lr:9.996393225699897e-05 tokens_per_second_per_gpu:5219.2760123609 mfu:24.620215450619554 +Step 519 | loss:1.5436575412750244 lr:9.996376739601939e-05 tokens_per_second_per_gpu:3576.0474665885267 mfu:16.868826036511177 +Step 520 | loss:1.361534833908081 lr:9.996360215925595e-05 tokens_per_second_per_gpu:5320.007595257505 mfu:25.09538351372311 +Step 521 | loss:1.5981637239456177 lr:9.996343654670988e-05 tokens_per_second_per_gpu:4607.4897263292505 mfu:21.734315157904387 +Step 522 | loss:1.8130524158477783 lr:9.996327055838244e-05 tokens_per_second_per_gpu:3988.053897983722 mfu:18.812330668948153 +Step 523 | loss:1.6073002815246582 lr:9.996310419427485e-05 tokens_per_second_per_gpu:4310.073886520125 mfu:20.331353897149878 +Step 524 | loss:1.4276680946350098 lr:9.996293745438838e-05 tokens_per_second_per_gpu:4260.9650569390915 mfu:20.09969916918582 +Step 525 | loss:1.6404269933700562 lr:9.99627703387243e-05 tokens_per_second_per_gpu:3799.5944557562084 mfu:17.92333532546446 +Step 526 | loss:1.3690496683120728 lr:9.996260284728383e-05 tokens_per_second_per_gpu:4799.7930783654 mfu:22.64144287979406 +Step 527 | loss:1.582012414932251 lr:9.996243498006826e-05 tokens_per_second_per_gpu:5341.13072256275 mfu:25.1950249091991 +Step 528 | loss:1.7796242237091064 lr:9.996226673707882e-05 tokens_per_second_per_gpu:5241.68509791503 mfu:24.725922930562643 +Step 529 | loss:1.4773002862930298 lr:9.996209811831683e-05 tokens_per_second_per_gpu:4760.743290527382 mfu:22.457238367981297 +Step 530 | loss:1.4163635969161987 lr:9.99619291237835e-05 tokens_per_second_per_gpu:5230.0832367374105 mfu:24.67119497190617 +Step 531 | loss:1.4736186265945435 lr:9.996175975348013e-05 tokens_per_second_per_gpu:4567.115479666822 mfu:21.543862947840907 +Step 532 | loss:1.701994776725769 lr:9.996159000740799e-05 tokens_per_second_per_gpu:5249.005490847771 mfu:24.760454473014203 +Step 533 | loss:1.6312142610549927 lr:9.996141988556836e-05 tokens_per_second_per_gpu:3474.389128642341 mfu:16.389286311718035 +Step 534 | loss:1.4183307886123657 lr:9.996124938796252e-05 tokens_per_second_per_gpu:4596.492987082526 mfu:21.682441662640404 +Step 535 | loss:1.391146183013916 lr:9.996107851459175e-05 tokens_per_second_per_gpu:5307.7617062956515 mfu:25.03761756612594 +Step 536 | loss:1.3032913208007812 lr:9.996090726545733e-05 tokens_per_second_per_gpu:5368.468344981125 mfu:25.32398113842592 +Step 537 | loss:1.4176400899887085 lr:9.996073564056054e-05 tokens_per_second_per_gpu:5341.907366470261 mfu:25.198688471019942 +Step 538 | loss:1.5004874467849731 lr:9.99605636399027e-05 tokens_per_second_per_gpu:5336.241194607266 mfu:25.17196017159359 +Step 539 | loss:1.359086036682129 lr:9.996039126348507e-05 tokens_per_second_per_gpu:4369.375984506685 mfu:20.611092011333735 +Step 540 | loss:1.59640634059906 lr:9.996021851130897e-05 tokens_per_second_per_gpu:4512.154120058535 mfu:21.284600837193707 +Step 541 | loss:1.6492400169372559 lr:9.996004538337569e-05 tokens_per_second_per_gpu:5164.616005954361 mfu:24.362374874441155 +Step 542 | loss:1.4635438919067383 lr:9.995987187968654e-05 tokens_per_second_per_gpu:4518.631380856524 mfu:21.31515518151247 +Step 543 | loss:1.6303563117980957 lr:9.99596980002428e-05 tokens_per_second_per_gpu:4732.762178748233 mfu:22.325246689649095 +Step 544 | loss:1.460953950881958 lr:9.995952374504581e-05 tokens_per_second_per_gpu:4133.084773607959 mfu:19.496466054086664 +Step 545 | loss:1.4712234735488892 lr:9.995934911409687e-05 tokens_per_second_per_gpu:3816.811394405324 mfu:18.004550562584978 +Step 546 | loss:1.5919684171676636 lr:9.995917410739728e-05 tokens_per_second_per_gpu:5065.744491878505 mfu:23.895981073325544 +Step 547 | loss:1.4935673475265503 lr:9.995899872494839e-05 tokens_per_second_per_gpu:5373.152071355096 mfu:25.346075074858067 +Step 548 | loss:1.4263869524002075 lr:9.995882296675148e-05 tokens_per_second_per_gpu:4605.035343964236 mfu:21.722737417527803 +Step 549 | loss:1.6298749446868896 lr:9.995864683280788e-05 tokens_per_second_per_gpu:4299.187648605827 mfu:20.28000165552442 +Step 550 | loss:1.3983356952667236 lr:9.995847032311892e-05 tokens_per_second_per_gpu:4608.012245625031 mfu:21.73677997057357 +Step 551 | loss:1.564378261566162 lr:9.995829343768594e-05 tokens_per_second_per_gpu:4838.449863716387 mfu:22.823793531822353 +Step 552 | loss:1.7055373191833496 lr:9.995811617651026e-05 tokens_per_second_per_gpu:3462.8010454413234 mfu:16.334623346127962 +Step 553 | loss:1.3496917486190796 lr:9.995793853959322e-05 tokens_per_second_per_gpu:3837.348337190379 mfu:18.10142682566584 +Step 554 | loss:1.5410743951797485 lr:9.995776052693614e-05 tokens_per_second_per_gpu:5316.829280211964 mfu:25.08039085937683 +Step 555 | loss:1.4656213521957397 lr:9.995758213854037e-05 tokens_per_second_per_gpu:4929.225647815671 mfu:23.251998393364513 +Step 556 | loss:1.7363373041152954 lr:9.995740337440726e-05 tokens_per_second_per_gpu:4279.1586405523685 mfu:20.1855214072351 +Step 557 | loss:1.4467570781707764 lr:9.995722423453813e-05 tokens_per_second_per_gpu:4889.497569688815 mfu:23.064594270531046 +Step 558 | loss:1.4625496864318848 lr:9.995704471893434e-05 tokens_per_second_per_gpu:4262.919952239294 mfu:20.108920743856405 +Step 559 | loss:1.3456463813781738 lr:9.995686482759725e-05 tokens_per_second_per_gpu:4389.101537553222 mfu:20.704140810580643 +Step 560 | loss:1.458852767944336 lr:9.995668456052821e-05 tokens_per_second_per_gpu:4375.7948786695215 mfu:20.641371030276108 +Step 561 | loss:1.4422543048858643 lr:9.995650391772857e-05 tokens_per_second_per_gpu:5369.281664226681 mfu:25.327817704073784 +Step 562 | loss:1.7294741868972778 lr:9.995632289919968e-05 tokens_per_second_per_gpu:3602.6528530736696 mfu:16.99432818390861 +Step 563 | loss:1.502153992652893 lr:9.995614150494293e-05 tokens_per_second_per_gpu:5065.594968193083 mfu:23.895275744590037 +Step 564 | loss:1.421813726425171 lr:9.995595973495964e-05 tokens_per_second_per_gpu:3208.6670930612127 mfu:15.135830133027593 +Step 565 | loss:1.5568281412124634 lr:9.995577758925122e-05 tokens_per_second_per_gpu:4539.377480310807 mfu:21.413018072287887 +Step 566 | loss:1.5113446712493896 lr:9.995559506781902e-05 tokens_per_second_per_gpu:5343.659786849236 mfu:25.20695493694132 +Step 567 | loss:1.46664559841156 lr:9.995541217066441e-05 tokens_per_second_per_gpu:3637.2948945902735 mfu:17.15774060428419 +Step 568 | loss:1.3735157251358032 lr:9.995522889778877e-05 tokens_per_second_per_gpu:5354.036445862728 mfu:25.255903407948544 +Step 569 | loss:1.570102572441101 lr:9.995504524919349e-05 tokens_per_second_per_gpu:4366.177529270803 mfu:20.59600435227334 +Step 570 | loss:1.4438406229019165 lr:9.995486122487993e-05 tokens_per_second_per_gpu:5359.4629326124295 mfu:25.281501071801 +Step 571 | loss:1.5706324577331543 lr:9.995467682484947e-05 tokens_per_second_per_gpu:5324.8571130298005 mfu:25.118259516468083 +Step 572 | loss:1.6317493915557861 lr:9.995449204910354e-05 tokens_per_second_per_gpu:4856.227149316022 mfu:22.907652021113524 +Step 573 | loss:1.4386409521102905 lr:9.995430689764349e-05 tokens_per_second_per_gpu:5318.682693399529 mfu:25.08913372560747 +Step 574 | loss:1.727479338645935 lr:9.995412137047072e-05 tokens_per_second_per_gpu:5020.490649480803 mfu:23.682510977633896 +Step 575 | loss:1.5198596715927124 lr:9.995393546758662e-05 tokens_per_second_per_gpu:4521.068306821581 mfu:21.32665057707209 +Step 576 | loss:1.3351908922195435 lr:9.995374918899262e-05 tokens_per_second_per_gpu:4422.23854574587 mfu:20.86045373198196 +Step 577 | loss:1.5175509452819824 lr:9.995356253469008e-05 tokens_per_second_per_gpu:4481.705420125312 mfu:21.140969124524783 +Step 578 | loss:1.659040927886963 lr:9.995337550468043e-05 tokens_per_second_per_gpu:4868.96626685222 mfu:22.967744612049344 +Step 579 | loss:1.6436080932617188 lr:9.995318809896508e-05 tokens_per_second_per_gpu:5089.8181778996495 mfu:24.00954075768166 +Step 580 | loss:1.5073193311691284 lr:9.995300031754541e-05 tokens_per_second_per_gpu:5310.238702001196 mfu:25.04930197748804 +Step 581 | loss:1.622155785560608 lr:9.995281216042285e-05 tokens_per_second_per_gpu:5331.008099915078 mfu:25.147274771072514 +Step 582 | loss:1.7155958414077759 lr:9.995262362759883e-05 tokens_per_second_per_gpu:4242.032696240774 mfu:20.010391993578036 +Step 583 | loss:1.373441219329834 lr:9.995243471907474e-05 tokens_per_second_per_gpu:4852.116578421599 mfu:22.888261756868815 +Step 584 | loss:1.4608663320541382 lr:9.995224543485202e-05 tokens_per_second_per_gpu:5327.054738295191 mfu:25.12862608979846 +Step 585 | loss:1.4698569774627686 lr:9.99520557749321e-05 tokens_per_second_per_gpu:5318.2392582816 mfu:25.087041966498365 +Step 586 | loss:1.5994106531143188 lr:9.995186573931637e-05 tokens_per_second_per_gpu:5330.408958772074 mfu:25.144448520076942 +Step 587 | loss:1.4276939630508423 lr:9.99516753280063e-05 tokens_per_second_per_gpu:4108.807400063329 mfu:19.381945540929593 +Step 588 | loss:1.5448493957519531 lr:9.99514845410033e-05 tokens_per_second_per_gpu:4447.752238527147 mfu:20.9808061739161 +Step 589 | loss:1.5614464282989502 lr:9.995129337830882e-05 tokens_per_second_per_gpu:5295.331232035971 mfu:24.978980898185423 +Step 590 | loss:1.4678095579147339 lr:9.995110183992429e-05 tokens_per_second_per_gpu:4836.649461171835 mfu:22.815300725839357 +Step 591 | loss:1.496068000793457 lr:9.995090992585113e-05 tokens_per_second_per_gpu:3635.0716512976896 mfu:17.14725318634865 +Step 592 | loss:1.6668192148208618 lr:9.995071763609081e-05 tokens_per_second_per_gpu:4929.7080386419475 mfu:23.25427391319642 +Step 593 | loss:1.3463358879089355 lr:9.995052497064479e-05 tokens_per_second_per_gpu:4399.806236220442 mfu:20.754636700603843 +Step 594 | loss:1.473853588104248 lr:9.995033192951448e-05 tokens_per_second_per_gpu:3988.6412739894545 mfu:18.815101421783876 +Step 595 | loss:1.4883118867874146 lr:9.995013851270134e-05 tokens_per_second_per_gpu:4548.667306776032 mfu:21.45683976873219 +Step 596 | loss:1.442442536354065 lr:9.994994472020685e-05 tokens_per_second_per_gpu:4001.2836025799 mfu:18.87473744274869 +Step 597 | loss:1.4787311553955078 lr:9.994975055203245e-05 tokens_per_second_per_gpu:4209.8458673672085 mfu:19.85856123014214 +Step 598 | loss:1.4291331768035889 lr:9.994955600817959e-05 tokens_per_second_per_gpu:3977.133439825909 mfu:18.76081700459546 +Step 599 | loss:1.6367565393447876 lr:9.994936108864976e-05 tokens_per_second_per_gpu:3801.67555959831 mfu:17.933152247360585 +Step 600 | loss:1.4224445819854736 lr:9.99491657934444e-05 tokens_per_second_per_gpu:4481.215733427231 mfu:21.138659188820768 +Step 601 | loss:1.545867919921875 lr:9.994897012256502e-05 tokens_per_second_per_gpu:4729.322956642977 mfu:22.309023292186808 +Step 602 | loss:1.5697028636932373 lr:9.994877407601303e-05 tokens_per_second_per_gpu:3739.5284908111357 mfu:17.63999392050816 +Step 603 | loss:1.399283528327942 lr:9.994857765378995e-05 tokens_per_second_per_gpu:4540.9656023969055 mfu:21.42050951512958 +Step 604 | loss:1.4430257081985474 lr:9.994838085589723e-05 tokens_per_second_per_gpu:4283.37526113243 mfu:20.205411925940908 +Step 605 | loss:1.453667163848877 lr:9.994818368233639e-05 tokens_per_second_per_gpu:5067.794352997157 mfu:23.9056506179646 +Step 606 | loss:1.5657998323440552 lr:9.994798613310887e-05 tokens_per_second_per_gpu:5380.030583944264 mfu:25.37852219233644 +Step 607 | loss:1.6304668188095093 lr:9.994778820821618e-05 tokens_per_second_per_gpu:4570.9672863152055 mfu:21.562032576987313 +Step 608 | loss:1.6398911476135254 lr:9.994758990765979e-05 tokens_per_second_per_gpu:3350.2482399019186 mfu:15.803692558909765 +Step 609 | loss:1.4503601789474487 lr:9.994739123144122e-05 tokens_per_second_per_gpu:4910.263251349103 mfu:23.162549533913253 +Step 610 | loss:1.5640630722045898 lr:9.994719217956192e-05 tokens_per_second_per_gpu:4259.7949509184855 mfu:20.09417957944595 +Step 611 | loss:1.411986231803894 lr:9.994699275202343e-05 tokens_per_second_per_gpu:4294.786709026048 mfu:20.25924167264888 +Step 612 | loss:1.5578091144561768 lr:9.994679294882724e-05 tokens_per_second_per_gpu:5327.763380501321 mfu:25.131968875991603 +Step 613 | loss:1.628849983215332 lr:9.994659276997485e-05 tokens_per_second_per_gpu:3407.41241157779 mfu:16.073345709918257 +Step 614 | loss:1.5731234550476074 lr:9.994639221546775e-05 tokens_per_second_per_gpu:4087.4115168459807 mfu:19.281017509279256 +Step 615 | loss:1.3884774446487427 lr:9.994619128530747e-05 tokens_per_second_per_gpu:4772.65585847273 mfu:22.513431983473183 +Step 616 | loss:1.5057753324508667 lr:9.994598997949551e-05 tokens_per_second_per_gpu:5156.571727478101 mfu:24.324428640373203 +Step 617 | loss:1.4624083042144775 lr:9.994578829803338e-05 tokens_per_second_per_gpu:5337.241041306898 mfu:25.17667661906719 +Step 618 | loss:1.659225583076477 lr:9.994558624092261e-05 tokens_per_second_per_gpu:3475.248609032917 mfu:16.393340627362857 +Step 619 | loss:1.6157715320587158 lr:9.99453838081647e-05 tokens_per_second_per_gpu:2974.070229076804 mfu:14.029196699260757 +Step 620 | loss:1.448649287223816 lr:9.99451809997612e-05 tokens_per_second_per_gpu:5340.698882752468 mfu:25.192987847134532 +Step 621 | loss:1.5510979890823364 lr:9.994497781571362e-05 tokens_per_second_per_gpu:4794.492112470327 mfu:22.616437319228954 +Step 622 | loss:1.5480544567108154 lr:9.994477425602348e-05 tokens_per_second_per_gpu:4281.823613874337 mfu:20.198092541086158 +Step 623 | loss:1.425801396369934 lr:9.994457032069233e-05 tokens_per_second_per_gpu:5318.132964647366 mfu:25.086540561290118 +Step 624 | loss:1.5092283487319946 lr:9.994436600972168e-05 tokens_per_second_per_gpu:3854.69922176528 mfu:18.183273908571824 +Step 625 | loss:1.4959975481033325 lr:9.994416132311309e-05 tokens_per_second_per_gpu:4295.636790973579 mfu:20.26325165423432 +Step 626 | loss:1.490522861480713 lr:9.994395626086809e-05 tokens_per_second_per_gpu:4009.6657222387016 mfu:18.914277331316338 +Step 627 | loss:1.2629133462905884 lr:9.994375082298822e-05 tokens_per_second_per_gpu:5359.283019771383 mfu:25.28065239223349 +Step 628 | loss:1.3456288576126099 lr:9.994354500947502e-05 tokens_per_second_per_gpu:5313.198746439761 mfu:25.06326501214041 +Step 629 | loss:1.6439021825790405 lr:9.994333882033007e-05 tokens_per_second_per_gpu:3231.680679655634 mfu:15.244389147516241 +Step 630 | loss:1.446799874305725 lr:9.994313225555487e-05 tokens_per_second_per_gpu:4757.078040050599 mfu:22.4399487561261 +Step 631 | loss:1.4092999696731567 lr:9.994292531515101e-05 tokens_per_second_per_gpu:4517.226507473255 mfu:21.308528154067474 +Step 632 | loss:1.3985376358032227 lr:9.994271799912002e-05 tokens_per_second_per_gpu:4523.105791600008 mfu:21.336261740402676 +Step 633 | loss:1.5415433645248413 lr:9.99425103074635e-05 tokens_per_second_per_gpu:4326.982636403193 mfu:20.41111535527853 +Step 634 | loss:1.3898361921310425 lr:9.994230224018297e-05 tokens_per_second_per_gpu:5372.813668934539 mfu:25.344478772902896 +Step 635 | loss:1.4092674255371094 lr:9.994209379728003e-05 tokens_per_second_per_gpu:5372.46099899366 mfu:25.342815168620067 +Step 636 | loss:1.6642663478851318 lr:9.994188497875622e-05 tokens_per_second_per_gpu:3604.419699164701 mfu:17.002662698375033 +Step 637 | loss:1.6966241598129272 lr:9.994167578461311e-05 tokens_per_second_per_gpu:3272.000462158139 mfu:15.43458444084482 +Step 638 | loss:1.4474818706512451 lr:9.99414662148523e-05 tokens_per_second_per_gpu:4505.078393444854 mfu:21.251223427513526 +Step 639 | loss:1.6160324811935425 lr:9.994125626947535e-05 tokens_per_second_per_gpu:4724.447268216 mfu:22.28602383799853 +Step 640 | loss:1.5209083557128906 lr:9.994104594848383e-05 tokens_per_second_per_gpu:4647.061514618439 mfu:21.92098203490778 +Step 641 | loss:1.365119457244873 lr:9.994083525187933e-05 tokens_per_second_per_gpu:4125.7973307999355 mfu:19.462089942995263 +Step 642 | loss:1.5897315740585327 lr:9.994062417966344e-05 tokens_per_second_per_gpu:5315.904956730492 mfu:25.076030667808872 +Step 643 | loss:1.3482712507247925 lr:9.994041273183774e-05 tokens_per_second_per_gpu:4731.001056477024 mfu:22.31693917541736 +Step 644 | loss:1.4826968908309937 lr:9.994020090840381e-05 tokens_per_second_per_gpu:3354.8711837072156 mfu:15.825499773594899 +Step 645 | loss:1.555200219154358 lr:9.993998870936329e-05 tokens_per_second_per_gpu:5218.075107806002 mfu:24.61455057893866 +Step 646 | loss:1.390417456626892 lr:9.993977613471773e-05 tokens_per_second_per_gpu:5390.859189436159 mfu:25.429602571992316 +Step 647 | loss:1.450840950012207 lr:9.993956318446873e-05 tokens_per_second_per_gpu:3918.189307838624 mfu:18.48276747710548 +Step 648 | loss:1.4328171014785767 lr:9.99393498586179e-05 tokens_per_second_per_gpu:4803.5557561330625 mfu:22.659192072803148 +Step 649 | loss:1.4492402076721191 lr:9.993913615716686e-05 tokens_per_second_per_gpu:4015.917886780947 mfu:18.943769858191377 +Step 650 | loss:1.72652006149292 lr:9.99389220801172e-05 tokens_per_second_per_gpu:5371.107038673405 mfu:25.336428306779975 +Step 651 | loss:1.6348607540130615 lr:9.993870762747053e-05 tokens_per_second_per_gpu:4745.367417786073 mfu:22.384707752866138 +Step 652 | loss:1.4264979362487793 lr:9.993849279922848e-05 tokens_per_second_per_gpu:4521.4999432694185 mfu:21.32868667984131 +Step 653 | loss:1.3872908353805542 lr:9.993827759539265e-05 tokens_per_second_per_gpu:4022.866460746873 mfu:18.97654746713762 +Step 654 | loss:1.5148814916610718 lr:9.993806201596465e-05 tokens_per_second_per_gpu:4781.62965596769 mfu:22.555762917344644 +Step 655 | loss:1.5659644603729248 lr:9.993784606094612e-05 tokens_per_second_per_gpu:5003.568462301112 mfu:23.60268613348444 +Step 656 | loss:1.5107142925262451 lr:9.993762973033869e-05 tokens_per_second_per_gpu:5312.199379987131 mfu:25.05855083007385 +Step 657 | loss:1.5981369018554688 lr:9.993741302414395e-05 tokens_per_second_per_gpu:5290.510753155505 mfu:24.956241876847756 +Step 658 | loss:1.3759665489196777 lr:9.993719594236357e-05 tokens_per_second_per_gpu:4962.828207891764 mfu:23.410507402431378 +Step 659 | loss:1.4984952211380005 lr:9.993697848499916e-05 tokens_per_second_per_gpu:4481.33706150202 mfu:21.139231514050255 +Step 660 | loss:1.419933557510376 lr:9.993676065205237e-05 tokens_per_second_per_gpu:5173.684071558138 mfu:24.405150487065708 +Step 661 | loss:1.619120478630066 lr:9.993654244352484e-05 tokens_per_second_per_gpu:3516.261011687167 mfu:16.5868031281204 +Step 662 | loss:1.5863521099090576 lr:9.99363238594182e-05 tokens_per_second_per_gpu:3975.1311618701684 mfu:18.75137191785438 +Step 663 | loss:1.4992353916168213 lr:9.993610489973408e-05 tokens_per_second_per_gpu:4400.872005724097 mfu:20.759664117191612 +Step 664 | loss:1.5865633487701416 lr:9.993588556447415e-05 tokens_per_second_per_gpu:4399.949304065105 mfu:20.755311576037005 +Step 665 | loss:1.5438122749328613 lr:9.993566585364006e-05 tokens_per_second_per_gpu:5310.956640036113 mfu:25.052688613691604 +Step 666 | loss:1.4085745811462402 lr:9.993544576723345e-05 tokens_per_second_per_gpu:5310.65436025936 mfu:25.051262708410416 +Step 667 | loss:1.5176565647125244 lr:9.993522530525599e-05 tokens_per_second_per_gpu:4389.734351935934 mfu:20.707125903993155 +Step 668 | loss:1.504757046699524 lr:9.993500446770931e-05 tokens_per_second_per_gpu:3728.941478535043 mfu:17.590053177271276 +Step 669 | loss:1.6370899677276611 lr:9.99347832545951e-05 tokens_per_second_per_gpu:5359.423985004203 mfu:25.281317349287846 +Step 670 | loss:1.60109281539917 lr:9.993456166591501e-05 tokens_per_second_per_gpu:3102.5076828239753 mfu:14.63505808227544 +Step 671 | loss:1.5752426385879517 lr:9.993433970167072e-05 tokens_per_second_per_gpu:5355.947811026135 mfu:25.26491964353686 +Step 672 | loss:1.299145221710205 lr:9.993411736186388e-05 tokens_per_second_per_gpu:4240.244752156627 mfu:20.0019579562784 +Step 673 | loss:1.332332730293274 lr:9.993389464649616e-05 tokens_per_second_per_gpu:4677.7184674123855 mfu:22.065596111852336 +Step 674 | loss:1.4310598373413086 lr:9.993367155556927e-05 tokens_per_second_per_gpu:4901.378632472045 mfu:23.12063927079788 +Step 675 | loss:1.361663818359375 lr:9.993344808908486e-05 tokens_per_second_per_gpu:4078.9345124560887 mfu:19.241030033246048 +Step 676 | loss:1.4475398063659668 lr:9.99332242470446e-05 tokens_per_second_per_gpu:3568.823668863914 mfu:16.834750150137193 +Step 677 | loss:1.5110599994659424 lr:9.99330000294502e-05 tokens_per_second_per_gpu:3906.229675200452 mfu:18.426351849427487 +Step 678 | loss:1.382034182548523 lr:9.993277543630334e-05 tokens_per_second_per_gpu:3601.2570307474207 mfu:16.98774385184385 +Step 679 | loss:1.4841328859329224 lr:9.99325504676057e-05 tokens_per_second_per_gpu:3821.9607608794554 mfu:18.028841002815973 +Step 680 | loss:1.4597811698913574 lr:9.993232512335897e-05 tokens_per_second_per_gpu:5361.231179267316 mfu:25.28984219296609 +Step 681 | loss:1.5956149101257324 lr:9.993209940356487e-05 tokens_per_second_per_gpu:5358.929853111315 mfu:25.2789864448415 +Step 682 | loss:1.483483076095581 lr:9.993187330822507e-05 tokens_per_second_per_gpu:5375.822333038558 mfu:25.35867115481258 +Step 683 | loss:1.3343780040740967 lr:9.993164683734126e-05 tokens_per_second_per_gpu:3940.0165619388586 mfu:18.585730358809858 +Step 684 | loss:1.475881576538086 lr:9.993141999091518e-05 tokens_per_second_per_gpu:4963.264041257252 mfu:23.412563302777226 +Step 685 | loss:1.5443334579467773 lr:9.993119276894851e-05 tokens_per_second_per_gpu:4996.344186228302 mfu:23.568607990660396 +Step 686 | loss:1.5392162799835205 lr:9.993096517144299e-05 tokens_per_second_per_gpu:5382.056310865979 mfu:25.38807789185082 +Step 687 | loss:1.6266226768493652 lr:9.993073719840029e-05 tokens_per_second_per_gpu:4114.674926338443 mfu:19.40962366347252 +Step 688 | loss:1.4669727087020874 lr:9.993050884982214e-05 tokens_per_second_per_gpu:5312.013066573312 mfu:25.057671957912525 +Step 689 | loss:1.458366870880127 lr:9.993028012571028e-05 tokens_per_second_per_gpu:5333.303132343072 mfu:25.15810083060827 +Step 690 | loss:1.4081376791000366 lr:9.99300510260664e-05 tokens_per_second_per_gpu:4462.966129428351 mfu:21.052572693054536 +Step 691 | loss:1.5906598567962646 lr:9.992982155089223e-05 tokens_per_second_per_gpu:5359.797406704562 mfu:25.28307884316075 +Step 692 | loss:1.3748722076416016 lr:9.992959170018952e-05 tokens_per_second_per_gpu:4644.984689731052 mfu:21.91118529757111 +Step 693 | loss:1.5953267812728882 lr:9.992936147395997e-05 tokens_per_second_per_gpu:3837.465068799593 mfu:18.10197746858297 +Step 694 | loss:1.5253548622131348 lr:9.992913087220532e-05 tokens_per_second_per_gpu:4312.309139528197 mfu:20.341897966962442 +Step 695 | loss:1.468902826309204 lr:9.99288998949273e-05 tokens_per_second_per_gpu:3759.3341038396934 mfu:17.7334203763499 +Step 696 | loss:1.568112850189209 lr:9.992866854212766e-05 tokens_per_second_per_gpu:5297.499069855184 mfu:24.98920695904992 +Step 697 | loss:1.503315806388855 lr:9.992843681380813e-05 tokens_per_second_per_gpu:5340.351362084072 mfu:25.191348532848565 +Step 698 | loss:1.465664029121399 lr:9.992820470997046e-05 tokens_per_second_per_gpu:4923.683430906742 mfu:23.22585481060538 +Step 699 | loss:1.5401784181594849 lr:9.992797223061639e-05 tokens_per_second_per_gpu:3987.7437544540935 mfu:18.810867669002953 +Step 700 | loss:1.4908329248428345 lr:9.992773937574766e-05 tokens_per_second_per_gpu:4332.415990741496 mfu:20.436745414718143 +Step 701 | loss:1.5889136791229248 lr:9.992750614536605e-05 tokens_per_second_per_gpu:4154.231036517193 mfu:19.596216584154014 +Step 702 | loss:1.505631923675537 lr:9.99272725394733e-05 tokens_per_second_per_gpu:4477.171944084673 mfu:21.119583944550318 +Step 703 | loss:1.483835220336914 lr:9.992703855807113e-05 tokens_per_second_per_gpu:5064.396904805128 mfu:23.88962427517836 +Step 704 | loss:1.3668854236602783 lr:9.992680420116136e-05 tokens_per_second_per_gpu:4634.681768919202 mfu:21.862584662242064 +Step 705 | loss:1.4583110809326172 lr:9.992656946874571e-05 tokens_per_second_per_gpu:5337.997287858166 mfu:25.180243963078482 +Step 706 | loss:1.4918352365493774 lr:9.992633436082596e-05 tokens_per_second_per_gpu:4994.274349525332 mfu:23.558844217781928 +Step 707 | loss:1.5751063823699951 lr:9.99260988774039e-05 tokens_per_second_per_gpu:5358.7025279951595 mfu:25.277914113482932 +Step 708 | loss:1.2685396671295166 lr:9.992586301848125e-05 tokens_per_second_per_gpu:4933.8725032694365 mfu:23.27391840337522 +Step 709 | loss:1.4512611627578735 lr:9.992562678405984e-05 tokens_per_second_per_gpu:4407.564644136344 mfu:20.79123443446404 +Step 710 | loss:1.2987602949142456 lr:9.99253901741414e-05 tokens_per_second_per_gpu:4192.07960143052 mfu:19.774754722481237 +Step 711 | loss:1.5945231914520264 lr:9.992515318872775e-05 tokens_per_second_per_gpu:4269.312031477437 mfu:20.139073272223904 +Step 712 | loss:1.481652855873108 lr:9.992491582782063e-05 tokens_per_second_per_gpu:5087.384539107869 mfu:23.99806086827924 +Step 713 | loss:1.6192872524261475 lr:9.992467809142187e-05 tokens_per_second_per_gpu:4077.8594800144247 mfu:19.235958921799266 +Step 714 | loss:1.5695619583129883 lr:9.992443997953321e-05 tokens_per_second_per_gpu:5331.753908384624 mfu:25.15079287687141 +Step 715 | loss:1.438236951828003 lr:9.992420149215648e-05 tokens_per_second_per_gpu:5327.028459894587 mfu:25.128502130099257 +Step 716 | loss:1.5717909336090088 lr:9.992396262929347e-05 tokens_per_second_per_gpu:3877.773794034958 mfu:18.29212060289588 +Step 717 | loss:1.6316299438476562 lr:9.992372339094595e-05 tokens_per_second_per_gpu:4323.786303260807 mfu:20.39603770649522 +Step 718 | loss:1.516409993171692 lr:9.992348377711575e-05 tokens_per_second_per_gpu:5338.3242671233775 mfu:25.181786380060906 +Step 719 | loss:1.6474218368530273 lr:9.992324378780464e-05 tokens_per_second_per_gpu:4479.322831454931 mfu:21.129730047255048 +Step 720 | loss:1.3388638496398926 lr:9.992300342301446e-05 tokens_per_second_per_gpu:4507.22031553519 mfu:21.26132723946323 +Step 721 | loss:1.3406926393508911 lr:9.9922762682747e-05 tokens_per_second_per_gpu:5350.767710757175 mfu:25.240484226751935 +Step 722 | loss:1.4314384460449219 lr:9.992252156700407e-05 tokens_per_second_per_gpu:4591.653571274481 mfu:21.659613312584376 +Step 723 | loss:1.5735440254211426 lr:9.992228007578748e-05 tokens_per_second_per_gpu:5156.591509417446 mfu:24.324521955156243 +Step 724 | loss:1.507493019104004 lr:9.992203820909906e-05 tokens_per_second_per_gpu:4263.1308609748285 mfu:20.109915636346667 +Step 725 | loss:1.6445355415344238 lr:9.99217959669406e-05 tokens_per_second_per_gpu:5306.9287649945145 mfu:25.03368843989404 +Step 726 | loss:1.4099631309509277 lr:9.992155334931397e-05 tokens_per_second_per_gpu:4164.039504846396 mfu:19.642484802760126 +Step 727 | loss:1.4722791910171509 lr:9.992131035622094e-05 tokens_per_second_per_gpu:5361.702468877776 mfu:25.292065346468927 +Step 728 | loss:1.6595864295959473 lr:9.992106698766339e-05 tokens_per_second_per_gpu:3792.997740546796 mfu:17.89221749430599 +Step 729 | loss:1.5996758937835693 lr:9.992082324364312e-05 tokens_per_second_per_gpu:4510.621982877863 mfu:21.277373484702917 +Step 730 | loss:1.5909101963043213 lr:9.992057912416195e-05 tokens_per_second_per_gpu:5099.696845395817 mfu:24.056140117735556 +Step 731 | loss:1.6098237037658691 lr:9.992033462922176e-05 tokens_per_second_per_gpu:3100.480317577898 mfu:14.625494654505513 +Step 732 | loss:1.5880376100540161 lr:9.992008975882435e-05 tokens_per_second_per_gpu:4815.466906258864 mfu:22.715379000198322 +Step 733 | loss:1.4911988973617554 lr:9.991984451297157e-05 tokens_per_second_per_gpu:3730.9366278228968 mfu:17.59946463686965 +Step 734 | loss:1.4198412895202637 lr:9.991959889166528e-05 tokens_per_second_per_gpu:4799.539006464897 mfu:22.64024437929025 +Step 735 | loss:1.4208405017852783 lr:9.991935289490731e-05 tokens_per_second_per_gpu:5348.272375307964 mfu:25.228713303689876 +Step 736 | loss:1.3981095552444458 lr:9.991910652269952e-05 tokens_per_second_per_gpu:5374.54557995508 mfu:25.352648492681276 +Step 737 | loss:1.6092491149902344 lr:9.991885977504376e-05 tokens_per_second_per_gpu:5049.291910727555 mfu:23.81837144093671 +Step 738 | loss:1.326593041419983 lr:9.991861265194189e-05 tokens_per_second_per_gpu:5334.949223641281 mfu:25.16586572411406 +Step 739 | loss:1.5628348588943481 lr:9.991836515339575e-05 tokens_per_second_per_gpu:4103.29300958158 mfu:19.35593322017515 +Step 740 | loss:1.3488599061965942 lr:9.991811727940723e-05 tokens_per_second_per_gpu:4376.372287665176 mfu:20.644094767024743 +Step 741 | loss:1.3948118686676025 lr:9.991786902997817e-05 tokens_per_second_per_gpu:5128.492366576242 mfu:24.191973504166146 +Step 742 | loss:1.5592793226242065 lr:9.991762040511046e-05 tokens_per_second_per_gpu:4008.657805170314 mfu:18.909522814536373 +Step 743 | loss:1.3589632511138916 lr:9.991737140480594e-05 tokens_per_second_per_gpu:5345.985726080373 mfu:25.21792678914074 +Step 744 | loss:1.2865005731582642 lr:9.991712202906652e-05 tokens_per_second_per_gpu:5301.349058985372 mfu:25.007368014652897 +Step 745 | loss:1.40797758102417 lr:9.991687227789406e-05 tokens_per_second_per_gpu:5328.041200039692 mfu:25.133279398155 +Step 746 | loss:1.455317735671997 lr:9.99166221512904e-05 tokens_per_second_per_gpu:3724.1813470601455 mfu:17.567598824942007 +Step 747 | loss:1.519628882408142 lr:9.99163716492575e-05 tokens_per_second_per_gpu:5288.585578161549 mfu:24.947160497932547 +Step 748 | loss:1.340059518814087 lr:9.991612077179717e-05 tokens_per_second_per_gpu:5170.658413759101 mfu:24.390877942997193 +Step 749 | loss:1.5980961322784424 lr:9.991586951891134e-05 tokens_per_second_per_gpu:3510.6597988493127 mfu:16.560381251498786 +Step 750 | loss:1.513377070426941 lr:9.991561789060186e-05 tokens_per_second_per_gpu:4862.799792835019 mfu:22.93865630200125 +Step 751 | loss:1.444960355758667 lr:9.991536588687067e-05 tokens_per_second_per_gpu:4306.425729996799 mfu:20.314144920389666 +Step 752 | loss:1.6093581914901733 lr:9.991511350771964e-05 tokens_per_second_per_gpu:3937.1688214391706 mfu:18.5722970809982 +Step 753 | loss:1.4575772285461426 lr:9.991486075315069e-05 tokens_per_second_per_gpu:5352.452196394353 mfu:25.248430232905392 +Step 754 | loss:1.3955490589141846 lr:9.991460762316568e-05 tokens_per_second_per_gpu:5313.039427864056 mfu:25.06251347923626 +Step 755 | loss:1.4367924928665161 lr:9.991435411776654e-05 tokens_per_second_per_gpu:5252.792099217321 mfu:24.77831655075538 +Step 756 | loss:1.4112962484359741 lr:9.991410023695518e-05 tokens_per_second_per_gpu:4560.808082257091 mfu:21.51410987810624 +Step 757 | loss:1.5443147420883179 lr:9.99138459807335e-05 tokens_per_second_per_gpu:3917.5748026608494 mfu:18.479868751336525 +Step 758 | loss:1.359242558479309 lr:9.991359134910341e-05 tokens_per_second_per_gpu:4627.726809936007 mfu:21.829776934079966 +Step 759 | loss:1.593669056892395 lr:9.991333634206683e-05 tokens_per_second_per_gpu:4002.8436691011966 mfu:18.88209654270444 +Step 760 | loss:1.5306469202041626 lr:9.99130809596257e-05 tokens_per_second_per_gpu:3789.128798737018 mfu:17.873967035679726 +Step 761 | loss:1.5327104330062866 lr:9.99128252017819e-05 tokens_per_second_per_gpu:4310.07871623277 mfu:20.331376679728457 +Step 762 | loss:1.4119277000427246 lr:9.991256906853737e-05 tokens_per_second_per_gpu:4093.645941659043 mfu:19.310426354825033 +Step 763 | loss:1.4954489469528198 lr:9.991231255989405e-05 tokens_per_second_per_gpu:3434.6840257682707 mfu:16.20199056707194 +Step 764 | loss:1.508272647857666 lr:9.991205567585385e-05 tokens_per_second_per_gpu:5195.3074030204725 mfu:24.507151430894314 +Step 765 | loss:1.3113423585891724 lr:9.991179841641871e-05 tokens_per_second_per_gpu:4677.125093322362 mfu:22.06279706502964 +Step 766 | loss:1.5106689929962158 lr:9.991154078159057e-05 tokens_per_second_per_gpu:4079.1831034883394 mfu:19.242202679559178 +Step 767 | loss:1.6487873792648315 lr:9.991128277137137e-05 tokens_per_second_per_gpu:3899.902362684452 mfu:18.39650483673868 +Step 768 | loss:1.4891937971115112 lr:9.991102438576303e-05 tokens_per_second_per_gpu:4172.14915092834 mfu:19.680739386976956 +Step 769 | loss:1.4126560688018799 lr:9.991076562476751e-05 tokens_per_second_per_gpu:5264.9627910125155 mfu:24.835727780487456 +Step 770 | loss:1.2479087114334106 lr:9.991050648838675e-05 tokens_per_second_per_gpu:5344.885045230441 mfu:25.2127346897761 +Step 771 | loss:1.5366331338882446 lr:9.991024697662272e-05 tokens_per_second_per_gpu:4808.010044737846 mfu:22.680203712131917 +Step 772 | loss:1.4193477630615234 lr:9.990998708947734e-05 tokens_per_second_per_gpu:4327.907378468996 mfu:20.415477521380186 +Step 773 | loss:1.4690288305282593 lr:9.990972682695258e-05 tokens_per_second_per_gpu:4092.4053277243556 mfu:19.304574167224548 +Step 774 | loss:1.5285284519195557 lr:9.99094661890504e-05 tokens_per_second_per_gpu:4058.2102538545 mfu:19.143270169499225 +Step 775 | loss:1.5696258544921875 lr:9.990920517577276e-05 tokens_per_second_per_gpu:5391.088167305825 mfu:25.430682699671184 +Step 776 | loss:1.646986961364746 lr:9.990894378712161e-05 tokens_per_second_per_gpu:4826.064910713062 mfu:22.765371595414795 +Step 777 | loss:1.5293054580688477 lr:9.990868202309892e-05 tokens_per_second_per_gpu:4738.926554685785 mfu:22.354325102697516 +Step 778 | loss:1.4598788022994995 lr:9.990841988370667e-05 tokens_per_second_per_gpu:5334.77079072522 mfu:25.165024025605202 +Step 779 | loss:1.4200788736343384 lr:9.990815736894683e-05 tokens_per_second_per_gpu:3761.4909994744603 mfu:17.74359482106344 +Step 780 | loss:1.5373433828353882 lr:9.990789447882137e-05 tokens_per_second_per_gpu:4694.297506631047 mfu:22.14380226852254 +Step 781 | loss:1.3414479494094849 lr:9.990763121333226e-05 tokens_per_second_per_gpu:5362.757607936977 mfu:25.297042617436464 +Step 782 | loss:1.4215402603149414 lr:9.990736757248149e-05 tokens_per_second_per_gpu:5326.979016800681 mfu:25.12826889858944 +Step 783 | loss:1.700514793395996 lr:9.990710355627104e-05 tokens_per_second_per_gpu:5361.116881594943 mfu:25.28930303134431 +Step 784 | loss:1.3592796325683594 lr:9.990683916470291e-05 tokens_per_second_per_gpu:4809.16740133857 mfu:22.685663160682928 +Step 785 | loss:1.4325053691864014 lr:9.990657439777906e-05 tokens_per_second_per_gpu:3223.629819856614 mfu:15.206411868225942 +Step 786 | loss:1.4772863388061523 lr:9.99063092555015e-05 tokens_per_second_per_gpu:5316.95804836559 mfu:25.080998280727595 +Step 787 | loss:1.4092093706130981 lr:9.990604373787223e-05 tokens_per_second_per_gpu:4279.706833260885 mfu:20.188107325772492 +Step 788 | loss:1.3189047574996948 lr:9.99057778448932e-05 tokens_per_second_per_gpu:5285.233879967247 mfu:24.93134996569134 +Step 789 | loss:1.3814822435379028 lr:9.99055115765665e-05 tokens_per_second_per_gpu:4124.432148154952 mfu:19.4556501435347 +Step 790 | loss:1.567383050918579 lr:9.990524493289403e-05 tokens_per_second_per_gpu:4221.486940493184 mfu:19.91347415824925 +Step 791 | loss:1.4396783113479614 lr:9.990497791387787e-05 tokens_per_second_per_gpu:4004.939329184062 mfu:18.891982128871792 +Step 792 | loss:1.450468897819519 lr:9.990471051952001e-05 tokens_per_second_per_gpu:4669.7253116688735 mfu:22.027891032437193 +Step 793 | loss:1.5753775835037231 lr:9.990444274982245e-05 tokens_per_second_per_gpu:5339.0578926137505 mfu:25.18524701666841 +Step 794 | loss:1.6911855936050415 lr:9.99041746047872e-05 tokens_per_second_per_gpu:5383.204073812025 mfu:25.393492085495872 +Step 795 | loss:1.3216288089752197 lr:9.990390608441627e-05 tokens_per_second_per_gpu:5352.88844811425 mfu:25.250488106700363 +Step 796 | loss:1.4539576768875122 lr:9.990363718871172e-05 tokens_per_second_per_gpu:5318.439252634639 mfu:25.087985373984207 +Step 797 | loss:1.6221623420715332 lr:9.990336791767554e-05 tokens_per_second_per_gpu:3914.1814319407667 mfu:18.46386164268115 +Step 798 | loss:1.6024926900863647 lr:9.990309827130977e-05 tokens_per_second_per_gpu:5255.4112441261095 mfu:24.790671504162336 +Step 799 | loss:1.4477663040161133 lr:9.990282824961641e-05 tokens_per_second_per_gpu:5183.839292654047 mfu:24.453054397634496 +Step 800 | loss:1.4238747358322144 lr:9.990255785259752e-05 tokens_per_second_per_gpu:4092.8313368970817 mfu:19.306583725176953 +Step 801 | loss:1.7793495655059814 lr:9.990228708025514e-05 tokens_per_second_per_gpu:4677.418714715789 mfu:22.06418212723941 +Step 802 | loss:1.4548916816711426 lr:9.990201593259127e-05 tokens_per_second_per_gpu:4923.69954586941 mfu:23.225930827633782 +Step 803 | loss:1.3536447286605835 lr:9.990174440960798e-05 tokens_per_second_per_gpu:5353.046838120428 mfu:25.251235259383627 +Step 804 | loss:1.3944090604782104 lr:9.990147251130731e-05 tokens_per_second_per_gpu:4229.075046857938 mfu:19.949268550636244 +Step 805 | loss:1.5407716035842896 lr:9.990120023769129e-05 tokens_per_second_per_gpu:5063.849510739451 mfu:23.887042123975508 +Step 806 | loss:1.6497595310211182 lr:9.990092758876197e-05 tokens_per_second_per_gpu:3617.2555004965852 mfu:17.063211363271066 +Step 807 | loss:1.3944917917251587 lr:9.990065456452142e-05 tokens_per_second_per_gpu:4049.56056873099 mfu:19.102468128983165 +Step 808 | loss:1.5762859582901 lr:9.990038116497167e-05 tokens_per_second_per_gpu:4410.141444825047 mfu:20.803389642960187 +Step 809 | loss:1.717095136642456 lr:9.990010739011477e-05 tokens_per_second_per_gpu:4913.446832276225 mfu:23.17756702832141 +Step 810 | loss:1.6439449787139893 lr:9.989983323995283e-05 tokens_per_second_per_gpu:4338.909142705513 mfu:20.467374720378373 +Step 811 | loss:1.461789846420288 lr:9.989955871448786e-05 tokens_per_second_per_gpu:4919.6676843158675 mfu:23.206911848779207 +Step 812 | loss:1.352271556854248 lr:9.989928381372193e-05 tokens_per_second_per_gpu:5340.017582628169 mfu:25.189774038206423 +Step 813 | loss:1.5266406536102295 lr:9.989900853765713e-05 tokens_per_second_per_gpu:4470.10502554089 mfu:21.086248084038658 +Step 814 | loss:1.5110217332839966 lr:9.989873288629553e-05 tokens_per_second_per_gpu:4582.312011096812 mfu:21.615547579391794 +Step 815 | loss:1.6428472995758057 lr:9.989845685963916e-05 tokens_per_second_per_gpu:4381.4043575566375 mfu:20.667831899262424 +Step 816 | loss:1.5042662620544434 lr:9.989818045769017e-05 tokens_per_second_per_gpu:4015.5135108237064 mfu:18.941862347807422 +Step 817 | loss:1.611359715461731 lr:9.989790368045058e-05 tokens_per_second_per_gpu:4556.2675092815125 mfu:21.492691220678804 +Step 818 | loss:1.599743127822876 lr:9.989762652792249e-05 tokens_per_second_per_gpu:3957.729480585731 mfu:18.66928521820271 +Step 819 | loss:1.4256173372268677 lr:9.989734900010797e-05 tokens_per_second_per_gpu:5221.330657252728 mfu:24.629907561132878 +Step 820 | loss:1.4982625246047974 lr:9.989707109700915e-05 tokens_per_second_per_gpu:3619.961903320195 mfu:17.07597792714997 +Step 821 | loss:1.3718540668487549 lr:9.989679281862807e-05 tokens_per_second_per_gpu:5322.882425865981 mfu:25.108944580199967 +Step 822 | loss:1.676112174987793 lr:9.989651416496684e-05 tokens_per_second_per_gpu:4799.9880836402845 mfu:22.642362752947022 +Step 823 | loss:1.3926951885223389 lr:9.989623513602757e-05 tokens_per_second_per_gpu:3813.1067389156246 mfu:17.987075070560664 +Step 824 | loss:1.5328117609024048 lr:9.989595573181235e-05 tokens_per_second_per_gpu:3150.5722935036924 mfu:14.861787051519872 +Step 825 | loss:1.643311619758606 lr:9.989567595232328e-05 tokens_per_second_per_gpu:5379.14321652911 mfu:25.374336328838663 +Step 826 | loss:1.4570138454437256 lr:9.989539579756247e-05 tokens_per_second_per_gpu:4118.27669223352 mfu:19.42661380772423 +Step 827 | loss:1.4551821947097778 lr:9.989511526753202e-05 tokens_per_second_per_gpu:5332.539206975611 mfu:25.154497264310848 +Step 828 | loss:1.5451467037200928 lr:9.989483436223402e-05 tokens_per_second_per_gpu:4097.06992719701 mfu:19.326577878799792 +Step 829 | loss:1.4643867015838623 lr:9.989455308167062e-05 tokens_per_second_per_gpu:5224.538737624054 mfu:24.645040623599375 +Step 830 | loss:1.439710021018982 lr:9.989427142584392e-05 tokens_per_second_per_gpu:5369.448936596249 mfu:25.328606756380644 +Step 831 | loss:1.4925227165222168 lr:9.989398939475605e-05 tokens_per_second_per_gpu:3812.4503401942493 mfu:17.983978725798767 +Step 832 | loss:1.552917242050171 lr:9.98937069884091e-05 tokens_per_second_per_gpu:3719.8073622548404 mfu:17.54696599233662 +Step 833 | loss:1.4380176067352295 lr:9.989342420680522e-05 tokens_per_second_per_gpu:5378.202390727227 mfu:25.369898293009747 +Step 834 | loss:1.294582724571228 lr:9.989314104994653e-05 tokens_per_second_per_gpu:5339.508409630738 mfu:25.18737218230367 +Step 835 | loss:1.447312355041504 lr:9.989285751783515e-05 tokens_per_second_per_gpu:5343.120596195104 mfu:25.204411482630363 +Step 836 | loss:1.5027756690979004 lr:9.989257361047323e-05 tokens_per_second_per_gpu:4520.193937276363 mfu:21.322526026744466 +Step 837 | loss:1.5416289567947388 lr:9.989228932786291e-05 tokens_per_second_per_gpu:4912.812749564002 mfu:23.174575951982185 +Step 838 | loss:1.5275640487670898 lr:9.98920046700063e-05 tokens_per_second_per_gpu:5319.627066570406 mfu:25.093588494981173 +Step 839 | loss:1.5449949502944946 lr:9.989171963690556e-05 tokens_per_second_per_gpu:4966.00173864522 mfu:23.425477488455613 +Step 840 | loss:1.3246716260910034 lr:9.989143422856284e-05 tokens_per_second_per_gpu:3868.96426424053 mfu:18.250564547794795 +Step 841 | loss:1.6072688102722168 lr:9.989114844498026e-05 tokens_per_second_per_gpu:5382.85300816967 mfu:25.391836048962908 +Step 842 | loss:1.489867925643921 lr:9.989086228616e-05 tokens_per_second_per_gpu:5155.968071679818 mfu:24.321581092978406 +Step 843 | loss:1.4686732292175293 lr:9.98905757521042e-05 tokens_per_second_per_gpu:3607.6075258993164 mfu:17.017700220981542 +Step 844 | loss:1.3862800598144531 lr:9.989028884281502e-05 tokens_per_second_per_gpu:5343.311491994754 mfu:25.205311970687493 +Step 845 | loss:1.8998146057128906 lr:9.989000155829458e-05 tokens_per_second_per_gpu:5374.4002585565095 mfu:25.35196298684983 +Step 846 | loss:1.568395972251892 lr:9.98897138985451e-05 tokens_per_second_per_gpu:4187.9209723182485 mfu:19.755137759423366 +Step 847 | loss:1.3404215574264526 lr:9.988942586356871e-05 tokens_per_second_per_gpu:3948.1002399948998 mfu:18.623862447417903 +Step 848 | loss:1.310103416442871 lr:9.988913745336758e-05 tokens_per_second_per_gpu:5347.302818180885 mfu:25.224139737297943 +Step 849 | loss:1.2952079772949219 lr:9.988884866794388e-05 tokens_per_second_per_gpu:4874.797885678107 mfu:22.995253353027827 +Step 850 | loss:1.4687070846557617 lr:9.988855950729979e-05 tokens_per_second_per_gpu:4082.643834712172 mfu:19.258527539203627 +Step 851 | loss:1.452724814414978 lr:9.988826997143747e-05 tokens_per_second_per_gpu:4579.312778743418 mfu:21.601399688659026 +Step 852 | loss:1.4617140293121338 lr:9.988798006035911e-05 tokens_per_second_per_gpu:3698.2337451988974 mfu:17.44519955984511 +Step 853 | loss:1.7070796489715576 lr:9.988768977406689e-05 tokens_per_second_per_gpu:3716.325312713438 mfu:17.53054057055091 +Step 854 | loss:1.5736373662948608 lr:9.988739911256297e-05 tokens_per_second_per_gpu:5325.957306300935 mfu:25.123449315840347 +Step 855 | loss:1.4427390098571777 lr:9.988710807584958e-05 tokens_per_second_per_gpu:5337.864737437189 mfu:25.179618700127307 +Step 856 | loss:1.531618356704712 lr:9.988681666392888e-05 tokens_per_second_per_gpu:5354.050578178407 mfu:25.255970072492826 +Step 857 | loss:1.5073434114456177 lr:9.988652487680306e-05 tokens_per_second_per_gpu:4022.5176301663264 mfu:18.974901973772546 +Step 858 | loss:1.627528190612793 lr:9.988623271447432e-05 tokens_per_second_per_gpu:4702.73261451123 mfu:22.18359211157117 +Step 859 | loss:1.2903072834014893 lr:9.988594017694487e-05 tokens_per_second_per_gpu:5363.932677745344 mfu:25.302585622210117 +Step 860 | loss:1.4518555402755737 lr:9.988564726421687e-05 tokens_per_second_per_gpu:5342.359235726344 mfu:25.20082001539773 +Step 861 | loss:1.295102596282959 lr:9.988535397629257e-05 tokens_per_second_per_gpu:5341.478154175943 mfu:25.196663803396476 +Step 862 | loss:1.560106873512268 lr:9.988506031317416e-05 tokens_per_second_per_gpu:3649.3159992225824 mfu:17.214446205846734 +Step 863 | loss:1.641174077987671 lr:9.988476627486384e-05 tokens_per_second_per_gpu:3017.4529183416175 mfu:14.233840246372925 +Step 864 | loss:1.5691109895706177 lr:9.988447186136383e-05 tokens_per_second_per_gpu:4245.037263352802 mfu:20.02456504927748 +Step 865 | loss:1.4877846240997314 lr:9.988417707267634e-05 tokens_per_second_per_gpu:5375.85107464889 mfu:25.35880673389234 +Step 866 | loss:1.3936728239059448 lr:9.98838819088036e-05 tokens_per_second_per_gpu:4282.7560486923285 mfu:20.20249099521241 +Step 867 | loss:1.4360507726669312 lr:9.98835863697478e-05 tokens_per_second_per_gpu:4522.630834638735 mfu:21.334021287380338 +Step 868 | loss:1.5814566612243652 lr:9.98832904555112e-05 tokens_per_second_per_gpu:5019.327224447999 mfu:23.67702290325338 +Step 869 | loss:1.4146116971969604 lr:9.9882994166096e-05 tokens_per_second_per_gpu:5327.211777370452 mfu:25.129366869910093 +Step 870 | loss:1.5604135990142822 lr:9.988269750150443e-05 tokens_per_second_per_gpu:4484.5936479377015 mfu:21.154593387898306 +Step 871 | loss:1.5328515768051147 lr:9.988240046173874e-05 tokens_per_second_per_gpu:4424.508212984718 mfu:20.87116013959276 +Step 872 | loss:1.211063265800476 lr:9.988210304680114e-05 tokens_per_second_per_gpu:5346.284762586238 mfu:25.219337395359798 +Step 873 | loss:1.3718396425247192 lr:9.988180525669389e-05 tokens_per_second_per_gpu:4980.197302012089 mfu:23.49244038286999 +Step 874 | loss:1.5821503400802612 lr:9.988150709141919e-05 tokens_per_second_per_gpu:3776.6802357050324 mfu:17.815245039913982 +Step 875 | loss:1.4830392599105835 lr:9.988120855097933e-05 tokens_per_second_per_gpu:4735.869992662286 mfu:22.339906778129414 +Step 876 | loss:1.4331774711608887 lr:9.988090963537654e-05 tokens_per_second_per_gpu:4168.716915985542 mfu:19.66454894915211 +Step 877 | loss:1.3785547018051147 lr:9.988061034461305e-05 tokens_per_second_per_gpu:5019.973158339216 mfu:23.680069883625936 +Step 878 | loss:1.4620507955551147 lr:9.988031067869115e-05 tokens_per_second_per_gpu:3438.2552417980132 mfu:16.21883660239652 +Step 879 | loss:1.3916348218917847 lr:9.988001063761305e-05 tokens_per_second_per_gpu:4987.870028010004 mfu:23.528633940504204 +Step 880 | loss:1.5580579042434692 lr:9.987971022138102e-05 tokens_per_second_per_gpu:3509.457009405384 mfu:16.554707488474733 +Step 881 | loss:1.545706868171692 lr:9.987940942999734e-05 tokens_per_second_per_gpu:5321.930816332887 mfu:25.104455675671037 +Step 882 | loss:1.3487783670425415 lr:9.987910826346424e-05 tokens_per_second_per_gpu:4398.480390058005 mfu:20.748382458042943 +Step 883 | loss:1.4061713218688965 lr:9.987880672178401e-05 tokens_per_second_per_gpu:4470.990154836845 mfu:21.09042338994631 +Step 884 | loss:1.4823857545852661 lr:9.987850480495891e-05 tokens_per_second_per_gpu:4279.258926076355 mfu:20.18599447116222 +Step 885 | loss:1.5907381772994995 lr:9.987820251299122e-05 tokens_per_second_per_gpu:4138.646235570192 mfu:19.52270041904578 +Step 886 | loss:1.6286160945892334 lr:9.98778998458832e-05 tokens_per_second_per_gpu:3960.2609559468665 mfu:18.681226619395417 +Step 887 | loss:1.3355433940887451 lr:9.987759680363711e-05 tokens_per_second_per_gpu:4575.117580161549 mfu:21.581610221173666 +Step 888 | loss:1.69623601436615 lr:9.987729338625528e-05 tokens_per_second_per_gpu:2934.060144023439 mfu:13.840462301639604 +Step 889 | loss:1.324407696723938 lr:9.987698959373995e-05 tokens_per_second_per_gpu:5353.172015253993 mfu:25.251825741280072 +Step 890 | loss:1.5352753400802612 lr:9.987668542609341e-05 tokens_per_second_per_gpu:3816.6778035326925 mfu:18.003920391645824 +Step 891 | loss:1.48747718334198 lr:9.987638088331797e-05 tokens_per_second_per_gpu:3921.1468676432014 mfu:18.49671879131541 +Step 892 | loss:1.8301739692687988 lr:9.987607596541589e-05 tokens_per_second_per_gpu:5277.729325860771 mfu:24.89594970356213 +Step 893 | loss:1.6182132959365845 lr:9.987577067238949e-05 tokens_per_second_per_gpu:3783.063745137658 mfu:17.845357143046694 +Step 894 | loss:1.4567562341690063 lr:9.987546500424105e-05 tokens_per_second_per_gpu:5319.649200718218 mfu:25.093692905532876 +Step 895 | loss:1.4847301244735718 lr:9.987515896097286e-05 tokens_per_second_per_gpu:5192.222978795702 mfu:24.492601675569034 +Step 896 | loss:1.3204094171524048 lr:9.987485254258727e-05 tokens_per_second_per_gpu:4175.607330639462 mfu:19.69705221070062 +Step 897 | loss:1.4395148754119873 lr:9.98745457490865e-05 tokens_per_second_per_gpu:5150.98855472553 mfu:24.298091861911825 +Step 898 | loss:1.5480892658233643 lr:9.987423858047295e-05 tokens_per_second_per_gpu:5056.914561204924 mfu:23.85432878379674 +Step 899 | loss:1.468989372253418 lr:9.987393103674887e-05 tokens_per_second_per_gpu:4907.764807634997 mfu:23.150763948635365 +Step 900 | loss:1.389864206314087 lr:9.98736231179166e-05 tokens_per_second_per_gpu:5328.8728987432005 mfu:25.13720266284188 +Step 901 | loss:1.4871981143951416 lr:9.987331482397843e-05 tokens_per_second_per_gpu:4064.7966243641367 mfu:19.174339202943866 +Step 902 | loss:1.30354642868042 lr:9.98730061549367e-05 tokens_per_second_per_gpu:4045.722250319518 mfu:19.084362126151603 +Step 903 | loss:1.3280410766601562 lr:9.987269711079374e-05 tokens_per_second_per_gpu:5172.044404493095 mfu:24.397415897763846 +Step 904 | loss:1.3832671642303467 lr:9.987238769155184e-05 tokens_per_second_per_gpu:4028.517817298732 mfu:19.003205880213592 +Step 905 | loss:1.304124355316162 lr:9.987207789721337e-05 tokens_per_second_per_gpu:5326.578954372526 mfu:25.12638173585814 +Step 906 | loss:1.3544343709945679 lr:9.987176772778065e-05 tokens_per_second_per_gpu:4549.319464930934 mfu:21.45991611001852 +Step 907 | loss:1.5617035627365112 lr:9.987145718325597e-05 tokens_per_second_per_gpu:3846.049394260457 mfu:18.142471196418818 +Step 908 | loss:1.380602240562439 lr:9.987114626364171e-05 tokens_per_second_per_gpu:5350.238834397275 mfu:25.237989426726678 +Step 909 | loss:1.5187227725982666 lr:9.987083496894018e-05 tokens_per_second_per_gpu:5153.037701381816 mfu:24.30775803631 +Step 910 | loss:1.5534697771072388 lr:9.987052329915376e-05 tokens_per_second_per_gpu:3696.7217673580953 mfu:17.43806730240003 +Step 911 | loss:1.398012399673462 lr:9.987021125428477e-05 tokens_per_second_per_gpu:4430.838689761355 mfu:20.901022078638874 +Step 912 | loss:1.465094804763794 lr:9.986989883433554e-05 tokens_per_second_per_gpu:4457.867876945584 mfu:21.02852336624175 +Step 913 | loss:1.3296607732772827 lr:9.986958603930845e-05 tokens_per_second_per_gpu:4941.400522365546 mfu:23.30942935386407 +Step 914 | loss:1.408515214920044 lr:9.986927286920583e-05 tokens_per_second_per_gpu:3398.238313682476 mfu:16.030069924883556 +Step 915 | loss:1.4244552850723267 lr:9.986895932403006e-05 tokens_per_second_per_gpu:5069.715381535067 mfu:23.914712437339183 +Step 916 | loss:1.4721837043762207 lr:9.986864540378346e-05 tokens_per_second_per_gpu:4788.275879685865 mfu:22.58711428858597 +Step 917 | loss:1.5183130502700806 lr:9.986833110846844e-05 tokens_per_second_per_gpu:5371.157358368605 mfu:25.33666567336766 +Step 918 | loss:1.4612681865692139 lr:9.986801643808733e-05 tokens_per_second_per_gpu:4557.5658269866135 mfu:21.498815606809604 +Step 919 | loss:1.6983661651611328 lr:9.98677013926425e-05 tokens_per_second_per_gpu:4942.889081017673 mfu:23.316451139000446 +Step 920 | loss:1.3147958517074585 lr:9.986738597213633e-05 tokens_per_second_per_gpu:3921.263327188553 mfu:18.497268150860883 +Step 921 | loss:1.7266000509262085 lr:9.986707017657119e-05 tokens_per_second_per_gpu:4767.060939554893 mfu:22.48703979634614 +Step 922 | loss:1.3617016077041626 lr:9.986675400594944e-05 tokens_per_second_per_gpu:5341.073637610858 mfu:25.194755629741326 +Step 923 | loss:1.5156513452529907 lr:9.986643746027349e-05 tokens_per_second_per_gpu:5340.639484565058 mfu:25.192707655750144 +Step 924 | loss:1.2761822938919067 lr:9.986612053954567e-05 tokens_per_second_per_gpu:5281.521431217151 mfu:24.913837711528068 +Step 925 | loss:1.3509398698806763 lr:9.986580324376842e-05 tokens_per_second_per_gpu:4724.767023980221 mfu:22.287532180494317 +Step 926 | loss:1.4481128454208374 lr:9.98654855729441e-05 tokens_per_second_per_gpu:5330.8527366235685 mfu:25.14654189591875 +Step 927 | loss:1.6254472732543945 lr:9.986516752707509e-05 tokens_per_second_per_gpu:3783.8129082920664 mfu:17.848891073466454 +Step 928 | loss:1.4918781518936157 lr:9.986484910616379e-05 tokens_per_second_per_gpu:3661.3371601348367 mfu:17.271152072891386 +Step 929 | loss:1.5390392541885376 lr:9.986453031021262e-05 tokens_per_second_per_gpu:5356.4439683406135 mfu:25.26726010224242 +Step 930 | loss:1.6005266904830933 lr:9.986421113922393e-05 tokens_per_second_per_gpu:3500.870218297592 mfu:16.514202129761642 +Step 931 | loss:1.4108344316482544 lr:9.986389159320015e-05 tokens_per_second_per_gpu:5385.224259126179 mfu:25.403021644301894 +Step 932 | loss:1.6137163639068604 lr:9.986357167214368e-05 tokens_per_second_per_gpu:5274.639612736221 mfu:24.88137499959421 +Step 933 | loss:1.3354289531707764 lr:9.986325137605692e-05 tokens_per_second_per_gpu:4570.861646921087 mfu:21.561534257941318 +Step 934 | loss:1.3923791646957397 lr:9.98629307049423e-05 tokens_per_second_per_gpu:5168.775023892919 mfu:24.381993671658968 +Step 935 | loss:1.7304073572158813 lr:9.98626096588022e-05 tokens_per_second_per_gpu:5313.087265850788 mfu:25.062739139181673 +Step 936 | loss:1.509167194366455 lr:9.986228823763905e-05 tokens_per_second_per_gpu:3902.5934573302393 mfu:18.40919918933095 +Step 937 | loss:1.4237734079360962 lr:9.986196644145529e-05 tokens_per_second_per_gpu:5303.563400681354 mfu:25.0178134422385 +Step 938 | loss:1.4748425483703613 lr:9.98616442702533e-05 tokens_per_second_per_gpu:5357.770251321906 mfu:25.27351640535239 +Step 939 | loss:1.4011013507843018 lr:9.986132172403553e-05 tokens_per_second_per_gpu:5355.385237609 mfu:25.26226588874331 +Step 940 | loss:1.3379535675048828 lr:9.986099880280438e-05 tokens_per_second_per_gpu:4857.446117665087 mfu:22.913402102793505 +Step 941 | loss:1.5146321058273315 lr:9.98606755065623e-05 tokens_per_second_per_gpu:4684.620994147449 mfu:22.09815650815413 +Step 942 | loss:1.3641244173049927 lr:9.986035183531172e-05 tokens_per_second_per_gpu:5218.082840839067 mfu:24.61458705697502 +Step 943 | loss:1.470343828201294 lr:9.986002778905508e-05 tokens_per_second_per_gpu:5299.975620840485 mfu:25.000889272590758 +Step 944 | loss:1.3829617500305176 lr:9.98597033677948e-05 tokens_per_second_per_gpu:5002.886376811981 mfu:23.599468619856424 +Step 945 | loss:1.3374351263046265 lr:9.985937857153334e-05 tokens_per_second_per_gpu:4102.569023954265 mfu:19.352518056446417 +Step 946 | loss:1.5833269357681274 lr:9.985905340027313e-05 tokens_per_second_per_gpu:5355.795702019373 mfu:25.264202119399492 +Step 947 | loss:1.3929150104522705 lr:9.985872785401659e-05 tokens_per_second_per_gpu:4629.929308459375 mfu:21.84016649539969 +Step 948 | loss:1.6643149852752686 lr:9.985840193276624e-05 tokens_per_second_per_gpu:4162.2883712055145 mfu:19.63422440660188 +Step 949 | loss:1.617537021636963 lr:9.985807563652445e-05 tokens_per_second_per_gpu:4864.941191624607 mfu:22.948757645452044 +Step 950 | loss:1.5630460977554321 lr:9.985774896529373e-05 tokens_per_second_per_gpu:5339.963596317214 mfu:25.189519375566633 +Step 951 | loss:1.5583497285842896 lr:9.985742191907652e-05 tokens_per_second_per_gpu:4648.7836066177 mfu:21.929105436687486 +Step 952 | loss:1.218157410621643 lr:9.985709449787525e-05 tokens_per_second_per_gpu:5337.8043292296325 mfu:25.179333744305556 +Step 953 | loss:1.4446593523025513 lr:9.985676670169244e-05 tokens_per_second_per_gpu:4558.9410141906255 mfu:21.505302599482064 +Step 954 | loss:1.41490638256073 lr:9.985643853053053e-05 tokens_per_second_per_gpu:4995.058890213433 mfu:23.562545030064147 +Step 955 | loss:1.5447646379470825 lr:9.985610998439197e-05 tokens_per_second_per_gpu:4940.406278978515 mfu:23.304739338172027 +Step 956 | loss:1.3762744665145874 lr:9.985578106327924e-05 tokens_per_second_per_gpu:4813.411471808115 mfu:22.70568316520086 +Step 957 | loss:1.4738197326660156 lr:9.985545176719483e-05 tokens_per_second_per_gpu:5057.228300148709 mfu:23.85580874392396 +Step 958 | loss:1.7192338705062866 lr:9.98551220961412e-05 tokens_per_second_per_gpu:3579.048032155351 mfu:16.88298021623916 +Step 959 | loss:1.4808845520019531 lr:9.985479205012084e-05 tokens_per_second_per_gpu:4767.167044476525 mfu:22.487540311364846 +Step 960 | loss:1.5164483785629272 lr:9.985446162913623e-05 tokens_per_second_per_gpu:5318.086001625897 mfu:25.086319028705134 +Step 961 | loss:1.480167031288147 lr:9.985413083318985e-05 tokens_per_second_per_gpu:4207.340545940939 mfu:19.84674320152251 +Step 962 | loss:1.3936469554901123 lr:9.985379966228418e-05 tokens_per_second_per_gpu:5370.77097943409 mfu:25.334843058010506 +Step 963 | loss:1.5308254957199097 lr:9.985346811642172e-05 tokens_per_second_per_gpu:4713.794422293868 mfu:22.235772546220893 +Step 964 | loss:1.5349514484405518 lr:9.985313619560499e-05 tokens_per_second_per_gpu:5245.484731517415 mfu:24.743846450549427 +Step 965 | loss:1.463502049446106 lr:9.985280389983643e-05 tokens_per_second_per_gpu:5329.39020200404 mfu:25.139642870584296 +Step 966 | loss:1.5273621082305908 lr:9.985247122911858e-05 tokens_per_second_per_gpu:4594.286469706631 mfu:21.67203314370762 +Step 967 | loss:1.3431161642074585 lr:9.985213818345394e-05 tokens_per_second_per_gpu:4808.153530900582 mfu:22.680880560842805 +Step 968 | loss:1.2698962688446045 lr:9.985180476284502e-05 tokens_per_second_per_gpu:5353.0286080435535 mfu:25.251149265000638 +Step 969 | loss:1.4283676147460938 lr:9.985147096729429e-05 tokens_per_second_per_gpu:4389.656376558487 mfu:20.706758080833772 +Step 970 | loss:1.3542100191116333 lr:9.98511367968043e-05 tokens_per_second_per_gpu:4869.0455056451065 mfu:22.968118394954924 +Step 971 | loss:1.270270824432373 lr:9.985080225137755e-05 tokens_per_second_per_gpu:4626.429082513318 mfu:21.82365533241201 +Step 972 | loss:1.3426204919815063 lr:9.985046733101655e-05 tokens_per_second_per_gpu:3617.0512991619926 mfu:17.062248110735453 +Step 973 | loss:1.6324387788772583 lr:9.985013203572383e-05 tokens_per_second_per_gpu:4723.591559502926 mfu:22.281987314 +Step 974 | loss:1.3624080419540405 lr:9.984979636550189e-05 tokens_per_second_per_gpu:5327.2001819303205 mfu:25.12931217224061 +Step 975 | loss:1.3228814601898193 lr:9.98494603203533e-05 tokens_per_second_per_gpu:4354.540679717233 mfu:20.541111347477884 +Step 976 | loss:1.4300600290298462 lr:9.984912390028052e-05 tokens_per_second_per_gpu:5353.017243627616 mfu:25.251095657111694 +Step 977 | loss:1.4646259546279907 lr:9.984878710528615e-05 tokens_per_second_per_gpu:4295.274366415506 mfu:20.261542035757884 +Step 978 | loss:1.407858967781067 lr:9.984844993537267e-05 tokens_per_second_per_gpu:4458.883636549879 mfu:21.033314877601676 +Step 979 | loss:1.608216404914856 lr:9.984811239054266e-05 tokens_per_second_per_gpu:5000.634276087944 mfu:23.588845076493136 +Step 980 | loss:1.2953217029571533 lr:9.984777447079862e-05 tokens_per_second_per_gpu:3765.976407412799 mfu:17.76475325559809 +Step 981 | loss:1.5546107292175293 lr:9.98474361761431e-05 tokens_per_second_per_gpu:3876.033933464108 mfu:18.28391338373236 +Step 982 | loss:1.525080680847168 lr:9.984709750657867e-05 tokens_per_second_per_gpu:4962.726646663162 mfu:23.410028320787927 +Step 983 | loss:1.5413309335708618 lr:9.984675846210785e-05 tokens_per_second_per_gpu:5297.889987046709 mfu:24.991050982139697 +Step 984 | loss:1.3179513216018677 lr:9.984641904273319e-05 tokens_per_second_per_gpu:4808.931337738521 mfu:22.684549608404893 +Step 985 | loss:1.4853001832962036 lr:9.984607924845728e-05 tokens_per_second_per_gpu:5234.24963392321 mfu:24.69084857829225 +Step 986 | loss:1.4760452508926392 lr:9.984573907928263e-05 tokens_per_second_per_gpu:4853.969333336577 mfu:22.89700151791535 +Step 987 | loss:1.3997281789779663 lr:9.984539853521182e-05 tokens_per_second_per_gpu:4792.694150509792 mfu:22.60795602589717 +Step 988 | loss:1.4427624940872192 lr:9.98450576162474e-05 tokens_per_second_per_gpu:4559.995724408451 mfu:21.510277847531665 +Step 989 | loss:1.3814692497253418 lr:9.984471632239193e-05 tokens_per_second_per_gpu:4972.208733954552 mfu:23.454756944352336 +Step 990 | loss:1.4909080266952515 lr:9.984437465364802e-05 tokens_per_second_per_gpu:3815.8637938028282 mfu:18.000080568865673 +Step 991 | loss:1.3674546480178833 lr:9.984403261001819e-05 tokens_per_second_per_gpu:5369.265258755843 mfu:25.32774031666878 +Step 992 | loss:1.5935183763504028 lr:9.984369019150504e-05 tokens_per_second_per_gpu:3409.9892446158424 mfu:16.085501071011805 +Step 993 | loss:1.4368481636047363 lr:9.984334739811112e-05 tokens_per_second_per_gpu:5332.829767115108 mfu:25.155867886063525 +Step 994 | loss:1.509032964706421 lr:9.984300422983903e-05 tokens_per_second_per_gpu:4231.730683270182 mfu:19.961795640691093 +Step 995 | loss:1.4823269844055176 lr:9.984266068669135e-05 tokens_per_second_per_gpu:5363.454594209271 mfu:25.300330420600996 +Step 996 | loss:1.2938828468322754 lr:9.984231676867066e-05 tokens_per_second_per_gpu:4057.151456489107 mfu:19.138275641677204 +Step 997 | loss:1.4171808958053589 lr:9.984197247577953e-05 tokens_per_second_per_gpu:5285.390034983772 mfu:24.932086575547082 +Step 998 | loss:1.4045178890228271 lr:9.984162780802059e-05 tokens_per_second_per_gpu:5326.790597488246 mfu:25.12738009254503 +Step 999 | loss:1.38722825050354 lr:9.984128276539639e-05 tokens_per_second_per_gpu:3447.4502204830856 mfu:16.262210885679146 +Step 1000 | loss:1.276262879371643 lr:9.984093734790956e-05 tokens_per_second_per_gpu:3959.3315916038273 mfu:18.676842649223396 +Step 1001 | loss:1.4867864847183228 lr:9.984059155556265e-05 tokens_per_second_per_gpu:4780.050569705343 mfu:22.548314098025255 +Step 1002 | loss:1.4442917108535767 lr:9.984024538835832e-05 tokens_per_second_per_gpu:3827.642704817069 mfu:18.0556437018093 +Step 1003 | loss:1.3762685060501099 lr:9.983989884629913e-05 tokens_per_second_per_gpu:4665.168505686327 mfu:22.00639580114625 +Step 1004 | loss:1.5158172845840454 lr:9.983955192938771e-05 tokens_per_second_per_gpu:5333.204733605328 mfu:25.15763666697378 +Step 1005 | loss:1.4802569150924683 lr:9.983920463762666e-05 tokens_per_second_per_gpu:4819.236387893433 mfu:22.733160288208474 +Step 1006 | loss:1.4483882188796997 lr:9.983885697101858e-05 tokens_per_second_per_gpu:4110.9420977360205 mfu:19.392015274068658 +Step 1007 | loss:1.4587736129760742 lr:9.98385089295661e-05 tokens_per_second_per_gpu:4567.911713487663 mfu:21.547618918625314 +Step 1008 | loss:1.5668548345565796 lr:9.983816051327184e-05 tokens_per_second_per_gpu:5030.874383631259 mfu:23.731492823269328 +Step 1009 | loss:1.4453752040863037 lr:9.983781172213843e-05 tokens_per_second_per_gpu:4183.023800672244 mfu:19.73203696522552 +Step 1010 | loss:1.3352482318878174 lr:9.983746255616846e-05 tokens_per_second_per_gpu:5329.984473056555 mfu:25.142446148532105 +Step 1011 | loss:1.4278268814086914 lr:9.983711301536458e-05 tokens_per_second_per_gpu:5334.948141547372 mfu:25.165860619692467 +Step 1012 | loss:1.4230936765670776 lr:9.98367630997294e-05 tokens_per_second_per_gpu:4937.2380073343365 mfu:23.289794060263556 +Step 1013 | loss:1.2303587198257446 lr:9.98364128092656e-05 tokens_per_second_per_gpu:4776.433153650286 mfu:22.531250129298726 +Step 1014 | loss:1.4748873710632324 lr:9.983606214397574e-05 tokens_per_second_per_gpu:3931.4090374935645 mfu:18.5451271974063 +Step 1015 | loss:1.5260229110717773 lr:9.983571110386252e-05 tokens_per_second_per_gpu:5351.773455113244 mfu:25.245228494477196 +Step 1016 | loss:1.6366533041000366 lr:9.983535968892855e-05 tokens_per_second_per_gpu:5161.233695548984 mfu:24.34641993917733 +Step 1017 | loss:1.4908559322357178 lr:9.983500789917648e-05 tokens_per_second_per_gpu:3835.211582040712 mfu:18.091347387056842 +Step 1018 | loss:1.5623724460601807 lr:9.983465573460897e-05 tokens_per_second_per_gpu:3828.6429462653105 mfu:18.060362011379443 +Step 1019 | loss:1.45371413230896 lr:9.983430319522864e-05 tokens_per_second_per_gpu:3922.767384927597 mfu:18.504363047834666 +Step 1020 | loss:1.4349507093429565 lr:9.983395028103816e-05 tokens_per_second_per_gpu:5156.922525977921 mfu:24.326083416749253 +Step 1021 | loss:1.4325737953186035 lr:9.983359699204018e-05 tokens_per_second_per_gpu:5205.159651674296 mfu:24.553626168761355 +Step 1022 | loss:1.411439061164856 lr:9.983324332823738e-05 tokens_per_second_per_gpu:3852.469409584153 mfu:18.172755504068256 +Step 1023 | loss:1.4893343448638916 lr:9.983288928963237e-05 tokens_per_second_per_gpu:4264.912543544681 mfu:20.118320137015054 +Step 1024 | loss:1.4544495344161987 lr:9.983253487622785e-05 tokens_per_second_per_gpu:5330.345177351816 mfu:25.144147652235358 +Step 1025 | loss:1.4252848625183105 lr:9.983218008802648e-05 tokens_per_second_per_gpu:3727.6109357069695 mfu:17.58377677973724 +Step 1026 | loss:1.4282504320144653 lr:9.983182492503093e-05 tokens_per_second_per_gpu:5379.0156886601135 mfu:25.373734758122108 +Step 1027 | loss:1.6422864198684692 lr:9.983146938724386e-05 tokens_per_second_per_gpu:4268.361309462184 mfu:20.134588554268056 +Step 1028 | loss:1.6780788898468018 lr:9.983111347466794e-05 tokens_per_second_per_gpu:3946.056431719412 mfu:18.614221455072776 +Step 1029 | loss:1.3107068538665771 lr:9.983075718730587e-05 tokens_per_second_per_gpu:5325.824815662481 mfu:25.12282433489353 +Step 1030 | loss:1.4667096138000488 lr:9.98304005251603e-05 tokens_per_second_per_gpu:5317.129517155548 mfu:25.081807128265414 +Step 1031 | loss:1.4628809690475464 lr:9.983004348823395e-05 tokens_per_second_per_gpu:3100.2828498730455 mfu:14.624563165650258 +Step 1032 | loss:1.4380550384521484 lr:9.982968607652949e-05 tokens_per_second_per_gpu:4437.718034928674 mfu:20.93347312353043 +Step 1033 | loss:1.2988851070404053 lr:9.982932829004959e-05 tokens_per_second_per_gpu:3027.460723808848 mfu:14.281048772269536 +Step 1034 | loss:1.4779181480407715 lr:9.982897012879695e-05 tokens_per_second_per_gpu:5272.366534965455 mfu:24.870652503922447 +Step 1035 | loss:1.4288066625595093 lr:9.982861159277426e-05 tokens_per_second_per_gpu:4946.682360872354 mfu:23.33434468727526 +Step 1036 | loss:1.5180423259735107 lr:9.982825268198423e-05 tokens_per_second_per_gpu:4219.769479631574 mfu:19.90537260233599 +Step 1037 | loss:1.464133858680725 lr:9.982789339642957e-05 tokens_per_second_per_gpu:5359.521568049106 mfu:25.281777665160035 +Step 1038 | loss:1.3051729202270508 lr:9.982753373611296e-05 tokens_per_second_per_gpu:5082.875609115346 mfu:23.976791476201722 +Step 1039 | loss:1.5224756002426147 lr:9.982717370103709e-05 tokens_per_second_per_gpu:5313.958226824705 mfu:25.066847610696907 +Step 1040 | loss:1.4162362813949585 lr:9.982681329120472e-05 tokens_per_second_per_gpu:3735.793178683866 mfu:17.622373815893862 +Step 1041 | loss:1.583907961845398 lr:9.982645250661851e-05 tokens_per_second_per_gpu:3933.9913469557964 mfu:18.557308391727073 +Step 1042 | loss:1.4124540090560913 lr:9.982609134728118e-05 tokens_per_second_per_gpu:5037.028531356135 mfu:23.76052298809345 +Step 1043 | loss:1.387663722038269 lr:9.982572981319547e-05 tokens_per_second_per_gpu:5014.592753346127 mfu:23.654689595281784 +Step 1044 | loss:1.391074776649475 lr:9.982536790436409e-05 tokens_per_second_per_gpu:4084.36703677676 mfu:19.266656177350878 +Step 1045 | loss:1.6101336479187012 lr:9.982500562078978e-05 tokens_per_second_per_gpu:5347.755231714383 mfu:25.22627384912516 +Step 1046 | loss:1.4827840328216553 lr:9.982464296247522e-05 tokens_per_second_per_gpu:5305.936321268754 mfu:25.029006913511264 +Step 1047 | loss:1.4462348222732544 lr:9.982427992942317e-05 tokens_per_second_per_gpu:4838.818010481912 mfu:22.82553014292765 +Step 1048 | loss:1.7224611043930054 lr:9.982391652163636e-05 tokens_per_second_per_gpu:5316.453994169603 mfu:25.07862057108461 +Step 1049 | loss:1.3302024602890015 lr:9.98235527391175e-05 tokens_per_second_per_gpu:4541.601625001494 mfu:21.423509742272067 +Step 1050 | loss:1.5767016410827637 lr:9.982318858186937e-05 tokens_per_second_per_gpu:4459.554083110498 mfu:21.03647748841809 +Step 1051 | loss:1.4268211126327515 lr:9.982282404989465e-05 tokens_per_second_per_gpu:4898.208890820127 mfu:23.105687058612087 +Step 1052 | loss:1.3815715312957764 lr:9.982245914319613e-05 tokens_per_second_per_gpu:5364.156984215215 mfu:25.303643714099127 +Step 1053 | loss:1.3835922479629517 lr:9.982209386177653e-05 tokens_per_second_per_gpu:4838.984859291288 mfu:22.826317197228818 +Step 1054 | loss:1.4169394969940186 lr:9.98217282056386e-05 tokens_per_second_per_gpu:5280.777575752731 mfu:24.910328818387566 +Step 1055 | loss:1.4063780307769775 lr:9.982136217478511e-05 tokens_per_second_per_gpu:4743.298510275761 mfu:22.374948363147976 +Step 1056 | loss:1.426688313484192 lr:9.98209957692188e-05 tokens_per_second_per_gpu:5238.234849425955 mfu:24.709647519766577 +Step 1057 | loss:1.3828580379486084 lr:9.98206289889424e-05 tokens_per_second_per_gpu:5310.659677939006 mfu:25.05128779281259 +Step 1058 | loss:1.3611527681350708 lr:9.982026183395872e-05 tokens_per_second_per_gpu:3903.431147966427 mfu:18.413150718986607 +Step 1059 | loss:1.510347843170166 lr:9.981989430427048e-05 tokens_per_second_per_gpu:4384.8306294205495 mfu:20.683994208226753 +Step 1060 | loss:1.7054311037063599 lr:9.981952639988046e-05 tokens_per_second_per_gpu:4226.095570387063 mfu:19.93521385177202 +Step 1061 | loss:1.5311654806137085 lr:9.981915812079142e-05 tokens_per_second_per_gpu:5301.280258424916 mfu:25.00704347067064 +Step 1062 | loss:1.4115033149719238 lr:9.981878946700613e-05 tokens_per_second_per_gpu:4435.890146856861 mfu:20.92485066362583 +Step 1063 | loss:1.4551002979278564 lr:9.981842043852738e-05 tokens_per_second_per_gpu:4048.8918538118564 mfu:19.099313686615083 +Step 1064 | loss:1.2952967882156372 lr:9.981805103535793e-05 tokens_per_second_per_gpu:5313.591084655883 mfu:25.06511573844566 +Step 1065 | loss:1.6148813962936401 lr:9.981768125750055e-05 tokens_per_second_per_gpu:3723.186708816286 mfu:17.562906946642936 +Step 1066 | loss:1.3819721937179565 lr:9.981731110495805e-05 tokens_per_second_per_gpu:3486.134910391757 mfu:16.444693168267044 +Step 1067 | loss:1.3023625612258911 lr:9.98169405777332e-05 tokens_per_second_per_gpu:5314.924614818606 mfu:25.071406227746934 +Step 1068 | loss:1.4563257694244385 lr:9.981656967582875e-05 tokens_per_second_per_gpu:3978.09460584549 mfu:18.76535098870166 +Step 1069 | loss:1.5185511112213135 lr:9.981619839924756e-05 tokens_per_second_per_gpu:4118.127199130042 mfu:19.425908623248965 +Step 1070 | loss:1.2993115186691284 lr:9.981582674799236e-05 tokens_per_second_per_gpu:4498.539437389343 mfu:21.220378056138735 +Step 1071 | loss:1.672562599182129 lr:9.9815454722066e-05 tokens_per_second_per_gpu:5286.52038151542 mfu:24.937418612993874 +Step 1072 | loss:1.595900297164917 lr:9.98150823214712e-05 tokens_per_second_per_gpu:3878.0734780712373 mfu:18.29353426362638 +Step 1073 | loss:1.3361194133758545 lr:9.981470954621087e-05 tokens_per_second_per_gpu:4734.5209289603745 mfu:22.333543014473495 +Step 1074 | loss:1.4240920543670654 lr:9.98143363962877e-05 tokens_per_second_per_gpu:3945.796838004425 mfu:18.612996907227927 +Step 1075 | loss:1.4161969423294067 lr:9.981396287170457e-05 tokens_per_second_per_gpu:4168.097386885293 mfu:19.661626524683673 +Step 1076 | loss:1.472750186920166 lr:9.981358897246428e-05 tokens_per_second_per_gpu:4243.1260329406105 mfu:20.015549449333637 +Step 1077 | loss:1.5319907665252686 lr:9.98132146985696e-05 tokens_per_second_per_gpu:4080.7756052677705 mfu:19.24971478216158 +Step 1078 | loss:1.43359375 lr:9.981284005002341e-05 tokens_per_second_per_gpu:3499.345503251928 mfu:16.507009788748036 +Step 1079 | loss:1.4015322923660278 lr:9.981246502682849e-05 tokens_per_second_per_gpu:4319.182275627285 mfu:20.37431972261978 +Step 1080 | loss:1.4383320808410645 lr:9.981208962898765e-05 tokens_per_second_per_gpu:5374.943981204575 mfu:25.354527819349403 +Step 1081 | loss:1.4238359928131104 lr:9.981171385650375e-05 tokens_per_second_per_gpu:5107.445680509353 mfu:24.092692695053707 +Step 1082 | loss:1.3046793937683105 lr:9.981133770937956e-05 tokens_per_second_per_gpu:4047.9249214040246 mfu:19.09475250147166 +Step 1083 | loss:1.3631263971328735 lr:9.981096118761796e-05 tokens_per_second_per_gpu:4682.788702523412 mfu:22.089513275942362 +Step 1084 | loss:1.4870257377624512 lr:9.981058429122176e-05 tokens_per_second_per_gpu:4403.522150534602 mfu:20.772165302424092 +Step 1085 | loss:1.2845544815063477 lr:9.981020702019382e-05 tokens_per_second_per_gpu:4139.940049463941 mfu:19.52880355993 +Step 1086 | loss:1.5487173795700073 lr:9.980982937453694e-05 tokens_per_second_per_gpu:3673.3871028954695 mfu:17.327993708825524 +Step 1087 | loss:1.350592851638794 lr:9.980945135425397e-05 tokens_per_second_per_gpu:4700.594346920898 mfu:22.1735055385215 +Step 1088 | loss:1.3458728790283203 lr:9.980907295934778e-05 tokens_per_second_per_gpu:4882.439218238451 mfu:23.03129882246074 +Step 1089 | loss:1.3778815269470215 lr:9.980869418982119e-05 tokens_per_second_per_gpu:4805.582165259653 mfu:22.668750990394326 +Step 1090 | loss:1.4711016416549683 lr:9.980831504567705e-05 tokens_per_second_per_gpu:5313.074265194839 mfu:25.062677812869424 +Step 1091 | loss:1.3585855960845947 lr:9.980793552691822e-05 tokens_per_second_per_gpu:5325.064968203407 mfu:25.11924000478573 +Step 1092 | loss:1.330926537513733 lr:9.980755563354756e-05 tokens_per_second_per_gpu:4600.4959280387275 mfu:21.70132421810296 +Step 1093 | loss:1.3694413900375366 lr:9.98071753655679e-05 tokens_per_second_per_gpu:3796.590930350096 mfu:17.909167183669553 +Step 1094 | loss:1.28037428855896 lr:9.980679472298214e-05 tokens_per_second_per_gpu:4133.901156572266 mfu:19.500317071818596 +Step 1095 | loss:1.4646732807159424 lr:9.980641370579309e-05 tokens_per_second_per_gpu:4283.858334309265 mfu:20.20769066453682 +Step 1096 | loss:1.4124243259429932 lr:9.980603231400367e-05 tokens_per_second_per_gpu:3888.517819465017 mfu:18.342802003969236 +Step 1097 | loss:1.3273746967315674 lr:9.980565054761672e-05 tokens_per_second_per_gpu:4362.27635398605 mfu:20.577601842846022 +Step 1098 | loss:1.6862610578536987 lr:9.980526840663512e-05 tokens_per_second_per_gpu:3638.1464452181813 mfu:17.16175751388555 +Step 1099 | loss:1.440786361694336 lr:9.980488589106174e-05 tokens_per_second_per_gpu:4950.668004575005 mfu:23.3531456486416 +Step 1100 | loss:1.2603815793991089 lr:9.980450300089946e-05 tokens_per_second_per_gpu:5336.1992251411075 mfu:25.171762194460985 +Step 1101 | loss:1.3580621480941772 lr:9.980411973615116e-05 tokens_per_second_per_gpu:5301.669954826829 mfu:25.008881735087527 +Step 1102 | loss:1.487776756286621 lr:9.980373609681972e-05 tokens_per_second_per_gpu:3114.756621786632 mfu:14.692838417246781 +Step 1103 | loss:1.6736348867416382 lr:9.980335208290801e-05 tokens_per_second_per_gpu:5007.509817886602 mfu:23.6212781802458 +Step 1104 | loss:1.4002314805984497 lr:9.980296769441894e-05 tokens_per_second_per_gpu:4731.133791361048 mfu:22.317565308512208 +Step 1105 | loss:1.4612321853637695 lr:9.980258293135541e-05 tokens_per_second_per_gpu:4987.888299258484 mfu:23.52872012910068 +Step 1106 | loss:1.4919090270996094 lr:9.98021977937203e-05 tokens_per_second_per_gpu:4406.522814375549 mfu:20.786319945727886 +Step 1107 | loss:1.3778587579727173 lr:9.980181228151649e-05 tokens_per_second_per_gpu:4987.158873010126 mfu:23.52527930103422 +Step 1108 | loss:1.2858701944351196 lr:9.98014263947469e-05 tokens_per_second_per_gpu:5082.135265423597 mfu:23.97329914869208 +Step 1109 | loss:1.4694944620132446 lr:9.980104013341443e-05 tokens_per_second_per_gpu:5320.868022897726 mfu:25.099442297706126 +Step 1110 | loss:1.504562497138977 lr:9.980065349752199e-05 tokens_per_second_per_gpu:5303.885708048124 mfu:25.01933382107886 +Step 1111 | loss:1.578790545463562 lr:9.980026648707247e-05 tokens_per_second_per_gpu:3648.4664355865766 mfu:17.210438669225194 +Step 1112 | loss:1.3734166622161865 lr:9.979987910206878e-05 tokens_per_second_per_gpu:5326.852572849397 mfu:25.127672440897665 +Step 1113 | loss:1.5388981103897095 lr:9.979949134251387e-05 tokens_per_second_per_gpu:4121.745802873437 mfu:19.44297819450403 +Step 1114 | loss:1.4580057859420776 lr:9.979910320841062e-05 tokens_per_second_per_gpu:4223.90624409955 mfu:19.92488642613624 +Step 1115 | loss:1.7212127447128296 lr:9.979871469976196e-05 tokens_per_second_per_gpu:4240.330330823768 mfu:20.002361645452527 +Step 1116 | loss:1.337880253791809 lr:9.979832581657081e-05 tokens_per_second_per_gpu:3226.110372862402 mfu:15.218113059980226 +Step 1117 | loss:1.625375747680664 lr:9.97979365588401e-05 tokens_per_second_per_gpu:3864.922512521532 mfu:18.23149891534233 +Step 1118 | loss:1.5873762369155884 lr:9.979754692657275e-05 tokens_per_second_per_gpu:4060.0223143888343 mfu:19.151817968209084 +Step 1119 | loss:1.4782333374023438 lr:9.97971569197717e-05 tokens_per_second_per_gpu:4465.229664674845 mfu:21.063250175011184 +Step 1120 | loss:1.3832045793533325 lr:9.979676653843988e-05 tokens_per_second_per_gpu:3972.935878703737 mfu:18.741016392604397 +Step 1121 | loss:1.461864948272705 lr:9.979637578258022e-05 tokens_per_second_per_gpu:4717.116091064507 mfu:22.25144142454713 +Step 1122 | loss:1.2889890670776367 lr:9.979598465219566e-05 tokens_per_second_per_gpu:5277.7581611755 mfu:24.89608572466152 +Step 1123 | loss:1.511808156967163 lr:9.979559314728915e-05 tokens_per_second_per_gpu:4962.36881086994 mfu:23.40834034829822 +Step 1124 | loss:1.439618706703186 lr:9.979520126786364e-05 tokens_per_second_per_gpu:4848.309903851167 mfu:22.870305023434433 +Step 1125 | loss:1.3804374933242798 lr:9.979480901392206e-05 tokens_per_second_per_gpu:5096.382667851116 mfu:24.040506576800205 +Step 1126 | loss:1.432431697845459 lr:9.979441638546737e-05 tokens_per_second_per_gpu:5314.248367668408 mfu:25.068216254559786 +Step 1127 | loss:1.4941606521606445 lr:9.97940233825025e-05 tokens_per_second_per_gpu:4153.054609650535 mfu:19.590667177904866 +Step 1128 | loss:1.5428173542022705 lr:9.979363000503046e-05 tokens_per_second_per_gpu:4705.80724192927 mfu:22.198095653687595 +Step 1129 | loss:1.5800132751464844 lr:9.979323625305415e-05 tokens_per_second_per_gpu:4434.354816691349 mfu:20.917608249280768 +Step 1130 | loss:1.5990309715270996 lr:9.979284212657657e-05 tokens_per_second_per_gpu:4941.495718961599 mfu:23.309878412449866 +Step 1131 | loss:1.3019343614578247 lr:9.979244762560067e-05 tokens_per_second_per_gpu:4032.7316138134856 mfu:19.02308307732151 +Step 1132 | loss:1.5810081958770752 lr:9.97920527501294e-05 tokens_per_second_per_gpu:5205.452267198255 mfu:24.555006486113225 +Step 1133 | loss:1.5944690704345703 lr:9.979165750016577e-05 tokens_per_second_per_gpu:4403.149789731295 mfu:20.770408812983586 +Step 1134 | loss:1.5684916973114014 lr:9.979126187571273e-05 tokens_per_second_per_gpu:5354.878256158188 mfu:25.25987437073988 +Step 1135 | loss:1.3684508800506592 lr:9.979086587677324e-05 tokens_per_second_per_gpu:4156.776543072847 mfu:19.608224172885233 +Step 1136 | loss:1.410111665725708 lr:9.979046950335029e-05 tokens_per_second_per_gpu:3812.8035827788535 mfu:17.985645031339743 +Step 1137 | loss:1.419413447380066 lr:9.97900727554469e-05 tokens_per_second_per_gpu:4006.909374752849 mfu:18.9012751699442 +Step 1138 | loss:1.3160353899002075 lr:9.978967563306598e-05 tokens_per_second_per_gpu:5331.620792057184 mfu:25.15016494444289 +Step 1139 | loss:1.5235339403152466 lr:9.978927813621057e-05 tokens_per_second_per_gpu:4838.674976175609 mfu:22.82485542570065 +Step 1140 | loss:1.3950504064559937 lr:9.978888026488365e-05 tokens_per_second_per_gpu:4777.939416214021 mfu:22.538355426806753 +Step 1141 | loss:1.4474047422409058 lr:9.978848201908822e-05 tokens_per_second_per_gpu:5361.5780101736045 mfu:25.291478253526503 +Step 1142 | loss:1.22795569896698 lr:9.978808339882725e-05 tokens_per_second_per_gpu:4564.448031881429 mfu:21.531280141523688 +Step 1143 | loss:1.3656930923461914 lr:9.978768440410376e-05 tokens_per_second_per_gpu:5357.266743427135 mfu:25.271141272705453 +Step 1144 | loss:1.6402753591537476 lr:9.978728503492071e-05 tokens_per_second_per_gpu:3761.4219869384915 mfu:17.74326927715652 +Step 1145 | loss:1.3202745914459229 lr:9.978688529128117e-05 tokens_per_second_per_gpu:4298.094258092649 mfu:20.274843945922104 +Step 1146 | loss:1.4906294345855713 lr:9.978648517318812e-05 tokens_per_second_per_gpu:4104.432354017347 mfu:19.36130770227087 +Step 1147 | loss:1.3471193313598633 lr:9.978608468064455e-05 tokens_per_second_per_gpu:5353.723282594069 mfu:25.25442616337912 +Step 1148 | loss:1.4588619470596313 lr:9.978568381365347e-05 tokens_per_second_per_gpu:5342.917199776544 mfu:25.203452027021008 +Step 1149 | loss:1.3511314392089844 lr:9.978528257221792e-05 tokens_per_second_per_gpu:5056.699907838148 mfu:23.853316227242015 +Step 1150 | loss:1.3940309286117554 lr:9.978488095634091e-05 tokens_per_second_per_gpu:4246.922405546901 mfu:20.033457586645472 +Step 1151 | loss:1.3881741762161255 lr:9.978447896602546e-05 tokens_per_second_per_gpu:3839.4080907101484 mfu:18.111143034448784 +Step 1152 | loss:1.5540354251861572 lr:9.978407660127459e-05 tokens_per_second_per_gpu:5349.645037568327 mfu:25.235188385772343 +Step 1153 | loss:1.556215763092041 lr:9.978367386209132e-05 tokens_per_second_per_gpu:5122.5170460884365 mfu:24.163786897930372 +Step 1154 | loss:1.4428001642227173 lr:9.978327074847868e-05 tokens_per_second_per_gpu:4733.859896515218 mfu:22.33042481164593 +Step 1155 | loss:1.4664123058319092 lr:9.978286726043973e-05 tokens_per_second_per_gpu:4224.717595372229 mfu:19.928713708520267 +Step 1156 | loss:1.35652494430542 lr:9.978246339797748e-05 tokens_per_second_per_gpu:5335.571006784408 mfu:25.168798781287332 +Step 1157 | loss:1.3368319272994995 lr:9.978205916109496e-05 tokens_per_second_per_gpu:5333.2378798588215 mfu:25.15779302350684 +Step 1158 | loss:1.4855449199676514 lr:9.978165454979522e-05 tokens_per_second_per_gpu:4215.620396467373 mfu:19.885800669143876 +Step 1159 | loss:1.4186558723449707 lr:9.97812495640813e-05 tokens_per_second_per_gpu:3893.523357406694 mfu:18.366413980472963 +Step 1160 | loss:1.423802375793457 lr:9.978084420395628e-05 tokens_per_second_per_gpu:5355.915924434596 mfu:25.26476922904416 +Step 1161 | loss:1.4084200859069824 lr:9.978043846942315e-05 tokens_per_second_per_gpu:5092.273246962141 mfu:24.021121737327405 +Step 1162 | loss:1.5356254577636719 lr:9.978003236048499e-05 tokens_per_second_per_gpu:5158.057918756144 mfu:24.33143925820211 +Step 1163 | loss:1.4763593673706055 lr:9.977962587714486e-05 tokens_per_second_per_gpu:5072.367026415117 mfu:23.92722069865565 +Step 1164 | loss:1.3856662511825562 lr:9.977921901940581e-05 tokens_per_second_per_gpu:5361.3726685576785 mfu:25.29050962208929 +Step 1165 | loss:1.3986729383468628 lr:9.977881178727091e-05 tokens_per_second_per_gpu:3961.734992204534 mfu:18.68817989991936 +Step 1166 | loss:1.2632898092269897 lr:9.97784041807432e-05 tokens_per_second_per_gpu:4232.640799896927 mfu:19.966088816101355 +Step 1167 | loss:1.3140949010849 lr:9.977799619982577e-05 tokens_per_second_per_gpu:5202.315378105177 mfu:24.54020924505271 +Step 1168 | loss:1.494553804397583 lr:9.97775878445217e-05 tokens_per_second_per_gpu:3254.9554194765697 mfu:15.354180066331695 +Step 1169 | loss:1.5625394582748413 lr:9.977717911483401e-05 tokens_per_second_per_gpu:5376.664999158964 mfu:25.3626461546762 +Step 1170 | loss:1.3345648050308228 lr:9.97767700107658e-05 tokens_per_second_per_gpu:5322.231041460884 mfu:25.105871888823486 +Step 1171 | loss:1.3954572677612305 lr:9.977636053232017e-05 tokens_per_second_per_gpu:4142.959028577534 mfu:19.543044599499773 +Step 1172 | loss:1.3987128734588623 lr:9.977595067950018e-05 tokens_per_second_per_gpu:3676.6343105040414 mfu:17.343311341145906 +Step 1173 | loss:1.438596248626709 lr:9.977554045230891e-05 tokens_per_second_per_gpu:5362.272641102066 mfu:25.294754946132315 +Step 1174 | loss:1.3594493865966797 lr:9.977512985074944e-05 tokens_per_second_per_gpu:4175.922456740362 mfu:19.69853871428462 +Step 1175 | loss:1.5850812196731567 lr:9.977471887482489e-05 tokens_per_second_per_gpu:3694.5315118995054 mfu:17.427735493705743 +Step 1176 | loss:1.4471791982650757 lr:9.97743075245383e-05 tokens_per_second_per_gpu:5332.947432061259 mfu:25.156422931690233 +Step 1177 | loss:1.4640101194381714 lr:9.977389579989281e-05 tokens_per_second_per_gpu:4024.478323218672 mfu:18.984150897429192 +Step 1178 | loss:1.5468227863311768 lr:9.97734837008915e-05 tokens_per_second_per_gpu:4783.70398349269 mfu:22.565547874196838 +Step 1179 | loss:1.6154512166976929 lr:9.977307122753745e-05 tokens_per_second_per_gpu:3846.526256885675 mfu:18.144720638783124 +Step 1180 | loss:1.4182602167129517 lr:9.97726583798338e-05 tokens_per_second_per_gpu:4024.6614813698707 mfu:18.9850148856784 +Step 1181 | loss:1.4457632303237915 lr:9.977224515778363e-05 tokens_per_second_per_gpu:4118.700572297046 mfu:19.428613322305278 +Step 1182 | loss:1.380250334739685 lr:9.977183156139004e-05 tokens_per_second_per_gpu:5080.3865468911 mfu:23.965050145012146 +Step 1183 | loss:1.4181199073791504 lr:9.977141759065616e-05 tokens_per_second_per_gpu:4559.560689571286 mfu:21.508225713980163 +Step 1184 | loss:1.4057700634002686 lr:9.97710032455851e-05 tokens_per_second_per_gpu:4721.672699173624 mfu:22.27293572243535 +Step 1185 | loss:1.293125867843628 lr:9.977058852617998e-05 tokens_per_second_per_gpu:5329.668982418492 mfu:25.140957925364113 +Step 1186 | loss:1.5010584592819214 lr:9.97701734324439e-05 tokens_per_second_per_gpu:4679.777521437912 mfu:22.075309021001587 +Step 1187 | loss:1.6563317775726318 lr:9.976975796438002e-05 tokens_per_second_per_gpu:3873.720147900855 mfu:18.272998862457975 +Step 1188 | loss:1.4634876251220703 lr:9.976934212199142e-05 tokens_per_second_per_gpu:3938.0407831234347 mfu:18.576410273034874 +Step 1189 | loss:1.3334234952926636 lr:9.976892590528124e-05 tokens_per_second_per_gpu:5348.3130959133605 mfu:25.228905389733185 +Step 1190 | loss:1.3495954275131226 lr:9.976850931425264e-05 tokens_per_second_per_gpu:4715.702164715115 mfu:22.24477169271634 +Step 1191 | loss:1.2970372438430786 lr:9.976809234890871e-05 tokens_per_second_per_gpu:5340.389365060004 mfu:25.19152779937001 +Step 1192 | loss:1.3748219013214111 lr:9.976767500925262e-05 tokens_per_second_per_gpu:4453.324866448456 mfu:21.00709321060928 +Step 1193 | loss:1.3475651741027832 lr:9.976725729528749e-05 tokens_per_second_per_gpu:4578.826767186344 mfu:21.599107089222187 +Step 1194 | loss:1.5558772087097168 lr:9.976683920701647e-05 tokens_per_second_per_gpu:4023.790982484097 mfu:18.98090859391178 +Step 1195 | loss:1.3555434942245483 lr:9.976642074444268e-05 tokens_per_second_per_gpu:5359.837491402 mfu:25.28326792951772 +Step 1196 | loss:1.3956663608551025 lr:9.976600190756932e-05 tokens_per_second_per_gpu:5218.340123343102 mfu:24.615800702443476 +Step 1197 | loss:1.3831501007080078 lr:9.97655826963995e-05 tokens_per_second_per_gpu:5330.271316462708 mfu:25.14379923781946 +Step 1198 | loss:1.398629069328308 lr:9.976516311093638e-05 tokens_per_second_per_gpu:3937.0894747754605 mfu:18.571922789247544 +Step 1199 | loss:1.2830302715301514 lr:9.976474315118313e-05 tokens_per_second_per_gpu:3910.313880221116 mfu:18.445617741347192 +Step 1200 | loss:1.5908459424972534 lr:9.976432281714288e-05 tokens_per_second_per_gpu:5370.36882317971 mfu:25.332946018343666 +Step 1201 | loss:1.346169114112854 lr:9.976390210881883e-05 tokens_per_second_per_gpu:5289.571407972715 mfu:24.951810825351725 +Step 1202 | loss:1.3215422630310059 lr:9.976348102621409e-05 tokens_per_second_per_gpu:5361.174418667359 mfu:25.289574443531734 +Step 1203 | loss:1.3631013631820679 lr:9.976305956933189e-05 tokens_per_second_per_gpu:4317.26700100041 mfu:20.365285045425267 +Step 1204 | loss:1.4140599966049194 lr:9.976263773817536e-05 tokens_per_second_per_gpu:4422.183543213727 mfu:20.860194275653434 +Step 1205 | loss:1.3458086252212524 lr:9.976221553274767e-05 tokens_per_second_per_gpu:5365.17957583861 mfu:25.308467453258803 +Step 1206 | loss:1.5268675088882446 lr:9.976179295305201e-05 tokens_per_second_per_gpu:3879.066137729421 mfu:18.298216808599236 +Step 1207 | loss:1.4999022483825684 lr:9.976136999909156e-05 tokens_per_second_per_gpu:5338.383475382285 mfu:25.182065675520143 +Step 1208 | loss:1.3215454816818237 lr:9.976094667086951e-05 tokens_per_second_per_gpu:4465.584417055299 mfu:21.064923602516785 +Step 1209 | loss:1.4073920249938965 lr:9.9760522968389e-05 tokens_per_second_per_gpu:3403.5415322139074 mfu:16.055086111518705 +Step 1210 | loss:1.433641791343689 lr:9.976009889165327e-05 tokens_per_second_per_gpu:3554.268622467317 mfu:16.766091512937454 +Step 1211 | loss:1.344834804534912 lr:9.975967444066548e-05 tokens_per_second_per_gpu:3127.8661459675 mfu:14.754678279523404 +Step 1212 | loss:1.3581064939498901 lr:9.975924961542884e-05 tokens_per_second_per_gpu:4566.777302366928 mfu:21.542267707818308 +Step 1213 | loss:1.2447245121002197 lr:9.975882441594652e-05 tokens_per_second_per_gpu:5362.371931429898 mfu:25.295223315549375 +Step 1214 | loss:1.3864994049072266 lr:9.975839884222173e-05 tokens_per_second_per_gpu:5314.107730175379 mfu:25.06755284350994 +Step 1215 | loss:1.2533930540084839 lr:9.975797289425768e-05 tokens_per_second_per_gpu:4309.177122644349 mfu:20.32712370894415 +Step 1216 | loss:1.4912582635879517 lr:9.975754657205758e-05 tokens_per_second_per_gpu:3890.6614792177666 mfu:18.352914012769023 +Step 1217 | loss:1.5743367671966553 lr:9.975711987562462e-05 tokens_per_second_per_gpu:5374.084079861385 mfu:25.350471517999285 +Step 1218 | loss:1.4620164632797241 lr:9.9756692804962e-05 tokens_per_second_per_gpu:5315.775380509637 mfu:25.075419434667122 +Step 1219 | loss:1.4811491966247559 lr:9.975626536007294e-05 tokens_per_second_per_gpu:3942.6671789352727 mfu:18.598233771423835 +Step 1220 | loss:1.58191978931427 lr:9.975583754096067e-05 tokens_per_second_per_gpu:5326.739618576551 mfu:25.12713961632067 +Step 1221 | loss:1.3402252197265625 lr:9.97554093476284e-05 tokens_per_second_per_gpu:5300.537192951428 mfu:25.003538304052153 +Step 1222 | loss:1.2550681829452515 lr:9.975498078007934e-05 tokens_per_second_per_gpu:4352.586563911955 mfu:20.53189344981785 +Step 1223 | loss:1.3411753177642822 lr:9.975455183831671e-05 tokens_per_second_per_gpu:5274.497047252674 mfu:24.88070249388414 +Step 1224 | loss:1.5172168016433716 lr:9.975412252234376e-05 tokens_per_second_per_gpu:4689.31646738588 mfu:22.120305856550157 +Step 1225 | loss:1.3990811109542847 lr:9.97536928321637e-05 tokens_per_second_per_gpu:4323.04585843505 mfu:20.392544901919212 +Step 1226 | loss:1.553388237953186 lr:9.975326276777978e-05 tokens_per_second_per_gpu:4271.082167056692 mfu:20.14742330376752 +Step 1227 | loss:1.472485899925232 lr:9.975283232919521e-05 tokens_per_second_per_gpu:3064.2177304639767 mfu:14.454437844053798 +Step 1228 | loss:1.5327445268630981 lr:9.975240151641325e-05 tokens_per_second_per_gpu:3979.7643516214544 mfu:18.773227464414884 +Step 1229 | loss:1.5118849277496338 lr:9.975197032943712e-05 tokens_per_second_per_gpu:4091.4170493468155 mfu:19.29991228950001 +Step 1230 | loss:1.5399601459503174 lr:9.975153876827008e-05 tokens_per_second_per_gpu:3682.863488954803 mfu:17.372695438705584 +Step 1231 | loss:1.5654951333999634 lr:9.975110683291536e-05 tokens_per_second_per_gpu:4517.583792032962 mfu:21.310213526294646 +Step 1232 | loss:1.298698902130127 lr:9.975067452337623e-05 tokens_per_second_per_gpu:3979.947657976213 mfu:18.774092151765572 +Step 1233 | loss:1.3692203760147095 lr:9.975024183965591e-05 tokens_per_second_per_gpu:3169.777048809176 mfu:14.952379158964975 +Step 1234 | loss:1.6475082635879517 lr:9.974980878175768e-05 tokens_per_second_per_gpu:5383.573725984378 mfu:25.39523579786202 +Step 1235 | loss:1.3442647457122803 lr:9.97493753496848e-05 tokens_per_second_per_gpu:4239.040955476823 mfu:19.99627944195087 +Step 1236 | loss:1.667319655418396 lr:9.974894154344053e-05 tokens_per_second_per_gpu:3524.877230771783 mfu:16.6274473036211 +Step 1237 | loss:1.2281088829040527 lr:9.97485073630281e-05 tokens_per_second_per_gpu:5359.489017736069 mfu:25.28162411977979 +Step 1238 | loss:1.493471622467041 lr:9.97480728084508e-05 tokens_per_second_per_gpu:5363.8685789090105 mfu:25.302283257062744 +Step 1239 | loss:1.4810537099838257 lr:9.974763787971189e-05 tokens_per_second_per_gpu:5195.6438927968575 mfu:24.508738710580598 +Step 1240 | loss:1.5451627969741821 lr:9.974720257681466e-05 tokens_per_second_per_gpu:5374.543552130022 mfu:25.3526389270845 +Step 1241 | loss:1.5358357429504395 lr:9.974676689976238e-05 tokens_per_second_per_gpu:4461.101270609471 mfu:21.043775835828153 +Step 1242 | loss:1.392395257949829 lr:9.974633084855831e-05 tokens_per_second_per_gpu:4999.875279491908 mfu:23.585264760051462 +Step 1243 | loss:1.38040030002594 lr:9.974589442320573e-05 tokens_per_second_per_gpu:4523.171776953273 mfu:21.33657300457229 +Step 1244 | loss:1.4656636714935303 lr:9.974545762370795e-05 tokens_per_second_per_gpu:3394.2912777358365 mfu:16.011451082889483 +Step 1245 | loss:1.4370323419570923 lr:9.974502045006822e-05 tokens_per_second_per_gpu:5026.057620456269 mfu:23.708771329548362 +Step 1246 | loss:1.3455767631530762 lr:9.974458290228984e-05 tokens_per_second_per_gpu:5314.42865858434 mfu:25.06906671757104 +Step 1247 | loss:1.4187252521514893 lr:9.97441449803761e-05 tokens_per_second_per_gpu:4672.703907539163 mfu:22.04194157735821 +Step 1248 | loss:1.5465279817581177 lr:9.974370668433031e-05 tokens_per_second_per_gpu:5328.629851306089 mfu:25.136056166612857 +Step 1249 | loss:1.6151363849639893 lr:9.974326801415576e-05 tokens_per_second_per_gpu:4191.308184588828 mfu:19.771115817621695 +Step 1250 | loss:1.280426263809204 lr:9.974282896985574e-05 tokens_per_second_per_gpu:4231.709754434422 mfu:19.961696915801987 +Step 1251 | loss:1.3444992303848267 lr:9.974238955143356e-05 tokens_per_second_per_gpu:4605.389299507109 mfu:21.72440708621458 +Step 1252 | loss:1.442343831062317 lr:9.974194975889251e-05 tokens_per_second_per_gpu:5356.192041883482 mfu:25.266071722161048 +Step 1253 | loss:1.5196537971496582 lr:9.974150959223591e-05 tokens_per_second_per_gpu:4055.996339600462 mfu:19.132826758230205 +Step 1254 | loss:1.5567761659622192 lr:9.974106905146708e-05 tokens_per_second_per_gpu:5343.970278779849 mfu:25.208419580353404 +Step 1255 | loss:1.390025019645691 lr:9.974062813658931e-05 tokens_per_second_per_gpu:4693.966487143417 mfu:22.142240793121637 +Step 1256 | loss:1.5269917249679565 lr:9.974018684760593e-05 tokens_per_second_per_gpu:5157.741014547836 mfu:24.32994436698209 +Step 1257 | loss:1.3931539058685303 lr:9.973974518452027e-05 tokens_per_second_per_gpu:3752.6472743777995 mfu:17.7018774608874 +Step 1258 | loss:1.554884433746338 lr:9.973930314733563e-05 tokens_per_second_per_gpu:5375.805517959098 mfu:25.35859183521382 +Step 1259 | loss:1.3700549602508545 lr:9.973886073605535e-05 tokens_per_second_per_gpu:3432.426604393661 mfu:16.191341925292104 +Step 1260 | loss:1.3792682886123657 lr:9.973841795068276e-05 tokens_per_second_per_gpu:4778.623668301102 mfu:22.541583160646965 +Step 1261 | loss:1.4646567106246948 lr:9.973797479122116e-05 tokens_per_second_per_gpu:5363.5274431119815 mfu:25.300674061304385 +Step 1262 | loss:1.4316593408584595 lr:9.97375312576739e-05 tokens_per_second_per_gpu:5346.861096269075 mfu:25.2220560596744 +Step 1263 | loss:1.7209479808807373 lr:9.973708735004435e-05 tokens_per_second_per_gpu:3329.1665320262814 mfu:15.704246545950122 +Step 1264 | loss:1.4018144607543945 lr:9.973664306833581e-05 tokens_per_second_per_gpu:5340.446433764663 mfu:25.191797002186814 +Step 1265 | loss:1.2465382814407349 lr:9.973619841255162e-05 tokens_per_second_per_gpu:5171.041681572312 mfu:24.392685882664345 +Step 1266 | loss:1.427857518196106 lr:9.973575338269514e-05 tokens_per_second_per_gpu:4960.075704958925 mfu:23.39752337647175 +Step 1267 | loss:1.307967185974121 lr:9.97353079787697e-05 tokens_per_second_per_gpu:4368.504974577949 mfu:20.606983308890264 +Step 1268 | loss:1.356313705444336 lr:9.97348622007787e-05 tokens_per_second_per_gpu:5314.280817974649 mfu:25.06836932819092 +Step 1269 | loss:1.3377645015716553 lr:9.973441604872542e-05 tokens_per_second_per_gpu:4943.216852607996 mfu:23.3179972935156 +Step 1270 | loss:1.4150828123092651 lr:9.973396952261326e-05 tokens_per_second_per_gpu:5348.826836255432 mfu:25.231328790579873 +Step 1271 | loss:1.4665403366088867 lr:9.973352262244558e-05 tokens_per_second_per_gpu:4129.506496050941 mfu:19.479586708333105 +Step 1272 | loss:1.4398056268692017 lr:9.973307534822572e-05 tokens_per_second_per_gpu:5313.688727261241 mfu:25.06557633527468 +Step 1273 | loss:1.4394047260284424 lr:9.973262769995706e-05 tokens_per_second_per_gpu:4499.507199931852 mfu:21.224943157169065 +Step 1274 | loss:1.3437680006027222 lr:9.973217967764296e-05 tokens_per_second_per_gpu:5096.785566605023 mfu:24.042407118964 +Step 1275 | loss:1.4493697881698608 lr:9.973173128128678e-05 tokens_per_second_per_gpu:4614.410623118744 mfu:21.766962208887186 +Step 1276 | loss:1.47506844997406 lr:9.973128251089192e-05 tokens_per_second_per_gpu:4419.735757173098 mfu:20.848647651264425 +Step 1277 | loss:1.4214668273925781 lr:9.973083336646172e-05 tokens_per_second_per_gpu:5328.2659472721925 mfu:25.134339569196538 +Step 1278 | loss:1.4730639457702637 lr:9.973038384799959e-05 tokens_per_second_per_gpu:5290.715883933904 mfu:24.957209513728383 +Step 1279 | loss:1.5306209325790405 lr:9.97299339555089e-05 tokens_per_second_per_gpu:4180.18329072773 mfu:19.718637795176278 +Step 1280 | loss:1.4278630018234253 lr:9.972948368899301e-05 tokens_per_second_per_gpu:4347.75967776419 mfu:20.509124204307145 +Step 1281 | loss:1.4139671325683594 lr:9.972903304845535e-05 tokens_per_second_per_gpu:4396.0355708605575 mfu:20.736849828759418 +Step 1282 | loss:1.5879279375076294 lr:9.972858203389927e-05 tokens_per_second_per_gpu:3079.791255094904 mfu:14.527900816855176 +Step 1283 | loss:1.5467960834503174 lr:9.97281306453282e-05 tokens_per_second_per_gpu:5319.588641097754 mfu:25.09340723547061 +Step 1284 | loss:1.5388461351394653 lr:9.97276788827455e-05 tokens_per_second_per_gpu:4289.307739681316 mfu:20.23339643944122 +Step 1285 | loss:1.5197480916976929 lr:9.972722674615458e-05 tokens_per_second_per_gpu:5345.109482914157 mfu:25.2137934006232 +Step 1286 | loss:1.4045164585113525 lr:9.972677423555886e-05 tokens_per_second_per_gpu:5142.520484404518 mfu:24.258146529405835 +Step 1287 | loss:1.2278801202774048 lr:9.972632135096171e-05 tokens_per_second_per_gpu:4246.035160616465 mfu:20.02927230092901 +Step 1288 | loss:1.3590651750564575 lr:9.972586809236654e-05 tokens_per_second_per_gpu:4295.630757472198 mfu:20.263223193178884 +Step 1289 | loss:1.4044110774993896 lr:9.97254144597768e-05 tokens_per_second_per_gpu:4333.559081019142 mfu:20.442137566589107 +Step 1290 | loss:1.4323666095733643 lr:9.972496045319585e-05 tokens_per_second_per_gpu:3724.8868991225604 mfu:17.57092703440001 +Step 1291 | loss:1.5953526496887207 lr:9.972450607262716e-05 tokens_per_second_per_gpu:4387.784005525589 mfu:20.69792578721215 +Step 1292 | loss:1.4740447998046875 lr:9.972405131807408e-05 tokens_per_second_per_gpu:5359.061812744432 mfu:25.27960892094555 +Step 1293 | loss:1.4695334434509277 lr:9.972359618954009e-05 tokens_per_second_per_gpu:4094.5570571844178 mfu:19.31472424221056 +Step 1294 | loss:1.3060637712478638 lr:9.972314068702856e-05 tokens_per_second_per_gpu:4840.711787045229 mfu:22.83446341008842 +Step 1295 | loss:1.4369248151779175 lr:9.972268481054297e-05 tokens_per_second_per_gpu:5377.679799862311 mfu:25.36743314273674 +Step 1296 | loss:1.3745012283325195 lr:9.97222285600867e-05 tokens_per_second_per_gpu:5330.043964779602 mfu:25.142726781140055 +Step 1297 | loss:1.5365265607833862 lr:9.972177193566321e-05 tokens_per_second_per_gpu:5339.233179978876 mfu:25.18607387707682 +Step 1298 | loss:1.3289302587509155 lr:9.972131493727594e-05 tokens_per_second_per_gpu:4568.271838623973 mfu:21.54931769033697 +Step 1299 | loss:1.314978837966919 lr:9.97208575649283e-05 tokens_per_second_per_gpu:4819.956832772812 mfu:22.736558749625075 +Step 1300 | loss:1.343983769416809 lr:9.972039981862378e-05 tokens_per_second_per_gpu:5347.301737316253 mfu:25.224134638675057 +Step 1301 | loss:1.528726577758789 lr:9.971994169836576e-05 tokens_per_second_per_gpu:5344.752688383994 mfu:25.212110339949 +Step 1302 | loss:1.5691673755645752 lr:9.971948320415772e-05 tokens_per_second_per_gpu:4706.254156966135 mfu:22.200203828168057 +Step 1303 | loss:1.3422445058822632 lr:9.97190243360031e-05 tokens_per_second_per_gpu:4283.95678106422 mfu:20.20815505467669 +Step 1304 | loss:1.2840240001678467 lr:9.971856509390535e-05 tokens_per_second_per_gpu:5363.938109199809 mfu:25.30261124330753 +Step 1305 | loss:1.3337433338165283 lr:9.971810547786793e-05 tokens_per_second_per_gpu:4238.857839644232 mfu:19.99541565332575 +Step 1306 | loss:1.3775771856307983 lr:9.97176454878943e-05 tokens_per_second_per_gpu:4933.48654471717 mfu:23.272097771032858 +Step 1307 | loss:1.522688865661621 lr:9.971718512398791e-05 tokens_per_second_per_gpu:2988.274219440076 mfu:14.096199345254965 +Step 1308 | loss:1.4693609476089478 lr:9.971672438615224e-05 tokens_per_second_per_gpu:5373.611538552413 mfu:25.348242460020995 +Step 1309 | loss:1.4178026914596558 lr:9.971626327439072e-05 tokens_per_second_per_gpu:3607.657588352774 mfu:17.017936374116033 +Step 1310 | loss:1.4135380983352661 lr:9.971580178870686e-05 tokens_per_second_per_gpu:4087.7441523373823 mfu:19.282586607645666 +Step 1311 | loss:1.3416721820831299 lr:9.97153399291041e-05 tokens_per_second_per_gpu:4538.733191796013 mfu:21.409978853436648 +Step 1312 | loss:1.3631232976913452 lr:9.971487769558593e-05 tokens_per_second_per_gpu:5372.163710793662 mfu:25.341412809458085 +Step 1313 | loss:1.3463200330734253 lr:9.971441508815582e-05 tokens_per_second_per_gpu:5328.011605454393 mfu:25.13313979544679 +Step 1314 | loss:1.4155290126800537 lr:9.971395210681724e-05 tokens_per_second_per_gpu:4724.872403008423 mfu:22.288029271349863 +Step 1315 | loss:1.4679828882217407 lr:9.971348875157369e-05 tokens_per_second_per_gpu:4600.414624996854 mfu:21.700940697782713 +Step 1316 | loss:1.259619116783142 lr:9.971302502242865e-05 tokens_per_second_per_gpu:5359.818556953763 mfu:25.28317861249484 +Step 1317 | loss:1.3606138229370117 lr:9.971256091938562e-05 tokens_per_second_per_gpu:5301.204686421379 mfu:25.006686984636044 +Step 1318 | loss:1.3950319290161133 lr:9.971209644244806e-05 tokens_per_second_per_gpu:4864.934688410946 mfu:22.948726968683644 +Step 1319 | loss:1.429618239402771 lr:9.971163159161947e-05 tokens_per_second_per_gpu:5301.520306920353 mfu:25.008175820380114 +Step 1320 | loss:1.404018521308899 lr:9.971116636690336e-05 tokens_per_second_per_gpu:4466.110195828639 mfu:21.067403790697735 +Step 1321 | loss:1.2490577697753906 lr:9.971070076830323e-05 tokens_per_second_per_gpu:3491.1374691856195 mfu:16.468291091622238 +Step 1322 | loss:1.3573579788208008 lr:9.971023479582257e-05 tokens_per_second_per_gpu:4122.326413807016 mfu:19.44571703534016 +Step 1323 | loss:1.350293517112732 lr:9.970976844946491e-05 tokens_per_second_per_gpu:5355.087940344537 mfu:25.260863486822696 +Step 1324 | loss:1.3509397506713867 lr:9.970930172923371e-05 tokens_per_second_per_gpu:5189.148750744595 mfu:24.478100017334572 +Step 1325 | loss:1.5681846141815186 lr:9.970883463513255e-05 tokens_per_second_per_gpu:4475.040524708869 mfu:21.109529675697008 +Step 1326 | loss:1.513684868812561 lr:9.970836716716486e-05 tokens_per_second_per_gpu:3608.280693074942 mfu:17.020875665403132 +Step 1327 | loss:1.3714370727539062 lr:9.970789932533421e-05 tokens_per_second_per_gpu:4791.105648666662 mfu:22.600462791676765 +Step 1328 | loss:1.3474040031433105 lr:9.970743110964412e-05 tokens_per_second_per_gpu:5014.357435726545 mfu:23.65357956192089 +Step 1329 | loss:1.3588149547576904 lr:9.970696252009809e-05 tokens_per_second_per_gpu:4188.268566024666 mfu:19.75677741824214 +Step 1330 | loss:1.3259518146514893 lr:9.970649355669966e-05 tokens_per_second_per_gpu:5368.33646229785 mfu:25.323359025307404 +Step 1331 | loss:1.4501092433929443 lr:9.970602421945231e-05 tokens_per_second_per_gpu:5373.722523670518 mfu:25.348765995759113 +Step 1332 | loss:1.417231559753418 lr:9.970555450835966e-05 tokens_per_second_per_gpu:4790.098393049377 mfu:22.5957113950745 +Step 1333 | loss:1.5024349689483643 lr:9.970508442342517e-05 tokens_per_second_per_gpu:4624.096367481043 mfu:21.812651517602543 +Step 1334 | loss:1.3512704372406006 lr:9.97046139646524e-05 tokens_per_second_per_gpu:3946.4826564175714 mfu:18.616232029695425 +Step 1335 | loss:1.2492432594299316 lr:9.970414313204488e-05 tokens_per_second_per_gpu:4665.922394091872 mfu:22.00995202138183 +Step 1336 | loss:1.3937782049179077 lr:9.970367192560617e-05 tokens_per_second_per_gpu:5384.160607707136 mfu:25.39800421908062 +Step 1337 | loss:1.3647762537002563 lr:9.970320034533978e-05 tokens_per_second_per_gpu:3653.443079977874 mfu:17.23391434992203 +Step 1338 | loss:1.5222880840301514 lr:9.97027283912493e-05 tokens_per_second_per_gpu:4682.965794838836 mfu:22.090348650609283 +Step 1339 | loss:1.3931907415390015 lr:9.970225606333825e-05 tokens_per_second_per_gpu:4861.970243397873 mfu:22.934743175770503 +Step 1340 | loss:1.348159670829773 lr:9.970178336161018e-05 tokens_per_second_per_gpu:4950.200495757829 mfu:23.35094032978575 +Step 1341 | loss:1.2647852897644043 lr:9.970131028606868e-05 tokens_per_second_per_gpu:4100.130588938527 mfu:19.34101554243727 +Step 1342 | loss:1.3053587675094604 lr:9.970083683671726e-05 tokens_per_second_per_gpu:5339.081869074457 mfu:25.185360117724368 +Step 1343 | loss:1.3684492111206055 lr:9.970036301355952e-05 tokens_per_second_per_gpu:5088.465442156618 mfu:24.003159672381265 +Step 1344 | loss:1.5412577390670776 lr:9.9699888816599e-05 tokens_per_second_per_gpu:3483.8426492740427 mfu:16.433880181474713 +Step 1345 | loss:1.3712347745895386 lr:9.969941424583926e-05 tokens_per_second_per_gpu:4979.96445657484 mfu:23.49134201121496 +Step 1346 | loss:1.465103030204773 lr:9.969893930128391e-05 tokens_per_second_per_gpu:3845.562767735858 mfu:18.14017569607524 +Step 1347 | loss:1.514612078666687 lr:9.969846398293647e-05 tokens_per_second_per_gpu:4670.531756374063 mfu:22.031695169705227 +Step 1348 | loss:1.5528717041015625 lr:9.969798829080057e-05 tokens_per_second_per_gpu:4648.891939004013 mfu:21.929616459037135 +Step 1349 | loss:1.236879825592041 lr:9.969751222487973e-05 tokens_per_second_per_gpu:5316.876929965114 mfu:25.080615631392323 +Step 1350 | loss:1.3990705013275146 lr:9.969703578517757e-05 tokens_per_second_per_gpu:5331.625857724384 mfu:25.15018884005936 +Step 1351 | loss:1.3303842544555664 lr:9.969655897169768e-05 tokens_per_second_per_gpu:5342.197653878883 mfu:25.200057806254552 +Step 1352 | loss:1.3599753379821777 lr:9.969608178444361e-05 tokens_per_second_per_gpu:4748.4657954099 mfu:22.399323328755465 +Step 1353 | loss:1.2253789901733398 lr:9.969560422341897e-05 tokens_per_second_per_gpu:4846.786249456451 mfu:22.86311768569231 +Step 1354 | loss:1.386418104171753 lr:9.969512628862736e-05 tokens_per_second_per_gpu:3381.7897253195424 mfu:15.952479127157035 +Step 1355 | loss:1.2997344732284546 lr:9.969464798007235e-05 tokens_per_second_per_gpu:5343.200389785141 mfu:25.204787882608755 +Step 1356 | loss:1.507775902748108 lr:9.969416929775756e-05 tokens_per_second_per_gpu:5354.466156686491 mfu:25.25793042722046 +Step 1357 | loss:1.3333661556243896 lr:9.969369024168657e-05 tokens_per_second_per_gpu:4545.156449003137 mfu:21.440278453603337 +Step 1358 | loss:1.2480919361114502 lr:9.9693210811863e-05 tokens_per_second_per_gpu:5344.862323841329 mfu:25.212627509106984 +Step 1359 | loss:1.3609648942947388 lr:9.969273100829046e-05 tokens_per_second_per_gpu:4608.175176328503 mfu:21.73754854250069 +Step 1360 | loss:1.657230257987976 lr:9.969225083097254e-05 tokens_per_second_per_gpu:3460.83818469088 mfu:16.325364197069547 +Step 1361 | loss:1.5810599327087402 lr:9.969177027991287e-05 tokens_per_second_per_gpu:5257.597271353566 mfu:24.80098336756895 +Step 1362 | loss:1.5246009826660156 lr:9.969128935511504e-05 tokens_per_second_per_gpu:4182.26908590423 mfu:19.72847684689718 +Step 1363 | loss:1.3836390972137451 lr:9.96908080565827e-05 tokens_per_second_per_gpu:4301.599970673029 mfu:20.291380990299995 +Step 1364 | loss:1.4254558086395264 lr:9.969032638431944e-05 tokens_per_second_per_gpu:5309.914123592211 mfu:25.04777088575407 +Step 1365 | loss:1.3016531467437744 lr:9.968984433832888e-05 tokens_per_second_per_gpu:4995.863503746855 mfu:23.566340529382522 +Step 1366 | loss:1.3979640007019043 lr:9.96893619186147e-05 tokens_per_second_per_gpu:3804.3536234658095 mfu:17.94578513154839 +Step 1367 | loss:1.3200713396072388 lr:9.968887912518046e-05 tokens_per_second_per_gpu:5346.623628848831 mfu:25.220935885336996 +Step 1368 | loss:1.3440911769866943 lr:9.968839595802982e-05 tokens_per_second_per_gpu:4107.656657192664 mfu:19.37651729047169 +Step 1369 | loss:1.5225614309310913 lr:9.968791241716642e-05 tokens_per_second_per_gpu:3360.11262198641 mfu:15.850224532239446 +Step 1370 | loss:1.6088025569915771 lr:9.968742850259389e-05 tokens_per_second_per_gpu:5375.998603653526 mfu:25.35950265337825 +Step 1371 | loss:1.6675093173980713 lr:9.968694421431589e-05 tokens_per_second_per_gpu:3853.461397084552 mfu:18.17743487835816 +Step 1372 | loss:1.5447747707366943 lr:9.968645955233601e-05 tokens_per_second_per_gpu:5339.699112234633 mfu:25.188271759023458 +Step 1373 | loss:1.3523383140563965 lr:9.968597451665793e-05 tokens_per_second_per_gpu:5081.1409857011895 mfu:23.96860896159987 +Step 1374 | loss:1.3346840143203735 lr:9.968548910728532e-05 tokens_per_second_per_gpu:4528.497940698295 mfu:21.3616974276948 +Step 1375 | loss:1.5810203552246094 lr:9.968500332422179e-05 tokens_per_second_per_gpu:4150.873121814799 mfu:19.580376727583495 +Step 1376 | loss:1.5521087646484375 lr:9.9684517167471e-05 tokens_per_second_per_gpu:5341.461160782355 mfu:25.19658364265955 +Step 1377 | loss:1.5051487684249878 lr:9.968403063703664e-05 tokens_per_second_per_gpu:4759.947899208725 mfu:22.45348637142344 +Step 1378 | loss:1.4343973398208618 lr:9.968354373292234e-05 tokens_per_second_per_gpu:5353.831686382133 mfu:25.25493752254301 +Step 1379 | loss:1.4319438934326172 lr:9.968305645513176e-05 tokens_per_second_per_gpu:4427.428477317219 mfu:20.884935524697646 +Step 1380 | loss:1.3347963094711304 lr:9.968256880366858e-05 tokens_per_second_per_gpu:4830.088947675354 mfu:22.784353664337427 +Step 1381 | loss:1.4165352582931519 lr:9.968208077853645e-05 tokens_per_second_per_gpu:4642.460352916317 mfu:21.899277570119732 +Step 1382 | loss:1.4228839874267578 lr:9.968159237973906e-05 tokens_per_second_per_gpu:5336.071077406405 mfu:25.171157699732138 +Step 1383 | loss:1.3647441864013672 lr:9.968110360728007e-05 tokens_per_second_per_gpu:3656.4327223288938 mfu:17.248017002977374 +Step 1384 | loss:1.4425538778305054 lr:9.968061446116315e-05 tokens_per_second_per_gpu:5086.776028950757 mfu:23.995190422045603 +Step 1385 | loss:1.498165488243103 lr:9.968012494139201e-05 tokens_per_second_per_gpu:5084.560384189694 mfu:23.98473884767969 +Step 1386 | loss:1.3488831520080566 lr:9.967963504797031e-05 tokens_per_second_per_gpu:5217.73787764979 mfu:24.612959806754407 +Step 1387 | loss:1.3805299997329712 lr:9.967914478090172e-05 tokens_per_second_per_gpu:5086.2852373646865 mfu:23.992875274396425 +Step 1388 | loss:1.3603582382202148 lr:9.967865414018995e-05 tokens_per_second_per_gpu:4768.097493124381 mfu:22.491929396387693 +Step 1389 | loss:1.4019885063171387 lr:9.967816312583869e-05 tokens_per_second_per_gpu:4957.662210401906 mfu:23.38613851086166 +Step 1390 | loss:1.3487679958343506 lr:9.967767173785162e-05 tokens_per_second_per_gpu:5301.043241091976 mfu:25.00592541947184 +Step 1391 | loss:1.4305919408798218 lr:9.967717997623245e-05 tokens_per_second_per_gpu:5377.747206356282 mfu:25.367751110669538 +Step 1392 | loss:1.262761116027832 lr:9.967668784098485e-05 tokens_per_second_per_gpu:3915.471658395643 mfu:18.469947861004133 +Step 1393 | loss:1.4928615093231201 lr:9.967619533211257e-05 tokens_per_second_per_gpu:4462.688671268114 mfu:21.051263875573454 +Step 1394 | loss:1.287206768989563 lr:9.967570244961926e-05 tokens_per_second_per_gpu:5358.883591888579 mfu:25.278768222760874 +Step 1395 | loss:1.4128044843673706 lr:9.967520919350866e-05 tokens_per_second_per_gpu:4106.3904065176675 mfu:19.37054416999309 +Step 1396 | loss:1.4575918912887573 lr:9.967471556378448e-05 tokens_per_second_per_gpu:3207.5703039062905 mfu:15.1306563914523 +Step 1397 | loss:1.249050498008728 lr:9.967422156045043e-05 tokens_per_second_per_gpu:3961.6585673109657 mfu:18.687819390656962 +Step 1398 | loss:1.3921570777893066 lr:9.96737271835102e-05 tokens_per_second_per_gpu:5354.041361938251 mfu:25.255926597915554 +Step 1399 | loss:1.3013505935668945 lr:9.967323243296756e-05 tokens_per_second_per_gpu:5349.079831845536 mfu:25.2325222139439 +Step 1400 | loss:1.3681581020355225 lr:9.967273730882618e-05 tokens_per_second_per_gpu:3701.3171360172328 mfu:17.45974444041592 +Step 1401 | loss:1.503650426864624 lr:9.96722418110898e-05 tokens_per_second_per_gpu:5291.749700172339 mfu:24.962086201312324 +Step 1402 | loss:1.1789448261260986 lr:9.967174593976216e-05 tokens_per_second_per_gpu:5202.772977351978 mfu:24.54236781877435 +Step 1403 | loss:1.4873135089874268 lr:9.967124969484698e-05 tokens_per_second_per_gpu:4856.223505959823 mfu:22.907634834780712 +Step 1404 | loss:1.4922752380371094 lr:9.9670753076348e-05 tokens_per_second_per_gpu:4221.062567141736 mfu:19.911472316744906 +Step 1405 | loss:1.5076136589050293 lr:9.967025608426894e-05 tokens_per_second_per_gpu:4603.586126932279 mfu:21.715901213526138 +Step 1406 | loss:1.4544761180877686 lr:9.966975871861355e-05 tokens_per_second_per_gpu:2887.9912931143876 mfu:13.62314767174513 +Step 1407 | loss:1.2971560955047607 lr:9.966926097938556e-05 tokens_per_second_per_gpu:5343.889301543904 mfu:25.20803759691531 +Step 1408 | loss:1.4090638160705566 lr:9.966876286658873e-05 tokens_per_second_per_gpu:5337.882614952721 mfu:25.17970303141829 +Step 1409 | loss:1.4731786251068115 lr:9.966826438022679e-05 tokens_per_second_per_gpu:5350.036002063703 mfu:25.23703263200245 +Step 1410 | loss:1.4089171886444092 lr:9.96677655203035e-05 tokens_per_second_per_gpu:5323.391341964056 mfu:25.111345224264632 +Step 1411 | loss:1.4105525016784668 lr:9.966726628682261e-05 tokens_per_second_per_gpu:3340.2893494182676 mfu:15.756714773336457 +Step 1412 | loss:1.3995970487594604 lr:9.966676667978786e-05 tokens_per_second_per_gpu:5364.7182087449255 mfu:25.30629110596055 +Step 1413 | loss:1.3361117839813232 lr:9.966626669920302e-05 tokens_per_second_per_gpu:4467.83250251676 mfu:21.075528205201422 +Step 1414 | loss:1.414633870124817 lr:9.966576634507186e-05 tokens_per_second_per_gpu:5358.799690672103 mfu:25.278372446407733 +Step 1415 | loss:1.4359205961227417 lr:9.966526561739814e-05 tokens_per_second_per_gpu:5335.0735004672315 mfu:25.16645195910588 +Step 1416 | loss:1.4815595149993896 lr:9.96647645161856e-05 tokens_per_second_per_gpu:3847.327243225463 mfu:18.148499028010267 +Step 1417 | loss:1.5276323556900024 lr:9.966426304143805e-05 tokens_per_second_per_gpu:4122.640226370276 mfu:19.447197342743 +Step 1418 | loss:1.3459593057632446 lr:9.966376119315923e-05 tokens_per_second_per_gpu:3815.491673844457 mfu:17.998325215531825 +Step 1419 | loss:1.4180480241775513 lr:9.966325897135293e-05 tokens_per_second_per_gpu:4763.072381503101 mfu:22.46822508749766 +Step 1420 | loss:1.3488348722457886 lr:9.966275637602291e-05 tokens_per_second_per_gpu:4794.638263744293 mfu:22.61712673972423 +Step 1421 | loss:1.6150497198104858 lr:9.9662253407173e-05 tokens_per_second_per_gpu:5298.2706474244815 mfu:24.992846622089008 +Step 1422 | loss:1.3811182975769043 lr:9.966175006480689e-05 tokens_per_second_per_gpu:5253.491556923715 mfu:24.781616012096226 +Step 1423 | loss:1.552683711051941 lr:9.966124634892846e-05 tokens_per_second_per_gpu:3927.8391888800957 mfu:18.528287612417586 +Step 1424 | loss:1.3726621866226196 lr:9.966074225954145e-05 tokens_per_second_per_gpu:4650.088633606519 mfu:21.93526147165431 +Step 1425 | loss:1.5091265439987183 lr:9.966023779664966e-05 tokens_per_second_per_gpu:4412.512507040143 mfu:20.814574347973718 +Step 1426 | loss:1.4883815050125122 lr:9.965973296025688e-05 tokens_per_second_per_gpu:5333.1081589503365 mfu:25.157181107848846 +Step 1427 | loss:1.297759771347046 lr:9.965922775036691e-05 tokens_per_second_per_gpu:4587.623575660405 mfu:21.64060313568462 +Step 1428 | loss:1.5367159843444824 lr:9.965872216698355e-05 tokens_per_second_per_gpu:4920.263028407577 mfu:23.209720188435977 +Step 1429 | loss:1.3509465456008911 lr:9.96582162101106e-05 tokens_per_second_per_gpu:4659.976949602471 mfu:21.981906345327896 +Step 1430 | loss:1.4865325689315796 lr:9.965770987975189e-05 tokens_per_second_per_gpu:4528.5249072877 mfu:21.361824633698 +Step 1431 | loss:1.3881356716156006 lr:9.96572031759112e-05 tokens_per_second_per_gpu:3445.318432732256 mfu:16.252154879137986 +Step 1432 | loss:1.537140965461731 lr:9.965669609859234e-05 tokens_per_second_per_gpu:3858.161656690077 mfu:18.199606804864402 +Step 1433 | loss:1.494384527206421 lr:9.965618864779915e-05 tokens_per_second_per_gpu:3661.604389597021 mfu:17.272412639858555 +Step 1434 | loss:1.4659653902053833 lr:9.96556808235354e-05 tokens_per_second_per_gpu:4687.807353088108 mfu:22.11318710692612 +Step 1435 | loss:1.4424471855163574 lr:9.965517262580497e-05 tokens_per_second_per_gpu:3648.110607483157 mfu:17.208760167350828 +Step 1436 | loss:1.468522071838379 lr:9.965466405461164e-05 tokens_per_second_per_gpu:4581.996403112126 mfu:21.614058802679757 +Step 1437 | loss:1.4961496591567993 lr:9.965415510995924e-05 tokens_per_second_per_gpu:4420.952447551576 mfu:20.854386987368493 +Step 1438 | loss:1.5914528369903564 lr:9.96536457918516e-05 tokens_per_second_per_gpu:3597.1778329204863 mfu:16.968501579711283 +Step 1439 | loss:1.3003332614898682 lr:9.965313610029255e-05 tokens_per_second_per_gpu:5300.55168504897 mfu:25.00360666574887 +Step 1440 | loss:1.534982681274414 lr:9.965262603528595e-05 tokens_per_second_per_gpu:5358.46421816644 mfu:25.27678996536101 +Step 1441 | loss:1.4266939163208008 lr:9.96521155968356e-05 tokens_per_second_per_gpu:4497.321235393666 mfu:21.21463158947843 +Step 1442 | loss:1.3244984149932861 lr:9.965160478494535e-05 tokens_per_second_per_gpu:5334.1145174677795 mfu:25.161928272678097 +Step 1443 | loss:1.4974250793457031 lr:9.965109359961903e-05 tokens_per_second_per_gpu:5065.610745846666 mfu:23.895350170473932 +Step 1444 | loss:1.3250157833099365 lr:9.965058204086053e-05 tokens_per_second_per_gpu:4608.415281032427 mfu:21.738681157355188 +Step 1445 | loss:1.3692203760147095 lr:9.965007010867366e-05 tokens_per_second_per_gpu:5046.906243423928 mfu:23.807117841228685 +Step 1446 | loss:1.2169398069381714 lr:9.964955780306227e-05 tokens_per_second_per_gpu:4821.076334236561 mfu:22.74183962903622 +Step 1447 | loss:1.5102674961090088 lr:9.964904512403023e-05 tokens_per_second_per_gpu:4169.894119219722 mfu:19.670102017659318 +Step 1448 | loss:1.5356007814407349 lr:9.964853207158138e-05 tokens_per_second_per_gpu:4102.547644186989 mfu:19.352417204436414 +Step 1449 | loss:1.5844513177871704 lr:9.964801864571958e-05 tokens_per_second_per_gpu:4091.6106626334513 mfu:19.300825596406902 +Step 1450 | loss:1.3662447929382324 lr:9.964750484644869e-05 tokens_per_second_per_gpu:5335.882909359521 mfu:25.170270078950207 +Step 1451 | loss:1.4584341049194336 lr:9.96469906737726e-05 tokens_per_second_per_gpu:4357.377997519211 mfu:20.55449545964615 +Step 1452 | loss:1.5559241771697998 lr:9.964647612769516e-05 tokens_per_second_per_gpu:5353.246202669609 mfu:25.252175695977215 +Step 1453 | loss:1.4315826892852783 lr:9.964596120822022e-05 tokens_per_second_per_gpu:5334.179005352594 mfu:25.16223247303353 +Step 1454 | loss:1.5586093664169312 lr:9.964544591535168e-05 tokens_per_second_per_gpu:3496.320734040895 mfu:16.492741436300747 +Step 1455 | loss:1.3795369863510132 lr:9.964493024909341e-05 tokens_per_second_per_gpu:5358.656910547019 mfu:25.277698928196916 +Step 1456 | loss:1.4747352600097656 lr:9.964441420944928e-05 tokens_per_second_per_gpu:3835.8783985920263 mfu:18.094492874500023 +Step 1457 | loss:1.434425711631775 lr:9.964389779642316e-05 tokens_per_second_per_gpu:5357.8364318547 mfu:25.273828590217146 +Step 1458 | loss:1.4096323251724243 lr:9.964338101001898e-05 tokens_per_second_per_gpu:5197.892780526704 mfu:24.519347097702095 +Step 1459 | loss:1.3029844760894775 lr:9.964286385024057e-05 tokens_per_second_per_gpu:4681.534973223512 mfu:22.08359922092646 +Step 1460 | loss:1.3142422437667847 lr:9.964234631709187e-05 tokens_per_second_per_gpu:4016.470306242743 mfu:18.94637571504515 +Step 1461 | loss:1.5317648649215698 lr:9.964182841057674e-05 tokens_per_second_per_gpu:4963.934227704052 mfu:23.41572468659238 +Step 1462 | loss:1.4079999923706055 lr:9.964131013069908e-05 tokens_per_second_per_gpu:3443.7976462233482 mfu:16.244981069702533 +Step 1463 | loss:1.660662055015564 lr:9.96407914774628e-05 tokens_per_second_per_gpu:4757.419496826166 mfu:22.44155946599498 +Step 1464 | loss:1.4225471019744873 lr:9.964027245087179e-05 tokens_per_second_per_gpu:5347.214492664548 mfu:25.223723090767493 +Step 1465 | loss:1.3455636501312256 lr:9.963975305092995e-05 tokens_per_second_per_gpu:5341.757976864595 mfu:25.19798377476152 +Step 1466 | loss:1.4530421495437622 lr:9.96392332776412e-05 tokens_per_second_per_gpu:4299.698470800904 mfu:20.28241129097385 +Step 1467 | loss:1.412163496017456 lr:9.963871313100944e-05 tokens_per_second_per_gpu:5343.847921052478 mfu:25.207842398077016 +Step 1468 | loss:1.3465595245361328 lr:9.963819261103858e-05 tokens_per_second_per_gpu:4227.0524597612175 mfu:19.939727662212068 +Step 1469 | loss:1.5583081245422363 lr:9.963767171773255e-05 tokens_per_second_per_gpu:3871.653897529999 mfu:18.26325200691998 +Step 1470 | loss:1.258327841758728 lr:9.963715045109525e-05 tokens_per_second_per_gpu:5113.994297045701 mfu:24.123583636564856 +Step 1471 | loss:1.3179517984390259 lr:9.96366288111306e-05 tokens_per_second_per_gpu:4171.179947801809 mfu:19.676167490466003 +Step 1472 | loss:1.6470588445663452 lr:9.963610679784255e-05 tokens_per_second_per_gpu:5189.824602231169 mfu:24.481288124108893 +Step 1473 | loss:1.2973586320877075 lr:9.9635584411235e-05 tokens_per_second_per_gpu:4283.615768863745 mfu:20.206546441944372 +Step 1474 | loss:1.5485506057739258 lr:9.96350616513119e-05 tokens_per_second_per_gpu:3617.515906628092 mfu:17.06443974342325 +Step 1475 | loss:1.4000308513641357 lr:9.963453851807714e-05 tokens_per_second_per_gpu:3668.35742286699 mfu:17.304267849979283 +Step 1476 | loss:1.4678289890289307 lr:9.963401501153469e-05 tokens_per_second_per_gpu:5368.385090252475 mfu:25.323588411666154 +Step 1477 | loss:1.2615411281585693 lr:9.963349113168849e-05 tokens_per_second_per_gpu:5328.945710379685 mfu:25.137546127754124 +Step 1478 | loss:1.4671251773834229 lr:9.963296687854246e-05 tokens_per_second_per_gpu:5352.167116267731 mfu:25.24708546130843 +Step 1479 | loss:1.4208688735961914 lr:9.963244225210055e-05 tokens_per_second_per_gpu:5338.565845072217 mfu:25.18292594446303 +Step 1480 | loss:1.4421354532241821 lr:9.963191725236672e-05 tokens_per_second_per_gpu:5324.427150523681 mfu:25.116231309967546 +Step 1481 | loss:1.6301841735839844 lr:9.96313918793449e-05 tokens_per_second_per_gpu:4570.940666972953 mfu:21.56190700900818 +Step 1482 | loss:1.5793125629425049 lr:9.963086613303906e-05 tokens_per_second_per_gpu:4190.736510282132 mfu:19.768419132379318 +Step 1483 | loss:1.3110004663467407 lr:9.963034001345312e-05 tokens_per_second_per_gpu:3949.0985108646782 mfu:18.628571461433218 +Step 1484 | loss:1.425008773803711 lr:9.962981352059109e-05 tokens_per_second_per_gpu:5364.619738544317 mfu:25.305826605223533 +Step 1485 | loss:1.3603216409683228 lr:9.962928665445688e-05 tokens_per_second_per_gpu:5352.165066214749 mfu:25.247075790858734 +Step 1486 | loss:1.5122898817062378 lr:9.962875941505447e-05 tokens_per_second_per_gpu:5342.0702917674325 mfu:25.199457017444622 +Step 1487 | loss:1.3846430778503418 lr:9.962823180238783e-05 tokens_per_second_per_gpu:4128.748762001572 mfu:19.47601234753947 +Step 1488 | loss:1.4654443264007568 lr:9.962770381646094e-05 tokens_per_second_per_gpu:3763.253978382042 mfu:17.75191109336572 +Step 1489 | loss:1.4705039262771606 lr:9.962717545727775e-05 tokens_per_second_per_gpu:5357.612177223021 mfu:25.272770742857727 +Step 1490 | loss:1.247074842453003 lr:9.962664672484223e-05 tokens_per_second_per_gpu:5365.072470920719 mfu:25.307962221087127 +Step 1491 | loss:1.485396385192871 lr:9.962611761915839e-05 tokens_per_second_per_gpu:4974.090265033053 mfu:23.463632447471923 +Step 1492 | loss:1.5028977394104004 lr:9.962558814023016e-05 tokens_per_second_per_gpu:5062.4784989643995 mfu:23.880574827513858 +Step 1493 | loss:1.500108242034912 lr:9.962505828806157e-05 tokens_per_second_per_gpu:3600.1381336404434 mfu:16.982465823286702 +Step 1494 | loss:1.4256598949432373 lr:9.962452806265657e-05 tokens_per_second_per_gpu:4056.9413665084967 mfu:19.13728461139362 +Step 1495 | loss:1.2967637777328491 lr:9.962399746401917e-05 tokens_per_second_per_gpu:5372.72367822446 mfu:25.34405427136945 +Step 1496 | loss:1.5240205526351929 lr:9.962346649215336e-05 tokens_per_second_per_gpu:4109.007280017364 mfu:19.382888408775813 +Step 1497 | loss:1.4325249195098877 lr:9.962293514706311e-05 tokens_per_second_per_gpu:5110.835727042964 mfu:24.10868412295433 +Step 1498 | loss:1.4493763446807861 lr:9.962240342875244e-05 tokens_per_second_per_gpu:4785.044129564309 mfu:22.57186957186916 +Step 1499 | loss:1.4227197170257568 lr:9.962187133722534e-05 tokens_per_second_per_gpu:4796.13458751473 mfu:22.624185154248543 +Step 1500 | loss:1.4250099658966064 lr:9.962133887248583e-05 tokens_per_second_per_gpu:5351.121605862836 mfu:25.242153610346072 +Step 1501 | loss:1.3592628240585327 lr:9.962080603453787e-05 tokens_per_second_per_gpu:4554.363098894677 mfu:21.48370779195818 +Step 1502 | loss:1.5331605672836304 lr:9.962027282338552e-05 tokens_per_second_per_gpu:5343.043553963327 mfu:25.20404806127831 +Step 1503 | loss:1.2370123863220215 lr:9.961973923903276e-05 tokens_per_second_per_gpu:4368.520670203326 mfu:20.607057347833116 +Step 1504 | loss:1.4669344425201416 lr:9.961920528148361e-05 tokens_per_second_per_gpu:4856.368962468401 mfu:22.908320977949113 +Step 1505 | loss:1.5711795091629028 lr:9.961867095074209e-05 tokens_per_second_per_gpu:5340.671795896186 mfu:25.192860073811957 +Step 1506 | loss:1.4912879467010498 lr:9.961813624681219e-05 tokens_per_second_per_gpu:4542.126946976742 mfu:21.425987775658488 +Step 1507 | loss:1.375012755393982 lr:9.961760116969798e-05 tokens_per_second_per_gpu:5027.912424225651 mfu:23.717520755390694 +Step 1508 | loss:1.3814630508422852 lr:9.961706571940346e-05 tokens_per_second_per_gpu:5234.638610263942 mfu:24.69268344605769 +Step 1509 | loss:1.4332760572433472 lr:9.961652989593264e-05 tokens_per_second_per_gpu:5334.755677736692 mfu:25.164952735059774 +Step 1510 | loss:1.4869756698608398 lr:9.961599369928959e-05 tokens_per_second_per_gpu:4277.790164695953 mfu:20.17906607313433 +Step 1511 | loss:1.3181464672088623 lr:9.96154571294783e-05 tokens_per_second_per_gpu:4921.601399881582 mfu:23.216033515028577 +Step 1512 | loss:1.4781287908554077 lr:9.961492018650284e-05 tokens_per_second_per_gpu:3330.5874210919264 mfu:15.710949122101441 +Step 1513 | loss:1.4161289930343628 lr:9.961438287036724e-05 tokens_per_second_per_gpu:5226.724655260728 mfu:24.655351970047573 +Step 1514 | loss:1.3550171852111816 lr:9.961384518107553e-05 tokens_per_second_per_gpu:5275.397130325853 mfu:24.884948339310334 +Step 1515 | loss:1.6363619565963745 lr:9.961330711863175e-05 tokens_per_second_per_gpu:3306.2662341140635 mfu:15.596221933504467 +Step 1516 | loss:1.3038239479064941 lr:9.961276868303996e-05 tokens_per_second_per_gpu:4241.948498853473 mfu:20.009994820136583 +Step 1517 | loss:1.3570975065231323 lr:9.961222987430421e-05 tokens_per_second_per_gpu:5097.55956270855 mfu:24.04605819064169 +Step 1518 | loss:1.6927186250686646 lr:9.961169069242855e-05 tokens_per_second_per_gpu:3913.583783824945 mfu:18.461042434549533 +Step 1519 | loss:1.4014261960983276 lr:9.961115113741701e-05 tokens_per_second_per_gpu:3740.6490794897627 mfu:17.645279928497388 +Step 1520 | loss:1.3022534847259521 lr:9.961061120927371e-05 tokens_per_second_per_gpu:5361.065647863558 mfu:25.28906135309199 +Step 1521 | loss:1.4492472410202026 lr:9.961007090800264e-05 tokens_per_second_per_gpu:4621.822995830013 mfu:21.801927635647434 +Step 1522 | loss:1.3271838426589966 lr:9.960953023360792e-05 tokens_per_second_per_gpu:5185.946912840985 mfu:24.462996401653363 +Step 1523 | loss:1.3544528484344482 lr:9.960898918609358e-05 tokens_per_second_per_gpu:5349.275826809979 mfu:25.233446755631363 +Step 1524 | loss:1.356529712677002 lr:9.96084477654637e-05 tokens_per_second_per_gpu:5318.721910387986 mfu:25.089318718833127 +Step 1525 | loss:1.5640848875045776 lr:9.960790597172238e-05 tokens_per_second_per_gpu:4204.677396885035 mfu:19.83418067304573 +Step 1526 | loss:1.3878248929977417 lr:9.960736380487365e-05 tokens_per_second_per_gpu:5182.653540936944 mfu:24.447461004474814 +Step 1527 | loss:1.5656182765960693 lr:9.96068212649216e-05 tokens_per_second_per_gpu:4886.196602938917 mfu:23.049023047168774 +Step 1528 | loss:1.3916172981262207 lr:9.960627835187032e-05 tokens_per_second_per_gpu:4823.004900968635 mfu:22.750937007358836 +Step 1529 | loss:1.4325145483016968 lr:9.96057350657239e-05 tokens_per_second_per_gpu:3572.4908697887327 mfu:16.852048962588558 +Step 1530 | loss:1.3380409479141235 lr:9.960519140648641e-05 tokens_per_second_per_gpu:4498.757696652834 mfu:21.22140762232349 +Step 1531 | loss:1.360102653503418 lr:9.960464737416196e-05 tokens_per_second_per_gpu:3886.425660916883 mfu:18.33293293513863 +Step 1532 | loss:1.3007420301437378 lr:9.96041029687546e-05 tokens_per_second_per_gpu:4781.5431792579975 mfu:22.555354991950573 +Step 1533 | loss:1.4565818309783936 lr:9.960355819026846e-05 tokens_per_second_per_gpu:4255.378339498634 mfu:20.073345669826896 +Step 1534 | loss:1.484337568283081 lr:9.960301303870765e-05 tokens_per_second_per_gpu:4471.198803257995 mfu:21.091407620147965 +Step 1535 | loss:1.411569356918335 lr:9.960246751407623e-05 tokens_per_second_per_gpu:4030.2671703255187 mfu:19.011457876911347 +Step 1536 | loss:1.4024348258972168 lr:9.960192161637833e-05 tokens_per_second_per_gpu:5197.571127449296 mfu:24.51782980525629 +Step 1537 | loss:1.3125420808792114 lr:9.960137534561804e-05 tokens_per_second_per_gpu:4874.938119594223 mfu:22.99591486033652 +Step 1538 | loss:1.279102087020874 lr:9.960082870179949e-05 tokens_per_second_per_gpu:4457.18015892859 mfu:21.025279283018797 +Step 1539 | loss:1.388414978981018 lr:9.960028168492676e-05 tokens_per_second_per_gpu:5349.814016380579 mfu:25.235985487660493 +Step 1540 | loss:1.5491361618041992 lr:9.9599734295004e-05 tokens_per_second_per_gpu:4261.333241303395 mfu:20.101435957650967 +Step 1541 | loss:1.6742091178894043 lr:9.959918653203531e-05 tokens_per_second_per_gpu:5308.243134345767 mfu:25.039888544378435 +Step 1542 | loss:1.250849723815918 lr:9.95986383960248e-05 tokens_per_second_per_gpu:3761.6993490472787 mfu:17.744577641546528 +Step 1543 | loss:1.4858883619308472 lr:9.959808988697661e-05 tokens_per_second_per_gpu:5053.744709355058 mfu:23.839376051788427 +Step 1544 | loss:1.4661533832550049 lr:9.959754100489484e-05 tokens_per_second_per_gpu:4491.646889065022 mfu:21.187864729702905 +Step 1545 | loss:1.3847779035568237 lr:9.959699174978366e-05 tokens_per_second_per_gpu:3189.2187453812894 mfu:15.044088958791116 +Step 1546 | loss:1.3752810955047607 lr:9.959644212164715e-05 tokens_per_second_per_gpu:5353.012039500264 mfu:25.251071108355056 +Step 1547 | loss:1.403080701828003 lr:9.95958921204895e-05 tokens_per_second_per_gpu:4148.904897404014 mfu:19.57109227722411 +Step 1548 | loss:1.4021797180175781 lr:9.95953417463148e-05 tokens_per_second_per_gpu:3754.286983531449 mfu:17.70961224872839 +Step 1549 | loss:1.344141960144043 lr:9.95947909991272e-05 tokens_per_second_per_gpu:4719.959258843508 mfu:22.264853132055126 +Step 1550 | loss:1.3745237588882446 lr:9.959423987893086e-05 tokens_per_second_per_gpu:5377.758851918244 mfu:25.367806044772248 +Step 1551 | loss:1.3500028848648071 lr:9.959368838572991e-05 tokens_per_second_per_gpu:4267.540631673325 mfu:20.130717277115934 +Step 1552 | loss:1.375867486000061 lr:9.959313651952849e-05 tokens_per_second_per_gpu:5355.846613376709 mfu:25.26444227695842 +Step 1553 | loss:1.443913221359253 lr:9.959258428033079e-05 tokens_per_second_per_gpu:3853.9329449231955 mfu:18.179659249967397 +Step 1554 | loss:1.5820348262786865 lr:9.95920316681409e-05 tokens_per_second_per_gpu:5331.014605262188 mfu:25.147305457904757 +Step 1555 | loss:1.5370088815689087 lr:9.959147868296303e-05 tokens_per_second_per_gpu:4221.04900245356 mfu:19.91140832979641 +Step 1556 | loss:1.4301806688308716 lr:9.959092532480132e-05 tokens_per_second_per_gpu:3730.382167985686 mfu:17.596849155216926 +Step 1557 | loss:1.3944381475448608 lr:9.959037159365994e-05 tokens_per_second_per_gpu:5347.181037554918 mfu:25.223565277307337 +Step 1558 | loss:1.5376635789871216 lr:9.958981748954303e-05 tokens_per_second_per_gpu:4597.95267026782 mfu:21.68932723727307 +Step 1559 | loss:1.4022220373153687 lr:9.958926301245479e-05 tokens_per_second_per_gpu:5329.530629158737 mfu:25.140305289432348 +Step 1560 | loss:1.3552228212356567 lr:9.958870816239938e-05 tokens_per_second_per_gpu:5358.66179267212 mfu:25.27772195801399 +Step 1561 | loss:1.241715908050537 lr:9.958815293938095e-05 tokens_per_second_per_gpu:4292.033027615432 mfu:20.24625208760851 +Step 1562 | loss:1.4272196292877197 lr:9.958759734340371e-05 tokens_per_second_per_gpu:5321.898699171851 mfu:25.104304173542676 +Step 1563 | loss:1.5049777030944824 lr:9.958704137447182e-05 tokens_per_second_per_gpu:5345.012203808375 mfu:25.213334518483702 +Step 1564 | loss:1.34462308883667 lr:9.958648503258946e-05 tokens_per_second_per_gpu:5341.1928358403 mfu:25.19531790812692 +Step 1565 | loss:1.3384398221969604 lr:9.958592831776083e-05 tokens_per_second_per_gpu:3968.7696991690086 mfu:18.721363812915556 +Step 1566 | loss:1.4094730615615845 lr:9.958537122999008e-05 tokens_per_second_per_gpu:5321.018142236596 mfu:25.100150436240575 +Step 1567 | loss:1.4486937522888184 lr:9.958481376928145e-05 tokens_per_second_per_gpu:5335.9140302492615 mfu:25.170416881496887 +Step 1568 | loss:1.3483216762542725 lr:9.95842559356391e-05 tokens_per_second_per_gpu:5327.093594324664 mfu:25.128809380319005 +Step 1569 | loss:1.390634536743164 lr:9.958369772906724e-05 tokens_per_second_per_gpu:4771.702588233376 mfu:22.508935245109477 +Step 1570 | loss:1.5687521696090698 lr:9.958313914957005e-05 tokens_per_second_per_gpu:3205.8821781192337 mfu:15.122693214090793 +Step 1571 | loss:1.5345847606658936 lr:9.958258019715175e-05 tokens_per_second_per_gpu:4201.866589068551 mfu:19.82092161300211 +Step 1572 | loss:1.3414803743362427 lr:9.958202087181653e-05 tokens_per_second_per_gpu:4589.666084050343 mfu:21.65023799624783 +Step 1573 | loss:1.4746829271316528 lr:9.958146117356862e-05 tokens_per_second_per_gpu:4369.173688292152 mfu:20.610137745574487 +Step 1574 | loss:1.416303277015686 lr:9.95809011024122e-05 tokens_per_second_per_gpu:5381.339767212085 mfu:25.384697833180446 +Step 1575 | loss:1.3962842226028442 lr:9.958034065835151e-05 tokens_per_second_per_gpu:4589.202724764977 mfu:21.64805225144151 +Step 1576 | loss:1.47483491897583 lr:9.957977984139075e-05 tokens_per_second_per_gpu:5334.375516047805 mfu:25.16315944751102 +Step 1577 | loss:1.479188084602356 lr:9.957921865153414e-05 tokens_per_second_per_gpu:4014.1022343361165 mfu:18.935205115826864 +Step 1578 | loss:1.3240455389022827 lr:9.957865708878589e-05 tokens_per_second_per_gpu:4646.841017850153 mfu:21.919941914031146 +Step 1579 | loss:1.3741496801376343 lr:9.957809515315024e-05 tokens_per_second_per_gpu:5337.0343860891335 mfu:25.17570179114598 +Step 1580 | loss:1.4470402002334595 lr:9.95775328446314e-05 tokens_per_second_per_gpu:5323.811311005805 mfu:25.11332628988867 +Step 1581 | loss:1.5174241065979004 lr:9.957697016323361e-05 tokens_per_second_per_gpu:5315.873129110483 mfu:25.075880531494946 +Step 1582 | loss:1.4035842418670654 lr:9.957640710896112e-05 tokens_per_second_per_gpu:5337.933386259714 mfu:25.17994252833585 +Step 1583 | loss:1.5058467388153076 lr:9.957584368181814e-05 tokens_per_second_per_gpu:3783.2867514517357 mfu:17.846409101879 +Step 1584 | loss:1.405059814453125 lr:9.95752798818089e-05 tokens_per_second_per_gpu:5096.456308384792 mfu:24.04085395176192 +Step 1585 | loss:1.5329452753067017 lr:9.957471570893766e-05 tokens_per_second_per_gpu:5348.772026056634 mfu:25.231070241520563 +Step 1586 | loss:1.4729658365249634 lr:9.957415116320867e-05 tokens_per_second_per_gpu:3656.4509220506493 mfu:17.24810285417009 +Step 1587 | loss:1.371480941772461 lr:9.957358624462614e-05 tokens_per_second_per_gpu:4226.5773958889395 mfu:19.937486704872278 +Step 1588 | loss:1.5614697933197021 lr:9.957302095319436e-05 tokens_per_second_per_gpu:5332.199099735477 mfu:25.15289292043086 +Step 1589 | loss:1.4468351602554321 lr:9.957245528891757e-05 tokens_per_second_per_gpu:5187.223284283981 mfu:24.4690172635214 +Step 1590 | loss:1.4276005029678345 lr:9.95718892518e-05 tokens_per_second_per_gpu:3742.0937901147504 mfu:17.652094875008505 +Step 1591 | loss:1.249566674232483 lr:9.957132284184593e-05 tokens_per_second_per_gpu:4487.475033496467 mfu:21.16818537519407 +Step 1592 | loss:1.377884030342102 lr:9.957075605905962e-05 tokens_per_second_per_gpu:5328.8857427734365 mfu:25.137263250323922 +Step 1593 | loss:1.3919222354888916 lr:9.957018890344533e-05 tokens_per_second_per_gpu:5367.5649300352225 mfu:25.319719575987616 +Step 1594 | loss:1.3368867635726929 lr:9.956962137500731e-05 tokens_per_second_per_gpu:5166.083903903757 mfu:24.369299199517673 +Step 1595 | loss:1.3741294145584106 lr:9.956905347374986e-05 tokens_per_second_per_gpu:5340.795192690854 mfu:25.193442157546063 +Step 1596 | loss:1.2662321329116821 lr:9.956848519967724e-05 tokens_per_second_per_gpu:5329.33785753909 mfu:25.139395952812244 +Step 1597 | loss:1.3459630012512207 lr:9.95679165527937e-05 tokens_per_second_per_gpu:5075.192009989738 mfu:23.940546628169034 +Step 1598 | loss:1.3609950542449951 lr:9.956734753310354e-05 tokens_per_second_per_gpu:5005.527521257856 mfu:23.611927348833326 +Step 1599 | loss:1.382182002067566 lr:9.956677814061104e-05 tokens_per_second_per_gpu:4450.349250149367 mfu:20.99305672083399 +Step 1600 | loss:1.4116263389587402 lr:9.956620837532047e-05 tokens_per_second_per_gpu:5020.084311333754 mfu:23.680594211265827 +Step 1601 | loss:1.4125159978866577 lr:9.956563823723612e-05 tokens_per_second_per_gpu:4120.312868338496 mfu:19.436218797818867 +Step 1602 | loss:1.4068549871444702 lr:9.956506772636228e-05 tokens_per_second_per_gpu:4950.618592832106 mfu:23.35291256501983 +Step 1603 | loss:1.3532812595367432 lr:9.956449684270325e-05 tokens_per_second_per_gpu:5372.386054925648 mfu:25.342461644664482 +Step 1604 | loss:1.5931782722473145 lr:9.956392558626331e-05 tokens_per_second_per_gpu:4847.825403449463 mfu:22.8680195523753 +Step 1605 | loss:1.3351620435714722 lr:9.956335395704676e-05 tokens_per_second_per_gpu:4994.101702585602 mfu:23.55802981271062 +Step 1606 | loss:1.399248480796814 lr:9.956278195505791e-05 tokens_per_second_per_gpu:3703.8109842104973 mfu:17.471508347837926 +Step 1607 | loss:1.3221715688705444 lr:9.956220958030102e-05 tokens_per_second_per_gpu:4900.23040527096 mfu:23.1152228871826 +Step 1608 | loss:1.5287551879882812 lr:9.956163683278046e-05 tokens_per_second_per_gpu:4880.132643798347 mfu:23.02041831728459 +Step 1609 | loss:1.4284496307373047 lr:9.956106371250049e-05 tokens_per_second_per_gpu:3770.3892715031475 mfu:17.785569488424517 +Step 1610 | loss:1.3907382488250732 lr:9.956049021946545e-05 tokens_per_second_per_gpu:5337.059391962092 mfu:25.175819748115167 +Step 1611 | loss:1.37810480594635 lr:9.955991635367962e-05 tokens_per_second_per_gpu:4924.061374696565 mfu:23.227637635945833 +Step 1612 | loss:1.3908029794692993 lr:9.955934211514736e-05 tokens_per_second_per_gpu:5077.353318690202 mfu:23.950741890066414 +Step 1613 | loss:1.400022268295288 lr:9.955876750387295e-05 tokens_per_second_per_gpu:5267.23162026763 mfu:24.846430235186023 +Step 1614 | loss:1.3632227182388306 lr:9.955819251986073e-05 tokens_per_second_per_gpu:4900.633969778403 mfu:23.11712656981942 +Step 1615 | loss:1.4510315656661987 lr:9.955761716311502e-05 tokens_per_second_per_gpu:5365.212462957503 mfu:25.308622587409744 +Step 1616 | loss:1.4654572010040283 lr:9.955704143364014e-05 tokens_per_second_per_gpu:4974.946942452271 mfu:23.467673541022616 +Step 1617 | loss:1.3686922788619995 lr:9.955646533144043e-05 tokens_per_second_per_gpu:3843.4717184352603 mfu:18.130311859754517 +Step 1618 | loss:1.401374340057373 lr:9.955588885652023e-05 tokens_per_second_per_gpu:5360.499271943909 mfu:25.28638966124531 +Step 1619 | loss:1.3400697708129883 lr:9.955531200888386e-05 tokens_per_second_per_gpu:4766.2248607106685 mfu:22.483095869788173 +Step 1620 | loss:1.3211501836776733 lr:9.955473478853567e-05 tokens_per_second_per_gpu:4271.562122244233 mfu:20.149687334275956 +Step 1621 | loss:1.2894630432128906 lr:9.955415719547998e-05 tokens_per_second_per_gpu:3518.0468182371123 mfu:16.595227082306508 +Step 1622 | loss:1.2424366474151611 lr:9.955357922972117e-05 tokens_per_second_per_gpu:5375.12194000362 mfu:25.3553672813674 +Step 1623 | loss:1.4227714538574219 lr:9.955300089126356e-05 tokens_per_second_per_gpu:3939.590342244139 mfu:18.583719807789286 +Step 1624 | loss:1.231078863143921 lr:9.95524221801115e-05 tokens_per_second_per_gpu:5340.318590990419 mfu:25.19119394600848 +Step 1625 | loss:1.5239819288253784 lr:9.955184309626936e-05 tokens_per_second_per_gpu:4355.982750705665 mfu:20.547913842373003 +Step 1626 | loss:1.3628504276275635 lr:9.955126363974148e-05 tokens_per_second_per_gpu:5316.859221800339 mfu:25.08053209895775 +Step 1627 | loss:1.4405789375305176 lr:9.955068381053223e-05 tokens_per_second_per_gpu:5345.185863646873 mfu:25.21415370157132 +Step 1628 | loss:1.3616505861282349 lr:9.955010360864596e-05 tokens_per_second_per_gpu:3252.311558311026 mfu:15.341708522124595 +Step 1629 | loss:1.468954086303711 lr:9.954952303408703e-05 tokens_per_second_per_gpu:4186.124056650042 mfu:19.746661401631005 +Step 1630 | loss:1.3303477764129639 lr:9.954894208685983e-05 tokens_per_second_per_gpu:5159.810468021096 mfu:24.33970633210063 +Step 1631 | loss:1.510000228881836 lr:9.954836076696871e-05 tokens_per_second_per_gpu:4888.475837342198 mfu:23.05977458472654 +Step 1632 | loss:1.4529544115066528 lr:9.954777907441803e-05 tokens_per_second_per_gpu:3148.68739934493 mfu:14.852895684176886 +Step 1633 | loss:1.3053562641143799 lr:9.95471970092122e-05 tokens_per_second_per_gpu:4314.19692585603 mfu:20.35080297716039 +Step 1634 | loss:1.379218339920044 lr:9.954661457135557e-05 tokens_per_second_per_gpu:3484.0464944770874 mfu:16.434841754077052 +Step 1635 | loss:1.499563455581665 lr:9.954603176085253e-05 tokens_per_second_per_gpu:4812.432171944957 mfu:22.70106364066056 +Step 1636 | loss:1.3664923906326294 lr:9.954544857770747e-05 tokens_per_second_per_gpu:4887.545776773677 mfu:23.055387330340128 +Step 1637 | loss:1.353930115699768 lr:9.954486502192477e-05 tokens_per_second_per_gpu:3569.4247550901882 mfu:16.837585576422413 +Step 1638 | loss:1.3397656679153442 lr:9.954428109350882e-05 tokens_per_second_per_gpu:5224.242909232113 mfu:24.64364515060106 +Step 1639 | loss:1.2078784704208374 lr:9.9543696792464e-05 tokens_per_second_per_gpu:3763.789316319975 mfu:17.754436373757024 +Step 1640 | loss:1.4573323726654053 lr:9.954311211879471e-05 tokens_per_second_per_gpu:3518.304401547693 mfu:16.596442146730762 +Step 1641 | loss:1.4003043174743652 lr:9.954252707250537e-05 tokens_per_second_per_gpu:3502.8616236969287 mfu:16.52359592879911 +Step 1642 | loss:1.5228931903839111 lr:9.954194165360035e-05 tokens_per_second_per_gpu:5153.086324115522 mfu:24.307987398040783 +Step 1643 | loss:1.5184613466262817 lr:9.954135586208406e-05 tokens_per_second_per_gpu:3895.842032358735 mfu:18.377351565828484 +Step 1644 | loss:1.3437695503234863 lr:9.954076969796092e-05 tokens_per_second_per_gpu:5351.94064131708 mfu:25.2460171403256 +Step 1645 | loss:1.314003348350525 lr:9.954018316123532e-05 tokens_per_second_per_gpu:4692.650466282515 mfu:22.13603289818445 +Step 1646 | loss:1.3924907445907593 lr:9.953959625191169e-05 tokens_per_second_per_gpu:5336.1597586797625 mfu:25.171576024429154 +Step 1647 | loss:1.504608392715454 lr:9.953900896999444e-05 tokens_per_second_per_gpu:3559.4935263481575 mfu:16.790738276003903 +Step 1648 | loss:1.3738462924957275 lr:9.953842131548799e-05 tokens_per_second_per_gpu:4598.056282551085 mfu:21.68981599409131 +Step 1649 | loss:1.3129626512527466 lr:9.953783328839673e-05 tokens_per_second_per_gpu:5335.118105683481 mfu:25.16666236952136 +Step 1650 | loss:1.766638994216919 lr:9.953724488872511e-05 tokens_per_second_per_gpu:4405.005598747641 mfu:20.779162980747362 +Step 1651 | loss:1.4223405122756958 lr:9.953665611647756e-05 tokens_per_second_per_gpu:4227.267208432586 mfu:19.940740668334815 +Step 1652 | loss:1.320273518562317 lr:9.953606697165849e-05 tokens_per_second_per_gpu:5306.1619970534 mfu:25.0300714646917 +Step 1653 | loss:1.4749246835708618 lr:9.953547745427234e-05 tokens_per_second_per_gpu:3934.322697325362 mfu:18.558871427954514 +Step 1654 | loss:1.2620704174041748 lr:9.953488756432354e-05 tokens_per_second_per_gpu:4838.791638129905 mfu:22.825405740043642 +Step 1655 | loss:1.319746971130371 lr:9.953429730181653e-05 tokens_per_second_per_gpu:3520.9607823145275 mfu:16.608972748033214 +Step 1656 | loss:1.4232637882232666 lr:9.953370666675576e-05 tokens_per_second_per_gpu:4019.342382885126 mfu:18.959923790594075 +Step 1657 | loss:1.449915885925293 lr:9.953311565914567e-05 tokens_per_second_per_gpu:3632.699092134034 mfu:17.136061419973352 +Step 1658 | loss:1.2463586330413818 lr:9.953252427899067e-05 tokens_per_second_per_gpu:4548.283796117622 mfu:21.455030683522804 +Step 1659 | loss:1.3151205778121948 lr:9.953193252629524e-05 tokens_per_second_per_gpu:5381.975846494233 mfu:25.38769832768066 +Step 1660 | loss:1.3287529945373535 lr:9.953134040106384e-05 tokens_per_second_per_gpu:5339.756430374275 mfu:25.188542138467636 +Step 1661 | loss:1.5778234004974365 lr:9.95307479033009e-05 tokens_per_second_per_gpu:4561.439321577965 mfu:21.51708754168333 +Step 1662 | loss:1.4477225542068481 lr:9.953015503301089e-05 tokens_per_second_per_gpu:4763.654903039847 mfu:22.47097294097492 +Step 1663 | loss:1.4363065958023071 lr:9.952956179019825e-05 tokens_per_second_per_gpu:4082.7765232959878 mfu:19.259153453892495 +Step 1664 | loss:1.5229302644729614 lr:9.952896817486748e-05 tokens_per_second_per_gpu:4902.692278327623 mfu:23.126835962429737 +Step 1665 | loss:1.4193960428237915 lr:9.952837418702301e-05 tokens_per_second_per_gpu:5110.309853113739 mfu:24.106203485906835 +Step 1666 | loss:1.2896173000335693 lr:9.952777982666931e-05 tokens_per_second_per_gpu:5327.958958887537 mfu:25.132891452608497 +Step 1667 | loss:1.4012244939804077 lr:9.952718509381085e-05 tokens_per_second_per_gpu:5343.8874899571165 mfu:25.208029051351346 +Step 1668 | loss:1.3687210083007812 lr:9.952658998845213e-05 tokens_per_second_per_gpu:5208.179414079171 mfu:24.567870903249744 +Step 1669 | loss:1.7307263612747192 lr:9.95259945105976e-05 tokens_per_second_per_gpu:3219.872169381607 mfu:15.18868638981405 +Step 1670 | loss:1.37649405002594 lr:9.952539866025175e-05 tokens_per_second_per_gpu:4373.316783448652 mfu:20.629681432312623 +Step 1671 | loss:1.4221715927124023 lr:9.952480243741905e-05 tokens_per_second_per_gpu:4560.22372932241 mfu:21.511353385612455 +Step 1672 | loss:1.3045741319656372 lr:9.9524205842104e-05 tokens_per_second_per_gpu:5320.057307943985 mfu:25.095618016947093 +Step 1673 | loss:1.2280097007751465 lr:9.952360887431107e-05 tokens_per_second_per_gpu:5323.095234600131 mfu:25.10994843530787 +Step 1674 | loss:1.457716941833496 lr:9.952301153404476e-05 tokens_per_second_per_gpu:3164.415583313774 mfu:14.927088211462822 +Step 1675 | loss:1.4503376483917236 lr:9.952241382130956e-05 tokens_per_second_per_gpu:4531.62957647903 mfu:21.376469887984687 +Step 1676 | loss:1.294408917427063 lr:9.952181573610997e-05 tokens_per_second_per_gpu:5334.076351440999 mfu:25.161748237017974 +Step 1677 | loss:1.5210968255996704 lr:9.952121727845048e-05 tokens_per_second_per_gpu:5357.090134596614 mfu:25.270308178719308 +Step 1678 | loss:1.4403314590454102 lr:9.95206184483356e-05 tokens_per_second_per_gpu:4845.9294894966215 mfu:22.85907620278347 +Step 1679 | loss:1.4392839670181274 lr:9.952001924576984e-05 tokens_per_second_per_gpu:4200.272979633852 mfu:19.81340428540083 +Step 1680 | loss:1.4337328672409058 lr:9.951941967075768e-05 tokens_per_second_per_gpu:4128.438796483787 mfu:19.474550187306512 +Step 1681 | loss:1.2746442556381226 lr:9.951881972330366e-05 tokens_per_second_per_gpu:5193.738137833316 mfu:24.499748939262243 +Step 1682 | loss:1.4538713693618774 lr:9.951821940341226e-05 tokens_per_second_per_gpu:3910.1461217252263 mfu:18.444826395899526 +Step 1683 | loss:1.5002710819244385 lr:9.951761871108802e-05 tokens_per_second_per_gpu:5360.273370153813 mfu:25.28532404395861 +Step 1684 | loss:1.322972059249878 lr:9.951701764633546e-05 tokens_per_second_per_gpu:4292.47576377732 mfu:20.248340549622714 +Step 1685 | loss:1.4107646942138672 lr:9.95164162091591e-05 tokens_per_second_per_gpu:5330.698803835595 mfu:25.145815768694238 +Step 1686 | loss:1.5004481077194214 lr:9.951581439956344e-05 tokens_per_second_per_gpu:5373.585228783775 mfu:25.348118352353424 +Step 1687 | loss:1.2269165515899658 lr:9.951521221755302e-05 tokens_per_second_per_gpu:4826.715049243936 mfu:22.768438409789997 +Step 1688 | loss:1.3252965211868286 lr:9.951460966313238e-05 tokens_per_second_per_gpu:4349.530570865328 mfu:20.51747780921073 +Step 1689 | loss:1.4206602573394775 lr:9.951400673630602e-05 tokens_per_second_per_gpu:5325.981254967012 mfu:25.1235622857843 +Step 1690 | loss:1.7123503684997559 lr:9.951340343707852e-05 tokens_per_second_per_gpu:2905.752383815902 mfu:13.706929766938726 +Step 1691 | loss:1.2707922458648682 lr:9.951279976545439e-05 tokens_per_second_per_gpu:5141.322950270725 mfu:24.25249755657634 +Step 1692 | loss:1.2703797817230225 lr:9.951219572143817e-05 tokens_per_second_per_gpu:4376.659944735874 mfu:20.64545169450658 +Step 1693 | loss:1.5110974311828613 lr:9.951159130503443e-05 tokens_per_second_per_gpu:5361.408504045674 mfu:25.29067866420033 +Step 1694 | loss:1.4567757844924927 lr:9.951098651624767e-05 tokens_per_second_per_gpu:3864.7703346018284 mfu:18.230781066130856 +Step 1695 | loss:1.3495887517929077 lr:9.951038135508246e-05 tokens_per_second_per_gpu:3948.0993592626305 mfu:18.623858292853516 +Step 1696 | loss:1.4576247930526733 lr:9.950977582154337e-05 tokens_per_second_per_gpu:3864.45806697463 mfu:18.22930804645501 +Step 1697 | loss:1.474303960800171 lr:9.950916991563492e-05 tokens_per_second_per_gpu:5362.480890070357 mfu:25.295737292046685 +Step 1698 | loss:1.4168109893798828 lr:9.95085636373617e-05 tokens_per_second_per_gpu:4663.873837936644 mfu:22.0002886324779 +Step 1699 | loss:1.4741606712341309 lr:9.950795698672826e-05 tokens_per_second_per_gpu:4431.848513203368 mfu:20.905785588108497 +Step 1700 | loss:1.2446670532226562 lr:9.950734996373915e-05 tokens_per_second_per_gpu:5346.240990774644 mfu:25.219130916256265 +Step 1701 | loss:1.3575273752212524 lr:9.950674256839895e-05 tokens_per_second_per_gpu:4324.530182849621 mfu:20.399546713434603 +Step 1702 | loss:1.3808155059814453 lr:9.950613480071221e-05 tokens_per_second_per_gpu:5013.156991503033 mfu:23.647916861701678 +Step 1703 | loss:1.6297377347946167 lr:9.950552666068352e-05 tokens_per_second_per_gpu:3529.3037904505145 mfu:16.648328140874636 +Step 1704 | loss:1.3524364233016968 lr:9.950491814831745e-05 tokens_per_second_per_gpu:4323.768163976023 mfu:20.395952140394016 +Step 1705 | loss:1.3152658939361572 lr:9.950430926361858e-05 tokens_per_second_per_gpu:3420.4976060511262 mfu:16.135070804813378 +Step 1706 | loss:1.3395966291427612 lr:9.950370000659146e-05 tokens_per_second_per_gpu:5346.986205291073 mfu:25.22264622027697 +Step 1707 | loss:1.5136873722076416 lr:9.950309037724071e-05 tokens_per_second_per_gpu:4323.396650381484 mfu:20.394199647381814 +Step 1708 | loss:1.3881949186325073 lr:9.95024803755709e-05 tokens_per_second_per_gpu:4576.941559595595 mfu:21.59021424336704 +Step 1709 | loss:1.3654218912124634 lr:9.950187000158662e-05 tokens_per_second_per_gpu:4076.442935718703 mfu:19.229276840668973 +Step 1710 | loss:1.5723682641983032 lr:9.950125925529245e-05 tokens_per_second_per_gpu:4015.564648088118 mfu:18.942103571008566 +Step 1711 | loss:1.4970701932907104 lr:9.950064813669302e-05 tokens_per_second_per_gpu:3701.305857246903 mfu:17.45969123653192 +Step 1712 | loss:1.4064291715621948 lr:9.950003664579288e-05 tokens_per_second_per_gpu:5045.115742814705 mfu:23.79867174436408 +Step 1713 | loss:1.4077961444854736 lr:9.949942478259663e-05 tokens_per_second_per_gpu:4498.184022939264 mfu:21.21870150553775 +Step 1714 | loss:1.7810695171356201 lr:9.949881254710892e-05 tokens_per_second_per_gpu:3311.1398845773765 mfu:15.619211774270429 +Step 1715 | loss:1.3274297714233398 lr:9.949819993933431e-05 tokens_per_second_per_gpu:4406.800997297133 mfu:20.787632182031896 +Step 1716 | loss:1.3013062477111816 lr:9.949758695927742e-05 tokens_per_second_per_gpu:3981.1101356895033 mfu:18.779575757478312 +Step 1717 | loss:1.5052909851074219 lr:9.949697360694288e-05 tokens_per_second_per_gpu:3709.1334169802935 mfu:17.496615171313728 +Step 1718 | loss:1.5566514730453491 lr:9.949635988233527e-05 tokens_per_second_per_gpu:4028.1481025840744 mfu:19.00146187282472 +Step 1719 | loss:1.4308884143829346 lr:9.949574578545922e-05 tokens_per_second_per_gpu:5311.340503387811 mfu:25.054499362615402 +Step 1720 | loss:1.3777616024017334 lr:9.949513131631936e-05 tokens_per_second_per_gpu:5257.394136834607 mfu:24.800025147384698 +Step 1721 | loss:1.4309325218200684 lr:9.949451647492029e-05 tokens_per_second_per_gpu:4893.998138962792 mfu:23.085824223672695 +Step 1722 | loss:1.220924973487854 lr:9.949390126126666e-05 tokens_per_second_per_gpu:4650.251306414898 mfu:21.936028827046275 +Step 1723 | loss:1.4523354768753052 lr:9.949328567536308e-05 tokens_per_second_per_gpu:4297.665329174018 mfu:20.2728206150301 +Step 1724 | loss:1.4160248041152954 lr:9.949266971721418e-05 tokens_per_second_per_gpu:4135.440115377077 mfu:19.507576603074337 +Step 1725 | loss:1.4885377883911133 lr:9.949205338682459e-05 tokens_per_second_per_gpu:5313.870577280039 mfu:25.066434152840163 +Step 1726 | loss:1.2756117582321167 lr:9.949143668419896e-05 tokens_per_second_per_gpu:4365.654176891286 mfu:20.593535609806036 +Step 1727 | loss:1.5050525665283203 lr:9.949081960934191e-05 tokens_per_second_per_gpu:4431.535651490422 mfu:20.904309766028817 +Step 1728 | loss:1.3175904750823975 lr:9.949020216225811e-05 tokens_per_second_per_gpu:5344.495339758109 mfu:25.210896382572496 +Step 1729 | loss:1.452455997467041 lr:9.948958434295217e-05 tokens_per_second_per_gpu:5034.868313736735 mfu:23.750332873011605 +Step 1730 | loss:1.3929197788238525 lr:9.948896615142874e-05 tokens_per_second_per_gpu:4655.191234062008 mfu:21.959331308596266 +Step 1731 | loss:1.5084052085876465 lr:9.94883475876925e-05 tokens_per_second_per_gpu:5357.752623798052 mfu:25.273433253314806 +Step 1732 | loss:1.5247148275375366 lr:9.948772865174807e-05 tokens_per_second_per_gpu:4529.989400951183 mfu:21.368732900177033 +Step 1733 | loss:1.3473262786865234 lr:9.948710934360012e-05 tokens_per_second_per_gpu:4376.590859775183 mfu:20.645125808959897 +Step 1734 | loss:1.391586184501648 lr:9.94864896632533e-05 tokens_per_second_per_gpu:3742.1765465575854 mfu:17.652485251269063 +Step 1735 | loss:1.5596381425857544 lr:9.94858696107123e-05 tokens_per_second_per_gpu:5377.704920107531 mfu:25.367551639219396 +Step 1736 | loss:1.4839656352996826 lr:9.948524918598175e-05 tokens_per_second_per_gpu:5325.668017958192 mfu:25.12208469336995 +Step 1737 | loss:1.4036742448806763 lr:9.948462838906633e-05 tokens_per_second_per_gpu:5339.465601144232 mfu:25.18717024736894 +Step 1738 | loss:1.453696370124817 lr:9.948400721997069e-05 tokens_per_second_per_gpu:4486.141464870653 mfu:21.16189470445584 +Step 1739 | loss:1.5263432264328003 lr:9.948338567869953e-05 tokens_per_second_per_gpu:3361.0665398284796 mfu:15.854724325455654 +Step 1740 | loss:1.357632040977478 lr:9.948276376525751e-05 tokens_per_second_per_gpu:5309.77339648489 mfu:25.04710705197832 +Step 1741 | loss:1.3507702350616455 lr:9.948214147964932e-05 tokens_per_second_per_gpu:3949.627427955967 mfu:18.631066453595388 +Step 1742 | loss:1.3951681852340698 lr:9.94815188218796e-05 tokens_per_second_per_gpu:4412.759473100718 mfu:20.81573932901821 +Step 1743 | loss:1.3551275730133057 lr:9.948089579195308e-05 tokens_per_second_per_gpu:5346.797359289329 mfu:25.22175540146629 +Step 1744 | loss:1.2907341718673706 lr:9.948027238987444e-05 tokens_per_second_per_gpu:4936.642786555601 mfu:23.286986302295123 +Step 1745 | loss:1.2443264722824097 lr:9.947964861564835e-05 tokens_per_second_per_gpu:5347.499929685162 mfu:25.22506954589431 +Step 1746 | loss:1.4525933265686035 lr:9.947902446927951e-05 tokens_per_second_per_gpu:4869.08829652675 mfu:22.968320246844513 +Step 1747 | loss:1.5490626096725464 lr:9.94783999507726e-05 tokens_per_second_per_gpu:3650.103881258273 mfu:17.218162779835623 +Step 1748 | loss:1.2926608324050903 lr:9.947777506013234e-05 tokens_per_second_per_gpu:3913.9800382324406 mfu:18.462911634197667 +Step 1749 | loss:1.366976261138916 lr:9.947714979736342e-05 tokens_per_second_per_gpu:4288.273799417803 mfu:20.22851916680983 +Step 1750 | loss:1.294357180595398 lr:9.947652416247056e-05 tokens_per_second_per_gpu:5362.549965920535 mfu:25.29606313461753 +Step 1751 | loss:1.378319263458252 lr:9.947589815545842e-05 tokens_per_second_per_gpu:4939.135552294485 mfu:23.29874510359513 +Step 1752 | loss:1.3727320432662964 lr:9.947527177633175e-05 tokens_per_second_per_gpu:4086.9131333322625 mfu:19.27866654921205 +Step 1753 | loss:1.5116184949874878 lr:9.947464502509524e-05 tokens_per_second_per_gpu:3998.097039037492 mfu:18.859705878835346 +Step 1754 | loss:1.3426135778427124 lr:9.947401790175364e-05 tokens_per_second_per_gpu:4851.228218223311 mfu:22.88407121024361 +Step 1755 | loss:1.3031085729599 lr:9.947339040631161e-05 tokens_per_second_per_gpu:5338.266508342979 mfu:25.1815139220391 +Step 1756 | loss:1.1928287744522095 lr:9.947276253877392e-05 tokens_per_second_per_gpu:5347.819386572049 mfu:25.226576478534717 +Step 1757 | loss:1.3768925666809082 lr:9.947213429914524e-05 tokens_per_second_per_gpu:4050.635471228202 mfu:19.10753862745979 +Step 1758 | loss:1.2942554950714111 lr:9.947150568743035e-05 tokens_per_second_per_gpu:4056.7699328842973 mfu:19.136475929738676 +Step 1759 | loss:1.3780137300491333 lr:9.947087670363395e-05 tokens_per_second_per_gpu:5341.995620562186 mfu:25.19910478062925 +Step 1760 | loss:1.3909424543380737 lr:9.947024734776076e-05 tokens_per_second_per_gpu:4159.373269733304 mfu:19.620473375589867 +Step 1761 | loss:1.3076164722442627 lr:9.946961761981554e-05 tokens_per_second_per_gpu:4973.812408414372 mfu:23.462321750394377 +Step 1762 | loss:1.4822801351547241 lr:9.946898751980301e-05 tokens_per_second_per_gpu:5374.6838303314335 mfu:25.353300643294478 +Step 1763 | loss:1.357478141784668 lr:9.94683570477279e-05 tokens_per_second_per_gpu:5345.911260532126 mfu:25.21757552244453 +Step 1764 | loss:1.4232432842254639 lr:9.946772620359498e-05 tokens_per_second_per_gpu:4264.5931603074905 mfu:20.116813551792823 +Step 1765 | loss:1.4618000984191895 lr:9.946709498740899e-05 tokens_per_second_per_gpu:4519.855124101776 mfu:21.32092778719245 +Step 1766 | loss:1.3071393966674805 lr:9.946646339917464e-05 tokens_per_second_per_gpu:4861.792086651189 mfu:22.933902779999716 +Step 1767 | loss:1.3654367923736572 lr:9.946583143889672e-05 tokens_per_second_per_gpu:5342.589057316703 mfu:25.201904123051058 +Step 1768 | loss:1.4936330318450928 lr:9.946519910657996e-05 tokens_per_second_per_gpu:3778.4491680572005 mfu:17.82358939563014 +Step 1769 | loss:1.408122181892395 lr:9.946456640222914e-05 tokens_per_second_per_gpu:4724.233910003446 mfu:22.285017390907246 +Step 1770 | loss:1.3976686000823975 lr:9.9463933325849e-05 tokens_per_second_per_gpu:4816.267592312983 mfu:22.71915597292673 +Step 1771 | loss:1.3937370777130127 lr:9.94632998774443e-05 tokens_per_second_per_gpu:5345.796263600747 mfu:25.217033062298714 +Step 1772 | loss:1.5122425556182861 lr:9.946266605701982e-05 tokens_per_second_per_gpu:5000.6990866128135 mfu:23.589150798796762 +Step 1773 | loss:1.2666865587234497 lr:9.94620318645803e-05 tokens_per_second_per_gpu:5036.581216901479 mfu:23.758412929491115 +Step 1774 | loss:1.2259489297866821 lr:9.946139730013054e-05 tokens_per_second_per_gpu:4389.078830439858 mfu:20.704033697252672 +Step 1775 | loss:1.5008325576782227 lr:9.94607623636753e-05 tokens_per_second_per_gpu:5162.073260834687 mfu:24.350380311874602 +Step 1776 | loss:1.5634104013442993 lr:9.946012705521936e-05 tokens_per_second_per_gpu:4816.317750126716 mfu:22.71939257589192 +Step 1777 | loss:1.4259201288223267 lr:9.945949137476749e-05 tokens_per_second_per_gpu:5147.65587322813 mfu:24.282371034674494 +Step 1778 | loss:1.5350364446640015 lr:9.945885532232446e-05 tokens_per_second_per_gpu:3718.5111473144457 mfu:17.540851525305705 +Step 1779 | loss:1.2571595907211304 lr:9.945821889789509e-05 tokens_per_second_per_gpu:5364.972631370725 mfu:25.30749126089542 +Step 1780 | loss:1.4997833967208862 lr:9.945758210148413e-05 tokens_per_second_per_gpu:5154.90893945555 mfu:24.316584985569943 +Step 1781 | loss:1.4126921892166138 lr:9.945694493309639e-05 tokens_per_second_per_gpu:5340.737835188894 mfu:25.19317159242312 +Step 1782 | loss:1.2957757711410522 lr:9.945630739273665e-05 tokens_per_second_per_gpu:5370.020589548655 mfu:25.331303340891196 +Step 1783 | loss:1.5955123901367188 lr:9.945566948040972e-05 tokens_per_second_per_gpu:5326.461534801044 mfu:25.125827847705615 +Step 1784 | loss:1.4616800546646118 lr:9.945503119612038e-05 tokens_per_second_per_gpu:4642.389958665513 mfu:21.8989455084282 +Step 1785 | loss:1.2888816595077515 lr:9.945439253987344e-05 tokens_per_second_per_gpu:4999.374610399442 mfu:23.582903018519342 +Step 1786 | loss:1.4825024604797363 lr:9.945375351167372e-05 tokens_per_second_per_gpu:4159.521571091297 mfu:19.621172938398338 +Step 1787 | loss:1.3645929098129272 lr:9.945311411152599e-05 tokens_per_second_per_gpu:5328.242924780717 mfu:25.1342309681762 +Step 1788 | loss:1.4819245338439941 lr:9.945247433943509e-05 tokens_per_second_per_gpu:4456.319788687548 mfu:21.021220769797505 +Step 1789 | loss:1.3251545429229736 lr:9.945183419540581e-05 tokens_per_second_per_gpu:4536.584826508017 mfu:21.399844647824256 +Step 1790 | loss:1.5220288038253784 lr:9.945119367944297e-05 tokens_per_second_per_gpu:5112.432602588832 mfu:24.11621685735925 +Step 1791 | loss:1.5478886365890503 lr:9.945055279155142e-05 tokens_per_second_per_gpu:5306.037308616031 mfu:25.02948328805859 +Step 1792 | loss:1.2349693775177002 lr:9.944991153173593e-05 tokens_per_second_per_gpu:5043.148280845816 mfu:23.789390890573202 +Step 1793 | loss:1.2881474494934082 lr:9.944926990000136e-05 tokens_per_second_per_gpu:5370.467353124173 mfu:25.33341080090264 +Step 1794 | loss:1.4590469598770142 lr:9.944862789635251e-05 tokens_per_second_per_gpu:4579.064006779309 mfu:21.600226188859455 +Step 1795 | loss:1.4921066761016846 lr:9.944798552079422e-05 tokens_per_second_per_gpu:4268.782753817244 mfu:20.136576579196696 +Step 1796 | loss:1.2970420122146606 lr:9.944734277333132e-05 tokens_per_second_per_gpu:4965.700968468741 mfu:23.424058704216474 +Step 1797 | loss:1.394189715385437 lr:9.944669965396864e-05 tokens_per_second_per_gpu:4800.926549074347 mfu:22.64678965451821 +Step 1798 | loss:1.3344017267227173 lr:9.944605616271105e-05 tokens_per_second_per_gpu:4056.1740169555387 mfu:19.133664892628982 +Step 1799 | loss:1.332026481628418 lr:9.944541229956334e-05 tokens_per_second_per_gpu:5391.41611506919 mfu:25.432229685224634 +Step 1800 | loss:1.4387491941452026 lr:9.944476806453038e-05 tokens_per_second_per_gpu:4535.261194512041 mfu:21.393600849864328 +Step 1801 | loss:1.283638596534729 lr:9.9444123457617e-05 tokens_per_second_per_gpu:3981.0874294225096 mfu:18.779468648142814 +Step 1802 | loss:1.5607843399047852 lr:9.944347847882807e-05 tokens_per_second_per_gpu:3314.951477583969 mfu:15.63719170880735 +Step 1803 | loss:1.2575318813323975 lr:9.944283312816844e-05 tokens_per_second_per_gpu:5083.18891138226 mfu:23.978269376449635 +Step 1804 | loss:1.3537366390228271 lr:9.944218740564293e-05 tokens_per_second_per_gpu:4688.760273168018 mfu:22.117682193527983 +Step 1805 | loss:1.3838963508605957 lr:9.944154131125642e-05 tokens_per_second_per_gpu:5348.360249601235 mfu:25.22912782172414 +Step 1806 | loss:1.291991114616394 lr:9.94408948450138e-05 tokens_per_second_per_gpu:5304.6662391595055 mfu:25.023015719503586 +Step 1807 | loss:1.437062382698059 lr:9.944024800691986e-05 tokens_per_second_per_gpu:4014.4352954934066 mfu:18.936776222131893 +Step 1808 | loss:1.336028814315796 lr:9.943960079697954e-05 tokens_per_second_per_gpu:3453.046442223681 mfu:16.28860921844342 +Step 1809 | loss:1.3552488088607788 lr:9.943895321519764e-05 tokens_per_second_per_gpu:5364.0521493867345 mfu:25.30314919032652 +Step 1810 | loss:1.4322197437286377 lr:9.94383052615791e-05 tokens_per_second_per_gpu:3867.140554224628 mfu:18.241961796492767 +Step 1811 | loss:1.374933123588562 lr:9.943765693612874e-05 tokens_per_second_per_gpu:5408.054685119764 mfu:25.510716658985974 +Step 1812 | loss:1.3649535179138184 lr:9.943700823885146e-05 tokens_per_second_per_gpu:4179.1478376044515 mfu:19.71375338612725 +Step 1813 | loss:1.3576449155807495 lr:9.943635916975214e-05 tokens_per_second_per_gpu:3420.796563829161 mfu:16.136481039659827 +Step 1814 | loss:1.4059630632400513 lr:9.943570972883562e-05 tokens_per_second_per_gpu:5330.248749836131 mfu:25.143692787191636 +Step 1815 | loss:1.4474461078643799 lr:9.943505991610685e-05 tokens_per_second_per_gpu:3425.4116940218732 mfu:16.158251396201255 +Step 1816 | loss:1.5609440803527832 lr:9.943440973157068e-05 tokens_per_second_per_gpu:4991.624603030214 mfu:23.546344911471085 +Step 1817 | loss:1.3266475200653076 lr:9.9433759175232e-05 tokens_per_second_per_gpu:3875.615928393375 mfu:18.281941582494692 +Step 1818 | loss:1.4144501686096191 lr:9.943310824709573e-05 tokens_per_second_per_gpu:4007.2684061489385 mfu:18.90296878229662 +Step 1819 | loss:1.3859338760375977 lr:9.943245694716672e-05 tokens_per_second_per_gpu:4382.3831283856825 mfu:20.672448928258554 +Step 1820 | loss:1.3514312505722046 lr:9.943180527544991e-05 tokens_per_second_per_gpu:5372.135768510468 mfu:25.34128100094081 +Step 1821 | loss:1.5011911392211914 lr:9.943115323195018e-05 tokens_per_second_per_gpu:5316.405476656887 mfu:25.078391705699186 +Step 1822 | loss:1.4533799886703491 lr:9.943050081667244e-05 tokens_per_second_per_gpu:3865.5937690351966 mfu:18.234665346845457 +Step 1823 | loss:1.402681827545166 lr:9.942984802962158e-05 tokens_per_second_per_gpu:5340.145256994689 mfu:25.190376299976787 +Step 1824 | loss:1.4763344526290894 lr:9.942919487080255e-05 tokens_per_second_per_gpu:3510.4078604922106 mfu:16.559192815283218 +Step 1825 | loss:1.4251673221588135 lr:9.942854134022023e-05 tokens_per_second_per_gpu:5346.283417300331 mfu:25.219331049416642 +Step 1826 | loss:1.4699469804763794 lr:9.942788743787954e-05 tokens_per_second_per_gpu:4147.432849144757 mfu:19.564148374427862 +Step 1827 | loss:1.447708249092102 lr:9.94272331637854e-05 tokens_per_second_per_gpu:4775.14438066705 mfu:22.52517076724946 +Step 1828 | loss:1.3390601873397827 lr:9.942657851794273e-05 tokens_per_second_per_gpu:4957.9130613475545 mfu:23.387321817571866 +Step 1829 | loss:1.4916903972625732 lr:9.942592350035647e-05 tokens_per_second_per_gpu:5303.774974500122 mfu:25.018811472039843 +Step 1830 | loss:1.2961816787719727 lr:9.942526811103152e-05 tokens_per_second_per_gpu:3974.6679575176586 mfu:18.749186903892785 +Step 1831 | loss:1.424452781677246 lr:9.942461234997283e-05 tokens_per_second_per_gpu:3527.4773896456472 mfu:16.639712696661967 +Step 1832 | loss:1.3697837591171265 lr:9.942395621718532e-05 tokens_per_second_per_gpu:5342.205604580301 mfu:25.20009531106967 +Step 1833 | loss:1.5029215812683105 lr:9.942329971267392e-05 tokens_per_second_per_gpu:4524.176139827666 mfu:21.341310755612778 +Step 1834 | loss:1.1838513612747192 lr:9.942264283644358e-05 tokens_per_second_per_gpu:5346.6261767682145 mfu:25.220947904307447 +Step 1835 | loss:1.3883368968963623 lr:9.942198558849925e-05 tokens_per_second_per_gpu:4403.028410282039 mfu:20.76983624541198 +Step 1836 | loss:1.5279277563095093 lr:9.942132796884584e-05 tokens_per_second_per_gpu:4830.987223614166 mfu:22.788590985202482 +Step 1837 | loss:1.1828800439834595 lr:9.942066997748833e-05 tokens_per_second_per_gpu:4669.927113390249 mfu:22.028842965587845 +Step 1838 | loss:1.291791558265686 lr:9.942001161443166e-05 tokens_per_second_per_gpu:4524.516936013242 mfu:21.342918349365966 +Step 1839 | loss:1.383912205696106 lr:9.941935287968076e-05 tokens_per_second_per_gpu:4562.0833969426985 mfu:21.520125755069373 +Step 1840 | loss:1.3436895608901978 lr:9.941869377324063e-05 tokens_per_second_per_gpu:5200.880378567742 mfu:24.53344010739846 +Step 1841 | loss:1.4388799667358398 lr:9.941803429511617e-05 tokens_per_second_per_gpu:5311.774581762827 mfu:25.056546984371767 +Step 1842 | loss:1.482878565788269 lr:9.941737444531238e-05 tokens_per_second_per_gpu:3637.9430607070217 mfu:17.160798114445434 +Step 1843 | loss:1.475581407546997 lr:9.941671422383421e-05 tokens_per_second_per_gpu:5310.949178100951 mfu:25.05265341447031 +Step 1844 | loss:1.3994659185409546 lr:9.941605363068663e-05 tokens_per_second_per_gpu:4987.236371766378 mfu:23.525644875890116 +Step 1845 | loss:1.3133918046951294 lr:9.941539266587458e-05 tokens_per_second_per_gpu:4749.332995156107 mfu:22.403414057917857 +Step 1846 | loss:1.5062294006347656 lr:9.941473132940309e-05 tokens_per_second_per_gpu:3652.742695420178 mfu:17.23061051646545 +Step 1847 | loss:1.4824037551879883 lr:9.941406962127708e-05 tokens_per_second_per_gpu:2904.4767833527094 mfu:13.700912541913965 +Step 1848 | loss:1.4302254915237427 lr:9.941340754150154e-05 tokens_per_second_per_gpu:5298.764676113836 mfu:24.99517703970668 +Step 1849 | loss:1.4893484115600586 lr:9.941274509008149e-05 tokens_per_second_per_gpu:4902.927931345468 mfu:23.127947577921464 +Step 1850 | loss:1.7253535985946655 lr:9.941208226702185e-05 tokens_per_second_per_gpu:5216.146979781681 mfu:24.605455270075375 +Step 1851 | loss:1.3623957633972168 lr:9.941141907232765e-05 tokens_per_second_per_gpu:3837.376909187819 mfu:18.10156160465265 +Step 1852 | loss:1.4760379791259766 lr:9.941075550600385e-05 tokens_per_second_per_gpu:5092.7134887084085 mfu:24.023198432757535 +Step 1853 | loss:1.4289159774780273 lr:9.941009156805545e-05 tokens_per_second_per_gpu:4466.880902004027 mfu:21.071039343223074 +Step 1854 | loss:1.455979347229004 lr:9.940942725848747e-05 tokens_per_second_per_gpu:4090.8850253170826 mfu:19.2974026413791 +Step 1855 | loss:1.5804132223129272 lr:9.940876257730487e-05 tokens_per_second_per_gpu:2970.873643259538 mfu:14.01411785856693 +Step 1856 | loss:1.4402501583099365 lr:9.940809752451266e-05 tokens_per_second_per_gpu:4717.203075467172 mfu:22.251851744815575 +Step 1857 | loss:1.4248534440994263 lr:9.940743210011584e-05 tokens_per_second_per_gpu:5331.899805890925 mfu:25.15148110030131 +Step 1858 | loss:1.5239298343658447 lr:9.940676630411942e-05 tokens_per_second_per_gpu:4650.089077872723 mfu:21.93526356733379 +Step 1859 | loss:1.4583412408828735 lr:9.940610013652841e-05 tokens_per_second_per_gpu:5322.056419744437 mfu:25.105048168393473 +Step 1860 | loss:1.5131638050079346 lr:9.940543359734782e-05 tokens_per_second_per_gpu:4616.5410101956 mfu:21.777011608210234 +Step 1861 | loss:1.431341290473938 lr:9.940476668658265e-05 tokens_per_second_per_gpu:4384.041505502931 mfu:20.68027177606892 +Step 1862 | loss:1.6000635623931885 lr:9.940409940423793e-05 tokens_per_second_per_gpu:5069.253572157461 mfu:23.912534003712214 +Step 1863 | loss:1.3217612504959106 lr:9.940343175031868e-05 tokens_per_second_per_gpu:5289.744254279334 mfu:24.95262617087067 +Step 1864 | loss:1.6325832605361938 lr:9.94027637248299e-05 tokens_per_second_per_gpu:4310.062199365332 mfu:20.331298766846686 +Step 1865 | loss:1.4028096199035645 lr:9.940209532777665e-05 tokens_per_second_per_gpu:4280.46348705714 mfu:20.19167659082788 +Step 1866 | loss:1.5169477462768555 lr:9.940142655916393e-05 tokens_per_second_per_gpu:4081.4520081404294 mfu:19.252905489918138 +Step 1867 | loss:1.3899295330047607 lr:9.940075741899677e-05 tokens_per_second_per_gpu:4205.8642737366445 mfu:19.839779373656846 +Step 1868 | loss:1.52861487865448 lr:9.940008790728023e-05 tokens_per_second_per_gpu:4177.219640918271 mfu:19.704657753374256 +Step 1869 | loss:1.4559708833694458 lr:9.93994180240193e-05 tokens_per_second_per_gpu:4034.1042270958997 mfu:19.02955792836663 +Step 1870 | loss:1.324895977973938 lr:9.939874776921906e-05 tokens_per_second_per_gpu:4427.4479179139 mfu:20.88502722930909 +Step 1871 | loss:1.515569806098938 lr:9.939807714288454e-05 tokens_per_second_per_gpu:5321.572254387184 mfu:25.102764277796183 +Step 1872 | loss:1.2658486366271973 lr:9.939740614502076e-05 tokens_per_second_per_gpu:4780.682301078273 mfu:22.55129408269585 +Step 1873 | loss:1.4687176942825317 lr:9.939673477563279e-05 tokens_per_second_per_gpu:4202.655943819254 mfu:19.824645134038988 +Step 1874 | loss:1.4675146341323853 lr:9.939606303472568e-05 tokens_per_second_per_gpu:5181.419243848973 mfu:24.441638614518055 +Step 1875 | loss:1.376399278640747 lr:9.939539092230448e-05 tokens_per_second_per_gpu:4639.017900409031 mfu:21.883038934299908 +Step 1876 | loss:1.2750760316848755 lr:9.939471843837424e-05 tokens_per_second_per_gpu:4823.64438437762 mfu:22.753953559706133 +Step 1877 | loss:1.4087923765182495 lr:9.939404558294001e-05 tokens_per_second_per_gpu:5060.728087242489 mfu:23.87231783677057 +Step 1878 | loss:1.4993199110031128 lr:9.939337235600687e-05 tokens_per_second_per_gpu:4764.219167750932 mfu:22.473634673890594 +Step 1879 | loss:1.477340579032898 lr:9.939269875757986e-05 tokens_per_second_per_gpu:3873.76087317675 mfu:18.273190970532816 +Step 1880 | loss:1.2644224166870117 lr:9.939202478766407e-05 tokens_per_second_per_gpu:4121.807939352974 mfu:19.44327130287958 +Step 1881 | loss:1.4457666873931885 lr:9.939135044626458e-05 tokens_per_second_per_gpu:4225.356634619738 mfu:19.931728165682102 +Step 1882 | loss:1.3364332914352417 lr:9.939067573338641e-05 tokens_per_second_per_gpu:4826.819278278095 mfu:22.768930075927432 +Step 1883 | loss:1.3754184246063232 lr:9.939000064903468e-05 tokens_per_second_per_gpu:4520.734729020278 mfu:21.3250770336691 +Step 1884 | loss:1.4167689085006714 lr:9.938932519321445e-05 tokens_per_second_per_gpu:4139.162465377652 mfu:19.525135563125453 +Step 1885 | loss:1.3976824283599854 lr:9.93886493659308e-05 tokens_per_second_per_gpu:4385.697315292041 mfu:20.688082513354438 +Step 1886 | loss:1.3676196336746216 lr:9.938797316718882e-05 tokens_per_second_per_gpu:4829.878312402399 mfu:22.78336006181682 +Step 1887 | loss:1.3175647258758545 lr:9.938729659699359e-05 tokens_per_second_per_gpu:3450.731804792201 mfu:16.277690678761257 +Step 1888 | loss:1.2898430824279785 lr:9.938661965535021e-05 tokens_per_second_per_gpu:5277.776630139513 mfu:24.8961728459159 +Step 1889 | loss:1.6047194004058838 lr:9.938594234226375e-05 tokens_per_second_per_gpu:4902.86403407631 mfu:23.12764616360085 +Step 1890 | loss:1.5894761085510254 lr:9.938526465773933e-05 tokens_per_second_per_gpu:5250.066408798934 mfu:24.76545900400111 +Step 1891 | loss:1.354753851890564 lr:9.938458660178201e-05 tokens_per_second_per_gpu:4383.962348245738 mfu:20.679898377781125 +Step 1892 | loss:1.4624555110931396 lr:9.938390817439693e-05 tokens_per_second_per_gpu:3558.755103463058 mfu:16.787255009266953 +Step 1893 | loss:1.4610005617141724 lr:9.938322937558915e-05 tokens_per_second_per_gpu:4200.322852264951 mfu:19.813639543110867 +Step 1894 | loss:1.3509877920150757 lr:9.938255020536383e-05 tokens_per_second_per_gpu:5327.965210839769 mfu:25.13292094413387 +Step 1895 | loss:1.4571841955184937 lr:9.938187066372604e-05 tokens_per_second_per_gpu:3432.7892555026388 mfu:16.19305261244781 +Step 1896 | loss:1.3920762538909912 lr:9.93811907506809e-05 tokens_per_second_per_gpu:3713.554176410401 mfu:17.51746864780488 +Step 1897 | loss:1.5918909311294556 lr:9.938051046623352e-05 tokens_per_second_per_gpu:3603.8890683458912 mfu:17.000159622267756 +Step 1898 | loss:1.5104836225509644 lr:9.937982981038902e-05 tokens_per_second_per_gpu:5350.731773256122 mfu:25.2403147034279 +Step 1899 | loss:1.653176188468933 lr:9.937914878315253e-05 tokens_per_second_per_gpu:5045.35165471422 mfu:23.799784581044808 +Step 1900 | loss:1.4331010580062866 lr:9.937846738452913e-05 tokens_per_second_per_gpu:5380.751362405784 mfu:25.381922227316913 +Step 1901 | loss:1.6466600894927979 lr:9.937778561452402e-05 tokens_per_second_per_gpu:2646.977006558453 mfu:12.486242160783208 +Step 1902 | loss:1.495093584060669 lr:9.937710347314227e-05 tokens_per_second_per_gpu:3883.8779990185503 mfu:18.320915179288257 +Step 1903 | loss:1.3505159616470337 lr:9.9376420960389e-05 tokens_per_second_per_gpu:4635.695536592054 mfu:21.867366777321504 +Step 1904 | loss:1.344626545906067 lr:9.937573807626939e-05 tokens_per_second_per_gpu:5336.860972176825 mfu:25.174883768134155 +Step 1905 | loss:1.4547374248504639 lr:9.937505482078855e-05 tokens_per_second_per_gpu:5100.617102397393 mfu:24.060481127024023 +Step 1906 | loss:1.4788188934326172 lr:9.937437119395161e-05 tokens_per_second_per_gpu:4675.811828067071 mfu:22.056602168754267 +Step 1907 | loss:1.2783615589141846 lr:9.937368719576374e-05 tokens_per_second_per_gpu:5325.955881971896 mfu:25.123442597037254 +Step 1908 | loss:1.5514557361602783 lr:9.937300282623005e-05 tokens_per_second_per_gpu:4821.25482225887 mfu:22.742681587490654 +Step 1909 | loss:1.4840571880340576 lr:9.937231808535573e-05 tokens_per_second_per_gpu:3696.117948608361 mfu:17.43521898633543 +Step 1910 | loss:1.4130393266677856 lr:9.937163297314588e-05 tokens_per_second_per_gpu:4394.034593420716 mfu:20.72741087677373 +Step 1911 | loss:1.3788529634475708 lr:9.93709474896057e-05 tokens_per_second_per_gpu:5344.374060396894 mfu:25.21032428713325 +Step 1912 | loss:1.4237637519836426 lr:9.937026163474033e-05 tokens_per_second_per_gpu:4361.8003216605275 mfu:20.57535631714704 +Step 1913 | loss:1.2682441473007202 lr:9.93695754085549e-05 tokens_per_second_per_gpu:5264.769907824804 mfu:24.83481791758176 +Step 1914 | loss:1.298755168914795 lr:9.93688888110546e-05 tokens_per_second_per_gpu:4254.9116555635965 mfu:20.071144242080834 +Step 1915 | loss:1.2985939979553223 lr:9.93682018422446e-05 tokens_per_second_per_gpu:4640.8698855536795 mfu:21.89177506420846 +Step 1916 | loss:1.5093367099761963 lr:9.936751450213003e-05 tokens_per_second_per_gpu:5274.564054761552 mfu:24.88101857973618 +Step 1917 | loss:1.4774636030197144 lr:9.936682679071609e-05 tokens_per_second_per_gpu:3734.590481664368 mfu:17.61670048885162 +Step 1918 | loss:1.3255853652954102 lr:9.936613870800797e-05 tokens_per_second_per_gpu:4942.591015020416 mfu:23.315045110836085 +Step 1919 | loss:1.4504412412643433 lr:9.93654502540108e-05 tokens_per_second_per_gpu:5214.873890292418 mfu:24.599449889743052 +Step 1920 | loss:1.2772927284240723 lr:9.936476142872979e-05 tokens_per_second_per_gpu:4205.069114923236 mfu:19.836028473865916 +Step 1921 | loss:1.3770159482955933 lr:9.936407223217011e-05 tokens_per_second_per_gpu:5353.220650677557 mfu:25.252055162871088 +Step 1922 | loss:1.4714949131011963 lr:9.936338266433693e-05 tokens_per_second_per_gpu:4043.7537585112864 mfu:19.075076414432836 +Step 1923 | loss:1.3678607940673828 lr:9.936269272523547e-05 tokens_per_second_per_gpu:3209.3594345769848 mfu:15.139096026083134 +Step 1924 | loss:1.5177125930786133 lr:9.936200241487088e-05 tokens_per_second_per_gpu:5389.312337055117 mfu:25.42230580539071 +Step 1925 | loss:1.4664355516433716 lr:9.936131173324839e-05 tokens_per_second_per_gpu:4953.789741265898 mfu:23.36787141323574 +Step 1926 | loss:1.262371301651001 lr:9.936062068037317e-05 tokens_per_second_per_gpu:4085.4273053534603 mfu:19.271657645127647 +Step 1927 | loss:1.362702488899231 lr:9.935992925625042e-05 tokens_per_second_per_gpu:3570.0379658823977 mfu:16.840478196353818 +Step 1928 | loss:1.3768101930618286 lr:9.935923746088534e-05 tokens_per_second_per_gpu:4972.798374875325 mfu:23.457538381178733 +Step 1929 | loss:1.4379688501358032 lr:9.935854529428315e-05 tokens_per_second_per_gpu:3968.2129568089126 mfu:18.718737564212944 +Step 1930 | loss:1.4735122919082642 lr:9.935785275644902e-05 tokens_per_second_per_gpu:4115.452398895019 mfu:19.413291134172383 +Step 1931 | loss:1.316447377204895 lr:9.93571598473882e-05 tokens_per_second_per_gpu:4921.373292787936 mfu:23.21495749494877 +Step 1932 | loss:1.3149045705795288 lr:9.935646656710588e-05 tokens_per_second_per_gpu:4490.383167635233 mfu:21.181903540105477 +Step 1933 | loss:1.5282293558120728 lr:9.935577291560728e-05 tokens_per_second_per_gpu:4452.009688204814 mfu:21.000889290441478 +Step 1934 | loss:1.508345127105713 lr:9.93550788928976e-05 tokens_per_second_per_gpu:4311.421456573421 mfu:20.337710615010433 +Step 1935 | loss:1.3042073249816895 lr:9.935438449898209e-05 tokens_per_second_per_gpu:4504.997219690088 mfu:21.250840517062624 +Step 1936 | loss:1.4493346214294434 lr:9.935368973386594e-05 tokens_per_second_per_gpu:5333.726680121038 mfu:25.160098777741638 +Step 1937 | loss:1.4551317691802979 lr:9.93529945975544e-05 tokens_per_second_per_gpu:4570.720916347099 mfu:21.560870407812704 +Step 1938 | loss:1.4209312200546265 lr:9.93522990900527e-05 tokens_per_second_per_gpu:5327.103585874543 mfu:25.128856512164518 +Step 1939 | loss:1.352137565612793 lr:9.935160321136605e-05 tokens_per_second_per_gpu:3676.7646276122446 mfu:17.34392606917969 +Step 1940 | loss:1.3559761047363281 lr:9.935090696149969e-05 tokens_per_second_per_gpu:4967.371431178227 mfu:23.431938561827447 +Step 1941 | loss:1.1811048984527588 lr:9.935021034045888e-05 tokens_per_second_per_gpu:4126.8518659367255 mfu:19.46706436515775 +Step 1942 | loss:1.411982774734497 lr:9.934951334824882e-05 tokens_per_second_per_gpu:3817.34534835035 mfu:18.007069314445605 +Step 1943 | loss:1.426064133644104 lr:9.934881598487479e-05 tokens_per_second_per_gpu:5324.999938346646 mfu:25.11893324785615 +Step 1944 | loss:1.2881616353988647 lr:9.9348118250342e-05 tokens_per_second_per_gpu:4420.391588459375 mfu:20.851741319338153 +Step 1945 | loss:1.3693113327026367 lr:9.934742014465573e-05 tokens_per_second_per_gpu:5219.684717012034 mfu:24.622143380191158 +Step 1946 | loss:1.4207643270492554 lr:9.934672166782121e-05 tokens_per_second_per_gpu:4520.427961499916 mfu:21.323629958935616 +Step 1947 | loss:1.4809694290161133 lr:9.93460228198437e-05 tokens_per_second_per_gpu:5316.30423506943 mfu:25.077914131857558 +Step 1948 | loss:1.5195423364639282 lr:9.934532360072847e-05 tokens_per_second_per_gpu:4245.222523225679 mfu:20.025438951706434 +Step 1949 | loss:1.3643012046813965 lr:9.934462401048074e-05 tokens_per_second_per_gpu:5360.601182900127 mfu:25.286870392613594 +Step 1950 | loss:1.5419334173202515 lr:9.934392404910582e-05 tokens_per_second_per_gpu:5346.952120855506 mfu:25.222485438178996 +Step 1951 | loss:1.3547782897949219 lr:9.934322371660893e-05 tokens_per_second_per_gpu:4817.090232056933 mfu:22.723036504955758 +Step 1952 | loss:1.4824268817901611 lr:9.934252301299538e-05 tokens_per_second_per_gpu:5353.946669189682 mfu:25.25547991606431 +Step 1953 | loss:1.4161194562911987 lr:9.934182193827041e-05 tokens_per_second_per_gpu:5345.989389465891 mfu:25.21794406995527 +Step 1954 | loss:1.4476546049118042 lr:9.93411204924393e-05 tokens_per_second_per_gpu:5334.803090940456 mfu:25.16517639123146 +Step 1955 | loss:1.477028250694275 lr:9.934041867550732e-05 tokens_per_second_per_gpu:5220.696625556888 mfu:24.626916725446637 +Step 1956 | loss:1.3897404670715332 lr:9.933971648747976e-05 tokens_per_second_per_gpu:5347.4900760404835 mfu:25.225023064571154 +Step 1957 | loss:1.3645132780075073 lr:9.93390139283619e-05 tokens_per_second_per_gpu:4705.9569887357 mfu:22.198802034922963 +Step 1958 | loss:1.4368325471878052 lr:9.933831099815902e-05 tokens_per_second_per_gpu:5350.951894365022 mfu:25.241353052255114 +Step 1959 | loss:1.365036129951477 lr:9.93376076968764e-05 tokens_per_second_per_gpu:4266.595749884317 mfu:20.12626010850373 +Step 1960 | loss:1.3682180643081665 lr:9.933690402451935e-05 tokens_per_second_per_gpu:4443.717470978665 mfu:20.961773486987582 +Step 1961 | loss:1.5040605068206787 lr:9.933619998109314e-05 tokens_per_second_per_gpu:5093.9876184400155 mfu:24.029208720090477 +Step 1962 | loss:1.2512098550796509 lr:9.933549556660306e-05 tokens_per_second_per_gpu:4845.883570613413 mfu:22.85885959557681 +Step 1963 | loss:1.4408291578292847 lr:9.933479078105446e-05 tokens_per_second_per_gpu:4314.952496467596 mfu:20.35436713264879 +Step 1964 | loss:1.4100641012191772 lr:9.933408562445256e-05 tokens_per_second_per_gpu:5351.162115944416 mfu:25.242344703312614 +Step 1965 | loss:1.3100799322128296 lr:9.933338009680273e-05 tokens_per_second_per_gpu:3752.4700697288854 mfu:17.70104155632424 +Step 1966 | loss:1.2303495407104492 lr:9.933267419811025e-05 tokens_per_second_per_gpu:4269.317471781661 mfu:20.139098935067143 +Step 1967 | loss:1.4341440200805664 lr:9.933196792838045e-05 tokens_per_second_per_gpu:4254.775160205871 mfu:20.070500370189933 +Step 1968 | loss:1.39264976978302 lr:9.93312612876186e-05 tokens_per_second_per_gpu:4368.05872868264 mfu:20.604878290864697 +Step 1969 | loss:1.2813720703125 lr:9.933055427583004e-05 tokens_per_second_per_gpu:4077.6603040998452 mfu:19.235019375027285 +Step 1970 | loss:1.4826045036315918 lr:9.93298468930201e-05 tokens_per_second_per_gpu:4063.7900909177556 mfu:19.169591212944972 +Step 1971 | loss:1.4936871528625488 lr:9.932913913919409e-05 tokens_per_second_per_gpu:5315.19670313914 mfu:25.07268971477794 +Step 1972 | loss:1.373641848564148 lr:9.932843101435731e-05 tokens_per_second_per_gpu:3410.7678173352056 mfu:16.089173731367243 +Step 1973 | loss:1.5756518840789795 lr:9.932772251851512e-05 tokens_per_second_per_gpu:4085.2690604274535 mfu:19.27091117681231 +Step 1974 | loss:1.4687490463256836 lr:9.932701365167283e-05 tokens_per_second_per_gpu:3988.9316691616145 mfu:18.81647126535748 +Step 1975 | loss:1.4008162021636963 lr:9.932630441383579e-05 tokens_per_second_per_gpu:4558.164081970674 mfu:21.50163767764224 +Step 1976 | loss:1.5101072788238525 lr:9.93255948050093e-05 tokens_per_second_per_gpu:3623.0626364128793 mfu:17.090604614187782 +Step 1977 | loss:1.4771842956542969 lr:9.932488482519873e-05 tokens_per_second_per_gpu:4703.108424774868 mfu:22.18536487270459 +Step 1978 | loss:1.4331703186035156 lr:9.932417447440941e-05 tokens_per_second_per_gpu:4933.133437655491 mfu:23.270432104776642 +Step 1979 | loss:1.3809337615966797 lr:9.932346375264667e-05 tokens_per_second_per_gpu:5019.247788224252 mfu:23.676648189032747 +Step 1980 | loss:1.3960963487625122 lr:9.932275265991586e-05 tokens_per_second_per_gpu:4746.485156568251 mfu:22.389980317407577 +Step 1981 | loss:1.1872425079345703 lr:9.932204119622235e-05 tokens_per_second_per_gpu:4343.191693135743 mfu:20.487576241436216 +Step 1982 | loss:1.5173054933547974 lr:9.932132936157147e-05 tokens_per_second_per_gpu:5119.486943219346 mfu:24.149493385708013 +Step 1983 | loss:1.4741705656051636 lr:9.932061715596856e-05 tokens_per_second_per_gpu:4280.155673923419 mfu:20.190224583757782 +Step 1984 | loss:1.3411791324615479 lr:9.931990457941902e-05 tokens_per_second_per_gpu:5319.9305939099295 mfu:25.09502028522881 +Step 1985 | loss:1.2995221614837646 lr:9.931919163192818e-05 tokens_per_second_per_gpu:4750.185985366887 mfu:22.40743775827724 +Step 1986 | loss:1.2879749536514282 lr:9.931847831350141e-05 tokens_per_second_per_gpu:5342.765615092754 mfu:25.20273697620456 +Step 1987 | loss:1.2432869672775269 lr:9.931776462414406e-05 tokens_per_second_per_gpu:3836.0786108094717 mfu:18.09543730968944 +Step 1988 | loss:1.4583407640457153 lr:9.931705056386151e-05 tokens_per_second_per_gpu:5002.8260063815005 mfu:23.59918384223559 +Step 1989 | loss:1.4690178632736206 lr:9.931633613265912e-05 tokens_per_second_per_gpu:4480.6568173134965 mfu:21.13602268615136 +Step 1990 | loss:1.4625097513198853 lr:9.931562133054228e-05 tokens_per_second_per_gpu:5076.0864769115215 mfu:23.944765981250878 +Step 1991 | loss:1.4158672094345093 lr:9.931490615751638e-05 tokens_per_second_per_gpu:5350.342498624726 mfu:25.238478428574503 +Step 1992 | loss:1.3740724325180054 lr:9.931419061358675e-05 tokens_per_second_per_gpu:5330.412613761315 mfu:25.144465761284795 +Step 1993 | loss:1.3008109331130981 lr:9.931347469875882e-05 tokens_per_second_per_gpu:5105.001291829075 mfu:24.081162096593122 +Step 1994 | loss:1.5691032409667969 lr:9.931275841303795e-05 tokens_per_second_per_gpu:5151.478480347029 mfu:24.30040292465837 +Step 1995 | loss:1.3287534713745117 lr:9.931204175642953e-05 tokens_per_second_per_gpu:5335.221343352499 mfu:25.167149359220286 +Step 1996 | loss:1.382727861404419 lr:9.931132472893895e-05 tokens_per_second_per_gpu:4903.816877349266 mfu:23.132140887891207 +Step 1997 | loss:1.2612344026565552 lr:9.93106073305716e-05 tokens_per_second_per_gpu:5155.820720369107 mfu:24.320886011705056 +Step 1998 | loss:1.2597354650497437 lr:9.930988956133289e-05 tokens_per_second_per_gpu:5220.16950166944 mfu:24.624430192132937 +Step 1999 | loss:1.2984287738800049 lr:9.93091714212282e-05 tokens_per_second_per_gpu:4169.411610505713 mfu:19.667825941730865 +Step 2000 | loss:1.3409628868103027 lr:9.930845291026294e-05 tokens_per_second_per_gpu:4533.360690756062 mfu:21.384635849387998 +Step 2001 | loss:1.278647541999817 lr:9.930773402844252e-05 tokens_per_second_per_gpu:4004.5425999776844 mfu:18.890110689516327 +Step 2002 | loss:1.369895577430725 lr:9.930701477577234e-05 tokens_per_second_per_gpu:4478.209999464183 mfu:21.124480628885976 +Step 2003 | loss:1.5359936952590942 lr:9.93062951522578e-05 tokens_per_second_per_gpu:5320.728280982091 mfu:25.098783111247645 +Step 2004 | loss:1.376957654953003 lr:9.930557515790433e-05 tokens_per_second_per_gpu:3827.494000070308 mfu:18.05494223614718 +Step 2005 | loss:1.2773903608322144 lr:9.930485479271735e-05 tokens_per_second_per_gpu:5331.450423197707 mfu:25.14936128546481 +Step 2006 | loss:1.4563796520233154 lr:9.930413405670225e-05 tokens_per_second_per_gpu:4316.85315847507 mfu:20.36333287962442 +Step 2007 | loss:1.394701600074768 lr:9.930341294986447e-05 tokens_per_second_per_gpu:4570.906974487729 mfu:21.561748075806847 +Step 2008 | loss:1.5353553295135498 lr:9.930269147220943e-05 tokens_per_second_per_gpu:3797.3557622911867 mfu:17.91277502642946 +Step 2009 | loss:1.3956892490386963 lr:9.930196962374256e-05 tokens_per_second_per_gpu:4071.186999611308 mfu:19.20448369329522 +Step 2010 | loss:1.3572055101394653 lr:9.930124740446928e-05 tokens_per_second_per_gpu:5387.401596669608 mfu:25.41329251698634 +Step 2011 | loss:1.3524807691574097 lr:9.930052481439504e-05 tokens_per_second_per_gpu:4339.508040594852 mfu:20.470199823904903 +Step 2012 | loss:1.5503547191619873 lr:9.929980185352526e-05 tokens_per_second_per_gpu:5346.93569391515 mfu:25.222407949498614 +Step 2013 | loss:1.5593154430389404 lr:9.929907852186536e-05 tokens_per_second_per_gpu:5320.046039653621 mfu:25.09556486249888 +Step 2014 | loss:1.3460043668746948 lr:9.929835481942082e-05 tokens_per_second_per_gpu:5363.26101052427 mfu:25.299417253330066 +Step 2015 | loss:1.4420835971832275 lr:9.929763074619705e-05 tokens_per_second_per_gpu:4732.88570097539 mfu:22.325829365069662 +Step 2016 | loss:1.4165884256362915 lr:9.929690630219953e-05 tokens_per_second_per_gpu:5325.811551056308 mfu:25.122761763483037 +Step 2017 | loss:1.4471313953399658 lr:9.929618148743365e-05 tokens_per_second_per_gpu:4269.3751421857805 mfu:20.139370976202958 +Step 2018 | loss:1.415595531463623 lr:9.929545630190493e-05 tokens_per_second_per_gpu:5225.160523724149 mfu:24.647973694721173 +Step 2019 | loss:1.4951698780059814 lr:9.929473074561878e-05 tokens_per_second_per_gpu:4783.33382790954 mfu:22.56380178715606 +Step 2020 | loss:1.2815519571304321 lr:9.929400481858066e-05 tokens_per_second_per_gpu:4778.179757203603 mfu:22.53948915626102 +Step 2021 | loss:1.4431195259094238 lr:9.929327852079607e-05 tokens_per_second_per_gpu:3617.808401190083 mfu:17.065819490176878 +Step 2022 | loss:1.5340487957000732 lr:9.929255185227042e-05 tokens_per_second_per_gpu:5049.960451118154 mfu:23.82152506002392 +Step 2023 | loss:1.301151156425476 lr:9.92918248130092e-05 tokens_per_second_per_gpu:4527.3898558463925 mfu:21.356470402389604 +Step 2024 | loss:1.4403536319732666 lr:9.929109740301787e-05 tokens_per_second_per_gpu:5066.652811868589 mfu:23.900265773694144 +Step 2025 | loss:1.3516701459884644 lr:9.929036962230192e-05 tokens_per_second_per_gpu:4736.147939354248 mfu:22.341217900097995 +Step 2026 | loss:1.4052575826644897 lr:9.928964147086681e-05 tokens_per_second_per_gpu:4670.6058409614425 mfu:22.032044639344033 +Step 2027 | loss:1.445024847984314 lr:9.928891294871801e-05 tokens_per_second_per_gpu:5312.309727679419 mfu:25.059071358965635 +Step 2028 | loss:1.460282564163208 lr:9.928818405586101e-05 tokens_per_second_per_gpu:5317.729788424754 mfu:25.084638710259103 +Step 2029 | loss:1.3448185920715332 lr:9.928745479230129e-05 tokens_per_second_per_gpu:5356.425646619955 mfu:25.26717367556007 +Step 2030 | loss:1.4391567707061768 lr:9.928672515804434e-05 tokens_per_second_per_gpu:5300.298375266334 mfu:25.00241176028507 +Step 2031 | loss:1.4002329111099243 lr:9.928599515309563e-05 tokens_per_second_per_gpu:4173.405850584923 mfu:19.68666745367213 +Step 2032 | loss:1.4385148286819458 lr:9.928526477746065e-05 tokens_per_second_per_gpu:4395.727093253041 mfu:20.735394687253056 +Step 2033 | loss:1.474013328552246 lr:9.928453403114493e-05 tokens_per_second_per_gpu:5382.804672800037 mfu:25.391608042777484 +Step 2034 | loss:1.3400498628616333 lr:9.928380291415394e-05 tokens_per_second_per_gpu:5344.924015655652 mfu:25.212918519920855 +Step 2035 | loss:1.355545163154602 lr:9.928307142649316e-05 tokens_per_second_per_gpu:4879.301027887421 mfu:23.016495443144446 +Step 2036 | loss:1.3395843505859375 lr:9.928233956816812e-05 tokens_per_second_per_gpu:5333.702871399348 mfu:25.15998646793924 +Step 2037 | loss:1.3070989847183228 lr:9.928160733918432e-05 tokens_per_second_per_gpu:3597.8447798268326 mfu:16.971647682061395 +Step 2038 | loss:1.4594981670379639 lr:9.928087473954726e-05 tokens_per_second_per_gpu:4212.812790390692 mfu:19.87255671225318 +Step 2039 | loss:1.6092934608459473 lr:9.928014176926244e-05 tokens_per_second_per_gpu:3531.9336830592365 mfu:16.660733793015925 +Step 2040 | loss:1.4035836458206177 lr:9.927940842833539e-05 tokens_per_second_per_gpu:5225.747592118737 mfu:24.650742996502636 +Step 2041 | loss:1.3996471166610718 lr:9.927867471677164e-05 tokens_per_second_per_gpu:5368.232900664918 mfu:25.32287050741543 +Step 2042 | loss:1.4358645677566528 lr:9.927794063457668e-05 tokens_per_second_per_gpu:3702.234429421282 mfu:17.464071469909214 +Step 2043 | loss:1.3182671070098877 lr:9.927720618175603e-05 tokens_per_second_per_gpu:4802.566367528598 mfu:22.654524957948784 +Step 2044 | loss:1.4019242525100708 lr:9.927647135831524e-05 tokens_per_second_per_gpu:4760.011483723644 mfu:22.45378631042895 +Step 2045 | loss:1.325217366218567 lr:9.92757361642598e-05 tokens_per_second_per_gpu:5350.089417750732 mfu:25.23728460291182 +Step 2046 | loss:1.4560552835464478 lr:9.92750005995953e-05 tokens_per_second_per_gpu:5259.813989415498 mfu:24.81144000487819 +Step 2047 | loss:1.3294751644134521 lr:9.927426466432718e-05 tokens_per_second_per_gpu:4092.55643417669 mfu:19.305286962141626 +Step 2048 | loss:1.537394642829895 lr:9.927352835846106e-05 tokens_per_second_per_gpu:4130.64475625308 mfu:19.48495607591381 +Step 2049 | loss:1.3289700746536255 lr:9.927279168200245e-05 tokens_per_second_per_gpu:5316.562617933939 mfu:25.079132967914106 +Step 2050 | loss:1.5304728746414185 lr:9.927205463495689e-05 tokens_per_second_per_gpu:5302.3717502805885 mfu:25.012192223981565 +Step 2051 | loss:1.3903511762619019 lr:9.92713172173299e-05 tokens_per_second_per_gpu:4942.589894119278 mfu:23.31503982335417 +Step 2052 | loss:1.3274487257003784 lr:9.927057942912705e-05 tokens_per_second_per_gpu:4867.926669008749 mfu:22.96284065164801 +Step 2053 | loss:1.5316611528396606 lr:9.926984127035389e-05 tokens_per_second_per_gpu:4651.0869419959645 mfu:21.9399706626563 +Step 2054 | loss:1.4264637231826782 lr:9.926910274101597e-05 tokens_per_second_per_gpu:4857.016528812078 mfu:22.91137565887845 +Step 2055 | loss:1.377486228942871 lr:9.926836384111884e-05 tokens_per_second_per_gpu:4736.069487652286 mfu:22.3408478300346 +Step 2056 | loss:1.3986438512802124 lr:9.926762457066805e-05 tokens_per_second_per_gpu:5169.383736897427 mfu:24.384865074758384 +Step 2057 | loss:1.438049554824829 lr:9.926688492966918e-05 tokens_per_second_per_gpu:3748.4530908754937 mfu:17.682092781705524 +Step 2058 | loss:1.4145451784133911 lr:9.926614491812779e-05 tokens_per_second_per_gpu:5167.606064068517 mfu:24.37647948872215 +Step 2059 | loss:1.3980070352554321 lr:9.926540453604941e-05 tokens_per_second_per_gpu:3600.967754153777 mfu:16.986379284795955 +Step 2060 | loss:1.3836309909820557 lr:9.926466378343965e-05 tokens_per_second_per_gpu:4759.795976158154 mfu:22.452769724473153 +Step 2061 | loss:1.4377270936965942 lr:9.926392266030406e-05 tokens_per_second_per_gpu:3771.179880465044 mfu:17.78929892578958 +Step 2062 | loss:1.3716087341308594 lr:9.926318116664824e-05 tokens_per_second_per_gpu:5248.4573363473955 mfu:24.75786873471113 +Step 2063 | loss:1.2194733619689941 lr:9.926243930247774e-05 tokens_per_second_per_gpu:4730.307236379832 mfu:22.31366630764099 +Step 2064 | loss:1.402060866355896 lr:9.926169706779815e-05 tokens_per_second_per_gpu:5371.475874310488 mfu:25.33816816740994 +Step 2065 | loss:1.441888689994812 lr:9.926095446261504e-05 tokens_per_second_per_gpu:5364.7466485120585 mfu:25.30642526119444 +Step 2066 | loss:1.447057843208313 lr:9.926021148693401e-05 tokens_per_second_per_gpu:3583.7825054876093 mfu:16.90531353473187 +Step 2067 | loss:1.585414171218872 lr:9.925946814076066e-05 tokens_per_second_per_gpu:5357.119067773597 mfu:25.27044466145165 +Step 2068 | loss:1.4117051362991333 lr:9.925872442410054e-05 tokens_per_second_per_gpu:5337.337951922419 mfu:25.177133762975256 +Step 2069 | loss:1.4325573444366455 lr:9.925798033695928e-05 tokens_per_second_per_gpu:4652.380693827247 mfu:21.946073510781286 +Step 2070 | loss:1.4941293001174927 lr:9.925723587934247e-05 tokens_per_second_per_gpu:4933.323889386494 mfu:23.271330498086197 +Step 2071 | loss:1.2964130640029907 lr:9.92564910512557e-05 tokens_per_second_per_gpu:5241.286887408301 mfu:24.72404450366068 +Step 2072 | loss:1.391052484512329 lr:9.925574585270456e-05 tokens_per_second_per_gpu:5285.006712047231 mfu:24.930278375853785 +Step 2073 | loss:1.4546079635620117 lr:9.92550002836947e-05 tokens_per_second_per_gpu:4466.087356620577 mfu:21.067296054256538 +Step 2074 | loss:1.2062656879425049 lr:9.925425434423168e-05 tokens_per_second_per_gpu:5372.101014545024 mfu:25.341117060556172 +Step 2075 | loss:1.4097588062286377 lr:9.925350803432112e-05 tokens_per_second_per_gpu:4433.664324341994 mfu:20.91435107905872 +Step 2076 | loss:1.5240201950073242 lr:9.925276135396866e-05 tokens_per_second_per_gpu:4555.009353286828 mfu:21.486756284188324 +Step 2077 | loss:1.21089506149292 lr:9.925201430317989e-05 tokens_per_second_per_gpu:5220.660381835419 mfu:24.62674575762865 +Step 2078 | loss:1.7426824569702148 lr:9.925126688196042e-05 tokens_per_second_per_gpu:5354.640980648416 mfu:25.25875510167691 +Step 2079 | loss:1.2119659185409546 lr:9.925051909031591e-05 tokens_per_second_per_gpu:5349.996934281029 mfu:25.236848342605928 +Step 2080 | loss:1.373219609260559 lr:9.924977092825197e-05 tokens_per_second_per_gpu:3385.951623956012 mfu:15.972111513118394 +Step 2081 | loss:1.4081975221633911 lr:9.924902239577418e-05 tokens_per_second_per_gpu:4672.1617355343315 mfu:22.039384059507338 +Step 2082 | loss:1.380661129951477 lr:9.924827349288823e-05 tokens_per_second_per_gpu:3747.962456488535 mfu:17.679778375591383 +Step 2083 | loss:1.359086036682129 lr:9.924752421959973e-05 tokens_per_second_per_gpu:4394.880105778456 mfu:20.731399302824475 +Step 2084 | loss:1.277417540550232 lr:9.924677457591433e-05 tokens_per_second_per_gpu:4696.751641692839 mfu:22.155378842327632 +Step 2085 | loss:1.4999287128448486 lr:9.924602456183764e-05 tokens_per_second_per_gpu:4322.99251591964 mfu:20.39229327617285 +Step 2086 | loss:1.403241753578186 lr:9.924527417737532e-05 tokens_per_second_per_gpu:4429.261121754116 mfu:20.89358042118787 +Step 2087 | loss:1.482536792755127 lr:9.924452342253299e-05 tokens_per_second_per_gpu:4663.402450632485 mfu:21.998065018137215 +Step 2088 | loss:1.4307358264923096 lr:9.924377229731633e-05 tokens_per_second_per_gpu:4482.4800510043815 mfu:21.14462319054701 +Step 2089 | loss:1.2966724634170532 lr:9.9243020801731e-05 tokens_per_second_per_gpu:4482.836652468833 mfu:21.14630534049731 +Step 2090 | loss:1.4381825923919678 lr:9.924226893578259e-05 tokens_per_second_per_gpu:3798.676270948677 mfu:17.919004091068853 +Step 2091 | loss:1.4074803590774536 lr:9.92415166994768e-05 tokens_per_second_per_gpu:4865.557828105723 mfu:22.951666424942502 +Step 2092 | loss:1.3245574235916138 lr:9.92407640928193e-05 tokens_per_second_per_gpu:4737.092737342463 mfu:22.34567467340757 +Step 2093 | loss:1.3171436786651611 lr:9.924001111581571e-05 tokens_per_second_per_gpu:5366.549180581525 mfu:25.3149281125098 +Step 2094 | loss:1.4791767597198486 lr:9.923925776847172e-05 tokens_per_second_per_gpu:5334.464299493269 mfu:25.163578254171295 +Step 2095 | loss:1.5100526809692383 lr:9.9238504050793e-05 tokens_per_second_per_gpu:5002.787384117115 mfu:23.599001654424846 +Step 2096 | loss:1.2922627925872803 lr:9.923774996278519e-05 tokens_per_second_per_gpu:5334.725419662046 mfu:25.16481000255911 +Step 2097 | loss:1.475292682647705 lr:9.923699550445397e-05 tokens_per_second_per_gpu:3806.1074533998344 mfu:17.95405824658143 +Step 2098 | loss:1.2251416444778442 lr:9.923624067580506e-05 tokens_per_second_per_gpu:5367.567691201736 mfu:25.319732600881164 +Step 2099 | loss:1.2342816591262817 lr:9.923548547684408e-05 tokens_per_second_per_gpu:3868.350588829702 mfu:18.247669736178096 +Step 2100 | loss:1.4240816831588745 lr:9.923472990757673e-05 tokens_per_second_per_gpu:4682.751705697782 mfu:22.08933875560356 +Step 2101 | loss:1.6648565530776978 lr:9.923397396800869e-05 tokens_per_second_per_gpu:4155.06287430935 mfu:19.60014050494578 +Step 2102 | loss:1.4724406003952026 lr:9.923321765814566e-05 tokens_per_second_per_gpu:3727.043908522784 mfu:17.58110201576462 +Step 2103 | loss:1.2400457859039307 lr:9.923246097799331e-05 tokens_per_second_per_gpu:3978.3888086803504 mfu:18.766738793669873 +Step 2104 | loss:1.375085473060608 lr:9.923170392755736e-05 tokens_per_second_per_gpu:5357.609604137497 mfu:25.272758605174296 +Step 2105 | loss:1.3244118690490723 lr:9.923094650684345e-05 tokens_per_second_per_gpu:5338.628946907892 mfu:25.18322360658819 +Step 2106 | loss:1.435118556022644 lr:9.923018871585734e-05 tokens_per_second_per_gpu:5251.258048834078 mfu:24.77108017336164 +Step 2107 | loss:1.376429557800293 lr:9.922943055460467e-05 tokens_per_second_per_gpu:5316.075316634534 mfu:25.076834284542638 +Step 2108 | loss:1.4246506690979004 lr:9.922867202309118e-05 tokens_per_second_per_gpu:4616.996350274234 mfu:21.779159525050176 +Step 2109 | loss:1.3176008462905884 lr:9.922791312132258e-05 tokens_per_second_per_gpu:3674.2220450763934 mfu:17.33193227354877 +Step 2110 | loss:1.4087369441986084 lr:9.922715384930456e-05 tokens_per_second_per_gpu:3618.0996562356736 mfu:17.067193389920167 +Step 2111 | loss:1.3212014436721802 lr:9.922639420704282e-05 tokens_per_second_per_gpu:4317.3092495964 mfu:20.365484339260906 +Step 2112 | loss:1.4677743911743164 lr:9.92256341945431e-05 tokens_per_second_per_gpu:3896.1749281260245 mfu:18.378921891960662 +Step 2113 | loss:1.522922158241272 lr:9.92248738118111e-05 tokens_per_second_per_gpu:5351.226070970465 mfu:25.242646390082342 +Step 2114 | loss:1.3586193323135376 lr:9.922411305885255e-05 tokens_per_second_per_gpu:3710.3243102527117 mfu:17.502232818067068 +Step 2115 | loss:1.5190465450286865 lr:9.922335193567316e-05 tokens_per_second_per_gpu:5151.909390309559 mfu:24.302435600472446 +Step 2116 | loss:1.4656327962875366 lr:9.922259044227867e-05 tokens_per_second_per_gpu:5048.6440075745995 mfu:23.815315171220544 +Step 2117 | loss:1.4448893070220947 lr:9.922182857867477e-05 tokens_per_second_per_gpu:4159.196562995845 mfu:19.619639819759993 +Step 2118 | loss:1.394780158996582 lr:9.922106634486723e-05 tokens_per_second_per_gpu:3467.271352783569 mfu:16.355710548574418 +Step 2119 | loss:1.4205151796340942 lr:9.922030374086177e-05 tokens_per_second_per_gpu:4786.626076236494 mfu:22.57933188422979 +Step 2120 | loss:1.3506040573120117 lr:9.921954076666413e-05 tokens_per_second_per_gpu:4272.863526780244 mfu:20.155826281514827 +Step 2121 | loss:1.3913933038711548 lr:9.921877742228005e-05 tokens_per_second_per_gpu:5092.488037820053 mfu:24.02213494245128 +Step 2122 | loss:1.4360592365264893 lr:9.921801370771524e-05 tokens_per_second_per_gpu:4797.454871029455 mfu:22.630413156851098 +Step 2123 | loss:1.5162599086761475 lr:9.921724962297548e-05 tokens_per_second_per_gpu:4195.87823567003 mfu:19.792673528302842 +Step 2124 | loss:1.396372675895691 lr:9.921648516806652e-05 tokens_per_second_per_gpu:3731.38031568073 mfu:17.601557588196005 +Step 2125 | loss:1.311449646949768 lr:9.921572034299407e-05 tokens_per_second_per_gpu:4765.071867939998 mfu:22.47765700616807 +Step 2126 | loss:1.6075494289398193 lr:9.921495514776393e-05 tokens_per_second_per_gpu:4873.00958317877 mfu:22.986817625022585 +Step 2127 | loss:1.4191405773162842 lr:9.921418958238182e-05 tokens_per_second_per_gpu:3751.1959630552474 mfu:17.695031377759673 +Step 2128 | loss:1.3717010021209717 lr:9.92134236468535e-05 tokens_per_second_per_gpu:4298.263988658343 mfu:20.2756445939595 +Step 2129 | loss:1.4427530765533447 lr:9.921265734118476e-05 tokens_per_second_per_gpu:4672.550183743346 mfu:22.041216435985355 +Step 2130 | loss:1.3237276077270508 lr:9.921189066538134e-05 tokens_per_second_per_gpu:4794.156200743385 mfu:22.61485276630052 +Step 2131 | loss:1.3546099662780762 lr:9.921112361944902e-05 tokens_per_second_per_gpu:5314.766247305149 mfu:25.07065918116567 +Step 2132 | loss:1.3226193189620972 lr:9.921035620339354e-05 tokens_per_second_per_gpu:5154.764033516965 mfu:24.31590143953413 +Step 2133 | loss:1.2350032329559326 lr:9.92095884172207e-05 tokens_per_second_per_gpu:3740.9319710994937 mfu:17.646614376486777 +Step 2134 | loss:1.551865816116333 lr:9.920882026093627e-05 tokens_per_second_per_gpu:5372.384548813968 mfu:25.342454540078712 +Step 2135 | loss:1.3452322483062744 lr:9.920805173454601e-05 tokens_per_second_per_gpu:4238.413441652441 mfu:19.99331935217636 +Step 2136 | loss:1.4410592317581177 lr:9.920728283805574e-05 tokens_per_second_per_gpu:3697.3390706788905 mfu:17.440979227487027 +Step 2137 | loss:1.248839020729065 lr:9.92065135714712e-05 tokens_per_second_per_gpu:4775.458528939431 mfu:22.5266526582495 +Step 2138 | loss:1.4679638147354126 lr:9.92057439347982e-05 tokens_per_second_per_gpu:4316.730538675591 mfu:20.36275446110918 +Step 2139 | loss:1.3412128686904907 lr:9.920497392804251e-05 tokens_per_second_per_gpu:3426.911178050999 mfu:16.165324718204996 +Step 2140 | loss:1.5364755392074585 lr:9.920420355120994e-05 tokens_per_second_per_gpu:5331.498561314783 mfu:25.149588361176264 +Step 2141 | loss:1.3180389404296875 lr:9.920343280430627e-05 tokens_per_second_per_gpu:5192.689004155657 mfu:24.494799996704067 +Step 2142 | loss:1.6108371019363403 lr:9.920266168733732e-05 tokens_per_second_per_gpu:5315.913222914395 mfu:25.076069660808702 +Step 2143 | loss:1.3378379344940186 lr:9.920189020030885e-05 tokens_per_second_per_gpu:5355.21945326674 mfu:25.261483855715145 +Step 2144 | loss:1.444989562034607 lr:9.92011183432267e-05 tokens_per_second_per_gpu:4569.56637108535 mfu:21.5554242208273 +Step 2145 | loss:1.39577317237854 lr:9.920034611609665e-05 tokens_per_second_per_gpu:3903.2951469639306 mfu:18.41250917905367 +Step 2146 | loss:1.4948447942733765 lr:9.919957351892453e-05 tokens_per_second_per_gpu:4777.685516712104 mfu:22.537157739536493 +Step 2147 | loss:1.5099927186965942 lr:9.919880055171614e-05 tokens_per_second_per_gpu:3127.936328071943 mfu:14.755009340484362 +Step 2148 | loss:1.3385566473007202 lr:9.919802721447729e-05 tokens_per_second_per_gpu:4733.320332832593 mfu:22.327879597695926 +Step 2149 | loss:1.463986873626709 lr:9.91972535072138e-05 tokens_per_second_per_gpu:4189.867338555181 mfu:19.7643191010473 +Step 2150 | loss:1.4777711629867554 lr:9.919647942993148e-05 tokens_per_second_per_gpu:5170.6304902879765 mfu:24.390746223219665 +Step 2151 | loss:1.3015117645263672 lr:9.919570498263617e-05 tokens_per_second_per_gpu:3773.9002519666533 mfu:17.802131382305912 +Step 2152 | loss:1.135724425315857 lr:9.919493016533369e-05 tokens_per_second_per_gpu:5353.210756391439 mfu:25.252008489835326 +Step 2153 | loss:1.4082982540130615 lr:9.919415497802987e-05 tokens_per_second_per_gpu:5336.829074453369 mfu:25.174733301130296 +Step 2154 | loss:1.3149927854537964 lr:9.919337942073051e-05 tokens_per_second_per_gpu:5167.737350577479 mfu:24.377098789584714 +Step 2155 | loss:1.205463171005249 lr:9.919260349344147e-05 tokens_per_second_per_gpu:5346.528064126226 mfu:25.220485090235446 +Step 2156 | loss:1.2906001806259155 lr:9.91918271961686e-05 tokens_per_second_per_gpu:5097.819662014398 mfu:24.047285123444894 +Step 2157 | loss:1.4260139465332031 lr:9.919105052891771e-05 tokens_per_second_per_gpu:5334.814742150847 mfu:25.165231351978775 +Step 2158 | loss:1.3696720600128174 lr:9.919027349169464e-05 tokens_per_second_per_gpu:5089.022870666337 mfu:24.00578915776872 +Step 2159 | loss:1.268959879875183 lr:9.918949608450526e-05 tokens_per_second_per_gpu:5049.631492963209 mfu:23.819973308280996 +Step 2160 | loss:1.3987270593643188 lr:9.918871830735541e-05 tokens_per_second_per_gpu:4665.140597197816 mfu:22.006264152044263 +Step 2161 | loss:1.5136408805847168 lr:9.918794016025092e-05 tokens_per_second_per_gpu:5316.317911841019 mfu:25.077978647522638 +Step 2162 | loss:1.4296517372131348 lr:9.918716164319766e-05 tokens_per_second_per_gpu:4387.07376324933 mfu:20.694575453215208 +Step 2163 | loss:1.1964582204818726 lr:9.918638275620148e-05 tokens_per_second_per_gpu:4630.455096068756 mfu:21.842646725261755 +Step 2164 | loss:1.388988733291626 lr:9.918560349926824e-05 tokens_per_second_per_gpu:4637.612163525059 mfu:21.876407833574767 +Step 2165 | loss:1.3061785697937012 lr:9.91848238724038e-05 tokens_per_second_per_gpu:4005.0960249447935 mfu:18.89272128951022 +Step 2166 | loss:1.3777068853378296 lr:9.918404387561402e-05 tokens_per_second_per_gpu:5170.642085219843 mfu:24.390800918491575 +Step 2167 | loss:1.2026245594024658 lr:9.918326350890476e-05 tokens_per_second_per_gpu:4257.480280802142 mfu:20.083260885583712 +Step 2168 | loss:1.4401915073394775 lr:9.918248277228192e-05 tokens_per_second_per_gpu:4066.485382712412 mfu:19.182305364204566 +Step 2169 | loss:1.3191226720809937 lr:9.918170166575136e-05 tokens_per_second_per_gpu:4931.499265601185 mfu:23.262723436377975 +Step 2170 | loss:1.466754674911499 lr:9.918092018931893e-05 tokens_per_second_per_gpu:3627.0035892168653 mfu:17.109194760960065 +Step 2171 | loss:1.5486135482788086 lr:9.918013834299051e-05 tokens_per_second_per_gpu:4019.143450729198 mfu:18.958985393673238 +Step 2172 | loss:1.405060887336731 lr:9.917935612677201e-05 tokens_per_second_per_gpu:5333.581680328314 mfu:25.159414788979845 +Step 2173 | loss:1.5230238437652588 lr:9.917857354066931e-05 tokens_per_second_per_gpu:4992.6234895115995 mfu:23.55105682943139 +Step 2174 | loss:1.363813877105713 lr:9.917779058468825e-05 tokens_per_second_per_gpu:4252.962519815318 mfu:20.061949836199354 +Step 2175 | loss:1.4595489501953125 lr:9.917700725883478e-05 tokens_per_second_per_gpu:4499.22583633304 mfu:21.223615917067885 +Step 2176 | loss:1.3227506875991821 lr:9.917622356311476e-05 tokens_per_second_per_gpu:5355.164728787279 mfu:25.261225711008574 +Step 2177 | loss:1.3716087341308594 lr:9.917543949753407e-05 tokens_per_second_per_gpu:4767.135707314401 mfu:22.487392488624437 +Step 2178 | loss:1.3921767473220825 lr:9.917465506209864e-05 tokens_per_second_per_gpu:4046.218166701733 mfu:19.08670144833857 +Step 2179 | loss:1.440216302871704 lr:9.917387025681435e-05 tokens_per_second_per_gpu:5361.517273251418 mfu:25.291191747101767 +Step 2180 | loss:1.3939889669418335 lr:9.91730850816871e-05 tokens_per_second_per_gpu:5310.86339781605 mfu:25.052248774230804 +Step 2181 | loss:1.2509167194366455 lr:9.917229953672282e-05 tokens_per_second_per_gpu:5312.748095035065 mfu:25.061139212575398 +Step 2182 | loss:1.4653252363204956 lr:9.91715136219274e-05 tokens_per_second_per_gpu:4474.541209011768 mfu:21.107174318361736 +Step 2183 | loss:1.431700348854065 lr:9.917072733730673e-05 tokens_per_second_per_gpu:4677.535551325497 mfu:22.064733265461783 +Step 2184 | loss:1.3584518432617188 lr:9.916994068286676e-05 tokens_per_second_per_gpu:4817.938138835232 mfu:22.72703622589681 +Step 2185 | loss:1.3597551584243774 lr:9.916915365861339e-05 tokens_per_second_per_gpu:5345.646422915781 mfu:25.216326238222273 +Step 2186 | loss:1.5421761274337769 lr:9.916836626455255e-05 tokens_per_second_per_gpu:3051.7377479553143 mfu:14.395567637255855 +Step 2187 | loss:1.3386164903640747 lr:9.916757850069014e-05 tokens_per_second_per_gpu:3685.3864170196107 mfu:17.3845965208432 +Step 2188 | loss:1.314652681350708 lr:9.916679036703211e-05 tokens_per_second_per_gpu:3759.0842064170874 mfu:17.732241567570572 +Step 2189 | loss:1.5171836614608765 lr:9.916600186358435e-05 tokens_per_second_per_gpu:3194.7508423347895 mfu:15.070184803993845 +Step 2190 | loss:1.2956056594848633 lr:9.916521299035286e-05 tokens_per_second_per_gpu:5307.357707256627 mfu:25.03571183372952 +Step 2191 | loss:1.5111267566680908 lr:9.916442374734349e-05 tokens_per_second_per_gpu:4211.95676167461 mfu:19.86851867874586 +Step 2192 | loss:1.368862271308899 lr:9.916363413456223e-05 tokens_per_second_per_gpu:4827.767302809864 mfu:22.773402069393452 +Step 2193 | loss:1.227571964263916 lr:9.916284415201501e-05 tokens_per_second_per_gpu:3842.746393204867 mfu:18.12689037688951 +Step 2194 | loss:1.333513855934143 lr:9.916205379970775e-05 tokens_per_second_per_gpu:4793.948341385101 mfu:22.613872258243028 +Step 2195 | loss:1.5779012441635132 lr:9.916126307764644e-05 tokens_per_second_per_gpu:4937.483923134455 mfu:23.290954087050277 +Step 2196 | loss:1.5279940366744995 lr:9.916047198583698e-05 tokens_per_second_per_gpu:3554.816693682034 mfu:16.768676858367865 +Step 2197 | loss:1.5445809364318848 lr:9.915968052428532e-05 tokens_per_second_per_gpu:5195.993145882846 mfu:24.510386196975926 +Step 2198 | loss:1.3696837425231934 lr:9.915888869299745e-05 tokens_per_second_per_gpu:4923.407285188701 mfu:23.224552184138066 +Step 2199 | loss:1.4457921981811523 lr:9.915809649197931e-05 tokens_per_second_per_gpu:5356.250996388475 mfu:25.266349820619777 +Step 2200 | loss:1.3619343042373657 lr:9.915730392123684e-05 tokens_per_second_per_gpu:4060.8110166348083 mfu:19.155538411269823 +Step 2201 | loss:1.379844307899475 lr:9.915651098077601e-05 tokens_per_second_per_gpu:5353.209710603045 mfu:25.252003556673042 +Step 2202 | loss:1.338866114616394 lr:9.91557176706028e-05 tokens_per_second_per_gpu:4297.32649672597 mfu:20.27122228456188 +Step 2203 | loss:1.4318864345550537 lr:9.915492399072317e-05 tokens_per_second_per_gpu:5324.11381269815 mfu:25.114753241983568 +Step 2204 | loss:1.5393931865692139 lr:9.915412994114308e-05 tokens_per_second_per_gpu:5322.713022441764 mfu:25.108145475344735 +Step 2205 | loss:1.2335340976715088 lr:9.915333552186851e-05 tokens_per_second_per_gpu:5321.559534123782 mfu:25.102704274143413 +Step 2206 | loss:1.4375154972076416 lr:9.915254073290542e-05 tokens_per_second_per_gpu:5302.477499508652 mfu:25.01269106113302 +Step 2207 | loss:1.258103370666504 lr:9.915174557425981e-05 tokens_per_second_per_gpu:4678.5877649657705 mfu:22.069696736729114 +Step 2208 | loss:1.364956259727478 lr:9.915095004593764e-05 tokens_per_second_per_gpu:4229.161116147152 mfu:19.9496745541584 +Step 2209 | loss:1.4429025650024414 lr:9.915015414794491e-05 tokens_per_second_per_gpu:5178.57156995178 mfu:24.428205650881726 +Step 2210 | loss:1.456344485282898 lr:9.91493578802876e-05 tokens_per_second_per_gpu:3899.1352558119415 mfu:18.39288626273912 +Step 2211 | loss:1.2551862001419067 lr:9.91485612429717e-05 tokens_per_second_per_gpu:3608.3014039277164 mfu:17.020973362029302 +Step 2212 | loss:1.3813765048980713 lr:9.914776423600318e-05 tokens_per_second_per_gpu:5214.1775033185495 mfu:24.596164913571446 +Step 2213 | loss:1.4940991401672363 lr:9.914696685938807e-05 tokens_per_second_per_gpu:4294.338936614961 mfu:20.257129453787936 +Step 2214 | loss:1.4543335437774658 lr:9.914616911313235e-05 tokens_per_second_per_gpu:3924.8822780218175 mfu:18.514339359398075 +Step 2215 | loss:1.4536058902740479 lr:9.914537099724203e-05 tokens_per_second_per_gpu:3757.9420123484742 mfu:17.72685364327052 +Step 2216 | loss:1.3632115125656128 lr:9.914457251172309e-05 tokens_per_second_per_gpu:3359.2921069951844 mfu:15.846354023031541 +Step 2217 | loss:1.4906959533691406 lr:9.914377365658156e-05 tokens_per_second_per_gpu:3829.4530946612485 mfu:18.06418361958865 +Step 2218 | loss:1.260799765586853 lr:9.914297443182342e-05 tokens_per_second_per_gpu:3392.8791109596737 mfu:16.004789651265828 +Step 2219 | loss:1.456935167312622 lr:9.914217483745472e-05 tokens_per_second_per_gpu:4347.921801242362 mfu:20.5098889684148 +Step 2220 | loss:1.279589295387268 lr:9.914137487348142e-05 tokens_per_second_per_gpu:4644.361322108917 mfu:21.90824476614002 +Step 2221 | loss:1.3716070652008057 lr:9.91405745399096e-05 tokens_per_second_per_gpu:5059.125666494287 mfu:23.864758944700124 +Step 2222 | loss:1.4570659399032593 lr:9.913977383674524e-05 tokens_per_second_per_gpu:3560.903566416334 mfu:16.797389675583137 +Step 2223 | loss:1.5161716938018799 lr:9.913897276399438e-05 tokens_per_second_per_gpu:5329.964164785445 mfu:25.142350350952334 +Step 2224 | loss:1.3213369846343994 lr:9.913817132166302e-05 tokens_per_second_per_gpu:5326.52458143737 mfu:25.12612524944603 +Step 2225 | loss:1.2682398557662964 lr:9.91373695097572e-05 tokens_per_second_per_gpu:5361.493929794452 mfu:25.29108163203245 +Step 2226 | loss:1.3455251455307007 lr:9.913656732828298e-05 tokens_per_second_per_gpu:4111.99434802106 mfu:19.396978918195604 +Step 2227 | loss:1.285656452178955 lr:9.913576477724634e-05 tokens_per_second_per_gpu:5330.227095873562 mfu:25.14359064175565 +Step 2228 | loss:1.3481755256652832 lr:9.913496185665337e-05 tokens_per_second_per_gpu:4649.795638253391 mfu:21.933879362583447 +Step 2229 | loss:1.5169730186462402 lr:9.913415856651005e-05 tokens_per_second_per_gpu:4386.369840495376 mfu:20.6912549294832 +Step 2230 | loss:1.4302749633789062 lr:9.913335490682248e-05 tokens_per_second_per_gpu:5221.403143897897 mfu:24.63024949300564 +Step 2231 | loss:1.403822660446167 lr:9.913255087759666e-05 tokens_per_second_per_gpu:4336.295237917127 mfu:20.455044485513415 +Step 2232 | loss:1.356137752532959 lr:9.913174647883867e-05 tokens_per_second_per_gpu:4186.2855429306455 mfu:19.747423159969014 +Step 2233 | loss:1.4822511672973633 lr:9.913094171055454e-05 tokens_per_second_per_gpu:5347.527629641973 mfu:25.225200211316686 +Step 2234 | loss:1.3434789180755615 lr:9.913013657275032e-05 tokens_per_second_per_gpu:5349.512576636898 mfu:25.23456354495863 +Step 2235 | loss:1.3557945489883423 lr:9.912933106543209e-05 tokens_per_second_per_gpu:5345.93910323689 mfu:25.21770686123336 +Step 2236 | loss:1.313239574432373 lr:9.912852518860588e-05 tokens_per_second_per_gpu:3890.703968601079 mfu:18.353114442439548 +Step 2237 | loss:1.4525818824768066 lr:9.912771894227777e-05 tokens_per_second_per_gpu:3951.6802993429633 mfu:18.640750198183827 +Step 2238 | loss:1.2924683094024658 lr:9.912691232645382e-05 tokens_per_second_per_gpu:4135.83585860312 mfu:19.509443391392345 +Step 2239 | loss:1.41913640499115 lr:9.912610534114008e-05 tokens_per_second_per_gpu:5347.200913376543 mfu:25.223659034949108 +Step 2240 | loss:1.3675462007522583 lr:9.912529798634263e-05 tokens_per_second_per_gpu:5355.8198478885315 mfu:25.264316019584026 +Step 2241 | loss:1.3875046968460083 lr:9.912449026206756e-05 tokens_per_second_per_gpu:4535.2555345283345 mfu:21.393574150755494 +Step 2242 | loss:1.1541575193405151 lr:9.912368216832093e-05 tokens_per_second_per_gpu:4816.4882782519 mfu:22.720196986153507 +Step 2243 | loss:1.2533891201019287 lr:9.912287370510882e-05 tokens_per_second_per_gpu:5063.51173457564 mfu:23.88544877618025 +Step 2244 | loss:1.20427405834198 lr:9.912206487243729e-05 tokens_per_second_per_gpu:5305.670580613918 mfu:25.027753369502374 +Step 2245 | loss:1.3649324178695679 lr:9.912125567031246e-05 tokens_per_second_per_gpu:4484.041257291136 mfu:21.151987666970893 +Step 2246 | loss:1.3218380212783813 lr:9.91204460987404e-05 tokens_per_second_per_gpu:4048.448265283423 mfu:19.097221203842455 +Step 2247 | loss:1.4087077379226685 lr:9.911963615772719e-05 tokens_per_second_per_gpu:5371.814821755564 mfu:25.339767040338764 +Step 2248 | loss:1.3403725624084473 lr:9.911882584727893e-05 tokens_per_second_per_gpu:3864.9691042267677 mfu:18.23171869636529 +Step 2249 | loss:1.3112266063690186 lr:9.911801516740171e-05 tokens_per_second_per_gpu:4908.081286529236 mfu:23.15225683357636 +Step 2250 | loss:1.4252561330795288 lr:9.911720411810163e-05 tokens_per_second_per_gpu:5386.215274657001 mfu:25.407696433645466 +Step 2251 | loss:1.2766667604446411 lr:9.91163926993848e-05 tokens_per_second_per_gpu:5243.373362294725 mfu:24.733886761689266 +Step 2252 | loss:1.330170750617981 lr:9.91155809112573e-05 tokens_per_second_per_gpu:4633.543947265771 mfu:21.857217363370435 +Step 2253 | loss:1.6181849241256714 lr:9.911476875372526e-05 tokens_per_second_per_gpu:4515.151581993934 mfu:21.298740376562726 +Step 2254 | loss:1.2833861112594604 lr:9.911395622679478e-05 tokens_per_second_per_gpu:5328.845393324321 mfu:25.13707291508812 +Step 2255 | loss:1.2314131259918213 lr:9.911314333047195e-05 tokens_per_second_per_gpu:4478.178048428506 mfu:21.12432991039928 +Step 2256 | loss:1.4040025472640991 lr:9.911233006476291e-05 tokens_per_second_per_gpu:3805.1132947638907 mfu:17.94936863067473 +Step 2257 | loss:1.4072856903076172 lr:9.911151642967378e-05 tokens_per_second_per_gpu:3311.847220831478 mfu:15.622548400065035 +Step 2258 | loss:1.5164886713027954 lr:9.911070242521065e-05 tokens_per_second_per_gpu:5325.9373490028 mfu:25.12335517385998 +Step 2259 | loss:1.4194681644439697 lr:9.910988805137968e-05 tokens_per_second_per_gpu:4678.651046904441 mfu:22.069995248430832 +Step 2260 | loss:1.48642897605896 lr:9.910907330818697e-05 tokens_per_second_per_gpu:5106.198469796701 mfu:24.086809389326035 +Step 2261 | loss:1.2924634218215942 lr:9.910825819563864e-05 tokens_per_second_per_gpu:4495.115767726998 mfu:21.20422802220317 +Step 2262 | loss:1.2289183139801025 lr:9.910744271374086e-05 tokens_per_second_per_gpu:4285.853356882397 mfu:20.217101526400235 +Step 2263 | loss:1.4759883880615234 lr:9.91066268624997e-05 tokens_per_second_per_gpu:4852.98718339975 mfu:22.89236854909128 +Step 2264 | loss:1.2629426717758179 lr:9.910581064192135e-05 tokens_per_second_per_gpu:5349.6933925587755 mfu:25.2354164845125 +Step 2265 | loss:1.2521576881408691 lr:9.910499405201195e-05 tokens_per_second_per_gpu:3924.0756666554153 mfu:18.510534435960665 +Step 2266 | loss:1.3068828582763672 lr:9.910417709277761e-05 tokens_per_second_per_gpu:4149.832012659378 mfu:19.575465638065836 +Step 2267 | loss:1.4343383312225342 lr:9.910335976422449e-05 tokens_per_second_per_gpu:4529.910395426751 mfu:21.368360217638557 +Step 2268 | loss:1.2508046627044678 lr:9.910254206635873e-05 tokens_per_second_per_gpu:3931.8609012100537 mfu:18.547258715652877 +Step 2269 | loss:1.4436569213867188 lr:9.910172399918649e-05 tokens_per_second_per_gpu:5336.982919481098 mfu:25.175459014374766 +Step 2270 | loss:1.6201680898666382 lr:9.910090556271392e-05 tokens_per_second_per_gpu:5341.526546529896 mfu:25.19689207838666 +Step 2271 | loss:1.4364057779312134 lr:9.910008675694718e-05 tokens_per_second_per_gpu:3749.7064248255747 mfu:17.688004971788878 +Step 2272 | loss:1.3225302696228027 lr:9.90992675818924e-05 tokens_per_second_per_gpu:4942.73486580946 mfu:23.31572367955148 +Step 2273 | loss:1.2047436237335205 lr:9.909844803755578e-05 tokens_per_second_per_gpu:5001.171040056376 mfu:23.591377083713606 +Step 2274 | loss:1.2155632972717285 lr:9.909762812394346e-05 tokens_per_second_per_gpu:5332.622205455693 mfu:25.154888782302397 +Step 2275 | loss:1.4659874439239502 lr:9.909680784106162e-05 tokens_per_second_per_gpu:5314.388373651113 mfu:25.068876686667693 +Step 2276 | loss:1.399086594581604 lr:9.909598718891643e-05 tokens_per_second_per_gpu:5342.345367856637 mfu:25.200754598290217 +Step 2277 | loss:1.3766478300094604 lr:9.909516616751405e-05 tokens_per_second_per_gpu:5136.572820271628 mfu:24.230090382914852 +Step 2278 | loss:1.2232176065444946 lr:9.909434477686066e-05 tokens_per_second_per_gpu:4333.394472277075 mfu:20.441361079067633 +Step 2279 | loss:1.4209980964660645 lr:9.909352301696243e-05 tokens_per_second_per_gpu:5350.1665803267015 mfu:25.237648591947963 +Step 2280 | loss:1.4584813117980957 lr:9.909270088782557e-05 tokens_per_second_per_gpu:4863.7471223627235 mfu:22.943125017014516 +Step 2281 | loss:1.288742184638977 lr:9.909187838945621e-05 tokens_per_second_per_gpu:4483.594744069476 mfu:21.14988138792131 +Step 2282 | loss:1.4037392139434814 lr:9.90910555218606e-05 tokens_per_second_per_gpu:4673.538161523901 mfu:22.045876895742627 +Step 2283 | loss:1.486325979232788 lr:9.909023228504488e-05 tokens_per_second_per_gpu:5025.893150738028 mfu:23.707995497825994 +Step 2284 | loss:1.4619554281234741 lr:9.908940867901526e-05 tokens_per_second_per_gpu:4094.5754761178923 mfu:19.314811127462352 +Step 2285 | loss:1.4275825023651123 lr:9.908858470377793e-05 tokens_per_second_per_gpu:4378.296466316661 mfu:20.653171446022533 +Step 2286 | loss:1.7139301300048828 lr:9.908776035933911e-05 tokens_per_second_per_gpu:3255.788203733473 mfu:15.358108451759152 +Step 2287 | loss:1.4437686204910278 lr:9.908693564570496e-05 tokens_per_second_per_gpu:3596.508536575732 mfu:16.965344394659343 +Step 2288 | loss:1.3245881795883179 lr:9.90861105628817e-05 tokens_per_second_per_gpu:5333.534453766746 mfu:25.159192013231248 +Step 2289 | loss:1.2832794189453125 lr:9.908528511087556e-05 tokens_per_second_per_gpu:4482.652963303532 mfu:21.145438847363984 +Step 2290 | loss:1.2756987810134888 lr:9.908445928969272e-05 tokens_per_second_per_gpu:5311.874561855678 mfu:25.0570186075281 +Step 2291 | loss:1.35626220703125 lr:9.908363309933939e-05 tokens_per_second_per_gpu:4850.055920196579 mfu:22.878541280436846 +Step 2292 | loss:1.2522085905075073 lr:9.90828065398218e-05 tokens_per_second_per_gpu:5137.844204055665 mfu:24.236087717144592 +Step 2293 | loss:1.429908275604248 lr:9.908197961114616e-05 tokens_per_second_per_gpu:4555.441923700618 mfu:21.488796792634297 +Step 2294 | loss:1.5582326650619507 lr:9.908115231331869e-05 tokens_per_second_per_gpu:3543.9544234694226 mfu:16.717437676480575 +Step 2295 | loss:1.2504569292068481 lr:9.908032464634561e-05 tokens_per_second_per_gpu:5357.983096143521 mfu:25.27452043069102 +Step 2296 | loss:1.4004517793655396 lr:9.907949661023315e-05 tokens_per_second_per_gpu:5339.7274672583535 mfu:25.18840551450822 +Step 2297 | loss:1.3798466920852661 lr:9.907866820498752e-05 tokens_per_second_per_gpu:5020.618136073998 mfu:23.683112353645456 +Step 2298 | loss:1.416965365409851 lr:9.907783943061497e-05 tokens_per_second_per_gpu:5326.131812935048 mfu:25.124272493407364 +Step 2299 | loss:1.46495521068573 lr:9.907701028712173e-05 tokens_per_second_per_gpu:4106.8987684115755 mfu:19.372942199782134 +Step 2300 | loss:1.25246262550354 lr:9.907618077451402e-05 tokens_per_second_per_gpu:4486.08496936454 mfu:21.16162820551437 +Step 2301 | loss:1.5171633958816528 lr:9.907535089279812e-05 tokens_per_second_per_gpu:5355.688460370109 mfu:25.263696242242606 +Step 2302 | loss:1.4090980291366577 lr:9.907452064198022e-05 tokens_per_second_per_gpu:5296.161962747416 mfu:24.982899596688583 +Step 2303 | loss:1.4890023469924927 lr:9.907369002206659e-05 tokens_per_second_per_gpu:3341.317970021211 mfu:15.76156695222131 +Step 2304 | loss:1.4661505222320557 lr:9.907285903306348e-05 tokens_per_second_per_gpu:3297.3263062660794 mfu:15.55405076853655 +Step 2305 | loss:1.5057239532470703 lr:9.907202767497714e-05 tokens_per_second_per_gpu:4446.383595398844 mfu:20.9743500507653 +Step 2306 | loss:1.3819624185562134 lr:9.90711959478138e-05 tokens_per_second_per_gpu:4200.7125387058895 mfu:19.81547776054014 +Step 2307 | loss:1.2355849742889404 lr:9.907036385157975e-05 tokens_per_second_per_gpu:5336.818819728743 mfu:25.174684927844602 +Step 2308 | loss:1.377549409866333 lr:9.906953138628123e-05 tokens_per_second_per_gpu:3833.89321796765 mfu:18.0851284387881 +Step 2309 | loss:1.4027421474456787 lr:9.906869855192451e-05 tokens_per_second_per_gpu:5346.499492217995 mfu:25.22035031166945 +Step 2310 | loss:1.4328064918518066 lr:9.906786534851584e-05 tokens_per_second_per_gpu:4355.071766779125 mfu:20.543616575761625 +Step 2311 | loss:1.472989559173584 lr:9.90670317760615e-05 tokens_per_second_per_gpu:4345.356241662741 mfu:20.497786786148126 +Step 2312 | loss:1.3808404207229614 lr:9.906619783456774e-05 tokens_per_second_per_gpu:4817.855145001866 mfu:22.72664472982461 +Step 2313 | loss:1.244429111480713 lr:9.906536352404085e-05 tokens_per_second_per_gpu:5029.041289420274 mfu:23.72284580511799 +Step 2314 | loss:1.50078284740448 lr:9.906452884448711e-05 tokens_per_second_per_gpu:5327.319504444196 mfu:25.12987503689708 +Step 2315 | loss:1.3842884302139282 lr:9.906369379591277e-05 tokens_per_second_per_gpu:4130.562456386066 mfu:19.4845678533993 +Step 2316 | loss:1.4443538188934326 lr:9.906285837832414e-05 tokens_per_second_per_gpu:4223.160641799197 mfu:19.9213692928726 +Step 2317 | loss:1.3130507469177246 lr:9.90620225917275e-05 tokens_per_second_per_gpu:5328.361876999511 mfu:25.1347920860876 +Step 2318 | loss:1.2539100646972656 lr:9.906118643612911e-05 tokens_per_second_per_gpu:4703.7326957092055 mfu:22.188309665213467 +Step 2319 | loss:1.4752846956253052 lr:9.906034991153528e-05 tokens_per_second_per_gpu:5333.966304264386 mfu:25.161229125709905 +Step 2320 | loss:1.390159249305725 lr:9.905951301795231e-05 tokens_per_second_per_gpu:5318.322534109025 mfu:25.087434792784652 +Step 2321 | loss:1.520365834236145 lr:9.905867575538648e-05 tokens_per_second_per_gpu:4016.2846441086135 mfu:18.945499915081232 +Step 2322 | loss:1.3794244527816772 lr:9.905783812384409e-05 tokens_per_second_per_gpu:3253.3370060401476 mfu:15.346545734022286 +Step 2323 | loss:1.4933812618255615 lr:9.905700012333141e-05 tokens_per_second_per_gpu:5375.498375432454 mfu:25.357142991511722 +Step 2324 | loss:1.3294529914855957 lr:9.905616175385481e-05 tokens_per_second_per_gpu:4526.2025140646665 mfu:21.350869508623806 +Step 2325 | loss:1.4412516355514526 lr:9.905532301542054e-05 tokens_per_second_per_gpu:4620.7814662004075 mfu:21.797014562681305 +Step 2326 | loss:1.468471646308899 lr:9.905448390803493e-05 tokens_per_second_per_gpu:4201.135333812706 mfu:19.81747215719528 +Step 2327 | loss:1.2989073991775513 lr:9.905364443170429e-05 tokens_per_second_per_gpu:4120.512290527075 mfu:19.437159506307324 +Step 2328 | loss:1.2331892251968384 lr:9.905280458643494e-05 tokens_per_second_per_gpu:5148.2830296921375 mfu:24.285329438719142 +Step 2329 | loss:1.4784401655197144 lr:9.905196437223316e-05 tokens_per_second_per_gpu:4085.0475894834776 mfu:19.269866460582733 +Step 2330 | loss:1.2527518272399902 lr:9.905112378910532e-05 tokens_per_second_per_gpu:5324.967071711095 mfu:25.11877821032863 +Step 2331 | loss:1.4361780881881714 lr:9.905028283705771e-05 tokens_per_second_per_gpu:5051.271871008509 mfu:23.827711251398032 +Step 2332 | loss:1.3274482488632202 lr:9.904944151609667e-05 tokens_per_second_per_gpu:4904.120631915853 mfu:23.133573750007667 +Step 2333 | loss:1.3298935890197754 lr:9.904859982622851e-05 tokens_per_second_per_gpu:4236.956183782741 mfu:19.986445218171156 +Step 2334 | loss:1.4322775602340698 lr:9.904775776745958e-05 tokens_per_second_per_gpu:5332.393062739112 mfu:25.153807877012504 +Step 2335 | loss:1.2124367952346802 lr:9.904691533979621e-05 tokens_per_second_per_gpu:3648.5695933188294 mfu:17.210925281848752 +Step 2336 | loss:1.504140019416809 lr:9.904607254324471e-05 tokens_per_second_per_gpu:5383.200879778795 mfu:25.393477018696167 +Step 2337 | loss:1.3956698179244995 lr:9.904522937781146e-05 tokens_per_second_per_gpu:5110.460077732491 mfu:24.10691212106445 +Step 2338 | loss:1.4272178411483765 lr:9.90443858435028e-05 tokens_per_second_per_gpu:5256.1664518737525 mfu:24.794233947959697 +Step 2339 | loss:1.454197883605957 lr:9.904354194032502e-05 tokens_per_second_per_gpu:5243.234870233305 mfu:24.73323347100636 +Step 2340 | loss:1.4885737895965576 lr:9.904269766828452e-05 tokens_per_second_per_gpu:5027.543532193253 mfu:23.715780628734397 +Step 2341 | loss:1.3095910549163818 lr:9.904185302738763e-05 tokens_per_second_per_gpu:4696.435385776034 mfu:22.15388700920924 +Step 2342 | loss:1.436242938041687 lr:9.904100801764071e-05 tokens_per_second_per_gpu:3778.030614837508 mfu:17.821615008680308 +Step 2343 | loss:1.4690473079681396 lr:9.904016263905011e-05 tokens_per_second_per_gpu:5084.637896625036 mfu:23.98510448706218 +Step 2344 | loss:1.3819547891616821 lr:9.903931689162218e-05 tokens_per_second_per_gpu:4500.77012257152 mfu:21.230900578737202 +Step 2345 | loss:1.2023793458938599 lr:9.90384707753633e-05 tokens_per_second_per_gpu:5355.368705012475 mfu:25.262187901664756 +Step 2346 | loss:1.2998429536819458 lr:9.903762429027981e-05 tokens_per_second_per_gpu:5350.986133979805 mfu:25.24151456635993 +Step 2347 | loss:1.2995394468307495 lr:9.90367774363781e-05 tokens_per_second_per_gpu:4302.10591895144 mfu:20.293767634652937 +Step 2348 | loss:1.3551826477050781 lr:9.903593021366454e-05 tokens_per_second_per_gpu:3939.011746187997 mfu:18.580990471473797 +Step 2349 | loss:1.3082280158996582 lr:9.903508262214547e-05 tokens_per_second_per_gpu:3112.905751407869 mfu:14.684107545879106 +Step 2350 | loss:1.3384371995925903 lr:9.903423466182731e-05 tokens_per_second_per_gpu:5321.237180578818 mfu:25.10118367747263 +Step 2351 | loss:1.3284821510314941 lr:9.90333863327164e-05 tokens_per_second_per_gpu:4206.053167618677 mfu:19.84067041832701 +Step 2352 | loss:1.3749629259109497 lr:9.903253763481915e-05 tokens_per_second_per_gpu:4150.03240426332 mfu:19.576410919451963 +Step 2353 | loss:1.2993228435516357 lr:9.903168856814191e-05 tokens_per_second_per_gpu:3883.733174187575 mfu:18.32023201585088 +Step 2354 | loss:1.3421655893325806 lr:9.903083913269109e-05 tokens_per_second_per_gpu:4830.838127000015 mfu:22.78788767103503 +Step 2355 | loss:1.339895486831665 lr:9.902998932847307e-05 tokens_per_second_per_gpu:5039.302168691054 mfu:23.77124812332479 +Step 2356 | loss:1.527738094329834 lr:9.902913915549423e-05 tokens_per_second_per_gpu:4730.373179855798 mfu:22.313977374268244 +Step 2357 | loss:1.4132441282272339 lr:9.902828861376101e-05 tokens_per_second_per_gpu:5370.188393196098 mfu:25.33209489932644 +Step 2358 | loss:1.3671133518218994 lr:9.902743770327976e-05 tokens_per_second_per_gpu:5328.370553698074 mfu:25.134833015555117 +Step 2359 | loss:1.2876272201538086 lr:9.90265864240569e-05 tokens_per_second_per_gpu:4466.761239273336 mfu:21.07047487370584 +Step 2360 | loss:1.3452178239822388 lr:9.90257347760988e-05 tokens_per_second_per_gpu:5375.203336079903 mfu:25.355751240546834 +Step 2361 | loss:1.3269178867340088 lr:9.902488275941193e-05 tokens_per_second_per_gpu:4891.6878657259895 mfu:23.074926270641434 +Step 2362 | loss:1.421006441116333 lr:9.902403037400264e-05 tokens_per_second_per_gpu:5154.633303494536 mfu:24.315284763713493 +Step 2363 | loss:1.36953604221344 lr:9.902317761987737e-05 tokens_per_second_per_gpu:5250.873557035391 mfu:24.769266459946163 +Step 2364 | loss:1.3035836219787598 lr:9.902232449704253e-05 tokens_per_second_per_gpu:4441.522137605111 mfu:20.951417724902033 +Step 2365 | loss:1.3112901449203491 lr:9.902147100550452e-05 tokens_per_second_per_gpu:5026.428954230491 mfu:23.710522974317445 +Step 2366 | loss:1.388155460357666 lr:9.902061714526978e-05 tokens_per_second_per_gpu:5369.857149835825 mfu:25.330532367880085 +Step 2367 | loss:1.4081956148147583 lr:9.901976291634472e-05 tokens_per_second_per_gpu:3937.4987937080195 mfu:18.573853616489433 +Step 2368 | loss:1.4334218502044678 lr:9.901890831873577e-05 tokens_per_second_per_gpu:3553.9870370613394 mfu:16.764763226534033 +Step 2369 | loss:1.5048853158950806 lr:9.901805335244936e-05 tokens_per_second_per_gpu:5348.944006315993 mfu:25.231881501746816 +Step 2370 | loss:1.3496222496032715 lr:9.901719801749191e-05 tokens_per_second_per_gpu:5322.252313586855 mfu:25.105972233071093 +Step 2371 | loss:1.225339412689209 lr:9.901634231386987e-05 tokens_per_second_per_gpu:5356.6524344433765 mfu:25.26824347241714 +Step 2372 | loss:1.3826642036437988 lr:9.901548624158965e-05 tokens_per_second_per_gpu:5343.033221791274 mfu:25.203999322659882 +Step 2373 | loss:1.2501558065414429 lr:9.90146298006577e-05 tokens_per_second_per_gpu:4069.2674073231883 mfu:19.195428648955943 +Step 2374 | loss:1.3073745965957642 lr:9.901377299108048e-05 tokens_per_second_per_gpu:3505.6289255188876 mfu:16.53664976364367 +Step 2375 | loss:1.3586548566818237 lr:9.901291581286441e-05 tokens_per_second_per_gpu:4137.228801071736 mfu:19.516014138677296 +Step 2376 | loss:1.3213238716125488 lr:9.901205826601593e-05 tokens_per_second_per_gpu:4788.241500892359 mfu:22.586952117951373 +Step 2377 | loss:1.5161585807800293 lr:9.901120035054154e-05 tokens_per_second_per_gpu:5255.191435278382 mfu:24.78963462832435 +Step 2378 | loss:1.3304847478866577 lr:9.901034206644764e-05 tokens_per_second_per_gpu:3889.4058857912196 mfu:18.346991164349674 +Step 2379 | loss:1.4761040210723877 lr:9.90094834137407e-05 tokens_per_second_per_gpu:5385.109204837781 mfu:25.402478913593235 +Step 2380 | loss:1.44933021068573 lr:9.900862439242719e-05 tokens_per_second_per_gpu:4796.444128987341 mfu:22.625645314188073 +Step 2381 | loss:1.337471604347229 lr:9.900776500251356e-05 tokens_per_second_per_gpu:5355.745236935683 mfu:25.263964066989498 +Step 2382 | loss:1.2696599960327148 lr:9.900690524400628e-05 tokens_per_second_per_gpu:5345.287275999461 mfu:25.21463208094121 +Step 2383 | loss:1.3884272575378418 lr:9.90060451169118e-05 tokens_per_second_per_gpu:3625.455224368688 mfu:17.10189085979293 +Step 2384 | loss:1.416945457458496 lr:9.900518462123662e-05 tokens_per_second_per_gpu:4262.8555461422875 mfu:20.108616929308013 +Step 2385 | loss:1.3604826927185059 lr:9.900432375698718e-05 tokens_per_second_per_gpu:4747.622575345735 mfu:22.39534571584462 +Step 2386 | loss:1.3160374164581299 lr:9.900346252416998e-05 tokens_per_second_per_gpu:5326.057020659352 mfu:25.123919685482026 +Step 2387 | loss:1.4934628009796143 lr:9.900260092279148e-05 tokens_per_second_per_gpu:5345.8497035003165 mfu:25.21728514742277 +Step 2388 | loss:1.1215318441390991 lr:9.900173895285816e-05 tokens_per_second_per_gpu:4894.74658030641 mfu:23.089354749187798 +Step 2389 | loss:1.3836230039596558 lr:9.900087661437651e-05 tokens_per_second_per_gpu:4935.256912604521 mfu:23.2804488984134 +Step 2390 | loss:1.3892779350280762 lr:9.9000013907353e-05 tokens_per_second_per_gpu:4005.7764149671098 mfu:18.895930805331993 +Step 2391 | loss:1.5270310640335083 lr:9.899915083179416e-05 tokens_per_second_per_gpu:5336.735959308195 mfu:25.17429406110343 +Step 2392 | loss:1.334453821182251 lr:9.899828738770645e-05 tokens_per_second_per_gpu:3518.8873675595482 mfu:16.59919209687294 +Step 2393 | loss:1.5006322860717773 lr:9.899742357509636e-05 tokens_per_second_per_gpu:3358.8084002593673 mfu:15.844072295829813 +Step 2394 | loss:1.3769631385803223 lr:9.89965593939704e-05 tokens_per_second_per_gpu:5052.8874708853145 mfu:23.835332311666683 +Step 2395 | loss:1.3548896312713623 lr:9.899569484433506e-05 tokens_per_second_per_gpu:3958.4047914069156 mfu:18.672470774566143 +Step 2396 | loss:1.2245017290115356 lr:9.899482992619684e-05 tokens_per_second_per_gpu:5399.795454711741 mfu:25.47175645998131 +Step 2397 | loss:1.3006452322006226 lr:9.899396463956226e-05 tokens_per_second_per_gpu:5097.155486615395 mfu:24.044152094767785 +Step 2398 | loss:1.5512348413467407 lr:9.899309898443783e-05 tokens_per_second_per_gpu:4579.9090767978405 mfu:21.60421252832068 +Step 2399 | loss:1.2817189693450928 lr:9.899223296083005e-05 tokens_per_second_per_gpu:4039.502354373579 mfu:19.05502181575093 +Step 2400 | loss:1.4831465482711792 lr:9.899136656874541e-05 tokens_per_second_per_gpu:4946.775181315511 mfu:23.33478253714243 +Step 2401 | loss:1.2628049850463867 lr:9.899049980819047e-05 tokens_per_second_per_gpu:4042.79946210717 mfu:19.070574835474865 +Step 2402 | loss:1.2273629903793335 lr:9.898963267917173e-05 tokens_per_second_per_gpu:5336.473442767161 mfu:25.173055725789435 +Step 2403 | loss:1.4433287382125854 lr:9.898876518169571e-05 tokens_per_second_per_gpu:3163.1548389960712 mfu:14.921141065411055 +Step 2404 | loss:1.3375766277313232 lr:9.898789731576894e-05 tokens_per_second_per_gpu:4047.475692697285 mfu:19.092633413015648 +Step 2405 | loss:1.470647931098938 lr:9.898702908139793e-05 tokens_per_second_per_gpu:3738.542964532491 mfu:17.635345024903675 +Step 2406 | loss:1.2773371934890747 lr:9.898616047858923e-05 tokens_per_second_per_gpu:3275.8399193796827 mfu:15.452695815637794 +Step 2407 | loss:1.465851068496704 lr:9.898529150734936e-05 tokens_per_second_per_gpu:2982.6519541214743 mfu:14.06967816048974 +Step 2408 | loss:1.3256847858428955 lr:9.898442216768488e-05 tokens_per_second_per_gpu:3966.2919909991633 mfu:18.70967604073779 +Step 2409 | loss:1.3206650018692017 lr:9.898355245960229e-05 tokens_per_second_per_gpu:5312.991392723629 mfu:25.06228688928323 +Step 2410 | loss:1.3451699018478394 lr:9.898268238310817e-05 tokens_per_second_per_gpu:5048.360488431707 mfu:23.81397776305017 +Step 2411 | loss:1.240430474281311 lr:9.898181193820903e-05 tokens_per_second_per_gpu:5313.683833303658 mfu:25.065553249641763 +Step 2412 | loss:1.4250743389129639 lr:9.898094112491146e-05 tokens_per_second_per_gpu:3221.1184324156925 mfu:15.194565225180233 +Step 2413 | loss:1.2891227006912231 lr:9.898006994322195e-05 tokens_per_second_per_gpu:4203.1243406528665 mfu:19.826854641819875 +Step 2414 | loss:1.3699419498443604 lr:9.897919839314709e-05 tokens_per_second_per_gpu:4152.305169759287 mfu:19.587131942070222 +Step 2415 | loss:1.3966869115829468 lr:9.897832647469344e-05 tokens_per_second_per_gpu:4332.60818881299 mfu:20.437652045813753 +Step 2416 | loss:1.4308160543441772 lr:9.897745418786754e-05 tokens_per_second_per_gpu:5309.861053187363 mfu:25.047520543599656 +Step 2417 | loss:1.5400686264038086 lr:9.897658153267596e-05 tokens_per_second_per_gpu:4206.00016089554 mfu:19.840420376570094 +Step 2418 | loss:1.3846664428710938 lr:9.897570850912525e-05 tokens_per_second_per_gpu:5298.157172374653 mfu:24.99231134091813 +Step 2419 | loss:1.364471673965454 lr:9.8974835117222e-05 tokens_per_second_per_gpu:3979.7722952519684 mfu:18.773264935875346 +Step 2420 | loss:1.5597585439682007 lr:9.897396135697276e-05 tokens_per_second_per_gpu:4958.827025762368 mfu:23.391633143654953 +Step 2421 | loss:1.2022666931152344 lr:9.897308722838412e-05 tokens_per_second_per_gpu:5200.475291538071 mfu:24.53152924276453 +Step 2422 | loss:1.2785863876342773 lr:9.897221273146263e-05 tokens_per_second_per_gpu:5336.672735466802 mfu:25.17399582345648 +Step 2423 | loss:1.2647160291671753 lr:9.897133786621489e-05 tokens_per_second_per_gpu:5343.658543519101 mfu:25.206949071940933 +Step 2424 | loss:1.4534904956817627 lr:9.897046263264745e-05 tokens_per_second_per_gpu:3784.701563588693 mfu:17.853083012120866 +Step 2425 | loss:1.405752182006836 lr:9.896958703076693e-05 tokens_per_second_per_gpu:3828.7603992301497 mfu:18.060916057054133 +Step 2426 | loss:1.2882211208343506 lr:9.896871106057989e-05 tokens_per_second_per_gpu:4796.251245427025 mfu:22.62473544952473 +Step 2427 | loss:1.387313723564148 lr:9.896783472209292e-05 tokens_per_second_per_gpu:4339.897702018345 mfu:20.472037923322613 +Step 2428 | loss:1.3133355379104614 lr:9.896695801531262e-05 tokens_per_second_per_gpu:5326.708946062678 mfu:25.126994928839455 +Step 2429 | loss:1.274561882019043 lr:9.896608094024558e-05 tokens_per_second_per_gpu:4304.472328773263 mfu:20.30493039353363 +Step 2430 | loss:1.4811246395111084 lr:9.896520349689841e-05 tokens_per_second_per_gpu:4414.534041959868 mfu:20.824110273099485 +Step 2431 | loss:1.4104703664779663 lr:9.896432568527768e-05 tokens_per_second_per_gpu:3959.5885853801265 mfu:18.678054932713856 +Step 2432 | loss:1.4475600719451904 lr:9.896344750539001e-05 tokens_per_second_per_gpu:5079.492295131574 mfu:23.960831806887377 +Step 2433 | loss:1.2408692836761475 lr:9.8962568957242e-05 tokens_per_second_per_gpu:5337.286260857072 mfu:25.176889927400612 +Step 2434 | loss:1.4165446758270264 lr:9.896169004084028e-05 tokens_per_second_per_gpu:4605.777227915421 mfu:21.726237010703997 +Step 2435 | loss:1.3847066164016724 lr:9.896081075619143e-05 tokens_per_second_per_gpu:4356.395581681774 mfu:20.549861236523533 +Step 2436 | loss:1.19801664352417 lr:9.895993110330208e-05 tokens_per_second_per_gpu:5315.9691046624985 mfu:25.076333264548918 +Step 2437 | loss:1.40044105052948 lr:9.895905108217885e-05 tokens_per_second_per_gpu:5344.282962835484 mfu:25.209894564393235 +Step 2438 | loss:1.6088852882385254 lr:9.895817069282833e-05 tokens_per_second_per_gpu:4660.508289810912 mfu:21.984412767747102 +Step 2439 | loss:1.4325329065322876 lr:9.895728993525717e-05 tokens_per_second_per_gpu:4492.81251287999 mfu:21.193363176114037 +Step 2440 | loss:1.3612122535705566 lr:9.895640880947198e-05 tokens_per_second_per_gpu:5319.199489000383 mfu:25.091571538631122 +Step 2441 | loss:1.3495701551437378 lr:9.895552731547938e-05 tokens_per_second_per_gpu:5300.1361180638505 mfu:25.001646365377024 +Step 2442 | loss:1.3172250986099243 lr:9.895464545328604e-05 tokens_per_second_per_gpu:4208.888934359713 mfu:19.85404721387618 +Step 2443 | loss:1.513741135597229 lr:9.895376322289855e-05 tokens_per_second_per_gpu:4540.338521066123 mfu:21.417551465501173 +Step 2444 | loss:1.2599235773086548 lr:9.895288062432357e-05 tokens_per_second_per_gpu:4943.943818238437 mfu:23.321426514427518 +Step 2445 | loss:1.415340542793274 lr:9.895199765756772e-05 tokens_per_second_per_gpu:5382.038165873362 mfu:25.38799229882479 +Step 2446 | loss:1.4323946237564087 lr:9.895111432263766e-05 tokens_per_second_per_gpu:3976.0056456808134 mfu:18.75549700719152 +Step 2447 | loss:1.4247512817382812 lr:9.895023061954e-05 tokens_per_second_per_gpu:4271.166006999391 mfu:20.147818791081804 +Step 2448 | loss:1.5368554592132568 lr:9.894934654828144e-05 tokens_per_second_per_gpu:2954.0399211034564 mfu:13.93471031902753 +Step 2449 | loss:1.3427590131759644 lr:9.894846210886856e-05 tokens_per_second_per_gpu:4993.155054630776 mfu:23.55356431279291 +Step 2450 | loss:1.4623631238937378 lr:9.894757730130808e-05 tokens_per_second_per_gpu:4403.181112595864 mfu:20.770556568279993 +Step 2451 | loss:1.4483858346939087 lr:9.894669212560661e-05 tokens_per_second_per_gpu:5318.52674584957 mfu:25.088398094406916 +Step 2452 | loss:1.3425015211105347 lr:9.894580658177084e-05 tokens_per_second_per_gpu:3219.7040498952456 mfu:15.187893341512742 +Step 2453 | loss:1.3209638595581055 lr:9.894492066980741e-05 tokens_per_second_per_gpu:4923.057185637865 mfu:23.222900704823626 +Step 2454 | loss:1.199514389038086 lr:9.894403438972297e-05 tokens_per_second_per_gpu:5340.737126649355 mfu:25.193168250121225 +Step 2455 | loss:1.3884220123291016 lr:9.89431477415242e-05 tokens_per_second_per_gpu:3904.6011706893405 mfu:18.41866991579705 +Step 2456 | loss:1.4114559888839722 lr:9.89422607252178e-05 tokens_per_second_per_gpu:5335.357040804506 mfu:25.167789467253776 +Step 2457 | loss:1.4167063236236572 lr:9.894137334081039e-05 tokens_per_second_per_gpu:5294.064049146086 mfu:24.973003380290145 +Step 2458 | loss:1.1382776498794556 lr:9.894048558830867e-05 tokens_per_second_per_gpu:5243.278980461317 mfu:24.733441546477877 +Step 2459 | loss:1.394051432609558 lr:9.893959746771931e-05 tokens_per_second_per_gpu:5384.6326873094 mfu:25.400231099109586 +Step 2460 | loss:1.2322503328323364 lr:9.8938708979049e-05 tokens_per_second_per_gpu:4403.01767210328 mfu:20.769785591590654 +Step 2461 | loss:1.4577851295471191 lr:9.893782012230441e-05 tokens_per_second_per_gpu:4884.240161746088 mfu:23.039794180258912 +Step 2462 | loss:1.3625942468643188 lr:9.893693089749223e-05 tokens_per_second_per_gpu:3908.3656183025346 mfu:18.43642745746948 +Step 2463 | loss:1.3980317115783691 lr:9.893604130461915e-05 tokens_per_second_per_gpu:4830.355120013113 mfu:22.78560924466832 +Step 2464 | loss:1.282510757446289 lr:9.893515134369187e-05 tokens_per_second_per_gpu:3267.3069825882008 mfu:15.412444496923232 +Step 2465 | loss:1.3099740743637085 lr:9.893426101471705e-05 tokens_per_second_per_gpu:4781.9519296711105 mfu:22.557283137389238 +Step 2466 | loss:1.299239158630371 lr:9.893337031770143e-05 tokens_per_second_per_gpu:5178.993360900677 mfu:24.43019531075263 +Step 2467 | loss:1.3807097673416138 lr:9.893247925265168e-05 tokens_per_second_per_gpu:4570.778306933068 mfu:21.561141128998564 +Step 2468 | loss:1.3611173629760742 lr:9.89315878195745e-05 tokens_per_second_per_gpu:5326.002402001681 mfu:25.123662039954887 +Step 2469 | loss:1.404395580291748 lr:9.893069601847662e-05 tokens_per_second_per_gpu:4246.672733316847 mfu:20.032279840137832 +Step 2470 | loss:1.2090301513671875 lr:9.892980384936472e-05 tokens_per_second_per_gpu:5298.876719839828 mfu:24.995705569078815 +Step 2471 | loss:1.416959285736084 lr:9.892891131224553e-05 tokens_per_second_per_gpu:5141.663011118099 mfu:24.25410168161377 +Step 2472 | loss:1.3075945377349854 lr:9.892801840712576e-05 tokens_per_second_per_gpu:4453.983669294784 mfu:21.010200896038768 +Step 2473 | loss:1.3486835956573486 lr:9.89271251340121e-05 tokens_per_second_per_gpu:5317.442065553736 mfu:25.083281472385924 +Step 2474 | loss:1.4182829856872559 lr:9.89262314929113e-05 tokens_per_second_per_gpu:5177.471193462163 mfu:24.423014987236616 +Step 2475 | loss:1.328498363494873 lr:9.892533748383007e-05 tokens_per_second_per_gpu:4484.445191400001 mfu:21.153893093080672 +Step 2476 | loss:1.2817442417144775 lr:9.892444310677514e-05 tokens_per_second_per_gpu:5348.849012400827 mfu:25.231433399241343 +Step 2477 | loss:1.4377480745315552 lr:9.892354836175321e-05 tokens_per_second_per_gpu:5335.390835177757 mfu:25.167948881078313 +Step 2478 | loss:1.4890851974487305 lr:9.892265324877106e-05 tokens_per_second_per_gpu:5335.506253835815 mfu:25.16849333058083 +Step 2479 | loss:1.1824452877044678 lr:9.892175776783536e-05 tokens_per_second_per_gpu:4720.669229785707 mfu:22.268202186123013 +Step 2480 | loss:1.3718308210372925 lr:9.892086191895291e-05 tokens_per_second_per_gpu:4709.986465251778 mfu:22.217809763148892 +Step 2481 | loss:1.4654786586761475 lr:9.89199657021304e-05 tokens_per_second_per_gpu:5318.870043229678 mfu:25.09001748672146 +Step 2482 | loss:1.3396894931793213 lr:9.891906911737459e-05 tokens_per_second_per_gpu:5075.547072640585 mfu:23.942221519273016 +Step 2483 | loss:1.593559980392456 lr:9.89181721646922e-05 tokens_per_second_per_gpu:3426.749907063356 mfu:16.16456397544062 +Step 2484 | loss:1.357062578201294 lr:9.891727484409001e-05 tokens_per_second_per_gpu:5256.200498568549 mfu:24.794394552028017 +Step 2485 | loss:1.3278931379318237 lr:9.891637715557476e-05 tokens_per_second_per_gpu:4788.605984078768 mfu:22.58867144732891 +Step 2486 | loss:1.3090366125106812 lr:9.891547909915318e-05 tokens_per_second_per_gpu:5353.127938383109 mfu:25.251617823159823 +Step 2487 | loss:1.3725550174713135 lr:9.891458067483205e-05 tokens_per_second_per_gpu:5050.908598166316 mfu:23.825997631420663 +Step 2488 | loss:1.2717037200927734 lr:9.891368188261811e-05 tokens_per_second_per_gpu:5311.128871623767 mfu:25.053501059476307 +Step 2489 | loss:1.3004193305969238 lr:9.891278272251814e-05 tokens_per_second_per_gpu:5315.047893317257 mfu:25.07198775345855 +Step 2490 | loss:1.3819234371185303 lr:9.891188319453888e-05 tokens_per_second_per_gpu:5321.8118564641845 mfu:25.103894521673723 +Step 2491 | loss:1.371935248374939 lr:9.891098329868712e-05 tokens_per_second_per_gpu:4592.626388735134 mfu:21.66420225852605 +Step 2492 | loss:1.259489893913269 lr:9.89100830349696e-05 tokens_per_second_per_gpu:3671.2330111568162 mfu:17.31783249056835 +Step 2493 | loss:1.1995176076889038 lr:9.890918240339311e-05 tokens_per_second_per_gpu:5316.138911232734 mfu:25.07713427111271 +Step 2494 | loss:1.3876975774765015 lr:9.89082814039644e-05 tokens_per_second_per_gpu:4379.279038547056 mfu:20.657806406877413 +Step 2495 | loss:1.3093963861465454 lr:9.890738003669029e-05 tokens_per_second_per_gpu:5365.733882132058 mfu:25.311082210619407 +Step 2496 | loss:1.4314217567443848 lr:9.890647830157752e-05 tokens_per_second_per_gpu:3967.411144166323 mfu:18.714955277174504 +Step 2497 | loss:1.2201228141784668 lr:9.890557619863289e-05 tokens_per_second_per_gpu:3269.4566608460755 mfu:15.422584896038558 +Step 2498 | loss:1.4748255014419556 lr:9.890467372786319e-05 tokens_per_second_per_gpu:5337.614366004856 mfu:25.178437655363215 +Step 2499 | loss:1.4113249778747559 lr:9.890377088927518e-05 tokens_per_second_per_gpu:4576.947355017176 mfu:21.590241581359443 +Step 2500 | loss:1.424532175064087 lr:9.890286768287567e-05 tokens_per_second_per_gpu:3860.240071974712 mfu:18.20941104437624 +Step 2501 | loss:1.4221293926239014 lr:9.890196410867147e-05 tokens_per_second_per_gpu:3932.7176127978328 mfu:18.551299970382498 +Step 2502 | loss:1.3329315185546875 lr:9.890106016666934e-05 tokens_per_second_per_gpu:5334.8329244177 mfu:25.165317120833734 +Step 2503 | loss:1.3320621252059937 lr:9.89001558568761e-05 tokens_per_second_per_gpu:4088.454696763409 mfu:19.285938366934612 +Step 2504 | loss:1.1604342460632324 lr:9.889925117929857e-05 tokens_per_second_per_gpu:5336.148706945074 mfu:25.171523891511004 +Step 2505 | loss:1.454651951789856 lr:9.88983461339435e-05 tokens_per_second_per_gpu:5192.040770363493 mfu:24.491742167306338 +Step 2506 | loss:1.2955424785614014 lr:9.889744072081776e-05 tokens_per_second_per_gpu:4858.957572049879 mfu:22.92053189100728 +Step 2507 | loss:1.2968252897262573 lr:9.889653493992811e-05 tokens_per_second_per_gpu:5346.125554000153 mfu:25.21858638129545 +Step 2508 | loss:1.194759726524353 lr:9.889562879128138e-05 tokens_per_second_per_gpu:4920.181726542525 mfu:23.209336673666996 +Step 2509 | loss:1.2860854864120483 lr:9.88947222748844e-05 tokens_per_second_per_gpu:4395.435608253933 mfu:20.734019702779793 +Step 2510 | loss:1.4676610231399536 lr:9.889381539074396e-05 tokens_per_second_per_gpu:3867.145426803714 mfu:18.24198478127967 +Step 2511 | loss:1.3993970155715942 lr:9.889290813886691e-05 tokens_per_second_per_gpu:4484.671868897924 mfu:21.154962369513544 +Step 2512 | loss:1.3716819286346436 lr:9.889200051926004e-05 tokens_per_second_per_gpu:4530.220869399424 mfu:21.369824776339975 +Step 2513 | loss:1.3918626308441162 lr:9.889109253193022e-05 tokens_per_second_per_gpu:4135.758224676912 mfu:19.509077178916705 +Step 2514 | loss:1.5032879114151 lr:9.889018417688424e-05 tokens_per_second_per_gpu:4024.4887033188043 mfu:18.98419986213254 +Step 2515 | loss:1.280495524406433 lr:9.888927545412894e-05 tokens_per_second_per_gpu:5365.220606668258 mfu:25.308661002682836 +Step 2516 | loss:1.538743257522583 lr:9.888836636367118e-05 tokens_per_second_per_gpu:3423.359450468705 mfu:16.148570613212105 +Step 2517 | loss:1.2522292137145996 lr:9.888745690551776e-05 tokens_per_second_per_gpu:4850.676239968823 mfu:22.88146743463999 +Step 2518 | loss:1.3727312088012695 lr:9.888654707967555e-05 tokens_per_second_per_gpu:4109.094068403804 mfu:19.38329780440228 +Step 2519 | loss:1.2865115404129028 lr:9.888563688615137e-05 tokens_per_second_per_gpu:5328.662106981469 mfu:25.136208322137012 +Step 2520 | loss:1.2832040786743164 lr:9.88847263249521e-05 tokens_per_second_per_gpu:5339.5732565172175 mfu:25.18767807613121 +Step 2521 | loss:1.2189356088638306 lr:9.888381539608454e-05 tokens_per_second_per_gpu:5371.435912196104 mfu:25.337979659298156 +Step 2522 | loss:1.5810129642486572 lr:9.888290409955559e-05 tokens_per_second_per_gpu:5351.424529936481 mfu:25.24358255488563 +Step 2523 | loss:1.1815904378890991 lr:9.888199243537206e-05 tokens_per_second_per_gpu:4675.6766930954 mfu:22.0559647140365 +Step 2524 | loss:1.1691045761108398 lr:9.888108040354083e-05 tokens_per_second_per_gpu:4758.54104445791 mfu:22.446849997529675 +Step 2525 | loss:1.3300496339797974 lr:9.888016800406876e-05 tokens_per_second_per_gpu:5170.414753660496 mfu:24.389728556739374 +Step 2526 | loss:1.3340342044830322 lr:9.887925523696271e-05 tokens_per_second_per_gpu:3836.327371344842 mfu:18.096610755577707 +Step 2527 | loss:1.270477533340454 lr:9.887834210222954e-05 tokens_per_second_per_gpu:5361.712850502072 mfu:25.292114318362017 +Step 2528 | loss:1.377009391784668 lr:9.887742859987614e-05 tokens_per_second_per_gpu:2963.821182595417 mfu:13.980850198340406 +Step 2529 | loss:1.2865527868270874 lr:9.887651472990935e-05 tokens_per_second_per_gpu:4936.326514092876 mfu:23.285494391126804 +Step 2530 | loss:1.2563302516937256 lr:9.887560049233605e-05 tokens_per_second_per_gpu:4966.3354376626985 mfu:23.42705160365598 +Step 2531 | loss:1.4098429679870605 lr:9.887468588716313e-05 tokens_per_second_per_gpu:2998.5430071982605 mfu:14.144638969146175 +Step 2532 | loss:1.532485008239746 lr:9.887377091439746e-05 tokens_per_second_per_gpu:4428.037625373438 mfu:20.8878089800102 +Step 2533 | loss:1.4524728059768677 lr:9.887285557404593e-05 tokens_per_second_per_gpu:4194.785468827216 mfu:19.787518760660042 +Step 2534 | loss:1.225804090499878 lr:9.88719398661154e-05 tokens_per_second_per_gpu:4617.8909484833375 mfu:21.783379497436712 +Step 2535 | loss:1.4458528757095337 lr:9.88710237906128e-05 tokens_per_second_per_gpu:4262.498755129734 mfu:20.106933885226972 +Step 2536 | loss:1.3105710744857788 lr:9.887010734754496e-05 tokens_per_second_per_gpu:5354.6436584518015 mfu:25.258767733332355 +Step 2537 | loss:1.3560465574264526 lr:9.886919053691883e-05 tokens_per_second_per_gpu:4459.033449561062 mfu:21.034021571136158 +Step 2538 | loss:1.4562745094299316 lr:9.886827335874127e-05 tokens_per_second_per_gpu:4497.409272065227 mfu:21.21504687347879 +Step 2539 | loss:1.4168176651000977 lr:9.88673558130192e-05 tokens_per_second_per_gpu:5376.433800848723 mfu:25.361555552800294 +Step 2540 | loss:1.3467239141464233 lr:9.886643789975952e-05 tokens_per_second_per_gpu:4674.888824229484 mfu:22.05224820217168 +Step 2541 | loss:1.4443196058273315 lr:9.88655196189691e-05 tokens_per_second_per_gpu:4138.434712112707 mfu:19.521702626807194 +Step 2542 | loss:1.264060139656067 lr:9.886460097065488e-05 tokens_per_second_per_gpu:5385.466498309025 mfu:25.404164327857643 +Step 2543 | loss:1.6847294569015503 lr:9.886368195482377e-05 tokens_per_second_per_gpu:4704.720373940103 mfu:22.192968711943895 +Step 2544 | loss:1.3652151823043823 lr:9.886276257148267e-05 tokens_per_second_per_gpu:5229.581151962409 mfu:24.668826552358 +Step 2545 | loss:1.414838194847107 lr:9.88618428206385e-05 tokens_per_second_per_gpu:3924.1939174842996 mfu:18.511092245296183 +Step 2546 | loss:1.4296953678131104 lr:9.886092270229817e-05 tokens_per_second_per_gpu:4610.442859960353 mfu:21.748245593098854 +Step 2547 | loss:1.4036668539047241 lr:9.886000221646861e-05 tokens_per_second_per_gpu:3773.84541843738 mfu:17.801872723192766 +Step 2548 | loss:1.4491462707519531 lr:9.885908136315674e-05 tokens_per_second_per_gpu:3029.522569243879 mfu:14.290774848973683 +Step 2549 | loss:1.4010825157165527 lr:9.885816014236948e-05 tokens_per_second_per_gpu:2942.512892873899 mfu:13.880335360154891 +Step 2550 | loss:1.3769750595092773 lr:9.885723855411377e-05 tokens_per_second_per_gpu:5360.196417169929 mfu:25.284961043604095 +Step 2551 | loss:1.4882088899612427 lr:9.885631659839651e-05 tokens_per_second_per_gpu:4919.334024437895 mfu:23.205337918206563 +Step 2552 | loss:1.4817248582839966 lr:9.88553942752247e-05 tokens_per_second_per_gpu:5341.104943941628 mfu:25.19490330704498 +Step 2553 | loss:1.4549137353897095 lr:9.88544715846052e-05 tokens_per_second_per_gpu:4167.7948462934955 mfu:19.660199389093414 +Step 2554 | loss:1.2984364032745361 lr:9.8853548526545e-05 tokens_per_second_per_gpu:3659.62288780589 mfu:17.263065557830725 +Step 2555 | loss:1.4332603216171265 lr:9.885262510105102e-05 tokens_per_second_per_gpu:5326.524897537125 mfu:25.126126740542507 +Step 2556 | loss:1.378706693649292 lr:9.885170130813023e-05 tokens_per_second_per_gpu:3766.956304631117 mfu:17.769375597964604 +Step 2557 | loss:1.3178173303604126 lr:9.885077714778954e-05 tokens_per_second_per_gpu:5320.559350349202 mfu:25.09798623662977 +Step 2558 | loss:1.317965030670166 lr:9.884985262003595e-05 tokens_per_second_per_gpu:4184.961078494331 mfu:19.74117543524598 +Step 2559 | loss:1.302720308303833 lr:9.884892772487634e-05 tokens_per_second_per_gpu:5328.588826963485 mfu:25.135862647789324 +Step 2560 | loss:1.3741329908370972 lr:9.884800246231776e-05 tokens_per_second_per_gpu:4976.294262984567 mfu:23.47402908185004 +Step 2561 | loss:1.2452199459075928 lr:9.884707683236709e-05 tokens_per_second_per_gpu:4490.207839018187 mfu:21.181076485104725 +Step 2562 | loss:1.3741180896759033 lr:9.884615083503134e-05 tokens_per_second_per_gpu:4953.784188484595 mfu:23.367845219818932 +Step 2563 | loss:1.4461069107055664 lr:9.884522447031745e-05 tokens_per_second_per_gpu:4823.2912833342925 mfu:22.75228792183951 +Step 2564 | loss:1.4510352611541748 lr:9.884429773823239e-05 tokens_per_second_per_gpu:4044.2710385004993 mfu:19.077516512399022 +Step 2565 | loss:1.2917340993881226 lr:9.884337063878312e-05 tokens_per_second_per_gpu:4318.534423228856 mfu:20.371263692321378 +Step 2566 | loss:1.3407113552093506 lr:9.884244317197666e-05 tokens_per_second_per_gpu:5095.589478914506 mfu:24.03676496925423 +Step 2567 | loss:1.5747759342193604 lr:9.884151533781993e-05 tokens_per_second_per_gpu:5332.829547261031 mfu:25.155866848974334 +Step 2568 | loss:1.4081019163131714 lr:9.884058713631995e-05 tokens_per_second_per_gpu:4082.4571850754924 mfu:19.257647081021698 +Step 2569 | loss:1.3818734884262085 lr:9.883965856748366e-05 tokens_per_second_per_gpu:4735.43470993266 mfu:22.337853475226876 +Step 2570 | loss:1.4549236297607422 lr:9.883872963131807e-05 tokens_per_second_per_gpu:5358.667549706362 mfu:25.27774911492676 +Step 2571 | loss:1.2969508171081543 lr:9.883780032783016e-05 tokens_per_second_per_gpu:5202.732791595637 mfu:24.542178255705295 +Step 2572 | loss:1.3147375583648682 lr:9.883687065702693e-05 tokens_per_second_per_gpu:4393.47941567361 mfu:20.724792008617538 +Step 2573 | loss:1.2541062831878662 lr:9.883594061891537e-05 tokens_per_second_per_gpu:4556.651604976184 mfu:21.49450306560405 +Step 2574 | loss:1.2939385175704956 lr:9.883501021350246e-05 tokens_per_second_per_gpu:4373.855168471437 mfu:20.632221086323053 +Step 2575 | loss:1.3636486530303955 lr:9.883407944079522e-05 tokens_per_second_per_gpu:4958.833545821773 mfu:23.391663899887604 +Step 2576 | loss:1.350821852684021 lr:9.883314830080062e-05 tokens_per_second_per_gpu:5343.670687925167 mfu:25.207006359176397 +Step 2577 | loss:1.4270648956298828 lr:9.88322167935257e-05 tokens_per_second_per_gpu:5365.51659177997 mfu:25.310057214955048 +Step 2578 | loss:1.2878923416137695 lr:9.883128491897741e-05 tokens_per_second_per_gpu:5353.431346648843 mfu:25.25304905171608 +Step 2579 | loss:1.2080328464508057 lr:9.883035267716282e-05 tokens_per_second_per_gpu:4636.836873422184 mfu:21.872750657881372 +Step 2580 | loss:1.4013879299163818 lr:9.882942006808893e-05 tokens_per_second_per_gpu:3398.07596225818 mfu:16.029304085515143 +Step 2581 | loss:1.32210111618042 lr:9.882848709176273e-05 tokens_per_second_per_gpu:4513.210000851357 mfu:21.289581607045285 +Step 2582 | loss:1.3411579132080078 lr:9.882755374819124e-05 tokens_per_second_per_gpu:4616.7516593497 mfu:21.778005276210575 +Step 2583 | loss:1.3710300922393799 lr:9.882662003738148e-05 tokens_per_second_per_gpu:3228.229046020209 mfu:15.228107202748518 +Step 2584 | loss:1.3260005712509155 lr:9.88256859593405e-05 tokens_per_second_per_gpu:4211.193576762754 mfu:19.864918605303863 +Step 2585 | loss:1.7798030376434326 lr:9.882475151407531e-05 tokens_per_second_per_gpu:5361.609647210206 mfu:25.29162749082599 +Step 2586 | loss:1.3565597534179688 lr:9.882381670159291e-05 tokens_per_second_per_gpu:5130.317903207913 mfu:24.20058487192641 +Step 2587 | loss:1.405405879020691 lr:9.882288152190038e-05 tokens_per_second_per_gpu:5350.77857738128 mfu:25.240535486471877 +Step 2588 | loss:1.1797207593917847 lr:9.882194597500472e-05 tokens_per_second_per_gpu:5173.712444992985 mfu:24.40528432939889 +Step 2589 | loss:1.2773702144622803 lr:9.882101006091295e-05 tokens_per_second_per_gpu:5232.789061004446 mfu:24.683958806635005 +Step 2590 | loss:1.2985306978225708 lr:9.882007377963216e-05 tokens_per_second_per_gpu:3807.7141337502726 mfu:17.96163723192238 +Step 2591 | loss:1.2099521160125732 lr:9.881913713116937e-05 tokens_per_second_per_gpu:4092.0198967617384 mfu:19.302756023612577 +Step 2592 | loss:1.3189619779586792 lr:9.88182001155316e-05 tokens_per_second_per_gpu:4374.969893904371 mfu:20.637479436381977 +Step 2593 | loss:1.410544514656067 lr:9.881726273272593e-05 tokens_per_second_per_gpu:5332.991445598103 mfu:25.156630551053013 +Step 2594 | loss:1.2858161926269531 lr:9.88163249827594e-05 tokens_per_second_per_gpu:4332.255501330558 mfu:20.43598835878443 +Step 2595 | loss:1.414647102355957 lr:9.881538686563905e-05 tokens_per_second_per_gpu:4167.439581418107 mfu:19.65852354406177 +Step 2596 | loss:1.236351728439331 lr:9.881444838137195e-05 tokens_per_second_per_gpu:5357.686750459667 mfu:25.273122517538493 +Step 2597 | loss:1.5399540662765503 lr:9.881350952996515e-05 tokens_per_second_per_gpu:4581.574493012806 mfu:21.612068580755977 +Step 2598 | loss:1.227557897567749 lr:9.881257031142572e-05 tokens_per_second_per_gpu:5346.256331379937 mfu:25.219203280508815 +Step 2599 | loss:1.311738133430481 lr:9.881163072576074e-05 tokens_per_second_per_gpu:5312.807070923544 mfu:25.061417411903683 +Step 2600 | loss:1.3686952590942383 lr:9.881069077297723e-05 tokens_per_second_per_gpu:4309.908051620102 mfu:20.330571625632118 +Step 2601 | loss:1.289513111114502 lr:9.88097504530823e-05 tokens_per_second_per_gpu:4500.6676049416055 mfu:21.23041698558558 +Step 2602 | loss:1.4834165573120117 lr:9.8808809766083e-05 tokens_per_second_per_gpu:3653.5409103752168 mfu:17.234375832597884 +Step 2603 | loss:1.320642113685608 lr:9.880786871198642e-05 tokens_per_second_per_gpu:5365.447634214774 mfu:25.30973193035464 +Step 2604 | loss:1.4133163690567017 lr:9.880692729079964e-05 tokens_per_second_per_gpu:4404.977058995516 mfu:20.779028353867204 +Step 2605 | loss:1.366877794265747 lr:9.880598550252972e-05 tokens_per_second_per_gpu:5334.276943487282 mfu:25.162694463924822 +Step 2606 | loss:1.4887391328811646 lr:9.880504334718375e-05 tokens_per_second_per_gpu:4173.785218218351 mfu:19.6884569955258 +Step 2607 | loss:1.3928608894348145 lr:9.880410082476883e-05 tokens_per_second_per_gpu:4103.691787185198 mfu:19.357814322169894 +Step 2608 | loss:1.3846981525421143 lr:9.880315793529205e-05 tokens_per_second_per_gpu:4370.730334669664 mfu:20.617480712128472 +Step 2609 | loss:1.4878910779953003 lr:9.880221467876048e-05 tokens_per_second_per_gpu:4498.075790337614 mfu:21.218190953889383 +Step 2610 | loss:1.4238203763961792 lr:9.880127105518122e-05 tokens_per_second_per_gpu:4214.561144398505 mfu:19.880803996407987 +Step 2611 | loss:1.4470759630203247 lr:9.880032706456139e-05 tokens_per_second_per_gpu:5360.065633034136 mfu:25.284344112521485 +Step 2612 | loss:1.483069896697998 lr:9.879938270690806e-05 tokens_per_second_per_gpu:4779.120441190763 mfu:22.54392652312569 +Step 2613 | loss:1.3183876276016235 lr:9.879843798222835e-05 tokens_per_second_per_gpu:5055.030218781429 mfu:23.845440019083142 +Step 2614 | loss:1.5095330476760864 lr:9.879749289052936e-05 tokens_per_second_per_gpu:4657.270623155532 mfu:21.96914014173078 +Step 2615 | loss:1.3552350997924805 lr:9.879654743181821e-05 tokens_per_second_per_gpu:5303.3333031831535 mfu:25.016728033080806 +Step 2616 | loss:1.3543342351913452 lr:9.879560160610199e-05 tokens_per_second_per_gpu:4970.192221214093 mfu:23.445244709703033 +Step 2617 | loss:1.4551255702972412 lr:9.879465541338782e-05 tokens_per_second_per_gpu:5280.29128155247 mfu:24.90803488567459 +Step 2618 | loss:1.2968884706497192 lr:9.879370885368282e-05 tokens_per_second_per_gpu:5099.26254275277 mfu:24.05409143806863 +Step 2619 | loss:1.3352900743484497 lr:9.879276192699411e-05 tokens_per_second_per_gpu:4297.280627640565 mfu:20.27100591225995 +Step 2620 | loss:1.3451112508773804 lr:9.879181463332882e-05 tokens_per_second_per_gpu:4550.290984963555 mfu:21.4644989357708 +Step 2621 | loss:1.3139469623565674 lr:9.879086697269406e-05 tokens_per_second_per_gpu:5344.603086013189 mfu:25.211404640041113 +Step 2622 | loss:1.3380709886550903 lr:9.878991894509697e-05 tokens_per_second_per_gpu:5072.610550567168 mfu:23.928369443630377 +Step 2623 | loss:1.4849499464035034 lr:9.878897055054468e-05 tokens_per_second_per_gpu:3772.424181398156 mfu:17.795168505590702 +Step 2624 | loss:1.6447049379348755 lr:9.87880217890443e-05 tokens_per_second_per_gpu:5310.982816298478 mfu:25.05281209158731 +Step 2625 | loss:1.2970768213272095 lr:9.878707266060299e-05 tokens_per_second_per_gpu:5327.667848255713 mfu:25.131518234089587 +Step 2626 | loss:1.2572048902511597 lr:9.878612316522788e-05 tokens_per_second_per_gpu:3536.974782301044 mfu:16.684513518239754 +Step 2627 | loss:1.305930256843567 lr:9.878517330292611e-05 tokens_per_second_per_gpu:4212.262726035486 mfu:19.869961964839014 +Step 2628 | loss:1.4697027206420898 lr:9.878422307370483e-05 tokens_per_second_per_gpu:5339.248515353384 mfu:25.186146216654855 +Step 2629 | loss:1.2715060710906982 lr:9.878327247757119e-05 tokens_per_second_per_gpu:4514.853993945451 mfu:21.297336602964275 +Step 2630 | loss:1.386638879776001 lr:9.878232151453233e-05 tokens_per_second_per_gpu:5356.140195266542 mfu:25.265827152822997 +Step 2631 | loss:1.247666835784912 lr:9.87813701845954e-05 tokens_per_second_per_gpu:5120.166340574007 mfu:24.15269821894709 +Step 2632 | loss:1.248504400253296 lr:9.878041848776756e-05 tokens_per_second_per_gpu:4911.423468071922 mfu:23.168022474148177 +Step 2633 | loss:1.218843698501587 lr:9.877946642405598e-05 tokens_per_second_per_gpu:4362.292461438767 mfu:20.577677824448703 +Step 2634 | loss:1.4980190992355347 lr:9.87785139934678e-05 tokens_per_second_per_gpu:5090.681428984412 mfu:24.01361286033431 +Step 2635 | loss:1.2507405281066895 lr:9.877756119601019e-05 tokens_per_second_per_gpu:5300.531804492213 mfu:25.00351288577067 +Step 2636 | loss:1.4899429082870483 lr:9.877660803169031e-05 tokens_per_second_per_gpu:4193.480354415446 mfu:19.78136231330424 +Step 2637 | loss:1.2190335988998413 lr:9.877565450051536e-05 tokens_per_second_per_gpu:5344.042292531878 mfu:25.208759281508538 +Step 2638 | loss:1.4132052659988403 lr:9.877470060249247e-05 tokens_per_second_per_gpu:4223.8608571749755 mfu:19.924672328269214 +Step 2639 | loss:1.291564702987671 lr:9.877374633762884e-05 tokens_per_second_per_gpu:3679.0230709844673 mfu:17.35457953189664 +Step 2640 | loss:1.3152133226394653 lr:9.877279170593163e-05 tokens_per_second_per_gpu:5126.6994209080285 mfu:24.18351587354111 +Step 2641 | loss:1.3632971048355103 lr:9.877183670740804e-05 tokens_per_second_per_gpu:4924.22771118344 mfu:23.228422273534356 +Step 2642 | loss:1.2857985496520996 lr:9.877088134206525e-05 tokens_per_second_per_gpu:5330.278794031589 mfu:25.14383451078768 +Step 2643 | loss:1.260108470916748 lr:9.876992560991043e-05 tokens_per_second_per_gpu:5366.519438241627 mfu:25.31478781281793 +Step 2644 | loss:1.4033695459365845 lr:9.876896951095077e-05 tokens_per_second_per_gpu:3969.7651350870356 mfu:18.72605945397977 +Step 2645 | loss:1.2926379442214966 lr:9.876801304519348e-05 tokens_per_second_per_gpu:5316.414705166509 mfu:25.078435238153617 +Step 2646 | loss:1.3567243814468384 lr:9.876705621264573e-05 tokens_per_second_per_gpu:5299.653959336862 mfu:24.999371940397054 +Step 2647 | loss:1.4623911380767822 lr:9.876609901331473e-05 tokens_per_second_per_gpu:3942.0617445517114 mfu:18.59537783413882 +Step 2648 | loss:1.4447194337844849 lr:9.876514144720767e-05 tokens_per_second_per_gpu:5330.030905231408 mfu:25.142665177023037 +Step 2649 | loss:1.4705406427383423 lr:9.876418351433177e-05 tokens_per_second_per_gpu:4927.517385196739 mfu:23.243940227131354 +Step 2650 | loss:1.4634381532669067 lr:9.876322521469423e-05 tokens_per_second_per_gpu:5330.599163768312 mfu:25.145345749496517 +Step 2651 | loss:1.3665592670440674 lr:9.876226654830224e-05 tokens_per_second_per_gpu:5368.2345731750875 mfu:25.32287839693126 +Step 2652 | loss:1.293166995048523 lr:9.876130751516302e-05 tokens_per_second_per_gpu:4800.922560619133 mfu:22.64677084029446 +Step 2653 | loss:1.3449311256408691 lr:9.876034811528379e-05 tokens_per_second_per_gpu:4136.434535924091 mfu:19.512267454460243 +Step 2654 | loss:1.4623932838439941 lr:9.875938834867175e-05 tokens_per_second_per_gpu:4034.7580217977065 mfu:19.03264198952393 +Step 2655 | loss:1.3970401287078857 lr:9.875842821533414e-05 tokens_per_second_per_gpu:3684.770776820627 mfu:17.381692440985542 +Step 2656 | loss:1.376155138015747 lr:9.875746771527816e-05 tokens_per_second_per_gpu:3635.833606542022 mfu:17.150847459239138 +Step 2657 | loss:1.3280305862426758 lr:9.875650684851105e-05 tokens_per_second_per_gpu:5365.145045099742 mfu:25.308304565872007 +Step 2658 | loss:1.4843553304672241 lr:9.875554561504003e-05 tokens_per_second_per_gpu:3890.1713241237535 mfu:18.35060186756148 +Step 2659 | loss:1.4810231924057007 lr:9.875458401487235e-05 tokens_per_second_per_gpu:5292.741220751979 mfu:24.966763373057344 +Step 2660 | loss:1.3398511409759521 lr:9.87536220480152e-05 tokens_per_second_per_gpu:4663.982030545069 mfu:22.000798995471406 +Step 2661 | loss:1.392621397972107 lr:9.875265971447583e-05 tokens_per_second_per_gpu:3902.945309517098 mfu:18.410858936128534 +Step 2662 | loss:1.5680179595947266 lr:9.87516970142615e-05 tokens_per_second_per_gpu:5313.814701121063 mfu:25.066170575464817 +Step 2663 | loss:1.3917207717895508 lr:9.875073394737942e-05 tokens_per_second_per_gpu:5370.320407872694 mfu:25.332717635079966 +Step 2664 | loss:1.3517638444900513 lr:9.874977051383687e-05 tokens_per_second_per_gpu:5336.539920577886 mfu:25.173369312964905 +Step 2665 | loss:1.2769148349761963 lr:9.874880671364106e-05 tokens_per_second_per_gpu:4068.4597535040607 mfu:19.191618808091167 +Step 2666 | loss:1.4558148384094238 lr:9.874784254679926e-05 tokens_per_second_per_gpu:4729.9876152098495 mfu:22.312158600049187 +Step 2667 | loss:1.175788402557373 lr:9.874687801331873e-05 tokens_per_second_per_gpu:2808.197852755481 mfu:13.246748399407393 +Step 2668 | loss:1.4437007904052734 lr:9.874591311320669e-05 tokens_per_second_per_gpu:4405.002717159576 mfu:20.779149387804807 +Step 2669 | loss:1.3746745586395264 lr:9.874494784647041e-05 tokens_per_second_per_gpu:3231.835083067171 mfu:15.2451174947524 +Step 2670 | loss:1.2721415758132935 lr:9.874398221311718e-05 tokens_per_second_per_gpu:4935.341357038664 mfu:23.280847237216957 +Step 2671 | loss:1.2841379642486572 lr:9.874301621315423e-05 tokens_per_second_per_gpu:5352.1089956763335 mfu:25.24681129656241 +Step 2672 | loss:1.5116742849349976 lr:9.874204984658884e-05 tokens_per_second_per_gpu:3030.15420001655 mfu:14.293754359095695 +Step 2673 | loss:1.3521065711975098 lr:9.874108311342827e-05 tokens_per_second_per_gpu:5060.382350733178 mfu:23.87068693866742 +Step 2674 | loss:1.4615596532821655 lr:9.874011601367979e-05 tokens_per_second_per_gpu:4366.63241097911 mfu:20.59815010690225 +Step 2675 | loss:1.4016838073730469 lr:9.87391485473507e-05 tokens_per_second_per_gpu:4568.750056151346 mfu:21.551573524005914 +Step 2676 | loss:1.3237930536270142 lr:9.873818071444822e-05 tokens_per_second_per_gpu:5310.634032296658 mfu:25.05116681794204 +Step 2677 | loss:1.3312795162200928 lr:9.87372125149797e-05 tokens_per_second_per_gpu:4545.703485196966 mfu:21.442858916663628 +Step 2678 | loss:1.4468603134155273 lr:9.873624394895236e-05 tokens_per_second_per_gpu:4055.018675319235 mfu:19.12821494900893 +Step 2679 | loss:1.5456432104110718 lr:9.873527501637352e-05 tokens_per_second_per_gpu:5356.602046906814 mfu:25.26800578581018 +Step 2680 | loss:1.164914846420288 lr:9.873430571725046e-05 tokens_per_second_per_gpu:3868.8592669471986 mfu:18.250069257647766 +Step 2681 | loss:1.534442663192749 lr:9.873333605159045e-05 tokens_per_second_per_gpu:3687.4212006434545 mfu:17.394194942366756 +Step 2682 | loss:1.297692894935608 lr:9.873236601940081e-05 tokens_per_second_per_gpu:5087.7666366948115 mfu:23.999863287789417 +Step 2683 | loss:1.3478691577911377 lr:9.873139562068885e-05 tokens_per_second_per_gpu:4438.069475979883 mfu:20.935130930931013 +Step 2684 | loss:1.211235523223877 lr:9.873042485546182e-05 tokens_per_second_per_gpu:4764.153367085023 mfu:22.473324280922746 +Step 2685 | loss:1.3690261840820312 lr:9.872945372372706e-05 tokens_per_second_per_gpu:3826.244631805224 mfu:18.049048752876573 +Step 2686 | loss:1.4340693950653076 lr:9.872848222549185e-05 tokens_per_second_per_gpu:4257.736726600441 mfu:20.084470584168454 +Step 2687 | loss:1.3375966548919678 lr:9.87275103607635e-05 tokens_per_second_per_gpu:4764.475130277868 mfu:22.474842092801858 +Step 2688 | loss:1.3632303476333618 lr:9.872653812954935e-05 tokens_per_second_per_gpu:4509.322596003227 mfu:21.271244055116995 +Step 2689 | loss:1.4625858068466187 lr:9.872556553185668e-05 tokens_per_second_per_gpu:4407.101684702351 mfu:20.78905057582487 +Step 2690 | loss:1.3565824031829834 lr:9.87245925676928e-05 tokens_per_second_per_gpu:3153.8736218147455 mfu:14.877359980427801 +Step 2691 | loss:1.3148980140686035 lr:9.872361923706506e-05 tokens_per_second_per_gpu:4823.433142041303 mfu:22.752957093564067 +Step 2692 | loss:1.2741644382476807 lr:9.872264553998076e-05 tokens_per_second_per_gpu:5309.753092894619 mfu:25.04701127647886 +Step 2693 | loss:1.3579175472259521 lr:9.872167147644722e-05 tokens_per_second_per_gpu:4382.158245658057 mfu:20.671388118063394 +Step 2694 | loss:1.5096279382705688 lr:9.872069704647177e-05 tokens_per_second_per_gpu:5357.2436882484635 mfu:25.27103251749404 +Step 2695 | loss:1.374257206916809 lr:9.871972225006175e-05 tokens_per_second_per_gpu:4375.957640557668 mfu:20.642138805872538 +Step 2696 | loss:1.3272982835769653 lr:9.871874708722448e-05 tokens_per_second_per_gpu:5375.108642200526 mfu:25.35530455336137 +Step 2697 | loss:1.3455616235733032 lr:9.87177715579673e-05 tokens_per_second_per_gpu:5314.621855784126 mfu:25.06997806172527 +Step 2698 | loss:1.4107255935668945 lr:9.871679566229753e-05 tokens_per_second_per_gpu:4350.627731005403 mfu:20.52265330078629 +Step 2699 | loss:1.250206470489502 lr:9.871581940022254e-05 tokens_per_second_per_gpu:4654.035699416284 mfu:21.953880454517844 +Step 2700 | loss:1.3681517839431763 lr:9.871484277174965e-05 tokens_per_second_per_gpu:4006.0561651303174 mfu:18.897250434582144 +Step 2701 | loss:1.2803736925125122 lr:9.87138657768862e-05 tokens_per_second_per_gpu:4607.391843604005 mfu:21.73385342838889 +Step 2702 | loss:1.3094193935394287 lr:9.871288841563956e-05 tokens_per_second_per_gpu:5355.357118724479 mfu:25.26213324716745 +Step 2703 | loss:1.3782294988632202 lr:9.871191068801707e-05 tokens_per_second_per_gpu:3973.192401828189 mfu:18.74222645594979 +Step 2704 | loss:1.361412763595581 lr:9.871093259402609e-05 tokens_per_second_per_gpu:5380.054184479992 mfu:25.37863352009013 +Step 2705 | loss:1.3531548976898193 lr:9.870995413367397e-05 tokens_per_second_per_gpu:5055.553891458435 mfu:23.847910272448352 +Step 2706 | loss:1.3276569843292236 lr:9.870897530696806e-05 tokens_per_second_per_gpu:5338.497541243977 mfu:25.182603743651384 +Step 2707 | loss:1.1770615577697754 lr:9.870799611391572e-05 tokens_per_second_per_gpu:5374.08567886038 mfu:25.350479060750352 +Step 2708 | loss:1.6050857305526733 lr:9.870701655452434e-05 tokens_per_second_per_gpu:3924.333261010416 mfu:18.511749552483334 +Step 2709 | loss:1.3347333669662476 lr:9.870603662880128e-05 tokens_per_second_per_gpu:5364.42057450609 mfu:25.304887114474326 +Step 2710 | loss:1.3143683671951294 lr:9.87050563367539e-05 tokens_per_second_per_gpu:5059.157312043354 mfu:23.864908222154366 +Step 2711 | loss:1.516165852546692 lr:9.870407567838958e-05 tokens_per_second_per_gpu:5335.204724675354 mfu:25.16707096608475 +Step 2712 | loss:1.3333784341812134 lr:9.870309465371567e-05 tokens_per_second_per_gpu:4618.0828585526415 mfu:21.784284769976853 +Step 2713 | loss:1.3996983766555786 lr:9.870211326273959e-05 tokens_per_second_per_gpu:4826.40167575418 mfu:22.76696017357179 +Step 2714 | loss:1.4644464254379272 lr:9.870113150546871e-05 tokens_per_second_per_gpu:5341.89794705472 mfu:25.19864403802972 +Step 2715 | loss:1.4053593873977661 lr:9.870014938191039e-05 tokens_per_second_per_gpu:5289.872306850734 mfu:24.953230216697982 +Step 2716 | loss:1.4579304456710815 lr:9.869916689207203e-05 tokens_per_second_per_gpu:3927.5862286508636 mfu:18.527094355856832 +Step 2717 | loss:1.3794511556625366 lr:9.869818403596103e-05 tokens_per_second_per_gpu:3922.0611258851864 mfu:18.5010315034314 +Step 2718 | loss:1.330643653869629 lr:9.869720081358478e-05 tokens_per_second_per_gpu:4700.270718173673 mfu:22.171978926504085 +Step 2719 | loss:1.3752002716064453 lr:9.869621722495067e-05 tokens_per_second_per_gpu:5333.254562038675 mfu:25.157871716195483 +Step 2720 | loss:1.3115859031677246 lr:9.869523327006607e-05 tokens_per_second_per_gpu:3702.838684544868 mfu:17.466921844423727 +Step 2721 | loss:1.3402966260910034 lr:9.869424894893842e-05 tokens_per_second_per_gpu:4564.352965853519 mfu:21.5308316988501 +Step 2722 | loss:1.1645276546478271 lr:9.869326426157512e-05 tokens_per_second_per_gpu:4922.312086329394 mfu:23.21938594425852 +Step 2723 | loss:1.2508435249328613 lr:9.869227920798357e-05 tokens_per_second_per_gpu:3591.8775797521776 mfu:16.943499381199928 +Step 2724 | loss:1.295508861541748 lr:9.869129378817115e-05 tokens_per_second_per_gpu:4390.269870733958 mfu:20.709652037532365 +Step 2725 | loss:1.3614164590835571 lr:9.869030800214532e-05 tokens_per_second_per_gpu:5351.13985876244 mfu:25.242239712387764 +Step 2726 | loss:1.3488010168075562 lr:9.868932184991346e-05 tokens_per_second_per_gpu:3929.674709617664 mfu:18.53694607690883 +Step 2727 | loss:1.3296613693237305 lr:9.8688335331483e-05 tokens_per_second_per_gpu:5366.675767654475 mfu:25.315525245330885 +Step 2728 | loss:1.3062176704406738 lr:9.868734844686136e-05 tokens_per_second_per_gpu:5344.64090109294 mfu:25.211583020224218 +Step 2729 | loss:1.480977177619934 lr:9.868636119605597e-05 tokens_per_second_per_gpu:5247.672106229089 mfu:24.754164670269006 +Step 2730 | loss:1.5134884119033813 lr:9.868537357907422e-05 tokens_per_second_per_gpu:4230.271952058256 mfu:19.954914556679228 +Step 2731 | loss:1.6431019306182861 lr:9.868438559592356e-05 tokens_per_second_per_gpu:4197.1397066037935 mfu:19.79862410192654 +Step 2732 | loss:1.2325377464294434 lr:9.868339724661145e-05 tokens_per_second_per_gpu:4699.210962754705 mfu:22.166979879379895 +Step 2733 | loss:1.1628979444503784 lr:9.86824085311453e-05 tokens_per_second_per_gpu:4999.796185904191 mfu:23.58489166210344 +Step 2734 | loss:1.247543215751648 lr:9.868141944953252e-05 tokens_per_second_per_gpu:4185.842712748275 mfu:19.74533425444415 +Step 2735 | loss:1.4483929872512817 lr:9.868043000178057e-05 tokens_per_second_per_gpu:4170.4172650688 mfu:19.672569785887305 +Step 2736 | loss:1.3159091472625732 lr:9.867944018789692e-05 tokens_per_second_per_gpu:4328.996406020152 mfu:20.420614650146245 +Step 2737 | loss:1.3256120681762695 lr:9.867845000788898e-05 tokens_per_second_per_gpu:4501.276274114569 mfu:21.233288181924117 +Step 2738 | loss:1.213085412979126 lr:9.86774594617642e-05 tokens_per_second_per_gpu:4422.911315826356 mfu:20.86362730323802 +Step 2739 | loss:1.7434369325637817 lr:9.867646854953003e-05 tokens_per_second_per_gpu:3203.5637140375356 mfu:15.111756623446661 +Step 2740 | loss:1.2081493139266968 lr:9.867547727119395e-05 tokens_per_second_per_gpu:5262.831321936999 mfu:24.82567327719212 +Step 2741 | loss:1.257467269897461 lr:9.867448562676338e-05 tokens_per_second_per_gpu:5298.0757999609705 mfu:24.99192749335922 +Step 2742 | loss:1.5064860582351685 lr:9.867349361624579e-05 tokens_per_second_per_gpu:4122.938623004193 mfu:19.448604930576902 +Step 2743 | loss:1.3124293088912964 lr:9.867250123964864e-05 tokens_per_second_per_gpu:4015.034726138362 mfu:18.939603838757563 +Step 2744 | loss:1.4264497756958008 lr:9.867150849697942e-05 tokens_per_second_per_gpu:4691.28146547155 mfu:22.129575087783508 +Step 2745 | loss:1.3729159832000732 lr:9.867051538824554e-05 tokens_per_second_per_gpu:4024.4276348434787 mfu:18.983911791715077 +Step 2746 | loss:1.4945629835128784 lr:9.866952191345454e-05 tokens_per_second_per_gpu:5370.519536427395 mfu:25.333656958447328 +Step 2747 | loss:1.4603782892227173 lr:9.866852807261383e-05 tokens_per_second_per_gpu:4741.844335844567 mfu:22.368088774248292 +Step 2748 | loss:1.431203007698059 lr:9.866753386573092e-05 tokens_per_second_per_gpu:4813.602564079529 mfu:22.706584580048826 +Step 2749 | loss:1.3812505006790161 lr:9.866653929281328e-05 tokens_per_second_per_gpu:5323.939715939708 mfu:25.113931997870026 +Step 2750 | loss:1.171600103378296 lr:9.866554435386838e-05 tokens_per_second_per_gpu:4291.1327234338505 mfu:20.24200519917626 +Step 2751 | loss:1.2988883256912231 lr:9.866454904890372e-05 tokens_per_second_per_gpu:4981.414194414424 mfu:23.49818067195506 +Step 2752 | loss:1.4095923900604248 lr:9.866355337792676e-05 tokens_per_second_per_gpu:5342.27999937436 mfu:25.200446244006304 +Step 2753 | loss:1.4051975011825562 lr:9.866255734094503e-05 tokens_per_second_per_gpu:4733.1262591919085 mfu:22.326964119220445 +Step 2754 | loss:1.1981449127197266 lr:9.866156093796598e-05 tokens_per_second_per_gpu:4846.241289216437 mfu:22.860547015260355 +Step 2755 | loss:1.5562665462493896 lr:9.866056416899712e-05 tokens_per_second_per_gpu:4805.153345802031 mfu:22.66672817584858 +Step 2756 | loss:1.3340550661087036 lr:9.865956703404596e-05 tokens_per_second_per_gpu:5331.105774631404 mfu:25.147735519374447 +Step 2757 | loss:1.2069567441940308 lr:9.865856953311998e-05 tokens_per_second_per_gpu:5329.599364100451 mfu:25.140629523879802 +Step 2758 | loss:1.1491522789001465 lr:9.865757166622669e-05 tokens_per_second_per_gpu:5324.25034115934 mfu:25.11539727002916 +Step 2759 | loss:1.39962899684906 lr:9.86565734333736e-05 tokens_per_second_per_gpu:3266.7697270338713 mfu:15.40991017080821 +Step 2760 | loss:1.4577306509017944 lr:9.86555748345682e-05 tokens_per_second_per_gpu:5338.134741489094 mfu:25.180892355307584 +Step 2761 | loss:1.392643928527832 lr:9.865457586981803e-05 tokens_per_second_per_gpu:3424.7875778157872 mfu:16.155307333571944 +Step 2762 | loss:1.3564436435699463 lr:9.865357653913057e-05 tokens_per_second_per_gpu:5340.098867265393 mfu:25.190157471708797 +Step 2763 | loss:1.377999186515808 lr:9.865257684251336e-05 tokens_per_second_per_gpu:4660.533870770512 mfu:21.984533437498094 +Step 2764 | loss:1.1776351928710938 lr:9.865157677997391e-05 tokens_per_second_per_gpu:4507.021674665563 mfu:21.260390216589578 +Step 2765 | loss:1.5291284322738647 lr:9.865057635151975e-05 tokens_per_second_per_gpu:4446.082122967234 mfu:20.972927953868894 +Step 2766 | loss:1.3848623037338257 lr:9.86495755571584e-05 tokens_per_second_per_gpu:5348.300231230884 mfu:25.228844704831 +Step 2767 | loss:1.3625223636627197 lr:9.864857439689736e-05 tokens_per_second_per_gpu:5356.366986643306 mfu:25.26689696644179 +Step 2768 | loss:1.31890869140625 lr:9.86475728707442e-05 tokens_per_second_per_gpu:5096.705081146103 mfu:24.04202745532204 +Step 2769 | loss:1.443038821220398 lr:9.864657097870644e-05 tokens_per_second_per_gpu:4773.23116410665 mfu:22.516145798305647 +Step 2770 | loss:1.4055819511413574 lr:9.86455687207916e-05 tokens_per_second_per_gpu:4412.9281731659585 mfu:20.81653511600915 +Step 2771 | loss:1.3882768154144287 lr:9.864456609700726e-05 tokens_per_second_per_gpu:5146.424546877648 mfu:24.27656265819333 +Step 2772 | loss:1.3528587818145752 lr:9.864356310736091e-05 tokens_per_second_per_gpu:4270.016039666637 mfu:20.142394198968542 +Step 2773 | loss:1.3882824182510376 lr:9.86425597518601e-05 tokens_per_second_per_gpu:5042.420969988993 mfu:23.78596004116954 +Step 2774 | loss:1.4199621677398682 lr:9.864155603051241e-05 tokens_per_second_per_gpu:4717.0140530130175 mfu:22.250960093648768 +Step 2775 | loss:1.3656246662139893 lr:9.864055194332537e-05 tokens_per_second_per_gpu:5305.607605849469 mfu:25.027456306793866 +Step 2776 | loss:1.2726812362670898 lr:9.863954749030652e-05 tokens_per_second_per_gpu:5287.387693457697 mfu:24.94150987140427 +Step 2777 | loss:1.2533938884735107 lr:9.863854267146346e-05 tokens_per_second_per_gpu:5325.174523207132 mfu:25.11975679443031 +Step 2778 | loss:1.47574782371521 lr:9.863753748680369e-05 tokens_per_second_per_gpu:5278.0999447817185 mfu:24.89769797624635 +Step 2779 | loss:1.4338128566741943 lr:9.863653193633479e-05 tokens_per_second_per_gpu:5349.510707554303 mfu:25.234554728177123 +Step 2780 | loss:1.3354699611663818 lr:9.863552602006435e-05 tokens_per_second_per_gpu:4294.3298651944915 mfu:20.257086662349835 +Step 2781 | loss:1.2360591888427734 lr:9.863451973799989e-05 tokens_per_second_per_gpu:5355.9936106521645 mfu:25.265135688187065 +Step 2782 | loss:1.2605907917022705 lr:9.8633513090149e-05 tokens_per_second_per_gpu:5302.387628046885 mfu:25.01226712211421 +Step 2783 | loss:1.3122105598449707 lr:9.863250607651926e-05 tokens_per_second_per_gpu:4189.452134794274 mfu:19.76236051406603 +Step 2784 | loss:1.4293646812438965 lr:9.863149869711823e-05 tokens_per_second_per_gpu:5336.746236379682 mfu:25.174342539803085 +Step 2785 | loss:1.266339898109436 lr:9.86304909519535e-05 tokens_per_second_per_gpu:5227.939654117744 mfu:24.661083326956298 +Step 2786 | loss:1.3366022109985352 lr:9.862948284103264e-05 tokens_per_second_per_gpu:4508.179112013268 mfu:21.265850046037876 +Step 2787 | loss:1.3059735298156738 lr:9.862847436436324e-05 tokens_per_second_per_gpu:4204.412266636376 mfu:19.832930008426924 +Step 2788 | loss:1.3332931995391846 lr:9.862746552195287e-05 tokens_per_second_per_gpu:5071.67294369196 mfu:23.923946592027793 +Step 2789 | loss:1.38021981716156 lr:9.862645631380914e-05 tokens_per_second_per_gpu:3458.719003522231 mfu:16.31536765792744 +Step 2790 | loss:1.2793464660644531 lr:9.862544673993961e-05 tokens_per_second_per_gpu:4565.057807991907 mfu:21.534156559475345 +Step 2791 | loss:1.2754591703414917 lr:9.862443680035191e-05 tokens_per_second_per_gpu:4071.3071698533568 mfu:19.20505055683012 +Step 2792 | loss:1.255043864250183 lr:9.862342649505361e-05 tokens_per_second_per_gpu:3958.024727313209 mfu:18.67067794739051 +Step 2793 | loss:1.3603261709213257 lr:9.86224158240523e-05 tokens_per_second_per_gpu:4100.69188376177 mfu:19.343663265885283 +Step 2794 | loss:1.4099808931350708 lr:9.862140478735562e-05 tokens_per_second_per_gpu:4531.219525081236 mfu:21.374535605579307 +Step 2795 | loss:1.3707090616226196 lr:9.862039338497114e-05 tokens_per_second_per_gpu:4197.454904496239 mfu:19.80011094416351 +Step 2796 | loss:1.2719593048095703 lr:9.861938161690646e-05 tokens_per_second_per_gpu:5126.0529755984835 mfu:24.180466480721044 +Step 2797 | loss:1.3775814771652222 lr:9.861836948316922e-05 tokens_per_second_per_gpu:4129.549160607333 mfu:19.47978796432493 +Step 2798 | loss:1.3571720123291016 lr:9.861735698376702e-05 tokens_per_second_per_gpu:4650.509391593179 mfu:21.937246258866104 +Step 2799 | loss:1.3628215789794922 lr:9.861634411870748e-05 tokens_per_second_per_gpu:5018.071905104337 mfu:23.67110134772976 +Step 2800 | loss:1.1937955617904663 lr:9.861533088799821e-05 tokens_per_second_per_gpu:4545.020511447553 mfu:21.43963721296031 +Step 2801 | loss:1.3099489212036133 lr:9.861431729164683e-05 tokens_per_second_per_gpu:4447.94626869653 mfu:20.98172144733004 +Step 2802 | loss:1.3684778213500977 lr:9.861330332966096e-05 tokens_per_second_per_gpu:4726.876356456763 mfu:22.2974822616754 +Step 2803 | loss:1.3660694360733032 lr:9.861228900204824e-05 tokens_per_second_per_gpu:5364.60308344732 mfu:25.305748040289348 +Step 2804 | loss:1.224008560180664 lr:9.86112743088163e-05 tokens_per_second_per_gpu:4781.133181130851 mfu:22.553420960831932 +Step 2805 | loss:1.4620558023452759 lr:9.861025924997276e-05 tokens_per_second_per_gpu:5303.99335508316 mfu:25.0198416105096 +Step 2806 | loss:1.4843674898147583 lr:9.860924382552524e-05 tokens_per_second_per_gpu:4749.959491056763 mfu:22.40636934597247 +Step 2807 | loss:1.1582032442092896 lr:9.860822803548141e-05 tokens_per_second_per_gpu:4178.154394232441 mfu:19.70906714424256 +Step 2808 | loss:1.1889718770980835 lr:9.86072118798489e-05 tokens_per_second_per_gpu:4432.286461587561 mfu:20.90785146535843 +Step 2809 | loss:1.328167200088501 lr:9.860619535863534e-05 tokens_per_second_per_gpu:4450.677491924198 mfu:20.994605093289685 +Step 2810 | loss:1.3501739501953125 lr:9.860517847184837e-05 tokens_per_second_per_gpu:5360.723916988372 mfu:25.28744935024844 +Step 2811 | loss:1.3208246231079102 lr:9.860416121949566e-05 tokens_per_second_per_gpu:5321.306541973148 mfu:25.10151086700396 +Step 2812 | loss:1.2809594869613647 lr:9.860314360158485e-05 tokens_per_second_per_gpu:4773.098635114238 mfu:22.51552063643666 +Step 2813 | loss:1.3720732927322388 lr:9.860212561812361e-05 tokens_per_second_per_gpu:3880.955310784663 mfu:18.307128360226976 +Step 2814 | loss:1.2132712602615356 lr:9.860110726911956e-05 tokens_per_second_per_gpu:4540.880788144187 mfu:21.420109431828703 +Step 2815 | loss:1.2736221551895142 lr:9.86000885545804e-05 tokens_per_second_per_gpu:5313.787776268674 mfu:25.066043566342245 +Step 2816 | loss:1.4633103609085083 lr:9.859906947451376e-05 tokens_per_second_per_gpu:4326.290156144363 mfu:20.407848807748312 +Step 2817 | loss:1.3684697151184082 lr:9.859805002892732e-05 tokens_per_second_per_gpu:4152.272789524566 mfu:19.5869791989784 +Step 2818 | loss:1.2942173480987549 lr:9.859703021782873e-05 tokens_per_second_per_gpu:5375.334611970096 mfu:25.356370491318852 +Step 2819 | loss:1.3112884759902954 lr:9.859601004122569e-05 tokens_per_second_per_gpu:5329.944707681993 mfu:25.14225856847563 +Step 2820 | loss:1.3660651445388794 lr:9.859498949912586e-05 tokens_per_second_per_gpu:3606.220696372353 mfu:17.011158309485335 +Step 2821 | loss:1.3485254049301147 lr:9.859396859153691e-05 tokens_per_second_per_gpu:5337.811026248904 mfu:25.179365335288093 +Step 2822 | loss:1.4146510362625122 lr:9.859294731846653e-05 tokens_per_second_per_gpu:3523.4035123114277 mfu:16.620495522201306 +Step 2823 | loss:1.402690052986145 lr:9.85919256799224e-05 tokens_per_second_per_gpu:5361.1888615372545 mfu:25.28964257301324 +Step 2824 | loss:1.2423427104949951 lr:9.859090367591217e-05 tokens_per_second_per_gpu:3762.859085395706 mfu:17.750048315773483 +Step 2825 | loss:1.3578860759735107 lr:9.858988130644355e-05 tokens_per_second_per_gpu:5155.355061594357 mfu:24.318689419814977 +Step 2826 | loss:1.366062879562378 lr:9.858885857152426e-05 tokens_per_second_per_gpu:4466.972153494791 mfu:21.07146979207412 +Step 2827 | loss:1.3229140043258667 lr:9.858783547116195e-05 tokens_per_second_per_gpu:4482.098115796946 mfu:21.142821537008672 +Step 2828 | loss:1.286385416984558 lr:9.858681200536432e-05 tokens_per_second_per_gpu:4630.487096185461 mfu:21.842797675272035 +Step 2829 | loss:1.4507551193237305 lr:9.85857881741391e-05 tokens_per_second_per_gpu:4217.882065384094 mfu:19.896469347304553 +Step 2830 | loss:1.4126167297363281 lr:9.858476397749396e-05 tokens_per_second_per_gpu:5347.894308467003 mfu:25.22692989789621 +Step 2831 | loss:1.371647834777832 lr:9.858373941543661e-05 tokens_per_second_per_gpu:3907.608253958446 mfu:18.432854840638733 +Step 2832 | loss:1.217326045036316 lr:9.858271448797475e-05 tokens_per_second_per_gpu:5135.062663643481 mfu:24.22296671644059 +Step 2833 | loss:1.3191958665847778 lr:9.85816891951161e-05 tokens_per_second_per_gpu:3874.5955755957966 mfu:18.277128404258377 +Step 2834 | loss:1.531026005744934 lr:9.858066353686836e-05 tokens_per_second_per_gpu:3545.747839181889 mfu:16.725897524384553 +Step 2835 | loss:1.2427421808242798 lr:9.857963751323925e-05 tokens_per_second_per_gpu:5301.784416053465 mfu:25.009421668222668 +Step 2836 | loss:1.4119504690170288 lr:9.857861112423649e-05 tokens_per_second_per_gpu:5316.749437025142 mfu:25.080014225441932 +Step 2837 | loss:1.4405481815338135 lr:9.857758436986782e-05 tokens_per_second_per_gpu:3363.16060525745 mfu:15.86460238936851 +Step 2838 | loss:1.1107769012451172 lr:9.85765572501409e-05 tokens_per_second_per_gpu:4192.976830351307 mfu:19.77898710438363 +Step 2839 | loss:1.3616522550582886 lr:9.857552976506352e-05 tokens_per_second_per_gpu:4813.641185381339 mfu:22.706766763318935 +Step 2840 | loss:1.3063368797302246 lr:9.857450191464337e-05 tokens_per_second_per_gpu:5219.1408700039365 mfu:24.619577961064095 +Step 2841 | loss:1.4818459749221802 lr:9.857347369888821e-05 tokens_per_second_per_gpu:4101.585714915878 mfu:19.34787961994253 +Step 2842 | loss:1.6449766159057617 lr:9.857244511780575e-05 tokens_per_second_per_gpu:4427.791061391134 mfu:20.88664589563752 +Step 2843 | loss:1.208860158920288 lr:9.857141617140372e-05 tokens_per_second_per_gpu:5330.184786528109 mfu:25.143391061354432 +Step 2844 | loss:1.307009220123291 lr:9.857038685968987e-05 tokens_per_second_per_gpu:5323.670473007123 mfu:25.112661933019446 +Step 2845 | loss:1.321173071861267 lr:9.856935718267196e-05 tokens_per_second_per_gpu:4064.4282087745405 mfu:19.17260132374956 +Step 2846 | loss:1.6153570413589478 lr:9.856832714035771e-05 tokens_per_second_per_gpu:3484.6373757430015 mfu:16.437629041823108 +Step 2847 | loss:1.4695093631744385 lr:9.856729673275487e-05 tokens_per_second_per_gpu:3043.0722372109776 mfu:14.354690944586228 +Step 2848 | loss:1.3740696907043457 lr:9.85662659598712e-05 tokens_per_second_per_gpu:4247.116633424969 mfu:20.03437379268501 +Step 2849 | loss:1.5017322301864624 lr:9.856523482171444e-05 tokens_per_second_per_gpu:3995.621108522158 mfu:18.84802649215721 +Step 2850 | loss:1.1841193437576294 lr:9.856420331829236e-05 tokens_per_second_per_gpu:5324.380717631675 mfu:25.11601227809368 +Step 2851 | loss:1.3904529809951782 lr:9.856317144961271e-05 tokens_per_second_per_gpu:3547.1525045720064 mfu:16.732523570690365 +Step 2852 | loss:1.3676519393920898 lr:9.856213921568325e-05 tokens_per_second_per_gpu:4852.9568962544045 mfu:22.892225679459163 +Step 2853 | loss:1.5282236337661743 lr:9.856110661651173e-05 tokens_per_second_per_gpu:5088.180588002477 mfu:24.001815966735048 +Step 2854 | loss:1.4923830032348633 lr:9.856007365210594e-05 tokens_per_second_per_gpu:5131.1528743040335 mfu:24.204523573047567 +Step 2855 | loss:1.270573377609253 lr:9.855904032247365e-05 tokens_per_second_per_gpu:5343.606147256438 mfu:25.20670190983023 +Step 2856 | loss:1.4011211395263672 lr:9.855800662762262e-05 tokens_per_second_per_gpu:5283.972201299373 mfu:24.925398412150404 +Step 2857 | loss:1.390272617340088 lr:9.855697256756061e-05 tokens_per_second_per_gpu:3419.3583206633757 mfu:16.129696601257226 +Step 2858 | loss:1.4542324542999268 lr:9.855593814229542e-05 tokens_per_second_per_gpu:4073.8127309058127 mfu:19.216869715807352 +Step 2859 | loss:1.412455677986145 lr:9.855490335183483e-05 tokens_per_second_per_gpu:5332.100694863818 mfu:25.152428727861636 +Step 2860 | loss:1.4071623086929321 lr:9.855386819618662e-05 tokens_per_second_per_gpu:5336.619566303466 mfu:25.173745015441433 +Step 2861 | loss:1.2377270460128784 lr:9.855283267535856e-05 tokens_per_second_per_gpu:4591.536035565719 mfu:21.65905887659249 +Step 2862 | loss:1.2674756050109863 lr:9.855179678935844e-05 tokens_per_second_per_gpu:5008.762206132972 mfu:23.627185909283398 +Step 2863 | loss:1.4961826801300049 lr:9.855076053819409e-05 tokens_per_second_per_gpu:3877.8152642187033 mfu:18.292316224828152 +Step 2864 | loss:1.455460548400879 lr:9.854972392187325e-05 tokens_per_second_per_gpu:4635.156276185835 mfu:21.864822993978056 +Step 2865 | loss:1.3699642419815063 lr:9.854868694040375e-05 tokens_per_second_per_gpu:3574.9402620640285 mfu:16.863603163860546 +Step 2866 | loss:1.55838143825531 lr:9.854764959379339e-05 tokens_per_second_per_gpu:3849.0760962754734 mfu:18.156748666232264 +Step 2867 | loss:1.435196876525879 lr:9.854661188204995e-05 tokens_per_second_per_gpu:2975.38180860225 mfu:14.035383643660827 +Step 2868 | loss:1.370891809463501 lr:9.854557380518125e-05 tokens_per_second_per_gpu:4126.637373627197 mfu:19.466052568337545 +Step 2869 | loss:1.5001003742218018 lr:9.85445353631951e-05 tokens_per_second_per_gpu:4390.307993092983 mfu:20.709831867204233 +Step 2870 | loss:1.2586132287979126 lr:9.854349655609931e-05 tokens_per_second_per_gpu:4941.067530361774 mfu:23.30785857376798 +Step 2871 | loss:1.2909990549087524 lr:9.854245738390167e-05 tokens_per_second_per_gpu:4459.151893004634 mfu:21.034580289068025 +Step 2872 | loss:1.4709879159927368 lr:9.854141784661003e-05 tokens_per_second_per_gpu:4214.239169003327 mfu:19.879285183534265 +Step 2873 | loss:1.2605148553848267 lr:9.854037794423219e-05 tokens_per_second_per_gpu:5361.6613202599665 mfu:25.291871241418043 +Step 2874 | loss:1.4870655536651611 lr:9.853933767677597e-05 tokens_per_second_per_gpu:4114.719589775256 mfu:19.409834348524335 +Step 2875 | loss:1.3497354984283447 lr:9.85382970442492e-05 tokens_per_second_per_gpu:3401.818655948971 mfu:16.046959010223997 +Step 2876 | loss:1.3120557069778442 lr:9.85372560466597e-05 tokens_per_second_per_gpu:5086.521003442063 mfu:23.993987423209255 +Step 2877 | loss:1.5048601627349854 lr:9.853621468401531e-05 tokens_per_second_per_gpu:4954.385578373351 mfu:23.370682078531893 +Step 2878 | loss:1.4507914781570435 lr:9.853517295632385e-05 tokens_per_second_per_gpu:4882.724612001907 mfu:23.032645073536163 +Step 2879 | loss:1.4132052659988403 lr:9.853413086359317e-05 tokens_per_second_per_gpu:4193.358354084671 mfu:19.780786816928572 +Step 2880 | loss:1.3982160091400146 lr:9.853308840583109e-05 tokens_per_second_per_gpu:4458.931728385265 mfu:21.033541734994504 +Step 2881 | loss:1.4221858978271484 lr:9.853204558304545e-05 tokens_per_second_per_gpu:5360.041289728575 mfu:25.284229280995756 +Step 2882 | loss:1.4319618940353394 lr:9.853100239524413e-05 tokens_per_second_per_gpu:4035.526253918039 mfu:19.036265871509457 +Step 2883 | loss:1.2129900455474854 lr:9.852995884243493e-05 tokens_per_second_per_gpu:5155.697326604855 mfu:24.32030394226593 +Step 2884 | loss:1.3381317853927612 lr:9.852891492462571e-05 tokens_per_second_per_gpu:4082.383017128758 mfu:19.257297218162595 +Step 2885 | loss:1.19155752658844 lr:9.852787064182434e-05 tokens_per_second_per_gpu:3890.150421738565 mfu:18.350503267444235 +Step 2886 | loss:1.2149025201797485 lr:9.852682599403867e-05 tokens_per_second_per_gpu:4628.3580995734055 mfu:21.832754835008693 +Step 2887 | loss:1.434700608253479 lr:9.852578098127654e-05 tokens_per_second_per_gpu:4434.987164729344 mfu:20.920591142865696 +Step 2888 | loss:1.155590534210205 lr:9.852473560354582e-05 tokens_per_second_per_gpu:4455.939745217733 mfu:21.019428039908284 +Step 2889 | loss:1.249976634979248 lr:9.852368986085436e-05 tokens_per_second_per_gpu:4664.835314332885 mfu:22.004824080683758 +Step 2890 | loss:1.2681066989898682 lr:9.852264375321006e-05 tokens_per_second_per_gpu:5361.533356850036 mfu:25.29126761618059 +Step 2891 | loss:1.2948161363601685 lr:9.852159728062076e-05 tokens_per_second_per_gpu:5272.909711464351 mfu:24.873214760143206 +Step 2892 | loss:1.245801329612732 lr:9.852055044309432e-05 tokens_per_second_per_gpu:3598.864483673741 mfu:16.976457799086674 +Step 2893 | loss:1.4339429140090942 lr:9.851950324063864e-05 tokens_per_second_per_gpu:5369.95966115532 mfu:25.331015931264375 +Step 2894 | loss:1.3124253749847412 lr:9.851845567326158e-05 tokens_per_second_per_gpu:4625.563377260212 mfu:21.819571653028568 +Step 2895 | loss:1.4335556030273438 lr:9.851740774097102e-05 tokens_per_second_per_gpu:5333.199543677574 mfu:25.15761218519907 +Step 2896 | loss:1.574563980102539 lr:9.851635944377486e-05 tokens_per_second_per_gpu:4649.5595835389795 mfu:21.932765852220477 +Step 2897 | loss:1.3524537086486816 lr:9.851531078168096e-05 tokens_per_second_per_gpu:4359.434058009613 mfu:20.56419424777947 +Step 2898 | loss:1.4173493385314941 lr:9.851426175469722e-05 tokens_per_second_per_gpu:5335.483469715906 mfu:25.168385853999844 +Step 2899 | loss:1.324965476989746 lr:9.851321236283153e-05 tokens_per_second_per_gpu:5335.188188619956 mfu:25.1669929626901 +Step 2900 | loss:1.2311776876449585 lr:9.851216260609177e-05 tokens_per_second_per_gpu:4638.100449649562 mfu:21.878711162534245 +Step 2901 | loss:1.2840715646743774 lr:9.851111248448586e-05 tokens_per_second_per_gpu:5349.176244134443 mfu:25.232977007160304 +Step 2902 | loss:1.5007965564727783 lr:9.851006199802167e-05 tokens_per_second_per_gpu:4435.222494020103 mfu:20.92170123128152 +Step 2903 | loss:1.3008710145950317 lr:9.850901114670712e-05 tokens_per_second_per_gpu:5346.9485552842925 mfu:25.22246861877123 +Step 2904 | loss:1.429583191871643 lr:9.85079599305501e-05 tokens_per_second_per_gpu:4699.934568083772 mfu:22.1703932491769 +Step 2905 | loss:1.2634342908859253 lr:9.850690834955853e-05 tokens_per_second_per_gpu:4485.831347453687 mfu:21.160431827688523 +Step 2906 | loss:1.3767911195755005 lr:9.85058564037403e-05 tokens_per_second_per_gpu:4965.185829240945 mfu:23.421628704587146 +Step 2907 | loss:1.4028123617172241 lr:9.850480409310336e-05 tokens_per_second_per_gpu:5334.2248338204445 mfu:25.162448653735524 +Step 2908 | loss:1.449942708015442 lr:9.850375141765557e-05 tokens_per_second_per_gpu:5106.57687377999 mfu:24.088594385477585 +Step 2909 | loss:1.4137992858886719 lr:9.85026983774049e-05 tokens_per_second_per_gpu:3678.977505392785 mfu:17.354364591226386 +Step 2910 | loss:1.298905611038208 lr:9.850164497235924e-05 tokens_per_second_per_gpu:4373.839928033691 mfu:20.63214919457792 +Step 2911 | loss:1.3510531187057495 lr:9.85005912025265e-05 tokens_per_second_per_gpu:4816.895611459955 mfu:22.722118446394116 +Step 2912 | loss:1.3460876941680908 lr:9.849953706791466e-05 tokens_per_second_per_gpu:3804.806306218794 mfu:17.947920513329862 +Step 2913 | loss:1.279772162437439 lr:9.84984825685316e-05 tokens_per_second_per_gpu:4154.666145688201 mfu:19.598269068350938 +Step 2914 | loss:1.4837133884429932 lr:9.849742770438524e-05 tokens_per_second_per_gpu:4492.216746318304 mfu:21.190552843594304 +Step 2915 | loss:1.485957384109497 lr:9.849637247548356e-05 tokens_per_second_per_gpu:5278.520243089833 mfu:24.89968059507601 +Step 2916 | loss:1.4790133237838745 lr:9.849531688183448e-05 tokens_per_second_per_gpu:5198.686580139159 mfu:24.523091585910226 +Step 2917 | loss:1.3053957223892212 lr:9.849426092344593e-05 tokens_per_second_per_gpu:4455.3449120954765 mfu:21.016622110581604 +Step 2918 | loss:1.466662049293518 lr:9.849320460032585e-05 tokens_per_second_per_gpu:3307.362567235428 mfu:15.601393523891662 +Step 2919 | loss:1.337132215499878 lr:9.849214791248218e-05 tokens_per_second_per_gpu:5350.433329350285 mfu:25.238906892604284 +Step 2920 | loss:1.4551905393600464 lr:9.84910908599229e-05 tokens_per_second_per_gpu:5334.363577222602 mfu:25.163103130035566 +Step 2921 | loss:1.3945406675338745 lr:9.849003344265593e-05 tokens_per_second_per_gpu:4497.24388021247 mfu:21.214266691890572 +Step 2922 | loss:1.4149943590164185 lr:9.848897566068923e-05 tokens_per_second_per_gpu:3912.019581970889 mfu:18.453663827523595 +Step 2923 | loss:1.3135113716125488 lr:9.848791751403074e-05 tokens_per_second_per_gpu:3798.2409044767214 mfu:17.91695039313946 +Step 2924 | loss:1.382861852645874 lr:9.848685900268846e-05 tokens_per_second_per_gpu:5234.633060441004 mfu:24.692657266595997 +Step 2925 | loss:1.2935534715652466 lr:9.848580012667031e-05 tokens_per_second_per_gpu:5296.5068614200545 mfu:24.9845265425732 +Step 2926 | loss:1.395658016204834 lr:9.848474088598428e-05 tokens_per_second_per_gpu:3296.286867429897 mfu:15.549147558199724 +Step 2927 | loss:1.395078420639038 lr:9.848368128063832e-05 tokens_per_second_per_gpu:5056.779119238366 mfu:23.8536898809317 +Step 2928 | loss:1.4891092777252197 lr:9.848262131064041e-05 tokens_per_second_per_gpu:4540.696004514918 mfu:21.419237775922102 +Step 2929 | loss:1.3081821203231812 lr:9.848156097599852e-05 tokens_per_second_per_gpu:4471.3824974792715 mfu:21.09227413713119 +Step 2930 | loss:1.207061529159546 lr:9.848050027672062e-05 tokens_per_second_per_gpu:4326.8278513290925 mfu:20.410385207674935 +Step 2931 | loss:1.3318594694137573 lr:9.847943921281471e-05 tokens_per_second_per_gpu:3849.8238252440133 mfu:18.160275831352585 +Step 2932 | loss:1.35016930103302 lr:9.847837778428872e-05 tokens_per_second_per_gpu:4100.073782703715 mfu:19.34074757773552 +Step 2933 | loss:1.3095674514770508 lr:9.847731599115068e-05 tokens_per_second_per_gpu:5316.0782162465075 mfu:25.076847962507042 +Step 2934 | loss:1.2680903673171997 lr:9.847625383340857e-05 tokens_per_second_per_gpu:5355.314145396763 mfu:25.261930534648606 +Step 2935 | loss:1.4394333362579346 lr:9.847519131107037e-05 tokens_per_second_per_gpu:4242.312373916501 mfu:20.01171128089237 +Step 2936 | loss:1.3703845739364624 lr:9.847412842414405e-05 tokens_per_second_per_gpu:4742.861939374765 mfu:22.372888983721303 +Step 2937 | loss:1.3382192850112915 lr:9.847306517263764e-05 tokens_per_second_per_gpu:5344.2022342558475 mfu:25.20951375390937 +Step 2938 | loss:1.2959481477737427 lr:9.847200155655911e-05 tokens_per_second_per_gpu:4586.400287085652 mfu:21.634832674762965 +Step 2939 | loss:1.2451244592666626 lr:9.847093757591649e-05 tokens_per_second_per_gpu:5342.707245811117 mfu:25.202461638344246 +Step 2940 | loss:1.2804009914398193 lr:9.846987323071775e-05 tokens_per_second_per_gpu:4313.5744549078345 mfu:20.347866675493258 +Step 2941 | loss:1.3497285842895508 lr:9.846880852097091e-05 tokens_per_second_per_gpu:5043.171494990011 mfu:23.78950039565211 +Step 2942 | loss:1.4699976444244385 lr:9.846774344668398e-05 tokens_per_second_per_gpu:3993.046944226011 mfu:18.835883719974564 +Step 2943 | loss:1.4423171281814575 lr:9.846667800786497e-05 tokens_per_second_per_gpu:4257.527236949727 mfu:20.083482385743412 +Step 2944 | loss:1.2802926301956177 lr:9.846561220452189e-05 tokens_per_second_per_gpu:4870.038707929381 mfu:22.97280349958763 +Step 2945 | loss:1.277111291885376 lr:9.846454603666276e-05 tokens_per_second_per_gpu:3961.48600285656 mfu:18.687005374682997 +Step 2946 | loss:1.2909040451049805 lr:9.846347950429557e-05 tokens_per_second_per_gpu:5355.736717602269 mfu:25.263923879840235 +Step 2947 | loss:1.4904346466064453 lr:9.846241260742839e-05 tokens_per_second_per_gpu:4310.7876246513 mfu:20.33472072169091 +Step 2948 | loss:1.475787878036499 lr:9.846134534606921e-05 tokens_per_second_per_gpu:5320.774594872712 mfu:25.099001581771674 +Step 2949 | loss:1.458770990371704 lr:9.846027772022608e-05 tokens_per_second_per_gpu:4684.471291972469 mfu:22.097450337452926 +Step 2950 | loss:1.3005667924880981 lr:9.845920972990701e-05 tokens_per_second_per_gpu:3949.5132111750268 mfu:18.630527673552304 +Step 2951 | loss:1.2975667715072632 lr:9.845814137512004e-05 tokens_per_second_per_gpu:3404.0390165393815 mfu:16.05743282996154 +Step 2952 | loss:1.4641345739364624 lr:9.84570726558732e-05 tokens_per_second_per_gpu:5389.9436868652065 mfu:25.42528399016429 +Step 2953 | loss:1.3004639148712158 lr:9.845600357217453e-05 tokens_per_second_per_gpu:4166.178321782723 mfu:19.65257396716858 +Step 2954 | loss:1.4088443517684937 lr:9.845493412403209e-05 tokens_per_second_per_gpu:3989.953108578343 mfu:18.821289569361724 +Step 2955 | loss:1.3721799850463867 lr:9.84538643114539e-05 tokens_per_second_per_gpu:4600.370491954744 mfu:21.700732514693204 +Step 2956 | loss:1.4556926488876343 lr:9.8452794134448e-05 tokens_per_second_per_gpu:4300.899474118552 mfu:20.28807662853549 +Step 2957 | loss:1.4329601526260376 lr:9.845172359302248e-05 tokens_per_second_per_gpu:4279.742101600044 mfu:20.188273692545707 +Step 2958 | loss:1.336975336074829 lr:9.845065268718537e-05 tokens_per_second_per_gpu:4375.621576171987 mfu:20.640553532826782 +Step 2959 | loss:1.3522342443466187 lr:9.844958141694469e-05 tokens_per_second_per_gpu:3437.025012566051 mfu:16.213033401210605 +Step 2960 | loss:1.3592774868011475 lr:9.844850978230853e-05 tokens_per_second_per_gpu:5317.972230826957 mfu:25.08578235243523 +Step 2961 | loss:1.5951143503189087 lr:9.844743778328495e-05 tokens_per_second_per_gpu:4660.09439594288 mfu:21.982460359754015 +Step 2962 | loss:1.32439124584198 lr:9.844636541988202e-05 tokens_per_second_per_gpu:5284.737953315677 mfu:24.929010595062074 +Step 2963 | loss:1.3551108837127686 lr:9.844529269210778e-05 tokens_per_second_per_gpu:4922.9873679329585 mfu:23.222571362797456 +Step 2964 | loss:1.3611160516738892 lr:9.844421959997032e-05 tokens_per_second_per_gpu:5312.2488377724485 mfu:25.058784130885567 +Step 2965 | loss:1.3893963098526 lr:9.84431461434777e-05 tokens_per_second_per_gpu:4116.345061808109 mfu:19.41750197743743 +Step 2966 | loss:1.4174503087997437 lr:9.8442072322638e-05 tokens_per_second_per_gpu:5340.332077101901 mfu:25.191257562297306 +Step 2967 | loss:1.3738881349563599 lr:9.844099813745928e-05 tokens_per_second_per_gpu:5066.843925189256 mfu:23.901167287835023 +Step 2968 | loss:1.5021437406539917 lr:9.843992358794965e-05 tokens_per_second_per_gpu:3995.8283564834023 mfu:18.84900411615055 +Step 2969 | loss:1.3334518671035767 lr:9.843884867411716e-05 tokens_per_second_per_gpu:3472.736398732442 mfu:16.38149010274824 +Step 2970 | loss:1.4816855192184448 lr:9.843777339596991e-05 tokens_per_second_per_gpu:3366.8404986787405 mfu:15.881961074491299 +Step 2971 | loss:1.4642363786697388 lr:9.843669775351598e-05 tokens_per_second_per_gpu:5334.196444927483 mfu:25.162314738483783 +Step 2972 | loss:1.4740359783172607 lr:9.843562174676348e-05 tokens_per_second_per_gpu:4347.634036277226 mfu:20.50853153197649 +Step 2973 | loss:1.3980668783187866 lr:9.843454537572048e-05 tokens_per_second_per_gpu:4407.68259804631 mfu:20.791790843182362 +Step 2974 | loss:1.249220371246338 lr:9.843346864039508e-05 tokens_per_second_per_gpu:4992.997078780318 mfu:23.552819113752836 +Step 2975 | loss:1.251694917678833 lr:9.84323915407954e-05 tokens_per_second_per_gpu:3248.4470973717907 mfu:15.323479200529196 +Step 2976 | loss:1.1749954223632812 lr:9.843131407692951e-05 tokens_per_second_per_gpu:4354.0719352342285 mfu:20.538900199775497 +Step 2977 | loss:1.1495503187179565 lr:9.843023624880553e-05 tokens_per_second_per_gpu:4305.571877834653 mfu:20.310117154058617 +Step 2978 | loss:1.2541272640228271 lr:9.842915805643155e-05 tokens_per_second_per_gpu:3999.024229817318 mfu:18.864079595938843 +Step 2979 | loss:1.39181387424469 lr:9.842807949981572e-05 tokens_per_second_per_gpu:4216.877605373583 mfu:19.891731138056425 +Step 2980 | loss:1.253255844116211 lr:9.84270005789661e-05 tokens_per_second_per_gpu:5341.818067463983 mfu:25.19826723237135 +Step 2981 | loss:1.2127060890197754 lr:9.842592129389085e-05 tokens_per_second_per_gpu:4282.437065114921 mfu:20.200986295253575 +Step 2982 | loss:1.3399592638015747 lr:9.842484164459804e-05 tokens_per_second_per_gpu:4590.8292075984655 mfu:21.65572464847366 +Step 2983 | loss:1.3891968727111816 lr:9.842376163109584e-05 tokens_per_second_per_gpu:3574.7133717536813 mfu:16.862532883555158 +Step 2984 | loss:1.1978139877319336 lr:9.842268125339235e-05 tokens_per_second_per_gpu:4202.002271555931 mfu:19.821561650444902 +Step 2985 | loss:1.306717038154602 lr:9.842160051149568e-05 tokens_per_second_per_gpu:4318.622245573718 mfu:20.371677965306127 +Step 2986 | loss:1.4581609964370728 lr:9.842051940541399e-05 tokens_per_second_per_gpu:4357.406878611283 mfu:20.554631696685075 +Step 2987 | loss:1.425193190574646 lr:9.841943793515538e-05 tokens_per_second_per_gpu:4534.4944105875065 mfu:21.389983799265472 +Step 2988 | loss:1.2438691854476929 lr:9.8418356100728e-05 tokens_per_second_per_gpu:5197.772957062791 mfu:24.518781869978827 +Step 2989 | loss:1.2356667518615723 lr:9.841727390214e-05 tokens_per_second_per_gpu:3994.808014296988 mfu:18.8441909879689 +Step 2990 | loss:1.3695058822631836 lr:9.84161913393995e-05 tokens_per_second_per_gpu:5345.148092824812 mfu:25.213975530159292 +Step 2991 | loss:1.42966628074646 lr:9.841510841251463e-05 tokens_per_second_per_gpu:5307.103769354512 mfu:25.03451396531897 +Step 2992 | loss:1.2414052486419678 lr:9.841402512149357e-05 tokens_per_second_per_gpu:3909.630179050451 mfu:18.442392606273714 +Step 2993 | loss:1.347751259803772 lr:9.841294146634444e-05 tokens_per_second_per_gpu:4880.547294578967 mfu:23.02237429576349 +Step 2994 | loss:1.4809707403182983 lr:9.841185744707542e-05 tokens_per_second_per_gpu:5339.715535136884 mfu:25.188349228655415 +Step 2995 | loss:1.445970058441162 lr:9.841077306369462e-05 tokens_per_second_per_gpu:4194.315962733748 mfu:19.785304020311152 +Step 2996 | loss:1.2861875295639038 lr:9.840968831621022e-05 tokens_per_second_per_gpu:3907.3166412734945 mfu:18.43147925384956 +Step 2997 | loss:1.171366810798645 lr:9.84086032046304e-05 tokens_per_second_per_gpu:5329.298833269008 mfu:25.13921186867204 +Step 2998 | loss:1.2445292472839355 lr:9.840751772896327e-05 tokens_per_second_per_gpu:3881.861212564292 mfu:18.311401653484268 +Step 2999 | loss:1.2926650047302246 lr:9.840643188921704e-05 tokens_per_second_per_gpu:5344.518553246023 mfu:25.211005884555615 +Step 3000 | loss:1.2632282972335815 lr:9.840534568539982e-05 tokens_per_second_per_gpu:5361.541120105447 mfu:25.29130423678091 +Step 3001 | loss:1.5096360445022583 lr:9.840425911751987e-05 tokens_per_second_per_gpu:4324.314860272122 mfu:20.39853100009872 +Step 3002 | loss:1.4673575162887573 lr:9.840317218558529e-05 tokens_per_second_per_gpu:4305.34306188394 mfu:20.309037790179083 +Step 3003 | loss:1.293782114982605 lr:9.840208488960427e-05 tokens_per_second_per_gpu:5156.5993477829115 mfu:24.324558930063507 +Step 3004 | loss:1.2398993968963623 lr:9.840099722958501e-05 tokens_per_second_per_gpu:3861.3933104773746 mfu:18.214851067156076 +Step 3005 | loss:1.4764690399169922 lr:9.839990920553565e-05 tokens_per_second_per_gpu:5320.227208465545 mfu:25.096419466695313 +Step 3006 | loss:1.3793187141418457 lr:9.839882081746443e-05 tokens_per_second_per_gpu:4749.264976774069 mfu:22.40309320360465 +Step 3007 | loss:1.4138725996017456 lr:9.839773206537948e-05 tokens_per_second_per_gpu:5304.550937202013 mfu:25.022471820497554 +Step 3008 | loss:1.3661655187606812 lr:9.839664294928901e-05 tokens_per_second_per_gpu:4939.597111222202 mfu:23.30092235580725 +Step 3009 | loss:1.3397128582000732 lr:9.839555346920122e-05 tokens_per_second_per_gpu:5349.662022883049 mfu:25.23526850839988 +Step 3010 | loss:1.3334980010986328 lr:9.83944636251243e-05 tokens_per_second_per_gpu:4619.951430177974 mfu:21.79309914114474 +Step 3011 | loss:1.4018949270248413 lr:9.839337341706644e-05 tokens_per_second_per_gpu:5164.395221819509 mfu:24.361333398007403 +Step 3012 | loss:1.327380895614624 lr:9.839228284503583e-05 tokens_per_second_per_gpu:4084.002402207007 mfu:19.264936133872283 +Step 3013 | loss:1.3483246564865112 lr:9.839119190904072e-05 tokens_per_second_per_gpu:5322.178903833765 mfu:25.10562594674077 +Step 3014 | loss:1.4645941257476807 lr:9.839010060908924e-05 tokens_per_second_per_gpu:4659.7353673643875 mfu:21.980766760692674 +Step 3015 | loss:1.3026678562164307 lr:9.838900894518968e-05 tokens_per_second_per_gpu:4586.288411892613 mfu:21.63430494038952 +Step 3016 | loss:1.4194049835205078 lr:9.838791691735017e-05 tokens_per_second_per_gpu:5347.173993571873 mfu:25.223532049637488 +Step 3017 | loss:1.3402546644210815 lr:9.838682452557899e-05 tokens_per_second_per_gpu:4991.12460614181 mfu:23.543986340840217 +Step 3018 | loss:1.3104403018951416 lr:9.838573176988431e-05 tokens_per_second_per_gpu:3433.033943052474 mfu:16.194206845368385 +Step 3019 | loss:1.3379993438720703 lr:9.838463865027439e-05 tokens_per_second_per_gpu:5339.989196967631 mfu:25.189640138202574 +Step 3020 | loss:1.4327739477157593 lr:9.838354516675741e-05 tokens_per_second_per_gpu:5326.431727145828 mfu:25.12568723991025 +Step 3021 | loss:1.2417919635772705 lr:9.838245131934161e-05 tokens_per_second_per_gpu:5197.23935644952 mfu:24.516264784844378 +Step 3022 | loss:1.3954901695251465 lr:9.838135710803523e-05 tokens_per_second_per_gpu:4188.527529539342 mfu:19.757998993324595 +Step 3023 | loss:1.276960015296936 lr:9.838026253284649e-05 tokens_per_second_per_gpu:4368.410702017453 mfu:20.606538609137854 +Step 3024 | loss:1.3466540575027466 lr:9.837916759378362e-05 tokens_per_second_per_gpu:4562.900400990136 mfu:21.5239797025564 +Step 3025 | loss:1.4544472694396973 lr:9.837807229085485e-05 tokens_per_second_per_gpu:5330.891588309121 mfu:25.146725165948286 +Step 3026 | loss:1.3314658403396606 lr:9.837697662406844e-05 tokens_per_second_per_gpu:4321.995161884886 mfu:20.387588587024773 +Step 3027 | loss:1.4644885063171387 lr:9.837588059343261e-05 tokens_per_second_per_gpu:4778.569275427159 mfu:22.54132658018019 +Step 3028 | loss:1.2883602380752563 lr:9.83747841989556e-05 tokens_per_second_per_gpu:3657.352801905666 mfu:17.252357175322775 +Step 3029 | loss:1.4918972253799438 lr:9.837368744064568e-05 tokens_per_second_per_gpu:4628.349232791233 mfu:21.832713008884355 +Step 3030 | loss:1.5241408348083496 lr:9.837259031851107e-05 tokens_per_second_per_gpu:5297.0796331649435 mfu:24.98722840461878 +Step 3031 | loss:1.3510422706604004 lr:9.837149283256004e-05 tokens_per_second_per_gpu:4952.273688748232 mfu:23.360719934844685 +Step 3032 | loss:1.1794495582580566 lr:9.837039498280086e-05 tokens_per_second_per_gpu:4747.668791658113 mfu:22.395563726075427 +Step 3033 | loss:1.2480266094207764 lr:9.836929676924174e-05 tokens_per_second_per_gpu:4461.964587844759 mfu:21.04784825052412 +Step 3034 | loss:1.2691024541854858 lr:9.836819819189098e-05 tokens_per_second_per_gpu:4725.270515452792 mfu:22.28990723567494 +Step 3035 | loss:1.276916742324829 lr:9.836709925075683e-05 tokens_per_second_per_gpu:5383.439479058492 mfu:25.394602532207735 +Step 3036 | loss:1.3430501222610474 lr:9.836599994584754e-05 tokens_per_second_per_gpu:5348.592109875752 mfu:25.230221546198475 +Step 3037 | loss:1.4603369235992432 lr:9.83649002771714e-05 tokens_per_second_per_gpu:5320.199039247658 mfu:25.096286587688375 +Step 3038 | loss:1.1936662197113037 lr:9.836380024473668e-05 tokens_per_second_per_gpu:5283.107055614667 mfu:24.921317372346568 +Step 3039 | loss:1.4250736236572266 lr:9.836269984855166e-05 tokens_per_second_per_gpu:5181.826006078859 mfu:24.44355738135722 +Step 3040 | loss:1.3628889322280884 lr:9.836159908862458e-05 tokens_per_second_per_gpu:3886.5392020131553 mfu:18.333468527861857 +Step 3041 | loss:1.2834824323654175 lr:9.836049796496375e-05 tokens_per_second_per_gpu:4315.113720051539 mfu:20.355127651801823 +Step 3042 | loss:1.5383754968643188 lr:9.835939647757745e-05 tokens_per_second_per_gpu:4458.365189178016 mfu:21.030869272892353 +Step 3043 | loss:1.3674226999282837 lr:9.835829462647394e-05 tokens_per_second_per_gpu:5345.029309123013 mfu:25.213415207171316 +Step 3044 | loss:1.251255989074707 lr:9.835719241166155e-05 tokens_per_second_per_gpu:5322.922725327705 mfu:25.10913467963672 +Step 3045 | loss:1.3624476194381714 lr:9.835608983314853e-05 tokens_per_second_per_gpu:4203.842674130901 mfu:19.830243143396096 +Step 3046 | loss:1.4862738847732544 lr:9.835498689094321e-05 tokens_per_second_per_gpu:4196.606628092949 mfu:19.796109479638314 +Step 3047 | loss:1.2459697723388672 lr:9.835388358505384e-05 tokens_per_second_per_gpu:4588.119965498907 mfu:21.642944691245678 +Step 3048 | loss:1.3852697610855103 lr:9.835277991548874e-05 tokens_per_second_per_gpu:5338.126064277363 mfu:25.180851423419362 +Step 3049 | loss:1.3666956424713135 lr:9.835167588225622e-05 tokens_per_second_per_gpu:3561.435601042759 mfu:16.799899373690454 +Step 3050 | loss:1.3035482168197632 lr:9.835057148536457e-05 tokens_per_second_per_gpu:5314.717625720471 mfu:25.070429824855047 +Step 3051 | loss:1.4583005905151367 lr:9.834946672482211e-05 tokens_per_second_per_gpu:4403.947726110574 mfu:20.774172815026635 +Step 3052 | loss:1.1851911544799805 lr:9.834836160063714e-05 tokens_per_second_per_gpu:4462.32485212933 mfu:21.049547678621863 +Step 3053 | loss:1.3637715578079224 lr:9.834725611281798e-05 tokens_per_second_per_gpu:4867.726907688072 mfu:22.96189834341558 +Step 3054 | loss:1.421657681465149 lr:9.834615026137292e-05 tokens_per_second_per_gpu:4158.627976375009 mfu:19.616957699658627 +Step 3055 | loss:1.2325867414474487 lr:9.834504404631031e-05 tokens_per_second_per_gpu:4821.090699964149 mfu:22.741907394624302 +Step 3056 | loss:1.4688479900360107 lr:9.834393746763845e-05 tokens_per_second_per_gpu:5350.755550345931 mfu:25.240426864017316 +Step 3057 | loss:1.345272183418274 lr:9.834283052536566e-05 tokens_per_second_per_gpu:5166.013129381467 mfu:24.36896534402065 +Step 3058 | loss:1.2832919359207153 lr:9.834172321950028e-05 tokens_per_second_per_gpu:4686.033628588851 mfu:22.104820145834513 +Step 3059 | loss:1.3309084177017212 lr:9.834061555005063e-05 tokens_per_second_per_gpu:5042.068727559687 mfu:23.78429845353175 +Step 3060 | loss:1.3948824405670166 lr:9.833950751702505e-05 tokens_per_second_per_gpu:5214.341055300846 mfu:24.596936416176785 +Step 3061 | loss:1.3903205394744873 lr:9.833839912043186e-05 tokens_per_second_per_gpu:4735.487700446049 mfu:22.33810344051965 +Step 3062 | loss:1.54743230342865 lr:9.833729036027942e-05 tokens_per_second_per_gpu:4703.534405124734 mfu:22.187374294694653 +Step 3063 | loss:1.2616580724716187 lr:9.833618123657601e-05 tokens_per_second_per_gpu:5360.346815383199 mfu:25.285670497636673 +Step 3064 | loss:1.2509334087371826 lr:9.833507174933004e-05 tokens_per_second_per_gpu:4751.807702295696 mfu:22.415087673723875 +Step 3065 | loss:1.488649606704712 lr:9.833396189854982e-05 tokens_per_second_per_gpu:3754.623695253247 mfu:17.71120057536906 +Step 3066 | loss:1.4353821277618408 lr:9.833285168424371e-05 tokens_per_second_per_gpu:4588.9978982748935 mfu:21.647086049940786 +Step 3067 | loss:1.374011516571045 lr:9.833174110642007e-05 tokens_per_second_per_gpu:4693.63069067008 mfu:22.140656783864124 +Step 3068 | loss:1.3442668914794922 lr:9.833063016508721e-05 tokens_per_second_per_gpu:5256.255597832022 mfu:24.794654464654727 +Step 3069 | loss:1.3439604043960571 lr:9.832951886025352e-05 tokens_per_second_per_gpu:4675.537201052462 mfu:22.05530670627016 +Step 3070 | loss:1.3078910112380981 lr:9.832840719192736e-05 tokens_per_second_per_gpu:4256.086860342829 mfu:20.076687883535087 +Step 3071 | loss:1.247544288635254 lr:9.832729516011709e-05 tokens_per_second_per_gpu:5336.19391931785 mfu:25.171737165987413 +Step 3072 | loss:1.4193952083587646 lr:9.832618276483105e-05 tokens_per_second_per_gpu:5059.5501185355115 mfu:23.86676115739757 +Step 3073 | loss:1.4364873170852661 lr:9.832507000607763e-05 tokens_per_second_per_gpu:5321.293901849792 mfu:25.101451241385462 +Step 3074 | loss:1.2676904201507568 lr:9.83239568838652e-05 tokens_per_second_per_gpu:5303.2615400878485 mfu:25.016389514316206 +Step 3075 | loss:1.2542486190795898 lr:9.83228433982021e-05 tokens_per_second_per_gpu:5175.882954678863 mfu:24.41552299391431 +Step 3076 | loss:1.2946163415908813 lr:9.832172954909675e-05 tokens_per_second_per_gpu:4081.103657156213 mfu:19.251262258890783 +Step 3077 | loss:1.4475377798080444 lr:9.83206153365575e-05 tokens_per_second_per_gpu:5187.148511873393 mfu:24.468664549303167 +Step 3078 | loss:1.2120013236999512 lr:9.831950076059275e-05 tokens_per_second_per_gpu:5313.771782460587 mfu:25.065968120820674 +Step 3079 | loss:1.2343469858169556 lr:9.831838582121084e-05 tokens_per_second_per_gpu:4985.505418575353 mfu:23.517479674356984 +Step 3080 | loss:1.4582183361053467 lr:9.83172705184202e-05 tokens_per_second_per_gpu:3549.6420393730614 mfu:16.744267131105378 +Step 3081 | loss:1.3033263683319092 lr:9.831615485222921e-05 tokens_per_second_per_gpu:4688.00555356124 mfu:22.114122052374366 +Step 3082 | loss:1.31960129737854 lr:9.831503882264625e-05 tokens_per_second_per_gpu:5254.977254696694 mfu:24.788624301977553 +Step 3083 | loss:1.4482440948486328 lr:9.831392242967972e-05 tokens_per_second_per_gpu:3650.261198821026 mfu:17.218904873620275 +Step 3084 | loss:1.4648628234863281 lr:9.8312805673338e-05 tokens_per_second_per_gpu:4545.650955593892 mfu:21.442611125563452 +Step 3085 | loss:1.3112894296646118 lr:9.831168855362952e-05 tokens_per_second_per_gpu:5299.872615394879 mfu:25.000403378329295 +Step 3086 | loss:1.3834835290908813 lr:9.831057107056268e-05 tokens_per_second_per_gpu:5347.632177173298 mfu:25.225693379859596 +Step 3087 | loss:1.410000205039978 lr:9.830945322414585e-05 tokens_per_second_per_gpu:5336.36508356838 mfu:25.1725445769608 +Step 3088 | loss:1.3911768198013306 lr:9.830833501438746e-05 tokens_per_second_per_gpu:5056.424522320774 mfu:23.852017186770947 +Step 3089 | loss:1.4578417539596558 lr:9.83072164412959e-05 tokens_per_second_per_gpu:4935.408122073966 mfu:23.281162179279768 +Step 3090 | loss:1.4252547025680542 lr:9.830609750487962e-05 tokens_per_second_per_gpu:4952.316964261133 mfu:23.360924072822595 +Step 3091 | loss:1.400916337966919 lr:9.830497820514701e-05 tokens_per_second_per_gpu:3777.4338723253354 mfu:17.818800072435618 +Step 3092 | loss:1.0844601392745972 lr:9.83038585421065e-05 tokens_per_second_per_gpu:3665.493021129048 mfu:17.29075597826408 +Step 3093 | loss:1.2777478694915771 lr:9.830273851576651e-05 tokens_per_second_per_gpu:4771.42115773778 mfu:22.507607689444754 +Step 3094 | loss:1.4936002492904663 lr:9.830161812613545e-05 tokens_per_second_per_gpu:4888.066699484742 mfu:23.0578446116468 +Step 3095 | loss:1.4043188095092773 lr:9.830049737322178e-05 tokens_per_second_per_gpu:4025.2303406990204 mfu:18.98769829219141 +Step 3096 | loss:1.239870309829712 lr:9.829937625703389e-05 tokens_per_second_per_gpu:4238.097507966835 mfu:19.991829039077512 +Step 3097 | loss:1.3602323532104492 lr:9.829825477758022e-05 tokens_per_second_per_gpu:4743.00354413509 mfu:22.37355695753602 +Step 3098 | loss:1.39316987991333 lr:9.829713293486922e-05 tokens_per_second_per_gpu:3750.867496262115 mfu:17.693481943854252 +Step 3099 | loss:1.3275631666183472 lr:9.829601072890933e-05 tokens_per_second_per_gpu:3859.720749812957 mfu:18.206961313133743 +Step 3100 | loss:1.2663745880126953 lr:9.829488815970896e-05 tokens_per_second_per_gpu:3783.178057687336 mfu:17.845896374847257 +Step 3101 | loss:1.494296908378601 lr:9.82937652272766e-05 tokens_per_second_per_gpu:5325.888087469446 mfu:25.123122798802264 +Step 3102 | loss:1.1838306188583374 lr:9.829264193162064e-05 tokens_per_second_per_gpu:5343.567165207513 mfu:25.206518024854425 +Step 3103 | loss:1.3191912174224854 lr:9.829151827274958e-05 tokens_per_second_per_gpu:5355.285472984914 mfu:25.2617952819899 +Step 3104 | loss:1.4849714040756226 lr:9.829039425067183e-05 tokens_per_second_per_gpu:4676.8770582887955 mfu:22.061627041457164 +Step 3105 | loss:1.3129944801330566 lr:9.828926986539588e-05 tokens_per_second_per_gpu:5347.95253264683 mfu:25.227204551286523 +Step 3106 | loss:1.234932780265808 lr:9.828814511693017e-05 tokens_per_second_per_gpu:4874.1127926659265 mfu:22.99202165240045 +Step 3107 | loss:1.3677992820739746 lr:9.828702000528315e-05 tokens_per_second_per_gpu:5198.989953287941 mfu:24.52452264881383 +Step 3108 | loss:1.4179348945617676 lr:9.828589453046329e-05 tokens_per_second_per_gpu:4882.93137375869 mfu:23.0336204040201 +Step 3109 | loss:1.3428040742874146 lr:9.828476869247906e-05 tokens_per_second_per_gpu:5364.3170183218035 mfu:25.304398622284538 +Step 3110 | loss:1.315192461013794 lr:9.828364249133892e-05 tokens_per_second_per_gpu:4964.388182978794 mfu:23.417866071076098 +Step 3111 | loss:1.4989231824874878 lr:9.828251592705136e-05 tokens_per_second_per_gpu:4234.968215047458 mfu:19.977067630464383 +Step 3112 | loss:1.4383115768432617 lr:9.828138899962482e-05 tokens_per_second_per_gpu:4670.655229251972 mfu:22.03227761233697 +Step 3113 | loss:1.2415375709533691 lr:9.828026170906779e-05 tokens_per_second_per_gpu:3775.768621476166 mfu:17.810944800058994 +Step 3114 | loss:1.210073471069336 lr:9.827913405538876e-05 tokens_per_second_per_gpu:3948.452842162538 mfu:18.62552573200278 +Step 3115 | loss:1.4720112085342407 lr:9.82780060385962e-05 tokens_per_second_per_gpu:4116.640830706727 mfu:19.41889716979559 +Step 3116 | loss:1.3549399375915527 lr:9.82768776586986e-05 tokens_per_second_per_gpu:5383.39700275387 mfu:25.394402164231625 +Step 3117 | loss:1.3283134698867798 lr:9.827574891570444e-05 tokens_per_second_per_gpu:5355.3223715729655 mfu:25.261969338925287 +Step 3118 | loss:1.4190938472747803 lr:9.827461980962221e-05 tokens_per_second_per_gpu:4228.589082823266 mfu:19.94697617537068 +Step 3119 | loss:1.4007277488708496 lr:9.82734903404604e-05 tokens_per_second_per_gpu:3679.872918587653 mfu:17.35858840803982 +Step 3120 | loss:1.4151362180709839 lr:9.827236050822753e-05 tokens_per_second_per_gpu:3165.458004541506 mfu:14.932005490249436 +Step 3121 | loss:1.4198505878448486 lr:9.827123031293206e-05 tokens_per_second_per_gpu:4625.151882180915 mfu:21.81763056053122 +Step 3122 | loss:1.1953002214431763 lr:9.82700997545825e-05 tokens_per_second_per_gpu:3331.4352074380167 mfu:15.714948274943215 +Step 3123 | loss:1.2697758674621582 lr:9.826896883318737e-05 tokens_per_second_per_gpu:5322.829875782062 mfu:25.10869669248816 +Step 3124 | loss:1.3824305534362793 lr:9.826783754875518e-05 tokens_per_second_per_gpu:4666.066298400091 mfu:22.010630842556306 +Step 3125 | loss:1.3104008436203003 lr:9.82667059012944e-05 tokens_per_second_per_gpu:5383.031812276944 mfu:25.392679498444046 +Step 3126 | loss:1.2333080768585205 lr:9.826557389081358e-05 tokens_per_second_per_gpu:4422.974598636344 mfu:20.863925819049893 +Step 3127 | loss:1.3704068660736084 lr:9.826444151732122e-05 tokens_per_second_per_gpu:4567.516978358685 mfu:21.545756885673168 +Step 3128 | loss:1.331868290901184 lr:9.826330878082584e-05 tokens_per_second_per_gpu:4030.5889964526395 mfu:19.01297598566218 +Step 3129 | loss:1.294663429260254 lr:9.826217568133595e-05 tokens_per_second_per_gpu:4026.232538914524 mfu:18.99242583216659 +Step 3130 | loss:1.4468108415603638 lr:9.826104221886008e-05 tokens_per_second_per_gpu:3983.725790903347 mfu:18.79191424939881 +Step 3131 | loss:1.3353253602981567 lr:9.825990839340676e-05 tokens_per_second_per_gpu:5254.406847242643 mfu:24.785933592695248 +Step 3132 | loss:1.279652714729309 lr:9.825877420498448e-05 tokens_per_second_per_gpu:4706.008891137663 mfu:22.19904686740849 +Step 3133 | loss:1.3573392629623413 lr:9.825763965360185e-05 tokens_per_second_per_gpu:4354.079229216834 mfu:20.538934606735932 +Step 3134 | loss:1.2216238975524902 lr:9.825650473926731e-05 tokens_per_second_per_gpu:4675.83724390111 mfu:22.056722059579823 +Step 3135 | loss:1.3919789791107178 lr:9.825536946198947e-05 tokens_per_second_per_gpu:4234.543205371981 mfu:19.975062787311025 +Step 3136 | loss:1.3683847188949585 lr:9.825423382177683e-05 tokens_per_second_per_gpu:5352.292486749803 mfu:25.247676855262775 +Step 3137 | loss:1.3461592197418213 lr:9.825309781863792e-05 tokens_per_second_per_gpu:4956.625631992312 mfu:23.381248793645092 +Step 3138 | loss:1.1903462409973145 lr:9.825196145258133e-05 tokens_per_second_per_gpu:3852.1710608788694 mfu:18.171348142321897 +Step 3139 | loss:1.2683522701263428 lr:9.825082472361557e-05 tokens_per_second_per_gpu:5256.106615319435 mfu:24.793951688723816 +Step 3140 | loss:1.3494136333465576 lr:9.82496876317492e-05 tokens_per_second_per_gpu:3976.529409322665 mfu:18.757967689653423 +Step 3141 | loss:1.3055121898651123 lr:9.82485501769908e-05 tokens_per_second_per_gpu:5207.48343974618 mfu:24.56458787357556 +Step 3142 | loss:1.3400181531906128 lr:9.824741235934887e-05 tokens_per_second_per_gpu:5327.34041464588 mfu:25.12997367388607 +Step 3143 | loss:1.3463364839553833 lr:9.824627417883199e-05 tokens_per_second_per_gpu:4700.822405417511 mfu:22.174581329360816 +Step 3144 | loss:1.2709389925003052 lr:9.824513563544873e-05 tokens_per_second_per_gpu:3205.061533086114 mfu:15.118822091453088 +Step 3145 | loss:1.1922998428344727 lr:9.824399672920766e-05 tokens_per_second_per_gpu:4458.988715049722 mfu:21.03381055081367 +Step 3146 | loss:1.5263620615005493 lr:9.824285746011733e-05 tokens_per_second_per_gpu:3422.9300399652093 mfu:16.14654501060213 +Step 3147 | loss:1.4111979007720947 lr:9.82417178281863e-05 tokens_per_second_per_gpu:4549.34665859188 mfu:21.46004438715692 +Step 3148 | loss:1.4395195245742798 lr:9.824057783342318e-05 tokens_per_second_per_gpu:3280.3881195838226 mfu:15.474150452003695 +Step 3149 | loss:1.2159440517425537 lr:9.82394374758365e-05 tokens_per_second_per_gpu:5232.215982818064 mfu:24.681255499052618 +Step 3150 | loss:1.5102190971374512 lr:9.823829675543486e-05 tokens_per_second_per_gpu:3146.1331874672865 mfu:14.840847030956557 +Step 3151 | loss:1.4916146993637085 lr:9.823715567222685e-05 tokens_per_second_per_gpu:3138.094070799798 mfu:14.802925146021273 +Step 3152 | loss:1.3105003833770752 lr:9.8236014226221e-05 tokens_per_second_per_gpu:3585.4673456430455 mfu:16.913261213208436 +Step 3153 | loss:1.2002755403518677 lr:9.823487241742596e-05 tokens_per_second_per_gpu:5159.724710639651 mfu:24.339301799900635 +Step 3154 | loss:1.4058902263641357 lr:9.823373024585029e-05 tokens_per_second_per_gpu:4634.74149135537 mfu:21.862866383163162 +Step 3155 | loss:1.3605530261993408 lr:9.823258771150257e-05 tokens_per_second_per_gpu:3128.6340983013106 mfu:14.758300841708166 +Step 3156 | loss:1.445682406425476 lr:9.823144481439142e-05 tokens_per_second_per_gpu:5298.236756785576 mfu:24.992686754162953 +Step 3157 | loss:1.2357251644134521 lr:9.82303015545254e-05 tokens_per_second_per_gpu:3539.651358406362 mfu:16.69713938439951 +Step 3158 | loss:1.3075920343399048 lr:9.822915793191312e-05 tokens_per_second_per_gpu:5352.149007792103 mfu:25.247000040539262 +Step 3159 | loss:1.3980315923690796 lr:9.822801394656322e-05 tokens_per_second_per_gpu:5174.715444376803 mfu:24.410015648621695 +Step 3160 | loss:1.1620652675628662 lr:9.822686959848425e-05 tokens_per_second_per_gpu:5331.5400787758845 mfu:25.14978420612389 +Step 3161 | loss:1.3095074892044067 lr:9.822572488768485e-05 tokens_per_second_per_gpu:4632.860674188728 mfu:21.853994247687616 +Step 3162 | loss:1.297208547592163 lr:9.822457981417362e-05 tokens_per_second_per_gpu:3905.7683533330655 mfu:18.424175715469396 +Step 3163 | loss:1.2618985176086426 lr:9.822343437795915e-05 tokens_per_second_per_gpu:5306.437286557532 mfu:25.031370052251923 +Step 3164 | loss:1.287602186203003 lr:9.82222885790501e-05 tokens_per_second_per_gpu:5303.257058735793 mfu:25.01636837501396 +Step 3165 | loss:1.3329758644104004 lr:9.822114241745506e-05 tokens_per_second_per_gpu:5339.684021552851 mfu:25.18820057370276 +Step 3166 | loss:1.3711872100830078 lr:9.821999589318265e-05 tokens_per_second_per_gpu:3528.2415990206778 mfu:16.64331760267155 +Step 3167 | loss:1.3047740459442139 lr:9.82188490062415e-05 tokens_per_second_per_gpu:4390.9339311936765 mfu:20.71278452402126 +Step 3168 | loss:1.3615392446517944 lr:9.821770175664025e-05 tokens_per_second_per_gpu:3953.4418711486883 mfu:18.649059832946573 +Step 3169 | loss:1.2419992685317993 lr:9.821655414438749e-05 tokens_per_second_per_gpu:5347.1328303623295 mfu:25.223337875754904 +Step 3170 | loss:1.3822314739227295 lr:9.821540616949187e-05 tokens_per_second_per_gpu:3274.767723221304 mfu:15.447638083423193 +Step 3171 | loss:1.3509711027145386 lr:9.821425783196206e-05 tokens_per_second_per_gpu:5356.46841108761 mfu:25.267375402850366 +Step 3172 | loss:1.267699956893921 lr:9.821310913180664e-05 tokens_per_second_per_gpu:5363.100143450193 mfu:25.29865841589489 +Step 3173 | loss:1.362392783164978 lr:9.821196006903428e-05 tokens_per_second_per_gpu:4494.461251627343 mfu:21.201140558089136 +Step 3174 | loss:1.1981511116027832 lr:9.82108106436536e-05 tokens_per_second_per_gpu:4630.946256352759 mfu:21.844963612121983 +Step 3175 | loss:1.5523368120193481 lr:9.820966085567329e-05 tokens_per_second_per_gpu:5153.298099612608 mfu:24.308986379193165 +Step 3176 | loss:1.3404210805892944 lr:9.820851070510195e-05 tokens_per_second_per_gpu:4243.581083542184 mfu:20.0176960006609 +Step 3177 | loss:1.2097346782684326 lr:9.820736019194825e-05 tokens_per_second_per_gpu:3615.1484913911745 mfu:17.05327224182802 +Step 3178 | loss:1.4157263040542603 lr:9.820620931622085e-05 tokens_per_second_per_gpu:5314.249822096327 mfu:25.068223115344434 +Step 3179 | loss:1.4635818004608154 lr:9.82050580779284e-05 tokens_per_second_per_gpu:5364.363460096629 mfu:25.304617696060173 +Step 3180 | loss:1.47164785861969 lr:9.820390647707955e-05 tokens_per_second_per_gpu:4774.367543060788 mfu:22.521506291718556 +Step 3181 | loss:1.2503650188446045 lr:9.820275451368297e-05 tokens_per_second_per_gpu:5192.556415034374 mfu:24.494174551196938 +Step 3182 | loss:1.3993297815322876 lr:9.820160218774732e-05 tokens_per_second_per_gpu:5310.715761911836 mfu:25.051552350481344 +Step 3183 | loss:1.5287781953811646 lr:9.820044949928129e-05 tokens_per_second_per_gpu:5012.554075432833 mfu:23.64507280373046 +Step 3184 | loss:1.4609802961349487 lr:9.819929644829351e-05 tokens_per_second_per_gpu:4681.493401273884 mfu:22.08340311894716 +Step 3185 | loss:1.297567367553711 lr:9.819814303479267e-05 tokens_per_second_per_gpu:5372.172997732259 mfu:25.34145661753196 +Step 3186 | loss:1.3934584856033325 lr:9.819698925878744e-05 tokens_per_second_per_gpu:3774.3204154878217 mfu:17.804113365322554 +Step 3187 | loss:1.3975470066070557 lr:9.819583512028651e-05 tokens_per_second_per_gpu:4699.263520214295 mfu:22.167227801884003 +Step 3188 | loss:1.3246886730194092 lr:9.819468061929858e-05 tokens_per_second_per_gpu:5046.9944605428345 mfu:23.807533976429962 +Step 3189 | loss:1.361432671546936 lr:9.819352575583228e-05 tokens_per_second_per_gpu:5196.4390667395055 mfu:24.512489681738735 +Step 3190 | loss:1.27724289894104 lr:9.819237052989633e-05 tokens_per_second_per_gpu:4041.684466188376 mfu:19.06531520948799 +Step 3191 | loss:1.323240041732788 lr:9.819121494149941e-05 tokens_per_second_per_gpu:5057.471433125691 mfu:23.85695564365915 +Step 3192 | loss:1.3142579793930054 lr:9.819005899065021e-05 tokens_per_second_per_gpu:4413.980541882202 mfu:20.821499318796285 +Step 3193 | loss:1.3014847040176392 lr:9.818890267735742e-05 tokens_per_second_per_gpu:5016.833178440076 mfu:23.66525806270588 +Step 3194 | loss:1.2702418565750122 lr:9.818774600162976e-05 tokens_per_second_per_gpu:5304.447026870212 mfu:25.021981657733512 +Step 3195 | loss:1.2711626291275024 lr:9.818658896347591e-05 tokens_per_second_per_gpu:4748.276228536301 mfu:22.398429109469255 +Step 3196 | loss:1.3271238803863525 lr:9.818543156290459e-05 tokens_per_second_per_gpu:5070.145075288028 mfu:23.916739376085218 +Step 3197 | loss:1.1429696083068848 lr:9.818427379992446e-05 tokens_per_second_per_gpu:5306.576343431126 mfu:25.032026007250213 +Step 3198 | loss:1.3695893287658691 lr:9.818311567454428e-05 tokens_per_second_per_gpu:4242.462637891407 mfu:20.012420101699682 +Step 3199 | loss:1.5106542110443115 lr:9.818195718677271e-05 tokens_per_second_per_gpu:3341.1214540773594 mfu:15.760639952985086 +Step 3200 | loss:1.468119740486145 lr:9.818079833661853e-05 tokens_per_second_per_gpu:3156.5797019175934 mfu:14.890125021977813 +Step 3201 | loss:1.3009449243545532 lr:9.81796391240904e-05 tokens_per_second_per_gpu:5329.960873749085 mfu:25.142334826572316 +Step 3202 | loss:1.2065306901931763 lr:9.817847954919707e-05 tokens_per_second_per_gpu:3954.1706825385977 mfu:18.65249776062082 +Step 3203 | loss:1.3941731452941895 lr:9.817731961194722e-05 tokens_per_second_per_gpu:4425.267123090767 mfu:20.874740048046416 +Step 3204 | loss:1.285933256149292 lr:9.817615931234962e-05 tokens_per_second_per_gpu:5293.246594777524 mfu:24.96914730856153 +Step 3205 | loss:1.3155536651611328 lr:9.817499865041299e-05 tokens_per_second_per_gpu:5257.238752190417 mfu:24.799292171506703 +Step 3206 | loss:1.3755019903182983 lr:9.817383762614603e-05 tokens_per_second_per_gpu:4522.244217120536 mfu:21.332197546583245 +Step 3207 | loss:1.3561575412750244 lr:9.81726762395575e-05 tokens_per_second_per_gpu:3716.6109562251804 mfu:17.53188799973161 +Step 3208 | loss:1.3805145025253296 lr:9.817151449065611e-05 tokens_per_second_per_gpu:5240.968388393082 mfu:24.722542089464355 +Step 3209 | loss:1.4876708984375 lr:9.817035237945063e-05 tokens_per_second_per_gpu:4920.056276781916 mfu:23.208744905742265 +Step 3210 | loss:1.3158639669418335 lr:9.816918990594976e-05 tokens_per_second_per_gpu:5329.41338617434 mfu:25.13975223427123 +Step 3211 | loss:1.2978923320770264 lr:9.81680270701623e-05 tokens_per_second_per_gpu:5324.906874484292 mfu:25.118494249739094 +Step 3212 | loss:1.419795274734497 lr:9.816686387209694e-05 tokens_per_second_per_gpu:4320.072966189528 mfu:20.378521261969368 +Step 3213 | loss:1.2312885522842407 lr:9.816570031176244e-05 tokens_per_second_per_gpu:5354.330631111918 mfu:25.25729112996215 +Step 3214 | loss:1.2958283424377441 lr:9.816453638916758e-05 tokens_per_second_per_gpu:3950.5066935468535 mfu:18.63521409940578 +Step 3215 | loss:1.2943174839019775 lr:9.81633721043211e-05 tokens_per_second_per_gpu:5314.3532265772365 mfu:25.068710891923505 +Step 3216 | loss:1.3162429332733154 lr:9.816220745723175e-05 tokens_per_second_per_gpu:4586.328830206152 mfu:21.63449560047056 +Step 3217 | loss:1.336057186126709 lr:9.816104244790828e-05 tokens_per_second_per_gpu:4846.109923804434 mfu:22.859927342198695 +Step 3218 | loss:1.3274785280227661 lr:9.815987707635948e-05 tokens_per_second_per_gpu:5096.033227592606 mfu:24.038858207480157 +Step 3219 | loss:1.3694288730621338 lr:9.815871134259409e-05 tokens_per_second_per_gpu:4226.497097344297 mfu:19.937107922936853 +Step 3220 | loss:1.3538743257522583 lr:9.815754524662089e-05 tokens_per_second_per_gpu:5068.9111351738575 mfu:23.91091867003478 +Step 3221 | loss:1.387708306312561 lr:9.815637878844863e-05 tokens_per_second_per_gpu:4815.075731008759 mfu:22.7135337598022 +Step 3222 | loss:1.3940011262893677 lr:9.815521196808612e-05 tokens_per_second_per_gpu:4842.360212054189 mfu:22.84223931210599 +Step 3223 | loss:1.3272655010223389 lr:9.81540447855421e-05 tokens_per_second_per_gpu:4894.385695295946 mfu:23.08765239302011 +Step 3224 | loss:1.2239657640457153 lr:9.81528772408254e-05 tokens_per_second_per_gpu:5027.219344123854 mfu:23.714251378298854 +Step 3225 | loss:1.3226120471954346 lr:9.815170933394473e-05 tokens_per_second_per_gpu:5012.1259251923775 mfu:23.643053145996262 +Step 3226 | loss:1.4402260780334473 lr:9.815054106490893e-05 tokens_per_second_per_gpu:5308.9757996398175 mfu:25.043344651576056 +Step 3227 | loss:1.3469840288162231 lr:9.814937243372675e-05 tokens_per_second_per_gpu:5154.274432748776 mfu:24.313591909176083 +Step 3228 | loss:1.5543619394302368 lr:9.8148203440407e-05 tokens_per_second_per_gpu:5352.87618196544 mfu:25.250430245183605 +Step 3229 | loss:1.5621342658996582 lr:9.814703408495848e-05 tokens_per_second_per_gpu:2993.114690294735 mfu:14.11903267214588 +Step 3230 | loss:1.6198161840438843 lr:9.814586436738998e-05 tokens_per_second_per_gpu:5345.941509169505 mfu:25.217718210427993 +Step 3231 | loss:1.2753621339797974 lr:9.814469428771026e-05 tokens_per_second_per_gpu:4868.256455636047 mfu:22.96439631143637 +Step 3232 | loss:1.3741703033447266 lr:9.814352384592818e-05 tokens_per_second_per_gpu:3335.891224134442 mfu:15.735968065975099 +Step 3233 | loss:1.259143352508545 lr:9.81423530420525e-05 tokens_per_second_per_gpu:5323.888766300625 mfu:25.113691659729568 +Step 3234 | loss:1.3247331380844116 lr:9.814118187609203e-05 tokens_per_second_per_gpu:4447.908516886074 mfu:20.981543365599 +Step 3235 | loss:1.2336702346801758 lr:9.81400103480556e-05 tokens_per_second_per_gpu:4932.668752789769 mfu:23.26824010698162 +Step 3236 | loss:1.4696757793426514 lr:9.813883845795202e-05 tokens_per_second_per_gpu:3764.472395775858 mfu:17.757658576095793 +Step 3237 | loss:1.3397585153579712 lr:9.813766620579008e-05 tokens_per_second_per_gpu:5291.146675644272 mfu:24.95924163172684 +Step 3238 | loss:1.341928243637085 lr:9.813649359157862e-05 tokens_per_second_per_gpu:5281.961694552272 mfu:24.9159145087965 +Step 3239 | loss:1.196207046508789 lr:9.813532061532644e-05 tokens_per_second_per_gpu:5334.499533532939 mfu:25.163744459147964 +Step 3240 | loss:1.3524805307388306 lr:9.813414727704236e-05 tokens_per_second_per_gpu:3273.5878332705956 mfu:15.442072341215717 +Step 3241 | loss:1.306585431098938 lr:9.813297357673524e-05 tokens_per_second_per_gpu:5002.360877695631 mfu:23.596989750865205 +Step 3242 | loss:1.2617343664169312 lr:9.813179951441387e-05 tokens_per_second_per_gpu:5317.430433032481 mfu:25.083226599798454 +Step 3243 | loss:1.1979475021362305 lr:9.81306250900871e-05 tokens_per_second_per_gpu:5331.14273819388 mfu:25.147909882805283 +Step 3244 | loss:1.351544976234436 lr:9.812945030376375e-05 tokens_per_second_per_gpu:4335.949435606997 mfu:20.45341327701657 +Step 3245 | loss:1.3629961013793945 lr:9.812827515545267e-05 tokens_per_second_per_gpu:4555.902979355047 mfu:21.490971670820446 +Step 3246 | loss:1.3993892669677734 lr:9.812709964516269e-05 tokens_per_second_per_gpu:4310.304829056251 mfu:20.332443292495643 +Step 3247 | loss:1.3471615314483643 lr:9.812592377290267e-05 tokens_per_second_per_gpu:4256.911889677595 mfu:20.080579687671698 +Step 3248 | loss:1.3928028345108032 lr:9.81247475386814e-05 tokens_per_second_per_gpu:4645.733655149456 mfu:21.914718295236405 +Step 3249 | loss:1.3782145977020264 lr:9.812357094250779e-05 tokens_per_second_per_gpu:4399.558673677205 mfu:20.753468905848866 +Step 3250 | loss:1.3384413719177246 lr:9.812239398439065e-05 tokens_per_second_per_gpu:5364.10347044314 mfu:25.30339128050586 +Step 3251 | loss:1.4776610136032104 lr:9.812121666433885e-05 tokens_per_second_per_gpu:5322.965695772066 mfu:25.109337378554287 +Step 3252 | loss:1.2555400133132935 lr:9.812003898236125e-05 tokens_per_second_per_gpu:5313.970596163317 mfu:25.066905958977568 +Step 3253 | loss:1.1612701416015625 lr:9.811886093846667e-05 tokens_per_second_per_gpu:4990.6347537188485 mfu:23.541675623383668 +Step 3254 | loss:1.4113909006118774 lr:9.8117682532664e-05 tokens_per_second_per_gpu:4357.989327387982 mfu:20.557379206940773 +Step 3255 | loss:1.1821905374526978 lr:9.811650376496213e-05 tokens_per_second_per_gpu:5342.504551736893 mfu:25.20150549581296 +Step 3256 | loss:1.2078384160995483 lr:9.811532463536986e-05 tokens_per_second_per_gpu:5242.616716895492 mfu:24.730317536244065 +Step 3257 | loss:1.1870684623718262 lr:9.811414514389612e-05 tokens_per_second_per_gpu:4829.9750182922435 mfu:22.783816239998938 +Step 3258 | loss:1.4302124977111816 lr:9.811296529054973e-05 tokens_per_second_per_gpu:4633.455654067338 mfu:21.856800869289057 +Step 3259 | loss:1.4779177904129028 lr:9.811178507533961e-05 tokens_per_second_per_gpu:4489.7509452263685 mfu:21.178921239134596 +Step 3260 | loss:1.419722557067871 lr:9.81106044982746e-05 tokens_per_second_per_gpu:5321.117633676246 mfu:25.100619754336396 +Step 3261 | loss:1.3212790489196777 lr:9.810942355936361e-05 tokens_per_second_per_gpu:4001.341492475365 mfu:18.875010519262627 +Step 3262 | loss:1.3061703443527222 lr:9.81082422586155e-05 tokens_per_second_per_gpu:5300.387161834943 mfu:25.002830581677753 +Step 3263 | loss:1.504902720451355 lr:9.810706059603917e-05 tokens_per_second_per_gpu:5365.271786455574 mfu:25.30890242647178 +Step 3264 | loss:1.3732221126556396 lr:9.810587857164348e-05 tokens_per_second_per_gpu:5361.154403004254 mfu:25.289480026233843 +Step 3265 | loss:1.3055362701416016 lr:9.810469618543736e-05 tokens_per_second_per_gpu:5345.595309446516 mfu:25.216085127266894 +Step 3266 | loss:1.3086771965026855 lr:9.810351343742968e-05 tokens_per_second_per_gpu:3916.4062379447355 mfu:18.474356432192916 +Step 3267 | loss:1.4453377723693848 lr:9.810233032762932e-05 tokens_per_second_per_gpu:5350.723162453885 mfu:25.240274084804593 +Step 3268 | loss:1.4222692251205444 lr:9.81011468560452e-05 tokens_per_second_per_gpu:5333.429201987558 mfu:25.158695522631014 +Step 3269 | loss:1.4665195941925049 lr:9.809996302268622e-05 tokens_per_second_per_gpu:4479.131412550963 mfu:21.128827091625602 +Step 3270 | loss:1.1846504211425781 lr:9.809877882756129e-05 tokens_per_second_per_gpu:4931.4826685260905 mfu:23.262645145143 +Step 3271 | loss:1.1869491338729858 lr:9.809759427067929e-05 tokens_per_second_per_gpu:4165.5843187453565 mfu:19.649771953493953 +Step 3272 | loss:1.3849279880523682 lr:9.809640935204916e-05 tokens_per_second_per_gpu:5046.575780361605 mfu:23.8055589905809 +Step 3273 | loss:1.2818381786346436 lr:9.809522407167979e-05 tokens_per_second_per_gpu:5344.825270463786 mfu:25.21245272200316 +Step 3274 | loss:1.2377384901046753 lr:9.80940384295801e-05 tokens_per_second_per_gpu:4866.321143505681 mfu:22.95526711391914 +Step 3275 | loss:1.3820748329162598 lr:9.809285242575903e-05 tokens_per_second_per_gpu:5347.221318962593 mfu:25.223755291580055 +Step 3276 | loss:1.3131110668182373 lr:9.809166606022547e-05 tokens_per_second_per_gpu:5326.442046817303 mfu:25.12573591956132 +Step 3277 | loss:1.3580090999603271 lr:9.809047933298834e-05 tokens_per_second_per_gpu:5211.707529963803 mfu:24.584513627836195 +Step 3278 | loss:1.247665286064148 lr:9.808929224405658e-05 tokens_per_second_per_gpu:3255.6647340133213 mfu:15.35752602402309 +Step 3279 | loss:1.2724401950836182 lr:9.808810479343913e-05 tokens_per_second_per_gpu:3954.4995743186073 mfu:18.654049199261852 +Step 3280 | loss:1.3477853536605835 lr:9.808691698114489e-05 tokens_per_second_per_gpu:3958.3629688062206 mfu:18.67227349022299 +Step 3281 | loss:1.2775038480758667 lr:9.808572880718284e-05 tokens_per_second_per_gpu:4838.031745751968 mfu:22.82182119804591 +Step 3282 | loss:1.2658933401107788 lr:9.808454027156187e-05 tokens_per_second_per_gpu:5331.844926742724 mfu:25.151222225995898 +Step 3283 | loss:1.32283616065979 lr:9.808335137429093e-05 tokens_per_second_per_gpu:4152.700447202155 mfu:19.588996533209514 +Step 3284 | loss:1.3669228553771973 lr:9.808216211537898e-05 tokens_per_second_per_gpu:4853.004749830849 mfu:22.89245141294391 +Step 3285 | loss:1.3129256963729858 lr:9.808097249483494e-05 tokens_per_second_per_gpu:5329.425729082662 mfu:25.1398104578757 +Step 3286 | loss:1.3267987966537476 lr:9.807978251266779e-05 tokens_per_second_per_gpu:5302.557558992984 mfu:25.01306871538017 +Step 3287 | loss:1.2337288856506348 lr:9.807859216888647e-05 tokens_per_second_per_gpu:5335.7632229882665 mfu:25.1697054979164 +Step 3288 | loss:1.4948076009750366 lr:9.807740146349988e-05 tokens_per_second_per_gpu:5313.614670831126 mfu:25.0652269984585 +Step 3289 | loss:1.239277720451355 lr:9.807621039651705e-05 tokens_per_second_per_gpu:4562.231641097684 mfu:21.5208250480415 +Step 3290 | loss:1.267935872077942 lr:9.807501896794691e-05 tokens_per_second_per_gpu:5074.255474097559 mfu:23.93612882857613 +Step 3291 | loss:1.2078937292099 lr:9.80738271777984e-05 tokens_per_second_per_gpu:4819.518480262019 mfu:22.73449096604101 +Step 3292 | loss:1.2492622137069702 lr:9.807263502608051e-05 tokens_per_second_per_gpu:5372.973657127817 mfu:25.345233464507384 +Step 3293 | loss:1.3178081512451172 lr:9.80714425128022e-05 tokens_per_second_per_gpu:3951.9540863301427 mfu:18.64204169811523 +Step 3294 | loss:1.266719102859497 lr:9.807024963797245e-05 tokens_per_second_per_gpu:4520.943579423794 mfu:21.32606221665604 +Step 3295 | loss:1.5267504453659058 lr:9.80690564016002e-05 tokens_per_second_per_gpu:3559.9237929552255 mfu:16.7927679170005 +Step 3296 | loss:1.3450971841812134 lr:9.806786280369444e-05 tokens_per_second_per_gpu:3674.5573264361583 mfu:17.333513852927677 +Step 3297 | loss:1.4399863481521606 lr:9.806666884426416e-05 tokens_per_second_per_gpu:5329.877412293874 mfu:25.141941124648092 +Step 3298 | loss:1.3568413257598877 lr:9.806547452331833e-05 tokens_per_second_per_gpu:4671.36645938573 mfu:22.03563260622628 +Step 3299 | loss:1.3691877126693726 lr:9.806427984086592e-05 tokens_per_second_per_gpu:5367.937665711315 mfu:25.32147783376701 +Step 3300 | loss:1.14998197555542 lr:9.806308479691595e-05 tokens_per_second_per_gpu:5325.909525605272 mfu:25.123223926146682 +Step 3301 | loss:1.4802517890930176 lr:9.806188939147738e-05 tokens_per_second_per_gpu:4792.25563279316 mfu:22.605887463008976 +Step 3302 | loss:1.41939115524292 lr:9.806069362455919e-05 tokens_per_second_per_gpu:3064.5526898282264 mfu:14.456017904524924 +Step 3303 | loss:1.2790783643722534 lr:9.80594974961704e-05 tokens_per_second_per_gpu:3574.809564820792 mfu:16.86298664266495 +Step 3304 | loss:1.3560991287231445 lr:9.805830100632001e-05 tokens_per_second_per_gpu:4440.889739964288 mfu:20.94843459733211 +Step 3305 | loss:1.2667148113250732 lr:9.8057104155017e-05 tokens_per_second_per_gpu:4464.626744045606 mfu:21.060406095534198 +Step 3306 | loss:1.2357027530670166 lr:9.805590694227037e-05 tokens_per_second_per_gpu:4566.637302850745 mfu:21.541607306214097 +Step 3307 | loss:1.1991404294967651 lr:9.805470936808914e-05 tokens_per_second_per_gpu:3852.5046396985167 mfu:18.172921690528586 +Step 3308 | loss:1.2111570835113525 lr:9.805351143248229e-05 tokens_per_second_per_gpu:4794.8225119827475 mfu:22.617995870101012 +Step 3309 | loss:1.3193210363388062 lr:9.805231313545888e-05 tokens_per_second_per_gpu:5338.801388584735 mfu:25.184037043399442 +Step 3310 | loss:1.2151188850402832 lr:9.805111447702786e-05 tokens_per_second_per_gpu:5335.058921296975 mfu:25.16638318667232 +Step 3311 | loss:1.4541689157485962 lr:9.804991545719831e-05 tokens_per_second_per_gpu:4443.912365381832 mfu:20.962692837139738 +Step 3312 | loss:1.4855926036834717 lr:9.804871607597919e-05 tokens_per_second_per_gpu:4684.205990218654 mfu:22.096198863815033 +Step 3313 | loss:1.5323859453201294 lr:9.804751633337954e-05 tokens_per_second_per_gpu:4093.478460337841 mfu:19.309636316857663 +Step 3314 | loss:1.338387370109558 lr:9.804631622940841e-05 tokens_per_second_per_gpu:5112.541541058698 mfu:24.116730738708434 +Step 3315 | loss:1.4637181758880615 lr:9.804511576407479e-05 tokens_per_second_per_gpu:4630.864865756657 mfu:21.844579678793504 +Step 3316 | loss:1.4335969686508179 lr:9.804391493738774e-05 tokens_per_second_per_gpu:4302.397360233959 mfu:20.295142412907595 +Step 3317 | loss:1.413590908050537 lr:9.804271374935626e-05 tokens_per_second_per_gpu:5068.742713522718 mfu:23.910124196370656 +Step 3318 | loss:1.2917155027389526 lr:9.80415121999894e-05 tokens_per_second_per_gpu:4371.53535009422 mfu:20.621278107235895 +Step 3319 | loss:1.2735871076583862 lr:9.80403102892962e-05 tokens_per_second_per_gpu:5315.253058103884 mfu:25.072955550761787 +Step 3320 | loss:1.2522960901260376 lr:9.80391080172857e-05 tokens_per_second_per_gpu:4544.643454207234 mfu:21.437858569625124 +Step 3321 | loss:1.2766846418380737 lr:9.803790538396693e-05 tokens_per_second_per_gpu:3805.058382092196 mfu:17.949109598233278 +Step 3322 | loss:1.3613739013671875 lr:9.803670238934895e-05 tokens_per_second_per_gpu:4258.141184134783 mfu:20.086378479365514 +Step 3323 | loss:1.1915690898895264 lr:9.80354990334408e-05 tokens_per_second_per_gpu:4619.295734823115 mfu:21.790006114282704 +Step 3324 | loss:1.2755687236785889 lr:9.803429531625153e-05 tokens_per_second_per_gpu:4120.413937118193 mfu:19.436695556496808 +Step 3325 | loss:1.3763049840927124 lr:9.80330912377902e-05 tokens_per_second_per_gpu:3882.5166003115974 mfu:18.314493229308987 +Step 3326 | loss:1.3798956871032715 lr:9.803188679806586e-05 tokens_per_second_per_gpu:4145.218267740907 mfu:19.553701816099352 +Step 3327 | loss:1.3485887050628662 lr:9.803068199708758e-05 tokens_per_second_per_gpu:4765.406452926149 mfu:22.479235300633828 +Step 3328 | loss:1.2586548328399658 lr:9.80294768348644e-05 tokens_per_second_per_gpu:3269.777295180201 mfu:15.424097382898486 +Step 3329 | loss:1.3980196714401245 lr:9.80282713114054e-05 tokens_per_second_per_gpu:5165.905489056704 mfu:24.368457586243412 +Step 3330 | loss:1.3959498405456543 lr:9.802706542671964e-05 tokens_per_second_per_gpu:4196.122122807955 mfu:19.79382398554387 +Step 3331 | loss:1.5348044633865356 lr:9.80258591808162e-05 tokens_per_second_per_gpu:3491.4754442107596 mfu:16.46988537748031 +Step 3332 | loss:1.4374781847000122 lr:9.802465257370415e-05 tokens_per_second_per_gpu:4209.540045844052 mfu:19.857118617838875 +Step 3333 | loss:1.1991046667099 lr:9.802344560539254e-05 tokens_per_second_per_gpu:4745.540023801699 mfu:22.38552196489147 +Step 3334 | loss:1.2624146938323975 lr:9.802223827589048e-05 tokens_per_second_per_gpu:3896.3859509586982 mfu:18.379917322666593 +Step 3335 | loss:1.3697155714035034 lr:9.802103058520704e-05 tokens_per_second_per_gpu:4713.837280504768 mfu:22.23597471571408 +Step 3336 | loss:1.2490462064743042 lr:9.80198225333513e-05 tokens_per_second_per_gpu:3925.73513118123 mfu:18.518362413262775 +Step 3337 | loss:1.2717818021774292 lr:9.801861412033233e-05 tokens_per_second_per_gpu:5353.493352510371 mfu:25.253341543944195 +Step 3338 | loss:1.1277098655700684 lr:9.801740534615926e-05 tokens_per_second_per_gpu:4816.110930449374 mfu:22.71841697218682 +Step 3339 | loss:1.3875219821929932 lr:9.801619621084114e-05 tokens_per_second_per_gpu:5367.42367051594 mfu:25.319053230733804 +Step 3340 | loss:1.1797564029693604 lr:9.801498671438709e-05 tokens_per_second_per_gpu:5006.325012618544 mfu:23.615689251647748 +Step 3341 | loss:1.3522025346755981 lr:9.80137768568062e-05 tokens_per_second_per_gpu:5373.7074550974185 mfu:25.348694914728842 +Step 3342 | loss:1.4337674379348755 lr:9.801256663810757e-05 tokens_per_second_per_gpu:5333.008255808416 mfu:25.156709847683175 +Step 3343 | loss:1.3320567607879639 lr:9.80113560583003e-05 tokens_per_second_per_gpu:5129.000335470786 mfu:24.194369680110142 +Step 3344 | loss:1.3771156072616577 lr:9.801014511739348e-05 tokens_per_second_per_gpu:4006.1925230780807 mfu:18.897893658286808 +Step 3345 | loss:1.439980149269104 lr:9.800893381539624e-05 tokens_per_second_per_gpu:4512.766343633428 mfu:21.287488800252586 +Step 3346 | loss:1.2333344221115112 lr:9.80077221523177e-05 tokens_per_second_per_gpu:5345.584100018254 mfu:25.216032250481256 +Step 3347 | loss:1.3551011085510254 lr:9.800651012816693e-05 tokens_per_second_per_gpu:3351.8349180286605 mfu:15.811177190348811 +Step 3348 | loss:1.26508367061615 lr:9.800529774295309e-05 tokens_per_second_per_gpu:4320.692989142233 mfu:20.381446016024096 +Step 3349 | loss:1.3166693449020386 lr:9.800408499668527e-05 tokens_per_second_per_gpu:3332.6637094931757 mfu:15.720743328741465 +Step 3350 | loss:1.3546687364578247 lr:9.80028718893726e-05 tokens_per_second_per_gpu:4725.978619941888 mfu:22.29324748536509 +Step 3351 | loss:1.317793846130371 lr:9.800165842102422e-05 tokens_per_second_per_gpu:3783.6022367287383 mfu:17.84789729975796 +Step 3352 | loss:1.2643635272979736 lr:9.800044459164923e-05 tokens_per_second_per_gpu:5297.261460570892 mfu:24.988086115515596 +Step 3353 | loss:1.3695684671401978 lr:9.799923040125677e-05 tokens_per_second_per_gpu:5331.071147605231 mfu:25.147572177784014 +Step 3354 | loss:1.4061241149902344 lr:9.799801584985598e-05 tokens_per_second_per_gpu:5314.220781848537 mfu:25.06808612754083 +Step 3355 | loss:1.409208059310913 lr:9.799680093745598e-05 tokens_per_second_per_gpu:3691.94865988963 mfu:17.415551740095168 +Step 3356 | loss:1.1596213579177856 lr:9.799558566406592e-05 tokens_per_second_per_gpu:3586.9701254804054 mfu:16.92035008210194 +Step 3357 | loss:1.6048773527145386 lr:9.799437002969494e-05 tokens_per_second_per_gpu:4280.512311486023 mfu:20.191906903989224 +Step 3358 | loss:1.157231330871582 lr:9.799315403435218e-05 tokens_per_second_per_gpu:4281.0936779227695 mfu:20.1946493086627 +Step 3359 | loss:1.356927752494812 lr:9.799193767804677e-05 tokens_per_second_per_gpu:4249.74293910895 mfu:20.046762524690305 +Step 3360 | loss:1.5519016981124878 lr:9.799072096078788e-05 tokens_per_second_per_gpu:3131.773947651865 mfu:14.77311204680827 +Step 3361 | loss:1.376152515411377 lr:9.798950388258466e-05 tokens_per_second_per_gpu:4059.0036735867075 mfu:19.14701286574741 +Step 3362 | loss:1.319272756576538 lr:9.798828644344626e-05 tokens_per_second_per_gpu:5335.9125979261735 mfu:25.170410124984496 +Step 3363 | loss:1.447716236114502 lr:9.798706864338183e-05 tokens_per_second_per_gpu:4053.364747092681 mfu:19.120413087375017 +Step 3364 | loss:1.2609440088272095 lr:9.798585048240052e-05 tokens_per_second_per_gpu:4882.693225168932 mfu:23.032497016489874 +Step 3365 | loss:1.3718401193618774 lr:9.798463196051153e-05 tokens_per_second_per_gpu:5312.88846919399 mfu:25.06180138143335 +Step 3366 | loss:1.1951587200164795 lr:9.798341307772399e-05 tokens_per_second_per_gpu:5307.4830128213425 mfu:25.036302921457615 +Step 3367 | loss:1.3241685628890991 lr:9.798219383404708e-05 tokens_per_second_per_gpu:4366.083117896703 mfu:20.595558997713468 +Step 3368 | loss:1.319393277168274 lr:9.798097422948995e-05 tokens_per_second_per_gpu:4563.889660444584 mfu:21.528646208188096 +Step 3369 | loss:1.2650120258331299 lr:9.79797542640618e-05 tokens_per_second_per_gpu:5356.359103002169 mfu:25.26685977796146 +Step 3370 | loss:1.4296609163284302 lr:9.79785339377718e-05 tokens_per_second_per_gpu:5140.145096044142 mfu:24.246941417226893 +Step 3371 | loss:1.3443827629089355 lr:9.797731325062913e-05 tokens_per_second_per_gpu:4225.71512700968 mfu:19.933419235450568 +Step 3372 | loss:1.2506213188171387 lr:9.797609220264297e-05 tokens_per_second_per_gpu:4994.210380451867 mfu:23.55854246474815 +Step 3373 | loss:1.4366750717163086 lr:9.797487079382249e-05 tokens_per_second_per_gpu:4233.334268576515 mfu:19.969360026180926 +Step 3374 | loss:1.3438953161239624 lr:9.797364902417687e-05 tokens_per_second_per_gpu:4019.7117197107145 mfu:18.961666015415737 +Step 3375 | loss:1.4144701957702637 lr:9.797242689371533e-05 tokens_per_second_per_gpu:5188.678202619728 mfu:24.475880361545784 +Step 3376 | loss:1.233168125152588 lr:9.797120440244704e-05 tokens_per_second_per_gpu:5018.464031521501 mfu:23.672951074943192 +Step 3377 | loss:1.2855308055877686 lr:9.79699815503812e-05 tokens_per_second_per_gpu:4471.456587010392 mfu:21.092623630090465 +Step 3378 | loss:1.4134591817855835 lr:9.7968758337527e-05 tokens_per_second_per_gpu:5092.436242496048 mfu:24.021890615070756 +Step 3379 | loss:1.407496690750122 lr:9.796753476389367e-05 tokens_per_second_per_gpu:5320.5310199196165 mfu:25.097852597159743 +Step 3380 | loss:1.314743161201477 lr:9.796631082949038e-05 tokens_per_second_per_gpu:4871.556968500982 mfu:22.979965393745836 +Step 3381 | loss:1.3671882152557373 lr:9.796508653432635e-05 tokens_per_second_per_gpu:4402.438457899806 mfu:20.767053339368733 +Step 3382 | loss:1.1713210344314575 lr:9.796386187841077e-05 tokens_per_second_per_gpu:4890.52887003329 mfu:23.06945909021392 +Step 3383 | loss:1.3519502878189087 lr:9.796263686175288e-05 tokens_per_second_per_gpu:3891.804604213863 mfu:18.358306328412848 +Step 3384 | loss:1.2094311714172363 lr:9.796141148436186e-05 tokens_per_second_per_gpu:5288.763941690788 mfu:24.948001869131918 +Step 3385 | loss:1.341475486755371 lr:9.796018574624697e-05 tokens_per_second_per_gpu:5208.825615739266 mfu:24.570919146733786 +Step 3386 | loss:1.3127483129501343 lr:9.795895964741738e-05 tokens_per_second_per_gpu:5046.717980813661 mfu:23.806229774375144 +Step 3387 | loss:1.1709896326065063 lr:9.795773318788235e-05 tokens_per_second_per_gpu:5018.038041646669 mfu:23.670941608022236 +Step 3388 | loss:1.2708913087844849 lr:9.795650636765108e-05 tokens_per_second_per_gpu:4526.486991160056 mfu:21.352211435619555 +Step 3389 | loss:1.466036081314087 lr:9.795527918673282e-05 tokens_per_second_per_gpu:4334.361181979558 mfu:20.445921213671927 +Step 3390 | loss:1.2738254070281982 lr:9.795405164513675e-05 tokens_per_second_per_gpu:5313.066818167926 mfu:25.06264268397288 +Step 3391 | loss:1.262272596359253 lr:9.795282374287219e-05 tokens_per_second_per_gpu:4436.3092600023765 mfu:20.926827691841304 +Step 3392 | loss:1.2874037027359009 lr:9.79515954799483e-05 tokens_per_second_per_gpu:5363.267642402451 mfu:25.299448537030962 +Step 3393 | loss:1.3984684944152832 lr:9.795036685637434e-05 tokens_per_second_per_gpu:4592.145032293011 mfu:21.66193161806068 +Step 3394 | loss:1.315826654434204 lr:9.794913787215954e-05 tokens_per_second_per_gpu:5381.615611501656 mfu:25.38599903775824 +Step 3395 | loss:1.2355138063430786 lr:9.794790852731317e-05 tokens_per_second_per_gpu:4856.588538969264 mfu:22.909356757767295 +Step 3396 | loss:1.2561743259429932 lr:9.794667882184445e-05 tokens_per_second_per_gpu:5311.760869016024 mfu:25.056482299005456 +Step 3397 | loss:1.3190687894821167 lr:9.794544875576266e-05 tokens_per_second_per_gpu:3795.61160451894 mfu:17.904547536633775 +Step 3398 | loss:1.4184828996658325 lr:9.794421832907702e-05 tokens_per_second_per_gpu:5085.002935086221 mfu:23.98682643576558 +Step 3399 | loss:1.188522219657898 lr:9.79429875417968e-05 tokens_per_second_per_gpu:4779.542230388888 mfu:22.545916174737904 +Step 3400 | loss:1.1637026071548462 lr:9.794175639393124e-05 tokens_per_second_per_gpu:4720.557132974535 mfu:22.267673406339025 +Step 3401 | loss:1.319901943206787 lr:9.794052488548963e-05 tokens_per_second_per_gpu:5326.942857776076 mfu:25.12809833030101 +Step 3402 | loss:1.336838722229004 lr:9.793929301648122e-05 tokens_per_second_per_gpu:5344.926454776918 mfu:25.212930025672023 +Step 3403 | loss:1.1631832122802734 lr:9.793806078691525e-05 tokens_per_second_per_gpu:4120.163408275724 mfu:19.435513769202252 +Step 3404 | loss:1.376941442489624 lr:9.793682819680101e-05 tokens_per_second_per_gpu:5362.6822134276745 mfu:25.296686968672457 +Step 3405 | loss:1.4804109334945679 lr:9.793559524614779e-05 tokens_per_second_per_gpu:5325.125209546622 mfu:25.119524173479906 +Step 3406 | loss:1.3183797597885132 lr:9.793436193496482e-05 tokens_per_second_per_gpu:5314.372982601468 mfu:25.06880408446053 +Step 3407 | loss:1.2866824865341187 lr:9.79331282632614e-05 tokens_per_second_per_gpu:4097.79839731059 mfu:19.33001419661537 +Step 3408 | loss:1.4732505083084106 lr:9.793189423104682e-05 tokens_per_second_per_gpu:5242.41234867526 mfu:24.729353496480417 +Step 3409 | loss:1.3139749765396118 lr:9.793065983833034e-05 tokens_per_second_per_gpu:4072.1555444760083 mfu:19.209052484672807 +Step 3410 | loss:1.267314076423645 lr:9.792942508512124e-05 tokens_per_second_per_gpu:4855.505348638893 mfu:22.90424716416761 +Step 3411 | loss:1.4132308959960938 lr:9.792818997142883e-05 tokens_per_second_per_gpu:3840.582364239059 mfu:18.11668228303638 +Step 3412 | loss:1.308490514755249 lr:9.79269544972624e-05 tokens_per_second_per_gpu:4836.718455449333 mfu:22.81562618361795 +Step 3413 | loss:1.219051718711853 lr:9.79257186626312e-05 tokens_per_second_per_gpu:5313.871862965572 mfu:25.066440217638185 +Step 3414 | loss:1.441996455192566 lr:9.792448246754457e-05 tokens_per_second_per_gpu:3440.0133159673073 mfu:16.227129738211467 +Step 3415 | loss:1.2991712093353271 lr:9.792324591201179e-05 tokens_per_second_per_gpu:4700.70373242961 mfu:22.17402152862919 +Step 3416 | loss:1.3401092290878296 lr:9.792200899604215e-05 tokens_per_second_per_gpu:3891.5266317767287 mfu:18.356995084999895 +Step 3417 | loss:1.26374089717865 lr:9.7920771719645e-05 tokens_per_second_per_gpu:4588.390232921586 mfu:21.644219588790836 +Step 3418 | loss:1.2817245721817017 lr:9.791953408282959e-05 tokens_per_second_per_gpu:5337.772680979696 mfu:25.17918445411086 +Step 3419 | loss:1.3385560512542725 lr:9.791829608560525e-05 tokens_per_second_per_gpu:4776.351337299918 mfu:22.53086418761473 +Step 3420 | loss:1.413676142692566 lr:9.791705772798128e-05 tokens_per_second_per_gpu:4745.481982585797 mfu:22.385248174573206 +Step 3421 | loss:1.0829262733459473 lr:9.791581900996702e-05 tokens_per_second_per_gpu:4222.745926611936 mfu:19.919413010576584 +Step 3422 | loss:1.2758949995040894 lr:9.791457993157176e-05 tokens_per_second_per_gpu:4434.281691319031 mfu:20.91726330442307 +Step 3423 | loss:1.3820223808288574 lr:9.791334049280485e-05 tokens_per_second_per_gpu:3087.941643602719 mfu:14.566347590046215 +Step 3424 | loss:1.2826131582260132 lr:9.791210069367557e-05 tokens_per_second_per_gpu:4102.017132098116 mfu:19.349914688398506 +Step 3425 | loss:1.3366938829421997 lr:9.791086053419328e-05 tokens_per_second_per_gpu:4705.2143851443425 mfu:22.195299047081296 +Step 3426 | loss:1.3401834964752197 lr:9.790962001436728e-05 tokens_per_second_per_gpu:4788.204740403222 mfu:22.58677871245213 +Step 3427 | loss:1.4356319904327393 lr:9.790837913420694e-05 tokens_per_second_per_gpu:4725.573673458065 mfu:22.2913372837099 +Step 3428 | loss:1.2901288270950317 lr:9.790713789372154e-05 tokens_per_second_per_gpu:5357.992144821917 mfu:25.274563114850874 +Step 3429 | loss:1.4064265489578247 lr:9.790589629292046e-05 tokens_per_second_per_gpu:4987.577761227106 mfu:23.527255268223417 +Step 3430 | loss:1.3856890201568604 lr:9.790465433181302e-05 tokens_per_second_per_gpu:4874.372400499666 mfu:22.993246266845862 +Step 3431 | loss:1.3782066106796265 lr:9.790341201040855e-05 tokens_per_second_per_gpu:4553.547193269721 mfu:21.479859025917406 +Step 3432 | loss:1.3307642936706543 lr:9.790216932871643e-05 tokens_per_second_per_gpu:5314.5748296451475 mfu:25.069756231404224 +Step 3433 | loss:1.349480390548706 lr:9.790092628674596e-05 tokens_per_second_per_gpu:5244.508272100379 mfu:24.739240324878114 +Step 3434 | loss:1.2836309671401978 lr:9.789968288450652e-05 tokens_per_second_per_gpu:5314.442410795986 mfu:25.06913158909974 +Step 3435 | loss:1.2511544227600098 lr:9.789843912200745e-05 tokens_per_second_per_gpu:5303.229929024711 mfu:25.01624039953798 +Step 3436 | loss:1.313387393951416 lr:9.789719499925812e-05 tokens_per_second_per_gpu:3452.038285045752 mfu:16.283853569025776 +Step 3437 | loss:1.2358983755111694 lr:9.789595051626785e-05 tokens_per_second_per_gpu:4852.453433366302 mfu:22.88985075911636 +Step 3438 | loss:1.3481630086898804 lr:9.789470567304603e-05 tokens_per_second_per_gpu:4120.630408514982 mfu:19.437716689009296 +Step 3439 | loss:1.3001028299331665 lr:9.789346046960204e-05 tokens_per_second_per_gpu:5239.399087074091 mfu:24.71513942739226 +Step 3440 | loss:1.4441232681274414 lr:9.78922149059452e-05 tokens_per_second_per_gpu:3534.714586790513 mfu:16.67385179038746 +Step 3441 | loss:1.5405523777008057 lr:9.78909689820849e-05 tokens_per_second_per_gpu:5332.975025109333 mfu:25.156553092805858 +Step 3442 | loss:1.2551360130310059 lr:9.788972269803053e-05 tokens_per_second_per_gpu:4235.962036349283 mfu:19.98175565510849 +Step 3443 | loss:1.456527829170227 lr:9.788847605379143e-05 tokens_per_second_per_gpu:4790.266326204636 mfu:22.59650356442021 +Step 3444 | loss:1.1500205993652344 lr:9.788722904937699e-05 tokens_per_second_per_gpu:3505.774308448082 mfu:16.537335559725566 +Step 3445 | loss:1.4548912048339844 lr:9.78859816847966e-05 tokens_per_second_per_gpu:5324.6703119870435 mfu:25.11737834407785 +Step 3446 | loss:1.3149588108062744 lr:9.788473396005963e-05 tokens_per_second_per_gpu:5342.194578078276 mfu:25.2000432971783 +Step 3447 | loss:1.1504912376403809 lr:9.788348587517546e-05 tokens_per_second_per_gpu:5339.572039397979 mfu:25.187672334772106 +Step 3448 | loss:1.350224494934082 lr:9.788223743015348e-05 tokens_per_second_per_gpu:4158.697735978982 mfu:19.617286767612786 +Step 3449 | loss:1.4217288494110107 lr:9.788098862500309e-05 tokens_per_second_per_gpu:5335.038618021941 mfu:25.166287412659887 +Step 3450 | loss:1.265555739402771 lr:9.787973945973368e-05 tokens_per_second_per_gpu:5361.3153265645105 mfu:25.290239130123965 +Step 3451 | loss:1.430267572402954 lr:9.787848993435463e-05 tokens_per_second_per_gpu:4065.724981844098 mfu:19.17871842357096 +Step 3452 | loss:1.1915011405944824 lr:9.787724004887535e-05 tokens_per_second_per_gpu:5110.991615980493 mfu:24.109419477670052 +Step 3453 | loss:1.3158222436904907 lr:9.787598980330523e-05 tokens_per_second_per_gpu:4498.166975138093 mfu:21.21862108815097 +Step 3454 | loss:1.1853867769241333 lr:9.78747391976537e-05 tokens_per_second_per_gpu:3495.654785232983 mfu:16.489600042151245 +Step 3455 | loss:1.2237753868103027 lr:9.787348823193013e-05 tokens_per_second_per_gpu:5006.27538063065 mfu:23.615455129092762 +Step 3456 | loss:1.4778348207473755 lr:9.787223690614395e-05 tokens_per_second_per_gpu:5221.394766981203 mfu:24.630209977660343 +Step 3457 | loss:1.3503644466400146 lr:9.787098522030457e-05 tokens_per_second_per_gpu:4806.302133208394 mfu:22.672147202047732 +Step 3458 | loss:1.324471354484558 lr:9.786973317442141e-05 tokens_per_second_per_gpu:4966.105444211278 mfu:23.4259666853047 +Step 3459 | loss:1.3844660520553589 lr:9.786848076850388e-05 tokens_per_second_per_gpu:5331.933294684702 mfu:25.151639072655332 +Step 3460 | loss:1.2272316217422485 lr:9.786722800256139e-05 tokens_per_second_per_gpu:5144.604789907391 mfu:24.267978554082074 +Step 3461 | loss:1.3841692209243774 lr:9.786597487660337e-05 tokens_per_second_per_gpu:5318.416070590674 mfu:25.08787602032756 +Step 3462 | loss:1.3106215000152588 lr:9.786472139063925e-05 tokens_per_second_per_gpu:4109.597378467716 mfu:19.385672003847304 +Step 3463 | loss:1.3548768758773804 lr:9.786346754467845e-05 tokens_per_second_per_gpu:5371.2034217638175 mfu:25.336882962262102 +Step 3464 | loss:1.2010315656661987 lr:9.78622133387304e-05 tokens_per_second_per_gpu:5325.05916733382 mfu:25.11921264109416 +Step 3465 | loss:1.3138450384140015 lr:9.786095877280453e-05 tokens_per_second_per_gpu:4840.335603017906 mfu:22.83268888584792 +Step 3466 | loss:1.5285497903823853 lr:9.78597038469103e-05 tokens_per_second_per_gpu:4104.2188668981025 mfu:19.360300647104737 +Step 3467 | loss:1.2648553848266602 lr:9.785844856105711e-05 tokens_per_second_per_gpu:5015.233641560712 mfu:23.65771277433679 +Step 3468 | loss:1.410003423690796 lr:9.785719291525443e-05 tokens_per_second_per_gpu:5319.084523871292 mfu:25.091029228502123 +Step 3469 | loss:1.3936208486557007 lr:9.78559369095117e-05 tokens_per_second_per_gpu:4732.066796788773 mfu:22.32196645430193 +Step 3470 | loss:1.3167628049850464 lr:9.785468054383837e-05 tokens_per_second_per_gpu:4123.6058588294945 mfu:19.45175239580702 +Step 3471 | loss:1.4319230318069458 lr:9.785342381824387e-05 tokens_per_second_per_gpu:3849.8481504949295 mfu:18.16039057771147 +Step 3472 | loss:1.38905668258667 lr:9.785216673273765e-05 tokens_per_second_per_gpu:3500.695556753575 mfu:16.513378221458183 +Step 3473 | loss:1.3226311206817627 lr:9.785090928732921e-05 tokens_per_second_per_gpu:4914.564186513768 mfu:23.18283777890019 +Step 3474 | loss:1.2273951768875122 lr:9.784965148202794e-05 tokens_per_second_per_gpu:5333.4585479495 mfu:25.15883395254043 +Step 3475 | loss:1.297452449798584 lr:9.784839331684337e-05 tokens_per_second_per_gpu:4563.975841418941 mfu:21.52905273854813 +Step 3476 | loss:1.2742170095443726 lr:9.784713479178491e-05 tokens_per_second_per_gpu:5316.348014481635 mfu:25.07812064681446 +Step 3477 | loss:1.4088177680969238 lr:9.784587590686204e-05 tokens_per_second_per_gpu:4002.2621164449115 mfu:18.879353259601626 +Step 3478 | loss:1.2945140600204468 lr:9.784461666208424e-05 tokens_per_second_per_gpu:4687.539125159918 mfu:22.111921830024436 +Step 3479 | loss:1.3726869821548462 lr:9.784335705746097e-05 tokens_per_second_per_gpu:3255.62561569295 mfu:15.357341496231625 +Step 3480 | loss:1.3373820781707764 lr:9.78420970930017e-05 tokens_per_second_per_gpu:3740.205776999922 mfu:17.643188795017604 +Step 3481 | loss:1.443371057510376 lr:9.784083676871592e-05 tokens_per_second_per_gpu:5316.478984295667 mfu:25.078738453773006 +Step 3482 | loss:1.2670044898986816 lr:9.783957608461311e-05 tokens_per_second_per_gpu:5305.61402550751 mfu:25.02748658941615 +Step 3483 | loss:1.218332052230835 lr:9.783831504070273e-05 tokens_per_second_per_gpu:4397.927493506377 mfu:20.74577435067504 +Step 3484 | loss:1.3034615516662598 lr:9.783705363699429e-05 tokens_per_second_per_gpu:4703.584768233133 mfu:22.187611866069805 +Step 3485 | loss:1.3766422271728516 lr:9.783579187349726e-05 tokens_per_second_per_gpu:5072.3409936575545 mfu:23.92709789769658 +Step 3486 | loss:1.343793272972107 lr:9.783452975022113e-05 tokens_per_second_per_gpu:5335.370338438811 mfu:25.1678521944636 +Step 3487 | loss:1.2948390245437622 lr:9.783326726717542e-05 tokens_per_second_per_gpu:4704.853556937818 mfu:22.193596958867477 +Step 3488 | loss:1.4605053663253784 lr:9.783200442436957e-05 tokens_per_second_per_gpu:5331.287778277422 mfu:25.14859406162574 +Step 3489 | loss:1.357327938079834 lr:9.783074122181314e-05 tokens_per_second_per_gpu:4937.200020786893 mfu:23.28961487123809 +Step 3490 | loss:1.427957534790039 lr:9.78294776595156e-05 tokens_per_second_per_gpu:4730.257354093561 mfu:22.31343100438585 +Step 3491 | loss:1.6111996173858643 lr:9.782821373748644e-05 tokens_per_second_per_gpu:5331.710492146964 mfu:25.15058807507079 +Step 3492 | loss:1.1658003330230713 lr:9.782694945573518e-05 tokens_per_second_per_gpu:4739.628327481355 mfu:22.357635484709007 +Step 3493 | loss:1.3194271326065063 lr:9.782568481427135e-05 tokens_per_second_per_gpu:5340.725636012653 mfu:25.193114046827397 +Step 3494 | loss:1.2753386497497559 lr:9.782441981310442e-05 tokens_per_second_per_gpu:5292.348954532271 mfu:24.96491298637146 +Step 3495 | loss:1.3262070417404175 lr:9.782315445224394e-05 tokens_per_second_per_gpu:3278.858654204369 mfu:15.46693570285511 +Step 3496 | loss:1.5232765674591064 lr:9.782188873169941e-05 tokens_per_second_per_gpu:5346.050937734093 mfu:25.2182344036376 +Step 3497 | loss:1.1924669742584229 lr:9.782062265148035e-05 tokens_per_second_per_gpu:4584.985651256497 mfu:21.62815959619508 +Step 3498 | loss:1.417507529258728 lr:9.781935621159628e-05 tokens_per_second_per_gpu:3795.2993868408025 mfu:17.90307475257617 +Step 3499 | loss:1.5032894611358643 lr:9.781808941205674e-05 tokens_per_second_per_gpu:4515.558130523635 mfu:21.300658135341653 +Step 3500 | loss:1.445537805557251 lr:9.781682225287126e-05 tokens_per_second_per_gpu:4965.003292522197 mfu:23.420767647740835 +Step 3501 | loss:1.2261065244674683 lr:9.781555473404933e-05 tokens_per_second_per_gpu:4703.473745257558 mfu:22.187088151751542 +Step 3502 | loss:1.4583158493041992 lr:9.781428685560052e-05 tokens_per_second_per_gpu:4146.061926859072 mfu:19.557681500102824 +Step 3503 | loss:1.2286522388458252 lr:9.781301861753437e-05 tokens_per_second_per_gpu:3785.4817335627195 mfu:17.856763206999076 +Step 3504 | loss:1.336634874343872 lr:9.78117500198604e-05 tokens_per_second_per_gpu:4629.315349154819 mfu:21.83727034460784 +Step 3505 | loss:1.2643768787384033 lr:9.781048106258816e-05 tokens_per_second_per_gpu:4683.165275334457 mfu:22.09128963414201 +Step 3506 | loss:1.3013643026351929 lr:9.780921174572718e-05 tokens_per_second_per_gpu:5156.510058171753 mfu:24.324137735733295 +Step 3507 | loss:1.2479290962219238 lr:9.780794206928703e-05 tokens_per_second_per_gpu:5337.301506394853 mfu:25.176961843203483 +Step 3508 | loss:1.538527250289917 lr:9.780667203327723e-05 tokens_per_second_per_gpu:5367.890176012942 mfu:25.321253816757206 +Step 3509 | loss:1.2562388181686401 lr:9.780540163770738e-05 tokens_per_second_per_gpu:5052.233559231055 mfu:23.83224769882471 +Step 3510 | loss:1.2871015071868896 lr:9.780413088258697e-05 tokens_per_second_per_gpu:5321.466086642274 mfu:25.102263466429356 +Step 3511 | loss:1.4078985452651978 lr:9.780285976792562e-05 tokens_per_second_per_gpu:3346.4901523180188 mfu:15.78596501858065 +Step 3512 | loss:1.3481768369674683 lr:9.780158829373285e-05 tokens_per_second_per_gpu:4578.020311038663 mfu:21.59530289797797 +Step 3513 | loss:1.4203954935073853 lr:9.780031646001823e-05 tokens_per_second_per_gpu:4427.335920921254 mfu:20.88449892038613 +Step 3514 | loss:1.324716567993164 lr:9.779904426679134e-05 tokens_per_second_per_gpu:3875.257807264583 mfu:18.280252264028416 +Step 3515 | loss:1.2576501369476318 lr:9.779777171406173e-05 tokens_per_second_per_gpu:5329.922841419635 mfu:25.142155421585315 +Step 3516 | loss:1.2232694625854492 lr:9.779649880183898e-05 tokens_per_second_per_gpu:4293.428027622232 mfu:20.25283254065213 +Step 3517 | loss:1.3419216871261597 lr:9.779522553013268e-05 tokens_per_second_per_gpu:3812.329910383391 mfu:17.9834106378335 +Step 3518 | loss:1.2815759181976318 lr:9.779395189895238e-05 tokens_per_second_per_gpu:3266.5417807385197 mfu:15.408834909241389 +Step 3519 | loss:1.480104923248291 lr:9.779267790830766e-05 tokens_per_second_per_gpu:3429.629554799521 mfu:16.178147765130674 +Step 3520 | loss:1.1991208791732788 lr:9.77914035582081e-05 tokens_per_second_per_gpu:5352.746332974488 mfu:25.24981772533876 +Step 3521 | loss:1.308083176612854 lr:9.779012884866332e-05 tokens_per_second_per_gpu:3541.553969915635 mfu:16.706114327507752 +Step 3522 | loss:1.534522533416748 lr:9.778885377968288e-05 tokens_per_second_per_gpu:4280.740214417343 mfu:20.192981961000513 +Step 3523 | loss:1.2959634065628052 lr:9.778757835127637e-05 tokens_per_second_per_gpu:4345.008469044626 mfu:20.496146283372347 +Step 3524 | loss:1.1679158210754395 lr:9.778630256345338e-05 tokens_per_second_per_gpu:4956.565502438756 mfu:23.380965152281806 +Step 3525 | loss:1.2484009265899658 lr:9.778502641622352e-05 tokens_per_second_per_gpu:3550.336875601864 mfu:16.747544792147924 +Step 3526 | loss:1.3894906044006348 lr:9.778374990959638e-05 tokens_per_second_per_gpu:3829.0389754461003 mfu:18.062230148594775 +Step 3527 | loss:1.4363625049591064 lr:9.778247304358154e-05 tokens_per_second_per_gpu:5336.214001542207 mfu:25.17183189726609 +Step 3528 | loss:1.3469711542129517 lr:9.778119581818865e-05 tokens_per_second_per_gpu:3261.6505918202884 mfu:15.38576233659121 +Step 3529 | loss:1.5995789766311646 lr:9.777991823342727e-05 tokens_per_second_per_gpu:5174.005334878694 mfu:24.40666594096194 +Step 3530 | loss:1.1957952976226807 lr:9.777864028930705e-05 tokens_per_second_per_gpu:4324.410801599957 mfu:20.398983571711398 +Step 3531 | loss:1.3542234897613525 lr:9.777736198583755e-05 tokens_per_second_per_gpu:4498.515375284634 mfu:21.2202645510854 +Step 3532 | loss:1.239364743232727 lr:9.777608332302843e-05 tokens_per_second_per_gpu:5241.927215115881 mfu:24.72706503870851 +Step 3533 | loss:1.2321499586105347 lr:9.77748043008893e-05 tokens_per_second_per_gpu:5349.854538247158 mfu:25.236176636218893 +Step 3534 | loss:1.5904302597045898 lr:9.777352491942976e-05 tokens_per_second_per_gpu:3806.1901890655067 mfu:17.954448524832568 +Step 3535 | loss:1.3467752933502197 lr:9.777224517865944e-05 tokens_per_second_per_gpu:5317.543215468176 mfu:25.083758613790295 +Step 3536 | loss:1.4210745096206665 lr:9.777096507858797e-05 tokens_per_second_per_gpu:4261.2544286445755 mfu:20.101064184892493 +Step 3537 | loss:1.390842318534851 lr:9.776968461922498e-05 tokens_per_second_per_gpu:3973.540299062491 mfu:18.743867546561386 +Step 3538 | loss:1.359187126159668 lr:9.77684038005801e-05 tokens_per_second_per_gpu:4135.380488668697 mfu:19.507295333717543 +Step 3539 | loss:1.3928754329681396 lr:9.776712262266295e-05 tokens_per_second_per_gpu:4806.221846049178 mfu:22.67176847381931 +Step 3540 | loss:1.4152023792266846 lr:9.776584108548317e-05 tokens_per_second_per_gpu:3969.489607810866 mfu:18.724759744808274 +Step 3541 | loss:1.2184468507766724 lr:9.776455918905043e-05 tokens_per_second_per_gpu:5094.1556414289 mfu:24.030001313196724 +Step 3542 | loss:1.403093934059143 lr:9.776327693337432e-05 tokens_per_second_per_gpu:3410.7209383643235 mfu:16.088952595263216 +Step 3543 | loss:1.362247347831726 lr:9.776199431846452e-05 tokens_per_second_per_gpu:5376.100181479478 mfu:25.35998181331423 +Step 3544 | loss:1.3190804719924927 lr:9.776071134433067e-05 tokens_per_second_per_gpu:4667.193060943033 mfu:22.015945973716878 +Step 3545 | loss:1.3307923078536987 lr:9.77594280109824e-05 tokens_per_second_per_gpu:4670.470091125404 mfu:22.03140428420613 +Step 3546 | loss:1.3514653444290161 lr:9.775814431842941e-05 tokens_per_second_per_gpu:4320.229428985983 mfu:20.379259323675548 +Step 3547 | loss:1.2083375453948975 lr:9.77568602666813e-05 tokens_per_second_per_gpu:4525.533743674152 mfu:21.347714804587753 +Step 3548 | loss:1.5073826313018799 lr:9.775557585574775e-05 tokens_per_second_per_gpu:4245.071418395611 mfu:20.024726164441862 +Step 3549 | loss:1.2996937036514282 lr:9.775429108563842e-05 tokens_per_second_per_gpu:4601.372939534372 mfu:21.705461230961504 +Step 3550 | loss:1.2539328336715698 lr:9.775300595636296e-05 tokens_per_second_per_gpu:5364.208672028566 mfu:25.303887534333555 +Step 3551 | loss:1.1179441213607788 lr:9.775172046793108e-05 tokens_per_second_per_gpu:5331.4343159541795 mfu:25.14928530484891 +Step 3552 | loss:1.2416661977767944 lr:9.77504346203524e-05 tokens_per_second_per_gpu:4392.721720802329 mfu:20.721217832633634 +Step 3553 | loss:1.2943803071975708 lr:9.774914841363661e-05 tokens_per_second_per_gpu:5086.904067317888 mfu:23.995794400868956 +Step 3554 | loss:1.3426930904388428 lr:9.774786184779338e-05 tokens_per_second_per_gpu:4075.2411659395952 mfu:19.223607887577852 +Step 3555 | loss:1.5721189975738525 lr:9.774657492283239e-05 tokens_per_second_per_gpu:5306.517459258504 mfu:25.031748240561278 +Step 3556 | loss:1.3255926370620728 lr:9.774528763876332e-05 tokens_per_second_per_gpu:3179.1396742479055 mfu:14.996544260588083 +Step 3557 | loss:1.3829346895217896 lr:9.774399999559585e-05 tokens_per_second_per_gpu:4216.45893623088 mfu:19.889756204277973 +Step 3558 | loss:1.1202940940856934 lr:9.774271199333965e-05 tokens_per_second_per_gpu:4719.1920233944065 mfu:22.261233951537847 +Step 3559 | loss:1.2890212535858154 lr:9.774142363200442e-05 tokens_per_second_per_gpu:4548.015594043117 mfu:21.45376552857737 +Step 3560 | loss:1.4513643980026245 lr:9.774013491159987e-05 tokens_per_second_per_gpu:5088.452727485783 mfu:24.003099695109587 +Step 3561 | loss:1.4224063158035278 lr:9.773884583213567e-05 tokens_per_second_per_gpu:4198.31302638108 mfu:19.804158851504994 +Step 3562 | loss:1.2234786748886108 lr:9.773755639362153e-05 tokens_per_second_per_gpu:3873.2992929307484 mfu:18.27101361775872 +Step 3563 | loss:1.346778392791748 lr:9.773626659606712e-05 tokens_per_second_per_gpu:4162.718611724951 mfu:19.636253924538664 +Step 3564 | loss:1.3592438697814941 lr:9.773497643948215e-05 tokens_per_second_per_gpu:4323.506510929174 mfu:20.394717878329413 +Step 3565 | loss:1.524492621421814 lr:9.773368592387634e-05 tokens_per_second_per_gpu:4924.979221853362 mfu:23.231967277585323 +Step 3566 | loss:1.1310577392578125 lr:9.77323950492594e-05 tokens_per_second_per_gpu:5338.480411289952 mfu:25.18252293873558 +Step 3567 | loss:1.378889560699463 lr:9.773110381564101e-05 tokens_per_second_per_gpu:4460.204223964525 mfu:21.039544313752003 +Step 3568 | loss:1.2008559703826904 lr:9.77298122230309e-05 tokens_per_second_per_gpu:5358.87448469725 mfu:25.27872526258553 +Step 3569 | loss:1.225696086883545 lr:9.772852027143879e-05 tokens_per_second_per_gpu:4904.446366082004 mfu:23.135110293645383 +Step 3570 | loss:1.2539198398590088 lr:9.772722796087437e-05 tokens_per_second_per_gpu:4441.94195662568 mfu:20.95339808285057 +Step 3571 | loss:1.3510252237319946 lr:9.77259352913474e-05 tokens_per_second_per_gpu:4366.2535669552735 mfu:20.596363035004355 +Step 3572 | loss:1.417973518371582 lr:9.772464226286757e-05 tokens_per_second_per_gpu:3708.585753032025 mfu:17.49403174702864 +Step 3573 | loss:1.1469565629959106 lr:9.77233488754446e-05 tokens_per_second_per_gpu:4343.217317226296 mfu:20.48769711464327 +Step 3574 | loss:1.3500258922576904 lr:9.772205512908828e-05 tokens_per_second_per_gpu:4397.812265191829 mfu:20.745230799055335 +Step 3575 | loss:1.354308009147644 lr:9.772076102380825e-05 tokens_per_second_per_gpu:5155.113996197805 mfu:24.317552273209444 +Step 3576 | loss:1.2289581298828125 lr:9.77194665596143e-05 tokens_per_second_per_gpu:3726.3304009962567 mfu:17.577736279025054 +Step 3577 | loss:1.2666620016098022 lr:9.771817173651616e-05 tokens_per_second_per_gpu:4763.240114994456 mfu:22.46901631499467 +Step 3578 | loss:1.4549716711044312 lr:9.771687655452356e-05 tokens_per_second_per_gpu:5342.612940569899 mfu:25.202016784431294 +Step 3579 | loss:1.410137414932251 lr:9.771558101364623e-05 tokens_per_second_per_gpu:4195.334701998028 mfu:19.790109587235605 +Step 3580 | loss:1.2752270698547363 lr:9.771428511389395e-05 tokens_per_second_per_gpu:3482.9273567536284 mfu:16.42956259049091 +Step 3581 | loss:1.294394612312317 lr:9.771298885527642e-05 tokens_per_second_per_gpu:5164.261630975753 mfu:24.360703227204457 +Step 3582 | loss:1.2113689184188843 lr:9.771169223780343e-05 tokens_per_second_per_gpu:3641.9923757915935 mfu:17.179899424583756 +Step 3583 | loss:1.4278331995010376 lr:9.771039526148471e-05 tokens_per_second_per_gpu:4183.137034231561 mfu:19.732571107244457 +Step 3584 | loss:1.3242552280426025 lr:9.770909792633002e-05 tokens_per_second_per_gpu:4015.956855023488 mfu:18.943953678040113 +Step 3585 | loss:1.3991583585739136 lr:9.770780023234911e-05 tokens_per_second_per_gpu:3315.802486419541 mfu:15.641206062681825 +Step 3586 | loss:1.2699521780014038 lr:9.770650217955176e-05 tokens_per_second_per_gpu:4817.265879982563 mfu:22.723865066189582 +Step 3587 | loss:1.4672843217849731 lr:9.770520376794769e-05 tokens_per_second_per_gpu:4875.522028925988 mfu:22.998669260279655 +Step 3588 | loss:1.5637298822402954 lr:9.770390499754672e-05 tokens_per_second_per_gpu:5321.905242031905 mfu:25.104335037329882 +Step 3589 | loss:1.410710334777832 lr:9.77026058683586e-05 tokens_per_second_per_gpu:3749.1169445956575 mfu:17.685224292968662 +Step 3590 | loss:1.3329722881317139 lr:9.770130638039308e-05 tokens_per_second_per_gpu:4879.848186810664 mfu:23.01907648514032 +Step 3591 | loss:1.3249843120574951 lr:9.770000653365996e-05 tokens_per_second_per_gpu:5316.914355322124 mfu:25.08079217318676 +Step 3592 | loss:1.4435406923294067 lr:9.769870632816898e-05 tokens_per_second_per_gpu:4094.1730716546635 mfu:19.312912916952023 +Step 3593 | loss:1.191636562347412 lr:9.769740576392997e-05 tokens_per_second_per_gpu:5109.419329576405 mfu:24.102002734442245 +Step 3594 | loss:1.3683595657348633 lr:9.769610484095266e-05 tokens_per_second_per_gpu:3977.5376804452794 mfu:18.76272387656759 +Step 3595 | loss:1.270459771156311 lr:9.769480355924687e-05 tokens_per_second_per_gpu:5228.64454389906 mfu:24.664408412321368 +Step 3596 | loss:1.4515661001205444 lr:9.769350191882238e-05 tokens_per_second_per_gpu:5333.401112695537 mfu:25.15856302064802 +Step 3597 | loss:1.2234702110290527 lr:9.769219991968898e-05 tokens_per_second_per_gpu:5326.080411261096 mfu:25.124030022941298 +Step 3598 | loss:1.2173328399658203 lr:9.769089756185646e-05 tokens_per_second_per_gpu:5339.23110255163 mfu:25.18606407749805 +Step 3599 | loss:1.3861790895462036 lr:9.76895948453346e-05 tokens_per_second_per_gpu:5346.803154236086 mfu:25.221782737218863 +Step 3600 | loss:1.5327770709991455 lr:9.768829177013323e-05 tokens_per_second_per_gpu:5286.059163598114 mfu:24.935242969386046 +Step 3601 | loss:1.4335248470306396 lr:9.768698833626213e-05 tokens_per_second_per_gpu:5315.419673878336 mfu:25.073741505797464 +Step 3602 | loss:1.3085380792617798 lr:9.768568454373109e-05 tokens_per_second_per_gpu:3256.9319392696066 mfu:15.3635036474245 +Step 3603 | loss:1.3434566259384155 lr:9.768438039254995e-05 tokens_per_second_per_gpu:4830.522685940885 mfu:22.786399681739315 +Step 3604 | loss:1.3957587480545044 lr:9.768307588272849e-05 tokens_per_second_per_gpu:5349.586366588795 mfu:25.234911624751586 +Step 3605 | loss:1.250243067741394 lr:9.768177101427654e-05 tokens_per_second_per_gpu:5039.230329069664 mfu:23.770909243573584 +Step 3606 | loss:1.3831429481506348 lr:9.768046578720389e-05 tokens_per_second_per_gpu:5003.8211075922345 mfu:23.603877904428398 +Step 3607 | loss:1.4321879148483276 lr:9.767916020152038e-05 tokens_per_second_per_gpu:5063.777113495728 mfu:23.886700613824623 +Step 3608 | loss:1.309730887413025 lr:9.767785425723582e-05 tokens_per_second_per_gpu:4641.446632449779 mfu:21.89449567772053 +Step 3609 | loss:1.3356448411941528 lr:9.767654795436005e-05 tokens_per_second_per_gpu:4738.156879868445 mfu:22.350694415264062 +Step 3610 | loss:1.2328300476074219 lr:9.767524129290287e-05 tokens_per_second_per_gpu:4585.228342787684 mfu:21.62930441355513 +Step 3611 | loss:1.3747931718826294 lr:9.76739342728741e-05 tokens_per_second_per_gpu:5054.063572288641 mfu:23.840880182649762 +Step 3612 | loss:1.3492854833602905 lr:9.76726268942836e-05 tokens_per_second_per_gpu:4033.756710446614 mfu:19.027918633039494 +Step 3613 | loss:1.3825933933258057 lr:9.76713191571412e-05 tokens_per_second_per_gpu:4233.82935662083 mfu:19.971695440956697 +Step 3614 | loss:1.3947404623031616 lr:9.767001106145673e-05 tokens_per_second_per_gpu:5356.977876306639 mfu:25.26977863721226 +Step 3615 | loss:1.410158634185791 lr:9.766870260724e-05 tokens_per_second_per_gpu:3791.1442282099188 mfu:17.883474160371126 +Step 3616 | loss:1.2120814323425293 lr:9.76673937945009e-05 tokens_per_second_per_gpu:4420.502800576752 mfu:20.85226592587056 +Step 3617 | loss:1.2227004766464233 lr:9.766608462324923e-05 tokens_per_second_per_gpu:5306.102776387136 mfu:25.029792110704847 +Step 3618 | loss:1.4511302709579468 lr:9.766477509349486e-05 tokens_per_second_per_gpu:4177.030461236046 mfu:19.703765360535822 +Step 3619 | loss:1.3324151039123535 lr:9.766346520524763e-05 tokens_per_second_per_gpu:5359.911770599126 mfu:25.283618317163903 +Step 3620 | loss:1.34725821018219 lr:9.76621549585174e-05 tokens_per_second_per_gpu:3533.934471018053 mfu:16.67017185118729 +Step 3621 | loss:1.1257853507995605 lr:9.766084435331402e-05 tokens_per_second_per_gpu:5347.927900936053 mfu:25.227088359304187 +Step 3622 | loss:1.2913140058517456 lr:9.765953338964735e-05 tokens_per_second_per_gpu:5335.077087430999 mfu:25.16646887942597 +Step 3623 | loss:1.2549902200698853 lr:9.765822206752726e-05 tokens_per_second_per_gpu:4975.802817896799 mfu:23.471710851521866 +Step 3624 | loss:1.2603296041488647 lr:9.765691038696358e-05 tokens_per_second_per_gpu:5332.551826224685 mfu:25.154556791461797 +Step 3625 | loss:1.734884262084961 lr:9.76555983479662e-05 tokens_per_second_per_gpu:5308.54793657744 mfu:25.041326348509855 +Step 3626 | loss:1.4173811674118042 lr:9.7654285950545e-05 tokens_per_second_per_gpu:3584.990388253121 mfu:16.91101132382284 +Step 3627 | loss:1.1845766305923462 lr:9.765297319470983e-05 tokens_per_second_per_gpu:5310.81344439531 mfu:25.05201313542224 +Step 3628 | loss:1.353530764579773 lr:9.765166008047055e-05 tokens_per_second_per_gpu:4937.998626958389 mfu:23.293382032805297 +Step 3629 | loss:1.2893891334533691 lr:9.765034660783708e-05 tokens_per_second_per_gpu:5361.604173082332 mfu:25.291601668430822 +Step 3630 | loss:1.3387631177902222 lr:9.764903277681925e-05 tokens_per_second_per_gpu:5359.469158339012 mfu:25.281530439615505 +Step 3631 | loss:1.2398691177368164 lr:9.764771858742697e-05 tokens_per_second_per_gpu:5350.1480769858845 mfu:25.237561308532335 +Step 3632 | loss:1.4837228059768677 lr:9.764640403967012e-05 tokens_per_second_per_gpu:5338.189602788739 mfu:25.18115114541831 +Step 3633 | loss:1.3343826532363892 lr:9.764508913355858e-05 tokens_per_second_per_gpu:4362.585348108255 mfu:20.579059420840295 +Step 3634 | loss:1.3729119300842285 lr:9.764377386910226e-05 tokens_per_second_per_gpu:3877.555190122666 mfu:18.291089410943925 +Step 3635 | loss:1.3951499462127686 lr:9.764245824631104e-05 tokens_per_second_per_gpu:4300.808310509859 mfu:20.287646594239174 +Step 3636 | loss:1.277830958366394 lr:9.764114226519479e-05 tokens_per_second_per_gpu:4749.612851369153 mfu:22.404734187423188 +Step 3637 | loss:1.351955771446228 lr:9.763982592576346e-05 tokens_per_second_per_gpu:5292.7063463591785 mfu:24.966598864596314 +Step 3638 | loss:1.255700945854187 lr:9.763850922802688e-05 tokens_per_second_per_gpu:5354.462374112951 mfu:25.2579125841757 +Step 3639 | loss:1.4257094860076904 lr:9.763719217199503e-05 tokens_per_second_per_gpu:5240.683015739463 mfu:24.72119593796764 +Step 3640 | loss:1.2757924795150757 lr:9.763587475767778e-05 tokens_per_second_per_gpu:4976.514796353655 mfu:23.475069375378904 +Step 3641 | loss:1.2401864528656006 lr:9.7634556985085e-05 tokens_per_second_per_gpu:4538.079540506393 mfu:21.406895468779098 +Step 3642 | loss:1.47670578956604 lr:9.763323885422666e-05 tokens_per_second_per_gpu:3780.6180900484164 mfu:17.833820570718956 +Step 3643 | loss:1.315980315208435 lr:9.763192036511265e-05 tokens_per_second_per_gpu:5361.235546522536 mfu:25.289862794054095 +Step 3644 | loss:1.3576548099517822 lr:9.76306015177529e-05 tokens_per_second_per_gpu:3532.456237921087 mfu:16.663198773456326 +Step 3645 | loss:1.247042179107666 lr:9.76292823121573e-05 tokens_per_second_per_gpu:4856.0860935662195 mfu:22.90698663707511 +Step 3646 | loss:1.4330689907073975 lr:9.762796274833581e-05 tokens_per_second_per_gpu:4771.405318755781 mfu:22.507532974264254 +Step 3647 | loss:1.2015608549118042 lr:9.762664282629833e-05 tokens_per_second_per_gpu:4842.424511296253 mfu:22.84254262260139 +Step 3648 | loss:1.3318661451339722 lr:9.762532254605478e-05 tokens_per_second_per_gpu:4612.392687510696 mfu:21.757443262328707 +Step 3649 | loss:1.3842805624008179 lr:9.762400190761511e-05 tokens_per_second_per_gpu:4467.0089206758175 mfu:21.07164322914015 +Step 3650 | loss:1.4613624811172485 lr:9.762268091098925e-05 tokens_per_second_per_gpu:3604.4631957996357 mfu:17.002867879423313 +Step 3651 | loss:1.3114149570465088 lr:9.762135955618713e-05 tokens_per_second_per_gpu:5371.456313038695 mfu:25.3380758935534 +Step 3652 | loss:1.3747190237045288 lr:9.762003784321869e-05 tokens_per_second_per_gpu:5192.291423365295 mfu:24.49292454028165 +Step 3653 | loss:1.1901614665985107 lr:9.761871577209384e-05 tokens_per_second_per_gpu:3595.63571750182 mfu:16.96122715816941 +Step 3654 | loss:1.4546538591384888 lr:9.76173933428226e-05 tokens_per_second_per_gpu:4894.11243118952 mfu:23.086363359606434 +Step 3655 | loss:1.4524977207183838 lr:9.761607055541484e-05 tokens_per_second_per_gpu:5003.240002684365 mfu:23.601136733431257 +Step 3656 | loss:1.4619860649108887 lr:9.761474740988053e-05 tokens_per_second_per_gpu:3689.477388999432 mfu:17.40389433366381 +Step 3657 | loss:1.2841798067092896 lr:9.761342390622965e-05 tokens_per_second_per_gpu:3262.4290446030463 mfu:15.389434431185217 +Step 3658 | loss:1.1970020532608032 lr:9.761210004447212e-05 tokens_per_second_per_gpu:5300.086631673966 mfu:25.00141292963269 +Step 3659 | loss:1.2694331407546997 lr:9.761077582461792e-05 tokens_per_second_per_gpu:5033.842144832965 mfu:23.74549225921005 +Step 3660 | loss:1.1138733625411987 lr:9.760945124667698e-05 tokens_per_second_per_gpu:5331.552667521344 mfu:25.149843589384083 +Step 3661 | loss:1.394654393196106 lr:9.76081263106593e-05 tokens_per_second_per_gpu:3354.391761138458 mfu:15.823238255540145 +Step 3662 | loss:1.3851624727249146 lr:9.760680101657482e-05 tokens_per_second_per_gpu:5361.47081044008 mfu:25.290972574093306 +Step 3663 | loss:1.4458470344543457 lr:9.760547536443352e-05 tokens_per_second_per_gpu:3854.247446187934 mfu:18.18114280609262 +Step 3664 | loss:1.0964981317520142 lr:9.760414935424537e-05 tokens_per_second_per_gpu:4476.773248358438 mfu:21.11770322878516 +Step 3665 | loss:1.4266180992126465 lr:9.760282298602032e-05 tokens_per_second_per_gpu:4831.516631604323 mfu:22.79108829301837 +Step 3666 | loss:1.296755313873291 lr:9.760149625976839e-05 tokens_per_second_per_gpu:5328.519595160821 mfu:25.13553606956399 +Step 3667 | loss:1.3849714994430542 lr:9.760016917549952e-05 tokens_per_second_per_gpu:5324.044507001561 mfu:25.114426315187988 +Step 3668 | loss:1.2064051628112793 lr:9.759884173322369e-05 tokens_per_second_per_gpu:5311.268965037849 mfu:25.05416190401293 +Step 3669 | loss:1.3962388038635254 lr:9.759751393295092e-05 tokens_per_second_per_gpu:4927.891832279979 mfu:23.245706557910257 +Step 3670 | loss:1.2802098989486694 lr:9.759618577469117e-05 tokens_per_second_per_gpu:5343.781248059475 mfu:25.207527890193433 +Step 3671 | loss:1.2771044969558716 lr:9.759485725845442e-05 tokens_per_second_per_gpu:4228.455747676244 mfu:19.946347210732853 +Step 3672 | loss:1.406121850013733 lr:9.759352838425069e-05 tokens_per_second_per_gpu:3730.411166307266 mfu:17.596985945247443 +Step 3673 | loss:1.3592840433120728 lr:9.759219915208994e-05 tokens_per_second_per_gpu:5347.703566259097 mfu:25.22603013435754 +Step 3674 | loss:1.266494870185852 lr:9.75908695619822e-05 tokens_per_second_per_gpu:4234.356982658351 mfu:19.97418434299796 +Step 3675 | loss:1.1753754615783691 lr:9.758953961393747e-05 tokens_per_second_per_gpu:5314.8245921331345 mfu:25.070934403673935 +Step 3676 | loss:1.4860600233078003 lr:9.758820930796572e-05 tokens_per_second_per_gpu:4919.754905444369 mfu:23.207323285723806 +Step 3677 | loss:1.4873011112213135 lr:9.758687864407698e-05 tokens_per_second_per_gpu:3751.1566286460015 mfu:17.69484583064012 +Step 3678 | loss:1.3551121950149536 lr:9.758554762228126e-05 tokens_per_second_per_gpu:5336.318039175118 mfu:25.172322660531183 +Step 3679 | loss:1.376791000366211 lr:9.758421624258855e-05 tokens_per_second_per_gpu:3741.6938772135327 mfu:17.650208417621286 +Step 3680 | loss:1.2781975269317627 lr:9.75828845050089e-05 tokens_per_second_per_gpu:4402.642474231871 mfu:20.768015719215857 +Step 3681 | loss:1.3452833890914917 lr:9.758155240955229e-05 tokens_per_second_per_gpu:4449.188210617914 mfu:20.98757989927035 +Step 3682 | loss:1.3078258037567139 lr:9.758021995622877e-05 tokens_per_second_per_gpu:4700.440916774311 mfu:22.172781782342167 +Step 3683 | loss:1.4015485048294067 lr:9.757888714504832e-05 tokens_per_second_per_gpu:5303.286844184927 mfu:25.016508878059447 +Step 3684 | loss:1.3950060606002808 lr:9.7577553976021e-05 tokens_per_second_per_gpu:5320.468668612908 mfu:25.09755847540777 +Step 3685 | loss:1.3223803043365479 lr:9.757622044915682e-05 tokens_per_second_per_gpu:4037.361451045691 mfu:19.04492280947751 +Step 3686 | loss:1.5343687534332275 lr:9.757488656446583e-05 tokens_per_second_per_gpu:5338.632362299447 mfu:25.18323971757288 +Step 3687 | loss:1.282593011856079 lr:9.757355232195804e-05 tokens_per_second_per_gpu:5306.804552672648 mfu:25.033102509178985 +Step 3688 | loss:1.580311894416809 lr:9.757221772164348e-05 tokens_per_second_per_gpu:3546.755483379045 mfu:16.730650753984364 +Step 3689 | loss:1.252991795539856 lr:9.757088276353221e-05 tokens_per_second_per_gpu:4507.601733290107 mfu:21.263126452089754 +Step 3690 | loss:1.3751802444458008 lr:9.756954744763426e-05 tokens_per_second_per_gpu:3046.0475432621574 mfu:14.368725970869203 +Step 3691 | loss:1.3479946851730347 lr:9.756821177395968e-05 tokens_per_second_per_gpu:4100.420956783234 mfu:19.342385257102617 +Step 3692 | loss:1.17744779586792 lr:9.756687574251851e-05 tokens_per_second_per_gpu:4599.118469609348 mfu:21.694826511672936 +Step 3693 | loss:1.2228068113327026 lr:9.756553935332079e-05 tokens_per_second_per_gpu:4338.283686755229 mfu:20.464424337947094 +Step 3694 | loss:1.247717261314392 lr:9.756420260637657e-05 tokens_per_second_per_gpu:3948.8017172466866 mfu:18.62717143529894 +Step 3695 | loss:1.552805781364441 lr:9.756286550169593e-05 tokens_per_second_per_gpu:5046.324549534447 mfu:23.804373891904792 +Step 3696 | loss:1.3867788314819336 lr:9.756152803928891e-05 tokens_per_second_per_gpu:4365.452721682039 mfu:20.592585311212215 +Step 3697 | loss:1.2542519569396973 lr:9.756019021916555e-05 tokens_per_second_per_gpu:5323.248701180466 mfu:25.11067236335141 +Step 3698 | loss:1.2666820287704468 lr:9.755885204133594e-05 tokens_per_second_per_gpu:4778.512325596833 mfu:22.5410579381139 +Step 3699 | loss:1.2576816082000732 lr:9.755751350581015e-05 tokens_per_second_per_gpu:5329.763825912978 mfu:25.141405318309573 +Step 3700 | loss:1.2172545194625854 lr:9.755617461259822e-05 tokens_per_second_per_gpu:5331.470175668021 mfu:25.149454461237394 +Step 3701 | loss:1.1915112733840942 lr:9.755483536171023e-05 tokens_per_second_per_gpu:5360.035536808908 mfu:25.284202143492138 +Step 3702 | loss:1.2719180583953857 lr:9.755349575315627e-05 tokens_per_second_per_gpu:4085.4454239009283 mfu:19.271743113407396 +Step 3703 | loss:1.4077873229980469 lr:9.755215578694639e-05 tokens_per_second_per_gpu:5286.048326357624 mfu:24.935191848273632 +Step 3704 | loss:1.199454426765442 lr:9.755081546309068e-05 tokens_per_second_per_gpu:4382.880230915836 mfu:20.674793845707725 +Step 3705 | loss:1.239391565322876 lr:9.754947478159922e-05 tokens_per_second_per_gpu:5332.474700544428 mfu:25.154192976469027 +Step 3706 | loss:1.3974549770355225 lr:9.754813374248209e-05 tokens_per_second_per_gpu:4611.1453559794545 mfu:21.751559386679133 +Step 3707 | loss:1.2758549451828003 lr:9.754679234574937e-05 tokens_per_second_per_gpu:5303.116829736644 mfu:25.01570689089944 +Step 3708 | loss:1.4810569286346436 lr:9.754545059141116e-05 tokens_per_second_per_gpu:5338.735298342602 mfu:25.183725284451143 +Step 3709 | loss:1.4528560638427734 lr:9.754410847947757e-05 tokens_per_second_per_gpu:4489.494268214902 mfu:21.177710449877896 +Step 3710 | loss:1.3198215961456299 lr:9.754276600995865e-05 tokens_per_second_per_gpu:4108.849172222596 mfu:19.382142587331924 +Step 3711 | loss:1.4362428188323975 lr:9.754142318286453e-05 tokens_per_second_per_gpu:3212.691403747851 mfu:15.154813493154672 +Step 3712 | loss:1.3772882223129272 lr:9.75400799982053e-05 tokens_per_second_per_gpu:2733.5144669310225 mfu:12.894454126173857 +Step 3713 | loss:1.3546490669250488 lr:9.753873645599107e-05 tokens_per_second_per_gpu:4111.262043535837 mfu:19.393524512994773 +Step 3714 | loss:1.4496251344680786 lr:9.753739255623192e-05 tokens_per_second_per_gpu:4017.65031810159 mfu:18.951942032314115 +Step 3715 | loss:1.3620589971542358 lr:9.753604829893799e-05 tokens_per_second_per_gpu:3515.042239915399 mfu:16.581053973729063 +Step 3716 | loss:1.324544072151184 lr:9.753470368411936e-05 tokens_per_second_per_gpu:5358.724650623354 mfu:25.278018469694633 +Step 3717 | loss:1.1844265460968018 lr:9.753335871178618e-05 tokens_per_second_per_gpu:5046.428720974204 mfu:23.804865286359608 +Step 3718 | loss:1.4964251518249512 lr:9.753201338194853e-05 tokens_per_second_per_gpu:5088.420099023596 mfu:24.00294578108646 +Step 3719 | loss:1.3321915864944458 lr:9.753066769461654e-05 tokens_per_second_per_gpu:4085.288483338922 mfu:19.271002797999596 +Step 3720 | loss:1.3300832509994507 lr:9.752932164980034e-05 tokens_per_second_per_gpu:4280.123760650697 mfu:20.19007404340538 +Step 3721 | loss:1.5076839923858643 lr:9.752797524751004e-05 tokens_per_second_per_gpu:4071.7622701042583 mfu:19.20719734236173 +Step 3722 | loss:1.4453818798065186 lr:9.752662848775578e-05 tokens_per_second_per_gpu:5333.794818257428 mfu:25.160420196956558 +Step 3723 | loss:1.3705496788024902 lr:9.752528137054767e-05 tokens_per_second_per_gpu:3705.783352125698 mfu:17.480812343814595 +Step 3724 | loss:1.3649518489837646 lr:9.752393389589587e-05 tokens_per_second_per_gpu:5182.218252807019 mfu:24.445407676098142 +Step 3725 | loss:1.4143904447555542 lr:9.75225860638105e-05 tokens_per_second_per_gpu:5051.16946392215 mfu:23.82722817969922 +Step 3726 | loss:1.3133442401885986 lr:9.752123787430168e-05 tokens_per_second_per_gpu:4097.6073434371765 mfu:19.329112962897327 +Step 3727 | loss:1.4559245109558105 lr:9.751988932737957e-05 tokens_per_second_per_gpu:3741.423912975133 mfu:17.648934950248105 +Step 3728 | loss:1.3924418687820435 lr:9.751854042305433e-05 tokens_per_second_per_gpu:3364.341680084261 mfu:15.870173720838723 +Step 3729 | loss:1.268329381942749 lr:9.751719116133606e-05 tokens_per_second_per_gpu:4935.113146932159 mfu:23.279770731207915 +Step 3730 | loss:1.2063016891479492 lr:9.751584154223495e-05 tokens_per_second_per_gpu:4330.568213926164 mfu:20.428029136216992 +Step 3731 | loss:1.209185242652893 lr:9.75144915657611e-05 tokens_per_second_per_gpu:5338.708804054824 mfu:25.183600306375304 +Step 3732 | loss:1.3962523937225342 lr:9.751314123192474e-05 tokens_per_second_per_gpu:4698.544744318386 mfu:22.163837213347186 +Step 3733 | loss:1.3766226768493652 lr:9.751179054073596e-05 tokens_per_second_per_gpu:4505.774263379458 mfu:21.254505964722707 +Step 3734 | loss:1.2881630659103394 lr:9.751043949220494e-05 tokens_per_second_per_gpu:5331.281667936114 mfu:25.148565238103256 +Step 3735 | loss:1.279756784439087 lr:9.750908808634185e-05 tokens_per_second_per_gpu:5336.671003239139 mfu:25.17398765224305 +Step 3736 | loss:1.2518630027770996 lr:9.750773632315684e-05 tokens_per_second_per_gpu:3857.695871915274 mfu:18.197409618610582 +Step 3737 | loss:1.1752411127090454 lr:9.750638420266008e-05 tokens_per_second_per_gpu:5105.133806857108 mfu:24.081787192589747 +Step 3738 | loss:1.366225004196167 lr:9.750503172486175e-05 tokens_per_second_per_gpu:2804.9678255512636 mfu:13.231511809985632 +Step 3739 | loss:1.334570050239563 lr:9.750367888977201e-05 tokens_per_second_per_gpu:4266.633283703856 mfu:20.126437161934426 +Step 3740 | loss:1.259983777999878 lr:9.750232569740104e-05 tokens_per_second_per_gpu:5275.583754841635 mfu:24.885828678992702 +Step 3741 | loss:1.2333269119262695 lr:9.750097214775902e-05 tokens_per_second_per_gpu:4322.426903371114 mfu:20.389625185278003 +Step 3742 | loss:1.476234793663025 lr:9.749961824085612e-05 tokens_per_second_per_gpu:4144.849080449963 mfu:19.551960296658123 +Step 3743 | loss:1.411948561668396 lr:9.749826397670254e-05 tokens_per_second_per_gpu:5300.900558086009 mfu:25.00525235938846 +Step 3744 | loss:1.327735424041748 lr:9.749690935530845e-05 tokens_per_second_per_gpu:4428.551912584036 mfu:20.89023496052904 +Step 3745 | loss:1.2393683195114136 lr:9.749555437668405e-05 tokens_per_second_per_gpu:5354.311377033957 mfu:25.25720030519133 +Step 3746 | loss:1.3157906532287598 lr:9.749419904083951e-05 tokens_per_second_per_gpu:5334.682296351474 mfu:25.16460658254555 +Step 3747 | loss:1.3785065412521362 lr:9.749284334778506e-05 tokens_per_second_per_gpu:4822.268837603214 mfu:22.74746487087878 +Step 3748 | loss:1.2089844942092896 lr:9.749148729753086e-05 tokens_per_second_per_gpu:5121.66624921115 mfu:24.15977354389929 +Step 3749 | loss:1.5346729755401611 lr:9.749013089008713e-05 tokens_per_second_per_gpu:4819.122618219594 mfu:22.732623617245867 +Step 3750 | loss:1.2870814800262451 lr:9.748877412546406e-05 tokens_per_second_per_gpu:5373.688570256665 mfu:25.348605831712916 +Step 3751 | loss:1.4317963123321533 lr:9.748741700367187e-05 tokens_per_second_per_gpu:3678.13700001884 mfu:17.35039978397248 +Step 3752 | loss:1.3298152685165405 lr:9.748605952472075e-05 tokens_per_second_per_gpu:3841.0458526481348 mfu:18.118868636941265 +Step 3753 | loss:1.4335215091705322 lr:9.748470168862092e-05 tokens_per_second_per_gpu:5380.706784121535 mfu:25.38171194394427 +Step 3754 | loss:1.217894434928894 lr:9.748334349538258e-05 tokens_per_second_per_gpu:5075.558597845841 mfu:23.94227588563261 +Step 3755 | loss:1.5277855396270752 lr:9.748198494501597e-05 tokens_per_second_per_gpu:3986.836794320288 mfu:18.806589383308438 +Step 3756 | loss:1.1839138269424438 lr:9.748062603753129e-05 tokens_per_second_per_gpu:5006.549379566042 mfu:23.616747628820043 +Step 3757 | loss:1.2903213500976562 lr:9.747926677293875e-05 tokens_per_second_per_gpu:4501.971354435052 mfu:21.236566994388767 +Step 3758 | loss:1.3408210277557373 lr:9.74779071512486e-05 tokens_per_second_per_gpu:4434.372404464084 mfu:20.917691213805544 +Step 3759 | loss:1.2570112943649292 lr:9.747654717247104e-05 tokens_per_second_per_gpu:4713.109208863253 mfu:22.23254027755922 +Step 3760 | loss:1.4853174686431885 lr:9.747518683661631e-05 tokens_per_second_per_gpu:5080.056598128362 mfu:23.963493720403243 +Step 3761 | loss:1.3585439920425415 lr:9.747382614369467e-05 tokens_per_second_per_gpu:5310.805893452242 mfu:25.051977516335466 +Step 3762 | loss:1.19890558719635 lr:9.74724650937163e-05 tokens_per_second_per_gpu:5217.538193988415 mfu:24.612017864854156 +Step 3763 | loss:1.400639533996582 lr:9.747110368669145e-05 tokens_per_second_per_gpu:3829.503660096318 mfu:18.06442214537323 +Step 3764 | loss:1.6034375429153442 lr:9.746974192263041e-05 tokens_per_second_per_gpu:4092.202742320749 mfu:19.303618537311102 +Step 3765 | loss:1.413107991218567 lr:9.746837980154335e-05 tokens_per_second_per_gpu:4075.064506549057 mfu:19.222774555091487 +Step 3766 | loss:1.3174196481704712 lr:9.746701732344057e-05 tokens_per_second_per_gpu:4113.180863716228 mfu:19.40257591517016 +Step 3767 | loss:1.1866791248321533 lr:9.74656544883323e-05 tokens_per_second_per_gpu:5206.744333390998 mfu:24.561101382793616 +Step 3768 | loss:1.4262943267822266 lr:9.746429129622876e-05 tokens_per_second_per_gpu:4113.24643921795 mfu:19.40288524600012 +Step 3769 | loss:1.280622124671936 lr:9.746292774714024e-05 tokens_per_second_per_gpu:4433.364553943781 mfu:20.912937010944248 +Step 3770 | loss:1.3570300340652466 lr:9.746156384107699e-05 tokens_per_second_per_gpu:3638.1460500054004 mfu:17.16175564959943 +Step 3771 | loss:1.2553026676177979 lr:9.746019957804926e-05 tokens_per_second_per_gpu:4100.507196328752 mfu:19.34279206375281 +Step 3772 | loss:1.3319323062896729 lr:9.745883495806731e-05 tokens_per_second_per_gpu:5335.043796968403 mfu:25.166311842633966 +Step 3773 | loss:1.3321049213409424 lr:9.74574699811414e-05 tokens_per_second_per_gpu:4600.211265515776 mfu:21.699981416413763 +Step 3774 | loss:1.3469719886779785 lr:9.745610464728182e-05 tokens_per_second_per_gpu:4058.6160272851057 mfu:19.14518427200348 +Step 3775 | loss:1.3070127964019775 lr:9.745473895649882e-05 tokens_per_second_per_gpu:5389.818911262708 mfu:25.42469540235839 +Step 3776 | loss:1.3817845582962036 lr:9.745337290880265e-05 tokens_per_second_per_gpu:3763.163672271244 mfu:17.751485104033275 +Step 3777 | loss:1.240740418434143 lr:9.745200650420363e-05 tokens_per_second_per_gpu:4326.327142569025 mfu:20.408023279023976 +Step 3778 | loss:1.3376976251602173 lr:9.745063974271201e-05 tokens_per_second_per_gpu:5348.6066671614335 mfu:25.230290215398767 +Step 3779 | loss:1.4595502614974976 lr:9.744927262433807e-05 tokens_per_second_per_gpu:4702.253178133128 mfu:22.18133052837538 +Step 3780 | loss:1.2335139513015747 lr:9.74479051490921e-05 tokens_per_second_per_gpu:4447.495237151679 mfu:20.979593854579065 +Step 3781 | loss:1.4114856719970703 lr:9.744653731698436e-05 tokens_per_second_per_gpu:5345.79925583001 mfu:25.217047177154676 +Step 3782 | loss:1.2395920753479004 lr:9.74451691280252e-05 tokens_per_second_per_gpu:5322.579827357801 mfu:25.107517171408258 +Step 3783 | loss:1.439650058746338 lr:9.744380058222483e-05 tokens_per_second_per_gpu:4218.8935359779 mfu:19.901240626670216 +Step 3784 | loss:1.2966597080230713 lr:9.74424316795936e-05 tokens_per_second_per_gpu:5349.420420545396 mfu:25.234128828951558 +Step 3785 | loss:1.113175868988037 lr:9.744106242014179e-05 tokens_per_second_per_gpu:5348.013635746233 mfu:25.22749278503167 +Step 3786 | loss:1.4236911535263062 lr:9.743969280387968e-05 tokens_per_second_per_gpu:5357.608674107192 mfu:25.27275421806267 +Step 3787 | loss:1.4189783334732056 lr:9.743832283081759e-05 tokens_per_second_per_gpu:5305.327403745804 mfu:25.026134545663467 +Step 3788 | loss:1.3002103567123413 lr:9.743695250096584e-05 tokens_per_second_per_gpu:3596.506552272992 mfu:16.965335034364763 +Step 3789 | loss:1.3096762895584106 lr:9.743558181433471e-05 tokens_per_second_per_gpu:4508.954450393247 mfu:21.2695074494626 +Step 3790 | loss:1.2152234315872192 lr:9.743421077093451e-05 tokens_per_second_per_gpu:4789.379841314993 mfu:22.592321863945845 +Step 3791 | loss:1.4185665845870972 lr:9.743283937077556e-05 tokens_per_second_per_gpu:5205.699667902123 mfu:24.556173517444286 +Step 3792 | loss:1.2826272249221802 lr:9.743146761386817e-05 tokens_per_second_per_gpu:4620.856376363768 mfu:21.79736792670287 +Step 3793 | loss:1.2981889247894287 lr:9.743009550022266e-05 tokens_per_second_per_gpu:4559.177848031654 mfu:21.506419785118215 +Step 3794 | loss:1.3005315065383911 lr:9.742872302984936e-05 tokens_per_second_per_gpu:5131.691721894213 mfu:24.207065409067322 +Step 3795 | loss:1.2835345268249512 lr:9.742735020275857e-05 tokens_per_second_per_gpu:5361.449851395589 mfu:25.29087370670446 +Step 3796 | loss:1.3465678691864014 lr:9.742597701896063e-05 tokens_per_second_per_gpu:3778.232012404178 mfu:17.822565035364256 +Step 3797 | loss:1.2761054039001465 lr:9.742460347846587e-05 tokens_per_second_per_gpu:4537.616130832861 mfu:21.404709486283206 +Step 3798 | loss:1.2302091121673584 lr:9.742322958128461e-05 tokens_per_second_per_gpu:4100.432910148758 mfu:19.34244164316725 +Step 3799 | loss:1.3020741939544678 lr:9.74218553274272e-05 tokens_per_second_per_gpu:5348.182842236292 mfu:25.228290960914883 +Step 3800 | loss:1.2384231090545654 lr:9.742048071690395e-05 tokens_per_second_per_gpu:4114.399100541927 mfu:19.408322546129618 +Step 3801 | loss:1.2029800415039062 lr:9.741910574972523e-05 tokens_per_second_per_gpu:5347.0381933023655 mfu:25.222891456596308 +Step 3802 | loss:1.3009922504425049 lr:9.741773042590134e-05 tokens_per_second_per_gpu:3891.161227429193 mfu:18.35527141034816 +Step 3803 | loss:1.3211253881454468 lr:9.741635474544267e-05 tokens_per_second_per_gpu:4540.42118175349 mfu:21.417941390066847 +Step 3804 | loss:1.3856091499328613 lr:9.741497870835954e-05 tokens_per_second_per_gpu:4221.4692750550485 mfu:19.913390827363532 +Step 3805 | loss:1.2144132852554321 lr:9.741360231466231e-05 tokens_per_second_per_gpu:3432.5784244091774 mfu:16.192058086208462 +Step 3806 | loss:1.564469575881958 lr:9.741222556436132e-05 tokens_per_second_per_gpu:5342.308102774942 mfu:25.20057881254179 +Step 3807 | loss:1.3742401599884033 lr:9.741084845746693e-05 tokens_per_second_per_gpu:4436.802164906868 mfu:20.929152808374035 +Step 3808 | loss:1.3910939693450928 lr:9.740947099398951e-05 tokens_per_second_per_gpu:3698.6514396657544 mfu:17.4471698959113 +Step 3809 | loss:1.1626428365707397 lr:9.74080931739394e-05 tokens_per_second_per_gpu:4844.597802723154 mfu:22.852794409063833 +Step 3810 | loss:1.4184759855270386 lr:9.740671499732698e-05 tokens_per_second_per_gpu:3795.379568734089 mfu:17.903452984247245 +Step 3811 | loss:1.2884927988052368 lr:9.740533646416258e-05 tokens_per_second_per_gpu:3565.2556960942834 mfu:16.817919413823915 +Step 3812 | loss:1.2651444673538208 lr:9.740395757445663e-05 tokens_per_second_per_gpu:4477.09070691829 mfu:21.119200734976 +Step 3813 | loss:1.1277731657028198 lr:9.740257832821946e-05 tokens_per_second_per_gpu:4572.000473128697 mfu:21.566906295466403 +Step 3814 | loss:1.2773820161819458 lr:9.740119872546145e-05 tokens_per_second_per_gpu:4774.926538835108 mfu:22.52414317016054 +Step 3815 | loss:1.4072664976119995 lr:9.739981876619295e-05 tokens_per_second_per_gpu:4800.907779345631 mfu:22.646701114505408 +Step 3816 | loss:1.3340251445770264 lr:9.73984384504244e-05 tokens_per_second_per_gpu:4500.611654981283 mfu:21.23015306007669 +Step 3817 | loss:1.2433794736862183 lr:9.739705777816612e-05 tokens_per_second_per_gpu:4968.036753416225 mfu:23.435077000339767 +Step 3818 | loss:1.2763426303863525 lr:9.739567674942854e-05 tokens_per_second_per_gpu:4539.4325357961625 mfu:21.413277778405742 +Step 3819 | loss:1.1701655387878418 lr:9.739429536422201e-05 tokens_per_second_per_gpu:5344.832878234731 mfu:25.212488609156726 +Step 3820 | loss:1.2901500463485718 lr:9.739291362255694e-05 tokens_per_second_per_gpu:4281.827199303416 mfu:20.198109454166865 +Step 3821 | loss:1.5069282054901123 lr:9.739153152444374e-05 tokens_per_second_per_gpu:2948.280856695137 mfu:13.907543829615777 +Step 3822 | loss:1.2641252279281616 lr:9.739014906989276e-05 tokens_per_second_per_gpu:5046.242203412455 mfu:23.803985451197658 +Step 3823 | loss:1.2053371667861938 lr:9.738876625891442e-05 tokens_per_second_per_gpu:4983.463445942582 mfu:23.507847341051747 +Step 3824 | loss:1.2739094495773315 lr:9.738738309151914e-05 tokens_per_second_per_gpu:5067.670740975815 mfu:23.905067518969542 +Step 3825 | loss:1.375390648841858 lr:9.73859995677173e-05 tokens_per_second_per_gpu:4493.101176323301 mfu:21.194724850827196 +Step 3826 | loss:1.283198356628418 lr:9.73846156875193e-05 tokens_per_second_per_gpu:5320.055909862533 mfu:25.095611421958345 +Step 3827 | loss:1.1873033046722412 lr:9.738323145093558e-05 tokens_per_second_per_gpu:5324.057150753475 mfu:25.114485957923012 +Step 3828 | loss:1.2869118452072144 lr:9.738184685797651e-05 tokens_per_second_per_gpu:4709.644646021833 mfu:22.216197343520864 +Step 3829 | loss:1.4499938488006592 lr:9.738046190865253e-05 tokens_per_second_per_gpu:4415.914263240713 mfu:20.830621012371267 +Step 3830 | loss:1.262621283531189 lr:9.737907660297405e-05 tokens_per_second_per_gpu:4059.639687959049 mfu:19.150013054056988 +Step 3831 | loss:1.2545984983444214 lr:9.73776909409515e-05 tokens_per_second_per_gpu:5080.531058195735 mfu:23.965731829492277 +Step 3832 | loss:1.4049988985061646 lr:9.737630492259528e-05 tokens_per_second_per_gpu:5332.722359241724 mfu:25.1553612247991 +Step 3833 | loss:1.3289660215377808 lr:9.737491854791582e-05 tokens_per_second_per_gpu:4148.648699657749 mfu:19.569883748743134 +Step 3834 | loss:1.307059645652771 lr:9.737353181692357e-05 tokens_per_second_per_gpu:5326.637400790391 mfu:25.126657437583198 +Step 3835 | loss:1.2477073669433594 lr:9.737214472962892e-05 tokens_per_second_per_gpu:4094.286508188081 mfu:19.31344801643442 +Step 3836 | loss:1.2229176759719849 lr:9.737075728604234e-05 tokens_per_second_per_gpu:4379.920242034477 mfu:20.660831073128175 +Step 3837 | loss:1.4305365085601807 lr:9.736936948617425e-05 tokens_per_second_per_gpu:4967.597243079633 mfu:23.433003755093864 +Step 3838 | loss:1.2077499628067017 lr:9.736798133003509e-05 tokens_per_second_per_gpu:5287.218039406338 mfu:24.94070958429804 +Step 3839 | loss:1.4348254203796387 lr:9.736659281763529e-05 tokens_per_second_per_gpu:5335.207224508694 mfu:25.16708275822513 +Step 3840 | loss:1.250447392463684 lr:9.736520394898529e-05 tokens_per_second_per_gpu:3947.5682614894226 mfu:18.621353014042672 +Step 3841 | loss:1.2261223793029785 lr:9.736381472409557e-05 tokens_per_second_per_gpu:4833.0778646116905 mfu:22.798452895487944 +Step 3842 | loss:1.1449223756790161 lr:9.736242514297655e-05 tokens_per_second_per_gpu:5020.233827602535 mfu:23.681299505015772 +Step 3843 | loss:1.3256804943084717 lr:9.736103520563869e-05 tokens_per_second_per_gpu:5358.936779548954 mfu:25.279019118029616 +Step 3844 | loss:1.2139853239059448 lr:9.735964491209243e-05 tokens_per_second_per_gpu:5338.014304570974 mfu:25.18032423381615 +Step 3845 | loss:1.2922704219818115 lr:9.735825426234824e-05 tokens_per_second_per_gpu:4421.2473271059125 mfu:20.855777984537937 +Step 3846 | loss:1.4347906112670898 lr:9.735686325641658e-05 tokens_per_second_per_gpu:5365.552258488722 mfu:25.3102254609056 +Step 3847 | loss:1.329646110534668 lr:9.735547189430791e-05 tokens_per_second_per_gpu:4039.311505540772 mfu:19.05412154924441 +Step 3848 | loss:1.4675054550170898 lr:9.735408017603269e-05 tokens_per_second_per_gpu:4315.296560991355 mfu:20.355990143710823 +Step 3849 | loss:1.214461088180542 lr:9.735268810160139e-05 tokens_per_second_per_gpu:4217.8449330361855 mfu:19.896294187684042 +Step 3850 | loss:1.1403194665908813 lr:9.735129567102447e-05 tokens_per_second_per_gpu:4428.823406814346 mfu:20.8915156451351 +Step 3851 | loss:1.224865436553955 lr:9.734990288431243e-05 tokens_per_second_per_gpu:5326.398550938927 mfu:25.125530742081853 +Step 3852 | loss:1.2163606882095337 lr:9.734850974147573e-05 tokens_per_second_per_gpu:4208.2574046350555 mfu:19.851068180414885 +Step 3853 | loss:1.386314034461975 lr:9.734711624252484e-05 tokens_per_second_per_gpu:3567.0867928151292 mfu:16.826557009473465 +Step 3854 | loss:1.2857403755187988 lr:9.734572238747025e-05 tokens_per_second_per_gpu:4585.455429702492 mfu:21.630375621277306 +Step 3855 | loss:1.400144338607788 lr:9.734432817632244e-05 tokens_per_second_per_gpu:5337.4330815151225 mfu:25.177582505494822 +Step 3856 | loss:1.292756199836731 lr:9.73429336090919e-05 tokens_per_second_per_gpu:5323.218003290679 mfu:25.110527556167686 +Step 3857 | loss:1.1719954013824463 lr:9.73415386857891e-05 tokens_per_second_per_gpu:3959.3059944466463 mfu:18.676721903065648 +Step 3858 | loss:1.4654393196105957 lr:9.734014340642457e-05 tokens_per_second_per_gpu:5344.065719619088 mfu:25.20886979107598 +Step 3859 | loss:1.2719271183013916 lr:9.733874777100875e-05 tokens_per_second_per_gpu:3750.704485283354 mfu:17.692712993254897 +Step 3860 | loss:1.3452210426330566 lr:9.733735177955219e-05 tokens_per_second_per_gpu:4377.718206019788 mfu:20.6504436935408 +Step 3861 | loss:1.2783246040344238 lr:9.733595543206538e-05 tokens_per_second_per_gpu:5308.143945978033 mfu:25.039420655924552 +Step 3862 | loss:1.3149151802062988 lr:9.733455872855878e-05 tokens_per_second_per_gpu:4202.64703717608 mfu:19.824603119883502 +Step 3863 | loss:1.4348692893981934 lr:9.733316166904294e-05 tokens_per_second_per_gpu:3380.177417183424 mfu:15.944873594590831 +Step 3864 | loss:1.4552944898605347 lr:9.733176425352835e-05 tokens_per_second_per_gpu:3606.1974827866893 mfu:17.011048807041117 +Step 3865 | loss:1.4656975269317627 lr:9.733036648202552e-05 tokens_per_second_per_gpu:5408.200099821738 mfu:25.51140260494549 +Step 3866 | loss:1.3635708093643188 lr:9.732896835454495e-05 tokens_per_second_per_gpu:4900.236152255162 mfu:23.11524999668762 +Step 3867 | loss:1.329546332359314 lr:9.732756987109718e-05 tokens_per_second_per_gpu:5314.3300512431 mfu:25.06860156991826 +Step 3868 | loss:1.3661978244781494 lr:9.732617103169272e-05 tokens_per_second_per_gpu:5331.128877733886 mfu:25.14784450065065 +Step 3869 | loss:1.401581883430481 lr:9.73247718363421e-05 tokens_per_second_per_gpu:5267.131267192871 mfu:24.84595685261038 +Step 3870 | loss:1.3465467691421509 lr:9.732337228505581e-05 tokens_per_second_per_gpu:5355.078075016647 mfu:25.26081695038784 +Step 3871 | loss:1.2010581493377686 lr:9.732197237784442e-05 tokens_per_second_per_gpu:4230.939074409007 mfu:19.958061486630527 +Step 3872 | loss:1.5088142156600952 lr:9.732057211471842e-05 tokens_per_second_per_gpu:5315.3543704722915 mfu:25.07343345848924 +Step 3873 | loss:1.269681453704834 lr:9.731917149568837e-05 tokens_per_second_per_gpu:4427.545524775907 mfu:20.885487657530632 +Step 3874 | loss:1.4368265867233276 lr:9.731777052076479e-05 tokens_per_second_per_gpu:4026.0883811365834 mfu:18.991745815332187 +Step 3875 | loss:1.2099475860595703 lr:9.731636918995821e-05 tokens_per_second_per_gpu:5330.628738648797 mfu:25.145485259253757 +Step 3876 | loss:1.175854206085205 lr:9.73149675032792e-05 tokens_per_second_per_gpu:5303.079399841491 mfu:25.015530327697807 +Step 3877 | loss:1.3126353025436401 lr:9.731356546073826e-05 tokens_per_second_per_gpu:4703.261874176282 mfu:22.18608871971223 +Step 3878 | loss:1.3124275207519531 lr:9.731216306234597e-05 tokens_per_second_per_gpu:4259.84679842906 mfu:20.094424152999427 +Step 3879 | loss:1.2958465814590454 lr:9.731076030811286e-05 tokens_per_second_per_gpu:3318.122883544257 mfu:15.652151771819684 +Step 3880 | loss:1.2001395225524902 lr:9.73093571980495e-05 tokens_per_second_per_gpu:5345.019776384081 mfu:25.21337023961532 +Step 3881 | loss:1.3448821306228638 lr:9.730795373216641e-05 tokens_per_second_per_gpu:5343.077844987528 mfu:25.20420981789012 +Step 3882 | loss:1.3056602478027344 lr:9.730654991047416e-05 tokens_per_second_per_gpu:3792.0184100836 mfu:17.887597825420087 +Step 3883 | loss:1.2577506303787231 lr:9.730514573298333e-05 tokens_per_second_per_gpu:3320.1759820541415 mfu:15.661836587785535 +Step 3884 | loss:1.3564348220825195 lr:9.730374119970445e-05 tokens_per_second_per_gpu:3713.3817453521215 mfu:17.516655261083205 +Step 3885 | loss:1.4289623498916626 lr:9.73023363106481e-05 tokens_per_second_per_gpu:4775.330173485136 mfu:22.526047183671892 +Step 3886 | loss:1.3054683208465576 lr:9.730093106582486e-05 tokens_per_second_per_gpu:5346.151619932846 mfu:25.218709338747267 +Step 3887 | loss:1.346134901046753 lr:9.729952546524525e-05 tokens_per_second_per_gpu:4636.607967677565 mfu:21.871670870428655 +Step 3888 | loss:1.4869657754898071 lr:9.72981195089199e-05 tokens_per_second_per_gpu:4030.860280036654 mfu:19.014255676613725 +Step 3889 | loss:1.2668287754058838 lr:9.729671319685933e-05 tokens_per_second_per_gpu:5316.265092868989 mfu:25.077729491419728 +Step 3890 | loss:1.3838077783584595 lr:9.729530652907417e-05 tokens_per_second_per_gpu:3942.4213615087297 mfu:18.597074208682056 +Step 3891 | loss:1.1666032075881958 lr:9.729389950557496e-05 tokens_per_second_per_gpu:4059.199516609766 mfu:19.147936690701442 +Step 3892 | loss:1.381232500076294 lr:9.72924921263723e-05 tokens_per_second_per_gpu:5402.363916607156 mfu:25.483872332966776 +Step 3893 | loss:1.305765986442566 lr:9.729108439147679e-05 tokens_per_second_per_gpu:4721.443256151575 mfu:22.271853400553326 +Step 3894 | loss:1.3104673624038696 lr:9.728967630089897e-05 tokens_per_second_per_gpu:3898.145117174641 mfu:18.3882156098519 +Step 3895 | loss:1.2358404397964478 lr:9.728826785464948e-05 tokens_per_second_per_gpu:2654.8927056284947 mfu:12.52358186385412 +Step 3896 | loss:1.4222772121429443 lr:9.728685905273889e-05 tokens_per_second_per_gpu:3135.468042968274 mfu:14.790537724693426 +Step 3897 | loss:1.3850104808807373 lr:9.72854498951778e-05 tokens_per_second_per_gpu:4515.7109725983055 mfu:21.301379117485425 +Step 3898 | loss:1.4547486305236816 lr:9.72840403819768e-05 tokens_per_second_per_gpu:5352.411772628232 mfu:25.24823954710359 +Step 3899 | loss:1.3320566415786743 lr:9.72826305131465e-05 tokens_per_second_per_gpu:4550.630319551022 mfu:21.466099634916333 +Step 3900 | loss:1.3668593168258667 lr:9.72812202886975e-05 tokens_per_second_per_gpu:4564.245978922565 mfu:21.530327023242936 +Step 3901 | loss:1.1524431705474854 lr:9.727980970864043e-05 tokens_per_second_per_gpu:5359.476814386104 mfu:25.281566554495896 +Step 3902 | loss:1.397202968597412 lr:9.727839877298586e-05 tokens_per_second_per_gpu:5359.986179235332 mfu:25.283969315396313 +Step 3903 | loss:1.35244882106781 lr:9.727698748174441e-05 tokens_per_second_per_gpu:3247.3277088751815 mfu:15.318198854003318 +Step 3904 | loss:1.469222068786621 lr:9.727557583492672e-05 tokens_per_second_per_gpu:4832.134247656766 mfu:22.79400169331391 +Step 3905 | loss:1.3510299921035767 lr:9.72741638325434e-05 tokens_per_second_per_gpu:4366.106857614657 mfu:20.595670982013473 +Step 3906 | loss:1.3563469648361206 lr:9.727275147460505e-05 tokens_per_second_per_gpu:5003.742728455675 mfu:23.603508176668388 +Step 3907 | loss:1.5446527004241943 lr:9.727133876112231e-05 tokens_per_second_per_gpu:5363.269772913148 mfu:25.29945858701342 +Step 3908 | loss:1.3950932025909424 lr:9.726992569210581e-05 tokens_per_second_per_gpu:4667.639949102863 mfu:22.01805402141378 +Step 3909 | loss:1.3842717409133911 lr:9.726851226756614e-05 tokens_per_second_per_gpu:5376.643206340055 mfu:25.36254335423117 +Step 3910 | loss:1.3585978746414185 lr:9.726709848751396e-05 tokens_per_second_per_gpu:4687.453902327842 mfu:22.111519819385023 +Step 3911 | loss:1.2879694700241089 lr:9.726568435195992e-05 tokens_per_second_per_gpu:4173.307909169358 mfu:19.68620544730438 +Step 3912 | loss:1.3832982778549194 lr:9.726426986091463e-05 tokens_per_second_per_gpu:4035.9019756839825 mfu:19.038038215184155 +Step 3913 | loss:1.2092642784118652 lr:9.726285501438872e-05 tokens_per_second_per_gpu:4376.26496072877 mfu:20.643588487553853 +Step 3914 | loss:1.2043417692184448 lr:9.726143981239286e-05 tokens_per_second_per_gpu:4408.916019902374 mfu:20.79760910452088 +Step 3915 | loss:1.3738402128219604 lr:9.726002425493769e-05 tokens_per_second_per_gpu:4828.097755983592 mfu:22.77496087339505 +Step 3916 | loss:1.3722857236862183 lr:9.725860834203383e-05 tokens_per_second_per_gpu:3340.7968840443154 mfu:15.759108900761861 +Step 3917 | loss:1.3433393239974976 lr:9.725719207369196e-05 tokens_per_second_per_gpu:4020.0853020180048 mfu:18.96342826689894 +Step 3918 | loss:1.2627440690994263 lr:9.725577544992272e-05 tokens_per_second_per_gpu:5387.092782910985 mfu:25.411835789797852 +Step 3919 | loss:1.4052139520645142 lr:9.725435847073675e-05 tokens_per_second_per_gpu:5040.001116700252 mfu:23.774545180337004 +Step 3920 | loss:1.3716814517974854 lr:9.725294113614472e-05 tokens_per_second_per_gpu:5350.409350090608 mfu:25.238793778345105 +Step 3921 | loss:1.3364092111587524 lr:9.72515234461573e-05 tokens_per_second_per_gpu:4265.170825759123 mfu:20.119538498288062 +Step 3922 | loss:1.3612384796142578 lr:9.725010540078514e-05 tokens_per_second_per_gpu:4167.895765786752 mfu:19.660675443561747 +Step 3923 | loss:1.4874231815338135 lr:9.72486870000389e-05 tokens_per_second_per_gpu:4484.101037238526 mfu:21.15226965918224 +Step 3924 | loss:1.3422014713287354 lr:9.724726824392927e-05 tokens_per_second_per_gpu:5375.964220620401 mfu:25.35934046274465 +Step 3925 | loss:1.3767342567443848 lr:9.72458491324669e-05 tokens_per_second_per_gpu:4990.950103295142 mfu:23.54316318113923 +Step 3926 | loss:1.329185128211975 lr:9.724442966566247e-05 tokens_per_second_per_gpu:4708.475996973121 mfu:22.210684626566007 +Step 3927 | loss:1.5080381631851196 lr:9.724300984352666e-05 tokens_per_second_per_gpu:5160.538105461953 mfu:24.3431387220569 +Step 3928 | loss:1.2605206966400146 lr:9.724158966607014e-05 tokens_per_second_per_gpu:4221.205503448452 mfu:19.91214657169114 +Step 3929 | loss:1.3241887092590332 lr:9.724016913330359e-05 tokens_per_second_per_gpu:5390.320312047932 mfu:25.427060595410087 +Step 3930 | loss:1.323577880859375 lr:9.723874824523771e-05 tokens_per_second_per_gpu:3507.344316544951 mfu:16.5447415557895 +Step 3931 | loss:1.5249241590499878 lr:9.723732700188318e-05 tokens_per_second_per_gpu:4231.264910571793 mfu:19.95959851140377 +Step 3932 | loss:1.3092095851898193 lr:9.723590540325068e-05 tokens_per_second_per_gpu:4500.8969372976335 mfu:21.2314987854371 +Step 3933 | loss:1.3262604475021362 lr:9.723448344935092e-05 tokens_per_second_per_gpu:4195.89234202599 mfu:19.792740070390696 +Step 3934 | loss:1.2716401815414429 lr:9.723306114019455e-05 tokens_per_second_per_gpu:3374.0085605239115 mfu:15.915774045212432 +Step 3935 | loss:1.2308135032653809 lr:9.723163847579233e-05 tokens_per_second_per_gpu:4361.406754454533 mfu:20.573499793486878 +Step 3936 | loss:1.2177306413650513 lr:9.723021545615491e-05 tokens_per_second_per_gpu:5350.273845927396 mfu:25.238154582087837 +Step 3937 | loss:1.3083003759384155 lr:9.722879208129304e-05 tokens_per_second_per_gpu:5363.038765430752 mfu:25.29836888530502 +Step 3938 | loss:1.3071914911270142 lr:9.722736835121738e-05 tokens_per_second_per_gpu:5287.023679630493 mfu:24.93979275607418 +Step 3939 | loss:1.3363364934921265 lr:9.722594426593865e-05 tokens_per_second_per_gpu:4946.566650095182 mfu:23.333798859797042 +Step 3940 | loss:1.3591455221176147 lr:9.722451982546757e-05 tokens_per_second_per_gpu:3592.29169237353 mfu:16.94545282108975 +Step 3941 | loss:1.3120386600494385 lr:9.722309502981486e-05 tokens_per_second_per_gpu:5128.628051804646 mfu:24.19261355453882 +Step 3942 | loss:1.2951018810272217 lr:9.722166987899122e-05 tokens_per_second_per_gpu:3982.704577065153 mfu:18.787097009487002 +Step 3943 | loss:1.3516145944595337 lr:9.722024437300738e-05 tokens_per_second_per_gpu:4844.50173355038 mfu:22.852341234384763 +Step 3944 | loss:1.2868069410324097 lr:9.721881851187406e-05 tokens_per_second_per_gpu:3800.319644776554 mfu:17.92675616580344 +Step 3945 | loss:1.4570125341415405 lr:9.721739229560197e-05 tokens_per_second_per_gpu:4772.741931095662 mfu:22.513838002721055 +Step 3946 | loss:1.3428750038146973 lr:9.721596572420186e-05 tokens_per_second_per_gpu:4955.570121860234 mfu:23.376269772263484 +Step 3947 | loss:1.4078716039657593 lr:9.721453879768443e-05 tokens_per_second_per_gpu:5194.239137717042 mfu:24.502112241193178 +Step 3948 | loss:1.4785488843917847 lr:9.721311151606046e-05 tokens_per_second_per_gpu:5349.348987240911 mfu:25.233791865866525 +Step 3949 | loss:1.177736520767212 lr:9.721168387934062e-05 tokens_per_second_per_gpu:5363.121887115301 mfu:25.29876098447306 +Step 3950 | loss:1.2147561311721802 lr:9.72102558875357e-05 tokens_per_second_per_gpu:5333.83342336351 mfu:25.16060230382865 +Step 3951 | loss:1.3026480674743652 lr:9.720882754065641e-05 tokens_per_second_per_gpu:5312.276654973121 mfu:25.058915349367243 +Step 3952 | loss:1.3568865060806274 lr:9.72073988387135e-05 tokens_per_second_per_gpu:5319.023012178013 mfu:25.09073906734995 +Step 3953 | loss:1.2517386674880981 lr:9.720596978171774e-05 tokens_per_second_per_gpu:3829.4774059574866 mfu:18.064298300120942 +Step 3954 | loss:1.4170893430709839 lr:9.720454036967986e-05 tokens_per_second_per_gpu:4331.578107113917 mfu:20.432792974689164 +Step 3955 | loss:1.5739779472351074 lr:9.720311060261058e-05 tokens_per_second_per_gpu:4436.722719409037 mfu:20.92877805040597 +Step 3956 | loss:1.2678439617156982 lr:9.72016804805207e-05 tokens_per_second_per_gpu:3749.9408906124586 mfu:17.6891109869096 +Step 3957 | loss:1.346739649772644 lr:9.720025000342095e-05 tokens_per_second_per_gpu:3383.2127187676465 mfu:15.95919163003946 +Step 3958 | loss:1.263018250465393 lr:9.719881917132211e-05 tokens_per_second_per_gpu:5342.173839228239 mfu:25.199945468484263 +Step 3959 | loss:1.4614288806915283 lr:9.719738798423492e-05 tokens_per_second_per_gpu:5350.853020720786 mfu:25.24088664840567 +Step 3960 | loss:1.4467947483062744 lr:9.719595644217015e-05 tokens_per_second_per_gpu:3972.7611505404957 mfu:18.74019217004669 +Step 3961 | loss:1.3313570022583008 lr:9.719452454513857e-05 tokens_per_second_per_gpu:5322.631089017932 mfu:25.107758981405233 +Step 3962 | loss:1.2813516855239868 lr:9.719309229315096e-05 tokens_per_second_per_gpu:5180.316429082683 mfu:24.43643644910586 +Step 3963 | loss:1.345284104347229 lr:9.719165968621807e-05 tokens_per_second_per_gpu:5318.7812715615355 mfu:25.08959873561682 +Step 3964 | loss:1.3220354318618774 lr:9.71902267243507e-05 tokens_per_second_per_gpu:4541.304739084632 mfu:21.422109280748273 +Step 3965 | loss:1.1978102922439575 lr:9.71887934075596e-05 tokens_per_second_per_gpu:4756.392081849501 mfu:22.43671297425498 +Step 3966 | loss:1.394883155822754 lr:9.718735973585555e-05 tokens_per_second_per_gpu:4540.5439491676125 mfu:21.418520504905544 +Step 3967 | loss:1.327789545059204 lr:9.718592570924938e-05 tokens_per_second_per_gpu:4960.625181049169 mfu:23.40011534893972 +Step 3968 | loss:1.4003386497497559 lr:9.718449132775182e-05 tokens_per_second_per_gpu:3390.88588767333 mfu:15.995387276945186 +Step 3969 | loss:1.3898199796676636 lr:9.718305659137367e-05 tokens_per_second_per_gpu:4242.1556181304495 mfu:20.01097183710427 +Step 3970 | loss:1.285556435585022 lr:9.718162150012575e-05 tokens_per_second_per_gpu:5377.57259144405 mfu:25.36692742233616 +Step 3971 | loss:1.22832453250885 lr:9.718018605401885e-05 tokens_per_second_per_gpu:4912.849685989097 mfu:23.17475018740136 +Step 3972 | loss:1.2696322202682495 lr:9.71787502530637e-05 tokens_per_second_per_gpu:4088.1074165903706 mfu:19.284300187106286 +Step 3973 | loss:1.1900453567504883 lr:9.71773140972712e-05 tokens_per_second_per_gpu:5405.12317154447 mfu:25.496888209283522 +Step 3974 | loss:1.3406792879104614 lr:9.717587758665207e-05 tokens_per_second_per_gpu:5355.5237528572925 mfu:25.26291928880265 +Step 3975 | loss:1.3918466567993164 lr:9.717444072121716e-05 tokens_per_second_per_gpu:4271.374320488433 mfu:20.148801441351562 +Step 3976 | loss:1.2592945098876953 lr:9.717300350097724e-05 tokens_per_second_per_gpu:5352.241082057464 mfu:25.247434370558466 +Step 3977 | loss:1.491381049156189 lr:9.717156592594316e-05 tokens_per_second_per_gpu:3697.890927532576 mfu:17.44358243042235 +Step 3978 | loss:1.2591047286987305 lr:9.717012799612572e-05 tokens_per_second_per_gpu:4842.57866858412 mfu:22.843269808830204 +Step 3979 | loss:1.2772759199142456 lr:9.716868971153572e-05 tokens_per_second_per_gpu:5045.153654826265 mfu:23.79885058179106 +Step 3980 | loss:1.3719991445541382 lr:9.716725107218398e-05 tokens_per_second_per_gpu:3182.2948584095334 mfu:15.011427802608411 +Step 3981 | loss:1.3692774772644043 lr:9.716581207808133e-05 tokens_per_second_per_gpu:4398.670654672382 mfu:20.74927996869145 +Step 3982 | loss:1.2792500257492065 lr:9.71643727292386e-05 tokens_per_second_per_gpu:5356.634005279089 mfu:25.268156538904858 +Step 3983 | loss:1.248940110206604 lr:9.716293302566659e-05 tokens_per_second_per_gpu:4604.82891278513 mfu:21.72176364643517 +Step 3984 | loss:1.3901981115341187 lr:9.716149296737615e-05 tokens_per_second_per_gpu:5335.225948253999 mfu:25.167171081326337 +Step 3985 | loss:1.341172218322754 lr:9.71600525543781e-05 tokens_per_second_per_gpu:5046.550456763244 mfu:23.80543953484678 +Step 3986 | loss:1.3011342287063599 lr:9.715861178668328e-05 tokens_per_second_per_gpu:5354.644147399757 mfu:25.258770039783283 +Step 3987 | loss:1.2783300876617432 lr:9.715717066430252e-05 tokens_per_second_per_gpu:4129.278718474114 mfu:19.478512242640353 +Step 3988 | loss:1.5078399181365967 lr:9.715572918724666e-05 tokens_per_second_per_gpu:4748.69818853995 mfu:22.400419566800267 +Step 3989 | loss:1.5020428895950317 lr:9.715428735552655e-05 tokens_per_second_per_gpu:3585.327843247835 mfu:16.912603156608665 +Step 3990 | loss:1.2741880416870117 lr:9.715284516915303e-05 tokens_per_second_per_gpu:4981.253849185025 mfu:23.497424296150264 +Step 3991 | loss:1.2506906986236572 lr:9.715140262813693e-05 tokens_per_second_per_gpu:4258.182181324291 mfu:20.086571870103267 +Step 3992 | loss:1.3088651895523071 lr:9.714995973248912e-05 tokens_per_second_per_gpu:5371.925706493004 mfu:25.340290102564158 +Step 3993 | loss:1.328740119934082 lr:9.714851648222046e-05 tokens_per_second_per_gpu:5009.825487424096 mfu:23.63220158854812 +Step 3994 | loss:1.253397822380066 lr:9.714707287734179e-05 tokens_per_second_per_gpu:4764.508080030763 mfu:22.474997522408074 +Step 3995 | loss:1.2697030305862427 lr:9.714562891786396e-05 tokens_per_second_per_gpu:5311.717352279283 mfu:25.05627702313353 +Step 3996 | loss:1.2037421464920044 lr:9.714418460379784e-05 tokens_per_second_per_gpu:5021.978141910075 mfu:23.68952773321515 +Step 3997 | loss:1.2799303531646729 lr:9.714273993515429e-05 tokens_per_second_per_gpu:5316.4341102047265 mfu:25.07852677502973 +Step 3998 | loss:1.326456904411316 lr:9.714129491194417e-05 tokens_per_second_per_gpu:4273.838205179542 mfu:20.160424005821596 +Step 3999 | loss:1.298627257347107 lr:9.713984953417838e-05 tokens_per_second_per_gpu:4650.419035846879 mfu:21.936820035394526 +Step 4000 | loss:1.4735724925994873 lr:9.713840380186773e-05 tokens_per_second_per_gpu:4412.312340942768 mfu:20.813630130340545 +Step 4001 | loss:1.3781660795211792 lr:9.713695771502317e-05 tokens_per_second_per_gpu:3715.1399278890844 mfu:17.52494890808693 +Step 4002 | loss:1.274774193763733 lr:9.713551127365552e-05 tokens_per_second_per_gpu:4965.529590503558 mfu:23.423250285114612 +Step 4003 | loss:1.3807517290115356 lr:9.713406447777567e-05 tokens_per_second_per_gpu:5351.507392921136 mfu:25.24397343372213 +Step 4004 | loss:1.334496021270752 lr:9.713261732739452e-05 tokens_per_second_per_gpu:3524.117513649784 mfu:16.623863588335524 +Step 4005 | loss:1.305191993713379 lr:9.713116982252292e-05 tokens_per_second_per_gpu:5342.2632815360375 mfu:25.20036738311058 +Step 4006 | loss:1.328021764755249 lr:9.712972196317178e-05 tokens_per_second_per_gpu:5327.357729691075 mfu:25.13005535190851 +Step 4007 | loss:1.2244353294372559 lr:9.7128273749352e-05 tokens_per_second_per_gpu:4561.974711435543 mfu:21.519613066988462 +Step 4008 | loss:1.4898645877838135 lr:9.712682518107445e-05 tokens_per_second_per_gpu:4687.648026928188 mfu:22.11243553824593 +Step 4009 | loss:1.322688102722168 lr:9.712537625835005e-05 tokens_per_second_per_gpu:3782.140456212638 mfu:17.84100183165804 +Step 4010 | loss:1.332214593887329 lr:9.712392698118967e-05 tokens_per_second_per_gpu:3695.0509486830188 mfu:17.43018576563863 +Step 4011 | loss:1.4259270429611206 lr:9.712247734960423e-05 tokens_per_second_per_gpu:5252.5400480273665 mfu:24.777127582287996 +Step 4012 | loss:1.4247709512710571 lr:9.712102736360462e-05 tokens_per_second_per_gpu:5306.304407017437 mfu:25.030743236789707 +Step 4013 | loss:1.2084821462631226 lr:9.711957702320175e-05 tokens_per_second_per_gpu:4787.9263165342245 mfu:22.58546533955806 +Step 4014 | loss:1.372344970703125 lr:9.711812632840652e-05 tokens_per_second_per_gpu:4841.518834671051 mfu:22.838270391435945 +Step 4015 | loss:1.3642239570617676 lr:9.711667527922987e-05 tokens_per_second_per_gpu:5248.531741593385 mfu:24.75821971695131 +Step 4016 | loss:1.4253441095352173 lr:9.711522387568268e-05 tokens_per_second_per_gpu:3996.137253607295 mfu:18.850461236587233 +Step 4017 | loss:1.2469450235366821 lr:9.711377211777587e-05 tokens_per_second_per_gpu:4579.736774600593 mfu:21.603399749458475 +Step 4018 | loss:1.3062901496887207 lr:9.711232000552037e-05 tokens_per_second_per_gpu:4410.396712790251 mfu:20.804593785505254 +Step 4019 | loss:1.34251070022583 lr:9.711086753892711e-05 tokens_per_second_per_gpu:5306.83137226741 mfu:25.033229021783374 +Step 4020 | loss:1.2357450723648071 lr:9.710941471800701e-05 tokens_per_second_per_gpu:5028.435706273787 mfu:23.719989166092873 +Step 4021 | loss:1.3855139017105103 lr:9.710796154277097e-05 tokens_per_second_per_gpu:4145.988621143205 mfu:19.557335704533838 +Step 4022 | loss:1.2079299688339233 lr:9.710650801322995e-05 tokens_per_second_per_gpu:5342.670141180418 mfu:25.202286609470537 +Step 4023 | loss:1.3591848611831665 lr:9.71050541293949e-05 tokens_per_second_per_gpu:3725.973853573729 mfu:17.576054383999725 +Step 4024 | loss:1.4186506271362305 lr:9.710359989127669e-05 tokens_per_second_per_gpu:5026.607205695039 mfu:23.71136381688855 +Step 4025 | loss:1.1356052160263062 lr:9.71021452988863e-05 tokens_per_second_per_gpu:4770.175967907757 mfu:22.501733916565243 +Step 4026 | loss:1.2782447338104248 lr:9.710069035223468e-05 tokens_per_second_per_gpu:5379.4343129863055 mfu:25.37570948049343 +Step 4027 | loss:1.3567540645599365 lr:9.709923505133275e-05 tokens_per_second_per_gpu:3763.7837164140883 mfu:17.754409958045514 +Step 4028 | loss:1.3048169612884521 lr:9.709777939619146e-05 tokens_per_second_per_gpu:4510.506309777264 mfu:21.276827834951522 +Step 4029 | loss:1.423851728439331 lr:9.709632338682178e-05 tokens_per_second_per_gpu:5401.035937031453 mfu:25.477608026731502 +Step 4030 | loss:1.3996331691741943 lr:9.709486702323461e-05 tokens_per_second_per_gpu:5052.786466621219 mfu:23.834855857319837 +Step 4031 | loss:1.2829124927520752 lr:9.709341030544096e-05 tokens_per_second_per_gpu:5374.703255837798 mfu:25.353392276722335 +Step 4032 | loss:1.3615161180496216 lr:9.709195323345175e-05 tokens_per_second_per_gpu:5134.828026794852 mfu:24.221859894392434 +Step 4033 | loss:1.3138623237609863 lr:9.709049580727796e-05 tokens_per_second_per_gpu:4816.5344065083 mfu:22.720414581009237 +Step 4034 | loss:1.3494771718978882 lr:9.708903802693054e-05 tokens_per_second_per_gpu:5293.833682769856 mfu:24.971916702788906 +Step 4035 | loss:1.4450483322143555 lr:9.708757989242046e-05 tokens_per_second_per_gpu:5347.397609270563 mfu:25.22458688304098 +Step 4036 | loss:1.438804030418396 lr:9.708612140375867e-05 tokens_per_second_per_gpu:5296.465713756383 mfu:24.984332442023145 +Step 4037 | loss:1.301110863685608 lr:9.708466256095617e-05 tokens_per_second_per_gpu:3153.2904659292763 mfu:14.87460913461939 +Step 4038 | loss:1.3559861183166504 lr:9.70832033640239e-05 tokens_per_second_per_gpu:5378.568639698474 mfu:25.371625951895638 +Step 4039 | loss:1.3461180925369263 lr:9.708174381297284e-05 tokens_per_second_per_gpu:5221.832409390155 mfu:24.632274411572862 +Step 4040 | loss:1.3359951972961426 lr:9.7080283907814e-05 tokens_per_second_per_gpu:3752.1080977549987 mfu:17.69933407276463 +Step 4041 | loss:1.238715410232544 lr:9.707882364855834e-05 tokens_per_second_per_gpu:4609.886900930766 mfu:21.745623039499904 +Step 4042 | loss:1.2817058563232422 lr:9.707736303521682e-05 tokens_per_second_per_gpu:5341.792872620196 mfu:25.19814838399452 +Step 4043 | loss:1.3162394762039185 lr:9.707590206780045e-05 tokens_per_second_per_gpu:4043.777700675546 mfu:19.075189353706605 +Step 4044 | loss:1.219988465309143 lr:9.707444074632022e-05 tokens_per_second_per_gpu:3695.761441717431 mfu:17.433537282504563 +Step 4045 | loss:1.1706902980804443 lr:9.707297907078711e-05 tokens_per_second_per_gpu:4772.265907589369 mfu:22.511592518623885 +Step 4046 | loss:1.330221176147461 lr:9.707151704121212e-05 tokens_per_second_per_gpu:5332.720016921156 mfu:25.15535017567334 +Step 4047 | loss:1.4086203575134277 lr:9.707005465760624e-05 tokens_per_second_per_gpu:4610.482686550233 mfu:21.748433461918435 +Step 4048 | loss:1.4451130628585815 lr:9.706859191998048e-05 tokens_per_second_per_gpu:4824.41945335673 mfu:22.757609692320997 +Step 4049 | loss:1.27457857131958 lr:9.706712882834583e-05 tokens_per_second_per_gpu:5329.782501782465 mfu:25.141493415572377 +Step 4050 | loss:1.3352041244506836 lr:9.706566538271332e-05 tokens_per_second_per_gpu:3957.1456536220994 mfu:18.666531206804105 +Step 4051 | loss:1.1596510410308838 lr:9.70642015830939e-05 tokens_per_second_per_gpu:4620.9523049624595 mfu:21.797820438270005 +Step 4052 | loss:1.219677448272705 lr:9.706273742949862e-05 tokens_per_second_per_gpu:3929.0499796259883 mfu:18.533999118948167 +Step 4053 | loss:1.2138304710388184 lr:9.706127292193849e-05 tokens_per_second_per_gpu:4967.04149059268 mfu:23.430382175791753 +Step 4054 | loss:1.2577519416809082 lr:9.705980806042454e-05 tokens_per_second_per_gpu:4778.342061174377 mfu:22.54025477178307 +Step 4055 | loss:1.3386200666427612 lr:9.705834284496775e-05 tokens_per_second_per_gpu:5132.515994310583 mfu:24.21095364269072 +Step 4056 | loss:1.3977608680725098 lr:9.705687727557917e-05 tokens_per_second_per_gpu:4936.359948276287 mfu:23.285652105874412 +Step 4057 | loss:1.5188844203948975 lr:9.70554113522698e-05 tokens_per_second_per_gpu:4519.499126118323 mfu:21.319248483964984 +Step 4058 | loss:1.3051508665084839 lr:9.70539450750507e-05 tokens_per_second_per_gpu:3612.962882150502 mfu:17.04296235013638 +Step 4059 | loss:1.4305802583694458 lr:9.705247844393285e-05 tokens_per_second_per_gpu:5354.553059566689 mfu:25.258340362933517 +Step 4060 | loss:1.310656189918518 lr:9.705101145892732e-05 tokens_per_second_per_gpu:5345.0684797802805 mfu:25.21359998184485 +Step 4061 | loss:1.294806718826294 lr:9.704954412004511e-05 tokens_per_second_per_gpu:5351.825508580316 mfu:25.24547403956261 +Step 4062 | loss:1.311553955078125 lr:9.70480764272973e-05 tokens_per_second_per_gpu:4539.072970425782 mfu:21.41158164720548 +Step 4063 | loss:1.369882583618164 lr:9.704660838069488e-05 tokens_per_second_per_gpu:5366.560995427918 mfu:25.314983845156057 +Step 4064 | loss:1.1505683660507202 lr:9.704513998024894e-05 tokens_per_second_per_gpu:4895.8656647023845 mfu:23.0946336612191 +Step 4065 | loss:1.4136818647384644 lr:9.704367122597046e-05 tokens_per_second_per_gpu:5162.412508970528 mfu:24.351980603213065 +Step 4066 | loss:1.0955047607421875 lr:9.704220211787056e-05 tokens_per_second_per_gpu:5318.386449384146 mfu:25.087736292042468 +Step 4067 | loss:1.3530718088150024 lr:9.704073265596025e-05 tokens_per_second_per_gpu:4872.476033755922 mfu:22.98430078136209 +Step 4068 | loss:1.4584416151046753 lr:9.703926284025057e-05 tokens_per_second_per_gpu:4230.415127816544 mfu:19.955589941159374 +Step 4069 | loss:1.3227335214614868 lr:9.703779267075262e-05 tokens_per_second_per_gpu:5336.581974727049 mfu:25.173567689561686 +Step 4070 | loss:1.2817124128341675 lr:9.703632214747742e-05 tokens_per_second_per_gpu:5304.133657629033 mfu:25.02050344155657 +Step 4071 | loss:1.3800041675567627 lr:9.703485127043604e-05 tokens_per_second_per_gpu:4204.417448850289 mfu:19.832954453814125 +Step 4072 | loss:1.3051677942276 lr:9.703338003963953e-05 tokens_per_second_per_gpu:5324.930622472634 mfu:25.11860627305193 +Step 4073 | loss:1.2042828798294067 lr:9.703190845509897e-05 tokens_per_second_per_gpu:3593.7515700555145 mfu:16.952339313196486 +Step 4074 | loss:1.1715589761734009 lr:9.703043651682542e-05 tokens_per_second_per_gpu:5365.543910849003 mfu:25.31018608366479 +Step 4075 | loss:1.3388618230819702 lr:9.702896422482998e-05 tokens_per_second_per_gpu:3811.4285606153803 mfu:17.979158817191667 +Step 4076 | loss:1.4035755395889282 lr:9.702749157912367e-05 tokens_per_second_per_gpu:5043.134934586891 mfu:23.789327933992745 +Step 4077 | loss:1.2733559608459473 lr:9.702601857971761e-05 tokens_per_second_per_gpu:5334.87842997047 mfu:25.1655317782902 +Step 4078 | loss:1.4707731008529663 lr:9.702454522662288e-05 tokens_per_second_per_gpu:5320.99305630937 mfu:25.10003210164175 +Step 4079 | loss:1.3641374111175537 lr:9.702307151985052e-05 tokens_per_second_per_gpu:5147.468162431935 mfu:24.281485570822074 +Step 4080 | loss:1.3356709480285645 lr:9.702159745941163e-05 tokens_per_second_per_gpu:3717.1195454387357 mfu:17.53428710182622 +Step 4081 | loss:1.3039904832839966 lr:9.702012304531733e-05 tokens_per_second_per_gpu:3843.5989072630655 mfu:18.130911831155927 +Step 4082 | loss:1.2493081092834473 lr:9.701864827757868e-05 tokens_per_second_per_gpu:4284.440529910313 mfu:20.210436980519777 +Step 4083 | loss:1.3389688730239868 lr:9.701717315620678e-05 tokens_per_second_per_gpu:5332.859893761291 mfu:25.156009998593706 +Step 4084 | loss:1.2711046934127808 lr:9.701569768121271e-05 tokens_per_second_per_gpu:5336.479119848591 mfu:25.17308250555115 +Step 4085 | loss:1.2247364521026611 lr:9.701422185260758e-05 tokens_per_second_per_gpu:5214.569673332118 mfu:24.598014846436556 +Step 4086 | loss:1.5143545866012573 lr:9.70127456704025e-05 tokens_per_second_per_gpu:3398.904318376442 mfu:16.033211582657113 +Step 4087 | loss:1.3860596418380737 lr:9.701126913460856e-05 tokens_per_second_per_gpu:5132.691728096646 mfu:24.211782608942855 +Step 4088 | loss:1.1969621181488037 lr:9.700979224523685e-05 tokens_per_second_per_gpu:5329.106956487702 mfu:25.138306753157064 +Step 4089 | loss:1.239471197128296 lr:9.70083150022985e-05 tokens_per_second_per_gpu:4067.7399182213903 mfu:19.188223222246243 +Step 4090 | loss:1.3627876043319702 lr:9.700683740580462e-05 tokens_per_second_per_gpu:3905.325148483959 mfu:18.422085042577617 +Step 4091 | loss:1.3821182250976562 lr:9.700535945576632e-05 tokens_per_second_per_gpu:3796.6756111200716 mfu:17.909566637309556 +Step 4092 | loss:1.379326343536377 lr:9.70038811521947e-05 tokens_per_second_per_gpu:4417.2745666458895 mfu:20.83703779562518 +Step 4093 | loss:1.2769306898117065 lr:9.70024024951009e-05 tokens_per_second_per_gpu:5237.655072541654 mfu:24.70691261328319 +Step 4094 | loss:1.277827501296997 lr:9.700092348449603e-05 tokens_per_second_per_gpu:4407.597478313137 mfu:20.791389318878057 +Step 4095 | loss:1.3682467937469482 lr:9.699944412039123e-05 tokens_per_second_per_gpu:3996.396420671052 mfu:18.851683771845437 +Step 4096 | loss:1.453372597694397 lr:9.69979644027976e-05 tokens_per_second_per_gpu:5382.516320707493 mfu:25.390247836759013 +Step 4097 | loss:1.3057266473770142 lr:9.699648433172629e-05 tokens_per_second_per_gpu:4636.544400177097 mfu:21.871371011683213 +Step 4098 | loss:1.383386492729187 lr:9.699500390718842e-05 tokens_per_second_per_gpu:3810.8703923080266 mfu:17.97652584205255 +Step 4099 | loss:1.217043161392212 lr:9.699352312919514e-05 tokens_per_second_per_gpu:5140.231023996466 mfu:24.247346754038826 +Step 4100 | loss:1.2750166654586792 lr:9.699204199775755e-05 tokens_per_second_per_gpu:5317.211016115134 mfu:25.08219157276294 +Step 4101 | loss:1.43381929397583 lr:9.699056051288684e-05 tokens_per_second_per_gpu:5323.896738322434 mfu:25.11372926511661 +Step 4102 | loss:1.3931381702423096 lr:9.698907867459413e-05 tokens_per_second_per_gpu:3756.6643626136756 mfu:17.720826751481955 +Step 4103 | loss:1.4315682649612427 lr:9.698759648289056e-05 tokens_per_second_per_gpu:5371.382126812034 mfu:25.337725944464765 +Step 4104 | loss:1.292296290397644 lr:9.698611393778727e-05 tokens_per_second_per_gpu:4833.8115631216015 mfu:22.801913876541064 +Step 4105 | loss:1.504551649093628 lr:9.698463103929542e-05 tokens_per_second_per_gpu:4897.635034672842 mfu:23.102980081255392 +Step 4106 | loss:1.152547001838684 lr:9.698314778742618e-05 tokens_per_second_per_gpu:4818.942932821022 mfu:22.73177601056331 +Step 4107 | loss:1.317635178565979 lr:9.698166418219068e-05 tokens_per_second_per_gpu:5337.436618213301 mfu:25.17759918870355 +Step 4108 | loss:1.3663538694381714 lr:9.698018022360009e-05 tokens_per_second_per_gpu:5349.501783475598 mfu:25.23451263177527 +Step 4109 | loss:1.1715619564056396 lr:9.697869591166558e-05 tokens_per_second_per_gpu:4629.5193373735665 mfu:21.838232591839652 +Step 4110 | loss:1.1606429815292358 lr:9.69772112463983e-05 tokens_per_second_per_gpu:5326.786558495606 mfu:25.127361039927628 +Step 4111 | loss:1.3981236219406128 lr:9.69757262278094e-05 tokens_per_second_per_gpu:3635.9872192845996 mfu:17.151572076754913 +Step 4112 | loss:1.29633629322052 lr:9.69742408559101e-05 tokens_per_second_per_gpu:4711.326130245704 mfu:22.224129191495656 +Step 4113 | loss:1.5340075492858887 lr:9.69727551307115e-05 tokens_per_second_per_gpu:4493.479108334966 mfu:21.196507620608074 +Step 4114 | loss:1.37973153591156 lr:9.697126905222484e-05 tokens_per_second_per_gpu:5380.446517774634 mfu:25.380484223179923 +Step 4115 | loss:1.476658582687378 lr:9.696978262046128e-05 tokens_per_second_per_gpu:4097.2350200590145 mfu:19.327356649997615 +Step 4116 | loss:1.2721827030181885 lr:9.696829583543198e-05 tokens_per_second_per_gpu:4514.151960812893 mfu:21.294024992898795 +Step 4117 | loss:1.234620213508606 lr:9.696680869714814e-05 tokens_per_second_per_gpu:4905.607511749101 mfu:23.140587615868963 +Step 4118 | loss:1.3346731662750244 lr:9.696532120562094e-05 tokens_per_second_per_gpu:3250.791187235104 mfu:15.334536672357418 +Step 4119 | loss:1.482377290725708 lr:9.696383336086156e-05 tokens_per_second_per_gpu:4574.807586686886 mfu:21.580147929063337 +Step 4120 | loss:1.363243579864502 lr:9.69623451628812e-05 tokens_per_second_per_gpu:5330.446423053194 mfu:25.14462524548305 +Step 4121 | loss:1.2479534149169922 lr:9.696085661169104e-05 tokens_per_second_per_gpu:4010.5971383298042 mfu:18.918670979934042 +Step 4122 | loss:1.3033052682876587 lr:9.69593677073023e-05 tokens_per_second_per_gpu:5295.042285350853 mfu:24.977617887372332 +Step 4123 | loss:1.271545171737671 lr:9.695787844972614e-05 tokens_per_second_per_gpu:4232.582779231817 mfu:19.965815122724685 +Step 4124 | loss:1.1964696645736694 lr:9.695638883897382e-05 tokens_per_second_per_gpu:4162.930193871845 mfu:19.637251993625203 +Step 4125 | loss:1.490460753440857 lr:9.69548988750565e-05 tokens_per_second_per_gpu:3360.122700589353 mfu:15.850272074729114 +Step 4126 | loss:1.1438102722167969 lr:9.695340855798535e-05 tokens_per_second_per_gpu:5345.377699655832 mfu:25.21505862475618 +Step 4127 | loss:1.317541241645813 lr:9.695191788777168e-05 tokens_per_second_per_gpu:4210.183582945362 mfu:19.86015429214451 +Step 4128 | loss:1.2837618589401245 lr:9.695042686442662e-05 tokens_per_second_per_gpu:4551.782962004578 mfu:21.471536846030837 +Step 4129 | loss:1.320813775062561 lr:9.69489354879614e-05 tokens_per_second_per_gpu:5270.39962460666 mfu:24.8613742521705 +Step 4130 | loss:1.488376498222351 lr:9.694744375838725e-05 tokens_per_second_per_gpu:4420.090363484985 mfu:20.85032038973967 +Step 4131 | loss:1.288504719734192 lr:9.694595167571539e-05 tokens_per_second_per_gpu:5347.209059268824 mfu:25.223697460512827 +Step 4132 | loss:1.3196511268615723 lr:9.694445923995704e-05 tokens_per_second_per_gpu:4991.923313407383 mfu:23.54775397928544 +Step 4133 | loss:1.114861011505127 lr:9.694296645112342e-05 tokens_per_second_per_gpu:5351.550650300419 mfu:25.24417748616068 +Step 4134 | loss:1.2659504413604736 lr:9.694147330922576e-05 tokens_per_second_per_gpu:3484.9291836416114 mfu:16.439005549468412 +Step 4135 | loss:1.1171685457229614 lr:9.693997981427529e-05 tokens_per_second_per_gpu:4290.946562526685 mfu:20.24112704641483 +Step 4136 | loss:1.2455850839614868 lr:9.693848596628323e-05 tokens_per_second_per_gpu:5372.713357290188 mfu:25.344005585761558 +Step 4137 | loss:1.3236671686172485 lr:9.693699176526084e-05 tokens_per_second_per_gpu:4395.281383904828 mfu:20.733292200211675 +Step 4138 | loss:1.3709075450897217 lr:9.693549721121935e-05 tokens_per_second_per_gpu:4777.032076100449 mfu:22.53407534868285 +Step 4139 | loss:1.2042183876037598 lr:9.693400230416997e-05 tokens_per_second_per_gpu:5318.0909178173315 mfu:25.08634221921892 +Step 4140 | loss:1.4402977228164673 lr:9.6932507044124e-05 tokens_per_second_per_gpu:5024.826998503782 mfu:23.702966275832726 +Step 4141 | loss:1.2623498439788818 lr:9.693101143109264e-05 tokens_per_second_per_gpu:3737.806177225393 mfu:17.631869473467216 +Step 4142 | loss:1.1698338985443115 lr:9.692951546508718e-05 tokens_per_second_per_gpu:5354.032826841737 mfu:25.255886336409063 +Step 4143 | loss:1.465228796005249 lr:9.69280191461188e-05 tokens_per_second_per_gpu:3667.568077573115 mfu:17.30054437355188 +Step 4144 | loss:1.304093599319458 lr:9.692652247419885e-05 tokens_per_second_per_gpu:4837.410971524933 mfu:22.818892900101982 +Step 4145 | loss:1.4623576402664185 lr:9.69250254493385e-05 tokens_per_second_per_gpu:3223.276774546818 mfu:15.20474649326389 +Step 4146 | loss:1.2475610971450806 lr:9.692352807154905e-05 tokens_per_second_per_gpu:3774.428835753114 mfu:17.80462480221234 +Step 4147 | loss:1.1615253686904907 lr:9.692203034084177e-05 tokens_per_second_per_gpu:5112.436309174317 mfu:24.116234341955394 +Step 4148 | loss:1.36104416847229 lr:9.69205322572279e-05 tokens_per_second_per_gpu:3450.4263399780702 mfu:16.276249749115333 +Step 4149 | loss:1.2334580421447754 lr:9.691903382071874e-05 tokens_per_second_per_gpu:5063.944826209798 mfu:23.887491743311628 +Step 4150 | loss:1.592707633972168 lr:9.691753503132553e-05 tokens_per_second_per_gpu:3534.0055745662335 mfu:16.670507258755574 +Step 4151 | loss:1.2164669036865234 lr:9.691603588905955e-05 tokens_per_second_per_gpu:3812.654532970029 mfu:17.984941937961 +Step 4152 | loss:1.272354245185852 lr:9.691453639393208e-05 tokens_per_second_per_gpu:5282.059653971433 mfu:24.916376600090288 +Step 4153 | loss:1.2301948070526123 lr:9.691303654595439e-05 tokens_per_second_per_gpu:4067.2823191973353 mfu:19.186064649575332 +Step 4154 | loss:1.354644775390625 lr:9.691153634513777e-05 tokens_per_second_per_gpu:4639.729674733448 mfu:21.8863964952294 +Step 4155 | loss:1.180277705192566 lr:9.691003579149349e-05 tokens_per_second_per_gpu:5355.781660856483 mfu:25.26413588483847 +Step 4156 | loss:1.5152759552001953 lr:9.690853488503286e-05 tokens_per_second_per_gpu:5353.680785153512 mfu:25.2542256957012 +Step 4157 | loss:1.2652486562728882 lr:9.690703362576714e-05 tokens_per_second_per_gpu:4302.240574722834 mfu:20.29440282890125 +Step 4158 | loss:1.454717993736267 lr:9.690553201370765e-05 tokens_per_second_per_gpu:4598.320663141207 mfu:21.69106312244289 +Step 4159 | loss:1.5404249429702759 lr:9.690403004886565e-05 tokens_per_second_per_gpu:5364.663316135432 mfu:25.306032168155923 +Step 4160 | loss:1.2771483659744263 lr:9.690252773125248e-05 tokens_per_second_per_gpu:4854.0004759774265 mfu:22.89714842306572 +Step 4161 | loss:1.2830405235290527 lr:9.69010250608794e-05 tokens_per_second_per_gpu:3767.692863407654 mfu:17.772850071383413 +Step 4162 | loss:1.2603222131729126 lr:9.689952203775772e-05 tokens_per_second_per_gpu:5355.2451906234055 mfu:25.261605263217785 +Step 4163 | loss:1.392055869102478 lr:9.689801866189876e-05 tokens_per_second_per_gpu:5375.133991348473 mfu:25.355424129617244 +Step 4164 | loss:1.2584236860275269 lr:9.68965149333138e-05 tokens_per_second_per_gpu:4761.055698300986 mfu:22.458712048751647 +Step 4165 | loss:1.1758785247802734 lr:9.689501085201419e-05 tokens_per_second_per_gpu:4773.865054127125 mfu:22.519135965685017 +Step 4166 | loss:1.4076075553894043 lr:9.68935064180112e-05 tokens_per_second_per_gpu:5333.189598769466 mfu:25.157565273370746 +Step 4167 | loss:1.3264204263687134 lr:9.689200163131618e-05 tokens_per_second_per_gpu:5334.81050842966 mfu:25.16521138079359 +Step 4168 | loss:1.296924352645874 lr:9.689049649194043e-05 tokens_per_second_per_gpu:3892.5881430150116 mfu:18.36200241462337 +Step 4169 | loss:1.362431287765503 lr:9.688899099989526e-05 tokens_per_second_per_gpu:4127.972165155259 mfu:19.472349007714328 +Step 4170 | loss:1.4556808471679688 lr:9.688748515519202e-05 tokens_per_second_per_gpu:4221.383309500523 mfu:19.912985313175597 +Step 4171 | loss:1.455062747001648 lr:9.688597895784201e-05 tokens_per_second_per_gpu:5255.592068127182 mfu:24.79152448182715 +Step 4172 | loss:1.3376511335372925 lr:9.688447240785658e-05 tokens_per_second_per_gpu:4719.794171896102 mfu:22.26407438875747 +Step 4173 | loss:1.2650423049926758 lr:9.688296550524705e-05 tokens_per_second_per_gpu:4583.613420546171 mfu:21.62168654980777 +Step 4174 | loss:1.3350789546966553 lr:9.688145825002474e-05 tokens_per_second_per_gpu:5185.527643216313 mfu:24.46101863529901 +Step 4175 | loss:1.3261058330535889 lr:9.687995064220102e-05 tokens_per_second_per_gpu:3710.1847732976175 mfu:17.501574598442428 +Step 4176 | loss:1.3577347993850708 lr:9.687844268178721e-05 tokens_per_second_per_gpu:4455.080025827098 mfu:21.01537259685941 +Step 4177 | loss:1.4615904092788696 lr:9.687693436879464e-05 tokens_per_second_per_gpu:3392.1664732612935 mfu:16.00142801765404 +Step 4178 | loss:1.2869985103607178 lr:9.687542570323466e-05 tokens_per_second_per_gpu:3491.96392381176 mfu:16.472189619101442 +Step 4179 | loss:1.228592038154602 lr:9.687391668511862e-05 tokens_per_second_per_gpu:4905.547548967871 mfu:23.140304761199197 +Step 4180 | loss:1.4325648546218872 lr:9.687240731445788e-05 tokens_per_second_per_gpu:5378.058042119171 mfu:25.36921737599713 +Step 4181 | loss:1.3034684658050537 lr:9.687089759126379e-05 tokens_per_second_per_gpu:4732.933038459795 mfu:22.32605266406021 +Step 4182 | loss:1.206697702407837 lr:9.686938751554768e-05 tokens_per_second_per_gpu:5348.1697131607125 mfu:25.22822902882532 +Step 4183 | loss:1.423640489578247 lr:9.686787708732095e-05 tokens_per_second_per_gpu:4929.975882979465 mfu:23.25553738063544 +Step 4184 | loss:1.3291457891464233 lr:9.686636630659491e-05 tokens_per_second_per_gpu:3886.2392338886666 mfu:18.332053527038802 +Step 4185 | loss:1.2578479051589966 lr:9.686485517338097e-05 tokens_per_second_per_gpu:4557.915007247029 mfu:21.50046274967428 +Step 4186 | loss:1.2594292163848877 lr:9.686334368769047e-05 tokens_per_second_per_gpu:4147.396742131277 mfu:19.563978051484785 +Step 4187 | loss:1.4606585502624512 lr:9.686183184953478e-05 tokens_per_second_per_gpu:5371.796667019484 mfu:25.339681401351157 +Step 4188 | loss:1.0803223848342896 lr:9.686031965892527e-05 tokens_per_second_per_gpu:5322.4040713888735 mfu:25.106688100515765 +Step 4189 | loss:1.3310962915420532 lr:9.685880711587332e-05 tokens_per_second_per_gpu:5331.1963411167 mfu:25.148162736937824 +Step 4190 | loss:1.2303671836853027 lr:9.685729422039029e-05 tokens_per_second_per_gpu:4336.502505599563 mfu:20.456022202534978 +Step 4191 | loss:1.3959381580352783 lr:9.68557809724876e-05 tokens_per_second_per_gpu:4388.166702719742 mfu:20.699731035171844 +Step 4192 | loss:1.4680408239364624 lr:9.685426737217658e-05 tokens_per_second_per_gpu:4759.800095045902 mfu:22.452789153969416 +Step 4193 | loss:1.410420298576355 lr:9.685275341946863e-05 tokens_per_second_per_gpu:5056.566591966547 mfu:23.852687353529642 +Step 4194 | loss:1.2720005512237549 lr:9.685123911437516e-05 tokens_per_second_per_gpu:5082.8689758497285 mfu:23.976760185956053 +Step 4195 | loss:1.3142037391662598 lr:9.684972445690754e-05 tokens_per_second_per_gpu:5357.647143367061 mfu:25.272935684125006 +Step 4196 | loss:1.3054866790771484 lr:9.684820944707716e-05 tokens_per_second_per_gpu:4368.864892638915 mfu:20.608681103792197 +Step 4197 | loss:1.374648094177246 lr:9.684669408489542e-05 tokens_per_second_per_gpu:3746.8606051391903 mfu:17.6745807547813 +Step 4198 | loss:1.2827421426773071 lr:9.684517837037371e-05 tokens_per_second_per_gpu:4255.871316153651 mfu:20.075671124819102 +Step 4199 | loss:1.2524110078811646 lr:9.684366230352345e-05 tokens_per_second_per_gpu:5317.013747993767 mfu:25.08126102537704 +Step 4200 | loss:1.2432478666305542 lr:9.6842145884356e-05 tokens_per_second_per_gpu:3382.640664273063 mfu:15.95649315138595 +Step 4201 | loss:1.2906948328018188 lr:9.684062911288283e-05 tokens_per_second_per_gpu:4524.925564400112 mfu:21.344845919186028 +Step 4202 | loss:1.5842863321304321 lr:9.68391119891153e-05 tokens_per_second_per_gpu:4962.455751315825 mfu:23.40875046121503 +Step 4203 | loss:1.3821215629577637 lr:9.683759451306482e-05 tokens_per_second_per_gpu:4352.795286667446 mfu:20.532878030667277 +Step 4204 | loss:1.2544766664505005 lr:9.683607668474282e-05 tokens_per_second_per_gpu:4329.834852538353 mfu:20.424569745426325 +Step 4205 | loss:1.3199043273925781 lr:9.683455850416073e-05 tokens_per_second_per_gpu:4665.403139523646 mfu:22.007502608989544 +Step 4206 | loss:1.3738716840744019 lr:9.683303997132992e-05 tokens_per_second_per_gpu:3812.6790007887935 mfu:17.98505735683675 +Step 4207 | loss:1.3909181356430054 lr:9.683152108626186e-05 tokens_per_second_per_gpu:4495.545049200573 mfu:21.206253016157 +Step 4208 | loss:1.5032278299331665 lr:9.683000184896794e-05 tokens_per_second_per_gpu:3444.679291216775 mfu:16.249139939560674 +Step 4209 | loss:1.516015887260437 lr:9.682848225945961e-05 tokens_per_second_per_gpu:5081.552159648564 mfu:23.970548539263145 +Step 4210 | loss:1.4283852577209473 lr:9.682696231774829e-05 tokens_per_second_per_gpu:3892.699542424312 mfu:18.362527904643738 +Step 4211 | loss:1.2952368259429932 lr:9.68254420238454e-05 tokens_per_second_per_gpu:5355.806327464616 mfu:25.26425224143759 +Step 4212 | loss:1.3631436824798584 lr:9.68239213777624e-05 tokens_per_second_per_gpu:5296.345035585291 mfu:24.983763182500287 +Step 4213 | loss:1.2597488164901733 lr:9.68224003795107e-05 tokens_per_second_per_gpu:4138.03258800881 mfu:19.51980573879967 +Step 4214 | loss:1.3777390718460083 lr:9.682087902910176e-05 tokens_per_second_per_gpu:5330.584833556487 mfu:25.14527815144235 +Step 4215 | loss:1.288090705871582 lr:9.681935732654701e-05 tokens_per_second_per_gpu:3713.238349798771 mfu:17.515978839792243 +Step 4216 | loss:1.4325636625289917 lr:9.681783527185791e-05 tokens_per_second_per_gpu:4186.983931055238 mfu:19.75071757591002 +Step 4217 | loss:1.3949658870697021 lr:9.681631286504589e-05 tokens_per_second_per_gpu:5349.911458005922 mfu:25.236445136432494 +Step 4218 | loss:1.2542505264282227 lr:9.68147901061224e-05 tokens_per_second_per_gpu:4518.358817878844 mfu:21.313869455442713 +Step 4219 | loss:1.276410460472107 lr:9.681326699509892e-05 tokens_per_second_per_gpu:5355.727976549782 mfu:25.26388264680428 +Step 4220 | loss:1.5219626426696777 lr:9.681174353198687e-05 tokens_per_second_per_gpu:4274.2964170800105 mfu:20.162585469535124 +Step 4221 | loss:1.3617802858352661 lr:9.681021971679772e-05 tokens_per_second_per_gpu:4484.384451600204 mfu:21.153606573081753 +Step 4222 | loss:1.1004552841186523 lr:9.680869554954296e-05 tokens_per_second_per_gpu:4955.28976820931 mfu:23.374947296258803 +Step 4223 | loss:1.3746877908706665 lr:9.680717103023398e-05 tokens_per_second_per_gpu:3807.164171965265 mfu:17.959042968349344 +Step 4224 | loss:1.3539836406707764 lr:9.680564615888233e-05 tokens_per_second_per_gpu:4344.985592075896 mfu:20.49603836880764 +Step 4225 | loss:1.3719717264175415 lr:9.680412093549944e-05 tokens_per_second_per_gpu:5357.608553366439 mfu:25.27275364850794 +Step 4226 | loss:1.2474921941757202 lr:9.680259536009677e-05 tokens_per_second_per_gpu:4404.635272649851 mfu:20.77741608935994 +Step 4227 | loss:1.2709380388259888 lr:9.680106943268583e-05 tokens_per_second_per_gpu:4878.631989089394 mfu:23.01333947298436 +Step 4228 | loss:1.4398566484451294 lr:9.679954315327806e-05 tokens_per_second_per_gpu:4915.163944074589 mfu:23.185666937643777 +Step 4229 | loss:1.3813453912734985 lr:9.679801652188497e-05 tokens_per_second_per_gpu:3899.0845058080345 mfu:18.392646866311846 +Step 4230 | loss:1.5854406356811523 lr:9.6796489538518e-05 tokens_per_second_per_gpu:4964.018325401568 mfu:23.41612138978011 +Step 4231 | loss:1.4126536846160889 lr:9.679496220318866e-05 tokens_per_second_per_gpu:4594.745461413589 mfu:21.6741982859014 +Step 4232 | loss:1.4537413120269775 lr:9.679343451590846e-05 tokens_per_second_per_gpu:5288.630555416938 mfu:24.94737266332012 +Step 4233 | loss:1.2650052309036255 lr:9.679190647668887e-05 tokens_per_second_per_gpu:5349.869217916978 mfu:25.236245882726042 +Step 4234 | loss:1.30319082736969 lr:9.679037808554136e-05 tokens_per_second_per_gpu:3963.7681833960614 mfu:18.69777080966798 +Step 4235 | loss:1.2514876127243042 lr:9.678884934247748e-05 tokens_per_second_per_gpu:4990.228808274903 mfu:23.539760709463142 +Step 4236 | loss:1.3158208131790161 lr:9.678732024750865e-05 tokens_per_second_per_gpu:4993.52740911424 mfu:23.555320772421634 +Step 4237 | loss:1.1690951585769653 lr:9.678579080064645e-05 tokens_per_second_per_gpu:5069.934811084451 mfu:23.91574752396217 +Step 4238 | loss:1.1529172658920288 lr:9.678426100190232e-05 tokens_per_second_per_gpu:4459.136378322189 mfu:21.03450710366385 +Step 4239 | loss:1.321101427078247 lr:9.678273085128782e-05 tokens_per_second_per_gpu:4501.605941842145 mfu:21.23484328084221 +Step 4240 | loss:1.2021030187606812 lr:9.678120034881441e-05 tokens_per_second_per_gpu:5365.6964111536 mfu:25.31090545362081 +Step 4241 | loss:1.2483280897140503 lr:9.677966949449363e-05 tokens_per_second_per_gpu:5045.276756249058 mfu:23.799431272205645 +Step 4242 | loss:1.3623729944229126 lr:9.677813828833699e-05 tokens_per_second_per_gpu:5348.790184869978 mfu:25.231155899741317 +Step 4243 | loss:1.254274845123291 lr:9.6776606730356e-05 tokens_per_second_per_gpu:4524.902536476362 mfu:21.344737292540724 +Step 4244 | loss:1.2520736455917358 lr:9.677507482056219e-05 tokens_per_second_per_gpu:4379.833011764927 mfu:20.660419593063686 +Step 4245 | loss:1.2956781387329102 lr:9.677354255896705e-05 tokens_per_second_per_gpu:4300.451044504021 mfu:20.28596130953377 +Step 4246 | loss:1.2870124578475952 lr:9.677200994558216e-05 tokens_per_second_per_gpu:4201.044170627666 mfu:19.817042124897405 +Step 4247 | loss:1.2957704067230225 lr:9.677047698041899e-05 tokens_per_second_per_gpu:4373.612052737147 mfu:20.631074267924586 +Step 4248 | loss:1.2743940353393555 lr:9.67689436634891e-05 tokens_per_second_per_gpu:4395.116196198195 mfu:20.732512981615564 +Step 4249 | loss:1.3543956279754639 lr:9.676740999480402e-05 tokens_per_second_per_gpu:4303.100126091699 mfu:20.29845747936277 +Step 4250 | loss:1.2383977174758911 lr:9.676587597437528e-05 tokens_per_second_per_gpu:4837.98109065004 mfu:22.821582249287463 +Step 4251 | loss:1.2632735967636108 lr:9.676434160221442e-05 tokens_per_second_per_gpu:3638.062841185847 mfu:17.16136313940071 +Step 4252 | loss:1.246924638748169 lr:9.6762806878333e-05 tokens_per_second_per_gpu:5101.553426305514 mfu:24.064897926651962 +Step 4253 | loss:1.1853901147842407 lr:9.676127180274251e-05 tokens_per_second_per_gpu:5325.004208089412 mfu:25.11895338896127 +Step 4254 | loss:1.3598345518112183 lr:9.675973637545455e-05 tokens_per_second_per_gpu:3928.3722820996136 mfu:18.530802304089132 +Step 4255 | loss:1.2016979455947876 lr:9.675820059648063e-05 tokens_per_second_per_gpu:5365.990252259592 mfu:25.312291552252233 +Step 4256 | loss:1.1892805099487305 lr:9.675666446583234e-05 tokens_per_second_per_gpu:4281.345843140249 mfu:20.195838815017375 +Step 4257 | loss:1.3693925142288208 lr:9.675512798352119e-05 tokens_per_second_per_gpu:3674.0525964256567 mfu:17.331132955352324 +Step 4258 | loss:1.2458209991455078 lr:9.675359114955875e-05 tokens_per_second_per_gpu:5321.300380448748 mfu:25.101481802042098 +Step 4259 | loss:1.2669904232025146 lr:9.67520539639566e-05 tokens_per_second_per_gpu:5333.11882624161 mfu:25.15723142728181 +Step 4260 | loss:1.2061101198196411 lr:9.675051642672629e-05 tokens_per_second_per_gpu:5152.181371770166 mfu:24.30371858342695 +Step 4261 | loss:1.2226436138153076 lr:9.674897853787936e-05 tokens_per_second_per_gpu:5024.3981667819335 mfu:23.70094340343453 +Step 4262 | loss:1.3342783451080322 lr:9.674744029742741e-05 tokens_per_second_per_gpu:3204.5221799436677 mfu:15.116277870657031 +Step 4263 | loss:1.2534185647964478 lr:9.674590170538202e-05 tokens_per_second_per_gpu:4839.899356822623 mfu:22.830631038113495 +Step 4264 | loss:1.2908662557601929 lr:9.674436276175471e-05 tokens_per_second_per_gpu:4504.636633431708 mfu:21.249139570159485 +Step 4265 | loss:1.1694209575653076 lr:9.674282346655709e-05 tokens_per_second_per_gpu:4016.6397938238006 mfu:18.94717521688256 +Step 4266 | loss:1.288796067237854 lr:9.674128381980072e-05 tokens_per_second_per_gpu:4586.792191694511 mfu:21.636681355668767 +Step 4267 | loss:1.470565915107727 lr:9.67397438214972e-05 tokens_per_second_per_gpu:3630.6297755993473 mfu:17.126300100817428 +Step 4268 | loss:1.293766975402832 lr:9.673820347165812e-05 tokens_per_second_per_gpu:5329.739411520477 mfu:25.14129015145462 +Step 4269 | loss:1.4234594106674194 lr:9.673666277029503e-05 tokens_per_second_per_gpu:3790.3241710698153 mfu:17.87960581092508 +Step 4270 | loss:1.2463610172271729 lr:9.673512171741952e-05 tokens_per_second_per_gpu:5252.1546640939405 mfu:24.775309660520744 +Step 4271 | loss:1.316145896911621 lr:9.673358031304322e-05 tokens_per_second_per_gpu:5356.30832795145 mfu:25.266620263384105 +Step 4272 | loss:1.4069164991378784 lr:9.67320385571777e-05 tokens_per_second_per_gpu:3392.41840231517 mfu:16.002616409984704 +Step 4273 | loss:1.2932766675949097 lr:9.673049644983453e-05 tokens_per_second_per_gpu:3641.851739298029 mfu:17.17923601824855 +Step 4274 | loss:1.3153228759765625 lr:9.672895399102535e-05 tokens_per_second_per_gpu:4822.428153339067 mfu:22.74821639038687 +Step 4275 | loss:1.3598421812057495 lr:9.672741118076175e-05 tokens_per_second_per_gpu:5342.298977208364 mfu:25.200535765687256 +Step 4276 | loss:1.2364075183868408 lr:9.672586801905532e-05 tokens_per_second_per_gpu:5334.238029501691 mfu:25.16251090001538 +Step 4277 | loss:1.1792510747909546 lr:9.672432450591768e-05 tokens_per_second_per_gpu:4408.761619902783 mfu:20.796880773379456 +Step 4278 | loss:1.184521198272705 lr:9.672278064136043e-05 tokens_per_second_per_gpu:5332.424204354305 mfu:25.153954777324678 +Step 4279 | loss:1.1494065523147583 lr:9.67212364253952e-05 tokens_per_second_per_gpu:5355.827824514106 mfu:25.264353646687812 +Step 4280 | loss:1.2931627035140991 lr:9.671969185803356e-05 tokens_per_second_per_gpu:5328.23238510984 mfu:25.134181250750427 +Step 4281 | loss:1.167805790901184 lr:9.671814693928717e-05 tokens_per_second_per_gpu:4331.704465745592 mfu:20.43338902991378 +Step 4282 | loss:1.5345927476882935 lr:9.671660166916763e-05 tokens_per_second_per_gpu:4138.006827252749 mfu:19.519684220918087 +Step 4283 | loss:1.4542227983474731 lr:9.671505604768658e-05 tokens_per_second_per_gpu:5151.786127577681 mfu:24.301854149135384 +Step 4284 | loss:1.4714926481246948 lr:9.671351007485562e-05 tokens_per_second_per_gpu:3217.104243188584 mfu:15.175629609704655 +Step 4285 | loss:1.2993714809417725 lr:9.671196375068641e-05 tokens_per_second_per_gpu:3861.2363728119344 mfu:18.21411076541187 +Step 4286 | loss:1.2781916856765747 lr:9.671041707519055e-05 tokens_per_second_per_gpu:5329.259325011071 mfu:25.13902550147855 +Step 4287 | loss:1.404160737991333 lr:9.670887004837968e-05 tokens_per_second_per_gpu:4241.683062839895 mfu:20.008742713172477 +Step 4288 | loss:1.466643214225769 lr:9.670732267026543e-05 tokens_per_second_per_gpu:3747.941072196509 mfu:17.679677502237357 +Step 4289 | loss:1.5212466716766357 lr:9.670577494085945e-05 tokens_per_second_per_gpu:3437.4415474916327 mfu:16.21499826752276 +Step 4290 | loss:1.3537437915802002 lr:9.670422686017338e-05 tokens_per_second_per_gpu:5337.574196818767 mfu:25.178248170458865 +Step 4291 | loss:1.2565370798110962 lr:9.670267842821886e-05 tokens_per_second_per_gpu:3634.133383454013 mfu:17.142827216845024 +Step 4292 | loss:1.316924810409546 lr:9.670112964500753e-05 tokens_per_second_per_gpu:4051.61496483379 mfu:19.112159065916195 +Step 4293 | loss:1.349094033241272 lr:9.669958051055106e-05 tokens_per_second_per_gpu:3662.012345119949 mfu:17.274337035664598 +Step 4294 | loss:1.4371812343597412 lr:9.669803102486108e-05 tokens_per_second_per_gpu:4055.298024298693 mfu:19.129532685806883 +Step 4295 | loss:1.2795262336730957 lr:9.669648118794925e-05 tokens_per_second_per_gpu:4877.541265942487 mfu:23.008194345803847 +Step 4296 | loss:1.2739180326461792 lr:9.669493099982723e-05 tokens_per_second_per_gpu:5176.8543367153925 mfu:24.420105168715487 +Step 4297 | loss:1.2444952726364136 lr:9.669338046050666e-05 tokens_per_second_per_gpu:5326.790442648426 mfu:25.12737936213918 +Step 4298 | loss:1.194441556930542 lr:9.669182956999923e-05 tokens_per_second_per_gpu:4967.106125037514 mfu:23.43068706749581 +Step 4299 | loss:1.340735673904419 lr:9.669027832831658e-05 tokens_per_second_per_gpu:4816.351268608013 mfu:22.719550688287036 +Step 4300 | loss:1.3653711080551147 lr:9.66887267354704e-05 tokens_per_second_per_gpu:3678.035770266265 mfu:17.349922265958007 +Step 4301 | loss:1.3617486953735352 lr:9.668717479147236e-05 tokens_per_second_per_gpu:5358.332921205479 mfu:25.276170615194584 +Step 4302 | loss:1.2658395767211914 lr:9.66856224963341e-05 tokens_per_second_per_gpu:4725.89843541284 mfu:22.292869241260675 +Step 4303 | loss:1.200963020324707 lr:9.668406985006732e-05 tokens_per_second_per_gpu:4900.236506641372 mfu:23.115251668387838 +Step 4304 | loss:1.398408055305481 lr:9.66825168526837e-05 tokens_per_second_per_gpu:3489.082946774233 mfu:16.458599558870056 +Step 4305 | loss:1.2542719841003418 lr:9.668096350419492e-05 tokens_per_second_per_gpu:4754.699520901969 mfu:22.428728875484797 +Step 4306 | loss:1.2683086395263672 lr:9.667940980461265e-05 tokens_per_second_per_gpu:4306.07622657523 mfu:20.312496253118688 +Step 4307 | loss:1.4560916423797607 lr:9.667785575394859e-05 tokens_per_second_per_gpu:4512.449609208896 mfu:21.285994709931604 +Step 4308 | loss:1.1726371049880981 lr:9.667630135221439e-05 tokens_per_second_per_gpu:4273.358537477824 mfu:20.158161331432844 +Step 4309 | loss:1.254008412361145 lr:9.66747465994218e-05 tokens_per_second_per_gpu:5321.585876179355 mfu:25.10282853411397 +Step 4310 | loss:1.2595711946487427 lr:9.667319149558248e-05 tokens_per_second_per_gpu:5001.999548936819 mfu:23.595285301460848 +Step 4311 | loss:1.1725642681121826 lr:9.667163604070814e-05 tokens_per_second_per_gpu:4899.3589910933015 mfu:23.111112277827996 +Step 4312 | loss:1.3821947574615479 lr:9.667008023481046e-05 tokens_per_second_per_gpu:4952.790900741193 mfu:23.36315971206094 +Step 4313 | loss:1.2517648935317993 lr:9.666852407790114e-05 tokens_per_second_per_gpu:4026.5097297893885 mfu:18.993733388816963 +Step 4314 | loss:1.387742280960083 lr:9.666696756999191e-05 tokens_per_second_per_gpu:3928.1345646409973 mfu:18.52968095027806 +Step 4315 | loss:1.2601723670959473 lr:9.666541071109446e-05 tokens_per_second_per_gpu:4188.894675221185 mfu:19.759730882147363 +Step 4316 | loss:1.3941634893417358 lr:9.66638535012205e-05 tokens_per_second_per_gpu:4882.570433300676 mfu:23.031917786296855 +Step 4317 | loss:1.351072072982788 lr:9.666229594038174e-05 tokens_per_second_per_gpu:4733.131707879271 mfu:22.32698982160837 +Step 4318 | loss:1.302720546722412 lr:9.66607380285899e-05 tokens_per_second_per_gpu:4518.62243599949 mfu:21.31511298709569 +Step 4319 | loss:1.2445776462554932 lr:9.665917976585671e-05 tokens_per_second_per_gpu:4918.117043821739 mfu:23.19959721300185 +Step 4320 | loss:1.2547180652618408 lr:9.665762115219386e-05 tokens_per_second_per_gpu:4627.7498753158225 mfu:21.829885737411914 +Step 4321 | loss:1.316580057144165 lr:9.66560621876131e-05 tokens_per_second_per_gpu:3988.024564508768 mfu:18.81219229794199 +Step 4322 | loss:1.2920581102371216 lr:9.665450287212613e-05 tokens_per_second_per_gpu:3372.7903485998268 mfu:15.910027531718137 +Step 4323 | loss:1.2735331058502197 lr:9.665294320574469e-05 tokens_per_second_per_gpu:4690.180037829505 mfu:22.124379465673044 +Step 4324 | loss:1.30197274684906 lr:9.665138318848052e-05 tokens_per_second_per_gpu:4385.092346252964 mfu:20.68522877117875 +Step 4325 | loss:1.2762004137039185 lr:9.664982282034534e-05 tokens_per_second_per_gpu:5149.837940517748 mfu:24.292664218378693 +Step 4326 | loss:1.3307266235351562 lr:9.664826210135088e-05 tokens_per_second_per_gpu:5338.053434027659 mfu:25.18050881413951 +Step 4327 | loss:1.1179468631744385 lr:9.664670103150891e-05 tokens_per_second_per_gpu:4013.8491067524074 mfu:18.934011069826486 +Step 4328 | loss:1.321427583694458 lr:9.664513961083112e-05 tokens_per_second_per_gpu:4949.72599417993 mfu:23.348702024884485 +Step 4329 | loss:1.3105823993682861 lr:9.664357783932931e-05 tokens_per_second_per_gpu:5310.238192276874 mfu:25.049299573031448 +Step 4330 | loss:1.1388107538223267 lr:9.664201571701519e-05 tokens_per_second_per_gpu:4552.641961647472 mfu:21.475588893906824 +Step 4331 | loss:1.2212365865707397 lr:9.664045324390052e-05 tokens_per_second_per_gpu:5320.839581356898 mfu:25.09930813410541 +Step 4332 | loss:1.3924648761749268 lr:9.663889041999706e-05 tokens_per_second_per_gpu:3335.3798382105874 mfu:15.733555771320841 +Step 4333 | loss:1.2898210287094116 lr:9.663732724531653e-05 tokens_per_second_per_gpu:5387.864626280716 mfu:25.415476706662933 +Step 4334 | loss:1.0294016599655151 lr:9.663576371987074e-05 tokens_per_second_per_gpu:3841.1091578788332 mfu:18.119167258515457 +Step 4335 | loss:1.2815560102462769 lr:9.663419984367139e-05 tokens_per_second_per_gpu:4202.730543877799 mfu:19.82499703524323 +Step 4336 | loss:1.3233259916305542 lr:9.663263561673029e-05 tokens_per_second_per_gpu:4034.103647935295 mfu:19.029555196367244 +Step 4337 | loss:1.5336554050445557 lr:9.663107103905917e-05 tokens_per_second_per_gpu:4163.29905479232 mfu:19.638991973521243 +Step 4338 | loss:1.3209956884384155 lr:9.662950611066983e-05 tokens_per_second_per_gpu:3444.115677813016 mfu:16.24648127897278 +Step 4339 | loss:1.2835605144500732 lr:9.662794083157403e-05 tokens_per_second_per_gpu:5383.895464472759 mfu:25.396753493205967 +Step 4340 | loss:1.327893614768982 lr:9.662637520178352e-05 tokens_per_second_per_gpu:4842.514544886711 mfu:22.84296732640889 +Step 4341 | loss:1.5208466053009033 lr:9.66248092213101e-05 tokens_per_second_per_gpu:3012.6976022953418 mfu:14.21140860924206 +Step 4342 | loss:1.2696677446365356 lr:9.662324289016555e-05 tokens_per_second_per_gpu:4240.53845278365 mfu:20.00334339224656 +Step 4343 | loss:1.2745252847671509 lr:9.662167620836164e-05 tokens_per_second_per_gpu:4202.878318917851 mfu:19.825694115320175 +Step 4344 | loss:1.279686450958252 lr:9.662010917591014e-05 tokens_per_second_per_gpu:4002.6993324991345 mfu:18.881415682326146 +Step 4345 | loss:1.1803900003433228 lr:9.661854179282286e-05 tokens_per_second_per_gpu:5275.304256925989 mfu:24.88451023963734 +Step 4346 | loss:1.1396372318267822 lr:9.661697405911157e-05 tokens_per_second_per_gpu:5342.513774546391 mfu:25.201549001378933 +Step 4347 | loss:1.2528598308563232 lr:9.661540597478807e-05 tokens_per_second_per_gpu:4635.876617459968 mfu:21.868220966671036 +Step 4348 | loss:1.2605026960372925 lr:9.661383753986417e-05 tokens_per_second_per_gpu:5081.234754316654 mfu:23.969051284157594 +Step 4349 | loss:1.1967908143997192 lr:9.661226875435163e-05 tokens_per_second_per_gpu:4054.9057792674653 mfu:19.12768239907068 +Step 4350 | loss:1.1723482608795166 lr:9.661069961826227e-05 tokens_per_second_per_gpu:5335.2240373645345 mfu:25.1671620673347 +Step 4351 | loss:1.3787693977355957 lr:9.660913013160792e-05 tokens_per_second_per_gpu:4546.325140563824 mfu:21.44579137109444 +Step 4352 | loss:1.4363698959350586 lr:9.660756029440032e-05 tokens_per_second_per_gpu:4083.601224437623 mfu:19.263043709886112 +Step 4353 | loss:1.3560422658920288 lr:9.660599010665133e-05 tokens_per_second_per_gpu:5325.91355656164 mfu:25.12324294085562 +Step 4354 | loss:1.3314250707626343 lr:9.660441956837273e-05 tokens_per_second_per_gpu:5355.639830638424 mfu:25.263466847501157 +Step 4355 | loss:1.2871227264404297 lr:9.660284867957636e-05 tokens_per_second_per_gpu:4809.301935242863 mfu:22.686297780063327 +Step 4356 | loss:1.3197505474090576 lr:9.660127744027401e-05 tokens_per_second_per_gpu:5342.844370942007 mfu:25.203108480982618 +Step 4357 | loss:1.4553405046463013 lr:9.65997058504775e-05 tokens_per_second_per_gpu:3618.457785568491 mfu:17.06888274708623 +Step 4358 | loss:1.2810477018356323 lr:9.659813391019867e-05 tokens_per_second_per_gpu:5317.961096758192 mfu:25.0857298311332 +Step 4359 | loss:1.4243919849395752 lr:9.659656161944932e-05 tokens_per_second_per_gpu:4348.540942389594 mfu:20.51280956284267 +Step 4360 | loss:1.197025179862976 lr:9.659498897824129e-05 tokens_per_second_per_gpu:4841.266587149198 mfu:22.837080496837498 +Step 4361 | loss:1.275671362876892 lr:9.659341598658639e-05 tokens_per_second_per_gpu:5330.187091092687 mfu:25.143401932378747 +Step 4362 | loss:1.3814294338226318 lr:9.659184264449648e-05 tokens_per_second_per_gpu:4999.64738977331 mfu:23.58418976536684 +Step 4363 | loss:1.2118163108825684 lr:9.659026895198336e-05 tokens_per_second_per_gpu:4214.233096901428 mfu:19.8792565403936 +Step 4364 | loss:1.3029996156692505 lr:9.658869490905889e-05 tokens_per_second_per_gpu:4304.582991600194 mfu:20.305452408969288 +Step 4365 | loss:1.3545867204666138 lr:9.658712051573489e-05 tokens_per_second_per_gpu:4671.716250351373 mfu:22.03728262989171 +Step 4366 | loss:1.3513461351394653 lr:9.658554577202322e-05 tokens_per_second_per_gpu:5363.078200495661 mfu:25.298554907234507 +Step 4367 | loss:1.2231944799423218 lr:9.658397067793571e-05 tokens_per_second_per_gpu:4802.416785083521 mfu:22.653819352033963 +Step 4368 | loss:1.3618295192718506 lr:9.658239523348422e-05 tokens_per_second_per_gpu:4389.102010776624 mfu:20.70414304285617 +Step 4369 | loss:1.316206455230713 lr:9.658081943868061e-05 tokens_per_second_per_gpu:3674.61163636131 mfu:17.33377004211085 +Step 4370 | loss:1.15505850315094 lr:9.65792432935367e-05 tokens_per_second_per_gpu:4884.915237791554 mfu:23.042978629145225 +Step 4371 | loss:1.3632014989852905 lr:9.657766679806435e-05 tokens_per_second_per_gpu:5270.795473798658 mfu:24.863241540347992 +Step 4372 | loss:1.3945869207382202 lr:9.657608995227541e-05 tokens_per_second_per_gpu:5109.263545279106 mfu:24.101267873332343 +Step 4373 | loss:1.491770625114441 lr:9.657451275618178e-05 tokens_per_second_per_gpu:3709.422179021107 mfu:17.49797731112813 +Step 4374 | loss:1.4398609399795532 lr:9.657293520979529e-05 tokens_per_second_per_gpu:4436.344645141477 mfu:20.926994609579765 +Step 4375 | loss:1.2981305122375488 lr:9.657135731312781e-05 tokens_per_second_per_gpu:4261.360321961288 mfu:20.101563701734687 +Step 4376 | loss:1.1448153257369995 lr:9.65697790661912e-05 tokens_per_second_per_gpu:4212.088018223041 mfu:19.869137838279485 +Step 4377 | loss:1.3677177429199219 lr:9.656820046899735e-05 tokens_per_second_per_gpu:4339.859829753844 mfu:20.471859273389285 +Step 4378 | loss:1.1926175355911255 lr:9.656662152155812e-05 tokens_per_second_per_gpu:3779.3545668623888 mfu:17.827860316271178 +Step 4379 | loss:1.3119173049926758 lr:9.656504222388538e-05 tokens_per_second_per_gpu:5314.386699205905 mfu:25.068868788023952 +Step 4380 | loss:1.3137797117233276 lr:9.656346257599102e-05 tokens_per_second_per_gpu:5183.461817859026 mfu:24.45127378462254 +Step 4381 | loss:1.3491902351379395 lr:9.656188257788693e-05 tokens_per_second_per_gpu:4357.667031619786 mfu:20.555858882812867 +Step 4382 | loss:1.2982704639434814 lr:9.656030222958497e-05 tokens_per_second_per_gpu:4329.857180676924 mfu:20.42467507106554 +Step 4383 | loss:1.3230586051940918 lr:9.655872153109704e-05 tokens_per_second_per_gpu:3675.538971044031 mfu:17.338144438028856 +Step 4384 | loss:1.489039659500122 lr:9.655714048243502e-05 tokens_per_second_per_gpu:5189.628477941327 mfu:24.480362972371456 +Step 4385 | loss:1.410139560699463 lr:9.65555590836108e-05 tokens_per_second_per_gpu:5311.508140533668 mfu:25.055290135634255 +Step 4386 | loss:1.1492434740066528 lr:9.65539773346363e-05 tokens_per_second_per_gpu:5323.456303785518 mfu:25.111651660260094 +Step 4387 | loss:1.2412387132644653 lr:9.655239523552337e-05 tokens_per_second_per_gpu:4236.53730777714 mfu:19.984469308585115 +Step 4388 | loss:1.2378218173980713 lr:9.655081278628394e-05 tokens_per_second_per_gpu:4380.660456196098 mfu:20.664322789622094 +Step 4389 | loss:1.1946320533752441 lr:9.654922998692992e-05 tokens_per_second_per_gpu:4472.672253798631 mfu:21.09835813774473 +Step 4390 | loss:1.2834084033966064 lr:9.654764683747319e-05 tokens_per_second_per_gpu:3037.207424776425 mfu:14.32702562369226 +Step 4391 | loss:1.2522687911987305 lr:9.654606333792568e-05 tokens_per_second_per_gpu:5303.3600995014485 mfu:25.016854435886128 +Step 4392 | loss:1.4033032655715942 lr:9.654447948829927e-05 tokens_per_second_per_gpu:4859.768420184289 mfu:22.924356800002254 +Step 4393 | loss:1.3065935373306274 lr:9.65428952886059e-05 tokens_per_second_per_gpu:4437.294977371112 mfu:20.93147748885035 +Step 4394 | loss:1.1721514463424683 lr:9.654131073885748e-05 tokens_per_second_per_gpu:4245.282528193185 mfu:20.02572200537606 +Step 4395 | loss:1.4042928218841553 lr:9.653972583906593e-05 tokens_per_second_per_gpu:3030.3259820972135 mfu:14.29456468447923 +Step 4396 | loss:1.1920268535614014 lr:9.653814058924316e-05 tokens_per_second_per_gpu:5377.122427845943 mfu:25.364803923838064 +Step 4397 | loss:1.1722718477249146 lr:9.653655498940109e-05 tokens_per_second_per_gpu:3795.6208501583583 mfu:17.90459114989238 +Step 4398 | loss:1.1671507358551025 lr:9.653496903955165e-05 tokens_per_second_per_gpu:4713.886161698221 mfu:22.236205296643586 +Step 4399 | loss:1.223537564277649 lr:9.653338273970677e-05 tokens_per_second_per_gpu:4251.702979903256 mfu:20.056008371534492 +Step 4400 | loss:1.4182378053665161 lr:9.653179608987839e-05 tokens_per_second_per_gpu:4588.520109182782 mfu:21.64483223727407 +Step 4401 | loss:1.5463413000106812 lr:9.653020909007842e-05 tokens_per_second_per_gpu:3825.0202315464658 mfu:18.043273047951782 +Step 4402 | loss:1.5334796905517578 lr:9.652862174031882e-05 tokens_per_second_per_gpu:5377.285105336799 mfu:25.36557130131807 +Step 4403 | loss:1.2262461185455322 lr:9.652703404061152e-05 tokens_per_second_per_gpu:4750.494542164515 mfu:22.408893273336872 +Step 4404 | loss:1.2566969394683838 lr:9.652544599096844e-05 tokens_per_second_per_gpu:5027.829946054277 mfu:23.71713169178409 +Step 4405 | loss:1.2553861141204834 lr:9.652385759140156e-05 tokens_per_second_per_gpu:4124.386532035273 mfu:19.455434964515263 +Step 4406 | loss:1.250346302986145 lr:9.652226884192283e-05 tokens_per_second_per_gpu:4732.0293517808705 mfu:22.321789819810874 +Step 4407 | loss:1.4061756134033203 lr:9.652067974254416e-05 tokens_per_second_per_gpu:5328.721249599591 mfu:25.136487307957463 +Step 4408 | loss:1.3112038373947144 lr:9.651909029327752e-05 tokens_per_second_per_gpu:5332.3088026286205 mfu:25.153410407695063 +Step 4409 | loss:1.2773069143295288 lr:9.651750049413486e-05 tokens_per_second_per_gpu:5214.984858698996 mfu:24.599973346650042 +Step 4410 | loss:1.4585226774215698 lr:9.651591034512816e-05 tokens_per_second_per_gpu:5336.619749236079 mfu:25.173745878365775 +Step 4411 | loss:1.2186386585235596 lr:9.651431984626935e-05 tokens_per_second_per_gpu:3432.089389632265 mfu:16.18975122572783 +Step 4412 | loss:1.3341772556304932 lr:9.651272899757041e-05 tokens_per_second_per_gpu:5322.938066083253 mfu:25.109207044598044 +Step 4413 | loss:1.304425835609436 lr:9.65111377990433e-05 tokens_per_second_per_gpu:4459.0560505602525 mfu:21.03412818390547 +Step 4414 | loss:1.3179641962051392 lr:9.65095462507e-05 tokens_per_second_per_gpu:5362.994355898343 mfu:25.298159397963595 +Step 4415 | loss:1.2337323427200317 lr:9.650795435255245e-05 tokens_per_second_per_gpu:3930.5365780885295 mfu:18.541011657535826 +Step 4416 | loss:1.251758098602295 lr:9.650636210461264e-05 tokens_per_second_per_gpu:4760.699846259846 mfu:22.457033433958973 +Step 4417 | loss:1.34142005443573 lr:9.650476950689256e-05 tokens_per_second_per_gpu:5340.449690747868 mfu:25.19181236593229 +Step 4418 | loss:1.374226450920105 lr:9.650317655940416e-05 tokens_per_second_per_gpu:3675.486655184355 mfu:17.33789765519276 +Step 4419 | loss:1.6141626834869385 lr:9.650158326215944e-05 tokens_per_second_per_gpu:4970.8656081433455 mfu:23.448421190739975 +Step 4420 | loss:1.2811025381088257 lr:9.64999896151704e-05 tokens_per_second_per_gpu:3426.1591016900034 mfu:16.161777045692723 +Step 4421 | loss:1.2971104383468628 lr:9.649839561844898e-05 tokens_per_second_per_gpu:3318.566574612091 mfu:15.654244738288158 +Step 4422 | loss:1.407135248184204 lr:9.649680127200719e-05 tokens_per_second_per_gpu:4851.672726651093 mfu:22.886168032338418 +Step 4423 | loss:1.1904988288879395 lr:9.649520657585704e-05 tokens_per_second_per_gpu:4812.120106971252 mfu:22.699591576935823 +Step 4424 | loss:1.3100371360778809 lr:9.64936115300105e-05 tokens_per_second_per_gpu:4198.472891618393 mfu:19.804912963105224 +Step 4425 | loss:1.2985891103744507 lr:9.649201613447957e-05 tokens_per_second_per_gpu:5366.243063720632 mfu:25.31348410704804 +Step 4426 | loss:1.2513518333435059 lr:9.649042038927627e-05 tokens_per_second_per_gpu:4453.172060240379 mfu:21.006372397654356 +Step 4427 | loss:1.2716418504714966 lr:9.648882429441257e-05 tokens_per_second_per_gpu:5351.3872878474385 mfu:25.24340687759746 +Step 4428 | loss:1.2773921489715576 lr:9.64872278499005e-05 tokens_per_second_per_gpu:4278.741911410104 mfu:20.183555624769713 +Step 4429 | loss:1.3048951625823975 lr:9.648563105575206e-05 tokens_per_second_per_gpu:5372.812745104521 mfu:25.344474415039077 +Step 4430 | loss:1.4016278982162476 lr:9.648403391197926e-05 tokens_per_second_per_gpu:4572.561441418238 mfu:21.569552478599228 +Step 4431 | loss:1.282878041267395 lr:9.64824364185941e-05 tokens_per_second_per_gpu:4281.413953448484 mfu:20.196160102962136 +Step 4432 | loss:1.4409253597259521 lr:9.64808385756086e-05 tokens_per_second_per_gpu:3876.763585817126 mfu:18.287355278373077 +Step 4433 | loss:1.541406512260437 lr:9.64792403830348e-05 tokens_per_second_per_gpu:3170.866232962223 mfu:14.957517026448876 +Step 4434 | loss:1.2792248725891113 lr:9.647764184088469e-05 tokens_per_second_per_gpu:4950.200622124899 mfu:23.35094092588078 +Step 4435 | loss:1.4418612718582153 lr:9.647604294917029e-05 tokens_per_second_per_gpu:5284.250429982558 mfu:24.92671086431969 +Step 4436 | loss:1.2413979768753052 lr:9.647444370790366e-05 tokens_per_second_per_gpu:5305.239645449988 mfu:25.02572057480899 +Step 4437 | loss:1.2622802257537842 lr:9.647284411709681e-05 tokens_per_second_per_gpu:5340.896272269183 mfu:25.193918967162997 +Step 4438 | loss:1.2923134565353394 lr:9.647124417676177e-05 tokens_per_second_per_gpu:5302.851088494893 mfu:25.01445334412197 +Step 4439 | loss:1.21271550655365 lr:9.646964388691056e-05 tokens_per_second_per_gpu:3820.2342148462335 mfu:18.02069659059814 +Step 4440 | loss:1.3137950897216797 lr:9.646804324755523e-05 tokens_per_second_per_gpu:5302.022710916839 mfu:25.010545745750502 +Step 4441 | loss:1.2612115144729614 lr:9.646644225870784e-05 tokens_per_second_per_gpu:4559.582816861825 mfu:21.508330092184938 +Step 4442 | loss:1.4326622486114502 lr:9.64648409203804e-05 tokens_per_second_per_gpu:4148.055084041321 mfu:19.567083562599176 +Step 4443 | loss:1.3689351081848145 lr:9.646323923258494e-05 tokens_per_second_per_gpu:4546.253323865261 mfu:21.445452599474127 +Step 4444 | loss:1.2210088968276978 lr:9.646163719533355e-05 tokens_per_second_per_gpu:5382.136540532529 mfu:25.388456348876527 +Step 4445 | loss:1.2959938049316406 lr:9.646003480863825e-05 tokens_per_second_per_gpu:2828.0235778637402 mfu:13.34026972736056 +Step 4446 | loss:1.407706379890442 lr:9.64584320725111e-05 tokens_per_second_per_gpu:5392.976345030541 mfu:25.43958955615501 +Step 4447 | loss:1.32221519947052 lr:9.645682898696416e-05 tokens_per_second_per_gpu:4871.298900649492 mfu:22.978748043659444 +Step 4448 | loss:1.3771272897720337 lr:9.645522555200948e-05 tokens_per_second_per_gpu:5277.410285608036 mfu:24.8944447362561 +Step 4449 | loss:1.4000911712646484 lr:9.645362176765911e-05 tokens_per_second_per_gpu:3800.7662646818785 mfu:17.928862948098068 +Step 4450 | loss:1.30512273311615 lr:9.645201763392513e-05 tokens_per_second_per_gpu:4512.129363393143 mfu:21.284484055779128 +Step 4451 | loss:1.2907332181930542 lr:9.645041315081961e-05 tokens_per_second_per_gpu:4103.24210990444 mfu:19.355693117713702 +Step 4452 | loss:1.387941837310791 lr:9.644880831835459e-05 tokens_per_second_per_gpu:5311.688519001464 mfu:25.05614101164259 +Step 4453 | loss:1.3359295129776 lr:9.644720313654217e-05 tokens_per_second_per_gpu:4537.823753059378 mfu:21.40568887574969 +Step 4454 | loss:1.3453258275985718 lr:9.64455976053944e-05 tokens_per_second_per_gpu:5230.295993822218 mfu:24.672198583375312 +Step 4455 | loss:1.3395081758499146 lr:9.644399172492336e-05 tokens_per_second_per_gpu:5320.045875552658 mfu:25.095564088406636 +Step 4456 | loss:1.378575086593628 lr:9.644238549514113e-05 tokens_per_second_per_gpu:3789.3904435999293 mfu:17.875201259139295 +Step 4457 | loss:1.382710576057434 lr:9.64407789160598e-05 tokens_per_second_per_gpu:5140.967416335602 mfu:24.250820442344022 +Step 4458 | loss:1.2122596502304077 lr:9.643917198769144e-05 tokens_per_second_per_gpu:4721.05208143833 mfu:22.270008162689663 +Step 4459 | loss:1.3977861404418945 lr:9.643756471004814e-05 tokens_per_second_per_gpu:3671.509658475694 mfu:17.319137483172394 +Step 4460 | loss:1.241584062576294 lr:9.643595708314201e-05 tokens_per_second_per_gpu:3675.8610846521174 mfu:17.339663902876445 +Step 4461 | loss:1.2864952087402344 lr:9.64343491069851e-05 tokens_per_second_per_gpu:5319.343849382681 mfu:25.09225251118484 +Step 4462 | loss:1.4792336225509644 lr:9.643274078158954e-05 tokens_per_second_per_gpu:3856.9053115709894 mfu:18.193680410583084 +Step 4463 | loss:1.4744629859924316 lr:9.64311321069674e-05 tokens_per_second_per_gpu:3539.805149456227 mfu:16.697864843021183 +Step 4464 | loss:1.342321515083313 lr:9.64295230831308e-05 tokens_per_second_per_gpu:4140.055516362948 mfu:19.529348236993304 +Step 4465 | loss:1.1462503671646118 lr:9.642791371009183e-05 tokens_per_second_per_gpu:4644.738427673765 mfu:21.91002363743025 +Step 4466 | loss:1.1236677169799805 lr:9.64263039878626e-05 tokens_per_second_per_gpu:4822.910111863368 mfu:22.750489870976597 +Step 4467 | loss:1.3454217910766602 lr:9.64246939164552e-05 tokens_per_second_per_gpu:4099.670989095852 mfu:19.33884753156355 +Step 4468 | loss:1.200859546661377 lr:9.642308349588177e-05 tokens_per_second_per_gpu:4859.094622804666 mfu:22.921178382800882 +Step 4469 | loss:1.280042290687561 lr:9.642147272615438e-05 tokens_per_second_per_gpu:5330.022709739534 mfu:25.142626517489568 +Step 4470 | loss:1.4800945520401 lr:9.64198616072852e-05 tokens_per_second_per_gpu:4353.835988774681 mfu:20.537787200069193 +Step 4471 | loss:1.1122663021087646 lr:9.641825013928629e-05 tokens_per_second_per_gpu:3932.5360484596054 mfu:18.55044350042101 +Step 4472 | loss:1.1749104261398315 lr:9.641663832216983e-05 tokens_per_second_per_gpu:4620.637418230865 mfu:21.796335063831854 +Step 4473 | loss:1.1963444948196411 lr:9.641502615594788e-05 tokens_per_second_per_gpu:5303.242464291605 mfu:25.016299530530656 +Step 4474 | loss:1.3882395029067993 lr:9.641341364063261e-05 tokens_per_second_per_gpu:4460.3430772017655 mfu:21.040199308162872 +Step 4475 | loss:1.4423234462738037 lr:9.641180077623613e-05 tokens_per_second_per_gpu:3718.5324078340614 mfu:17.540951814804153 +Step 4476 | loss:1.4301741123199463 lr:9.641018756277058e-05 tokens_per_second_per_gpu:3816.5005646780464 mfu:18.00308432572839 +Step 4477 | loss:1.311987280845642 lr:9.640857400024809e-05 tokens_per_second_per_gpu:5333.950327211774 mfu:25.161153759226767 +Step 4478 | loss:1.2812572717666626 lr:9.640696008868077e-05 tokens_per_second_per_gpu:3694.5847629507975 mfu:17.427986688000257 +Step 4479 | loss:1.5517125129699707 lr:9.640534582808081e-05 tokens_per_second_per_gpu:4455.195741944607 mfu:21.0159184495289 +Step 4480 | loss:1.408205270767212 lr:9.640373121846031e-05 tokens_per_second_per_gpu:4849.58045319689 mfu:22.876298421476278 +Step 4481 | loss:1.453473687171936 lr:9.640211625983142e-05 tokens_per_second_per_gpu:3938.221710898821 mfu:18.577263740220836 +Step 4482 | loss:1.3646132946014404 lr:9.640050095220631e-05 tokens_per_second_per_gpu:4346.582679082046 mfu:20.503572100706787 +Step 4483 | loss:1.261462926864624 lr:9.63988852955971e-05 tokens_per_second_per_gpu:5310.39832213603 mfu:25.05005493289806 +Step 4484 | loss:1.2380352020263672 lr:9.639726929001595e-05 tokens_per_second_per_gpu:4096.558992152325 mfu:19.324167711019506 +Step 4485 | loss:1.291033148765564 lr:9.639565293547503e-05 tokens_per_second_per_gpu:4680.675136638655 mfu:22.07954322505244 +Step 4486 | loss:1.2593032121658325 lr:9.639403623198647e-05 tokens_per_second_per_gpu:5390.727715815489 mfu:25.428982388491903 +Step 4487 | loss:1.2693331241607666 lr:9.639241917956245e-05 tokens_per_second_per_gpu:4261.40572852488 mfu:20.10177789224231 +Step 4488 | loss:1.1301605701446533 lr:9.639080177821514e-05 tokens_per_second_per_gpu:5336.15483006659 mfu:25.171552775319906 +Step 4489 | loss:1.2883977890014648 lr:9.638918402795667e-05 tokens_per_second_per_gpu:4747.517340799962 mfu:22.394849306537342 +Step 4490 | loss:1.2978906631469727 lr:9.638756592879922e-05 tokens_per_second_per_gpu:4496.803992479554 mfu:21.212191666401825 +Step 4491 | loss:1.1920255422592163 lr:9.638594748075499e-05 tokens_per_second_per_gpu:4646.842564920961 mfu:21.919949211828108 +Step 4492 | loss:1.367895483970642 lr:9.638432868383612e-05 tokens_per_second_per_gpu:3293.3397946510468 mfu:15.535245713079423 +Step 4493 | loss:1.2554430961608887 lr:9.638270953805479e-05 tokens_per_second_per_gpu:5214.352620216072 mfu:24.59699096985508 +Step 4494 | loss:1.399181604385376 lr:9.63810900434232e-05 tokens_per_second_per_gpu:3644.0427540155997 mfu:17.18957140849699 +Step 4495 | loss:1.2970043420791626 lr:9.63794701999535e-05 tokens_per_second_per_gpu:3886.1263172944623 mfu:18.331520880198386 +Step 4496 | loss:1.4875506162643433 lr:9.637785000765789e-05 tokens_per_second_per_gpu:5366.987310935464 mfu:25.316994848142073 +Step 4497 | loss:1.2730923891067505 lr:9.637622946654855e-05 tokens_per_second_per_gpu:4392.337255953932 mfu:20.719404246347278 +Step 4498 | loss:1.334471344947815 lr:9.637460857663767e-05 tokens_per_second_per_gpu:3735.8724842866795 mfu:17.622747913953265 +Step 4499 | loss:1.1380517482757568 lr:9.637298733793745e-05 tokens_per_second_per_gpu:5362.436342664957 mfu:25.295527154334838 +Step 4500 | loss:1.4318417310714722 lr:9.637136575046006e-05 tokens_per_second_per_gpu:4388.039912752783 mfu:20.699132945265095 +Step 4501 | loss:1.4199365377426147 lr:9.636974381421772e-05 tokens_per_second_per_gpu:5069.128316286135 mfu:23.91194315039631 +Step 4502 | loss:1.483965277671814 lr:9.636812152922264e-05 tokens_per_second_per_gpu:5134.679530984943 mfu:24.221159414321097 +Step 4503 | loss:1.294458031654358 lr:9.636649889548697e-05 tokens_per_second_per_gpu:5341.770589501186 mfu:25.19804327072026 +Step 4504 | loss:1.392783284187317 lr:9.636487591302295e-05 tokens_per_second_per_gpu:3722.8056597395985 mfu:17.561109473134525 +Step 4505 | loss:1.1995352506637573 lr:9.63632525818428e-05 tokens_per_second_per_gpu:5395.047278824785 mfu:25.449358504200198 +Step 4506 | loss:1.1546112298965454 lr:9.63616289019587e-05 tokens_per_second_per_gpu:5319.241709371491 mfu:25.091770699325178 +Step 4507 | loss:1.2255370616912842 lr:9.636000487338287e-05 tokens_per_second_per_gpu:4656.18395915247 mfu:21.96401416222434 +Step 4508 | loss:1.2199440002441406 lr:9.635838049612753e-05 tokens_per_second_per_gpu:5328.739659181085 mfu:25.136574149094365 +Step 4509 | loss:1.2889297008514404 lr:9.635675577020489e-05 tokens_per_second_per_gpu:5326.485461716676 mfu:25.125940715049 +Step 4510 | loss:1.2907932996749878 lr:9.635513069562718e-05 tokens_per_second_per_gpu:5334.136537821953 mfu:25.162032146445792 +Step 4511 | loss:1.2988873720169067 lr:9.635350527240663e-05 tokens_per_second_per_gpu:4067.5118582572763 mfu:19.187147424484476 +Step 4512 | loss:1.2110323905944824 lr:9.635187950055545e-05 tokens_per_second_per_gpu:5360.487224064677 mfu:25.286332829343394 +Step 4513 | loss:1.3766645193099976 lr:9.635025338008584e-05 tokens_per_second_per_gpu:4154.314726794728 mfu:19.596611365472178 +Step 4514 | loss:1.2686196565628052 lr:9.634862691101009e-05 tokens_per_second_per_gpu:4192.456005618193 mfu:19.776530285255667 +Step 4515 | loss:1.2527445554733276 lr:9.634700009334039e-05 tokens_per_second_per_gpu:3870.5178267551646 mfu:18.257892967241425 +Step 4516 | loss:1.1928656101226807 lr:9.634537292708899e-05 tokens_per_second_per_gpu:5287.231281291483 mfu:24.94077204852957 +Step 4517 | loss:1.3271199464797974 lr:9.634374541226813e-05 tokens_per_second_per_gpu:5344.964061696691 mfu:25.213107423928996 +Step 4518 | loss:1.2314865589141846 lr:9.634211754889005e-05 tokens_per_second_per_gpu:3902.394747945923 mfu:18.408261843261812 +Step 4519 | loss:1.3854715824127197 lr:9.634048933696697e-05 tokens_per_second_per_gpu:4114.968098215408 mfu:19.411006605236444 +Step 4520 | loss:1.439704418182373 lr:9.633886077651117e-05 tokens_per_second_per_gpu:5318.9731065389715 mfu:25.09050365393581 +Step 4521 | loss:1.2759672403335571 lr:9.633723186753489e-05 tokens_per_second_per_gpu:4067.415241242353 mfu:19.186691665540522 +Step 4522 | loss:1.4618337154388428 lr:9.633560261005038e-05 tokens_per_second_per_gpu:4559.603399955243 mfu:21.508427186148424 +Step 4523 | loss:1.412157654762268 lr:9.633397300406988e-05 tokens_per_second_per_gpu:5360.419558653762 mfu:25.286013640055195 +Step 4524 | loss:1.4736430644989014 lr:9.633234304960566e-05 tokens_per_second_per_gpu:4607.305392926627 mfu:21.733445625793696 +Step 4525 | loss:1.1880028247833252 lr:9.633071274666999e-05 tokens_per_second_per_gpu:4551.843032115522 mfu:21.471820206993176 +Step 4526 | loss:1.168884038925171 lr:9.63290820952751e-05 tokens_per_second_per_gpu:4384.686162319075 mfu:20.68331273226047 +Step 4527 | loss:1.476828694343567 lr:9.632745109543327e-05 tokens_per_second_per_gpu:3901.8285250634863 mfu:18.40559087331777 +Step 4528 | loss:1.107367753982544 lr:9.632581974715678e-05 tokens_per_second_per_gpu:4912.784763164169 mfu:23.1744439353592 +Step 4529 | loss:1.2261536121368408 lr:9.632418805045789e-05 tokens_per_second_per_gpu:5206.7844088235215 mfu:24.561290425446405 +Step 4530 | loss:1.28826904296875 lr:9.632255600534885e-05 tokens_per_second_per_gpu:4167.004797948674 mfu:19.656472596254748 +Step 4531 | loss:1.343187928199768 lr:9.632092361184197e-05 tokens_per_second_per_gpu:5299.701823532859 mfu:24.999597723976045 +Step 4532 | loss:1.486954927444458 lr:9.631929086994952e-05 tokens_per_second_per_gpu:3979.835330173624 mfu:18.77356228235566 +Step 4533 | loss:1.1092873811721802 lr:9.631765777968376e-05 tokens_per_second_per_gpu:5347.528357864884 mfu:25.225203646468405 +Step 4534 | loss:1.3690088987350464 lr:9.631602434105699e-05 tokens_per_second_per_gpu:4500.17289549448 mfu:21.228083356717356 +Step 4535 | loss:1.2062145471572876 lr:9.631439055408149e-05 tokens_per_second_per_gpu:4981.947591774771 mfu:23.50069679831079 +Step 4536 | loss:1.285193681716919 lr:9.631275641876956e-05 tokens_per_second_per_gpu:5326.6622627082625 mfu:25.126774715491678 +Step 4537 | loss:1.2724826335906982 lr:9.631112193513348e-05 tokens_per_second_per_gpu:5027.671287636148 mfu:23.716383272955863 +Step 4538 | loss:1.1790164709091187 lr:9.630948710318552e-05 tokens_per_second_per_gpu:5085.777976093125 mfu:23.990482436430774 +Step 4539 | loss:1.2608352899551392 lr:9.630785192293801e-05 tokens_per_second_per_gpu:5133.030426038341 mfu:24.213380304921653 +Step 4540 | loss:1.3762056827545166 lr:9.630621639440325e-05 tokens_per_second_per_gpu:4826.796758020769 mfu:22.76882384403127 +Step 4541 | loss:1.439659833908081 lr:9.63045805175935e-05 tokens_per_second_per_gpu:3589.491396144259 mfu:16.932243345996472 +Step 4542 | loss:1.401132583618164 lr:9.630294429252112e-05 tokens_per_second_per_gpu:4829.46772680286 mfu:22.781423259490662 +Step 4543 | loss:1.3932380676269531 lr:9.630130771919836e-05 tokens_per_second_per_gpu:5344.154071032161 mfu:25.209286559765747 +Step 4544 | loss:1.4151346683502197 lr:9.629967079763758e-05 tokens_per_second_per_gpu:5189.990441858039 mfu:24.482070417923996 +Step 4545 | loss:1.292719841003418 lr:9.629803352785106e-05 tokens_per_second_per_gpu:5353.450273873102 mfu:25.253138334662225 +Step 4546 | loss:1.423032522201538 lr:9.629639590985113e-05 tokens_per_second_per_gpu:3877.2608872669643 mfu:18.28970113416034 +Step 4547 | loss:1.1640516519546509 lr:9.629475794365009e-05 tokens_per_second_per_gpu:4008.8705263260654 mfu:18.910526256522022 +Step 4548 | loss:1.1897952556610107 lr:9.629311962926027e-05 tokens_per_second_per_gpu:5061.704548678398 mfu:23.876923971965255 +Step 4549 | loss:1.2881317138671875 lr:9.629148096669398e-05 tokens_per_second_per_gpu:5280.549645086249 mfu:24.909253630544793 +Step 4550 | loss:1.466040015220642 lr:9.628984195596357e-05 tokens_per_second_per_gpu:3822.4010292271223 mfu:18.030917823729443 +Step 4551 | loss:1.2327375411987305 lr:9.628820259708135e-05 tokens_per_second_per_gpu:5329.0484316490365 mfu:25.13803068150771 +Step 4552 | loss:1.3011544942855835 lr:9.628656289005964e-05 tokens_per_second_per_gpu:4292.088233659683 mfu:20.246512503938366 +Step 4553 | loss:1.3836565017700195 lr:9.62849228349108e-05 tokens_per_second_per_gpu:3962.2876899880994 mfu:18.6907870696644 +Step 4554 | loss:1.3536248207092285 lr:9.628328243164713e-05 tokens_per_second_per_gpu:4826.095858006065 mfu:22.76551757907604 +Step 4555 | loss:1.323733925819397 lr:9.6281641680281e-05 tokens_per_second_per_gpu:3859.674985174091 mfu:18.206745433523963 +Step 4556 | loss:1.194644570350647 lr:9.628000058082474e-05 tokens_per_second_per_gpu:4777.604828761951 mfu:22.536777120708837 +Step 4557 | loss:1.315338134765625 lr:9.627835913329068e-05 tokens_per_second_per_gpu:4025.2975814954307 mfu:18.98801547850054 +Step 4558 | loss:1.3189703226089478 lr:9.627671733769118e-05 tokens_per_second_per_gpu:3698.0103212530616 mfu:17.44414563097255 +Step 4559 | loss:1.3434659242630005 lr:9.62750751940386e-05 tokens_per_second_per_gpu:5333.370485162044 mfu:25.158418545346937 +Step 4560 | loss:1.2872132062911987 lr:9.627343270234526e-05 tokens_per_second_per_gpu:5317.7082103395705 mfu:25.08453692274969 +Step 4561 | loss:1.214113473892212 lr:9.627178986262354e-05 tokens_per_second_per_gpu:5054.139871830302 mfu:23.841240100605823 +Step 4562 | loss:1.4310603141784668 lr:9.627014667488576e-05 tokens_per_second_per_gpu:4712.341988775405 mfu:22.228921169503597 +Step 4563 | loss:1.3146436214447021 lr:9.62685031391443e-05 tokens_per_second_per_gpu:4910.303223069116 mfu:23.162738087336436 +Step 4564 | loss:1.6119412183761597 lr:9.626685925541155e-05 tokens_per_second_per_gpu:4502.608124248172 mfu:21.23957074624139 +Step 4565 | loss:1.1779205799102783 lr:9.626521502369984e-05 tokens_per_second_per_gpu:5281.701082583646 mfu:24.914685157675507 +Step 4566 | loss:1.226333737373352 lr:9.626357044402155e-05 tokens_per_second_per_gpu:5347.964714461848 mfu:25.227262014986398 +Step 4567 | loss:1.3601561784744263 lr:9.626192551638902e-05 tokens_per_second_per_gpu:3780.7397275456133 mfu:17.83439435554673 +Step 4568 | loss:1.223354697227478 lr:9.626028024081467e-05 tokens_per_second_per_gpu:4389.477423461312 mfu:20.705913928541854 +Step 4569 | loss:1.1934064626693726 lr:9.625863461731084e-05 tokens_per_second_per_gpu:5004.826538472364 mfu:23.608620693434165 +Step 4570 | loss:1.2260719537734985 lr:9.625698864588992e-05 tokens_per_second_per_gpu:5060.7121048021645 mfu:23.872242444872676 +Step 4571 | loss:1.334921956062317 lr:9.625534232656428e-05 tokens_per_second_per_gpu:4350.540210723595 mfu:20.522240452684503 +Step 4572 | loss:1.234244704246521 lr:9.62536956593463e-05 tokens_per_second_per_gpu:5039.700853439193 mfu:23.773128787304397 +Step 4573 | loss:1.2621495723724365 lr:9.625204864424839e-05 tokens_per_second_per_gpu:4843.798150944921 mfu:22.84902231519125 +Step 4574 | loss:1.3665167093276978 lr:9.625040128128291e-05 tokens_per_second_per_gpu:5365.492745507193 mfu:25.309944728017765 +Step 4575 | loss:1.2642672061920166 lr:9.624875357046227e-05 tokens_per_second_per_gpu:3791.808864010969 mfu:17.886609360843238 +Step 4576 | loss:1.1040942668914795 lr:9.624710551179886e-05 tokens_per_second_per_gpu:5264.209622156952 mfu:24.832174954491908 +Step 4577 | loss:1.3630318641662598 lr:9.624545710530503e-05 tokens_per_second_per_gpu:4900.893486322289 mfu:23.118350753635017 +Step 4578 | loss:1.2633401155471802 lr:9.624380835099325e-05 tokens_per_second_per_gpu:4428.325440135686 mfu:20.88916665134999 +Step 4579 | loss:1.1232181787490845 lr:9.624215924887589e-05 tokens_per_second_per_gpu:4866.068908935849 mfu:22.95407728041759 +Step 4580 | loss:1.330931305885315 lr:9.624050979896533e-05 tokens_per_second_per_gpu:3758.3741737900195 mfu:17.728892222524635 +Step 4581 | loss:1.2019866704940796 lr:9.623886000127401e-05 tokens_per_second_per_gpu:3649.6035947923956 mfu:17.21580284321841 +Step 4582 | loss:1.2750853300094604 lr:9.62372098558143e-05 tokens_per_second_per_gpu:5331.628557801647 mfu:25.15020157678449 +Step 4583 | loss:1.2639859914779663 lr:9.623555936259867e-05 tokens_per_second_per_gpu:5347.504090918499 mfu:25.22508917514194 +Step 4584 | loss:1.326798439025879 lr:9.623390852163947e-05 tokens_per_second_per_gpu:4199.748950623413 mfu:19.810932351150022 +Step 4585 | loss:1.2807459831237793 lr:9.623225733294916e-05 tokens_per_second_per_gpu:3080.6679565202676 mfu:14.532036367059304 +Step 4586 | loss:1.2026758193969727 lr:9.623060579654013e-05 tokens_per_second_per_gpu:3628.0548728709705 mfu:17.11415384532397 +Step 4587 | loss:1.3489373922348022 lr:9.622895391242484e-05 tokens_per_second_per_gpu:4416.959747377545 mfu:20.83555273942269 +Step 4588 | loss:1.189315915107727 lr:9.622730168061567e-05 tokens_per_second_per_gpu:5384.4414512956755 mfu:25.399329006204617 +Step 4589 | loss:1.1060761213302612 lr:9.622564910112506e-05 tokens_per_second_per_gpu:4604.958835751433 mfu:21.722376515234366 +Step 4590 | loss:1.1179150342941284 lr:9.622399617396545e-05 tokens_per_second_per_gpu:3909.1265498810917 mfu:18.440016901553847 +Step 4591 | loss:1.2493458986282349 lr:9.622234289914926e-05 tokens_per_second_per_gpu:4964.221012798537 mfu:23.41707750081365 +Step 4592 | loss:1.3144872188568115 lr:9.622068927668895e-05 tokens_per_second_per_gpu:4778.6804696334075 mfu:22.541851102222754 +Step 4593 | loss:1.2594977617263794 lr:9.621903530659693e-05 tokens_per_second_per_gpu:4231.524095621431 mfu:19.960821131504446 +Step 4594 | loss:1.3004987239837646 lr:9.621738098888564e-05 tokens_per_second_per_gpu:5335.697691820212 mfu:25.169396376215914 +Step 4595 | loss:1.3156754970550537 lr:9.621572632356755e-05 tokens_per_second_per_gpu:4716.4722765874985 mfu:22.24840444181301 +Step 4596 | loss:1.1891402006149292 lr:9.621407131065505e-05 tokens_per_second_per_gpu:3547.502242878993 mfu:16.734173345955927 +Step 4597 | loss:1.3368407487869263 lr:9.621241595016065e-05 tokens_per_second_per_gpu:5333.709323650276 mfu:25.16001690430774 +Step 4598 | loss:1.4323437213897705 lr:9.621076024209676e-05 tokens_per_second_per_gpu:5321.864144959028 mfu:25.104141175425234 +Step 4599 | loss:1.3295629024505615 lr:9.620910418647587e-05 tokens_per_second_per_gpu:3929.462089250515 mfu:18.53594311036021 +Step 4600 | loss:1.1825823783874512 lr:9.620744778331039e-05 tokens_per_second_per_gpu:5358.22732489984 mfu:25.275672499404298 +Step 4601 | loss:1.265910029411316 lr:9.62057910326128e-05 tokens_per_second_per_gpu:5337.318444184207 mfu:25.17704174164575 +Step 4602 | loss:1.3428736925125122 lr:9.620413393439556e-05 tokens_per_second_per_gpu:5344.474324881725 mfu:25.210797251815087 +Step 4603 | loss:1.2449018955230713 lr:9.620247648867114e-05 tokens_per_second_per_gpu:4474.3912387564615 mfu:21.10646688307989 +Step 4604 | loss:1.1505919694900513 lr:9.620081869545199e-05 tokens_per_second_per_gpu:5363.834143160159 mfu:25.302120817760066 +Step 4605 | loss:1.2836270332336426 lr:9.619916055475058e-05 tokens_per_second_per_gpu:5311.91898850245 mfu:25.057228175600894 +Step 4606 | loss:1.1528278589248657 lr:9.61975020665794e-05 tokens_per_second_per_gpu:4708.211453559868 mfu:22.209436730149925 +Step 4607 | loss:1.3210253715515137 lr:9.619584323095092e-05 tokens_per_second_per_gpu:3386.8218231857704 mfu:15.976216391356461 +Step 4608 | loss:1.5026756525039673 lr:9.61941840478776e-05 tokens_per_second_per_gpu:4664.744466732041 mfu:22.004395537050375 +Step 4609 | loss:1.3339552879333496 lr:9.619252451737194e-05 tokens_per_second_per_gpu:4207.716718745549 mfu:19.848517672823558 +Step 4610 | loss:1.283016324043274 lr:9.619086463944638e-05 tokens_per_second_per_gpu:4081.67998063126 mfu:19.253980875053504 +Step 4611 | loss:1.3554733991622925 lr:9.618920441411345e-05 tokens_per_second_per_gpu:4844.809182296036 mfu:22.85379152257419 +Step 4612 | loss:1.2233928442001343 lr:9.618754384138561e-05 tokens_per_second_per_gpu:5312.226620795306 mfu:25.05867932961378 +Step 4613 | loss:1.1815721988677979 lr:9.618588292127539e-05 tokens_per_second_per_gpu:5315.286823111043 mfu:25.073114826061463 +Step 4614 | loss:1.2745994329452515 lr:9.618422165379521e-05 tokens_per_second_per_gpu:4487.559775231444 mfu:21.168585116416388 +Step 4615 | loss:1.3018471002578735 lr:9.618256003895764e-05 tokens_per_second_per_gpu:4503.772387670359 mfu:21.24506277544767 +Step 4616 | loss:1.219115138053894 lr:9.618089807677513e-05 tokens_per_second_per_gpu:4968.71035264528 mfu:23.4382544828311 +Step 4617 | loss:1.3167657852172852 lr:9.61792357672602e-05 tokens_per_second_per_gpu:4657.884490664714 mfu:21.972035859508185 +Step 4618 | loss:1.3125076293945312 lr:9.617757311042533e-05 tokens_per_second_per_gpu:3278.459724379059 mfu:15.465053882804675 +Step 4619 | loss:1.391906976699829 lr:9.617591010628304e-05 tokens_per_second_per_gpu:5357.641711592204 mfu:25.272910061516246 +Step 4620 | loss:1.3212870359420776 lr:9.617424675484585e-05 tokens_per_second_per_gpu:5189.281780415484 mfu:24.478727540989354 +Step 4621 | loss:1.3282784223556519 lr:9.617258305612625e-05 tokens_per_second_per_gpu:3629.9056165843 mfu:17.12288411918918 +Step 4622 | loss:1.248788595199585 lr:9.617091901013677e-05 tokens_per_second_per_gpu:5344.6984422879195 mfu:25.211854451858475 +Step 4623 | loss:1.2913681268692017 lr:9.61692546168899e-05 tokens_per_second_per_gpu:4117.199488965504 mfu:19.42153245611917 +Step 4624 | loss:1.2702029943466187 lr:9.616758987639819e-05 tokens_per_second_per_gpu:3883.8827746108695 mfu:18.320937706572025 +Step 4625 | loss:1.2680537700653076 lr:9.616592478867415e-05 tokens_per_second_per_gpu:5343.887668518484 mfu:25.20802989365578 +Step 4626 | loss:1.1996170282363892 lr:9.616425935373027e-05 tokens_per_second_per_gpu:3904.218266251581 mfu:18.41686369023391 +Step 4627 | loss:1.3249019384384155 lr:9.616259357157914e-05 tokens_per_second_per_gpu:4554.680475556556 mfu:21.485204911822308 +Step 4628 | loss:1.3120733499526978 lr:9.616092744223324e-05 tokens_per_second_per_gpu:3673.570949068453 mfu:17.32886094259088 +Step 4629 | loss:1.2426015138626099 lr:9.615926096570511e-05 tokens_per_second_per_gpu:5330.763052470215 mfu:25.14611884046569 +Step 4630 | loss:1.2178330421447754 lr:9.615759414200729e-05 tokens_per_second_per_gpu:4504.103959402547 mfu:21.246626855880372 +Step 4631 | loss:1.204998254776001 lr:9.615592697115232e-05 tokens_per_second_per_gpu:4756.883021648137 mfu:22.43902882104812 +Step 4632 | loss:1.2262221574783325 lr:9.615425945315273e-05 tokens_per_second_per_gpu:4465.3369561269 mfu:21.063756287096336 +Step 4633 | loss:1.4522058963775635 lr:9.615259158802105e-05 tokens_per_second_per_gpu:5279.42329661756 mfu:24.90394045264312 +Step 4634 | loss:1.3967676162719727 lr:9.615092337576988e-05 tokens_per_second_per_gpu:5276.04001691334 mfu:24.887980945031273 +Step 4635 | loss:1.4672695398330688 lr:9.614925481641168e-05 tokens_per_second_per_gpu:3681.0866080438905 mfu:17.364313588281668 +Step 4636 | loss:1.3082703351974487 lr:9.614758590995907e-05 tokens_per_second_per_gpu:5333.051667316832 mfu:25.156914627175137 +Step 4637 | loss:1.2699558734893799 lr:9.614591665642459e-05 tokens_per_second_per_gpu:3795.7526857067296 mfu:17.90521304066751 +Step 4638 | loss:1.306145429611206 lr:9.614424705582075e-05 tokens_per_second_per_gpu:4405.4448094318695 mfu:20.781234812481937 +Step 4639 | loss:1.239027738571167 lr:9.614257710816017e-05 tokens_per_second_per_gpu:4546.639552589477 mfu:21.447274506263668 +Step 4640 | loss:1.347747802734375 lr:9.614090681345535e-05 tokens_per_second_per_gpu:4787.66248366682 mfu:22.58422079490798 +Step 4641 | loss:1.1898548603057861 lr:9.613923617171892e-05 tokens_per_second_per_gpu:5040.324917899108 mfu:23.776072605838188 +Step 4642 | loss:1.1876193284988403 lr:9.613756518296337e-05 tokens_per_second_per_gpu:5366.026435935875 mfu:25.312462236826835 +Step 4643 | loss:1.2841453552246094 lr:9.613589384720133e-05 tokens_per_second_per_gpu:5308.448262678014 mfu:25.04085616972014 +Step 4644 | loss:1.132686972618103 lr:9.613422216444534e-05 tokens_per_second_per_gpu:4591.271730446355 mfu:21.65781210424934 +Step 4645 | loss:1.1108391284942627 lr:9.613255013470796e-05 tokens_per_second_per_gpu:5317.580135856267 mfu:25.083932773560186 +Step 4646 | loss:1.414383053779602 lr:9.613087775800181e-05 tokens_per_second_per_gpu:5314.929677712121 mfu:25.071430110279458 +Step 4647 | loss:1.343973994255066 lr:9.612920503433943e-05 tokens_per_second_per_gpu:5327.959344543335 mfu:25.132893271812694 +Step 4648 | loss:1.2515164613723755 lr:9.612753196373342e-05 tokens_per_second_per_gpu:3343.602400089026 mfu:15.772342998615157 +Step 4649 | loss:1.3662186861038208 lr:9.612585854619634e-05 tokens_per_second_per_gpu:3604.6594589435094 mfu:17.00379368615891 +Step 4650 | loss:1.1967753171920776 lr:9.612418478174082e-05 tokens_per_second_per_gpu:4485.665410824584 mfu:21.15964907629733 +Step 4651 | loss:1.1698509454727173 lr:9.61225106703794e-05 tokens_per_second_per_gpu:4154.154315479966 mfu:19.595854677931452 +Step 4652 | loss:1.319189190864563 lr:9.612083621212469e-05 tokens_per_second_per_gpu:4713.717846566145 mfu:22.235411325448066 +Step 4653 | loss:1.23470938205719 lr:9.611916140698929e-05 tokens_per_second_per_gpu:4730.156126754807 mfu:22.31295349775779 +Step 4654 | loss:1.2987316846847534 lr:9.61174862549858e-05 tokens_per_second_per_gpu:3891.0793829326367 mfu:18.35488533589379 +Step 4655 | loss:1.3394538164138794 lr:9.61158107561268e-05 tokens_per_second_per_gpu:4957.1786112158825 mfu:23.383857290990658 +Step 4656 | loss:1.2457642555236816 lr:9.611413491042491e-05 tokens_per_second_per_gpu:4735.430155802641 mfu:22.337831992618565 +Step 4657 | loss:1.50356125831604 lr:9.611245871789272e-05 tokens_per_second_per_gpu:3931.4778017544813 mfu:18.545451570157436 +Step 4658 | loss:1.3388864994049072 lr:9.611078217854284e-05 tokens_per_second_per_gpu:4391.391598309491 mfu:20.714943417892616 +Step 4659 | loss:1.3142521381378174 lr:9.61091052923879e-05 tokens_per_second_per_gpu:3752.7220995865164 mfu:17.70223042416341 +Step 4660 | loss:1.294373869895935 lr:9.610742805944048e-05 tokens_per_second_per_gpu:4079.450214165169 mfu:19.243462686195564 +Step 4661 | loss:1.2975738048553467 lr:9.610575047971322e-05 tokens_per_second_per_gpu:4287.628316943438 mfu:20.225474315847578 +Step 4662 | loss:1.3124836683273315 lr:9.610407255321874e-05 tokens_per_second_per_gpu:5346.805188808145 mfu:25.2217923346424 +Step 4663 | loss:1.2645792961120605 lr:9.610239427996962e-05 tokens_per_second_per_gpu:4583.528665395884 mfu:21.62128674530314 +Step 4664 | loss:1.306638240814209 lr:9.610071565997852e-05 tokens_per_second_per_gpu:4137.327082541037 mfu:19.516477749136538 +Step 4665 | loss:1.3702654838562012 lr:9.609903669325805e-05 tokens_per_second_per_gpu:4629.101789630203 mfu:21.836262947893214 +Step 4666 | loss:1.354661226272583 lr:9.609735737982086e-05 tokens_per_second_per_gpu:5324.468419334604 mfu:25.116425981989853 +Step 4667 | loss:1.2047815322875977 lr:9.609567771967955e-05 tokens_per_second_per_gpu:5323.131653884217 mfu:25.11012023128467 +Step 4668 | loss:1.2729703187942505 lr:9.609399771284678e-05 tokens_per_second_per_gpu:5338.474328190116 mfu:25.182494243715766 +Step 4669 | loss:1.2422633171081543 lr:9.609231735933516e-05 tokens_per_second_per_gpu:4992.6810366931495 mfu:23.55132828930532 +Step 4670 | loss:1.5498018264770508 lr:9.609063665915734e-05 tokens_per_second_per_gpu:5123.034897566014 mfu:24.16622969170467 +Step 4671 | loss:1.4093656539916992 lr:9.608895561232596e-05 tokens_per_second_per_gpu:4251.741018968516 mfu:20.056187808295455 +Step 4672 | loss:1.3143430948257446 lr:9.608727421885367e-05 tokens_per_second_per_gpu:5363.110466723308 mfu:25.298707112535503 +Step 4673 | loss:1.2935066223144531 lr:9.608559247875311e-05 tokens_per_second_per_gpu:5313.180680852597 mfu:25.06317979368345 +Step 4674 | loss:1.3896769285202026 lr:9.608391039203691e-05 tokens_per_second_per_gpu:3646.5687428471783 mfu:17.20148692879368 +Step 4675 | loss:1.2760698795318604 lr:9.608222795871776e-05 tokens_per_second_per_gpu:5147.924400857419 mfu:24.283637725317387 +Step 4676 | loss:1.3252018690109253 lr:9.60805451788083e-05 tokens_per_second_per_gpu:3646.7188019225873 mfu:17.202194783055106 +Step 4677 | loss:1.3802821636199951 lr:9.607886205232115e-05 tokens_per_second_per_gpu:5111.173202953273 mfu:24.110276054402583 +Step 4678 | loss:1.2422587871551514 lr:9.607717857926902e-05 tokens_per_second_per_gpu:4024.286505707561 mfu:18.98324606150184 +Step 4679 | loss:1.3626477718353271 lr:9.607549475966453e-05 tokens_per_second_per_gpu:3591.629958395357 mfu:16.942331309011244 +Step 4680 | loss:1.2580926418304443 lr:9.607381059352038e-05 tokens_per_second_per_gpu:4502.000584704345 mfu:21.23670487855622 +Step 4681 | loss:1.296704649925232 lr:9.607212608084924e-05 tokens_per_second_per_gpu:3920.6151303002544 mfu:18.494210495544984 +Step 4682 | loss:1.3187143802642822 lr:9.607044122166373e-05 tokens_per_second_per_gpu:5320.178884275467 mfu:25.096191513245756 +Step 4683 | loss:1.192511796951294 lr:9.606875601597656e-05 tokens_per_second_per_gpu:4213.222348169546 mfu:19.874488666173797 +Step 4684 | loss:1.255070686340332 lr:9.606707046380042e-05 tokens_per_second_per_gpu:5317.551072058102 mfu:25.083795674665453 +Step 4685 | loss:1.0871440172195435 lr:9.606538456514794e-05 tokens_per_second_per_gpu:4262.304067066338 mfu:20.106015508414725 +Step 4686 | loss:1.3957852125167847 lr:9.606369832003184e-05 tokens_per_second_per_gpu:4209.53980309859 mfu:19.85711747276712 +Step 4687 | loss:1.2951130867004395 lr:9.606201172846476e-05 tokens_per_second_per_gpu:3397.231356403695 mfu:16.02531993559492 +Step 4688 | loss:1.3114969730377197 lr:9.606032479045943e-05 tokens_per_second_per_gpu:5171.340164104741 mfu:24.394093875696182 +Step 4689 | loss:1.1193181276321411 lr:9.605863750602852e-05 tokens_per_second_per_gpu:5344.384808522881 mfu:25.21037498787735 +Step 4690 | loss:1.2688624858856201 lr:9.605694987518474e-05 tokens_per_second_per_gpu:4784.406493257914 mfu:22.568861732619922 +Step 4691 | loss:1.1537286043167114 lr:9.605526189794074e-05 tokens_per_second_per_gpu:4443.682830275278 mfu:20.961610080879392 +Step 4692 | loss:1.3081984519958496 lr:9.605357357430925e-05 tokens_per_second_per_gpu:5203.633130528302 mfu:24.54642530806434 +Step 4693 | loss:1.1961005926132202 lr:9.605188490430294e-05 tokens_per_second_per_gpu:5356.433797927235 mfu:25.267212126667268 +Step 4694 | loss:1.2221015691757202 lr:9.605019588793454e-05 tokens_per_second_per_gpu:4744.34449527184 mfu:22.379882452837926 +Step 4695 | loss:1.3875092267990112 lr:9.604850652521673e-05 tokens_per_second_per_gpu:4215.075230879042 mfu:19.883229030048874 +Step 4696 | loss:1.4691698551177979 lr:9.604681681616224e-05 tokens_per_second_per_gpu:4114.214674512156 mfu:19.407452577080864 +Step 4697 | loss:1.2819833755493164 lr:9.604512676078376e-05 tokens_per_second_per_gpu:4088.354799348482 mfu:19.285467133784145 +Step 4698 | loss:1.39738130569458 lr:9.6043436359094e-05 tokens_per_second_per_gpu:3466.60331016127 mfu:16.352559277545208 +Step 4699 | loss:1.1151683330535889 lr:9.604174561110569e-05 tokens_per_second_per_gpu:5339.5941955928665 mfu:25.187776849323622 +Step 4700 | loss:1.570248007774353 lr:9.604005451683153e-05 tokens_per_second_per_gpu:5248.319486757742 mfu:24.757218474677174 +Step 4701 | loss:1.2807517051696777 lr:9.603836307628425e-05 tokens_per_second_per_gpu:5338.897890859586 mfu:25.184492261094903 +Step 4702 | loss:1.1975291967391968 lr:9.603667128947657e-05 tokens_per_second_per_gpu:3819.2901513220113 mfu:18.016243281880534 +Step 4703 | loss:1.435559868812561 lr:9.603497915642121e-05 tokens_per_second_per_gpu:4680.766064911514 mfu:22.07997214922948 +Step 4704 | loss:1.2414299249649048 lr:9.60332866771309e-05 tokens_per_second_per_gpu:5324.032090720431 mfu:25.114367745471462 +Step 4705 | loss:1.3691301345825195 lr:9.603159385161835e-05 tokens_per_second_per_gpu:4545.207790366103 mfu:21.44052063957219 +Step 4706 | loss:1.4912835359573364 lr:9.602990067989632e-05 tokens_per_second_per_gpu:5321.056892548777 mfu:25.100333228074618 +Step 4707 | loss:1.212082028388977 lr:9.602820716197754e-05 tokens_per_second_per_gpu:5195.130089371719 mfu:24.506315012160314 +Step 4708 | loss:1.2607192993164062 lr:9.602651329787474e-05 tokens_per_second_per_gpu:5320.139849415769 mfu:25.096007379152525 +Step 4709 | loss:1.4172066450119019 lr:9.602481908760064e-05 tokens_per_second_per_gpu:5367.922261031627 mfu:25.321405167264615 +Step 4710 | loss:1.2580302953720093 lr:9.602312453116802e-05 tokens_per_second_per_gpu:3716.4562517148397 mfu:17.531158232161022 +Step 4711 | loss:1.3584489822387695 lr:9.60214296285896e-05 tokens_per_second_per_gpu:3744.909315299982 mfu:17.665376187685375 +Step 4712 | loss:1.4401904344558716 lr:9.601973437987815e-05 tokens_per_second_per_gpu:4027.925671474825 mfu:19.00041262733173 +Step 4713 | loss:1.2623130083084106 lr:9.601803878504638e-05 tokens_per_second_per_gpu:4122.049061061511 mfu:19.44440871511835 +Step 4714 | loss:1.1417796611785889 lr:9.601634284410709e-05 tokens_per_second_per_gpu:4994.048330700667 mfu:23.5577780484232 +Step 4715 | loss:1.1473745107650757 lr:9.601464655707301e-05 tokens_per_second_per_gpu:4330.605156827824 mfu:20.42820340218723 +Step 4716 | loss:1.256861925125122 lr:9.601294992395689e-05 tokens_per_second_per_gpu:3725.038053572354 mfu:17.571640055729297 +Step 4717 | loss:1.1522331237792969 lr:9.601125294477152e-05 tokens_per_second_per_gpu:5203.5357946091135 mfu:24.545966157926305 +Step 4718 | loss:1.3409656286239624 lr:9.600955561952963e-05 tokens_per_second_per_gpu:3792.741824482123 mfu:17.89101029456514 +Step 4719 | loss:1.2819966077804565 lr:9.6007857948244e-05 tokens_per_second_per_gpu:3728.1902573356183 mfu:17.586509538702035 +Step 4720 | loss:1.3893938064575195 lr:9.600615993092742e-05 tokens_per_second_per_gpu:4567.879345799036 mfu:21.547466234715564 +Step 4721 | loss:1.4779026508331299 lr:9.600446156759263e-05 tokens_per_second_per_gpu:3718.9820375972513 mfu:17.543072795111865 +Step 4722 | loss:1.5738776922225952 lr:9.600276285825242e-05 tokens_per_second_per_gpu:3630.152474348422 mfu:17.124048589380823 +Step 4723 | loss:1.3605362176895142 lr:9.600106380291955e-05 tokens_per_second_per_gpu:4163.579126902304 mfu:19.640313121449267 +Step 4724 | loss:1.3322886228561401 lr:9.599936440160682e-05 tokens_per_second_per_gpu:5328.512768109995 mfu:25.13550386520043 +Step 4725 | loss:1.2472871541976929 lr:9.599766465432697e-05 tokens_per_second_per_gpu:5344.881552242748 mfu:25.21271821275718 +Step 4726 | loss:1.267042636871338 lr:9.599596456109285e-05 tokens_per_second_per_gpu:5362.595942281118 mfu:25.296280012954774 +Step 4727 | loss:1.3526054620742798 lr:9.599426412191721e-05 tokens_per_second_per_gpu:5350.234242595135 mfu:25.237967766412538 +Step 4728 | loss:1.2648833990097046 lr:9.599256333681282e-05 tokens_per_second_per_gpu:4955.234176446895 mfu:23.37468506043056 +Step 4729 | loss:1.3419264554977417 lr:9.599086220579251e-05 tokens_per_second_per_gpu:4102.060471936099 mfu:19.35011912980881 +Step 4730 | loss:1.3673008680343628 lr:9.598916072886905e-05 tokens_per_second_per_gpu:4106.979329303667 mfu:19.373322219255236 +Step 4731 | loss:1.0206563472747803 lr:9.598745890605523e-05 tokens_per_second_per_gpu:4716.288140853796 mfu:22.247535842140337 +Step 4732 | loss:1.3779046535491943 lr:9.59857567373639e-05 tokens_per_second_per_gpu:4387.0297282432575 mfu:20.694367732578417 +Step 4733 | loss:1.235617995262146 lr:9.598405422280778e-05 tokens_per_second_per_gpu:5217.024147832182 mfu:24.609593021429887 +Step 4734 | loss:1.1742759943008423 lr:9.598235136239976e-05 tokens_per_second_per_gpu:4806.658425290059 mfu:22.673827892586647 +Step 4735 | loss:1.1346876621246338 lr:9.598064815615258e-05 tokens_per_second_per_gpu:4642.674119384891 mfu:21.90028594302432 +Step 4736 | loss:1.3279749155044556 lr:9.59789446040791e-05 tokens_per_second_per_gpu:4879.216032134276 mfu:23.01609450367509 +Step 4737 | loss:1.2339496612548828 lr:9.59772407061921e-05 tokens_per_second_per_gpu:4107.151956578626 mfu:19.374136531564464 +Step 4738 | loss:1.5445442199707031 lr:9.597553646250441e-05 tokens_per_second_per_gpu:5339.922698028799 mfu:25.189326451362213 +Step 4739 | loss:1.2421298027038574 lr:9.597383187302885e-05 tokens_per_second_per_gpu:5348.114041718761 mfu:25.22796641713511 +Step 4740 | loss:1.2305748462677002 lr:9.597212693777821e-05 tokens_per_second_per_gpu:3359.5119200699382 mfu:15.847390918809133 +Step 4741 | loss:1.5216625928878784 lr:9.597042165676535e-05 tokens_per_second_per_gpu:5315.028745690285 mfu:25.071897430835204 +Step 4742 | loss:1.3773314952850342 lr:9.596871603000308e-05 tokens_per_second_per_gpu:4167.627951319804 mfu:19.65941211702729 +Step 4743 | loss:1.2609302997589111 lr:9.596701005750424e-05 tokens_per_second_per_gpu:4491.852032255993 mfu:21.188832425135754 +Step 4744 | loss:1.4129142761230469 lr:9.596530373928166e-05 tokens_per_second_per_gpu:4788.337684276372 mfu:22.58740583138434 +Step 4745 | loss:1.364784836769104 lr:9.596359707534815e-05 tokens_per_second_per_gpu:3255.687218175861 mfu:15.357632085653984 +Step 4746 | loss:1.352891445159912 lr:9.596189006571655e-05 tokens_per_second_per_gpu:5349.052753364149 mfu:25.232394480127127 +Step 4747 | loss:1.3347676992416382 lr:9.596018271039973e-05 tokens_per_second_per_gpu:4212.345998025262 mfu:19.870354773022456 +Step 4748 | loss:1.292605996131897 lr:9.595847500941051e-05 tokens_per_second_per_gpu:4808.159761395136 mfu:22.680909951148646 +Step 4749 | loss:1.2705931663513184 lr:9.595676696276172e-05 tokens_per_second_per_gpu:5347.131770694753 mfu:25.223332877122147 +Step 4750 | loss:1.3733388185501099 lr:9.595505857046624e-05 tokens_per_second_per_gpu:4824.272740135658 mfu:22.756917621025348 +Step 4751 | loss:1.196567177772522 lr:9.59533498325369e-05 tokens_per_second_per_gpu:4663.801062892861 mfu:21.999945340179675 +Step 4752 | loss:1.3575308322906494 lr:9.595164074898653e-05 tokens_per_second_per_gpu:3520.9222026740044 mfu:16.608790761286503 +Step 4753 | loss:1.2938151359558105 lr:9.594993131982803e-05 tokens_per_second_per_gpu:3883.0943397734736 mfu:18.31721852492317 +Step 4754 | loss:1.2816591262817383 lr:9.594822154507422e-05 tokens_per_second_per_gpu:5350.784550218777 mfu:25.240563661365425 +Step 4755 | loss:1.4044381380081177 lr:9.594651142473799e-05 tokens_per_second_per_gpu:4551.8932229983375 mfu:21.472056965950866 +Step 4756 | loss:1.195671796798706 lr:9.594480095883217e-05 tokens_per_second_per_gpu:5358.747277822715 mfu:25.278125206054618 +Step 4757 | loss:1.2797234058380127 lr:9.594309014736964e-05 tokens_per_second_per_gpu:3915.074982370684 mfu:18.468076672514496 +Step 4758 | loss:1.2915724515914917 lr:9.594137899036326e-05 tokens_per_second_per_gpu:5353.9931564353155 mfu:25.255699204333503 +Step 4759 | loss:1.3239305019378662 lr:9.593966748782592e-05 tokens_per_second_per_gpu:4753.264792932099 mfu:22.421961018861087 +Step 4760 | loss:1.3282346725463867 lr:9.593795563977047e-05 tokens_per_second_per_gpu:3419.836439232717 mfu:16.131951968124252 +Step 4761 | loss:1.138810157775879 lr:9.593624344620979e-05 tokens_per_second_per_gpu:5062.944347967208 mfu:23.88277231674214 +Step 4762 | loss:1.5508112907409668 lr:9.593453090715679e-05 tokens_per_second_per_gpu:3929.82817591098 mfu:18.537670003598446 +Step 4763 | loss:1.4626660346984863 lr:9.593281802262429e-05 tokens_per_second_per_gpu:4921.053862047032 mfu:23.213450685643423 +Step 4764 | loss:1.1792174577713013 lr:9.593110479262522e-05 tokens_per_second_per_gpu:4770.327729979409 mfu:22.502449804150505 +Step 4765 | loss:1.4889551401138306 lr:9.592939121717244e-05 tokens_per_second_per_gpu:5365.690203682237 mfu:25.310876171919283 +Step 4766 | loss:1.3602720499038696 lr:9.592767729627885e-05 tokens_per_second_per_gpu:3588.0334562245384 mfu:16.92536599464495 +Step 4767 | loss:1.3673946857452393 lr:9.592596302995736e-05 tokens_per_second_per_gpu:4066.5881898791645 mfu:19.182790323150847 +Step 4768 | loss:1.2760539054870605 lr:9.592424841822081e-05 tokens_per_second_per_gpu:5387.916031074228 mfu:25.415719191844495 +Step 4769 | loss:1.4474190473556519 lr:9.592253346108215e-05 tokens_per_second_per_gpu:4299.6321580230015 mfu:20.282098482286358 +Step 4770 | loss:1.3642667531967163 lr:9.592081815855426e-05 tokens_per_second_per_gpu:5367.985836002699 mfu:25.321705061250167 +Step 4771 | loss:1.3172792196273804 lr:9.591910251065001e-05 tokens_per_second_per_gpu:5354.852522328375 mfu:25.259752979874005 +Step 4772 | loss:1.5265535116195679 lr:9.591738651738233e-05 tokens_per_second_per_gpu:5035.573100770844 mfu:23.753657473700564 +Step 4773 | loss:1.3319618701934814 lr:9.591567017876415e-05 tokens_per_second_per_gpu:4836.195801088647 mfu:22.813160733824567 +Step 4774 | loss:1.0851479768753052 lr:9.591395349480835e-05 tokens_per_second_per_gpu:5211.639305042999 mfu:24.584191799244394 +Step 4775 | loss:1.342628002166748 lr:9.591223646552785e-05 tokens_per_second_per_gpu:3312.5978553580867 mfu:15.626089271198506 +Step 4776 | loss:1.365940809249878 lr:9.591051909093555e-05 tokens_per_second_per_gpu:4545.116905568312 mfu:21.44009192047446 +Step 4777 | loss:1.1103209257125854 lr:9.590880137104438e-05 tokens_per_second_per_gpu:5339.070868426532 mfu:25.185308225791204 +Step 4778 | loss:1.409409523010254 lr:9.590708330586727e-05 tokens_per_second_per_gpu:3624.296362617167 mfu:17.0964243111888 +Step 4779 | loss:1.2838387489318848 lr:9.590536489541712e-05 tokens_per_second_per_gpu:4138.219251419622 mfu:19.52068626195653 +Step 4780 | loss:1.230956792831421 lr:9.590364613970686e-05 tokens_per_second_per_gpu:5355.035596816603 mfu:25.260616573470706 +Step 4781 | loss:1.376744270324707 lr:9.590192703874942e-05 tokens_per_second_per_gpu:3807.2980566908445 mfu:17.959674525442992 +Step 4782 | loss:1.397676944732666 lr:9.590020759255773e-05 tokens_per_second_per_gpu:4996.017728729047 mfu:23.56706803493702 +Step 4783 | loss:1.2470226287841797 lr:9.589848780114473e-05 tokens_per_second_per_gpu:5093.372731268869 mfu:24.02630819239361 +Step 4784 | loss:1.2444953918457031 lr:9.589676766452333e-05 tokens_per_second_per_gpu:5352.990408681866 mfu:25.250969072094076 +Step 4785 | loss:1.2155718803405762 lr:9.58950471827065e-05 tokens_per_second_per_gpu:4496.12450270209 mfu:21.208986397188614 +Step 4786 | loss:1.2046494483947754 lr:9.589332635570716e-05 tokens_per_second_per_gpu:5374.996473243077 mfu:25.354775433251085 +Step 4787 | loss:1.3183265924453735 lr:9.589160518353826e-05 tokens_per_second_per_gpu:3723.781291211228 mfu:17.565711693246097 +Step 4788 | loss:1.3423373699188232 lr:9.588988366621273e-05 tokens_per_second_per_gpu:3278.6406073937005 mfu:15.465907138846564 +Step 4789 | loss:1.231278419494629 lr:9.588816180374354e-05 tokens_per_second_per_gpu:4761.81097505425 mfu:22.462274818060298 +Step 4790 | loss:1.6567853689193726 lr:9.588643959614364e-05 tokens_per_second_per_gpu:5367.682370557099 mfu:25.320273562966726 +Step 4791 | loss:1.1988205909729004 lr:9.588471704342594e-05 tokens_per_second_per_gpu:5342.088426193602 mfu:25.199542560626917 +Step 4792 | loss:1.4647506475448608 lr:9.588299414560346e-05 tokens_per_second_per_gpu:5341.238307249437 mfu:25.195532404522037 +Step 4793 | loss:1.2031431198120117 lr:9.58812709026891e-05 tokens_per_second_per_gpu:5330.3223387706885 mfu:25.14403991875151 +Step 4794 | loss:1.3279294967651367 lr:9.587954731469586e-05 tokens_per_second_per_gpu:3858.3294646885474 mfu:18.200398383824183 +Step 4795 | loss:1.1277767419815063 lr:9.587782338163669e-05 tokens_per_second_per_gpu:4628.449742124669 mfu:21.83318712855883 +Step 4796 | loss:1.4259517192840576 lr:9.587609910352457e-05 tokens_per_second_per_gpu:5240.20431810898 mfu:24.718937839570415 +Step 4797 | loss:1.5485864877700806 lr:9.587437448037244e-05 tokens_per_second_per_gpu:3841.8828565457816 mfu:18.122816927134053 +Step 4798 | loss:1.366048812866211 lr:9.587264951219328e-05 tokens_per_second_per_gpu:3814.499046342073 mfu:17.993642822243363 +Step 4799 | loss:1.3178201913833618 lr:9.587092419900007e-05 tokens_per_second_per_gpu:5120.094075833336 mfu:24.15235733383571 +Step 4800 | loss:1.4300239086151123 lr:9.58691985408058e-05 tokens_per_second_per_gpu:5282.539920667773 mfu:24.918642100038866 +Step 4801 | loss:1.2226392030715942 lr:9.586747253762343e-05 tokens_per_second_per_gpu:3557.868433454722 mfu:16.783072435556132 +Step 4802 | loss:1.192613124847412 lr:9.586574618946595e-05 tokens_per_second_per_gpu:5348.054194707433 mfu:25.227684108571726 +Step 4803 | loss:1.1786407232284546 lr:9.586401949634634e-05 tokens_per_second_per_gpu:5323.395518919438 mfu:25.111364927675826 +Step 4804 | loss:1.3125685453414917 lr:9.586229245827756e-05 tokens_per_second_per_gpu:5336.68895208629 mfu:25.174072320017444 +Step 4805 | loss:1.2655911445617676 lr:9.586056507527266e-05 tokens_per_second_per_gpu:5204.374974750749 mfu:24.5499247138333 +Step 4806 | loss:1.3409847021102905 lr:9.585883734734458e-05 tokens_per_second_per_gpu:5299.064939701673 mfu:24.996593434280758 +Step 4807 | loss:1.3550593852996826 lr:9.585710927450634e-05 tokens_per_second_per_gpu:5338.036617912193 mfu:25.18042948965373 +Step 4808 | loss:1.425782322883606 lr:9.585538085677092e-05 tokens_per_second_per_gpu:5058.697538034429 mfu:23.862739389708732 +Step 4809 | loss:1.2685695886611938 lr:9.585365209415135e-05 tokens_per_second_per_gpu:4654.308809987654 mfu:21.955168763680465 +Step 4810 | loss:1.3182631731033325 lr:9.585192298666059e-05 tokens_per_second_per_gpu:3451.621525132062 mfu:16.281887641406325 +Step 4811 | loss:1.1508522033691406 lr:9.585019353431168e-05 tokens_per_second_per_gpu:5358.513661432707 mfu:25.27702319768388 +Step 4812 | loss:1.3649394512176514 lr:9.584846373711762e-05 tokens_per_second_per_gpu:4552.684861741203 mfu:21.47579126096846 +Step 4813 | loss:1.3016289472579956 lr:9.58467335950914e-05 tokens_per_second_per_gpu:5100.311483024362 mfu:24.059039468297396 +Step 4814 | loss:1.3669284582138062 lr:9.584500310824604e-05 tokens_per_second_per_gpu:4213.1772010136065 mfu:19.874275699336327 +Step 4815 | loss:1.3717665672302246 lr:9.584327227659458e-05 tokens_per_second_per_gpu:5173.992059548639 mfu:24.406603318965068 +Step 4816 | loss:1.134092092514038 lr:9.584154110015002e-05 tokens_per_second_per_gpu:4531.791770270038 mfu:21.377234983769984 +Step 4817 | loss:1.3822991847991943 lr:9.583980957892537e-05 tokens_per_second_per_gpu:4107.559521327672 mfu:19.376059084023435 +Step 4818 | loss:1.4167780876159668 lr:9.583807771293365e-05 tokens_per_second_per_gpu:5290.562432551361 mfu:24.956485657375424 +Step 4819 | loss:1.302032470703125 lr:9.583634550218792e-05 tokens_per_second_per_gpu:4516.966279924171 mfu:21.30730061631905 +Step 4820 | loss:1.3279078006744385 lr:9.583461294670117e-05 tokens_per_second_per_gpu:3725.496708164419 mfu:17.57380360769461 +Step 4821 | loss:1.3295788764953613 lr:9.583288004648648e-05 tokens_per_second_per_gpu:4724.068921048834 mfu:22.284239109858326 +Step 4822 | loss:1.4026014804840088 lr:9.583114680155682e-05 tokens_per_second_per_gpu:5006.372841069307 mfu:23.61591486661037 +Step 4823 | loss:1.2253080606460571 lr:9.582941321192527e-05 tokens_per_second_per_gpu:5345.045469385552 mfu:25.213491437886937 +Step 4824 | loss:1.4223703145980835 lr:9.582767927760484e-05 tokens_per_second_per_gpu:4459.249221764487 mfu:21.035039405434244 +Step 4825 | loss:1.3962820768356323 lr:9.582594499860859e-05 tokens_per_second_per_gpu:5338.262789119494 mfu:25.181496377827358 +Step 4826 | loss:1.3069342374801636 lr:9.582421037494958e-05 tokens_per_second_per_gpu:4746.902225542807 mfu:22.391947702919857 +Step 4827 | loss:1.3111937046051025 lr:9.582247540664082e-05 tokens_per_second_per_gpu:4904.654809023765 mfu:23.136093554565704 +Step 4828 | loss:1.3351092338562012 lr:9.582074009369537e-05 tokens_per_second_per_gpu:5202.54758227587 mfu:24.541304591743984 +Step 4829 | loss:1.2648265361785889 lr:9.58190044361263e-05 tokens_per_second_per_gpu:4301.428391361611 mfu:20.29057162141359 +Step 4830 | loss:1.394818663597107 lr:9.581726843394666e-05 tokens_per_second_per_gpu:5126.318355706992 mfu:24.1817183239714 +Step 4831 | loss:1.3387305736541748 lr:9.581553208716947e-05 tokens_per_second_per_gpu:5315.496082577923 mfu:25.074101938670083 +Step 4832 | loss:1.257900595664978 lr:9.581379539580784e-05 tokens_per_second_per_gpu:4206.936234046798 mfu:19.844835993335128 +Step 4833 | loss:1.418221354484558 lr:9.58120583598748e-05 tokens_per_second_per_gpu:5354.236666330021 mfu:25.2568478820539 +Step 4834 | loss:1.3419326543807983 lr:9.581032097938343e-05 tokens_per_second_per_gpu:5341.612632528625 mfu:25.19729816073045 +Step 4835 | loss:1.3653383255004883 lr:9.580858325434679e-05 tokens_per_second_per_gpu:3866.0985663728334 mfu:18.237046562014395 +Step 4836 | loss:1.4256322383880615 lr:9.580684518477794e-05 tokens_per_second_per_gpu:3981.62909756006 mfu:18.782023789165752 +Step 4837 | loss:1.379689335823059 lr:9.580510677068997e-05 tokens_per_second_per_gpu:4476.2841394534735 mfu:21.11539601862986 +Step 4838 | loss:1.2844791412353516 lr:9.580336801209594e-05 tokens_per_second_per_gpu:4021.41718284805 mfu:18.96971097601656 +Step 4839 | loss:1.2705192565917969 lr:9.580162890900895e-05 tokens_per_second_per_gpu:4725.988234532286 mfu:22.293292839028414 +Step 4840 | loss:1.3134899139404297 lr:9.579988946144204e-05 tokens_per_second_per_gpu:3688.8874970447396 mfu:17.401111712667582 +Step 4841 | loss:1.2912228107452393 lr:9.579814966940833e-05 tokens_per_second_per_gpu:4122.893679056528 mfu:19.448392922307534 +Step 4842 | loss:1.3162823915481567 lr:9.57964095329209e-05 tokens_per_second_per_gpu:4400.763547989466 mfu:20.759152503552556 +Step 4843 | loss:1.3022798299789429 lr:9.579466905199281e-05 tokens_per_second_per_gpu:5355.288265548261 mfu:25.26180845498767 +Step 4844 | loss:1.3694897890090942 lr:9.579292822663719e-05 tokens_per_second_per_gpu:4138.422784141849 mfu:19.521646360533538 +Step 4845 | loss:1.3003205060958862 lr:9.579118705686711e-05 tokens_per_second_per_gpu:4558.550717534746 mfu:21.5034615035648 +Step 4846 | loss:1.4087555408477783 lr:9.578944554269566e-05 tokens_per_second_per_gpu:3089.238275579676 mfu:14.572464024309749 +Step 4847 | loss:1.2350589036941528 lr:9.578770368413597e-05 tokens_per_second_per_gpu:5381.064096042402 mfu:25.383397445238696 +Step 4848 | loss:1.3257791996002197 lr:9.57859614812011e-05 tokens_per_second_per_gpu:5333.399358337299 mfu:25.158554745040888 +Step 4849 | loss:1.3623355627059937 lr:9.578421893390417e-05 tokens_per_second_per_gpu:3851.8842831303514 mfu:18.169995362752807 +Step 4850 | loss:1.2325358390808105 lr:9.578247604225829e-05 tokens_per_second_per_gpu:5372.6298838842995 mfu:25.343611827463953 +Step 4851 | loss:1.2772387266159058 lr:9.578073280627658e-05 tokens_per_second_per_gpu:5330.73842054789 mfu:25.146002647485442 +Step 4852 | loss:1.362486720085144 lr:9.577898922597211e-05 tokens_per_second_per_gpu:5133.265653729016 mfu:24.21448991407255 +Step 4853 | loss:1.3126332759857178 lr:9.577724530135803e-05 tokens_per_second_per_gpu:5331.933606578645 mfu:25.151640543912276 +Step 4854 | loss:1.423683762550354 lr:9.577550103244746e-05 tokens_per_second_per_gpu:4117.014838966568 mfu:19.420661430569652 +Step 4855 | loss:1.3139004707336426 lr:9.57737564192535e-05 tokens_per_second_per_gpu:5355.882469603744 mfu:25.264611416899047 +Step 4856 | loss:1.5975042581558228 lr:9.577201146178927e-05 tokens_per_second_per_gpu:3520.5919018524755 mfu:16.60723267595622 +Step 4857 | loss:1.407867670059204 lr:9.57702661600679e-05 tokens_per_second_per_gpu:5049.73048853763 mfu:23.820440287295924 +Step 4858 | loss:1.3226027488708496 lr:9.576852051410252e-05 tokens_per_second_per_gpu:4572.952970795162 mfu:21.57139938947198 +Step 4859 | loss:1.2292002439498901 lr:9.576677452390628e-05 tokens_per_second_per_gpu:5340.822241914586 mfu:25.193569753349486 +Step 4860 | loss:1.3330843448638916 lr:9.576502818949227e-05 tokens_per_second_per_gpu:5338.7696231989175 mfu:25.183887200654812 +Step 4861 | loss:1.3737683296203613 lr:9.576328151087364e-05 tokens_per_second_per_gpu:4187.377977557501 mfu:19.752576360492462 +Step 4862 | loss:1.4181848764419556 lr:9.576153448806353e-05 tokens_per_second_per_gpu:5325.3267538402215 mfu:25.120474892299818 +Step 4863 | loss:1.4627918004989624 lr:9.575978712107507e-05 tokens_per_second_per_gpu:4387.849871068772 mfu:20.698236486217166 +Step 4864 | loss:1.2062219381332397 lr:9.575803940992143e-05 tokens_per_second_per_gpu:4491.8941801921055 mfu:21.189031244141574 +Step 4865 | loss:1.1967225074768066 lr:9.575629135461575e-05 tokens_per_second_per_gpu:3935.963085894606 mfu:18.56660942071479 +Step 4866 | loss:1.3000789880752563 lr:9.575454295517113e-05 tokens_per_second_per_gpu:3537.108822240642 mfu:16.68514580750441 +Step 4867 | loss:1.2608323097229004 lr:9.575279421160077e-05 tokens_per_second_per_gpu:3945.437519534793 mfu:18.611301940700617 +Step 4868 | loss:1.3464030027389526 lr:9.575104512391781e-05 tokens_per_second_per_gpu:5345.032995557578 mfu:25.213432596712128 +Step 4869 | loss:1.3452684879302979 lr:9.57492956921354e-05 tokens_per_second_per_gpu:4566.181526808911 mfu:21.539457332861353 +Step 4870 | loss:1.316631555557251 lr:9.57475459162667e-05 tokens_per_second_per_gpu:4249.112203637682 mfu:20.043787237857764 +Step 4871 | loss:1.3436782360076904 lr:9.574579579632487e-05 tokens_per_second_per_gpu:5373.435623032026 mfu:25.34741263649705 +Step 4872 | loss:1.142277479171753 lr:9.574404533232306e-05 tokens_per_second_per_gpu:4376.739936209061 mfu:20.64582902793356 +Step 4873 | loss:1.3361670970916748 lr:9.574229452427447e-05 tokens_per_second_per_gpu:3504.95384553086 mfu:16.53346529615962 +Step 4874 | loss:1.2668687105178833 lr:9.574054337219223e-05 tokens_per_second_per_gpu:4595.304405136201 mfu:21.6768349188066 +Step 4875 | loss:1.4091631174087524 lr:9.573879187608955e-05 tokens_per_second_per_gpu:3860.7969300056507 mfu:18.212037838717954 +Step 4876 | loss:1.3228427171707153 lr:9.573704003597955e-05 tokens_per_second_per_gpu:3786.710215373598 mfu:17.8625581653016 +Step 4877 | loss:1.1616286039352417 lr:9.573528785187544e-05 tokens_per_second_per_gpu:5343.081429740234 mfu:25.204226727780256 +Step 4878 | loss:1.2138375043869019 lr:9.57335353237904e-05 tokens_per_second_per_gpu:5325.117950722494 mfu:25.119489932368015 +Step 4879 | loss:1.5063450336456299 lr:9.573178245173762e-05 tokens_per_second_per_gpu:4905.772677050008 mfu:23.141366728773445 +Step 4880 | loss:1.2700138092041016 lr:9.573002923573024e-05 tokens_per_second_per_gpu:4319.962710472706 mfu:20.378001166941548 +Step 4881 | loss:1.2567030191421509 lr:9.57282756757815e-05 tokens_per_second_per_gpu:5328.145422716482 mfu:25.133771034303646 +Step 4882 | loss:1.2747021913528442 lr:9.572652177190456e-05 tokens_per_second_per_gpu:5316.839450500565 mfu:25.080438834363388 +Step 4883 | loss:1.2609813213348389 lr:9.57247675241126e-05 tokens_per_second_per_gpu:3876.0942754917214 mfu:18.284198027372025 +Step 4884 | loss:1.134188175201416 lr:9.572301293241883e-05 tokens_per_second_per_gpu:4381.679352999625 mfu:20.669129099686053 +Step 4885 | loss:1.306747555732727 lr:9.572125799683646e-05 tokens_per_second_per_gpu:4896.09616930889 mfu:23.09572099077613 +Step 4886 | loss:1.3940823078155518 lr:9.571950271737867e-05 tokens_per_second_per_gpu:4372.661903330131 mfu:20.626592251059492 +Step 4887 | loss:1.1954854726791382 lr:9.571774709405866e-05 tokens_per_second_per_gpu:5357.036830245133 mfu:25.270056732998714 +Step 4888 | loss:1.2870731353759766 lr:9.571599112688965e-05 tokens_per_second_per_gpu:4614.302536577528 mfu:21.766452346230807 +Step 4889 | loss:1.3007564544677734 lr:9.571423481588483e-05 tokens_per_second_per_gpu:5327.087124195068 mfu:25.128778859613785 +Step 4890 | loss:1.1684540510177612 lr:9.571247816105741e-05 tokens_per_second_per_gpu:3605.876586403875 mfu:17.009535084052814 +Step 4891 | loss:1.2632324695587158 lr:9.571072116242062e-05 tokens_per_second_per_gpu:5313.065754707441 mfu:25.062637667448328 +Step 4892 | loss:1.2083760499954224 lr:9.570896381998766e-05 tokens_per_second_per_gpu:3903.694259178587 mfu:18.414391859465766 +Step 4893 | loss:1.3840126991271973 lr:9.570720613377175e-05 tokens_per_second_per_gpu:3414.004434893862 mfu:16.104441408615326 +Step 4894 | loss:1.1567392349243164 lr:9.570544810378613e-05 tokens_per_second_per_gpu:4771.077628478075 mfu:22.50598720331459 +Step 4895 | loss:1.374629020690918 lr:9.570368973004398e-05 tokens_per_second_per_gpu:4194.394423709375 mfu:19.785674134120004 +Step 4896 | loss:1.2799369096755981 lr:9.570193101255855e-05 tokens_per_second_per_gpu:3743.7104238424845 mfu:17.659720812128384 +Step 4897 | loss:1.2295249700546265 lr:9.570017195134307e-05 tokens_per_second_per_gpu:5362.942738895349 mfu:25.2979159117537 +Step 4898 | loss:1.2903482913970947 lr:9.569841254641076e-05 tokens_per_second_per_gpu:5321.372862962684 mfu:25.101823714427127 +Step 4899 | loss:1.2659143209457397 lr:9.569665279777486e-05 tokens_per_second_per_gpu:4667.073271076755 mfu:22.01538090447939 +Step 4900 | loss:1.3205033540725708 lr:9.569489270544861e-05 tokens_per_second_per_gpu:5322.308232078143 mfu:25.10623601013518 +Step 4901 | loss:1.2648899555206299 lr:9.569313226944523e-05 tokens_per_second_per_gpu:5312.302144838423 mfu:25.059035589410893 +Step 4902 | loss:1.400564193725586 lr:9.569137148977797e-05 tokens_per_second_per_gpu:5323.260798095445 mfu:25.110729426563314 +Step 4903 | loss:1.3559949398040771 lr:9.568961036646009e-05 tokens_per_second_per_gpu:3256.7832216090737 mfu:15.362802120845917 +Step 4904 | loss:1.4823174476623535 lr:9.568784889950482e-05 tokens_per_second_per_gpu:3639.5902962514288 mfu:17.168568405555767 +Step 4905 | loss:1.1578994989395142 lr:9.568608708892539e-05 tokens_per_second_per_gpu:4662.0056128077285 mfu:21.991475895792732 +Step 4906 | loss:1.2711955308914185 lr:9.56843249347351e-05 tokens_per_second_per_gpu:5336.669725924322 mfu:25.173981626931123 +Step 4907 | loss:1.39446222782135 lr:9.568256243694714e-05 tokens_per_second_per_gpu:4372.488370939744 mfu:20.625773669166353 +Step 4908 | loss:1.218764066696167 lr:9.568079959557482e-05 tokens_per_second_per_gpu:5128.362115301118 mfu:24.191359086678233 +Step 4909 | loss:1.2483339309692383 lr:9.567903641063136e-05 tokens_per_second_per_gpu:4658.453566095966 mfu:21.974720285411685 +Step 4910 | loss:1.2277497053146362 lr:9.567727288213005e-05 tokens_per_second_per_gpu:4684.88492634167 mfu:22.099401521345747 +Step 4911 | loss:1.331480860710144 lr:9.567550901008414e-05 tokens_per_second_per_gpu:4470.024143371896 mfu:21.08586654904759 +Step 4912 | loss:1.2644120454788208 lr:9.56737447945069e-05 tokens_per_second_per_gpu:5331.882403363546 mfu:25.151399009610483 +Step 4913 | loss:1.5320501327514648 lr:9.56719802354116e-05 tokens_per_second_per_gpu:4554.047553917729 mfu:21.48221931246439 +Step 4914 | loss:1.3284388780593872 lr:9.56702153328115e-05 tokens_per_second_per_gpu:4988.951887550196 mfu:23.53373725653956 +Step 4915 | loss:1.4228640794754028 lr:9.56684500867199e-05 tokens_per_second_per_gpu:2975.5604194748544 mfu:14.036226181621057 +Step 4916 | loss:1.2893390655517578 lr:9.566668449715004e-05 tokens_per_second_per_gpu:3797.0029270310665 mfu:17.911110642307573 +Step 4917 | loss:1.1868315935134888 lr:9.566491856411523e-05 tokens_per_second_per_gpu:4674.569801394464 mfu:22.050743317027983 +Step 4918 | loss:1.3476186990737915 lr:9.566315228762872e-05 tokens_per_second_per_gpu:3005.510108804117 mfu:14.17750397613111 +Step 4919 | loss:1.2484211921691895 lr:9.566138566770385e-05 tokens_per_second_per_gpu:4681.296023779254 mfu:22.082472055628916 +Step 4920 | loss:1.2993370294570923 lr:9.565961870435385e-05 tokens_per_second_per_gpu:3830.492101945688 mfu:18.069084794222274 +Step 4921 | loss:1.2411105632781982 lr:9.565785139759202e-05 tokens_per_second_per_gpu:4331.952838035794 mfu:20.43456064438335 +Step 4922 | loss:1.2938040494918823 lr:9.565608374743169e-05 tokens_per_second_per_gpu:5120.364505486585 mfu:24.153632996650135 +Step 4923 | loss:1.285014271736145 lr:9.565431575388612e-05 tokens_per_second_per_gpu:5326.073996804298 mfu:25.123999764854165 +Step 4924 | loss:1.44034743309021 lr:9.56525474169686e-05 tokens_per_second_per_gpu:5026.158379317357 mfu:23.709246626287346 +Step 4925 | loss:1.264910340309143 lr:9.565077873669247e-05 tokens_per_second_per_gpu:3939.966540864417 mfu:18.58549440086724 +Step 4926 | loss:1.2211815118789673 lr:9.5649009713071e-05 tokens_per_second_per_gpu:5331.913653519974 mfu:25.151546421930185 +Step 4927 | loss:1.184226155281067 lr:9.56472403461175e-05 tokens_per_second_per_gpu:4881.934477917088 mfu:23.02891787624729 +Step 4928 | loss:1.210592269897461 lr:9.564547063584527e-05 tokens_per_second_per_gpu:4370.438567784508 mfu:20.61610439795063 +Step 4929 | loss:1.321668028831482 lr:9.564370058226765e-05 tokens_per_second_per_gpu:5314.645494584375 mfu:25.070089569979245 +Step 4930 | loss:1.2663743495941162 lr:9.564193018539793e-05 tokens_per_second_per_gpu:5325.890704899454 mfu:25.123135145666193 +Step 4931 | loss:1.424592137336731 lr:9.564015944524942e-05 tokens_per_second_per_gpu:4327.933919706121 mfu:20.415602720924106 +Step 4932 | loss:1.2615766525268555 lr:9.563838836183546e-05 tokens_per_second_per_gpu:4753.357275954924 mfu:22.422397277058984 +Step 4933 | loss:1.515641212463379 lr:9.563661693516934e-05 tokens_per_second_per_gpu:4161.2774194729045 mfu:19.629455574793006 +Step 4934 | loss:1.2617329359054565 lr:9.56348451652644e-05 tokens_per_second_per_gpu:5327.782502205912 mfu:25.132059076334663 +Step 4935 | loss:1.369783878326416 lr:9.563307305213399e-05 tokens_per_second_per_gpu:5204.142628348019 mfu:24.548828696209235 +Step 4936 | loss:1.4163275957107544 lr:9.56313005957914e-05 tokens_per_second_per_gpu:3495.124561358143 mfu:16.487098885668104 +Step 4937 | loss:1.4537988901138306 lr:9.562952779624997e-05 tokens_per_second_per_gpu:5375.544552737557 mfu:25.357360817738552 +Step 4938 | loss:1.2664718627929688 lr:9.562775465352305e-05 tokens_per_second_per_gpu:4544.198279591044 mfu:21.435758605005937 +Step 4939 | loss:1.3291347026824951 lr:9.562598116762396e-05 tokens_per_second_per_gpu:5357.75503221713 mfu:25.273444614238517 +Step 4940 | loss:1.323433756828308 lr:9.562420733856605e-05 tokens_per_second_per_gpu:4374.2886731617255 mfu:20.634266001910675 +Step 4941 | loss:1.3644121885299683 lr:9.562243316636264e-05 tokens_per_second_per_gpu:3729.7699900215885 mfu:17.593961407311944 +Step 4942 | loss:1.3999499082565308 lr:9.56206586510271e-05 tokens_per_second_per_gpu:3599.661427665265 mfu:16.980217119867362 +Step 4943 | loss:1.2322295904159546 lr:9.561888379257277e-05 tokens_per_second_per_gpu:3683.485706041644 mfu:17.375630542865398 +Step 4944 | loss:1.2449232339859009 lr:9.561710859101299e-05 tokens_per_second_per_gpu:5323.901125522471 mfu:25.11374996028772 +Step 4945 | loss:1.1631065607070923 lr:9.56153330463611e-05 tokens_per_second_per_gpu:5321.483009388189 mfu:25.102343293909158 +Step 4946 | loss:1.2799427509307861 lr:9.561355715863049e-05 tokens_per_second_per_gpu:5338.197663051657 mfu:25.18118916705372 +Step 4947 | loss:1.38394033908844 lr:9.561178092783449e-05 tokens_per_second_per_gpu:4103.087518784391 mfu:19.354963885024464 +Step 4948 | loss:1.1987167596817017 lr:9.561000435398646e-05 tokens_per_second_per_gpu:5265.107969051159 mfu:24.836412610065665 +Step 4949 | loss:1.2671071290969849 lr:9.560822743709978e-05 tokens_per_second_per_gpu:5267.6753614327135 mfu:24.84852343797177 +Step 4950 | loss:1.3059247732162476 lr:9.560645017718778e-05 tokens_per_second_per_gpu:5311.01980661731 mfu:25.052986581232442 +Step 4951 | loss:1.291015863418579 lr:9.560467257426386e-05 tokens_per_second_per_gpu:5166.5023623168 mfu:24.371273139248363 +Step 4952 | loss:1.2935025691986084 lr:9.560289462834139e-05 tokens_per_second_per_gpu:5307.914248576783 mfu:25.038337134092437 +Step 4953 | loss:1.2862473726272583 lr:9.560111633943371e-05 tokens_per_second_per_gpu:5344.7896092246865 mfu:25.21228450185389 +Step 4954 | loss:1.3634601831436157 lr:9.559933770755424e-05 tokens_per_second_per_gpu:4690.796924415091 mfu:22.12728942494896 +Step 4955 | loss:1.348739504814148 lr:9.559755873271632e-05 tokens_per_second_per_gpu:3716.6345627130777 mfu:17.531999355562704 +Step 4956 | loss:1.2977166175842285 lr:9.559577941493335e-05 tokens_per_second_per_gpu:5325.318384201477 mfu:25.120435411285854 +Step 4957 | loss:1.342897891998291 lr:9.559399975421869e-05 tokens_per_second_per_gpu:3737.913930014456 mfu:17.63237776175775 +Step 4958 | loss:1.1449692249298096 lr:9.559221975058575e-05 tokens_per_second_per_gpu:5315.4130446020545 mfu:25.07371023437017 +Step 4959 | loss:1.4368486404418945 lr:9.559043940404791e-05 tokens_per_second_per_gpu:2823.481461695908 mfu:13.318843790431966 +Step 4960 | loss:1.3725924491882324 lr:9.558865871461854e-05 tokens_per_second_per_gpu:3948.040927391404 mfu:18.623582659747438 +Step 4961 | loss:1.187408447265625 lr:9.558687768231105e-05 tokens_per_second_per_gpu:5345.694955387234 mfu:25.216555174170608 +Step 4962 | loss:1.2669868469238281 lr:9.558509630713886e-05 tokens_per_second_per_gpu:5174.3181111176145 mfu:24.40814135984606 +Step 4963 | loss:1.3273030519485474 lr:9.558331458911533e-05 tokens_per_second_per_gpu:4216.245158121401 mfu:19.888747776461255 +Step 4964 | loss:1.277328610420227 lr:9.558153252825388e-05 tokens_per_second_per_gpu:5338.67368617517 mfu:25.183434649345244 +Step 4965 | loss:1.383501648902893 lr:9.55797501245679e-05 tokens_per_second_per_gpu:4825.468620166174 mfu:22.762558791167468 +Step 4966 | loss:1.3712329864501953 lr:9.55779673780708e-05 tokens_per_second_per_gpu:4943.376359004749 mfu:23.318749712403683 +Step 4967 | loss:1.1475708484649658 lr:9.557618428877602e-05 tokens_per_second_per_gpu:4228.737564706546 mfu:19.947676589747733 +Step 4968 | loss:1.1734195947647095 lr:9.55744008566969e-05 tokens_per_second_per_gpu:5296.102606348572 mfu:24.98261960242799 +Step 4969 | loss:1.1807377338409424 lr:9.557261708184692e-05 tokens_per_second_per_gpu:4658.019582547862 mfu:21.97267311097005 +Step 4970 | loss:1.262101650238037 lr:9.557083296423948e-05 tokens_per_second_per_gpu:3011.5956132679717 mfu:14.20621033898124 +Step 4971 | loss:1.5432442426681519 lr:9.556904850388798e-05 tokens_per_second_per_gpu:3977.1820359371723 mfu:18.76104624074364 +Step 4972 | loss:1.3153282403945923 lr:9.556726370080585e-05 tokens_per_second_per_gpu:3539.7472022914935 mfu:16.69759149635833 +Step 4973 | loss:1.1602145433425903 lr:9.556547855500653e-05 tokens_per_second_per_gpu:4704.300106090955 mfu:22.19098623679437 +Step 4974 | loss:1.3715673685073853 lr:9.556369306650342e-05 tokens_per_second_per_gpu:5156.476909672844 mfu:24.323981368608223 +Step 4975 | loss:1.5061182975769043 lr:9.556190723530995e-05 tokens_per_second_per_gpu:5032.31463228957 mfu:23.738286721922343 +Step 4976 | loss:1.227296233177185 lr:9.556012106143957e-05 tokens_per_second_per_gpu:4711.126423796405 mfu:22.223187142100883 +Step 4977 | loss:1.1494438648223877 lr:9.55583345449057e-05 tokens_per_second_per_gpu:4643.969756668518 mfu:21.906397685149702 +Step 4978 | loss:1.2687904834747314 lr:9.55565476857218e-05 tokens_per_second_per_gpu:5303.176896596777 mfu:25.015990236527152 +Step 4979 | loss:1.2805638313293457 lr:9.555476048390129e-05 tokens_per_second_per_gpu:5305.644806843585 mfu:25.027631790230252 +Step 4980 | loss:1.2780733108520508 lr:9.555297293945759e-05 tokens_per_second_per_gpu:4897.8708224249685 mfu:23.104092332311712 +Step 4981 | loss:1.2542567253112793 lr:9.55511850524042e-05 tokens_per_second_per_gpu:4758.1909605918345 mfu:22.445198592202928 +Step 4982 | loss:1.2424015998840332 lr:9.55493968227545e-05 tokens_per_second_per_gpu:4802.12681715931 mfu:22.652451523861828 +Step 4983 | loss:1.076867938041687 lr:9.5547608250522e-05 tokens_per_second_per_gpu:3921.5968154664924 mfu:18.498841271966892 +Step 4984 | loss:1.349073886871338 lr:9.554581933572011e-05 tokens_per_second_per_gpu:4110.1478521323725 mfu:19.38826868204398 +Step 4985 | loss:1.4586999416351318 lr:9.554403007836232e-05 tokens_per_second_per_gpu:4487.090047036431 mfu:21.166369328376618 +Step 4986 | loss:1.260979175567627 lr:9.554224047846206e-05 tokens_per_second_per_gpu:5371.8574927577365 mfu:25.339968326736436 +Step 4987 | loss:1.2722716331481934 lr:9.55404505360328e-05 tokens_per_second_per_gpu:3347.219429380484 mfu:15.789405142911798 +Step 4988 | loss:1.2401310205459595 lr:9.553866025108798e-05 tokens_per_second_per_gpu:5345.1336696663975 mfu:25.213907493660205 +Step 4989 | loss:1.218979835510254 lr:9.55368696236411e-05 tokens_per_second_per_gpu:4813.057934448926 mfu:22.704015469157877 +Step 4990 | loss:1.2769951820373535 lr:9.553507865370562e-05 tokens_per_second_per_gpu:3720.9082543649643 mfu:17.55215908825164 +Step 4991 | loss:1.335829257965088 lr:9.553328734129501e-05 tokens_per_second_per_gpu:4037.1810022219515 mfu:19.044071601588186 +Step 4992 | loss:1.175960898399353 lr:9.553149568642271e-05 tokens_per_second_per_gpu:4353.109186968448 mfu:20.534358751944016 +Step 4993 | loss:1.1887266635894775 lr:9.552970368910224e-05 tokens_per_second_per_gpu:5352.843777994771 mfu:25.25027739012527 +Step 4994 | loss:1.3330481052398682 lr:9.552791134934706e-05 tokens_per_second_per_gpu:4016.5766853973364 mfu:18.946877523667524 +Step 4995 | loss:1.3128498792648315 lr:9.552611866717063e-05 tokens_per_second_per_gpu:5347.821395906762 mfu:25.226585956909386 +Step 4996 | loss:1.2233495712280273 lr:9.552432564258647e-05 tokens_per_second_per_gpu:4111.60314038676 mfu:19.395133525037707 +Step 4997 | loss:1.1175519227981567 lr:9.552253227560804e-05 tokens_per_second_per_gpu:5088.762601807956 mfu:24.00456142515729 +Step 4998 | loss:1.210557460784912 lr:9.552073856624884e-05 tokens_per_second_per_gpu:4741.900504592137 mfu:22.368353731813954 +Step 4999 | loss:1.2580375671386719 lr:9.551894451452237e-05 tokens_per_second_per_gpu:4311.295643422719 mfu:20.33711713291264 +Step 5000 | loss:1.2871479988098145 lr:9.551715012044209e-05 tokens_per_second_per_gpu:3679.308094518119 mfu:17.355924036534905 +Step 5001 | loss:1.2744349241256714 lr:9.551535538402152e-05 tokens_per_second_per_gpu:5391.394359473525 mfu:25.432127060367996 +Step 5002 | loss:1.1851409673690796 lr:9.551356030527416e-05 tokens_per_second_per_gpu:5330.665414540783 mfu:25.145658265693807 +Step 5003 | loss:1.3963302373886108 lr:9.551176488421349e-05 tokens_per_second_per_gpu:4484.389124100422 mfu:21.15362861406249 +Step 5004 | loss:1.4308642148971558 lr:9.550996912085305e-05 tokens_per_second_per_gpu:4343.631583112018 mfu:20.489651277507274 +Step 5005 | loss:1.3510128259658813 lr:9.550817301520631e-05 tokens_per_second_per_gpu:4835.256899988047 mfu:22.80873177713999 +Step 5006 | loss:1.3014992475509644 lr:9.550637656728681e-05 tokens_per_second_per_gpu:5327.180529492365 mfu:25.129219468337723 +Step 5007 | loss:1.3671622276306152 lr:9.550457977710802e-05 tokens_per_second_per_gpu:4155.332392239414 mfu:19.60141186700639 +Step 5008 | loss:1.3364702463150024 lr:9.550278264468351e-05 tokens_per_second_per_gpu:4220.473608497992 mfu:19.908694098335666 +Step 5009 | loss:1.3875057697296143 lr:9.550098517002675e-05 tokens_per_second_per_gpu:3717.7984520156683 mfu:17.537489619983017 +Step 5010 | loss:1.2447702884674072 lr:9.549918735315126e-05 tokens_per_second_per_gpu:5365.616438484757 mfu:25.310528208897132 +Step 5011 | loss:1.4071717262268066 lr:9.54973891940706e-05 tokens_per_second_per_gpu:3768.2306311706743 mfu:17.775386813674224 +Step 5012 | loss:1.4700976610183716 lr:9.549559069279827e-05 tokens_per_second_per_gpu:4373.377017340269 mfu:20.629965565856345 +Step 5013 | loss:1.2590396404266357 lr:9.549379184934779e-05 tokens_per_second_per_gpu:4307.1485464147745 mfu:20.31755456875845 +Step 5014 | loss:1.2800538539886475 lr:9.549199266373268e-05 tokens_per_second_per_gpu:4176.596832521735 mfu:19.70171985990573 +Step 5015 | loss:1.1350576877593994 lr:9.549019313596651e-05 tokens_per_second_per_gpu:4883.3951886225195 mfu:23.035808297867757 +Step 5016 | loss:1.1205130815505981 lr:9.548839326606279e-05 tokens_per_second_per_gpu:4348.539992844697 mfu:20.512805083677378 +Step 5017 | loss:1.362656831741333 lr:9.548659305403504e-05 tokens_per_second_per_gpu:3747.4300767003674 mfu:17.67726704929724 +Step 5018 | loss:1.3499436378479004 lr:9.548479249989685e-05 tokens_per_second_per_gpu:3791.8306369849197 mfu:17.886712067676218 +Step 5019 | loss:1.344973087310791 lr:9.548299160366171e-05 tokens_per_second_per_gpu:5357.850592397024 mfu:25.273895387911324 +Step 5020 | loss:1.3540958166122437 lr:9.54811903653432e-05 tokens_per_second_per_gpu:5349.336081789415 mfu:25.233730988649924 +Step 5021 | loss:1.3084628582000732 lr:9.547938878495483e-05 tokens_per_second_per_gpu:4768.8402076662815 mfu:22.495432907602005 +Step 5022 | loss:1.306153416633606 lr:9.54775868625102e-05 tokens_per_second_per_gpu:5080.031035863785 mfu:23.963373138839867 +Step 5023 | loss:1.253936767578125 lr:9.547578459802283e-05 tokens_per_second_per_gpu:5295.173965498067 mfu:24.978239045093694 +Step 5024 | loss:1.394081473350525 lr:9.547398199150628e-05 tokens_per_second_per_gpu:4220.822663616692 mfu:19.910340650885512 +Step 5025 | loss:1.3124974966049194 lr:9.547217904297411e-05 tokens_per_second_per_gpu:5289.657465571907 mfu:24.95221677373005 +Step 5026 | loss:1.5229417085647583 lr:9.547037575243987e-05 tokens_per_second_per_gpu:5114.555971106662 mfu:24.126233148941882 +Step 5027 | loss:1.2736268043518066 lr:9.546857211991714e-05 tokens_per_second_per_gpu:5370.508705878039 mfu:25.333605868898132 +Step 5028 | loss:1.2031636238098145 lr:9.546676814541949e-05 tokens_per_second_per_gpu:5333.398253981507 mfu:25.15854953560619 +Step 5029 | loss:1.335052728652954 lr:9.546496382896044e-05 tokens_per_second_per_gpu:4445.394049576669 mfu:20.969682194288175 +Step 5030 | loss:1.2894551753997803 lr:9.546315917055361e-05 tokens_per_second_per_gpu:5293.408229730968 mfu:24.9699097682147 +Step 5031 | loss:1.4964599609375 lr:9.546135417021258e-05 tokens_per_second_per_gpu:3932.6550269477816 mfu:18.55100474224956 +Step 5032 | loss:1.3053686618804932 lr:9.54595488279509e-05 tokens_per_second_per_gpu:5035.9230096742685 mfu:23.755308053698634 +Step 5033 | loss:1.2197495698928833 lr:9.545774314378213e-05 tokens_per_second_per_gpu:5377.906815899666 mfu:25.36850401611788 +Step 5034 | loss:1.1221332550048828 lr:9.545593711771989e-05 tokens_per_second_per_gpu:5108.150432783455 mfu:24.09601713177378 +Step 5035 | loss:1.3242632150650024 lr:9.545413074977774e-05 tokens_per_second_per_gpu:4021.048482280202 mfu:18.96797175253122 +Step 5036 | loss:1.3911775350570679 lr:9.545232403996928e-05 tokens_per_second_per_gpu:5153.435825912538 mfu:24.309636057648465 +Step 5037 | loss:1.2201553583145142 lr:9.545051698830808e-05 tokens_per_second_per_gpu:4396.368445547113 mfu:20.738420055450177 +Step 5038 | loss:1.3129935264587402 lr:9.544870959480776e-05 tokens_per_second_per_gpu:4006.2516063886833 mfu:18.898172364343573 +Step 5039 | loss:1.177807092666626 lr:9.544690185948188e-05 tokens_per_second_per_gpu:4460.324774344205 mfu:21.04011297046097 +Step 5040 | loss:1.4084397554397583 lr:9.544509378234405e-05 tokens_per_second_per_gpu:4043.5418124531075 mfu:19.074076628714682 +Step 5041 | loss:1.34611177444458 lr:9.544328536340787e-05 tokens_per_second_per_gpu:4688.563825807588 mfu:22.11675551781146 +Step 5042 | loss:1.712697148323059 lr:9.544147660268696e-05 tokens_per_second_per_gpu:5356.80246094213 mfu:25.268951173008922 +Step 5043 | loss:1.4082107543945312 lr:9.543966750019488e-05 tokens_per_second_per_gpu:3689.2000976665686 mfu:17.402586303135916 +Step 5044 | loss:1.3842828273773193 lr:9.54378580559453e-05 tokens_per_second_per_gpu:5122.210319952875 mfu:24.16234001841605 +Step 5045 | loss:1.3601317405700684 lr:9.543604826995176e-05 tokens_per_second_per_gpu:3797.302052756703 mfu:17.91252166939151 +Step 5046 | loss:1.215771198272705 lr:9.543423814222791e-05 tokens_per_second_per_gpu:5376.897203394788 mfu:25.363741501674713 +Step 5047 | loss:1.2211562395095825 lr:9.543242767278737e-05 tokens_per_second_per_gpu:5064.133984673116 mfu:23.888384036056863 +Step 5048 | loss:1.2430527210235596 lr:9.543061686164373e-05 tokens_per_second_per_gpu:5263.641092383058 mfu:24.829493102547268 +Step 5049 | loss:1.2933080196380615 lr:9.542880570881063e-05 tokens_per_second_per_gpu:4851.287351351399 mfu:22.884350151297955 +Step 5050 | loss:1.2825292348861694 lr:9.542699421430168e-05 tokens_per_second_per_gpu:3991.3737387315955 mfu:18.827990924179115 +Step 5051 | loss:1.155613660812378 lr:9.542518237813052e-05 tokens_per_second_per_gpu:4980.551945624695 mfu:23.494113297299094 +Step 5052 | loss:1.3834812641143799 lr:9.542337020031075e-05 tokens_per_second_per_gpu:4319.65179282506 mfu:20.376534515349583 +Step 5053 | loss:1.3910962343215942 lr:9.542155768085604e-05 tokens_per_second_per_gpu:5288.550679539846 mfu:24.946995875179653 +Step 5054 | loss:1.1337828636169434 lr:9.541974481977998e-05 tokens_per_second_per_gpu:4768.6885114047145 mfu:22.494717330454144 +Step 5055 | loss:1.2599018812179565 lr:9.541793161709622e-05 tokens_per_second_per_gpu:5318.093766575978 mfu:25.086355657299503 +Step 5056 | loss:1.4559305906295776 lr:9.541611807281842e-05 tokens_per_second_per_gpu:4179.767536734468 mfu:19.716676612655114 +Step 5057 | loss:1.3908627033233643 lr:9.541430418696018e-05 tokens_per_second_per_gpu:3661.50924687813 mfu:17.271963835420525 +Step 5058 | loss:1.366532564163208 lr:9.541248995953518e-05 tokens_per_second_per_gpu:5300.1071990488645 mfu:25.001509949449243 +Step 5059 | loss:1.3974710702896118 lr:9.541067539055705e-05 tokens_per_second_per_gpu:4888.426484473449 mfu:23.059541778824183 +Step 5060 | loss:1.1833162307739258 lr:9.540886048003943e-05 tokens_per_second_per_gpu:5332.1556132167425 mfu:25.152687787102423 +Step 5061 | loss:1.2508121728897095 lr:9.540704522799597e-05 tokens_per_second_per_gpu:5321.405965324958 mfu:25.10197986391782 +Step 5062 | loss:1.3793116807937622 lr:9.540522963444034e-05 tokens_per_second_per_gpu:4294.671805634089 mfu:20.25869965374447 +Step 5063 | loss:1.2726463079452515 lr:9.540341369938617e-05 tokens_per_second_per_gpu:5338.103949801592 mfu:25.18074710566403 +Step 5064 | loss:1.2514537572860718 lr:9.540159742284714e-05 tokens_per_second_per_gpu:5151.760667938284 mfu:24.301734051672486 +Step 5065 | loss:1.4428349733352661 lr:9.539978080483691e-05 tokens_per_second_per_gpu:3847.7118491394417 mfu:18.15031327972704 +Step 5066 | loss:1.4149725437164307 lr:9.539796384536913e-05 tokens_per_second_per_gpu:4283.142783967438 mfu:20.204315291488946 +Step 5067 | loss:1.1760807037353516 lr:9.539614654445747e-05 tokens_per_second_per_gpu:4121.364655513635 mfu:19.441180257377397 +Step 5068 | loss:1.3618673086166382 lr:9.539432890211559e-05 tokens_per_second_per_gpu:3972.0416566851486 mfu:18.736798194772707 +Step 5069 | loss:1.396715521812439 lr:9.539251091835719e-05 tokens_per_second_per_gpu:5367.036091287967 mfu:25.317224953387615 +Step 5070 | loss:1.2529109716415405 lr:9.539069259319591e-05 tokens_per_second_per_gpu:5356.397893545487 mfu:25.267042759572643 +Step 5071 | loss:1.347155213356018 lr:9.538887392664544e-05 tokens_per_second_per_gpu:5256.831013729234 mfu:24.797368799617946 +Step 5072 | loss:1.1950503587722778 lr:9.538705491871946e-05 tokens_per_second_per_gpu:4826.929410316907 mfu:22.76944958754497 +Step 5073 | loss:1.2719961404800415 lr:9.538523556943165e-05 tokens_per_second_per_gpu:4709.374490382861 mfu:22.214922973278533 +Step 5074 | loss:1.3611412048339844 lr:9.53834158787957e-05 tokens_per_second_per_gpu:5089.293681167942 mfu:24.007066617109825 +Step 5075 | loss:1.3199563026428223 lr:9.538159584682526e-05 tokens_per_second_per_gpu:4391.676068511358 mfu:20.71628531237044 +Step 5076 | loss:1.3627190589904785 lr:9.537977547353407e-05 tokens_per_second_per_gpu:5327.648587005307 mfu:25.131427375485114 +Step 5077 | loss:1.3153109550476074 lr:9.537795475893579e-05 tokens_per_second_per_gpu:4919.211244211449 mfu:23.204758742929954 +Step 5078 | loss:1.3096920251846313 lr:9.537613370304412e-05 tokens_per_second_per_gpu:5330.398938668195 mfu:25.14440125353734 +Step 5079 | loss:1.2433266639709473 lr:9.537431230587277e-05 tokens_per_second_per_gpu:3420.579041730803 mfu:16.13545495080877 +Step 5080 | loss:1.249512791633606 lr:9.537249056743542e-05 tokens_per_second_per_gpu:5341.19473154799 mfu:25.195326850503534 +Step 5081 | loss:1.5151786804199219 lr:9.537066848774578e-05 tokens_per_second_per_gpu:2831.154531765629 mfu:13.355038971111592 +Step 5082 | loss:1.3311432600021362 lr:9.536884606681756e-05 tokens_per_second_per_gpu:5074.973283011967 mfu:23.939514855696224 +Step 5083 | loss:1.2309143543243408 lr:9.536702330466444e-05 tokens_per_second_per_gpu:4123.637049328804 mfu:19.451899526713913 +Step 5084 | loss:1.2332127094268799 lr:9.536520020130015e-05 tokens_per_second_per_gpu:4682.14207407076 mfu:22.086463019205013 +Step 5085 | loss:1.2695653438568115 lr:9.536337675673843e-05 tokens_per_second_per_gpu:4185.278944026712 mfu:19.74267486119568 +Step 5086 | loss:1.128849744796753 lr:9.536155297099294e-05 tokens_per_second_per_gpu:3847.542007959084 mfu:18.149512109901398 +Step 5087 | loss:1.2704856395721436 lr:9.535972884407742e-05 tokens_per_second_per_gpu:5351.777398324927 mfu:25.245247095279485 +Step 5088 | loss:1.3626222610473633 lr:9.53579043760056e-05 tokens_per_second_per_gpu:3895.297107654435 mfu:18.374781063024557 +Step 5089 | loss:1.2881051301956177 lr:9.53560795667912e-05 tokens_per_second_per_gpu:3949.130088948491 mfu:18.628720420642342 +Step 5090 | loss:1.1418014764785767 lr:9.535425441644793e-05 tokens_per_second_per_gpu:5239.803881607566 mfu:24.717048912271732 +Step 5091 | loss:1.2336280345916748 lr:9.535242892498952e-05 tokens_per_second_per_gpu:4421.135464878087 mfu:20.85525031132362 +Step 5092 | loss:1.3917779922485352 lr:9.535060309242972e-05 tokens_per_second_per_gpu:3777.776195993056 mfu:17.820414871582695 +Step 5093 | loss:1.2728602886199951 lr:9.534877691878223e-05 tokens_per_second_per_gpu:4052.4061770292133 mfu:19.115891348836723 +Step 5094 | loss:1.2905954122543335 lr:9.534695040406082e-05 tokens_per_second_per_gpu:4307.114681140083 mfu:20.31739482047972 +Step 5095 | loss:1.3789029121398926 lr:9.534512354827921e-05 tokens_per_second_per_gpu:5030.927501535004 mfu:23.731743389484144 +Step 5096 | loss:1.3626853227615356 lr:9.534329635145112e-05 tokens_per_second_per_gpu:4916.464900295014 mfu:23.191803770101536 +Step 5097 | loss:1.2544176578521729 lr:9.534146881359032e-05 tokens_per_second_per_gpu:5299.9035347970585 mfu:25.000549230424767 +Step 5098 | loss:1.3503618240356445 lr:9.533964093471055e-05 tokens_per_second_per_gpu:5356.119525748277 mfu:25.26572965117876 +Step 5099 | loss:1.2799123525619507 lr:9.533781271482556e-05 tokens_per_second_per_gpu:5310.736460841101 mfu:25.051649990862284 +Step 5100 | loss:1.352291464805603 lr:9.533598415394908e-05 tokens_per_second_per_gpu:4148.93662800547 mfu:19.571241955884975 +Step 5101 | loss:1.261667251586914 lr:9.53341552520949e-05 tokens_per_second_per_gpu:5311.007407743735 mfu:25.05292809363032 +Step 5102 | loss:1.277232050895691 lr:9.533232600927675e-05 tokens_per_second_per_gpu:3998.0942573268044 mfu:18.85969275703142 +Step 5103 | loss:1.2475025653839111 lr:9.533049642550838e-05 tokens_per_second_per_gpu:3534.0022519797435 mfu:16.67049158554821 +Step 5104 | loss:1.266832947731018 lr:9.532866650080357e-05 tokens_per_second_per_gpu:4570.264673673312 mfu:21.55871823327743 +Step 5105 | loss:1.320934534072876 lr:9.532683623517606e-05 tokens_per_second_per_gpu:3725.0561009316134 mfu:17.571725188202123 +Step 5106 | loss:1.3870657682418823 lr:9.532500562863964e-05 tokens_per_second_per_gpu:4031.8811258305745 mfu:19.019071180397905 +Step 5107 | loss:1.335331916809082 lr:9.532317468120808e-05 tokens_per_second_per_gpu:4533.243648660915 mfu:21.38408374185575 +Step 5108 | loss:1.354462742805481 lr:9.532134339289514e-05 tokens_per_second_per_gpu:5034.678303358994 mfu:23.749436561640685 +Step 5109 | loss:1.1669747829437256 lr:9.531951176371459e-05 tokens_per_second_per_gpu:5332.203075570326 mfu:25.152911675122187 +Step 5110 | loss:1.2130814790725708 lr:9.531767979368019e-05 tokens_per_second_per_gpu:3462.5099222735685 mfu:16.333250068474932 +Step 5111 | loss:1.514606237411499 lr:9.531584748280575e-05 tokens_per_second_per_gpu:3788.9611269963375 mfu:17.873176099471188 +Step 5112 | loss:1.4368761777877808 lr:9.531401483110504e-05 tokens_per_second_per_gpu:5077.539227894846 mfu:23.951618855504098 +Step 5113 | loss:1.248558521270752 lr:9.531218183859186e-05 tokens_per_second_per_gpu:3654.8273511257617 mfu:17.24044419310763 +Step 5114 | loss:1.4523327350616455 lr:9.531034850527995e-05 tokens_per_second_per_gpu:4419.213483076702 mfu:20.846183995242576 +Step 5115 | loss:1.247030258178711 lr:9.530851483118315e-05 tokens_per_second_per_gpu:5056.540750151698 mfu:23.852565453279933 +Step 5116 | loss:1.1793529987335205 lr:9.530668081631521e-05 tokens_per_second_per_gpu:5022.240725774019 mfu:23.690766386102798 +Step 5117 | loss:1.5182887315750122 lr:9.530484646068996e-05 tokens_per_second_per_gpu:3784.874451758056 mfu:17.853898555113517 +Step 5118 | loss:1.3644490242004395 lr:9.530301176432116e-05 tokens_per_second_per_gpu:4230.621879625767 mfu:19.956565224718975 +Step 5119 | loss:1.4071135520935059 lr:9.530117672722263e-05 tokens_per_second_per_gpu:4875.647270162747 mfu:22.999260044561808 +Step 5120 | loss:1.4264380931854248 lr:9.52993413494082e-05 tokens_per_second_per_gpu:5078.638356522051 mfu:23.95680363276931 +Step 5121 | loss:1.3889045715332031 lr:9.52975056308916e-05 tokens_per_second_per_gpu:5291.099677566771 mfu:24.959019933776567 +Step 5122 | loss:1.2971934080123901 lr:9.52956695716867e-05 tokens_per_second_per_gpu:5353.191810092157 mfu:25.25191911690904 +Step 5123 | loss:1.2349056005477905 lr:9.529383317180729e-05 tokens_per_second_per_gpu:5335.197937113793 mfu:25.16703894799879 +Step 5124 | loss:1.4090501070022583 lr:9.529199643126719e-05 tokens_per_second_per_gpu:4912.352159370812 mfu:23.17240326945612 +Step 5125 | loss:1.0049057006835938 lr:9.529015935008019e-05 tokens_per_second_per_gpu:4982.877412276579 mfu:23.505082920261554 +Step 5126 | loss:1.3251338005065918 lr:9.528832192826012e-05 tokens_per_second_per_gpu:5379.585649600031 mfu:25.37642336112106 +Step 5127 | loss:1.3426836729049683 lr:9.528648416582082e-05 tokens_per_second_per_gpu:5319.235251867788 mfu:25.091740238178442 +Step 5128 | loss:1.3359163999557495 lr:9.528464606277608e-05 tokens_per_second_per_gpu:3875.446951011458 mfu:18.28114448735368 +Step 5129 | loss:1.3393360376358032 lr:9.528280761913975e-05 tokens_per_second_per_gpu:4653.954200406965 mfu:21.953496009785724 +Step 5130 | loss:1.4192394018173218 lr:9.528096883492564e-05 tokens_per_second_per_gpu:4740.509298582589 mfu:22.361791175702695 +Step 5131 | loss:1.1630384922027588 lr:9.527912971014758e-05 tokens_per_second_per_gpu:3959.4180557065074 mfu:18.677250515148035 +Step 5132 | loss:1.371578335762024 lr:9.52772902448194e-05 tokens_per_second_per_gpu:4116.0045231472695 mfu:19.415895598472364 +Step 5133 | loss:1.4540612697601318 lr:9.527545043895496e-05 tokens_per_second_per_gpu:4552.06907502993 mfu:21.472886489987335 +Step 5134 | loss:1.2503743171691895 lr:9.527361029256808e-05 tokens_per_second_per_gpu:4023.943082516817 mfu:18.98162607571706 +Step 5135 | loss:1.2800387144088745 lr:9.527176980567258e-05 tokens_per_second_per_gpu:4869.550514327342 mfu:22.970500607072395 +Step 5136 | loss:1.1557732820510864 lr:9.526992897828235e-05 tokens_per_second_per_gpu:5294.506579349343 mfu:24.97509087076234 +Step 5137 | loss:1.2028040885925293 lr:9.526808781041119e-05 tokens_per_second_per_gpu:4738.785053502885 mfu:22.353657617475186 +Step 5138 | loss:1.2290868759155273 lr:9.526624630207297e-05 tokens_per_second_per_gpu:3965.170688955289 mfu:18.704386667682975 +Step 5139 | loss:1.2867335081100464 lr:9.526440445328154e-05 tokens_per_second_per_gpu:5286.140310099121 mfu:24.93562575127632 +Step 5140 | loss:1.2364579439163208 lr:9.526256226405075e-05 tokens_per_second_per_gpu:5174.5546156373175 mfu:24.40925699201733 +Step 5141 | loss:1.2657841444015503 lr:9.526071973439444e-05 tokens_per_second_per_gpu:4739.262009771875 mfu:22.35590750157314 +Step 5142 | loss:1.1653186082839966 lr:9.525887686432648e-05 tokens_per_second_per_gpu:3790.792652649748 mfu:17.881815718468822 +Step 5143 | loss:1.15753972530365 lr:9.525703365386074e-05 tokens_per_second_per_gpu:5361.020927137624 mfu:25.288850397797617 +Step 5144 | loss:1.3913400173187256 lr:9.52551901030111e-05 tokens_per_second_per_gpu:4104.788156906004 mfu:19.362986085202895 +Step 5145 | loss:1.2517887353897095 lr:9.525334621179136e-05 tokens_per_second_per_gpu:4132.376431484663 mfu:19.49312468343545 +Step 5146 | loss:1.2261748313903809 lr:9.525150198021545e-05 tokens_per_second_per_gpu:4889.567060696616 mfu:23.06492207147123 +Step 5147 | loss:1.6205506324768066 lr:9.524965740829722e-05 tokens_per_second_per_gpu:4696.367262711881 mfu:22.153565661092635 +Step 5148 | loss:1.2667412757873535 lr:9.524781249605053e-05 tokens_per_second_per_gpu:4078.5293542283457 mfu:19.23911883275866 +Step 5149 | loss:1.2079136371612549 lr:9.524596724348928e-05 tokens_per_second_per_gpu:4680.703825193107 mfu:22.07967855385856 +Step 5150 | loss:1.30995512008667 lr:9.524412165062734e-05 tokens_per_second_per_gpu:5049.613361062224 mfu:23.819887777010422 +Step 5151 | loss:1.313443899154663 lr:9.524227571747857e-05 tokens_per_second_per_gpu:5363.655267875918 mfu:25.30127703252503 +Step 5152 | loss:1.234026551246643 lr:9.524042944405688e-05 tokens_per_second_per_gpu:5251.4300387838975 mfu:24.771891479299445 +Step 5153 | loss:1.2608486413955688 lr:9.523858283037616e-05 tokens_per_second_per_gpu:3585.7237314438216 mfu:16.914470628774623 +Step 5154 | loss:1.3274855613708496 lr:9.523673587645028e-05 tokens_per_second_per_gpu:4310.074702098732 mfu:20.331357744373317 +Step 5155 | loss:1.1072580814361572 lr:9.523488858229313e-05 tokens_per_second_per_gpu:4017.5924459553667 mfu:18.951669039526376 +Step 5156 | loss:1.157441258430481 lr:9.52330409479186e-05 tokens_per_second_per_gpu:4035.046215716825 mfu:19.034001449411367 +Step 5157 | loss:1.2956913709640503 lr:9.523119297334062e-05 tokens_per_second_per_gpu:4889.228268786198 mfu:23.06332393222593 +Step 5158 | loss:1.193707823753357 lr:9.522934465857308e-05 tokens_per_second_per_gpu:4291.4062449736875 mfu:20.243295446947254 +Step 5159 | loss:1.2335259914398193 lr:9.522749600362984e-05 tokens_per_second_per_gpu:3659.7823141896697 mfu:17.263817599283957 +Step 5160 | loss:1.223817229270935 lr:9.522564700852484e-05 tokens_per_second_per_gpu:4147.951786074559 mfu:19.566596288465508 +Step 5161 | loss:1.2236047983169556 lr:9.522379767327196e-05 tokens_per_second_per_gpu:5207.026924073468 mfu:24.562434411257772 +Step 5162 | loss:1.3113800287246704 lr:9.522194799788514e-05 tokens_per_second_per_gpu:3975.082002798719 mfu:18.75114002612695 +Step 5163 | loss:1.329919457435608 lr:9.522009798237829e-05 tokens_per_second_per_gpu:5348.915616752025 mfu:25.23174758332982 +Step 5164 | loss:1.360365629196167 lr:9.52182476267653e-05 tokens_per_second_per_gpu:3696.380845028243 mfu:17.43645911360277 +Step 5165 | loss:1.108350157737732 lr:9.52163969310601e-05 tokens_per_second_per_gpu:4884.088608147619 mfu:23.03907927607722 +Step 5166 | loss:1.2816131114959717 lr:9.521454589527661e-05 tokens_per_second_per_gpu:5370.994848088269 mfu:25.335899084648197 +Step 5167 | loss:1.3725597858428955 lr:9.521269451942875e-05 tokens_per_second_per_gpu:4998.564639673847 mfu:23.579082248411453 +Step 5168 | loss:1.4097930192947388 lr:9.521084280353045e-05 tokens_per_second_per_gpu:5316.581836469546 mfu:25.079223625025584 +Step 5169 | loss:1.2478877305984497 lr:9.520899074759565e-05 tokens_per_second_per_gpu:5378.590699727509 mfu:25.371730012816425 +Step 5170 | loss:1.3633677959442139 lr:9.520713835163824e-05 tokens_per_second_per_gpu:3683.84260038804 mfu:17.3773140743898 +Step 5171 | loss:1.4326900243759155 lr:9.520528561567217e-05 tokens_per_second_per_gpu:5085.058228680055 mfu:23.98708726508175 +Step 5172 | loss:1.381731390953064 lr:9.520343253971138e-05 tokens_per_second_per_gpu:3350.3022763991303 mfu:15.803947458286691 +Step 5173 | loss:1.2960153818130493 lr:9.520157912376981e-05 tokens_per_second_per_gpu:4993.8828870236575 mfu:23.556997622371142 +Step 5174 | loss:1.3350104093551636 lr:9.51997253678614e-05 tokens_per_second_per_gpu:4682.0274435358015 mfu:22.085922287413627 +Step 5175 | loss:1.3003021478652954 lr:9.519787127200008e-05 tokens_per_second_per_gpu:5352.856532790436 mfu:25.25033755667262 +Step 5176 | loss:1.2065463066101074 lr:9.519601683619981e-05 tokens_per_second_per_gpu:4678.119518643094 mfu:22.067487938933986 +Step 5177 | loss:1.253722071647644 lr:9.519416206047453e-05 tokens_per_second_per_gpu:3995.0059226576436 mfu:18.84512455547276 +Step 5178 | loss:1.1604201793670654 lr:9.519230694483818e-05 tokens_per_second_per_gpu:5317.061283465666 mfu:25.081485258308376 +Step 5179 | loss:1.2284456491470337 lr:9.519045148930474e-05 tokens_per_second_per_gpu:5280.545874760489 mfu:24.909235845274896 +Step 5180 | loss:1.3936398029327393 lr:9.518859569388814e-05 tokens_per_second_per_gpu:3236.343991919662 mfu:15.266386787108779 +Step 5181 | loss:1.277349829673767 lr:9.518673955860235e-05 tokens_per_second_per_gpu:5263.307215392973 mfu:24.827918147819748 +Step 5182 | loss:1.2804052829742432 lr:9.518488308346131e-05 tokens_per_second_per_gpu:5308.228538438417 mfu:25.03981969299314 +Step 5183 | loss:1.3008681535720825 lr:9.518302626847901e-05 tokens_per_second_per_gpu:3906.8282535787475 mfu:18.42917544576591 +Step 5184 | loss:1.2700082063674927 lr:9.518116911366941e-05 tokens_per_second_per_gpu:3934.3800108292667 mfu:18.559141785531125 +Step 5185 | loss:1.2424708604812622 lr:9.517931161904648e-05 tokens_per_second_per_gpu:4232.38112789797 mfu:19.96486389897766 +Step 5186 | loss:1.374300479888916 lr:9.517745378462416e-05 tokens_per_second_per_gpu:3846.9509148500397 mfu:18.146723822855154 +Step 5187 | loss:1.2241913080215454 lr:9.517559561041645e-05 tokens_per_second_per_gpu:4906.5442697090075 mfu:23.14500646299323 +Step 5188 | loss:1.3759300708770752 lr:9.517373709643733e-05 tokens_per_second_per_gpu:4779.165953507039 mfu:22.544141212486768 +Step 5189 | loss:1.3341224193572998 lr:9.517187824270075e-05 tokens_per_second_per_gpu:5350.200972604367 mfu:25.23781082618964 +Step 5190 | loss:1.1205167770385742 lr:9.517001904922074e-05 tokens_per_second_per_gpu:4935.983465095445 mfu:23.28387617047573 +Step 5191 | loss:1.2520145177841187 lr:9.516815951601123e-05 tokens_per_second_per_gpu:5160.69766713752 mfu:24.343891401704578 +Step 5192 | loss:1.270857334136963 lr:9.516629964308623e-05 tokens_per_second_per_gpu:4641.234725715916 mfu:21.89349607750188 +Step 5193 | loss:1.2294057607650757 lr:9.516443943045974e-05 tokens_per_second_per_gpu:3774.270305510903 mfu:17.803876988011783 +Step 5194 | loss:1.1615761518478394 lr:9.516257887814573e-05 tokens_per_second_per_gpu:4666.860003038386 mfu:22.01437488275512 +Step 5195 | loss:1.3106402158737183 lr:9.51607179861582e-05 tokens_per_second_per_gpu:4789.029206351892 mfu:22.5906678590004 +Step 5196 | loss:1.2345865964889526 lr:9.515885675451115e-05 tokens_per_second_per_gpu:5279.055483093068 mfu:24.90220541349263 +Step 5197 | loss:1.282821536064148 lr:9.515699518321857e-05 tokens_per_second_per_gpu:4236.6109823769475 mfu:19.98481684424259 +Step 5198 | loss:1.5476678609848022 lr:9.515513327229448e-05 tokens_per_second_per_gpu:3840.808104012703 mfu:18.117747136063834 +Step 5199 | loss:1.3278907537460327 lr:9.515327102175287e-05 tokens_per_second_per_gpu:5162.883400724354 mfu:24.354201879958282 +Step 5200 | loss:1.2991418838500977 lr:9.515140843160772e-05 tokens_per_second_per_gpu:4972.238363623176 mfu:23.454896712554575 +Step 5201 | loss:1.19594407081604 lr:9.514954550187309e-05 tokens_per_second_per_gpu:4875.237791787631 mfu:22.997328465202173 +Step 5202 | loss:1.3199256658554077 lr:9.514768223256297e-05 tokens_per_second_per_gpu:5346.494378749575 mfu:25.220326190566464 +Step 5203 | loss:1.2629761695861816 lr:9.514581862369136e-05 tokens_per_second_per_gpu:3507.721544660502 mfu:16.54652100517231 +Step 5204 | loss:1.336922287940979 lr:9.51439546752723e-05 tokens_per_second_per_gpu:4528.26298443857 mfu:21.360589098930006 +Step 5205 | loss:1.2298238277435303 lr:9.514209038731979e-05 tokens_per_second_per_gpu:4896.76700871708 mfu:23.098885454721565 +Step 5206 | loss:1.2537732124328613 lr:9.514022575984784e-05 tokens_per_second_per_gpu:4611.331209318145 mfu:21.752436088587586 +Step 5207 | loss:1.332983136177063 lr:9.513836079287052e-05 tokens_per_second_per_gpu:5377.59738520944 mfu:25.36704437875776 +Step 5208 | loss:1.1608431339263916 lr:9.513649548640182e-05 tokens_per_second_per_gpu:4170.923471490045 mfu:19.67495764794404 +Step 5209 | loss:1.4019449949264526 lr:9.513462984045577e-05 tokens_per_second_per_gpu:5390.904071563801 mfu:25.42981428864609 +Step 5210 | loss:1.201140284538269 lr:9.513276385504642e-05 tokens_per_second_per_gpu:4455.859301835596 mfu:21.019048574749686 +Step 5211 | loss:1.4035319089889526 lr:9.513089753018778e-05 tokens_per_second_per_gpu:4682.868370385896 mfu:22.08988908284244 +Step 5212 | loss:1.2984609603881836 lr:9.512903086589391e-05 tokens_per_second_per_gpu:4550.228238343491 mfu:21.46420294925828 +Step 5213 | loss:1.326453447341919 lr:9.512716386217884e-05 tokens_per_second_per_gpu:4098.612803930378 mfu:19.333855891593004 +Step 5214 | loss:1.148146629333496 lr:9.512529651905662e-05 tokens_per_second_per_gpu:4377.744965385814 mfu:20.65056992203596 +Step 5215 | loss:1.1654884815216064 lr:9.512342883654129e-05 tokens_per_second_per_gpu:5005.522614965553 mfu:23.611904205015435 +Step 5216 | loss:1.0633832216262817 lr:9.512156081464686e-05 tokens_per_second_per_gpu:4485.759937845984 mfu:21.16009497638525 +Step 5217 | loss:1.3314367532730103 lr:9.511969245338746e-05 tokens_per_second_per_gpu:5356.396052266297 mfu:25.26703407394455 +Step 5218 | loss:1.3256438970565796 lr:9.511782375277707e-05 tokens_per_second_per_gpu:4349.926273193235 mfu:20.51934440460525 +Step 5219 | loss:1.1894383430480957 lr:9.511595471282978e-05 tokens_per_second_per_gpu:5311.624765422742 mfu:25.055840275134276 +Step 5220 | loss:1.4440505504608154 lr:9.51140853335596e-05 tokens_per_second_per_gpu:5344.766497306524 mfu:25.21217547899262 +Step 5221 | loss:1.3045529127120972 lr:9.511221561498067e-05 tokens_per_second_per_gpu:4485.091248414292 mfu:21.156940654245776 +Step 5222 | loss:1.2754132747650146 lr:9.5110345557107e-05 tokens_per_second_per_gpu:3576.871255309843 mfu:16.872711988462473 +Step 5223 | loss:1.1538652181625366 lr:9.510847515995267e-05 tokens_per_second_per_gpu:3337.5670748457146 mfu:15.743873339709788 +Step 5224 | loss:1.1927645206451416 lr:9.510660442353174e-05 tokens_per_second_per_gpu:5340.05564731303 mfu:25.189953595819418 +Step 5225 | loss:1.3671358823776245 lr:9.510473334785827e-05 tokens_per_second_per_gpu:5309.485924996987 mfu:25.045750999922184 +Step 5226 | loss:1.3571282625198364 lr:9.510286193294635e-05 tokens_per_second_per_gpu:5306.166689567114 mfu:25.030093600079514 +Step 5227 | loss:1.1672611236572266 lr:9.510099017881007e-05 tokens_per_second_per_gpu:3477.762108530624 mfu:16.405197233340775 +Step 5228 | loss:1.1549924612045288 lr:9.509911808546345e-05 tokens_per_second_per_gpu:5311.305832025753 mfu:25.054335811885036 +Step 5229 | loss:1.4535815715789795 lr:9.509724565292064e-05 tokens_per_second_per_gpu:5321.194334608867 mfu:25.100981565722098 +Step 5230 | loss:1.3205372095108032 lr:9.509537288119568e-05 tokens_per_second_per_gpu:3337.0967731734668 mfu:15.741654846480055 +Step 5231 | loss:1.1950498819351196 lr:9.509349977030266e-05 tokens_per_second_per_gpu:4110.402021805455 mfu:19.389467643757726 +Step 5232 | loss:1.29832124710083 lr:9.50916263202557e-05 tokens_per_second_per_gpu:3792.771687949264 mfu:17.89115116563488 +Step 5233 | loss:1.3293143510818481 lr:9.508975253106884e-05 tokens_per_second_per_gpu:5356.3835466055325 mfu:25.266975082609036 +Step 5234 | loss:1.246116280555725 lr:9.50878784027562e-05 tokens_per_second_per_gpu:3915.064889869028 mfu:18.468029064462225 +Step 5235 | loss:1.3744139671325684 lr:9.50860039353319e-05 tokens_per_second_per_gpu:5206.088031807731 mfu:24.558005496248754 +Step 5236 | loss:1.3122204542160034 lr:9.508412912880998e-05 tokens_per_second_per_gpu:5331.071528865177 mfu:25.147573976252225 +Step 5237 | loss:1.3009068965911865 lr:9.508225398320457e-05 tokens_per_second_per_gpu:4582.7595067353905 mfu:21.617658492669598 +Step 5238 | loss:1.3222217559814453 lr:9.508037849852979e-05 tokens_per_second_per_gpu:4701.118627554212 mfu:22.175978659720347 +Step 5239 | loss:1.1935468912124634 lr:9.507850267479972e-05 tokens_per_second_per_gpu:3502.493315116155 mfu:16.52185855438368 +Step 5240 | loss:1.2732518911361694 lr:9.507662651202849e-05 tokens_per_second_per_gpu:4403.871131230971 mfu:20.77381150391111 +Step 5241 | loss:1.2424196004867554 lr:9.50747500102302e-05 tokens_per_second_per_gpu:3572.203977138845 mfu:16.850695641010105 +Step 5242 | loss:1.2964328527450562 lr:9.507287316941896e-05 tokens_per_second_per_gpu:4745.265143705291 mfu:22.38422530857733 +Step 5243 | loss:1.373334288597107 lr:9.507099598960889e-05 tokens_per_second_per_gpu:3867.35149425388 mfu:18.242956836600833 +Step 5244 | loss:1.1982874870300293 lr:9.50691184708141e-05 tokens_per_second_per_gpu:3972.9925548165297 mfu:18.741283743498382 +Step 5245 | loss:1.2520594596862793 lr:9.506724061304872e-05 tokens_per_second_per_gpu:5342.3804059437825 mfu:25.2009198789254 +Step 5246 | loss:1.1851470470428467 lr:9.506536241632689e-05 tokens_per_second_per_gpu:3934.505549909158 mfu:18.559733974790152 +Step 5247 | loss:1.2475554943084717 lr:9.50634838806627e-05 tokens_per_second_per_gpu:4244.9714664642925 mfu:20.024254674128276 +Step 5248 | loss:1.3543195724487305 lr:9.506160500607032e-05 tokens_per_second_per_gpu:4861.684869387527 mfu:22.933397017873876 +Step 5249 | loss:1.3420639038085938 lr:9.505972579256384e-05 tokens_per_second_per_gpu:4386.876830011371 mfu:20.693646485531477 +Step 5250 | loss:1.2889176607131958 lr:9.505784624015742e-05 tokens_per_second_per_gpu:4678.45778949966 mfu:22.06908362027937 +Step 5251 | loss:1.255897045135498 lr:9.505596634886519e-05 tokens_per_second_per_gpu:4384.597627183177 mfu:20.682895096918788 +Step 5252 | loss:1.1683826446533203 lr:9.505408611870128e-05 tokens_per_second_per_gpu:4150.969369016837 mfu:19.580830742056975 +Step 5253 | loss:1.2547547817230225 lr:9.505220554967984e-05 tokens_per_second_per_gpu:5334.962727628253 mfu:25.165929424724624 +Step 5254 | loss:1.1288474798202515 lr:9.505032464181502e-05 tokens_per_second_per_gpu:5354.8947319373965 mfu:25.259952089801462 +Step 5255 | loss:1.285030484199524 lr:9.504844339512095e-05 tokens_per_second_per_gpu:5362.227573240661 mfu:25.294542353340695 +Step 5256 | loss:1.292941927909851 lr:9.50465618096118e-05 tokens_per_second_per_gpu:5346.097277202162 mfu:25.218452994814808 +Step 5257 | loss:1.138081669807434 lr:9.504467988530171e-05 tokens_per_second_per_gpu:5326.685608478502 mfu:25.1268848414731 +Step 5258 | loss:1.2571364641189575 lr:9.504279762220483e-05 tokens_per_second_per_gpu:5324.326131737933 mfu:25.115754787119588 +Step 5259 | loss:1.312484622001648 lr:9.504091502033532e-05 tokens_per_second_per_gpu:4286.507946201436 mfu:20.2201893359026 +Step 5260 | loss:1.2272913455963135 lr:9.503903207970735e-05 tokens_per_second_per_gpu:5312.050516520233 mfu:25.057848615701772 +Step 5261 | loss:1.1356865167617798 lr:9.503714880033507e-05 tokens_per_second_per_gpu:5123.304404062961 mfu:24.167500999833322 +Step 5262 | loss:1.1800706386566162 lr:9.503526518223265e-05 tokens_per_second_per_gpu:4558.883407111969 mfu:21.50503085706326 +Step 5263 | loss:1.257054328918457 lr:9.503338122541424e-05 tokens_per_second_per_gpu:4574.505206480707 mfu:21.57872155003941 +Step 5264 | loss:1.2641091346740723 lr:9.503149692989404e-05 tokens_per_second_per_gpu:4018.5278212928683 mfu:18.956081364584858 +Step 5265 | loss:1.3158321380615234 lr:9.502961229568619e-05 tokens_per_second_per_gpu:5280.7383839330505 mfu:24.91014394388732 +Step 5266 | loss:1.225516438484192 lr:9.50277273228049e-05 tokens_per_second_per_gpu:5046.06077043029 mfu:23.803129600865127 +Step 5267 | loss:1.3686294555664062 lr:9.50258420112643e-05 tokens_per_second_per_gpu:5131.174927309303 mfu:24.20462760083605 +Step 5268 | loss:1.2638862133026123 lr:9.502395636107862e-05 tokens_per_second_per_gpu:5273.26666825807 mfu:24.874898586242292 +Step 5269 | loss:1.313948154449463 lr:9.502207037226201e-05 tokens_per_second_per_gpu:3774.9788516407684 mfu:17.807219320996793 +Step 5270 | loss:1.2893420457839966 lr:9.502018404482866e-05 tokens_per_second_per_gpu:4782.436246741844 mfu:22.5595677436449 +Step 5271 | loss:1.1625195741653442 lr:9.501829737879276e-05 tokens_per_second_per_gpu:5194.362446494967 mfu:24.5026939097373 +Step 5272 | loss:1.2898601293563843 lr:9.501641037416851e-05 tokens_per_second_per_gpu:5127.9738633336965 mfu:24.189527635905286 +Step 5273 | loss:1.2807469367980957 lr:9.501452303097009e-05 tokens_per_second_per_gpu:4345.032002353531 mfu:20.496257294005655 +Step 5274 | loss:1.3287912607192993 lr:9.50126353492117e-05 tokens_per_second_per_gpu:4187.309018612644 mfu:19.75225106938396 +Step 5275 | loss:1.2959104776382446 lr:9.501074732890754e-05 tokens_per_second_per_gpu:4933.510197308561 mfu:23.272209344342 +Step 5276 | loss:1.1246925592422485 lr:9.500885897007178e-05 tokens_per_second_per_gpu:5324.198632577755 mfu:25.11515335182743 +Step 5277 | loss:1.4974088668823242 lr:9.500697027271867e-05 tokens_per_second_per_gpu:5317.724897012413 mfu:25.08461563663253 +Step 5278 | loss:1.174311637878418 lr:9.50050812368624e-05 tokens_per_second_per_gpu:5335.444227438309 mfu:25.168200741480995 +Step 5279 | loss:1.3647606372833252 lr:9.500319186251718e-05 tokens_per_second_per_gpu:5327.496642920247 mfu:25.13071062931155 +Step 5280 | loss:1.2785060405731201 lr:9.500130214969718e-05 tokens_per_second_per_gpu:4824.169334323626 mfu:22.75642983816714 +Step 5281 | loss:1.2188295125961304 lr:9.499941209841666e-05 tokens_per_second_per_gpu:4559.611661942414 mfu:21.508466159351553 +Step 5282 | loss:1.2868096828460693 lr:9.499752170868982e-05 tokens_per_second_per_gpu:5328.365655063742 mfu:25.134809907861182 +Step 5283 | loss:1.276444911956787 lr:9.499563098053088e-05 tokens_per_second_per_gpu:5267.7016859539035 mfu:24.848647615229645 +Step 5284 | loss:1.1369967460632324 lr:9.499373991395406e-05 tokens_per_second_per_gpu:4795.766354793172 mfu:22.62244813767397 +Step 5285 | loss:1.3031193017959595 lr:9.499184850897357e-05 tokens_per_second_per_gpu:4096.257056089472 mfu:19.322743427095425 +Step 5286 | loss:1.2383390665054321 lr:9.498995676560365e-05 tokens_per_second_per_gpu:4632.807364220855 mfu:21.85374277547354 +Step 5287 | loss:1.3532675504684448 lr:9.498806468385852e-05 tokens_per_second_per_gpu:4291.367925597035 mfu:20.243114687909625 +Step 5288 | loss:1.1375480890274048 lr:9.498617226375243e-05 tokens_per_second_per_gpu:4692.74317962266 mfu:22.136470242828445 +Step 5289 | loss:1.3109760284423828 lr:9.498427950529959e-05 tokens_per_second_per_gpu:4514.281495507339 mfu:21.294636030153395 +Step 5290 | loss:1.467441439628601 lr:9.498238640851424e-05 tokens_per_second_per_gpu:3998.288096204314 mfu:18.860607128088958 +Step 5291 | loss:1.265242576599121 lr:9.498049297341063e-05 tokens_per_second_per_gpu:4332.592552332111 mfu:20.43757828586559 +Step 5292 | loss:1.2737258672714233 lr:9.497859920000298e-05 tokens_per_second_per_gpu:5321.8984914878265 mfu:25.1043031938617 +Step 5293 | loss:1.3829524517059326 lr:9.497670508830557e-05 tokens_per_second_per_gpu:3498.7214560765888 mfu:16.50406605174853 +Step 5294 | loss:1.329988718032837 lr:9.49748106383326e-05 tokens_per_second_per_gpu:5342.662784568752 mfu:25.20225190707814 +Step 5295 | loss:1.3738653659820557 lr:9.497291585009834e-05 tokens_per_second_per_gpu:5274.931269556587 mfu:24.882750794577664 +Step 5296 | loss:1.2987993955612183 lr:9.497102072361706e-05 tokens_per_second_per_gpu:5319.941300250575 mfu:25.095070788864327 +Step 5297 | loss:1.326724648475647 lr:9.496912525890299e-05 tokens_per_second_per_gpu:5358.559181773724 mfu:25.277237924899033 +Step 5298 | loss:1.2679170370101929 lr:9.496722945597037e-05 tokens_per_second_per_gpu:5364.658869677116 mfu:25.306011193453404 +Step 5299 | loss:1.2577661275863647 lr:9.49653333148335e-05 tokens_per_second_per_gpu:5095.826142069192 mfu:24.037881349733784 +Step 5300 | loss:1.2156559228897095 lr:9.496343683550662e-05 tokens_per_second_per_gpu:4856.917755769216 mfu:22.910909729582848 +Step 5301 | loss:1.279384970664978 lr:9.496154001800397e-05 tokens_per_second_per_gpu:4757.649130496126 mfu:22.44264268719573 +Step 5302 | loss:1.277551293373108 lr:9.495964286233986e-05 tokens_per_second_per_gpu:4765.562858063116 mfu:22.479973090350374 +Step 5303 | loss:1.2238264083862305 lr:9.495774536852854e-05 tokens_per_second_per_gpu:4952.635055529058 mfu:23.362424563605366 +Step 5304 | loss:1.329986333847046 lr:9.495584753658428e-05 tokens_per_second_per_gpu:5345.024674291353 mfu:25.21339334387959 +Step 5305 | loss:1.2461191415786743 lr:9.495394936652135e-05 tokens_per_second_per_gpu:4294.607066335003 mfu:20.258394267425018 +Step 5306 | loss:1.3973503112792969 lr:9.495205085835404e-05 tokens_per_second_per_gpu:5003.533921231679 mfu:23.60252319736661 +Step 5307 | loss:1.3026303052902222 lr:9.49501520120966e-05 tokens_per_second_per_gpu:4236.281741202133 mfu:19.983263757446657 +Step 5308 | loss:1.285213589668274 lr:9.494825282776334e-05 tokens_per_second_per_gpu:4977.39231060303 mfu:23.479208759812575 +Step 5309 | loss:1.39806067943573 lr:9.494635330536854e-05 tokens_per_second_per_gpu:5304.403331982047 mfu:25.02177554149093 +Step 5310 | loss:1.347615361213684 lr:9.494445344492648e-05 tokens_per_second_per_gpu:5377.622409341715 mfu:25.36716242185926 +Step 5311 | loss:1.3771286010742188 lr:9.494255324645145e-05 tokens_per_second_per_gpu:5294.064591291972 mfu:24.973005937684785 +Step 5312 | loss:1.2186212539672852 lr:9.494065270995775e-05 tokens_per_second_per_gpu:4999.6492936073855 mfu:23.584198746077 +Step 5313 | loss:1.3598904609680176 lr:9.493875183545966e-05 tokens_per_second_per_gpu:3690.352415704725 mfu:17.40802198392763 +Step 5314 | loss:1.2024030685424805 lr:9.493685062297149e-05 tokens_per_second_per_gpu:5335.899799303832 mfu:25.170349751699245 +Step 5315 | loss:1.2983564138412476 lr:9.493494907250751e-05 tokens_per_second_per_gpu:4434.574979614808 mfu:20.918646795354395 +Step 5316 | loss:1.3009848594665527 lr:9.493304718408207e-05 tokens_per_second_per_gpu:5327.61998096565 mfu:25.131292435915363 +Step 5317 | loss:1.25174880027771 lr:9.493114495770943e-05 tokens_per_second_per_gpu:5259.529491733546 mfu:24.81009798077211 +Step 5318 | loss:1.3068125247955322 lr:9.492924239340393e-05 tokens_per_second_per_gpu:5309.5161794707365 mfu:25.045893715436808 +Step 5319 | loss:1.3570343255996704 lr:9.492733949117985e-05 tokens_per_second_per_gpu:3099.5703279134536 mfu:14.621202077998083 +Step 5320 | loss:1.276048183441162 lr:9.492543625105152e-05 tokens_per_second_per_gpu:5122.423078748898 mfu:24.163343637957283 +Step 5321 | loss:1.1758410930633545 lr:9.492353267303327e-05 tokens_per_second_per_gpu:5350.38796811444 mfu:25.238692915915376 +Step 5322 | loss:1.2233248949050903 lr:9.492162875713938e-05 tokens_per_second_per_gpu:5323.260690173324 mfu:25.110728917476255 +Step 5323 | loss:1.1920350790023804 lr:9.491972450338418e-05 tokens_per_second_per_gpu:5352.685084223514 mfu:25.249528804530307 +Step 5324 | loss:1.267871379852295 lr:9.491781991178202e-05 tokens_per_second_per_gpu:4613.01822802529 mfu:21.760394043664263 +Step 5325 | loss:1.246514916419983 lr:9.49159149823472e-05 tokens_per_second_per_gpu:3792.7974687101796 mfu:17.89127277788277 +Step 5326 | loss:1.3235584497451782 lr:9.491400971509405e-05 tokens_per_second_per_gpu:4688.813671877259 mfu:22.11793408435023 +Step 5327 | loss:1.3845279216766357 lr:9.49121041100369e-05 tokens_per_second_per_gpu:4719.594533502634 mfu:22.263132660393662 +Step 5328 | loss:0.9917855262756348 lr:9.49101981671901e-05 tokens_per_second_per_gpu:5375.981514759593 mfu:25.3594220421499 +Step 5329 | loss:1.302285075187683 lr:9.490829188656796e-05 tokens_per_second_per_gpu:3587.18255684833 mfu:16.92135215710897 +Step 5330 | loss:1.1543935537338257 lr:9.49063852681848e-05 tokens_per_second_per_gpu:5138.770672214749 mfu:24.24045802552276 +Step 5331 | loss:1.3325148820877075 lr:9.490447831205503e-05 tokens_per_second_per_gpu:4413.753841871194 mfu:20.82042993616535 +Step 5332 | loss:1.3803489208221436 lr:9.490257101819291e-05 tokens_per_second_per_gpu:5303.0488782802895 mfu:25.015386352286104 +Step 5333 | loss:1.3660168647766113 lr:9.490066338661286e-05 tokens_per_second_per_gpu:5382.839405266694 mfu:25.39177188174867 +Step 5334 | loss:1.2980446815490723 lr:9.489875541732915e-05 tokens_per_second_per_gpu:4669.004935438965 mfu:22.024492894851853 +Step 5335 | loss:1.298326849937439 lr:9.489684711035621e-05 tokens_per_second_per_gpu:5366.744296738295 mfu:25.315848508711532 +Step 5336 | loss:1.2339353561401367 lr:9.489493846570833e-05 tokens_per_second_per_gpu:4744.335778196712 mfu:22.37984133290726 +Step 5337 | loss:1.3366429805755615 lr:9.48930294833999e-05 tokens_per_second_per_gpu:5313.675789465965 mfu:25.065515305486947 +Step 5338 | loss:1.3881072998046875 lr:9.489112016344526e-05 tokens_per_second_per_gpu:4807.724293713089 mfu:22.67885577579398 +Step 5339 | loss:1.3283029794692993 lr:9.488921050585879e-05 tokens_per_second_per_gpu:3902.413841383019 mfu:18.408351910262272 +Step 5340 | loss:1.3408427238464355 lr:9.488730051065481e-05 tokens_per_second_per_gpu:4836.604408324453 mfu:22.815088203871447 +Step 5341 | loss:1.2783291339874268 lr:9.488539017784774e-05 tokens_per_second_per_gpu:4461.735093296527 mfu:21.0467656855843 +Step 5342 | loss:1.1558467149734497 lr:9.488347950745194e-05 tokens_per_second_per_gpu:3352.742911394416 mfu:15.815460349974721 +Step 5343 | loss:1.2244927883148193 lr:9.488156849948173e-05 tokens_per_second_per_gpu:4072.5243306570615 mfu:19.210792112010246 +Step 5344 | loss:1.2512176036834717 lr:9.487965715395153e-05 tokens_per_second_per_gpu:4555.8707825490455 mfu:21.490819792993193 +Step 5345 | loss:1.3478161096572876 lr:9.48777454708757e-05 tokens_per_second_per_gpu:3696.3170521633156 mfu:17.43615819177472 +Step 5346 | loss:1.5069330930709839 lr:9.487583345026863e-05 tokens_per_second_per_gpu:3953.396623009152 mfu:18.64884638975225 +Step 5347 | loss:1.280951738357544 lr:9.487392109214467e-05 tokens_per_second_per_gpu:5149.384091276402 mfu:24.290523334072596 +Step 5348 | loss:1.2751965522766113 lr:9.487200839651823e-05 tokens_per_second_per_gpu:5317.470858137968 mfu:25.083417291918273 +Step 5349 | loss:1.357877492904663 lr:9.48700953634037e-05 tokens_per_second_per_gpu:3925.345604149867 mfu:18.516524947795673 +Step 5350 | loss:1.3738850355148315 lr:9.486818199281544e-05 tokens_per_second_per_gpu:4467.023282966526 mfu:21.071710978515878 +Step 5351 | loss:1.2420353889465332 lr:9.486626828476788e-05 tokens_per_second_per_gpu:3851.864011245228 mfu:18.16989973681194 +Step 5352 | loss:1.2437005043029785 lr:9.486435423927538e-05 tokens_per_second_per_gpu:4177.746446073399 mfu:19.707142783173225 +Step 5353 | loss:1.3022704124450684 lr:9.486243985635235e-05 tokens_per_second_per_gpu:5343.136042233129 mfu:25.204484344227094 +Step 5354 | loss:1.2328275442123413 lr:9.486052513601318e-05 tokens_per_second_per_gpu:3976.152064428738 mfu:18.756187689407096 +Step 5355 | loss:1.199703335762024 lr:9.485861007827228e-05 tokens_per_second_per_gpu:3563.0926247193775 mfu:16.807715837090452 +Step 5356 | loss:1.344970703125 lr:9.485669468314405e-05 tokens_per_second_per_gpu:3954.1603469857273 mfu:18.652449006054503 +Step 5357 | loss:1.1726726293563843 lr:9.485477895064289e-05 tokens_per_second_per_gpu:5336.710564685349 mfu:25.1741742703347 +Step 5358 | loss:1.2227058410644531 lr:9.485286288078322e-05 tokens_per_second_per_gpu:3424.6727286267546 mfu:16.154765570351984 +Step 5359 | loss:1.440138339996338 lr:9.485094647357944e-05 tokens_per_second_per_gpu:3737.2149877126926 mfu:17.629080731668477 +Step 5360 | loss:1.408022403717041 lr:9.484902972904597e-05 tokens_per_second_per_gpu:3981.9019062124394 mfu:18.783310674124984 +Step 5361 | loss:1.3863121271133423 lr:9.484711264719723e-05 tokens_per_second_per_gpu:3147.8198284879536 mfu:14.848803204421676 +Step 5362 | loss:1.206794261932373 lr:9.484519522804763e-05 tokens_per_second_per_gpu:5198.643624849225 mfu:24.522888958478678 +Step 5363 | loss:1.3125016689300537 lr:9.484327747161159e-05 tokens_per_second_per_gpu:4720.061077796395 mfu:22.265333429427997 +Step 5364 | loss:1.3332675695419312 lr:9.484135937790354e-05 tokens_per_second_per_gpu:4336.417349913633 mfu:20.455620508635384 +Step 5365 | loss:1.1351354122161865 lr:9.48394409469379e-05 tokens_per_second_per_gpu:5349.199766677501 mfu:25.23308796700927 +Step 5366 | loss:1.268237829208374 lr:9.483752217872912e-05 tokens_per_second_per_gpu:3952.026696491841 mfu:18.64238421263653 +Step 5367 | loss:1.5252145528793335 lr:9.48356030732916e-05 tokens_per_second_per_gpu:3848.10821324658 mfu:18.1521829968471 +Step 5368 | loss:1.3171770572662354 lr:9.483368363063979e-05 tokens_per_second_per_gpu:5351.36239496287 mfu:25.243289453613787 +Step 5369 | loss:1.5328905582427979 lr:9.483176385078812e-05 tokens_per_second_per_gpu:5344.361917664272 mfu:25.210267007791714 +Step 5370 | loss:1.240839958190918 lr:9.482984373375105e-05 tokens_per_second_per_gpu:5115.285014214344 mfu:24.1296721696684 +Step 5371 | loss:1.267557978630066 lr:9.4827923279543e-05 tokens_per_second_per_gpu:4074.401863412121 mfu:19.21964875435598 +Step 5372 | loss:1.3763171434402466 lr:9.48260024881784e-05 tokens_per_second_per_gpu:4522.070826527849 mfu:21.331379633574056 +Step 5373 | loss:1.2090423107147217 lr:9.482408135967174e-05 tokens_per_second_per_gpu:3948.480989498921 mfu:18.62565850779093 +Step 5374 | loss:1.2250621318817139 lr:9.482215989403742e-05 tokens_per_second_per_gpu:5319.696308781872 mfu:25.0939151223066 +Step 5375 | loss:1.2936584949493408 lr:9.482023809128993e-05 tokens_per_second_per_gpu:4470.268815177577 mfu:21.087020707700304 +Step 5376 | loss:1.140746831893921 lr:9.481831595144373e-05 tokens_per_second_per_gpu:4649.4556045279405 mfu:21.93227536548476 +Step 5377 | loss:1.3652268648147583 lr:9.481639347451322e-05 tokens_per_second_per_gpu:3883.931307705428 mfu:18.321166645459655 +Step 5378 | loss:1.3538910150527954 lr:9.481447066051293e-05 tokens_per_second_per_gpu:5371.505989795898 mfu:25.338310227292858 +Step 5379 | loss:1.2102622985839844 lr:9.481254750945725e-05 tokens_per_second_per_gpu:3931.0981043433017 mfu:18.543660472685843 +Step 5380 | loss:1.3294405937194824 lr:9.481062402136072e-05 tokens_per_second_per_gpu:3676.0662701843617 mfu:17.340631798040306 +Step 5381 | loss:1.1245758533477783 lr:9.480870019623775e-05 tokens_per_second_per_gpu:4844.602708717174 mfu:22.852817551474672 +Step 5382 | loss:1.2947417497634888 lr:9.480677603410282e-05 tokens_per_second_per_gpu:5356.740572912503 mfu:25.268659236613985 +Step 5383 | loss:1.165379524230957 lr:9.480485153497043e-05 tokens_per_second_per_gpu:4332.487388846746 mfu:20.437082211762384 +Step 5384 | loss:1.2434402704238892 lr:9.480292669885503e-05 tokens_per_second_per_gpu:5340.8489209485815 mfu:25.193695602904754 +Step 5385 | loss:1.3359532356262207 lr:9.480100152577109e-05 tokens_per_second_per_gpu:4722.79344074775 mfu:22.27822245165873 +Step 5386 | loss:1.054400086402893 lr:9.479907601573311e-05 tokens_per_second_per_gpu:4339.395709883076 mfu:20.46966994077186 +Step 5387 | loss:1.2435939311981201 lr:9.479715016875554e-05 tokens_per_second_per_gpu:4382.893976716767 mfu:20.67485868699599 +Step 5388 | loss:1.2512236833572388 lr:9.479522398485293e-05 tokens_per_second_per_gpu:4141.600561609518 mfu:19.536636479033277 +Step 5389 | loss:1.2631036043167114 lr:9.479329746403969e-05 tokens_per_second_per_gpu:5364.614760490052 mfu:25.30580312289222 +Step 5390 | loss:1.1933947801589966 lr:9.479137060633035e-05 tokens_per_second_per_gpu:5379.874765386141 mfu:25.377787169612382 +Step 5391 | loss:1.4046988487243652 lr:9.47894434117394e-05 tokens_per_second_per_gpu:5309.077531475826 mfu:25.04382453800375 +Step 5392 | loss:1.2334965467453003 lr:9.478751588028134e-05 tokens_per_second_per_gpu:4587.593286970762 mfu:21.6404602587678 +Step 5393 | loss:1.3262975215911865 lr:9.478558801197065e-05 tokens_per_second_per_gpu:5248.845669143866 mfu:24.75970056676849 +Step 5394 | loss:1.1701968908309937 lr:9.478365980682184e-05 tokens_per_second_per_gpu:4657.79602883666 mfu:21.97161856997211 +Step 5395 | loss:1.3087642192840576 lr:9.47817312648494e-05 tokens_per_second_per_gpu:3735.7505892685326 mfu:17.622172914355232 +Step 5396 | loss:1.3532252311706543 lr:9.477980238606786e-05 tokens_per_second_per_gpu:5333.257952774461 mfu:25.1578877108747 +Step 5397 | loss:1.2448810338974 lr:9.47778731704917e-05 tokens_per_second_per_gpu:5209.750791442158 mfu:24.57528335837558 +Step 5398 | loss:1.1423701047897339 lr:9.477594361813545e-05 tokens_per_second_per_gpu:4287.546002656832 mfu:20.22508602531339 +Step 5399 | loss:1.2367969751358032 lr:9.477401372901362e-05 tokens_per_second_per_gpu:5335.3102705880465 mfu:25.16756884416291 +Step 5400 | loss:1.1280649900436401 lr:9.47720835031407e-05 tokens_per_second_per_gpu:5314.903298745709 mfu:25.07130567619418 +Step 5401 | loss:1.2065430879592896 lr:9.477015294053124e-05 tokens_per_second_per_gpu:4620.789389878663 mfu:21.797051940023557 +Step 5402 | loss:1.3296051025390625 lr:9.476822204119976e-05 tokens_per_second_per_gpu:4740.13496896167 mfu:22.3600253990136 +Step 5403 | loss:1.3876062631607056 lr:9.476629080516075e-05 tokens_per_second_per_gpu:4229.294300433949 mfu:19.9503028071628 +Step 5404 | loss:1.1725356578826904 lr:9.476435923242877e-05 tokens_per_second_per_gpu:5353.722674877869 mfu:25.25442329667812 +Step 5405 | loss:1.2795026302337646 lr:9.476242732301832e-05 tokens_per_second_per_gpu:4888.550473982787 mfu:23.06012665849452 +Step 5406 | loss:1.299597144126892 lr:9.476049507694394e-05 tokens_per_second_per_gpu:5320.088661199539 mfu:25.09576591560296 +Step 5407 | loss:1.3824236392974854 lr:9.475856249422018e-05 tokens_per_second_per_gpu:5008.08962276963 mfu:23.62401321880396 +Step 5408 | loss:1.328253984451294 lr:9.475662957486155e-05 tokens_per_second_per_gpu:4212.030208834739 mfu:19.86886514153157 +Step 5409 | loss:1.289535641670227 lr:9.475469631888258e-05 tokens_per_second_per_gpu:3855.9255788609294 mfu:18.189058844230097 +Step 5410 | loss:1.3203638792037964 lr:9.475276272629785e-05 tokens_per_second_per_gpu:5188.133950692214 mfu:24.473313032344745 +Step 5411 | loss:1.231030821800232 lr:9.475082879712189e-05 tokens_per_second_per_gpu:4760.567936233907 mfu:22.45641119186045 +Step 5412 | loss:1.2134350538253784 lr:9.474889453136922e-05 tokens_per_second_per_gpu:5289.3799556713875 mfu:24.95090771218123 +Step 5413 | loss:1.345362663269043 lr:9.474695992905441e-05 tokens_per_second_per_gpu:4414.181905477969 mfu:20.82244918523369 +Step 5414 | loss:1.3872730731964111 lr:9.4745024990192e-05 tokens_per_second_per_gpu:3707.93299185394 mfu:17.49095256117907 +Step 5415 | loss:1.3173013925552368 lr:9.474308971479656e-05 tokens_per_second_per_gpu:5331.793135178781 mfu:25.150977916352225 +Step 5416 | loss:1.3623104095458984 lr:9.474115410288263e-05 tokens_per_second_per_gpu:3746.150879459791 mfu:17.671232857659884 +Step 5417 | loss:1.343534231185913 lr:9.473921815446476e-05 tokens_per_second_per_gpu:5336.16243097925 mfu:25.17158863012177 +Step 5418 | loss:1.2145673036575317 lr:9.473728186955753e-05 tokens_per_second_per_gpu:4002.2923101191504 mfu:18.879495688314584 +Step 5419 | loss:1.3733975887298584 lr:9.473534524817547e-05 tokens_per_second_per_gpu:5336.82020882554 mfu:25.174691480451198 +Step 5420 | loss:1.3432118892669678 lr:9.47334082903332e-05 tokens_per_second_per_gpu:5022.633768791279 mfu:23.692620437075274 +Step 5421 | loss:1.3242380619049072 lr:9.473147099604525e-05 tokens_per_second_per_gpu:4248.033623766966 mfu:20.038699392582895 +Step 5422 | loss:1.1988614797592163 lr:9.47295333653262e-05 tokens_per_second_per_gpu:5285.685194809564 mfu:24.9334788948048 +Step 5423 | loss:1.1582658290863037 lr:9.472759539819063e-05 tokens_per_second_per_gpu:4971.291356236419 mfu:23.450429517135124 +Step 5424 | loss:1.2127782106399536 lr:9.472565709465309e-05 tokens_per_second_per_gpu:4357.67394950094 mfu:20.55589151563858 +Step 5425 | loss:1.3735731840133667 lr:9.472371845472819e-05 tokens_per_second_per_gpu:4122.590180609507 mfu:19.446961268350766 +Step 5426 | loss:1.2390758991241455 lr:9.472177947843048e-05 tokens_per_second_per_gpu:5276.959368553671 mfu:24.89231768357628 +Step 5427 | loss:1.201267123222351 lr:9.471984016577457e-05 tokens_per_second_per_gpu:4626.786010335591 mfu:21.82533902184784 +Step 5428 | loss:1.1743649244308472 lr:9.471790051677503e-05 tokens_per_second_per_gpu:4695.477682488791 mfu:22.149369359402282 +Step 5429 | loss:1.2400000095367432 lr:9.471596053144647e-05 tokens_per_second_per_gpu:3771.3706686977143 mfu:17.790198906434917 +Step 5430 | loss:1.0890729427337646 lr:9.471402020980345e-05 tokens_per_second_per_gpu:5348.547659583737 mfu:25.23001186658704 +Step 5431 | loss:1.156641960144043 lr:9.471207955186059e-05 tokens_per_second_per_gpu:5332.018013207945 mfu:25.152038704383944 +Step 5432 | loss:1.1847267150878906 lr:9.471013855763247e-05 tokens_per_second_per_gpu:5359.519299395497 mfu:25.281766963533894 +Step 5433 | loss:1.445219874382019 lr:9.470819722713369e-05 tokens_per_second_per_gpu:3763.112094763824 mfu:17.751241804130746 +Step 5434 | loss:1.2518000602722168 lr:9.470625556037886e-05 tokens_per_second_per_gpu:5311.444488296575 mfu:25.0549898771717 +Step 5435 | loss:1.3680378198623657 lr:9.470431355738257e-05 tokens_per_second_per_gpu:5316.504263861138 mfu:25.078857701796473 +Step 5436 | loss:1.3586137294769287 lr:9.470237121815943e-05 tokens_per_second_per_gpu:5132.292222688921 mfu:24.209898073771964 +Step 5437 | loss:1.3591172695159912 lr:9.470042854272405e-05 tokens_per_second_per_gpu:5301.473838143366 mfu:25.007956619230676 +Step 5438 | loss:1.2115256786346436 lr:9.469848553109106e-05 tokens_per_second_per_gpu:5191.974304906521 mfu:24.491428638405626 +Step 5439 | loss:1.3000706434249878 lr:9.469654218327504e-05 tokens_per_second_per_gpu:5356.579785224261 mfu:25.26790077365534 +Step 5440 | loss:1.2438253164291382 lr:9.469459849929063e-05 tokens_per_second_per_gpu:5320.596758371123 mfu:25.098162696651634 +Step 5441 | loss:1.2092819213867188 lr:9.469265447915245e-05 tokens_per_second_per_gpu:4816.771446991999 mfu:22.721532741414123 +Step 5442 | loss:1.3078179359436035 lr:9.46907101228751e-05 tokens_per_second_per_gpu:3719.0144031173086 mfu:17.5432254687921 +Step 5443 | loss:1.3364145755767822 lr:9.468876543047322e-05 tokens_per_second_per_gpu:4171.577221117141 mfu:19.67804149647624 +Step 5444 | loss:1.135186791419983 lr:9.468682040196143e-05 tokens_per_second_per_gpu:4306.505081123282 mfu:20.314519233191984 +Step 5445 | loss:1.087819218635559 lr:9.468487503735439e-05 tokens_per_second_per_gpu:4602.738458161217 mfu:21.71190261530583 +Step 5446 | loss:1.1821644306182861 lr:9.468292933666668e-05 tokens_per_second_per_gpu:5097.739522181294 mfu:24.046907090178877 +Step 5447 | loss:1.3467020988464355 lr:9.468098329991295e-05 tokens_per_second_per_gpu:4386.72958984864 mfu:20.692951928562117 +Step 5448 | loss:1.386459231376648 lr:9.467903692710785e-05 tokens_per_second_per_gpu:4639.3997104507125 mfu:21.884839997410026 +Step 5449 | loss:1.2336293458938599 lr:9.4677090218266e-05 tokens_per_second_per_gpu:3600.5142305870554 mfu:16.98423993675263 +Step 5450 | loss:1.2499462366104126 lr:9.467514317340207e-05 tokens_per_second_per_gpu:4854.806707668901 mfu:22.900951555511718 +Step 5451 | loss:1.3262094259262085 lr:9.467319579253069e-05 tokens_per_second_per_gpu:4324.876917996895 mfu:20.40118232228341 +Step 5452 | loss:1.225767970085144 lr:9.46712480756665e-05 tokens_per_second_per_gpu:3469.5130577490345 mfu:16.366285053370824 +Step 5453 | loss:1.368664026260376 lr:9.466930002282414e-05 tokens_per_second_per_gpu:5226.484176823479 mfu:24.654217592229216 +Step 5454 | loss:1.4299365282058716 lr:9.466735163401828e-05 tokens_per_second_per_gpu:4544.569675589155 mfu:21.437510543295716 +Step 5455 | loss:1.1800347566604614 lr:9.466540290926357e-05 tokens_per_second_per_gpu:4819.621396891355 mfu:22.73497644134098 +Step 5456 | loss:1.1081855297088623 lr:9.466345384857467e-05 tokens_per_second_per_gpu:4518.898014052777 mfu:21.31641293579151 +Step 5457 | loss:1.2457256317138672 lr:9.466150445196625e-05 tokens_per_second_per_gpu:4866.506249494187 mfu:22.956140290449262 +Step 5458 | loss:1.2333283424377441 lr:9.465955471945294e-05 tokens_per_second_per_gpu:5290.886190672974 mfu:24.95801287967391 +Step 5459 | loss:1.1488624811172485 lr:9.465760465104942e-05 tokens_per_second_per_gpu:5346.48905243348 mfu:25.220301065424678 +Step 5460 | loss:1.2337918281555176 lr:9.465565424677035e-05 tokens_per_second_per_gpu:5366.615905596966 mfu:25.315242865792094 +Step 5461 | loss:1.3269184827804565 lr:9.465370350663041e-05 tokens_per_second_per_gpu:4612.5387027504985 mfu:21.758132041127624 +Step 5462 | loss:1.325208067893982 lr:9.465175243064428e-05 tokens_per_second_per_gpu:4800.855392782357 mfu:22.64645399814837 +Step 5463 | loss:1.281101942062378 lr:9.464980101882661e-05 tokens_per_second_per_gpu:4480.626848363059 mfu:21.13588131749892 +Step 5464 | loss:1.215728759765625 lr:9.464784927119209e-05 tokens_per_second_per_gpu:5335.843533352252 mfu:25.17008433560568 +Step 5465 | loss:1.2445170879364014 lr:9.46458971877554e-05 tokens_per_second_per_gpu:4674.2524035124525 mfu:22.049246097064874 +Step 5466 | loss:1.3011797666549683 lr:9.464394476853123e-05 tokens_per_second_per_gpu:4456.562389816127 mfu:21.022365160713033 +Step 5467 | loss:1.546140432357788 lr:9.464199201353424e-05 tokens_per_second_per_gpu:3996.6548464475727 mfu:18.852902810325265 +Step 5468 | loss:1.2268513441085815 lr:9.464003892277915e-05 tokens_per_second_per_gpu:4265.788287149884 mfu:20.12245116901806 +Step 5469 | loss:1.1946876049041748 lr:9.463808549628061e-05 tokens_per_second_per_gpu:5322.239957188635 mfu:25.105913945832473 +Step 5470 | loss:1.3176387548446655 lr:9.463613173405335e-05 tokens_per_second_per_gpu:4513.4723339172515 mfu:21.290819076875746 +Step 5471 | loss:1.1802780628204346 lr:9.463417763611203e-05 tokens_per_second_per_gpu:4227.768623789751 mfu:19.94310593012484 +Step 5472 | loss:1.4662067890167236 lr:9.463222320247138e-05 tokens_per_second_per_gpu:3243.4796508368927 mfu:15.300046907690307 +Step 5473 | loss:1.2186404466629028 lr:9.463026843314608e-05 tokens_per_second_per_gpu:5350.080010134438 mfu:25.237240225580663 +Step 5474 | loss:1.4330296516418457 lr:9.462831332815084e-05 tokens_per_second_per_gpu:4816.925059063238 mfu:22.72225735576308 +Step 5475 | loss:1.3227969408035278 lr:9.462635788750035e-05 tokens_per_second_per_gpu:4459.041996620246 mfu:21.034061889072607 +Step 5476 | loss:1.2730821371078491 lr:9.462440211120933e-05 tokens_per_second_per_gpu:3807.125194026381 mfu:17.958859102761306 +Step 5477 | loss:1.1501964330673218 lr:9.46224459992925e-05 tokens_per_second_per_gpu:4725.323562478728 mfu:22.290157467547033 +Step 5478 | loss:1.3424450159072876 lr:9.462048955176455e-05 tokens_per_second_per_gpu:4350.036468609038 mfu:20.519864215182867 +Step 5479 | loss:1.3358006477355957 lr:9.461853276864021e-05 tokens_per_second_per_gpu:5384.70762238481 mfu:25.400584580645535 +Step 5480 | loss:1.4467157125473022 lr:9.461657564993418e-05 tokens_per_second_per_gpu:4272.7160454071545 mfu:20.155130586715046 +Step 5481 | loss:1.2880784273147583 lr:9.461461819566119e-05 tokens_per_second_per_gpu:3980.8331502983656 mfu:18.778269170129736 +Step 5482 | loss:1.190380573272705 lr:9.461266040583598e-05 tokens_per_second_per_gpu:3886.195061470985 mfu:18.331845158208008 +Step 5483 | loss:1.1618716716766357 lr:9.461070228047327e-05 tokens_per_second_per_gpu:4485.387517367977 mfu:21.158338205449006 +Step 5484 | loss:1.2124079465866089 lr:9.460874381958774e-05 tokens_per_second_per_gpu:4876.196112551556 mfu:23.001849027751987 +Step 5485 | loss:1.383445382118225 lr:9.460678502319418e-05 tokens_per_second_per_gpu:4864.960297237745 mfu:22.94884776988897 +Step 5486 | loss:1.1297798156738281 lr:9.460482589130728e-05 tokens_per_second_per_gpu:4337.449786638435 mfu:20.460490688818147 +Step 5487 | loss:1.1878483295440674 lr:9.46028664239418e-05 tokens_per_second_per_gpu:5324.392706889735 mfu:25.11606883346953 +Step 5488 | loss:1.1117922067642212 lr:9.460090662111249e-05 tokens_per_second_per_gpu:5367.539711184982 mfu:25.31960061436825 +Step 5489 | loss:1.2265673875808716 lr:9.459894648283405e-05 tokens_per_second_per_gpu:4363.669442725551 mfu:20.584173280115696 +Step 5490 | loss:1.2269037961959839 lr:9.459698600912125e-05 tokens_per_second_per_gpu:4618.268418281102 mfu:21.785160086875752 +Step 5491 | loss:1.275399088859558 lr:9.459502519998882e-05 tokens_per_second_per_gpu:4027.211233813742 mfu:18.99704250298877 +Step 5492 | loss:1.2961729764938354 lr:9.459306405545152e-05 tokens_per_second_per_gpu:5333.177749329742 mfu:25.157509377541846 +Step 5493 | loss:1.2691503763198853 lr:9.459110257552409e-05 tokens_per_second_per_gpu:4915.250547978475 mfu:23.186075463034552 +Step 5494 | loss:1.1983925104141235 lr:9.45891407602213e-05 tokens_per_second_per_gpu:3943.846167669852 mfu:18.603795262441068 +Step 5495 | loss:1.1902010440826416 lr:9.458717860955788e-05 tokens_per_second_per_gpu:4014.869914742683 mfu:18.93882639528442 +Step 5496 | loss:1.2613900899887085 lr:9.45852161235486e-05 tokens_per_second_per_gpu:5207.941422968433 mfu:24.566748258574926 +Step 5497 | loss:1.292553186416626 lr:9.458325330220823e-05 tokens_per_second_per_gpu:5357.080108547193 mfu:25.27026088413357 +Step 5498 | loss:1.2298003435134888 lr:9.458129014555153e-05 tokens_per_second_per_gpu:5074.621049510753 mfu:23.937853310173764 +Step 5499 | loss:1.218857765197754 lr:9.457932665359323e-05 tokens_per_second_per_gpu:3967.237731848016 mfu:18.714137261681856 +Step 5500 | loss:1.3492950201034546 lr:9.457736282634815e-05 tokens_per_second_per_gpu:5377.564242524186 mfu:25.366888039056693 +Step 5501 | loss:1.1352585554122925 lr:9.457539866383103e-05 tokens_per_second_per_gpu:5309.922185240183 mfu:25.047808913923017 +Step 5502 | loss:1.2994916439056396 lr:9.457343416605664e-05 tokens_per_second_per_gpu:4009.488081950517 mfu:18.913439371768412 +Step 5503 | loss:1.3188824653625488 lr:9.457146933303977e-05 tokens_per_second_per_gpu:4619.574888465218 mfu:21.791322929641026 +Step 5504 | loss:1.4018856287002563 lr:9.45695041647952e-05 tokens_per_second_per_gpu:5312.055988589741 mfu:25.057874428387272 +Step 5505 | loss:1.260377049446106 lr:9.45675386613377e-05 tokens_per_second_per_gpu:4233.946884404469 mfu:19.972249839564423 +Step 5506 | loss:1.1284751892089844 lr:9.456557282268204e-05 tokens_per_second_per_gpu:5359.4987507394 mfu:25.28167003201712 +Step 5507 | loss:1.339682698249817 lr:9.456360664884304e-05 tokens_per_second_per_gpu:5314.933377970627 mfu:25.07144756503016 +Step 5508 | loss:1.266196846961975 lr:9.456164013983546e-05 tokens_per_second_per_gpu:5299.005722520467 mfu:24.996314096733517 +Step 5509 | loss:1.2001469135284424 lr:9.45596732956741e-05 tokens_per_second_per_gpu:5157.542355173519 mfu:24.329007256818652 +Step 5510 | loss:1.436735987663269 lr:9.455770611637374e-05 tokens_per_second_per_gpu:5266.495667566095 mfu:24.842958620726403 +Step 5511 | loss:1.3046467304229736 lr:9.455573860194919e-05 tokens_per_second_per_gpu:4691.721890309663 mfu:22.131652646889112 +Step 5512 | loss:1.3191487789154053 lr:9.455377075241525e-05 tokens_per_second_per_gpu:4926.6804821241285 mfu:23.23999241254746 +Step 5513 | loss:1.2238892316818237 lr:9.455180256778673e-05 tokens_per_second_per_gpu:5356.112260878563 mfu:25.265695381548802 +Step 5514 | loss:1.2987526655197144 lr:9.45498340480784e-05 tokens_per_second_per_gpu:4398.486445401629 mfu:20.748411022131968 +Step 5515 | loss:1.3863158226013184 lr:9.454786519330508e-05 tokens_per_second_per_gpu:2912.6302511749354 mfu:13.739373840756466 +Step 5516 | loss:1.4245768785476685 lr:9.454589600348158e-05 tokens_per_second_per_gpu:3813.924979996703 mfu:17.990934853346435 +Step 5517 | loss:1.4037442207336426 lr:9.454392647862274e-05 tokens_per_second_per_gpu:5396.001505426784 mfu:25.453859753889674 +Step 5518 | loss:1.2144695520401 lr:9.454195661874333e-05 tokens_per_second_per_gpu:3339.3434399187104 mfu:15.752252756838242 +Step 5519 | loss:1.2453179359436035 lr:9.453998642385816e-05 tokens_per_second_per_gpu:4784.081289134258 mfu:22.567327689283093 +Step 5520 | loss:1.457252860069275 lr:9.453801589398208e-05 tokens_per_second_per_gpu:4458.99615758626 mfu:21.03384565852835 +Step 5521 | loss:1.3050267696380615 lr:9.453604502912991e-05 tokens_per_second_per_gpu:4548.496652692481 mfu:21.45603476430349 +Step 5522 | loss:1.3089017868041992 lr:9.453407382931645e-05 tokens_per_second_per_gpu:5069.448562306428 mfu:23.91345380551364 +Step 5523 | loss:1.2384308576583862 lr:9.453210229455656e-05 tokens_per_second_per_gpu:5358.554239443903 mfu:25.277214611086013 +Step 5524 | loss:1.3193068504333496 lr:9.453013042486503e-05 tokens_per_second_per_gpu:3418.700240357178 mfu:16.12659232417281 +Step 5525 | loss:1.2852146625518799 lr:9.45281582202567e-05 tokens_per_second_per_gpu:5341.700417868315 mfu:25.19771225915549 +Step 5526 | loss:1.3438373804092407 lr:9.452618568074641e-05 tokens_per_second_per_gpu:4951.235286747441 mfu:23.3558216154373 +Step 5527 | loss:1.3962833881378174 lr:9.4524212806349e-05 tokens_per_second_per_gpu:4559.984346815319 mfu:21.510224177483643 +Step 5528 | loss:1.353981375694275 lr:9.452223959707932e-05 tokens_per_second_per_gpu:3650.1877250880784 mfu:17.218558285486047 +Step 5529 | loss:1.4381730556488037 lr:9.45202660529522e-05 tokens_per_second_per_gpu:5118.06653597107 mfu:24.142793082371426 +Step 5530 | loss:1.1792091131210327 lr:9.451829217398244e-05 tokens_per_second_per_gpu:4772.881154889415 mfu:22.514494745110216 +Step 5531 | loss:1.1391319036483765 lr:9.451631796018494e-05 tokens_per_second_per_gpu:4106.128870683934 mfu:19.369310460842755 +Step 5532 | loss:1.3148502111434937 lr:9.451434341157455e-05 tokens_per_second_per_gpu:5366.422191866065 mfu:25.314329085072465 +Step 5533 | loss:1.3185349702835083 lr:9.451236852816608e-05 tokens_per_second_per_gpu:4564.74065264188 mfu:21.532660483576983 +Step 5534 | loss:1.4242373704910278 lr:9.451039330997442e-05 tokens_per_second_per_gpu:5321.784216214028 mfu:25.10376413789782 +Step 5535 | loss:1.2603799104690552 lr:9.450841775701442e-05 tokens_per_second_per_gpu:3994.5585222324366 mfu:18.843014091332677 +Step 5536 | loss:1.359689712524414 lr:9.450644186930091e-05 tokens_per_second_per_gpu:3925.5782159578666 mfu:18.517622217381675 +Step 5537 | loss:1.2775306701660156 lr:9.450446564684879e-05 tokens_per_second_per_gpu:4183.8878744112435 mfu:19.73611294847856 +Step 5538 | loss:1.2056506872177124 lr:9.450248908967289e-05 tokens_per_second_per_gpu:4986.342815746899 mfu:23.521429819689853 +Step 5539 | loss:1.2282718420028687 lr:9.450051219778811e-05 tokens_per_second_per_gpu:4507.698530299713 mfu:21.263583060099336 +Step 5540 | loss:1.3111491203308105 lr:9.449853497120929e-05 tokens_per_second_per_gpu:5280.396466291559 mfu:24.90853106003524 +Step 5541 | loss:1.2878551483154297 lr:9.44965574099513e-05 tokens_per_second_per_gpu:5179.644516601964 mfu:24.43326692329422 +Step 5542 | loss:1.2076407670974731 lr:9.449457951402903e-05 tokens_per_second_per_gpu:4959.713168099812 mfu:23.395813228252145 +Step 5543 | loss:1.280596137046814 lr:9.449260128345735e-05 tokens_per_second_per_gpu:5258.372171732704 mfu:24.804638704868985 +Step 5544 | loss:1.1901435852050781 lr:9.449062271825115e-05 tokens_per_second_per_gpu:3844.5198825436187 mfu:18.13525622868888 +Step 5545 | loss:1.2399959564208984 lr:9.448864381842528e-05 tokens_per_second_per_gpu:5296.36024904098 mfu:24.98383494696646 +Step 5546 | loss:1.1970385313034058 lr:9.448666458399465e-05 tokens_per_second_per_gpu:5295.347267770697 mfu:24.9790565414821 +Step 5547 | loss:1.2584989070892334 lr:9.448468501497415e-05 tokens_per_second_per_gpu:4471.502485771591 mfu:21.092840142378176 +Step 5548 | loss:1.3040410280227661 lr:9.448270511137865e-05 tokens_per_second_per_gpu:5150.188735121112 mfu:24.294318976374488 +Step 5549 | loss:1.4152708053588867 lr:9.448072487322304e-05 tokens_per_second_per_gpu:4499.766403184449 mfu:21.2261658631362 +Step 5550 | loss:1.2275798320770264 lr:9.447874430052224e-05 tokens_per_second_per_gpu:5334.011314542083 mfu:25.161441446869347 +Step 5551 | loss:1.295509696006775 lr:9.447676339329111e-05 tokens_per_second_per_gpu:4092.433844643275 mfu:19.304708686396573 +Step 5552 | loss:1.4504170417785645 lr:9.447478215154457e-05 tokens_per_second_per_gpu:5359.97223276616 mfu:25.283903527521716 +Step 5553 | loss:1.214911937713623 lr:9.447280057529754e-05 tokens_per_second_per_gpu:5351.645175284594 mfu:25.244623376638657 +Step 5554 | loss:1.2357304096221924 lr:9.447081866456489e-05 tokens_per_second_per_gpu:5346.823077508372 mfu:25.2218767186935 +Step 5555 | loss:1.279484748840332 lr:9.446883641936152e-05 tokens_per_second_per_gpu:3913.1282648138276 mfu:18.458893673654536 +Step 5556 | loss:1.3680795431137085 lr:9.446685383970238e-05 tokens_per_second_per_gpu:5127.43523983569 mfu:24.186986856966545 +Step 5557 | loss:1.2360302209854126 lr:9.446487092560234e-05 tokens_per_second_per_gpu:5350.865909079746 mfu:25.240947444993864 +Step 5558 | loss:1.2016295194625854 lr:9.446288767707635e-05 tokens_per_second_per_gpu:5310.319600338151 mfu:25.049683588746134 +Step 5559 | loss:1.3200767040252686 lr:9.44609040941393e-05 tokens_per_second_per_gpu:4010.2817021269766 mfu:18.91718301354634 +Step 5560 | loss:1.5186749696731567 lr:9.445892017680613e-05 tokens_per_second_per_gpu:4019.7501523070973 mfu:18.961847308530153 +Step 5561 | loss:1.516387939453125 lr:9.445693592509173e-05 tokens_per_second_per_gpu:5341.043873748276 mfu:25.194615228523293 +Step 5562 | loss:1.1713366508483887 lr:9.445495133901104e-05 tokens_per_second_per_gpu:4731.4705233441255 mfu:22.319153730728786 +Step 5563 | loss:1.0789077281951904 lr:9.4452966418579e-05 tokens_per_second_per_gpu:4775.0102277441165 mfu:22.52453794502313 +Step 5564 | loss:1.201022744178772 lr:9.445098116381052e-05 tokens_per_second_per_gpu:5223.535247064135 mfu:24.64030698741509 +Step 5565 | loss:1.3786377906799316 lr:9.444899557472055e-05 tokens_per_second_per_gpu:4336.490069968022 mfu:20.45596354153926 +Step 5566 | loss:1.2413160800933838 lr:9.444700965132401e-05 tokens_per_second_per_gpu:3505.084162460646 mfu:16.53408002335178 +Step 5567 | loss:1.2221219539642334 lr:9.444502339363583e-05 tokens_per_second_per_gpu:5343.787260475182 mfu:25.207556251784144 +Step 5568 | loss:1.3197877407073975 lr:9.444303680167097e-05 tokens_per_second_per_gpu:5307.840932522805 mfu:25.037991289756814 +Step 5569 | loss:1.3976553678512573 lr:9.444104987544435e-05 tokens_per_second_per_gpu:4525.198678256268 mfu:21.34613424384354 +Step 5570 | loss:1.2967597246170044 lr:9.443906261497093e-05 tokens_per_second_per_gpu:5331.161428574936 mfu:25.14799804852163 +Step 5571 | loss:1.2703778743743896 lr:9.443707502026564e-05 tokens_per_second_per_gpu:5348.11155904105 mfu:25.22795470592076 +Step 5572 | loss:1.1563704013824463 lr:9.443508709134343e-05 tokens_per_second_per_gpu:4180.86437536468 mfu:19.72185058761379 +Step 5573 | loss:1.13885498046875 lr:9.443309882821929e-05 tokens_per_second_per_gpu:5344.746992061858 mfu:25.21208346942559 +Step 5574 | loss:1.2954634428024292 lr:9.443111023090812e-05 tokens_per_second_per_gpu:5331.52734959415 mfu:25.149724160401828 +Step 5575 | loss:1.382248878479004 lr:9.442912129942492e-05 tokens_per_second_per_gpu:3455.8609152865124 mfu:16.301885568079417 +Step 5576 | loss:1.154535174369812 lr:9.44271320337846e-05 tokens_per_second_per_gpu:4361.320405197648 mfu:20.573092469309447 +Step 5577 | loss:1.1472432613372803 lr:9.442514243400218e-05 tokens_per_second_per_gpu:4809.692713842415 mfu:22.68814114939112 +Step 5578 | loss:1.1835980415344238 lr:9.442315250009256e-05 tokens_per_second_per_gpu:5340.702189559944 mfu:25.19300344590959 +Step 5579 | loss:1.24229896068573 lr:9.442116223207078e-05 tokens_per_second_per_gpu:4556.34648241777 mfu:21.493063750436736 +Step 5580 | loss:1.2602481842041016 lr:9.441917162995175e-05 tokens_per_second_per_gpu:4952.226672670696 mfu:23.360498151985173 +Step 5581 | loss:1.5337709188461304 lr:9.441718069375048e-05 tokens_per_second_per_gpu:4622.930255519588 mfu:21.807150768521144 +Step 5582 | loss:1.3080230951309204 lr:9.441518942348191e-05 tokens_per_second_per_gpu:5335.252686287848 mfu:25.16729720919398 +Step 5583 | loss:1.3852888345718384 lr:9.441319781916104e-05 tokens_per_second_per_gpu:5374.852790959969 mfu:25.354097659406936 +Step 5584 | loss:1.3129702806472778 lr:9.441120588080283e-05 tokens_per_second_per_gpu:4739.956496657293 mfu:22.359183514703346 +Step 5585 | loss:1.2560511827468872 lr:9.440921360842229e-05 tokens_per_second_per_gpu:4347.939001920407 mfu:20.509970106947875 +Step 5586 | loss:1.253968596458435 lr:9.440722100203439e-05 tokens_per_second_per_gpu:5324.9080745052 mfu:25.118499910442463 +Step 5587 | loss:1.409247875213623 lr:9.440522806165409e-05 tokens_per_second_per_gpu:3957.126616514686 mfu:18.666441405520374 +Step 5588 | loss:1.2140076160430908 lr:9.440323478729643e-05 tokens_per_second_per_gpu:3573.227571906217 mfu:16.855524112171075 +Step 5589 | loss:1.3517355918884277 lr:9.440124117897635e-05 tokens_per_second_per_gpu:4862.289350868424 mfu:22.936248460154438 +Step 5590 | loss:1.2214281558990479 lr:9.439924723670889e-05 tokens_per_second_per_gpu:5323.514440056176 mfu:25.111925898967858 +Step 5591 | loss:1.3302518129348755 lr:9.439725296050902e-05 tokens_per_second_per_gpu:5321.636690490169 mfu:25.103068233887893 +Step 5592 | loss:1.187744379043579 lr:9.439525835039173e-05 tokens_per_second_per_gpu:5231.6213184934895 mfu:24.678450365973408 +Step 5593 | loss:1.3447948694229126 lr:9.439326340637206e-05 tokens_per_second_per_gpu:4894.3832825652735 mfu:23.087641011757874 +Step 5594 | loss:1.3123339414596558 lr:9.4391268128465e-05 tokens_per_second_per_gpu:5138.817707567326 mfu:24.240679899305924 +Step 5595 | loss:1.3558138608932495 lr:9.438927251668552e-05 tokens_per_second_per_gpu:4889.1052301242335 mfu:23.062743537864883 +Step 5596 | loss:1.4070086479187012 lr:9.438727657104867e-05 tokens_per_second_per_gpu:3923.001488217909 mfu:18.505467352997204 +Step 5597 | loss:1.2445147037506104 lr:9.438528029156944e-05 tokens_per_second_per_gpu:5318.1910425831475 mfu:25.086814524822316 +Step 5598 | loss:1.459125280380249 lr:9.438328367826287e-05 tokens_per_second_per_gpu:4827.844364960535 mfu:22.773765584706325 +Step 5599 | loss:1.3534129858016968 lr:9.438128673114394e-05 tokens_per_second_per_gpu:4645.079143339465 mfu:21.911630851357444 +Step 5600 | loss:1.2411657571792603 lr:9.437928945022771e-05 tokens_per_second_per_gpu:4920.672019001389 mfu:23.21164946684798 +Step 5601 | loss:1.2839796543121338 lr:9.437729183552917e-05 tokens_per_second_per_gpu:5319.043285405208 mfu:25.0908346996216 +Step 5602 | loss:1.279923439025879 lr:9.437529388706337e-05 tokens_per_second_per_gpu:4447.05451459852 mfu:20.97751489110079 +Step 5603 | loss:1.3083864450454712 lr:9.437329560484529e-05 tokens_per_second_per_gpu:4591.941550999454 mfu:21.66097176208148 +Step 5604 | loss:1.27206289768219 lr:9.437129698889002e-05 tokens_per_second_per_gpu:4746.508795847737 mfu:22.390091827922188 +Step 5605 | loss:1.221975326538086 lr:9.436929803921255e-05 tokens_per_second_per_gpu:5367.681335484521 mfu:25.320268680352772 +Step 5606 | loss:1.3080826997756958 lr:9.436729875582794e-05 tokens_per_second_per_gpu:4465.043056426009 mfu:21.062369912062774 +Step 5607 | loss:1.118611216545105 lr:9.43652991387512e-05 tokens_per_second_per_gpu:4782.173942937997 mfu:22.55833041184851 +Step 5608 | loss:1.2785080671310425 lr:9.43632991879974e-05 tokens_per_second_per_gpu:4698.186352247127 mfu:22.162146616799067 +Step 5609 | loss:1.2884868383407593 lr:9.436129890358155e-05 tokens_per_second_per_gpu:4339.90865864877 mfu:20.472089607617722 +Step 5610 | loss:1.2348196506500244 lr:9.435929828551873e-05 tokens_per_second_per_gpu:5219.070778227721 mfu:24.619247326196803 +Step 5611 | loss:1.3224241733551025 lr:9.435729733382395e-05 tokens_per_second_per_gpu:5288.628597003182 mfu:24.9473634251483 +Step 5612 | loss:1.1754074096679688 lr:9.435529604851227e-05 tokens_per_second_per_gpu:5327.637865754791 mfu:25.1313768015172 +Step 5613 | loss:1.2536861896514893 lr:9.435329442959877e-05 tokens_per_second_per_gpu:3620.8158525606946 mfu:17.08000615141617 +Step 5614 | loss:1.304714322090149 lr:9.435129247709848e-05 tokens_per_second_per_gpu:4047.1498008772514 mfu:19.09109612569767 +Step 5615 | loss:1.3157914876937866 lr:9.434929019102646e-05 tokens_per_second_per_gpu:5338.776674219988 mfu:25.183920461524227 +Step 5616 | loss:1.2289096117019653 lr:9.434728757139775e-05 tokens_per_second_per_gpu:5326.07433381283 mfu:25.12400135458091 +Step 5617 | loss:1.280871868133545 lr:9.434528461822744e-05 tokens_per_second_per_gpu:3094.8617817172385 mfu:14.598991062229715 +Step 5618 | loss:1.3535799980163574 lr:9.434328133153058e-05 tokens_per_second_per_gpu:3364.4809047884337 mfu:15.87083046752249 +Step 5619 | loss:1.4547975063323975 lr:9.434127771132226e-05 tokens_per_second_per_gpu:2811.587421724891 mfu:13.262737574556203 +Step 5620 | loss:1.2415064573287964 lr:9.433927375761753e-05 tokens_per_second_per_gpu:4578.416304577325 mfu:21.5971708670633 +Step 5621 | loss:1.3003249168395996 lr:9.433726947043147e-05 tokens_per_second_per_gpu:5353.415896826194 mfu:25.252976172266628 +Step 5622 | loss:1.4506844282150269 lr:9.433526484977912e-05 tokens_per_second_per_gpu:4690.457539375111 mfu:22.12568848781032 +Step 5623 | loss:1.0794109106063843 lr:9.433325989567562e-05 tokens_per_second_per_gpu:5354.084914492469 mfu:25.256132042744635 +Step 5624 | loss:1.2457228899002075 lr:9.4331254608136e-05 tokens_per_second_per_gpu:5032.272245905754 mfu:23.738086778118017 +Step 5625 | loss:1.3455140590667725 lr:9.432924898717536e-05 tokens_per_second_per_gpu:4674.550758147498 mfu:22.050653486782927 +Step 5626 | loss:1.13655686378479 lr:9.432724303280878e-05 tokens_per_second_per_gpu:4665.934203703082 mfu:22.010007729332838 +Step 5627 | loss:1.1952005624771118 lr:9.432523674505134e-05 tokens_per_second_per_gpu:3775.7666987038933 mfu:17.810935730014123 +Step 5628 | loss:1.1549718379974365 lr:9.432323012391816e-05 tokens_per_second_per_gpu:4003.192935829311 mfu:18.883744093452005 +Step 5629 | loss:1.2553470134735107 lr:9.432122316942429e-05 tokens_per_second_per_gpu:4768.560407355397 mfu:22.49411304179671 +Step 5630 | loss:1.4114147424697876 lr:9.431921588158486e-05 tokens_per_second_per_gpu:4020.5043728676196 mfu:18.965405095597433 +Step 5631 | loss:1.2920138835906982 lr:9.431720826041493e-05 tokens_per_second_per_gpu:4452.369350044801 mfu:21.002585876705332 +Step 5632 | loss:1.109196424484253 lr:9.431520030592965e-05 tokens_per_second_per_gpu:3889.3247540587786 mfu:18.34660845212525 +Step 5633 | loss:1.1965364217758179 lr:9.431319201814406e-05 tokens_per_second_per_gpu:3990.3137960117906 mfu:18.82299099352492 +Step 5634 | loss:1.3992664813995361 lr:9.431118339707332e-05 tokens_per_second_per_gpu:4856.236739976515 mfu:22.907697261895404 +Step 5635 | loss:1.3345720767974854 lr:9.430917444273252e-05 tokens_per_second_per_gpu:4001.3775135196793 mfu:18.875180436674484 +Step 5636 | loss:1.3501861095428467 lr:9.430716515513676e-05 tokens_per_second_per_gpu:4229.581178243547 mfu:19.951656058737075 +Step 5637 | loss:1.2833878993988037 lr:9.430515553430116e-05 tokens_per_second_per_gpu:5315.4617736693035 mfu:25.073940097694422 +Step 5638 | loss:1.361431360244751 lr:9.430314558024082e-05 tokens_per_second_per_gpu:4286.914215595662 mfu:20.22210577795187 +Step 5639 | loss:1.1675388813018799 lr:9.430113529297088e-05 tokens_per_second_per_gpu:3888.393922801604 mfu:18.342217562268967 +Step 5640 | loss:1.252982258796692 lr:9.429912467250646e-05 tokens_per_second_per_gpu:3211.106603161511 mfu:15.147337718394043 +Step 5641 | loss:1.3749003410339355 lr:9.429711371886265e-05 tokens_per_second_per_gpu:5143.090410925273 mfu:24.260834970043728 +Step 5642 | loss:1.4611457586288452 lr:9.42951024320546e-05 tokens_per_second_per_gpu:4296.583763668465 mfu:20.2677186860063 +Step 5643 | loss:1.2775593996047974 lr:9.429309081209743e-05 tokens_per_second_per_gpu:3779.7468449957996 mfu:17.82971075915603 +Step 5644 | loss:1.1465866565704346 lr:9.429107885900628e-05 tokens_per_second_per_gpu:5344.371219132143 mfu:25.21031088440265 +Step 5645 | loss:1.2448381185531616 lr:9.428906657279628e-05 tokens_per_second_per_gpu:3668.46984474698 mfu:17.304798163168346 +Step 5646 | loss:1.416074514389038 lr:9.428705395348254e-05 tokens_per_second_per_gpu:4695.874916615964 mfu:22.151243180555287 +Step 5647 | loss:1.3060394525527954 lr:9.428504100108023e-05 tokens_per_second_per_gpu:4219.961995186928 mfu:19.90628073105714 +Step 5648 | loss:1.311699628829956 lr:9.428302771560446e-05 tokens_per_second_per_gpu:3281.308419075633 mfu:15.478491661725851 +Step 5649 | loss:1.1417670249938965 lr:9.428101409707042e-05 tokens_per_second_per_gpu:5387.6629240859975 mfu:25.4145252429965 +Step 5650 | loss:1.2893142700195312 lr:9.42790001454932e-05 tokens_per_second_per_gpu:4637.561785620529 mfu:21.876170192403738 +Step 5651 | loss:1.3873618841171265 lr:9.427698586088796e-05 tokens_per_second_per_gpu:5173.343536741796 mfu:24.40354412623579 +Step 5652 | loss:1.2641870975494385 lr:9.427497124326988e-05 tokens_per_second_per_gpu:5134.676846755074 mfu:24.221146752350823 +Step 5653 | loss:1.218204140663147 lr:9.42729562926541e-05 tokens_per_second_per_gpu:4980.713349161587 mfu:23.494874665320896 +Step 5654 | loss:1.2102097272872925 lr:9.427094100905574e-05 tokens_per_second_per_gpu:5303.535287733997 mfu:25.017680828669253 +Step 5655 | loss:1.2062625885009766 lr:9.426892539249e-05 tokens_per_second_per_gpu:4427.280848236942 mfu:20.884239133137857 +Step 5656 | loss:1.372400164604187 lr:9.4266909442972e-05 tokens_per_second_per_gpu:5382.64343490681 mfu:25.390847456125112 +Step 5657 | loss:1.2931104898452759 lr:9.426489316051694e-05 tokens_per_second_per_gpu:5254.178821795947 mfu:24.78485795775803 +Step 5658 | loss:1.2532674074172974 lr:9.426287654513999e-05 tokens_per_second_per_gpu:4297.786187754535 mfu:20.273390725575 +Step 5659 | loss:1.3164886236190796 lr:9.426085959685628e-05 tokens_per_second_per_gpu:4796.394133108601 mfu:22.625409475097875 +Step 5660 | loss:1.152605414390564 lr:9.4258842315681e-05 tokens_per_second_per_gpu:4472.641654715355 mfu:21.098213796648174 +Step 5661 | loss:1.3453154563903809 lr:9.425682470162931e-05 tokens_per_second_per_gpu:4741.944953756197 mfu:22.368563406104634 +Step 5662 | loss:1.2147853374481201 lr:9.425480675471641e-05 tokens_per_second_per_gpu:4455.427793279722 mfu:21.01701307526868 +Step 5663 | loss:1.255200743675232 lr:9.425278847495745e-05 tokens_per_second_per_gpu:4285.380693310339 mfu:20.214871891686116 +Step 5664 | loss:1.3018823862075806 lr:9.425076986236762e-05 tokens_per_second_per_gpu:4018.612834680822 mfu:18.956482387240616 +Step 5665 | loss:1.211829423904419 lr:9.42487509169621e-05 tokens_per_second_per_gpu:5359.586524073715 mfu:25.282084073810758 +Step 5666 | loss:1.1812739372253418 lr:9.42467316387561e-05 tokens_per_second_per_gpu:5345.661996851742 mfu:25.216399703135384 +Step 5667 | loss:1.3263640403747559 lr:9.424471202776477e-05 tokens_per_second_per_gpu:5342.483692858534 mfu:25.201407100924847 +Step 5668 | loss:1.2494282722473145 lr:9.424269208400332e-05 tokens_per_second_per_gpu:4376.109917599496 mfu:20.64285712265998 +Step 5669 | loss:1.2834625244140625 lr:9.424067180748692e-05 tokens_per_second_per_gpu:5033.969311877298 mfu:23.746092127855107 +Step 5670 | loss:1.2130545377731323 lr:9.423865119823081e-05 tokens_per_second_per_gpu:3966.1169804080414 mfu:18.708850485919584 +Step 5671 | loss:1.1936115026474 lr:9.423663025625014e-05 tokens_per_second_per_gpu:4101.10451275033 mfu:19.34560970722608 +Step 5672 | loss:1.2400015592575073 lr:9.423460898156013e-05 tokens_per_second_per_gpu:5330.630032456358 mfu:25.145491362364773 +Step 5673 | loss:1.2703626155853271 lr:9.423258737417599e-05 tokens_per_second_per_gpu:4684.668095525142 mfu:22.098378693389037 +Step 5674 | loss:1.2086983919143677 lr:9.42305654341129e-05 tokens_per_second_per_gpu:4936.693618038657 mfu:23.287226083073445 +Step 5675 | loss:1.5069539546966553 lr:9.42285431613861e-05 tokens_per_second_per_gpu:5172.121913865849 mfu:24.397781522699592 +Step 5676 | loss:1.223968744277954 lr:9.422652055601079e-05 tokens_per_second_per_gpu:5367.415846000402 mfu:25.31901632115901 +Step 5677 | loss:1.2642176151275635 lr:9.422449761800217e-05 tokens_per_second_per_gpu:5310.299846229742 mfu:25.049590405246374 +Step 5678 | loss:1.1643977165222168 lr:9.422247434737546e-05 tokens_per_second_per_gpu:4253.584331385853 mfu:20.064883027469563 +Step 5679 | loss:1.3028839826583862 lr:9.422045074414585e-05 tokens_per_second_per_gpu:5036.334214583474 mfu:23.757247777414147 +Step 5680 | loss:1.3403385877609253 lr:9.421842680832861e-05 tokens_per_second_per_gpu:4619.854522255312 mfu:21.79264200993935 +Step 5681 | loss:1.2580562829971313 lr:9.421640253993896e-05 tokens_per_second_per_gpu:3600.6779867266873 mfu:16.98501240240295 +Step 5682 | loss:1.253494381904602 lr:9.421437793899208e-05 tokens_per_second_per_gpu:3954.5913424091686 mfu:18.654482085002282 +Step 5683 | loss:1.3966970443725586 lr:9.421235300550322e-05 tokens_per_second_per_gpu:4663.571796981777 mfu:21.99886385376029 +Step 5684 | loss:1.4196275472640991 lr:9.421032773948762e-05 tokens_per_second_per_gpu:5346.55543390886 mfu:25.220614198169898 +Step 5685 | loss:1.2610112428665161 lr:9.42083021409605e-05 tokens_per_second_per_gpu:4312.139601316623 mfu:20.341098226292246 +Step 5686 | loss:1.3273998498916626 lr:9.420627620993711e-05 tokens_per_second_per_gpu:5180.951523213165 mfu:24.43943229647758 +Step 5687 | loss:1.23846435546875 lr:9.420424994643266e-05 tokens_per_second_per_gpu:5369.666425926421 mfu:25.329632690658574 +Step 5688 | loss:1.2839444875717163 lr:9.420222335046241e-05 tokens_per_second_per_gpu:5338.112443525328 mfu:25.180787172008074 +Step 5689 | loss:1.3017956018447876 lr:9.42001964220416e-05 tokens_per_second_per_gpu:4620.324697230277 mfu:21.79485990551638 +Step 5690 | loss:1.13176429271698 lr:9.419816916118546e-05 tokens_per_second_per_gpu:5331.79102313854 mfu:25.150967953498064 +Step 5691 | loss:1.249684453010559 lr:9.419614156790928e-05 tokens_per_second_per_gpu:4793.726543351923 mfu:22.612825999077884 +Step 5692 | loss:1.3358253240585327 lr:9.419411364222825e-05 tokens_per_second_per_gpu:5345.439057407448 mfu:25.21534805973926 +Step 5693 | loss:1.399304747581482 lr:9.419208538415767e-05 tokens_per_second_per_gpu:4470.514301314418 mfu:21.088178707692038 +Step 5694 | loss:1.3303625583648682 lr:9.419005679371278e-05 tokens_per_second_per_gpu:4482.537483752048 mfu:21.14489411061679 +Step 5695 | loss:1.3113206624984741 lr:9.418802787090881e-05 tokens_per_second_per_gpu:5277.1026964442235 mfu:24.892993785690273 +Step 5696 | loss:1.3043659925460815 lr:9.418599861576107e-05 tokens_per_second_per_gpu:4347.103143925885 mfu:20.506027222175636 +Step 5697 | loss:1.222459077835083 lr:9.418396902828478e-05 tokens_per_second_per_gpu:4458.4260054831175 mfu:21.031156153779847 +Step 5698 | loss:1.202539086341858 lr:9.418193910849522e-05 tokens_per_second_per_gpu:4801.46525293729 mfu:22.649330812552066 +Step 5699 | loss:1.2419040203094482 lr:9.417990885640768e-05 tokens_per_second_per_gpu:4782.898253720105 mfu:22.561747109387873 +Step 5700 | loss:1.214859962463379 lr:9.417787827203739e-05 tokens_per_second_per_gpu:5057.088154358822 mfu:23.855147652321723 +Step 5701 | loss:1.2490766048431396 lr:9.417584735539963e-05 tokens_per_second_per_gpu:5331.587059968213 mfu:25.150005824423925 +Step 5702 | loss:1.2996456623077393 lr:9.41738161065097e-05 tokens_per_second_per_gpu:4490.759005450878 mfu:21.18367643120642 +Step 5703 | loss:1.393039345741272 lr:9.417178452538287e-05 tokens_per_second_per_gpu:3776.068508375141 mfu:17.81235941772744 +Step 5704 | loss:1.5461013317108154 lr:9.41697526120344e-05 tokens_per_second_per_gpu:5055.54104462306 mfu:23.847849671733993 +Step 5705 | loss:1.3778852224349976 lr:9.416772036647958e-05 tokens_per_second_per_gpu:3519.8490757264326 mfu:16.603728638380467 +Step 5706 | loss:1.1198234558105469 lr:9.416568778873371e-05 tokens_per_second_per_gpu:4837.828903879026 mfu:22.82086435832285 +Step 5707 | loss:1.3739521503448486 lr:9.416365487881207e-05 tokens_per_second_per_gpu:4603.660995342897 mfu:21.716254380592893 +Step 5708 | loss:1.2980356216430664 lr:9.416162163672995e-05 tokens_per_second_per_gpu:4519.3431003456835 mfu:21.3185124837747 +Step 5709 | loss:1.2959738969802856 lr:9.415958806250263e-05 tokens_per_second_per_gpu:3411.4177150523615 mfu:16.092239409783055 +Step 5710 | loss:1.239527940750122 lr:9.41575541561454e-05 tokens_per_second_per_gpu:5303.957732908736 mfu:25.019673574635075 +Step 5711 | loss:1.172510027885437 lr:9.415551991767358e-05 tokens_per_second_per_gpu:4707.959820265722 mfu:22.208249732968373 +Step 5712 | loss:1.2879064083099365 lr:9.415348534710246e-05 tokens_per_second_per_gpu:5345.567735084594 mfu:25.21595505429703 +Step 5713 | loss:1.2918729782104492 lr:9.415145044444735e-05 tokens_per_second_per_gpu:5357.418693633871 mfu:25.271858047756005 +Step 5714 | loss:1.3205734491348267 lr:9.414941520972355e-05 tokens_per_second_per_gpu:4370.939619595203 mfu:20.61846794482993 +Step 5715 | loss:1.3029460906982422 lr:9.414737964294636e-05 tokens_per_second_per_gpu:5335.664001985225 mfu:25.169237455516203 +Step 5716 | loss:1.2686409950256348 lr:9.414534374413109e-05 tokens_per_second_per_gpu:4740.402935757261 mfu:22.361289444108255 +Step 5717 | loss:1.3763821125030518 lr:9.414330751329305e-05 tokens_per_second_per_gpu:5378.8560114454585 mfu:25.372981533456837 +Step 5718 | loss:1.1700727939605713 lr:9.414127095044758e-05 tokens_per_second_per_gpu:3683.631706884054 mfu:17.376319253749358 +Step 5719 | loss:1.1202831268310547 lr:9.413923405560996e-05 tokens_per_second_per_gpu:4404.227112136229 mfu:20.77549072657782 +Step 5720 | loss:1.2430994510650635 lr:9.413719682879554e-05 tokens_per_second_per_gpu:3970.7830542269326 mfu:18.730861152237576 +Step 5721 | loss:1.1588196754455566 lr:9.413515927001964e-05 tokens_per_second_per_gpu:5317.909979666912 mfu:25.085488703092153 +Step 5722 | loss:1.189553141593933 lr:9.413312137929755e-05 tokens_per_second_per_gpu:4200.639222578793 mfu:19.815131915859606 +Step 5723 | loss:1.2857131958007812 lr:9.413108315664464e-05 tokens_per_second_per_gpu:5325.802729865021 mfu:25.122720152418687 +Step 5724 | loss:1.254012942314148 lr:9.412904460207622e-05 tokens_per_second_per_gpu:5254.382884985185 mfu:24.785820558638434 +Step 5725 | loss:1.122441053390503 lr:9.412700571560762e-05 tokens_per_second_per_gpu:5351.318834599931 mfu:25.243083971949638 +Step 5726 | loss:1.2139348983764648 lr:9.412496649725419e-05 tokens_per_second_per_gpu:5368.159443932427 mfu:25.32252399947526 +Step 5727 | loss:1.0625669956207275 lr:9.412292694703125e-05 tokens_per_second_per_gpu:4971.706287654457 mfu:23.452386819428607 +Step 5728 | loss:1.208935260772705 lr:9.412088706495414e-05 tokens_per_second_per_gpu:4056.956222758031 mfu:19.13735469085741 +Step 5729 | loss:1.227144718170166 lr:9.411884685103821e-05 tokens_per_second_per_gpu:5309.094022912626 mfu:25.043902330924865 +Step 5730 | loss:1.3073034286499023 lr:9.41168063052988e-05 tokens_per_second_per_gpu:4598.788600982833 mfu:21.693270465080268 +Step 5731 | loss:1.4272772073745728 lr:9.411476542775128e-05 tokens_per_second_per_gpu:4715.908631822384 mfu:22.24574563328825 +Step 5732 | loss:1.2773761749267578 lr:9.411272421841096e-05 tokens_per_second_per_gpu:4890.1540079128135 mfu:23.067690801633557 +Step 5733 | loss:1.1485214233398438 lr:9.411068267729323e-05 tokens_per_second_per_gpu:5163.202021618943 mfu:24.355704869080384 +Step 5734 | loss:1.3009434938430786 lr:9.410864080441341e-05 tokens_per_second_per_gpu:5328.462592244134 mfu:25.135267177080276 +Step 5735 | loss:1.30572509765625 lr:9.410659859978688e-05 tokens_per_second_per_gpu:4119.420363197809 mfu:19.43200869879317 +Step 5736 | loss:1.3269871473312378 lr:9.4104556063429e-05 tokens_per_second_per_gpu:3747.025263440346 mfu:17.67535747608145 +Step 5737 | loss:1.267071008682251 lr:9.410251319535512e-05 tokens_per_second_per_gpu:5325.04138689396 mfu:25.119128767725684 +Step 5738 | loss:1.2985045909881592 lr:9.410046999558062e-05 tokens_per_second_per_gpu:5320.46275843687 mfu:25.09753059609902 +Step 5739 | loss:1.3051341772079468 lr:9.409842646412083e-05 tokens_per_second_per_gpu:3950.1768796382457 mfu:18.633658310926823 +Step 5740 | loss:1.2944011688232422 lr:9.409638260099119e-05 tokens_per_second_per_gpu:5297.619764144055 mfu:24.989776294603093 +Step 5741 | loss:1.3469947576522827 lr:9.409433840620701e-05 tokens_per_second_per_gpu:5330.510090604339 mfu:25.144925576184608 +Step 5742 | loss:1.2995939254760742 lr:9.409229387978366e-05 tokens_per_second_per_gpu:3827.387288618767 mfu:18.05443886002341 +Step 5743 | loss:1.2590296268463135 lr:9.409024902173657e-05 tokens_per_second_per_gpu:4587.2982551090045 mfu:21.63906854313826 +Step 5744 | loss:1.3965166807174683 lr:9.40882038320811e-05 tokens_per_second_per_gpu:5320.87436688499 mfu:25.099472223376445 +Step 5745 | loss:1.478893518447876 lr:9.40861583108326e-05 tokens_per_second_per_gpu:4860.071348022512 mfu:22.925785762299967 +Step 5746 | loss:1.1680797338485718 lr:9.408411245800648e-05 tokens_per_second_per_gpu:4785.284040311616 mfu:22.5730012717972 +Step 5747 | loss:1.2751739025115967 lr:9.408206627361813e-05 tokens_per_second_per_gpu:3292.450393261817 mfu:15.531050254978904 +Step 5748 | loss:1.282746434211731 lr:9.408001975768295e-05 tokens_per_second_per_gpu:4898.323386607736 mfu:23.10622715477724 +Step 5749 | loss:1.2173800468444824 lr:9.40779729102163e-05 tokens_per_second_per_gpu:4505.030142749842 mfu:21.25099582075273 +Step 5750 | loss:1.2432819604873657 lr:9.407592573123358e-05 tokens_per_second_per_gpu:4725.66403196733 mfu:22.291763520214854 +Step 5751 | loss:1.3266983032226562 lr:9.407387822075022e-05 tokens_per_second_per_gpu:4888.6738853435245 mfu:23.060708810939268 +Step 5752 | loss:1.4352445602416992 lr:9.407183037878159e-05 tokens_per_second_per_gpu:4388.361963839283 mfu:20.7006521151877 +Step 5753 | loss:1.3454086780548096 lr:9.406978220534311e-05 tokens_per_second_per_gpu:5307.7400152614555 mfu:25.037515245816763 +Step 5754 | loss:1.0534090995788574 lr:9.406773370045016e-05 tokens_per_second_per_gpu:5332.81240678175 mfu:25.1557859944091 +Step 5755 | loss:1.0236470699310303 lr:9.406568486411815e-05 tokens_per_second_per_gpu:4208.433074794898 mfu:19.851896846531282 +Step 5756 | loss:1.179816484451294 lr:9.406363569636253e-05 tokens_per_second_per_gpu:5354.211378648409 mfu:25.256728595745212 +Step 5757 | loss:1.3397045135498047 lr:9.406158619719867e-05 tokens_per_second_per_gpu:5374.098505362103 mfu:25.350539565547415 +Step 5758 | loss:1.3698036670684814 lr:9.4059536366642e-05 tokens_per_second_per_gpu:5355.6020064939175 mfu:25.263288424558056 +Step 5759 | loss:1.3570910692214966 lr:9.405748620470793e-05 tokens_per_second_per_gpu:5359.401399438723 mfu:25.281210809321987 +Step 5760 | loss:1.2827949523925781 lr:9.405543571141189e-05 tokens_per_second_per_gpu:4869.455510713408 mfu:22.97005245881618 +Step 5761 | loss:1.3062413930892944 lr:9.40533848867693e-05 tokens_per_second_per_gpu:3281.8968694173436 mfu:15.48126748238768 +Step 5762 | loss:1.349683165550232 lr:9.405133373079557e-05 tokens_per_second_per_gpu:5371.144775486611 mfu:25.336606317766435 +Step 5763 | loss:1.326054334640503 lr:9.404928224350615e-05 tokens_per_second_per_gpu:3080.5731362439938 mfu:14.531589083638114 +Step 5764 | loss:1.333400845527649 lr:9.404723042491645e-05 tokens_per_second_per_gpu:4181.190118324432 mfu:19.72338717273242 +Step 5765 | loss:1.2656917572021484 lr:9.404517827504192e-05 tokens_per_second_per_gpu:4848.787340582109 mfu:22.872557173953393 +Step 5766 | loss:1.3104993104934692 lr:9.404312579389796e-05 tokens_per_second_per_gpu:4126.194810310178 mfu:19.463964921662075 +Step 5767 | loss:1.4076342582702637 lr:9.404107298150005e-05 tokens_per_second_per_gpu:4784.014577598355 mfu:22.567012999586076 +Step 5768 | loss:1.2354817390441895 lr:9.403901983786361e-05 tokens_per_second_per_gpu:4952.516796024016 mfu:23.36186671334293 +Step 5769 | loss:1.3958812952041626 lr:9.403696636300408e-05 tokens_per_second_per_gpu:4538.822877059338 mfu:21.410401914125632 +Step 5770 | loss:1.3638678789138794 lr:9.40349125569369e-05 tokens_per_second_per_gpu:5116.4376457971375 mfu:24.135109329503663 +Step 5771 | loss:1.1609944105148315 lr:9.403285841967753e-05 tokens_per_second_per_gpu:5328.086113758969 mfu:25.133491263831903 +Step 5772 | loss:1.3403412103652954 lr:9.403080395124141e-05 tokens_per_second_per_gpu:4525.227690831257 mfu:21.34627110110982 +Step 5773 | loss:1.2632219791412354 lr:9.4028749151644e-05 tokens_per_second_per_gpu:4775.098416924712 mfu:22.524953948434625 +Step 5774 | loss:1.1701550483703613 lr:9.402669402090075e-05 tokens_per_second_per_gpu:5340.101370066196 mfu:25.190169277847207 +Step 5775 | loss:1.0832358598709106 lr:9.402463855902711e-05 tokens_per_second_per_gpu:5311.211259471755 mfu:25.05388969701208 +Step 5776 | loss:1.1868938207626343 lr:9.402258276603855e-05 tokens_per_second_per_gpu:3843.405264338231 mfu:18.129998384440594 +Step 5777 | loss:1.2841483354568481 lr:9.402052664195053e-05 tokens_per_second_per_gpu:4137.350830098661 mfu:19.516589770417603 +Step 5778 | loss:1.1106576919555664 lr:9.401847018677852e-05 tokens_per_second_per_gpu:4801.812572006794 mfu:22.650969175861658 +Step 5779 | loss:1.3270541429519653 lr:9.401641340053795e-05 tokens_per_second_per_gpu:4854.87148735929 mfu:22.901257132363835 +Step 5780 | loss:1.3290928602218628 lr:9.401435628324436e-05 tokens_per_second_per_gpu:4507.418038886343 mfu:21.2622599342457 +Step 5781 | loss:1.3883459568023682 lr:9.401229883491313e-05 tokens_per_second_per_gpu:5311.829943487912 mfu:25.05680813507467 +Step 5782 | loss:1.1886581182479858 lr:9.401024105555982e-05 tokens_per_second_per_gpu:4016.591455903629 mfu:18.94694719866581 +Step 5783 | loss:1.3254649639129639 lr:9.400818294519987e-05 tokens_per_second_per_gpu:5334.780384006853 mfu:25.16506927875145 +Step 5784 | loss:1.2683864831924438 lr:9.400612450384875e-05 tokens_per_second_per_gpu:5058.158402949849 mfu:23.860196197528836 +Step 5785 | loss:1.3733506202697754 lr:9.400406573152197e-05 tokens_per_second_per_gpu:4779.073547975939 mfu:22.543705319830558 +Step 5786 | loss:1.3343969583511353 lr:9.400200662823497e-05 tokens_per_second_per_gpu:5359.5588884543895 mfu:25.28195371187928 +Step 5787 | loss:1.2154541015625 lr:9.399994719400326e-05 tokens_per_second_per_gpu:3901.3578630913753 mfu:18.403370680492113 +Step 5788 | loss:1.3355522155761719 lr:9.399788742884235e-05 tokens_per_second_per_gpu:4753.7558763616225 mfu:22.42427754318562 +Step 5789 | loss:1.3965413570404053 lr:9.399582733276772e-05 tokens_per_second_per_gpu:5341.600851048661 mfu:25.19724258547933 +Step 5790 | loss:1.4092929363250732 lr:9.399376690579486e-05 tokens_per_second_per_gpu:5211.284082906454 mfu:24.582516155819167 +Step 5791 | loss:1.1552636623382568 lr:9.399170614793925e-05 tokens_per_second_per_gpu:5091.925112115983 mfu:24.01947952586015 +Step 5792 | loss:1.4249345064163208 lr:9.398964505921641e-05 tokens_per_second_per_gpu:3777.9255772089678 mfu:17.821119528265214 +Step 5793 | loss:1.2401176691055298 lr:9.398758363964184e-05 tokens_per_second_per_gpu:4816.807761002265 mfu:22.721704040796357 +Step 5794 | loss:1.578601360321045 lr:9.398552188923104e-05 tokens_per_second_per_gpu:5330.7713003140825 mfu:25.14615774695244 +Step 5795 | loss:1.1273194551467896 lr:9.398345980799952e-05 tokens_per_second_per_gpu:5161.636806395698 mfu:24.348321481819085 +Step 5796 | loss:1.3394275903701782 lr:9.398139739596278e-05 tokens_per_second_per_gpu:2900.813863821068 mfu:13.683633925524697 +Step 5797 | loss:1.2394767999649048 lr:9.397933465313633e-05 tokens_per_second_per_gpu:5354.841316421388 mfu:25.259700119698824 +Step 5798 | loss:1.2056013345718384 lr:9.397727157953571e-05 tokens_per_second_per_gpu:4236.892739428383 mfu:19.986145940326924 +Step 5799 | loss:1.1436820030212402 lr:9.397520817517642e-05 tokens_per_second_per_gpu:5347.803772011916 mfu:25.226502821990454 +Step 5800 | loss:1.3914309740066528 lr:9.397314444007397e-05 tokens_per_second_per_gpu:5111.912513087489 mfu:24.113763506445 +Step 5801 | loss:1.21663236618042 lr:9.39710803742439e-05 tokens_per_second_per_gpu:4907.756023868581 mfu:23.150722514110516 +Step 5802 | loss:1.2489181756973267 lr:9.396901597770169e-05 tokens_per_second_per_gpu:4266.994028816561 mfu:20.128138858180517 +Step 5803 | loss:1.2647666931152344 lr:9.396695125046294e-05 tokens_per_second_per_gpu:4212.293992803317 mfu:19.870109455517536 +Step 5804 | loss:1.1772961616516113 lr:9.39648861925431e-05 tokens_per_second_per_gpu:5299.671944231058 mfu:24.999456778211513 +Step 5805 | loss:1.4374160766601562 lr:9.396282080395778e-05 tokens_per_second_per_gpu:4438.220100815169 mfu:20.935841453978302 +Step 5806 | loss:1.2503031492233276 lr:9.396075508472247e-05 tokens_per_second_per_gpu:4759.3436367540835 mfu:22.45063596232709 +Step 5807 | loss:1.4294888973236084 lr:9.395868903485269e-05 tokens_per_second_per_gpu:5314.314052096699 mfu:25.068526099214953 +Step 5808 | loss:1.3827911615371704 lr:9.3956622654364e-05 tokens_per_second_per_gpu:4490.100283947883 mfu:21.180569129487317 +Step 5809 | loss:1.3150851726531982 lr:9.395455594327196e-05 tokens_per_second_per_gpu:5305.90407200734 mfu:25.028854788242885 +Step 5810 | loss:1.2270087003707886 lr:9.39524889015921e-05 tokens_per_second_per_gpu:5337.214377255338 mfu:25.176550840186625 +Step 5811 | loss:1.1492127180099487 lr:9.395042152933995e-05 tokens_per_second_per_gpu:5274.99655973369 mfu:24.883058779482713 +Step 5812 | loss:1.4046366214752197 lr:9.394835382653107e-05 tokens_per_second_per_gpu:3721.1291852257027 mfu:17.553201256816433 +Step 5813 | loss:1.3690824508666992 lr:9.394628579318102e-05 tokens_per_second_per_gpu:5041.3171085181475 mfu:23.78075293827337 +Step 5814 | loss:1.3317581415176392 lr:9.394421742930533e-05 tokens_per_second_per_gpu:4348.007015506143 mfu:20.51029093863611 +Step 5815 | loss:1.2154496908187866 lr:9.394214873491959e-05 tokens_per_second_per_gpu:4354.78452728684 mfu:20.542261618068405 +Step 5816 | loss:1.3984063863754272 lr:9.394007971003934e-05 tokens_per_second_per_gpu:5333.301091865072 mfu:25.1580912053254 +Step 5817 | loss:1.2409645318984985 lr:9.393801035468015e-05 tokens_per_second_per_gpu:4613.412075127266 mfu:21.76225188764153 +Step 5818 | loss:1.2134950160980225 lr:9.393594066885756e-05 tokens_per_second_per_gpu:4940.502809438013 mfu:23.305194688819277 +Step 5819 | loss:1.4197442531585693 lr:9.393387065258716e-05 tokens_per_second_per_gpu:3697.9472008782204 mfu:17.443847881394976 +Step 5820 | loss:1.1544933319091797 lr:9.393180030588453e-05 tokens_per_second_per_gpu:4883.893053675098 mfu:23.038156812265306 +Step 5821 | loss:1.304005742073059 lr:9.39297296287652e-05 tokens_per_second_per_gpu:5366.838735427407 mfu:25.316293992120976 +Step 5822 | loss:1.2835848331451416 lr:9.392765862124478e-05 tokens_per_second_per_gpu:4773.033478993714 mfu:22.51521328389937 +Step 5823 | loss:1.2337989807128906 lr:9.392558728333884e-05 tokens_per_second_per_gpu:5311.833917874817 mfu:25.056826882935805 +Step 5824 | loss:1.1087034940719604 lr:9.392351561506295e-05 tokens_per_second_per_gpu:5361.4513562844895 mfu:25.290880805522168 +Step 5825 | loss:1.2839092016220093 lr:9.392144361643269e-05 tokens_per_second_per_gpu:4827.661923621828 mfu:22.772904977783938 +Step 5826 | loss:1.2825149297714233 lr:9.391937128746364e-05 tokens_per_second_per_gpu:4585.7682374476335 mfu:21.631851188781642 +Step 5827 | loss:1.2276705503463745 lr:9.391729862817141e-05 tokens_per_second_per_gpu:5380.369139166117 mfu:25.380119215081397 +Step 5828 | loss:1.221702218055725 lr:9.391522563857159e-05 tokens_per_second_per_gpu:5267.523163344785 mfu:24.847805493623337 +Step 5829 | loss:1.3329107761383057 lr:9.391315231867972e-05 tokens_per_second_per_gpu:4221.604730632996 mfu:19.91402979443581 +Step 5830 | loss:1.38478684425354 lr:9.391107866851143e-05 tokens_per_second_per_gpu:4713.3716589599535 mfu:22.23377829944396 +Step 5831 | loss:1.3541440963745117 lr:9.390900468808233e-05 tokens_per_second_per_gpu:4511.209457652903 mfu:21.280144703450315 +Step 5832 | loss:1.3605320453643799 lr:9.390693037740799e-05 tokens_per_second_per_gpu:4924.296370919341 mfu:23.22874615322316 +Step 5833 | loss:1.4168007373809814 lr:9.390485573650402e-05 tokens_per_second_per_gpu:3865.763192758668 mfu:18.235464547455738 +Step 5834 | loss:1.1797701120376587 lr:9.390278076538606e-05 tokens_per_second_per_gpu:4426.644451605847 mfu:20.881237141648523 +Step 5835 | loss:1.3270187377929688 lr:9.390070546406965e-05 tokens_per_second_per_gpu:4666.142522769306 mfu:22.010990405911116 +Step 5836 | loss:1.4466842412948608 lr:9.389862983257044e-05 tokens_per_second_per_gpu:4196.99645836052 mfu:19.797948375522346 +Step 5837 | loss:1.3639928102493286 lr:9.389655387090405e-05 tokens_per_second_per_gpu:4388.426135068604 mfu:20.70095482182515 +Step 5838 | loss:1.1880879402160645 lr:9.389447757908607e-05 tokens_per_second_per_gpu:3922.5094652475145 mfu:18.503146396698078 +Step 5839 | loss:1.2303659915924072 lr:9.389240095713211e-05 tokens_per_second_per_gpu:4580.393904392352 mfu:21.606499542804105 +Step 5840 | loss:1.302827000617981 lr:9.389032400505782e-05 tokens_per_second_per_gpu:4746.423510835111 mfu:22.389689523966506 +Step 5841 | loss:1.2075021266937256 lr:9.388824672287879e-05 tokens_per_second_per_gpu:4367.088822931408 mfu:20.600303079953143 +Step 5842 | loss:1.243321418762207 lr:9.388616911061065e-05 tokens_per_second_per_gpu:4187.035647984431 mfu:19.75096153348884 +Step 5843 | loss:1.2565157413482666 lr:9.388409116826904e-05 tokens_per_second_per_gpu:5361.604235176854 mfu:25.29160196134128 +Step 5844 | loss:1.091528058052063 lr:9.388201289586959e-05 tokens_per_second_per_gpu:4297.919927714309 mfu:20.27402159978364 +Step 5845 | loss:1.2626949548721313 lr:9.38799342934279e-05 tokens_per_second_per_gpu:4616.9821444117615 mfu:21.779092513573126 +Step 5846 | loss:1.4051673412322998 lr:9.387785536095966e-05 tokens_per_second_per_gpu:5349.0756616537155 mfu:25.23250254243757 +Step 5847 | loss:1.376824975013733 lr:9.387577609848044e-05 tokens_per_second_per_gpu:5160.48375664007 mfu:24.342882349391214 +Step 5848 | loss:1.118667721748352 lr:9.387369650600592e-05 tokens_per_second_per_gpu:5323.096263823421 mfu:25.109953290329734 +Step 5849 | loss:1.4118314981460571 lr:9.387161658355172e-05 tokens_per_second_per_gpu:4361.28111455982 mfu:20.57290712866714 +Step 5850 | loss:1.3408445119857788 lr:9.386953633113351e-05 tokens_per_second_per_gpu:5348.791354154769 mfu:25.231161415457176 +Step 5851 | loss:1.2148405313491821 lr:9.38674557487669e-05 tokens_per_second_per_gpu:5350.626084126677 mfu:25.23981614977188 +Step 5852 | loss:1.3099833726882935 lr:9.386537483646758e-05 tokens_per_second_per_gpu:3340.7542699337214 mfu:15.75890788273147 +Step 5853 | loss:1.4240100383758545 lr:9.386329359425117e-05 tokens_per_second_per_gpu:4427.841232461958 mfu:20.886882561138666 +Step 5854 | loss:1.420681357383728 lr:9.386121202213333e-05 tokens_per_second_per_gpu:4079.038243395995 mfu:19.241519349787907 +Step 5855 | loss:1.2381441593170166 lr:9.385913012012973e-05 tokens_per_second_per_gpu:4425.616124340699 mfu:20.876386346488093 +Step 5856 | loss:1.1661733388900757 lr:9.385704788825599e-05 tokens_per_second_per_gpu:5330.258536813099 mfu:25.143738954031825 +Step 5857 | loss:1.1318355798721313 lr:9.38549653265278e-05 tokens_per_second_per_gpu:5301.504381614788 mfu:25.008100697996632 +Step 5858 | loss:1.2308359146118164 lr:9.385288243496085e-05 tokens_per_second_per_gpu:5350.780627999735 mfu:25.240545159589 +Step 5859 | loss:1.3838868141174316 lr:9.385079921357075e-05 tokens_per_second_per_gpu:5308.71986779286 mfu:25.042137377387483 +Step 5860 | loss:1.4071060419082642 lr:9.38487156623732e-05 tokens_per_second_per_gpu:3805.0373185345215 mfu:17.949010237837914 +Step 5861 | loss:1.299002766609192 lr:9.384663178138386e-05 tokens_per_second_per_gpu:5247.945524460551 mfu:24.75545443071677 +Step 5862 | loss:1.2392396926879883 lr:9.384454757061841e-05 tokens_per_second_per_gpu:4571.258787442393 mfu:21.56340763753948 +Step 5863 | loss:1.3587063550949097 lr:9.384246303009253e-05 tokens_per_second_per_gpu:4403.869348372555 mfu:20.773803093863773 +Step 5864 | loss:1.208803653717041 lr:9.384037815982188e-05 tokens_per_second_per_gpu:5322.691168772889 mfu:25.108042387860028 +Step 5865 | loss:1.2651656866073608 lr:9.383829295982216e-05 tokens_per_second_per_gpu:4125.667603540816 mfu:19.461477997380246 +Step 5866 | loss:1.2872381210327148 lr:9.383620743010905e-05 tokens_per_second_per_gpu:3926.1534361443155 mfu:18.520335629143833 +Step 5867 | loss:1.3782267570495605 lr:9.383412157069822e-05 tokens_per_second_per_gpu:5315.252579535166 mfu:25.072953293271496 +Step 5868 | loss:1.3386284112930298 lr:9.383203538160537e-05 tokens_per_second_per_gpu:3556.008789128056 mfu:16.774300175979207 +Step 5869 | loss:1.2274143695831299 lr:9.382994886284619e-05 tokens_per_second_per_gpu:5332.586170216794 mfu:25.15471879793226 +Step 5870 | loss:1.4176877737045288 lr:9.382786201443637e-05 tokens_per_second_per_gpu:3102.7752353056317 mfu:14.636320172980572 +Step 5871 | loss:1.2661020755767822 lr:9.382577483639161e-05 tokens_per_second_per_gpu:5360.215192599205 mfu:25.285049610507407 +Step 5872 | loss:1.4550528526306152 lr:9.382368732872761e-05 tokens_per_second_per_gpu:4699.581354791555 mfu:22.168727081813113 +Step 5873 | loss:1.2382774353027344 lr:9.382159949146005e-05 tokens_per_second_per_gpu:5321.449397407857 mfu:25.10218474046313 +Step 5874 | loss:1.2309736013412476 lr:9.381951132460466e-05 tokens_per_second_per_gpu:5323.524482621858 mfu:25.11197327146362 +Step 5875 | loss:1.2820985317230225 lr:9.381742282817713e-05 tokens_per_second_per_gpu:5365.8957468645585 mfu:25.31184575417958 +Step 5876 | loss:1.3493082523345947 lr:9.381533400219318e-05 tokens_per_second_per_gpu:4558.8648900688395 mfu:21.504943509011483 +Step 5877 | loss:1.3733314275741577 lr:9.381324484666851e-05 tokens_per_second_per_gpu:4236.803592167329 mfu:19.985725417486442 +Step 5878 | loss:1.2779501676559448 lr:9.381115536161885e-05 tokens_per_second_per_gpu:5303.939269884151 mfu:25.01958648139799 +Step 5879 | loss:1.363088607788086 lr:9.380906554705985e-05 tokens_per_second_per_gpu:5297.342042148236 mfu:24.98846623256456 +Step 5880 | loss:1.1544742584228516 lr:9.380697540300732e-05 tokens_per_second_per_gpu:5348.984402114441 mfu:25.23207205562033 +Step 5881 | loss:1.2544443607330322 lr:9.380488492947693e-05 tokens_per_second_per_gpu:4153.039637146832 mfu:19.59059655005033 +Step 5882 | loss:1.1516852378845215 lr:9.38027941264844e-05 tokens_per_second_per_gpu:4725.063565420308 mfu:22.288931017062396 +Step 5883 | loss:1.1988608837127686 lr:9.380070299404548e-05 tokens_per_second_per_gpu:4736.398306552721 mfu:22.34239892489024 +Step 5884 | loss:1.2416168451309204 lr:9.379861153217589e-05 tokens_per_second_per_gpu:4420.699219931534 mfu:20.853192469479666 +Step 5885 | loss:1.3884925842285156 lr:9.379651974089134e-05 tokens_per_second_per_gpu:5351.165773406029 mfu:25.24236195618307 +Step 5886 | loss:1.2189568281173706 lr:9.379442762020758e-05 tokens_per_second_per_gpu:5052.579419655385 mfu:23.833879181456144 +Step 5887 | loss:1.3099623918533325 lr:9.379233517014034e-05 tokens_per_second_per_gpu:3863.127832100273 mfu:18.223033101591195 +Step 5888 | loss:1.3561664819717407 lr:9.379024239070536e-05 tokens_per_second_per_gpu:5210.9682866835765 mfu:24.58102649115546 +Step 5889 | loss:1.3860552310943604 lr:9.378814928191837e-05 tokens_per_second_per_gpu:3815.3755635387856 mfu:17.99777750340951 +Step 5890 | loss:1.2080856561660767 lr:9.378605584379514e-05 tokens_per_second_per_gpu:4461.114454891221 mfu:21.043838028334637 +Step 5891 | loss:1.2928426265716553 lr:9.378396207635139e-05 tokens_per_second_per_gpu:4342.462512237616 mfu:20.48413657072866 +Step 5892 | loss:1.193134069442749 lr:9.378186797960287e-05 tokens_per_second_per_gpu:5005.501701927965 mfu:23.611805554649 +Step 5893 | loss:1.2609446048736572 lr:9.377977355356533e-05 tokens_per_second_per_gpu:5333.361382019343 mfu:25.15837560426966 +Step 5894 | loss:1.388479232788086 lr:9.377767879825454e-05 tokens_per_second_per_gpu:5315.923897027833 mfu:25.07612001242298 +Step 5895 | loss:1.4327023029327393 lr:9.377558371368625e-05 tokens_per_second_per_gpu:4168.9047886817 mfu:19.665435176714272 +Step 5896 | loss:1.2220059633255005 lr:9.377348829987617e-05 tokens_per_second_per_gpu:4614.526744564239 mfu:21.76750997355795 +Step 5897 | loss:1.4610669612884521 lr:9.377139255684014e-05 tokens_per_second_per_gpu:4236.543662437827 mfu:19.984499284603793 +Step 5898 | loss:1.3423168659210205 lr:9.376929648459386e-05 tokens_per_second_per_gpu:3286.441257770804 mfu:15.502704137603557 +Step 5899 | loss:1.194193959236145 lr:9.376720008315312e-05 tokens_per_second_per_gpu:5337.847022003144 mfu:25.17953513340235 +Step 5900 | loss:1.3528193235397339 lr:9.376510335253368e-05 tokens_per_second_per_gpu:5371.8865758274005 mfu:25.340105516538127 +Step 5901 | loss:1.246229887008667 lr:9.376300629275132e-05 tokens_per_second_per_gpu:5337.552177222747 mfu:25.17814430026751 +Step 5902 | loss:1.349797248840332 lr:9.37609089038218e-05 tokens_per_second_per_gpu:5291.913463084025 mfu:24.962858698907585 +Step 5903 | loss:1.3160349130630493 lr:9.37588111857609e-05 tokens_per_second_per_gpu:3966.041195889263 mfu:18.70849299741433 +Step 5904 | loss:1.4163494110107422 lr:9.37567131385844e-05 tokens_per_second_per_gpu:5340.427258336327 mfu:25.19170654841969 +Step 5905 | loss:1.360189437866211 lr:9.375461476230809e-05 tokens_per_second_per_gpu:5366.630530053471 mfu:25.31531185184862 +Step 5906 | loss:1.2132984399795532 lr:9.375251605694772e-05 tokens_per_second_per_gpu:5306.347441256269 mfu:25.030946236636677 +Step 5907 | loss:1.3684072494506836 lr:9.37504170225191e-05 tokens_per_second_per_gpu:5322.237086650583 mfu:25.105900405014715 +Step 5908 | loss:1.2987327575683594 lr:9.374831765903801e-05 tokens_per_second_per_gpu:3939.1635446472424 mfu:18.581706530705546 +Step 5909 | loss:1.2607709169387817 lr:9.374621796652025e-05 tokens_per_second_per_gpu:3464.8694506422207 mfu:16.344380366366916 +Step 5910 | loss:1.2986435890197754 lr:9.37441179449816e-05 tokens_per_second_per_gpu:4811.183864190963 mfu:22.695175160001703 +Step 5911 | loss:1.230865478515625 lr:9.374201759443784e-05 tokens_per_second_per_gpu:4724.080602276072 mfu:22.284294212200226 +Step 5912 | loss:1.160887360572815 lr:9.373991691490481e-05 tokens_per_second_per_gpu:5367.831615168267 mfu:25.320977575261363 +Step 5913 | loss:1.3941630125045776 lr:9.373781590639826e-05 tokens_per_second_per_gpu:4710.4821519997 mfu:22.22014800211172 +Step 5914 | loss:1.2266062498092651 lr:9.373571456893404e-05 tokens_per_second_per_gpu:4494.824224095483 mfu:21.202852761150844 +Step 5915 | loss:1.3136547803878784 lr:9.373361290252792e-05 tokens_per_second_per_gpu:4600.612968200588 mfu:21.701876316515563 +Step 5916 | loss:1.2526662349700928 lr:9.37315109071957e-05 tokens_per_second_per_gpu:4067.4815725471444 mfu:19.187004561622512 +Step 5917 | loss:1.1740055084228516 lr:9.372940858295322e-05 tokens_per_second_per_gpu:4493.656682350762 mfu:21.19734526753783 +Step 5918 | loss:1.5017682313919067 lr:9.372730592981626e-05 tokens_per_second_per_gpu:4789.445458070041 mfu:22.59263138937478 +Step 5919 | loss:1.4023361206054688 lr:9.372520294780067e-05 tokens_per_second_per_gpu:3543.2877772144857 mfu:16.71429299235446 +Step 5920 | loss:1.327921986579895 lr:9.372309963692224e-05 tokens_per_second_per_gpu:4780.112205695818 mfu:22.54860484550851 +Step 5921 | loss:1.2904572486877441 lr:9.372099599719683e-05 tokens_per_second_per_gpu:5357.0861101430455 mfu:25.270289194685184 +Step 5922 | loss:1.21104896068573 lr:9.37188920286402e-05 tokens_per_second_per_gpu:4472.188468348381 mfu:21.096076039233594 +Step 5923 | loss:1.358747124671936 lr:9.371678773126822e-05 tokens_per_second_per_gpu:4020.4339153889573 mfu:18.965072735649297 +Step 5924 | loss:1.363813042640686 lr:9.371468310509669e-05 tokens_per_second_per_gpu:4617.365787449217 mfu:21.780902223237057 +Step 5925 | loss:1.2557538747787476 lr:9.371257815014145e-05 tokens_per_second_per_gpu:4140.156719540519 mfu:19.529825629648947 +Step 5926 | loss:1.324832558631897 lr:9.371047286641834e-05 tokens_per_second_per_gpu:4317.765828787053 mfu:20.367638101203656 +Step 5927 | loss:1.2570273876190186 lr:9.370836725394318e-05 tokens_per_second_per_gpu:4582.53256840481 mfu:21.61658798584458 +Step 5928 | loss:1.3938204050064087 lr:9.370626131273182e-05 tokens_per_second_per_gpu:4379.923174557332 mfu:20.660844906338816 +Step 5929 | loss:1.2800356149673462 lr:9.370415504280007e-05 tokens_per_second_per_gpu:3334.0422469282666 mfu:15.727246125025529 +Step 5930 | loss:1.2877895832061768 lr:9.370204844416382e-05 tokens_per_second_per_gpu:5334.037365985919 mfu:25.161564335974756 +Step 5931 | loss:1.2202351093292236 lr:9.369994151683887e-05 tokens_per_second_per_gpu:5248.032402207836 mfu:24.755864247873586 +Step 5932 | loss:1.2119067907333374 lr:9.36978342608411e-05 tokens_per_second_per_gpu:4361.987167687215 mfu:20.576237701734147 +Step 5933 | loss:1.306500792503357 lr:9.369572667618634e-05 tokens_per_second_per_gpu:3132.9162634193267 mfu:14.778500545182556 +Step 5934 | loss:1.2700821161270142 lr:9.369361876289043e-05 tokens_per_second_per_gpu:4124.66961439582 mfu:19.45677031230879 +Step 5935 | loss:1.3295936584472656 lr:9.369151052096924e-05 tokens_per_second_per_gpu:5118.742409197105 mfu:24.145981291694497 +Step 5936 | loss:1.2446037530899048 lr:9.368940195043862e-05 tokens_per_second_per_gpu:5330.773930066769 mfu:25.146170151944542 +Step 5937 | loss:1.2307952642440796 lr:9.368729305131446e-05 tokens_per_second_per_gpu:5363.775773809504 mfu:25.301845479574187 +Step 5938 | loss:1.1561189889907837 lr:9.368518382361255e-05 tokens_per_second_per_gpu:4472.705879589928 mfu:21.09851675633943 +Step 5939 | loss:1.2917128801345825 lr:9.368307426734884e-05 tokens_per_second_per_gpu:4942.936805505024 mfu:23.316676263549915 +Step 5940 | loss:1.3892863988876343 lr:9.368096438253914e-05 tokens_per_second_per_gpu:5355.968280281895 mfu:25.2650162005087 +Step 5941 | loss:1.2732832431793213 lr:9.367885416919933e-05 tokens_per_second_per_gpu:3519.6770147123857 mfu:16.60291699722057 +Step 5942 | loss:1.279044270515442 lr:9.367674362734529e-05 tokens_per_second_per_gpu:4645.797236499416 mfu:21.91501821931226 +Step 5943 | loss:1.335170030593872 lr:9.367463275699289e-05 tokens_per_second_per_gpu:3745.3479084799114 mfu:17.667445106546815 +Step 5944 | loss:1.246062994003296 lr:9.3672521558158e-05 tokens_per_second_per_gpu:5041.911332789679 mfu:23.783555995547303 +Step 5945 | loss:1.1982016563415527 lr:9.367041003085649e-05 tokens_per_second_per_gpu:5356.44977280007 mfu:25.26728748286802 +Step 5946 | loss:1.084038496017456 lr:9.366829817510428e-05 tokens_per_second_per_gpu:4514.016879641989 mfu:21.293387791968417 +Step 5947 | loss:1.3498433828353882 lr:9.36661859909172e-05 tokens_per_second_per_gpu:4409.568783012916 mfu:20.800688299486183 +Step 5948 | loss:1.2474696636199951 lr:9.366407347831118e-05 tokens_per_second_per_gpu:5173.645806953448 mfu:24.404969986396768 +Step 5949 | loss:1.4153083562850952 lr:9.366196063730208e-05 tokens_per_second_per_gpu:4616.467446501412 mfu:21.776664595713488 +Step 5950 | loss:1.2357428073883057 lr:9.365984746790582e-05 tokens_per_second_per_gpu:4590.5431125671275 mfu:21.654375089398 +Step 5951 | loss:1.3541839122772217 lr:9.365773397013826e-05 tokens_per_second_per_gpu:3638.294480101377 mfu:17.162455819686937 +Step 5952 | loss:1.327804446220398 lr:9.365562014401532e-05 tokens_per_second_per_gpu:3568.528628515772 mfu:16.833358394475933 +Step 5953 | loss:1.2354358434677124 lr:9.365350598955289e-05 tokens_per_second_per_gpu:5262.214929082461 mfu:24.8227656469298 +Step 5954 | loss:1.340340495109558 lr:9.365139150676687e-05 tokens_per_second_per_gpu:5140.608687455516 mfu:24.249128257011193 +Step 5955 | loss:1.3229774236679077 lr:9.364927669567317e-05 tokens_per_second_per_gpu:5331.867474566978 mfu:25.151328587929957 +Step 5956 | loss:1.3811609745025635 lr:9.364716155628768e-05 tokens_per_second_per_gpu:3738.717912991009 mfu:17.63617028663179 +Step 5957 | loss:1.1937867403030396 lr:9.364504608862633e-05 tokens_per_second_per_gpu:5337.599745413506 mfu:25.178368687539287 +Step 5958 | loss:1.3982796669006348 lr:9.364293029270501e-05 tokens_per_second_per_gpu:4039.8157722557066 mfu:19.056500261375533 +Step 5959 | loss:1.1693495512008667 lr:9.364081416853965e-05 tokens_per_second_per_gpu:3595.484163126621 mfu:16.960512250323752 +Step 5960 | loss:1.1661319732666016 lr:9.363869771614616e-05 tokens_per_second_per_gpu:5310.6526480874145 mfu:25.051254631803218 +Step 5961 | loss:1.2880277633666992 lr:9.363658093554043e-05 tokens_per_second_per_gpu:3490.160854298455 mfu:16.4636842325718 +Step 5962 | loss:1.2273836135864258 lr:9.363446382673843e-05 tokens_per_second_per_gpu:4647.448489517396 mfu:21.922807461530887 +Step 5963 | loss:1.2798370122909546 lr:9.363234638975606e-05 tokens_per_second_per_gpu:5345.7016364844085 mfu:25.216586690045894 +Step 5964 | loss:1.6326706409454346 lr:9.363022862460924e-05 tokens_per_second_per_gpu:5342.188210728043 mfu:25.20001326130087 +Step 5965 | loss:1.3538039922714233 lr:9.36281105313139e-05 tokens_per_second_per_gpu:4763.168864608194 mfu:22.468680214766103 +Step 5966 | loss:1.351493239402771 lr:9.362599210988597e-05 tokens_per_second_per_gpu:4162.312502788777 mfu:19.634338239397405 +Step 5967 | loss:1.265920877456665 lr:9.36238733603414e-05 tokens_per_second_per_gpu:3294.9785226048857 mfu:15.542975872433502 +Step 5968 | loss:1.0830879211425781 lr:9.362175428269611e-05 tokens_per_second_per_gpu:5368.506048001511 mfu:25.32415899000588 +Step 5969 | loss:1.307037115097046 lr:9.361963487696603e-05 tokens_per_second_per_gpu:5357.074227681453 mfu:25.270233143086497 +Step 5970 | loss:1.3793699741363525 lr:9.361751514316712e-05 tokens_per_second_per_gpu:4678.863365530393 mfu:22.070996791614743 +Step 5971 | loss:1.3136775493621826 lr:9.361539508131531e-05 tokens_per_second_per_gpu:5126.62468082874 mfu:24.183163311835134 +Step 5972 | loss:1.337027907371521 lr:9.361327469142656e-05 tokens_per_second_per_gpu:4848.6975978879655 mfu:22.87213384235343 +Step 5973 | loss:1.3430694341659546 lr:9.361115397351678e-05 tokens_per_second_per_gpu:3639.5750958023136 mfu:17.168496702443825 +Step 5974 | loss:1.1991876363754272 lr:9.360903292760197e-05 tokens_per_second_per_gpu:5355.440110062726 mfu:25.262524731469743 +Step 5975 | loss:1.1793996095657349 lr:9.360691155369804e-05 tokens_per_second_per_gpu:4642.889437703176 mfu:21.90130163626876 +Step 5976 | loss:1.2177785634994507 lr:9.360478985182097e-05 tokens_per_second_per_gpu:5346.3069062845225 mfu:25.21944185096271 +Step 5977 | loss:1.3131760358810425 lr:9.360266782198672e-05 tokens_per_second_per_gpu:5337.591401930723 mfu:25.17832932990746 +Step 5978 | loss:1.2453051805496216 lr:9.360054546421125e-05 tokens_per_second_per_gpu:5090.085782428251 mfu:24.010803093901114 +Step 5979 | loss:1.289080262184143 lr:9.359842277851049e-05 tokens_per_second_per_gpu:5326.958491086337 mfu:25.12817207529283 +Step 5980 | loss:1.3372961282730103 lr:9.359629976490043e-05 tokens_per_second_per_gpu:4936.489296439023 mfu:23.286262263227112 +Step 5981 | loss:1.2999902963638306 lr:9.359417642339705e-05 tokens_per_second_per_gpu:5041.1696612826245 mfu:23.780057404506316 +Step 5982 | loss:1.2004499435424805 lr:9.35920527540163e-05 tokens_per_second_per_gpu:4285.107453793077 mfu:20.213582974263726 +Step 5983 | loss:1.0900994539260864 lr:9.358992875677415e-05 tokens_per_second_per_gpu:5027.603357629373 mfu:23.71606283552385 +Step 5984 | loss:1.3576927185058594 lr:9.35878044316866e-05 tokens_per_second_per_gpu:4284.289063910763 mfu:20.209722489557134 +Step 5985 | loss:1.1597239971160889 lr:9.358567977876962e-05 tokens_per_second_per_gpu:5346.322606341103 mfu:25.21951591080831 +Step 5986 | loss:1.2962226867675781 lr:9.358355479803915e-05 tokens_per_second_per_gpu:3666.788007945787 mfu:17.296864652026173 +Step 5987 | loss:1.3396732807159424 lr:9.358142948951123e-05 tokens_per_second_per_gpu:4450.7943127602675 mfu:20.995156157105118 +Step 5988 | loss:1.2563197612762451 lr:9.35793038532018e-05 tokens_per_second_per_gpu:5394.922941581476 mfu:25.448771984209387 +Step 5989 | loss:1.2659574747085571 lr:9.357717788912688e-05 tokens_per_second_per_gpu:5294.790846607901 mfu:24.976431807921955 +Step 5990 | loss:1.238812804222107 lr:9.357505159730244e-05 tokens_per_second_per_gpu:5282.51742695781 mfu:24.91853599337115 +Step 5991 | loss:1.1144979000091553 lr:9.357292497774447e-05 tokens_per_second_per_gpu:5135.324780802042 mfu:24.224203167799324 +Step 5992 | loss:1.3012948036193848 lr:9.357079803046898e-05 tokens_per_second_per_gpu:3610.0724730320567 mfu:17.029327797170023 +Step 5993 | loss:1.293454647064209 lr:9.356867075549194e-05 tokens_per_second_per_gpu:5312.702329288511 mfu:25.06092332774046 +Step 5994 | loss:1.3451032638549805 lr:9.356654315282939e-05 tokens_per_second_per_gpu:3794.190761592798 mfu:17.89784517813036 +Step 5995 | loss:1.1702415943145752 lr:9.356441522249731e-05 tokens_per_second_per_gpu:5380.056818657485 mfu:25.3786459459548 +Step 5996 | loss:1.3118302822113037 lr:9.356228696451168e-05 tokens_per_second_per_gpu:4080.671860860915 mfu:19.249225402093817 +Step 5997 | loss:1.1786850690841675 lr:9.356015837888855e-05 tokens_per_second_per_gpu:4527.680574508983 mfu:21.35784177192276 +Step 5998 | loss:1.1878165006637573 lr:9.355802946564392e-05 tokens_per_second_per_gpu:5295.504900923958 mfu:24.979800123961027 +Step 5999 | loss:1.3684635162353516 lr:9.355590022479377e-05 tokens_per_second_per_gpu:5319.796852743549 mfu:25.094389405328396 +Step 6000 | loss:1.503734827041626 lr:9.355377065635414e-05 tokens_per_second_per_gpu:4563.984869139875 mfu:21.529095323848065 +Step 6001 | loss:1.1963889598846436 lr:9.355164076034105e-05 tokens_per_second_per_gpu:4143.951849381098 mfu:19.547727904622878 +Step 6002 | loss:1.1821749210357666 lr:9.354951053677051e-05 tokens_per_second_per_gpu:5352.937897774215 mfu:25.250721369183402 +Step 6003 | loss:1.344520926475525 lr:9.354737998565854e-05 tokens_per_second_per_gpu:4896.474462014621 mfu:23.097505462012453 +Step 6004 | loss:1.2883328199386597 lr:9.354524910702118e-05 tokens_per_second_per_gpu:5343.720403952899 mfu:25.207240878161638 +Step 6005 | loss:1.2394105195999146 lr:9.354311790087444e-05 tokens_per_second_per_gpu:3845.941917633222 mfu:18.141964210830746 +Step 6006 | loss:1.296968936920166 lr:9.354098636723437e-05 tokens_per_second_per_gpu:5372.564065883386 mfu:25.343301352723934 +Step 6007 | loss:1.3629601001739502 lr:9.353885450611696e-05 tokens_per_second_per_gpu:4884.825101942657 mfu:23.042553442967215 +Step 6008 | loss:1.3593448400497437 lr:9.353672231753827e-05 tokens_per_second_per_gpu:3478.6554914666194 mfu:16.409411473076712 +Step 6009 | loss:1.1939051151275635 lr:9.353458980151435e-05 tokens_per_second_per_gpu:5337.873452670561 mfu:25.17965981137009 +Step 6010 | loss:1.2264608144760132 lr:9.353245695806122e-05 tokens_per_second_per_gpu:4600.264352728303 mfu:21.70023183785286 +Step 6011 | loss:1.2716548442840576 lr:9.353032378719493e-05 tokens_per_second_per_gpu:5351.426125246916 mfu:25.24359008023714 +Step 6012 | loss:1.3694584369659424 lr:9.352819028893152e-05 tokens_per_second_per_gpu:4220.469100654282 mfu:19.90867283406776 +Step 6013 | loss:1.296020746231079 lr:9.352605646328702e-05 tokens_per_second_per_gpu:5315.94236358626 mfu:25.076207122329805 +Step 6014 | loss:1.1173924207687378 lr:9.352392231027751e-05 tokens_per_second_per_gpu:5330.299459009471 mfu:25.14393199101416 +Step 6015 | loss:1.1939893960952759 lr:9.352178782991903e-05 tokens_per_second_per_gpu:4888.570742078454 mfu:23.060222266559876 +Step 6016 | loss:1.1208142042160034 lr:9.351965302222763e-05 tokens_per_second_per_gpu:3409.007194023928 mfu:16.080868570814634 +Step 6017 | loss:1.3710911273956299 lr:9.351751788721936e-05 tokens_per_second_per_gpu:2781.7563523636904 mfu:13.122019330674151 +Step 6018 | loss:1.1973321437835693 lr:9.351538242491029e-05 tokens_per_second_per_gpu:3203.370020329344 mfu:15.110842937177566 +Step 6019 | loss:1.2281336784362793 lr:9.351324663531647e-05 tokens_per_second_per_gpu:5377.250250563701 mfu:25.365406885406532 +Step 6020 | loss:1.3053022623062134 lr:9.351111051845399e-05 tokens_per_second_per_gpu:3525.2166893512453 mfu:16.62904858765803 +Step 6021 | loss:1.2833377122879028 lr:9.350897407433888e-05 tokens_per_second_per_gpu:4970.893444040432 mfu:23.448552497415825 +Step 6022 | loss:1.0609506368637085 lr:9.350683730298722e-05 tokens_per_second_per_gpu:4341.664046570066 mfu:20.480370071942534 +Step 6023 | loss:1.1949048042297363 lr:9.350470020441508e-05 tokens_per_second_per_gpu:4870.657701648218 mfu:22.9757233985706 +Step 6024 | loss:1.218271255493164 lr:9.350256277863855e-05 tokens_per_second_per_gpu:3575.3952597205002 mfu:16.86574946543673 +Step 6025 | loss:1.2625325918197632 lr:9.350042502567368e-05 tokens_per_second_per_gpu:3701.81283045095 mfu:17.462082715633958 +Step 6026 | loss:1.3273061513900757 lr:9.349828694553658e-05 tokens_per_second_per_gpu:5388.248677724527 mfu:25.41728834284932 +Step 6027 | loss:1.2436197996139526 lr:9.349614853824331e-05 tokens_per_second_per_gpu:5326.417967646064 mfu:25.12562233400225 +Step 6028 | loss:1.2834075689315796 lr:9.349400980380995e-05 tokens_per_second_per_gpu:4284.141072157378 mfu:20.209024387206423 +Step 6029 | loss:1.3692611455917358 lr:9.349187074225261e-05 tokens_per_second_per_gpu:3955.631063926715 mfu:18.65938662879533 +Step 6030 | loss:1.1572498083114624 lr:9.348973135358734e-05 tokens_per_second_per_gpu:4863.610036166063 mfu:22.94247835803555 +Step 6031 | loss:1.3226584196090698 lr:9.348759163783025e-05 tokens_per_second_per_gpu:4655.878455387627 mfu:21.962573048841257 +Step 6032 | loss:1.2185722589492798 lr:9.348545159499744e-05 tokens_per_second_per_gpu:4365.305583951796 mfu:20.59189123743437 +Step 6033 | loss:1.1593390703201294 lr:9.348331122510501e-05 tokens_per_second_per_gpu:4423.239496055538 mfu:20.865175385372392 +Step 6034 | loss:1.21170973777771 lr:9.348117052816904e-05 tokens_per_second_per_gpu:3740.5696507891294 mfu:17.64490524976457 +Step 6035 | loss:1.291185736656189 lr:9.347902950420564e-05 tokens_per_second_per_gpu:3221.1689002211897 mfu:15.194803290429459 +Step 6036 | loss:1.1498932838439941 lr:9.34768881532309e-05 tokens_per_second_per_gpu:3950.170181039734 mfu:18.633626712494742 +Step 6037 | loss:1.287579894065857 lr:9.347474647526095e-05 tokens_per_second_per_gpu:5018.2261024918025 mfu:23.671828723115222 +Step 6038 | loss:1.2188626527786255 lr:9.347260447031189e-05 tokens_per_second_per_gpu:4592.073329111787 mfu:21.66159338192102 +Step 6039 | loss:1.3512810468673706 lr:9.347046213839981e-05 tokens_per_second_per_gpu:5363.840841873017 mfu:25.30215241673153 +Step 6040 | loss:1.0431944131851196 lr:9.346831947954084e-05 tokens_per_second_per_gpu:5226.482105133094 mfu:24.6542078197122 +Step 6041 | loss:1.3405230045318604 lr:9.34661764937511e-05 tokens_per_second_per_gpu:4826.697309047351 mfu:22.768354726255914 +Step 6042 | loss:1.2430394887924194 lr:9.34640331810467e-05 tokens_per_second_per_gpu:4995.801006525765 mfu:23.566045719327516 +Step 6043 | loss:1.3160666227340698 lr:9.346188954144375e-05 tokens_per_second_per_gpu:3913.93780705456 mfu:18.46271242252622 +Step 6044 | loss:1.2490973472595215 lr:9.345974557495838e-05 tokens_per_second_per_gpu:4478.924526439611 mfu:21.127851174539945 +Step 6045 | loss:1.5042365789413452 lr:9.345760128160674e-05 tokens_per_second_per_gpu:3844.325232799565 mfu:18.13433803263551 +Step 6046 | loss:1.375234842300415 lr:9.345545666140492e-05 tokens_per_second_per_gpu:4013.448642444164 mfu:18.932122011358246 +Step 6047 | loss:1.4757972955703735 lr:9.345331171436907e-05 tokens_per_second_per_gpu:4117.111747650294 mfu:19.421118565365113 +Step 6048 | loss:1.214393973350525 lr:9.34511664405153e-05 tokens_per_second_per_gpu:4454.0338979102335 mfu:21.01043783298773 +Step 6049 | loss:1.4973080158233643 lr:9.344902083985979e-05 tokens_per_second_per_gpu:3639.546502899815 mfu:17.16836182484429 +Step 6050 | loss:1.2502819299697876 lr:9.344687491241864e-05 tokens_per_second_per_gpu:5003.67282102396 mfu:23.60317841138555 +Step 6051 | loss:1.4642128944396973 lr:9.344472865820799e-05 tokens_per_second_per_gpu:3915.657322816079 mfu:18.470823671752946 +Step 6052 | loss:1.410826325416565 lr:9.344258207724401e-05 tokens_per_second_per_gpu:5378.461267578249 mfu:25.371119459284962 +Step 6053 | loss:1.3042805194854736 lr:9.344043516954282e-05 tokens_per_second_per_gpu:3730.31865438353 mfu:17.59654955071902 +Step 6054 | loss:1.233804702758789 lr:9.343828793512057e-05 tokens_per_second_per_gpu:3684.8835760878005 mfu:17.38222453437434 +Step 6055 | loss:1.2734006643295288 lr:9.343614037399341e-05 tokens_per_second_per_gpu:3975.1380772911907 mfu:18.75140453907524 +Step 6056 | loss:1.1223615407943726 lr:9.34339924861775e-05 tokens_per_second_per_gpu:5071.336821212842 mfu:23.922361044945383 +Step 6057 | loss:1.3202100992202759 lr:9.3431844271689e-05 tokens_per_second_per_gpu:2441.234787261703 mfu:11.515721009118092 +Step 6058 | loss:1.3539237976074219 lr:9.342969573054405e-05 tokens_per_second_per_gpu:5329.126360659597 mfu:25.138398285946586 +Step 6059 | loss:1.3207488059997559 lr:9.342754686275881e-05 tokens_per_second_per_gpu:3139.8346611494903 mfu:14.81113580767611 +Step 6060 | loss:1.2727285623550415 lr:9.342539766834946e-05 tokens_per_second_per_gpu:5375.033405826425 mfu:25.354949650548086 +Step 6061 | loss:1.381258487701416 lr:9.342324814733214e-05 tokens_per_second_per_gpu:3934.0216607384227 mfu:18.557451387011785 +Step 6062 | loss:1.151709794998169 lr:9.342109829972303e-05 tokens_per_second_per_gpu:5358.4113532531555 mfu:25.276540592545363 +Step 6063 | loss:1.2128173112869263 lr:9.34189481255383e-05 tokens_per_second_per_gpu:5331.544247523404 mfu:25.149803870817202 +Step 6064 | loss:1.4207509756088257 lr:9.341679762479414e-05 tokens_per_second_per_gpu:3581.7173445159897 mfu:16.895571818075947 +Step 6065 | loss:1.348691463470459 lr:9.341464679750669e-05 tokens_per_second_per_gpu:4715.451966168797 mfu:22.243591463485085 +Step 6066 | loss:1.2952266931533813 lr:9.341249564369214e-05 tokens_per_second_per_gpu:4084.512443299875 mfu:19.267342084729325 +Step 6067 | loss:1.2764838933944702 lr:9.341034416336668e-05 tokens_per_second_per_gpu:5344.169236070507 mfu:25.209358095839054 +Step 6068 | loss:1.1693578958511353 lr:9.340819235654647e-05 tokens_per_second_per_gpu:3880.736531875656 mfu:18.306096342785658 +Step 6069 | loss:1.301455020904541 lr:9.340604022324773e-05 tokens_per_second_per_gpu:4950.548100138192 mfu:23.35258003895535 +Step 6070 | loss:1.3668172359466553 lr:9.340388776348658e-05 tokens_per_second_per_gpu:5358.710234085821 mfu:25.277950464427377 +Step 6071 | loss:1.2971971035003662 lr:9.340173497727928e-05 tokens_per_second_per_gpu:5308.460313243203 mfu:25.04091301429218 +Step 6072 | loss:1.2737071514129639 lr:9.3399581864642e-05 tokens_per_second_per_gpu:4634.277906412598 mfu:21.86067957389236 +Step 6073 | loss:1.3771361112594604 lr:9.339742842559092e-05 tokens_per_second_per_gpu:2917.3597886582616 mfu:13.761683876007618 +Step 6074 | loss:1.2130582332611084 lr:9.339527466014224e-05 tokens_per_second_per_gpu:4770.396588336867 mfu:22.50277462077108 +Step 6075 | loss:1.3253545761108398 lr:9.339312056831216e-05 tokens_per_second_per_gpu:4016.1642744743053 mfu:18.944932110979025 +Step 6076 | loss:1.0683820247650146 lr:9.339096615011689e-05 tokens_per_second_per_gpu:3848.8638069660788 mfu:18.15574726133956 +Step 6077 | loss:1.2422375679016113 lr:9.338881140557264e-05 tokens_per_second_per_gpu:4821.453151856479 mfu:22.743617142041092 +Step 6078 | loss:1.3848296403884888 lr:9.338665633469557e-05 tokens_per_second_per_gpu:3647.7806139548597 mfu:17.207203531575146 +Step 6079 | loss:1.220335841178894 lr:9.338450093750193e-05 tokens_per_second_per_gpu:4571.0304685265355 mfu:21.562330618258155 +Step 6080 | loss:1.1488219499588013 lr:9.338234521400794e-05 tokens_per_second_per_gpu:5114.89551952713 mfu:24.127834856774165 +Step 6081 | loss:1.3575679063796997 lr:9.338018916422978e-05 tokens_per_second_per_gpu:5123.4409189884245 mfu:24.16814496402858 +Step 6082 | loss:1.319593071937561 lr:9.33780327881837e-05 tokens_per_second_per_gpu:4354.353465867748 mfu:20.540228227807887 +Step 6083 | loss:1.1109353303909302 lr:9.337587608588588e-05 tokens_per_second_per_gpu:5339.768135646589 mfu:25.188597354234265 +Step 6084 | loss:1.1656899452209473 lr:9.337371905735258e-05 tokens_per_second_per_gpu:4222.185458105465 mfu:19.916769185005773 +Step 6085 | loss:1.3366001844406128 lr:9.337156170259998e-05 tokens_per_second_per_gpu:4666.747923612534 mfu:22.013846184980654 +Step 6086 | loss:1.4684656858444214 lr:9.336940402164435e-05 tokens_per_second_per_gpu:4523.017629954765 mfu:21.33584586688014 +Step 6087 | loss:1.3626078367233276 lr:9.336724601450189e-05 tokens_per_second_per_gpu:4047.1804558948215 mfu:19.091240730645836 +Step 6088 | loss:1.4543241262435913 lr:9.336508768118885e-05 tokens_per_second_per_gpu:5320.708546272224 mfu:25.098690019254114 +Step 6089 | loss:1.2399096488952637 lr:9.336292902172144e-05 tokens_per_second_per_gpu:4799.12283229301 mfu:22.638281214714823 +Step 6090 | loss:1.3162968158721924 lr:9.336077003611591e-05 tokens_per_second_per_gpu:4518.857597384328 mfu:21.31622228347064 +Step 6091 | loss:1.282788872718811 lr:9.33586107243885e-05 tokens_per_second_per_gpu:5297.439477702226 mfu:24.98892585269695 +Step 6092 | loss:1.3786072731018066 lr:9.335645108655544e-05 tokens_per_second_per_gpu:5302.763445331855 mfu:25.014039916368457 +Step 6093 | loss:1.2463715076446533 lr:9.335429112263299e-05 tokens_per_second_per_gpu:3171.4314929102893 mfu:14.960183454067197 +Step 6094 | loss:1.371659755706787 lr:9.335213083263736e-05 tokens_per_second_per_gpu:5368.967796501416 mfu:25.326337136462353 +Step 6095 | loss:1.125028133392334 lr:9.334997021658485e-05 tokens_per_second_per_gpu:4314.96994630445 mfu:20.35444944650648 +Step 6096 | loss:1.3727302551269531 lr:9.334780927449167e-05 tokens_per_second_per_gpu:5384.654602628248 mfu:25.40033447740773 +Step 6097 | loss:1.2740414142608643 lr:9.334564800637409e-05 tokens_per_second_per_gpu:3736.002998551068 mfu:17.623363572006415 +Step 6098 | loss:1.2847528457641602 lr:9.334348641224835e-05 tokens_per_second_per_gpu:5314.556815708614 mfu:25.0696712565921 +Step 6099 | loss:1.2054047584533691 lr:9.334132449213073e-05 tokens_per_second_per_gpu:4268.180301647487 mfu:20.133734709523008 +Step 6100 | loss:1.3451735973358154 lr:9.333916224603747e-05 tokens_per_second_per_gpu:4873.062941447779 mfu:22.987069325081357 +Step 6101 | loss:1.194890022277832 lr:9.333699967398484e-05 tokens_per_second_per_gpu:4458.171203629981 mfu:21.029954209964927 +Step 6102 | loss:1.2381539344787598 lr:9.333483677598911e-05 tokens_per_second_per_gpu:5331.179605334505 mfu:25.14808379139779 +Step 6103 | loss:1.2415286302566528 lr:9.333267355206655e-05 tokens_per_second_per_gpu:5366.619867436709 mfu:25.31526155446612 +Step 6104 | loss:1.1718780994415283 lr:9.333051000223341e-05 tokens_per_second_per_gpu:4869.809112081395 mfu:22.971720456799588 +Step 6105 | loss:1.4833191633224487 lr:9.332834612650598e-05 tokens_per_second_per_gpu:3267.256786395064 mfu:15.412207712915853 +Step 6106 | loss:1.3482927083969116 lr:9.332618192490054e-05 tokens_per_second_per_gpu:5371.367069721551 mfu:25.337654917599952 +Step 6107 | loss:1.2924954891204834 lr:9.332401739743335e-05 tokens_per_second_per_gpu:5299.305728320185 mfu:24.997729275277052 +Step 6108 | loss:1.3554680347442627 lr:9.332185254412069e-05 tokens_per_second_per_gpu:4814.225530170101 mfu:22.70952321738703 +Step 6109 | loss:1.3347142934799194 lr:9.331968736497885e-05 tokens_per_second_per_gpu:5324.7573237597 mfu:25.117788793455208 +Step 6110 | loss:1.2407031059265137 lr:9.331752186002412e-05 tokens_per_second_per_gpu:3994.1978619563865 mfu:18.84131279527556 +Step 6111 | loss:1.3750064373016357 lr:9.331535602927278e-05 tokens_per_second_per_gpu:3920.9129127247757 mfu:18.495615186048372 +Step 6112 | loss:1.309953212738037 lr:9.331318987274111e-05 tokens_per_second_per_gpu:4984.7881270464295 mfu:23.5140960878325 +Step 6113 | loss:1.3655611276626587 lr:9.331102339044543e-05 tokens_per_second_per_gpu:3586.013722574702 mfu:16.915838566416575 +Step 6114 | loss:1.2919212579727173 lr:9.330885658240201e-05 tokens_per_second_per_gpu:4942.956704606682 mfu:23.316770131007576 +Step 6115 | loss:1.4098271131515503 lr:9.330668944862715e-05 tokens_per_second_per_gpu:4266.212617830977 mfu:20.124452809239056 +Step 6116 | loss:1.2481498718261719 lr:9.330452198913715e-05 tokens_per_second_per_gpu:5303.690108560508 mfu:25.01841114492306 +Step 6117 | loss:1.354213833808899 lr:9.330235420394833e-05 tokens_per_second_per_gpu:5206.251177262525 mfu:24.5587750811945 +Step 6118 | loss:1.4291824102401733 lr:9.330018609307696e-05 tokens_per_second_per_gpu:4785.346525779693 mfu:22.573296026411242 +Step 6119 | loss:1.3038766384124756 lr:9.329801765653937e-05 tokens_per_second_per_gpu:3729.786649666893 mfu:17.594039993701276 +Step 6120 | loss:1.1719586849212646 lr:9.329584889435187e-05 tokens_per_second_per_gpu:5358.431088930193 mfu:25.276633689101207 +Step 6121 | loss:1.2414329051971436 lr:9.329367980653076e-05 tokens_per_second_per_gpu:5351.062466281735 mfu:25.241874638851765 +Step 6122 | loss:1.2482677698135376 lr:9.329151039309236e-05 tokens_per_second_per_gpu:4225.439842656088 mfu:19.932120672186322 +Step 6123 | loss:1.2667567729949951 lr:9.328934065405299e-05 tokens_per_second_per_gpu:4489.141686629463 mfu:21.17604726238276 +Step 6124 | loss:1.1997836828231812 lr:9.328717058942897e-05 tokens_per_second_per_gpu:5374.566723611633 mfu:25.352748230916717 +Step 6125 | loss:1.212009072303772 lr:9.328500019923661e-05 tokens_per_second_per_gpu:4870.797607273085 mfu:22.976383357273463 +Step 6126 | loss:1.3041850328445435 lr:9.328282948349223e-05 tokens_per_second_per_gpu:5301.089747047521 mfu:25.00614479599888 +Step 6127 | loss:1.37030029296875 lr:9.328065844221219e-05 tokens_per_second_per_gpu:4955.528430905935 mfu:23.376073108918842 +Step 6128 | loss:1.430503010749817 lr:9.327848707541278e-05 tokens_per_second_per_gpu:3439.3720952512663 mfu:16.224104990690364 +Step 6129 | loss:1.1721529960632324 lr:9.327631538311034e-05 tokens_per_second_per_gpu:5343.6431301690245 mfu:25.206876364538836 +Step 6130 | loss:1.3102202415466309 lr:9.327414336532122e-05 tokens_per_second_per_gpu:4175.981001875362 mfu:19.698814881675247 +Step 6131 | loss:1.2757827043533325 lr:9.327197102206174e-05 tokens_per_second_per_gpu:5069.215076247791 mfu:23.91235241193822 +Step 6132 | loss:1.2740315198898315 lr:9.326979835334826e-05 tokens_per_second_per_gpu:5324.781855311235 mfu:25.117904512969314 +Step 6133 | loss:1.2409141063690186 lr:9.326762535919708e-05 tokens_per_second_per_gpu:5357.841464239827 mfu:25.2738523288364 +Step 6134 | loss:1.3769984245300293 lr:9.326545203962459e-05 tokens_per_second_per_gpu:5311.628651296585 mfu:25.055858605464195 +Step 6135 | loss:1.1459356546401978 lr:9.32632783946471e-05 tokens_per_second_per_gpu:4121.94990647045 mfu:19.4439409859948 +Step 6136 | loss:1.3042933940887451 lr:9.326110442428099e-05 tokens_per_second_per_gpu:5353.849653087678 mfu:25.255022274558502 +Step 6137 | loss:1.1826274394989014 lr:9.32589301285426e-05 tokens_per_second_per_gpu:3971.03685273869 mfu:18.73205836312003 +Step 6138 | loss:1.2723572254180908 lr:9.325675550744823e-05 tokens_per_second_per_gpu:4535.609979769652 mfu:21.39524612943338 +Step 6139 | loss:1.168246865272522 lr:9.325458056101433e-05 tokens_per_second_per_gpu:5333.580257342247 mfu:25.15940807651178 +Step 6140 | loss:1.218328595161438 lr:9.325240528925719e-05 tokens_per_second_per_gpu:5262.809319069408 mfu:24.82556948591162 +Step 6141 | loss:1.162437081336975 lr:9.325022969219319e-05 tokens_per_second_per_gpu:5302.2742914178125 mfu:25.011732493897654 +Step 6142 | loss:1.4109551906585693 lr:9.324805376983869e-05 tokens_per_second_per_gpu:3374.382210363167 mfu:15.917536615255438 +Step 6143 | loss:1.311252474784851 lr:9.324587752221008e-05 tokens_per_second_per_gpu:3982.762294250168 mfu:18.78736927129628 +Step 6144 | loss:1.3790855407714844 lr:9.324370094932369e-05 tokens_per_second_per_gpu:4369.969760020525 mfu:20.613892951741068 +Step 6145 | loss:1.4170819520950317 lr:9.324152405119591e-05 tokens_per_second_per_gpu:3813.5344131644206 mfu:17.989092482960984 +Step 6146 | loss:1.3215157985687256 lr:9.32393468278431e-05 tokens_per_second_per_gpu:5348.280570109237 mfu:25.22875195996566 +Step 6147 | loss:1.2507286071777344 lr:9.323716927928166e-05 tokens_per_second_per_gpu:5120.712855821416 mfu:24.155276224614234 +Step 6148 | loss:1.2019412517547607 lr:9.323499140552795e-05 tokens_per_second_per_gpu:4743.0038560605 mfu:22.373558428941397 +Step 6149 | loss:1.3082927465438843 lr:9.323281320659835e-05 tokens_per_second_per_gpu:4402.435013206518 mfu:20.767037090162766 +Step 6150 | loss:1.2714643478393555 lr:9.323063468250925e-05 tokens_per_second_per_gpu:3420.0207418535997 mfu:16.132821355032224 +Step 6151 | loss:1.153998613357544 lr:9.322845583327702e-05 tokens_per_second_per_gpu:5341.793271635377 mfu:25.19815026621721 +Step 6152 | loss:1.2467690706253052 lr:9.322627665891808e-05 tokens_per_second_per_gpu:5336.303953374597 mfu:25.172256215406836 +Step 6153 | loss:1.1563929319381714 lr:9.322409715944878e-05 tokens_per_second_per_gpu:5307.915511943172 mfu:25.038343093607242 +Step 6154 | loss:1.471877932548523 lr:9.322191733488552e-05 tokens_per_second_per_gpu:5135.364477067559 mfu:24.224390421856647 +Step 6155 | loss:1.4102739095687866 lr:9.321973718524472e-05 tokens_per_second_per_gpu:4849.354283591026 mfu:22.875231540856902 +Step 6156 | loss:1.069934368133545 lr:9.321755671054275e-05 tokens_per_second_per_gpu:5323.211439484128 mfu:25.110496593572282 +Step 6157 | loss:1.1567089557647705 lr:9.321537591079603e-05 tokens_per_second_per_gpu:5139.064125945114 mfu:24.241842296837262 +Step 6158 | loss:1.214603066444397 lr:9.321319478602095e-05 tokens_per_second_per_gpu:4402.719482980957 mfu:20.768378982624345 +Step 6159 | loss:1.1492514610290527 lr:9.321101333623393e-05 tokens_per_second_per_gpu:4791.825685505419 mfu:22.603859328296128 +Step 6160 | loss:1.2272816896438599 lr:9.320883156145135e-05 tokens_per_second_per_gpu:5345.730935721181 mfu:25.216724899544715 +Step 6161 | loss:1.134235143661499 lr:9.320664946168964e-05 tokens_per_second_per_gpu:4289.141205299563 mfu:20.232610868348793 +Step 6162 | loss:1.390600562095642 lr:9.320446703696521e-05 tokens_per_second_per_gpu:5296.762055716064 mfu:24.985730337608118 +Step 6163 | loss:1.2052699327468872 lr:9.320228428729448e-05 tokens_per_second_per_gpu:5341.470387283171 mfu:25.1966271656381 +Step 6164 | loss:1.3336526155471802 lr:9.320010121269384e-05 tokens_per_second_per_gpu:4667.21395687566 mfu:22.016044543396294 +Step 6165 | loss:1.2169249057769775 lr:9.319791781317973e-05 tokens_per_second_per_gpu:3718.409693543906 mfu:17.540372950560272 +Step 6166 | loss:1.3876464366912842 lr:9.319573408876856e-05 tokens_per_second_per_gpu:4367.630692228036 mfu:20.602859169877778 +Step 6167 | loss:1.4079645872116089 lr:9.319355003947678e-05 tokens_per_second_per_gpu:4344.229464747283 mfu:20.492471587189947 +Step 6168 | loss:1.2280961275100708 lr:9.319136566532079e-05 tokens_per_second_per_gpu:5319.2977463652205 mfu:25.092035035385493 +Step 6169 | loss:1.125193476676941 lr:9.318918096631701e-05 tokens_per_second_per_gpu:5033.59793054627 mfu:23.74434025875246 +Step 6170 | loss:1.1967872381210327 lr:9.318699594248191e-05 tokens_per_second_per_gpu:4528.928300065983 mfu:21.363727506259067 +Step 6171 | loss:1.3076882362365723 lr:9.31848105938319e-05 tokens_per_second_per_gpu:4769.262374307435 mfu:22.49742433967738 +Step 6172 | loss:1.2949331998825073 lr:9.318262492038342e-05 tokens_per_second_per_gpu:4072.8718833755493 mfu:19.212431577481958 +Step 6173 | loss:1.4040988683700562 lr:9.318043892215291e-05 tokens_per_second_per_gpu:3198.4039515276277 mfu:15.087417143341163 +Step 6174 | loss:1.1950111389160156 lr:9.31782525991568e-05 tokens_per_second_per_gpu:3991.7094953520023 mfu:18.82957474544356 +Step 6175 | loss:1.1453604698181152 lr:9.317606595141154e-05 tokens_per_second_per_gpu:4456.417431220238 mfu:21.021681366283737 +Step 6176 | loss:1.181519627571106 lr:9.31738789789336e-05 tokens_per_second_per_gpu:4429.507678282601 mfu:20.894743470399614 +Step 6177 | loss:1.1786984205245972 lr:9.317169168173938e-05 tokens_per_second_per_gpu:4997.112689910831 mfu:23.572233153651133 +Step 6178 | loss:1.2424203157424927 lr:9.316950405984538e-05 tokens_per_second_per_gpu:4596.924949316918 mfu:21.684479302200387 +Step 6179 | loss:1.2940312623977661 lr:9.316731611326802e-05 tokens_per_second_per_gpu:5312.537582337935 mfu:25.060146188265872 +Step 6180 | loss:1.292370319366455 lr:9.316512784202378e-05 tokens_per_second_per_gpu:4267.916209831763 mfu:20.13248894336953 +Step 6181 | loss:1.3879659175872803 lr:9.316293924612909e-05 tokens_per_second_per_gpu:5109.977186943237 mfu:24.104634242821874 +Step 6182 | loss:1.422288179397583 lr:9.316075032560043e-05 tokens_per_second_per_gpu:4486.030333328621 mfu:21.161370478011072 +Step 6183 | loss:1.2105958461761475 lr:9.315856108045427e-05 tokens_per_second_per_gpu:5328.228897716056 mfu:25.134164800118935 +Step 6184 | loss:1.2717596292495728 lr:9.315637151070705e-05 tokens_per_second_per_gpu:4388.461843602394 mfu:20.70112326507154 +Step 6185 | loss:1.135650873184204 lr:9.315418161637526e-05 tokens_per_second_per_gpu:4928.935542846664 mfu:23.250629918728695 +Step 6186 | loss:1.2139924764633179 lr:9.315199139747536e-05 tokens_per_second_per_gpu:4637.925948960565 mfu:21.87788801301136 +Step 6187 | loss:1.1954283714294434 lr:9.314980085402383e-05 tokens_per_second_per_gpu:5314.022608000371 mfu:25.067151307687087 +Step 6188 | loss:1.2857056856155396 lr:9.314760998603713e-05 tokens_per_second_per_gpu:2952.0130079699393 mfu:13.925149023949688 +Step 6189 | loss:1.2659004926681519 lr:9.314541879353179e-05 tokens_per_second_per_gpu:4045.1748609974356 mfu:19.081779997325093 +Step 6190 | loss:1.2091407775878906 lr:9.31432272765242e-05 tokens_per_second_per_gpu:3712.3750823952346 mfu:17.511906660161156 +Step 6191 | loss:1.2096166610717773 lr:9.314103543503092e-05 tokens_per_second_per_gpu:4922.138128080587 mfu:23.21856535351856 +Step 6192 | loss:1.232871413230896 lr:9.31388432690684e-05 tokens_per_second_per_gpu:4481.642867377981 mfu:21.140674052543027 +Step 6193 | loss:1.3470503091812134 lr:9.313665077865313e-05 tokens_per_second_per_gpu:3898.4887961399063 mfu:18.389836802168738 +Step 6194 | loss:1.2084400653839111 lr:9.313445796380161e-05 tokens_per_second_per_gpu:5374.761436315733 mfu:25.353666723963364 +Step 6195 | loss:1.2238510847091675 lr:9.313226482453034e-05 tokens_per_second_per_gpu:4723.45683600388 mfu:22.2813518002688 +Step 6196 | loss:1.4043279886245728 lr:9.313007136085578e-05 tokens_per_second_per_gpu:4453.24846385536 mfu:21.006732806542022 +Step 6197 | loss:1.3015624284744263 lr:9.312787757279447e-05 tokens_per_second_per_gpu:4019.5586249843836 mfu:18.960943841470982 +Step 6198 | loss:1.2468422651290894 lr:9.312568346036288e-05 tokens_per_second_per_gpu:5341.2066263231945 mfu:25.195382960187633 +Step 6199 | loss:1.2891781330108643 lr:9.312348902357753e-05 tokens_per_second_per_gpu:4001.272257850748 mfu:18.874683927725666 +Step 6200 | loss:1.2572731971740723 lr:9.31212942624549e-05 tokens_per_second_per_gpu:5181.721452709372 mfu:24.443064185274707 +Step 6201 | loss:1.2954633235931396 lr:9.311909917701152e-05 tokens_per_second_per_gpu:4352.7818296389305 mfu:20.532814551567764 +Step 6202 | loss:1.2906947135925293 lr:9.31169037672639e-05 tokens_per_second_per_gpu:5044.9418331895495 mfu:23.797851382990174 +Step 6203 | loss:1.2235803604125977 lr:9.311470803322855e-05 tokens_per_second_per_gpu:4073.7702000942413 mfu:19.216669090712667 +Step 6204 | loss:1.3016966581344604 lr:9.311251197492197e-05 tokens_per_second_per_gpu:4774.675292624976 mfu:22.52295799892032 +Step 6205 | loss:1.4391628503799438 lr:9.311031559236067e-05 tokens_per_second_per_gpu:5311.450703992511 mfu:25.05501919766994 +Step 6206 | loss:1.3946433067321777 lr:9.310811888556122e-05 tokens_per_second_per_gpu:4679.117706066814 mfu:22.07219655932001 +Step 6207 | loss:1.536920428276062 lr:9.310592185454007e-05 tokens_per_second_per_gpu:5349.566541469959 mfu:25.234818106283523 +Step 6208 | loss:1.450600028038025 lr:9.310372449931382e-05 tokens_per_second_per_gpu:4437.712821399168 mfu:20.933448530422236 +Step 6209 | loss:1.2612947225570679 lr:9.310152681989892e-05 tokens_per_second_per_gpu:4170.265850500388 mfu:19.67185553753412 +Step 6210 | loss:1.174144983291626 lr:9.309932881631196e-05 tokens_per_second_per_gpu:2971.6731950419476 mfu:14.01788948074216 +Step 6211 | loss:1.3732948303222656 lr:9.309713048856945e-05 tokens_per_second_per_gpu:4230.450793026562 mfu:19.955758180040142 +Step 6212 | loss:1.2356427907943726 lr:9.309493183668792e-05 tokens_per_second_per_gpu:5354.535999175941 mfu:25.258259886159554 +Step 6213 | loss:1.4602082967758179 lr:9.309273286068389e-05 tokens_per_second_per_gpu:3331.0852530929506 mfu:15.71329748058917 +Step 6214 | loss:1.32646906375885 lr:9.309053356057394e-05 tokens_per_second_per_gpu:5375.738633227211 mfu:25.358276328521796 +Step 6215 | loss:1.2635748386383057 lr:9.308833393637458e-05 tokens_per_second_per_gpu:5319.81795713055 mfu:25.09448895832266 +Step 6216 | loss:1.281707525253296 lr:9.308613398810237e-05 tokens_per_second_per_gpu:4537.970380624172 mfu:21.406380542990068 +Step 6217 | loss:1.3874237537384033 lr:9.308393371577384e-05 tokens_per_second_per_gpu:4902.210156529458 mfu:23.12456171164933 +Step 6218 | loss:1.1982214450836182 lr:9.308173311940555e-05 tokens_per_second_per_gpu:5180.755016158686 mfu:24.438505339174043 +Step 6219 | loss:1.2168819904327393 lr:9.307953219901405e-05 tokens_per_second_per_gpu:3719.6980214956593 mfu:17.546450213320107 +Step 6220 | loss:1.4154855012893677 lr:9.307733095461589e-05 tokens_per_second_per_gpu:4928.324487553491 mfu:23.247747466654015 +Step 6221 | loss:1.232370376586914 lr:9.307512938622762e-05 tokens_per_second_per_gpu:5343.367383221249 mfu:25.205575619138894 +Step 6222 | loss:1.2383564710617065 lr:9.307292749386581e-05 tokens_per_second_per_gpu:5335.4645338006785 mfu:25.168296530056914 +Step 6223 | loss:1.3050023317337036 lr:9.307072527754702e-05 tokens_per_second_per_gpu:4528.8168961774 mfu:21.363201995109154 +Step 6224 | loss:1.480495572090149 lr:9.30685227372878e-05 tokens_per_second_per_gpu:5032.446320710736 mfu:23.73890791867334 +Step 6225 | loss:1.29445219039917 lr:9.306631987310475e-05 tokens_per_second_per_gpu:3705.2515069147826 mfu:17.478303539212558 +Step 6226 | loss:1.1401927471160889 lr:9.306411668501438e-05 tokens_per_second_per_gpu:4423.950189345541 mfu:20.868527846878116 +Step 6227 | loss:1.2557889223098755 lr:9.30619131730333e-05 tokens_per_second_per_gpu:3411.173556578672 mfu:16.091087672604928 +Step 6228 | loss:1.457529902458191 lr:9.30597093371781e-05 tokens_per_second_per_gpu:4381.577886378448 mfu:20.668650464321974 +Step 6229 | loss:1.3904545307159424 lr:9.30575051774653e-05 tokens_per_second_per_gpu:4370.657849846879 mfu:20.617138788852206 +Step 6230 | loss:1.1973859071731567 lr:9.305530069391153e-05 tokens_per_second_per_gpu:3572.144673667408 mfu:16.850415896417122 +Step 6231 | loss:1.185925841331482 lr:9.305309588653335e-05 tokens_per_second_per_gpu:3772.768315668428 mfu:17.796791845658937 +Step 6232 | loss:1.2417709827423096 lr:9.305089075534734e-05 tokens_per_second_per_gpu:4638.188354648198 mfu:21.879125825410952 +Step 6233 | loss:1.2704004049301147 lr:9.304868530037007e-05 tokens_per_second_per_gpu:5346.146676094242 mfu:25.218686017817067 +Step 6234 | loss:1.1934638023376465 lr:9.304647952161818e-05 tokens_per_second_per_gpu:3912.793566189 mfu:18.45731484313575 +Step 6235 | loss:1.0123180150985718 lr:9.30442734191082e-05 tokens_per_second_per_gpu:5307.030400663068 mfu:25.034167872683433 +Step 6236 | loss:1.5054097175598145 lr:9.304206699285675e-05 tokens_per_second_per_gpu:5318.336213228191 mfu:25.087499319523655 +Step 6237 | loss:1.3279125690460205 lr:9.303986024288043e-05 tokens_per_second_per_gpu:5076.311142691875 mfu:23.945825768068744 +Step 6238 | loss:1.23964262008667 lr:9.303765316919581e-05 tokens_per_second_per_gpu:4682.874369022842 mfu:22.08991737943639 +Step 6239 | loss:1.3451077938079834 lr:9.303544577181953e-05 tokens_per_second_per_gpu:4764.9040926681455 mfu:22.47686558158533 +Step 6240 | loss:1.252069115638733 lr:9.303323805076816e-05 tokens_per_second_per_gpu:5345.1152709547305 mfu:25.213820703798127 +Step 6241 | loss:1.412802815437317 lr:9.30310300060583e-05 tokens_per_second_per_gpu:4911.149765829159 mfu:23.16673137397065 +Step 6242 | loss:1.2705483436584473 lr:9.30288216377066e-05 tokens_per_second_per_gpu:4941.5811917434 mfu:23.31028160214478 +Step 6243 | loss:1.147121787071228 lr:9.302661294572963e-05 tokens_per_second_per_gpu:4025.9295499816517 mfu:18.990996581674985 +Step 6244 | loss:1.2555651664733887 lr:9.302440393014402e-05 tokens_per_second_per_gpu:4322.860913692366 mfu:20.391672486013146 +Step 6245 | loss:1.2143877744674683 lr:9.302219459096636e-05 tokens_per_second_per_gpu:3846.3936746882596 mfu:18.14409522593706 +Step 6246 | loss:1.339715838432312 lr:9.301998492821331e-05 tokens_per_second_per_gpu:4376.12361684904 mfu:20.642921744357416 +Step 6247 | loss:1.4515818357467651 lr:9.301777494190144e-05 tokens_per_second_per_gpu:5349.588835995196 mfu:25.234923273362902 +Step 6248 | loss:1.2109675407409668 lr:9.301556463204742e-05 tokens_per_second_per_gpu:5341.549429324741 mfu:25.197000020434153 +Step 6249 | loss:1.2454527616500854 lr:9.301335399866783e-05 tokens_per_second_per_gpu:5294.614749492389 mfu:24.97560112778282 +Step 6250 | loss:1.2088545560836792 lr:9.301114304177933e-05 tokens_per_second_per_gpu:4064.4611556203427 mfu:19.17275673964252 +Step 6251 | loss:1.2535842657089233 lr:9.300893176139853e-05 tokens_per_second_per_gpu:3530.1980105049884 mfu:16.65254632944126 +Step 6252 | loss:1.1899722814559937 lr:9.300672015754208e-05 tokens_per_second_per_gpu:4685.138220294491 mfu:22.10059635213761 +Step 6253 | loss:1.1909998655319214 lr:9.300450823022657e-05 tokens_per_second_per_gpu:4822.122485611321 mfu:22.746774503562797 +Step 6254 | loss:1.1817671060562134 lr:9.300229597946869e-05 tokens_per_second_per_gpu:5369.335250230451 mfu:25.328070478396278 +Step 6255 | loss:1.400282859802246 lr:9.300008340528505e-05 tokens_per_second_per_gpu:3182.735750034195 mfu:15.013507563625806 +Step 6256 | loss:1.1757386922836304 lr:9.299787050769229e-05 tokens_per_second_per_gpu:3203.9041863786647 mfu:15.113362689570337 +Step 6257 | loss:1.1499781608581543 lr:9.299565728670705e-05 tokens_per_second_per_gpu:3742.1540807845854 mfu:17.652379276384767 +Step 6258 | loss:1.3182317018508911 lr:9.2993443742346e-05 tokens_per_second_per_gpu:3218.1425924555547 mfu:15.18052768035761 +Step 6259 | loss:1.3155120611190796 lr:9.299122987462577e-05 tokens_per_second_per_gpu:5353.985007666295 mfu:25.255660765199714 +Step 6260 | loss:1.313596487045288 lr:9.298901568356302e-05 tokens_per_second_per_gpu:4461.868225567753 mfu:21.047393693222382 +Step 6261 | loss:1.2952238321304321 lr:9.29868011691744e-05 tokens_per_second_per_gpu:5335.868840421805 mfu:25.170203713370594 +Step 6262 | loss:1.2187163829803467 lr:9.298458633147655e-05 tokens_per_second_per_gpu:4865.555386564591 mfu:22.951654907776412 +Step 6263 | loss:1.1981573104858398 lr:9.298237117048614e-05 tokens_per_second_per_gpu:4386.649961480149 mfu:20.692576307961943 +Step 6264 | loss:1.278093695640564 lr:9.298015568621985e-05 tokens_per_second_per_gpu:3993.5451079950276 mfu:18.83823364346852 +Step 6265 | loss:1.3720401525497437 lr:9.29779398786943e-05 tokens_per_second_per_gpu:3660.902941185018 mfu:17.269103788019695 +Step 6266 | loss:1.1569323539733887 lr:9.29757237479262e-05 tokens_per_second_per_gpu:5123.67835289508 mfu:24.16926498027668 +Step 6267 | loss:1.1735053062438965 lr:9.29735072939322e-05 tokens_per_second_per_gpu:4634.6263425813595 mfu:21.862323206749746 +Step 6268 | loss:1.288081169128418 lr:9.297129051672895e-05 tokens_per_second_per_gpu:4758.699456011675 mfu:22.447597251856582 +Step 6269 | loss:1.2130184173583984 lr:9.296907341633315e-05 tokens_per_second_per_gpu:5323.240626399507 mfu:25.110634273231927 +Step 6270 | loss:1.1206973791122437 lr:9.296685599276147e-05 tokens_per_second_per_gpu:4778.7497039440605 mfu:22.5421776922787 +Step 6271 | loss:1.1070504188537598 lr:9.296463824603058e-05 tokens_per_second_per_gpu:4833.0507619434975 mfu:22.798325047577883 +Step 6272 | loss:1.3534107208251953 lr:9.296242017615715e-05 tokens_per_second_per_gpu:4349.851342567537 mfu:20.51899094405935 +Step 6273 | loss:1.4311710596084595 lr:9.29602017831579e-05 tokens_per_second_per_gpu:4023.130902867562 mfu:18.97779488573939 +Step 6274 | loss:1.3015551567077637 lr:9.295798306704947e-05 tokens_per_second_per_gpu:5327.8534913835465 mfu:25.13239394439756 +Step 6275 | loss:1.4135648012161255 lr:9.295576402784858e-05 tokens_per_second_per_gpu:5351.305065049761 mfu:25.243019018632157 +Step 6276 | loss:1.3169504404067993 lr:9.295354466557191e-05 tokens_per_second_per_gpu:4143.823197050965 mfu:19.547121029631292 +Step 6277 | loss:1.2327708005905151 lr:9.295132498023616e-05 tokens_per_second_per_gpu:4382.755447056189 mfu:20.674205218951446 +Step 6278 | loss:1.3584201335906982 lr:9.294910497185798e-05 tokens_per_second_per_gpu:3909.963938573666 mfu:18.443967006889967 +Step 6279 | loss:1.3770549297332764 lr:9.294688464045414e-05 tokens_per_second_per_gpu:4261.999958357957 mfu:20.104580975751002 +Step 6280 | loss:1.4973479509353638 lr:9.294466398604128e-05 tokens_per_second_per_gpu:3150.2062520168283 mfu:14.860060371373217 +Step 6281 | loss:1.2708293199539185 lr:9.29424430086361e-05 tokens_per_second_per_gpu:5202.546819172418 mfu:24.541300992054804 +Step 6282 | loss:1.2020899057388306 lr:9.294022170825535e-05 tokens_per_second_per_gpu:3762.710775213706 mfu:17.749348711340662 +Step 6283 | loss:1.1767754554748535 lr:9.293800008491573e-05 tokens_per_second_per_gpu:4247.484591844648 mfu:20.036109515330814 +Step 6284 | loss:1.336080551147461 lr:9.293577813863391e-05 tokens_per_second_per_gpu:4499.805036059604 mfu:21.226348100999754 +Step 6285 | loss:1.3625627756118774 lr:9.293355586942662e-05 tokens_per_second_per_gpu:5363.924603363176 mfu:25.302547533971797 +Step 6286 | loss:1.4000906944274902 lr:9.29313332773106e-05 tokens_per_second_per_gpu:5357.204536337328 mfu:25.27084783124921 +Step 6287 | loss:1.2630798816680908 lr:9.292911036230253e-05 tokens_per_second_per_gpu:5339.772704639005 mfu:25.188618906951042 +Step 6288 | loss:1.2480188608169556 lr:9.292688712441914e-05 tokens_per_second_per_gpu:4837.5883228994035 mfu:22.819729496794622 +Step 6289 | loss:1.1995899677276611 lr:9.292466356367717e-05 tokens_per_second_per_gpu:5378.719338871835 mfu:25.37233682560831 +Step 6290 | loss:1.2553164958953857 lr:9.292243968009331e-05 tokens_per_second_per_gpu:5349.179568702031 mfu:25.232992689712844 +Step 6291 | loss:1.3016020059585571 lr:9.292021547368432e-05 tokens_per_second_per_gpu:3882.57885343001 mfu:18.314786887890016 +Step 6292 | loss:1.1958045959472656 lr:9.29179909444669e-05 tokens_per_second_per_gpu:4809.8427585457775 mfu:22.68884893585715 +Step 6293 | loss:1.1405627727508545 lr:9.29157660924578e-05 tokens_per_second_per_gpu:4930.063252715833 mfu:23.255949518588707 +Step 6294 | loss:1.2841463088989258 lr:9.291354091767375e-05 tokens_per_second_per_gpu:5019.641111657635 mfu:23.678503562775912 +Step 6295 | loss:1.3230668306350708 lr:9.291131542013148e-05 tokens_per_second_per_gpu:5352.299332532829 mfu:25.24770914798932 +Step 6296 | loss:1.4192537069320679 lr:9.290908959984772e-05 tokens_per_second_per_gpu:3791.027275708747 mfu:17.88292247546894 +Step 6297 | loss:1.1077028512954712 lr:9.290686345683923e-05 tokens_per_second_per_gpu:5338.693176226076 mfu:25.183526587240717 +Step 6298 | loss:1.1335771083831787 lr:9.290463699112275e-05 tokens_per_second_per_gpu:4692.935999790628 mfu:22.13737980845926 +Step 6299 | loss:1.1354366540908813 lr:9.290241020271502e-05 tokens_per_second_per_gpu:3167.5916953114674 mfu:14.942070473656528 +Step 6300 | loss:1.245556354522705 lr:9.290018309163278e-05 tokens_per_second_per_gpu:5367.677494910169 mfu:25.32025056370829 +Step 6301 | loss:1.145395040512085 lr:9.289795565789279e-05 tokens_per_second_per_gpu:5358.152354734742 mfu:25.275318852344306 +Step 6302 | loss:1.2094289064407349 lr:9.289572790151181e-05 tokens_per_second_per_gpu:4114.076986459317 mfu:19.40680307904362 +Step 6303 | loss:1.3728001117706299 lr:9.289349982250658e-05 tokens_per_second_per_gpu:4784.14657911304 mfu:22.567635673252628 +Step 6304 | loss:1.2212542295455933 lr:9.289127142089386e-05 tokens_per_second_per_gpu:4772.168501737736 mfu:22.51113303860258 +Step 6305 | loss:1.2052289247512817 lr:9.288904269669043e-05 tokens_per_second_per_gpu:5366.94627763939 mfu:25.316801287083493 +Step 6306 | loss:1.2430542707443237 lr:9.288681364991303e-05 tokens_per_second_per_gpu:4615.749547494511 mfu:21.773278143611734 +Step 6307 | loss:1.3236894607543945 lr:9.288458428057843e-05 tokens_per_second_per_gpu:3822.1562778046587 mfu:18.029763289510786 +Step 6308 | loss:1.2659791707992554 lr:9.288235458870341e-05 tokens_per_second_per_gpu:5173.04261666546 mfu:24.40212463489345 +Step 6309 | loss:1.2254844903945923 lr:9.288012457430472e-05 tokens_per_second_per_gpu:5346.233599955245 mfu:25.21909605250016 +Step 6310 | loss:1.264251470565796 lr:9.287789423739915e-05 tokens_per_second_per_gpu:4804.715310004905 mfu:22.664661886256923 +Step 6311 | loss:1.217689037322998 lr:9.287566357800345e-05 tokens_per_second_per_gpu:4408.013580368 mfu:20.793352143265178 +Step 6312 | loss:1.328580617904663 lr:9.287343259613441e-05 tokens_per_second_per_gpu:3633.5038240256386 mfu:17.13985747760747 +Step 6313 | loss:1.1937006711959839 lr:9.287120129180882e-05 tokens_per_second_per_gpu:3491.413501213468 mfu:16.46959318179352 +Step 6314 | loss:1.1876246929168701 lr:9.286896966504347e-05 tokens_per_second_per_gpu:5043.047978364875 mfu:23.788917746657233 +Step 6315 | loss:1.0690475702285767 lr:9.286673771585512e-05 tokens_per_second_per_gpu:3911.7676298317815 mfu:18.45247532629596 +Step 6316 | loss:1.3588231801986694 lr:9.286450544426057e-05 tokens_per_second_per_gpu:3890.7217861322197 mfu:18.35319849077392 +Step 6317 | loss:1.1841378211975098 lr:9.28622728502766e-05 tokens_per_second_per_gpu:4160.300425220416 mfu:19.62484692621161 +Step 6318 | loss:1.1852082014083862 lr:9.286003993392e-05 tokens_per_second_per_gpu:5329.133958122629 mfu:25.13843412447597 +Step 6319 | loss:1.3905373811721802 lr:9.285780669520758e-05 tokens_per_second_per_gpu:3835.4191340180414 mfu:18.092326445145922 +Step 6320 | loss:1.3466023206710815 lr:9.285557313415612e-05 tokens_per_second_per_gpu:4156.890985295117 mfu:19.60876401637354 +Step 6321 | loss:1.1379493474960327 lr:9.285333925078243e-05 tokens_per_second_per_gpu:3564.0045063647394 mfu:16.81201733839468 +Step 6322 | loss:1.298216700553894 lr:9.285110504510331e-05 tokens_per_second_per_gpu:5359.947376641886 mfu:25.28378627694262 +Step 6323 | loss:1.4089607000350952 lr:9.284887051713555e-05 tokens_per_second_per_gpu:3449.490628776845 mfu:16.27183583973036 +Step 6324 | loss:1.2712955474853516 lr:9.284663566689598e-05 tokens_per_second_per_gpu:4486.256215969176 mfu:21.16243600496613 +Step 6325 | loss:1.107686996459961 lr:9.284440049440139e-05 tokens_per_second_per_gpu:5322.188938140959 mfu:25.105673280279834 +Step 6326 | loss:1.3928357362747192 lr:9.284216499966861e-05 tokens_per_second_per_gpu:3793.3438357914097 mfu:17.89385008462477 +Step 6327 | loss:1.2656042575836182 lr:9.283992918271441e-05 tokens_per_second_per_gpu:5072.05352563508 mfu:23.925741861987465 +Step 6328 | loss:1.1043851375579834 lr:9.283769304355566e-05 tokens_per_second_per_gpu:4798.658911612243 mfu:22.636092821710616 +Step 6329 | loss:1.1462818384170532 lr:9.283545658220915e-05 tokens_per_second_per_gpu:5126.729130368349 mfu:24.1836560181344 +Step 6330 | loss:1.2819757461547852 lr:9.28332197986917e-05 tokens_per_second_per_gpu:4866.610198005347 mfu:22.95663063331185 +Step 6331 | loss:1.270341396331787 lr:9.283098269302014e-05 tokens_per_second_per_gpu:4524.2261416403935 mfu:21.341546622694604 +Step 6332 | loss:1.3200435638427734 lr:9.28287452652113e-05 tokens_per_second_per_gpu:4764.871720539182 mfu:22.476712876729756 +Step 6333 | loss:1.22586989402771 lr:9.282650751528199e-05 tokens_per_second_per_gpu:5306.774057637217 mfu:25.032958658893868 +Step 6334 | loss:1.222019076347351 lr:9.282426944324906e-05 tokens_per_second_per_gpu:3811.9858286046265 mfu:17.981787545376665 +Step 6335 | loss:1.3133692741394043 lr:9.282203104912935e-05 tokens_per_second_per_gpu:3513.6172479065713 mfu:16.574332043294717 +Step 6336 | loss:1.2150304317474365 lr:9.281979233293965e-05 tokens_per_second_per_gpu:5309.4692755687365 mfu:25.045672461728444 +Step 6337 | loss:1.202580213546753 lr:9.281755329469686e-05 tokens_per_second_per_gpu:4810.032135890289 mfu:22.689742261102317 +Step 6338 | loss:1.3899166584014893 lr:9.281531393441775e-05 tokens_per_second_per_gpu:4698.621777742188 mfu:22.164200593150518 +Step 6339 | loss:1.2719472646713257 lr:9.281307425211923e-05 tokens_per_second_per_gpu:5335.870813273087 mfu:25.170213019646695 +Step 6340 | loss:1.3664213418960571 lr:9.281083424781811e-05 tokens_per_second_per_gpu:5357.219054077863 mfu:25.27091631390829 +Step 6341 | loss:1.26027512550354 lr:9.280859392153126e-05 tokens_per_second_per_gpu:4037.1916549462057 mfu:19.044121852306034 +Step 6342 | loss:1.2867043018341064 lr:9.280635327327548e-05 tokens_per_second_per_gpu:5347.493840642115 mfu:25.225040822839357 +Step 6343 | loss:1.445313811302185 lr:9.280411230306767e-05 tokens_per_second_per_gpu:4250.957720836843 mfu:20.052492857363912 +Step 6344 | loss:1.3295758962631226 lr:9.280187101092466e-05 tokens_per_second_per_gpu:4809.450988793744 mfu:22.687000891093994 +Step 6345 | loss:1.1558517217636108 lr:9.279962939686333e-05 tokens_per_second_per_gpu:3935.009622875695 mfu:18.562111772977023 +Step 6346 | loss:1.3554245233535767 lr:9.279738746090052e-05 tokens_per_second_per_gpu:4942.071537618074 mfu:23.31259464729725 +Step 6347 | loss:1.2973688840866089 lr:9.279514520305311e-05 tokens_per_second_per_gpu:5314.85996651831 mfu:25.0711012706843 +Step 6348 | loss:1.2688041925430298 lr:9.279290262333793e-05 tokens_per_second_per_gpu:4694.039231109651 mfu:22.14258393882177 +Step 6349 | loss:1.2408244609832764 lr:9.279065972177188e-05 tokens_per_second_per_gpu:5355.476588525355 mfu:25.26269680660183 +Step 6350 | loss:1.3125996589660645 lr:9.278841649837182e-05 tokens_per_second_per_gpu:4831.071082830563 mfu:22.788986563434083 +Step 6351 | loss:1.2206281423568726 lr:9.278617295315462e-05 tokens_per_second_per_gpu:4452.6899341308645 mfu:21.00409812653658 +Step 6352 | loss:1.472980260848999 lr:9.278392908613715e-05 tokens_per_second_per_gpu:3045.11344801792 mfu:14.364319684229052 +Step 6353 | loss:1.2325074672698975 lr:9.278168489733629e-05 tokens_per_second_per_gpu:5275.512291966801 mfu:24.885491576419287 +Step 6354 | loss:1.194760799407959 lr:9.277944038676891e-05 tokens_per_second_per_gpu:5320.475027738845 mfu:25.097588472489797 +Step 6355 | loss:1.1330698728561401 lr:9.277719555445193e-05 tokens_per_second_per_gpu:5337.04701205035 mfu:25.175761349959245 +Step 6356 | loss:1.2231669425964355 lr:9.277495040040218e-05 tokens_per_second_per_gpu:5352.551520000149 mfu:25.248898759300282 +Step 6357 | loss:1.254112720489502 lr:9.277270492463657e-05 tokens_per_second_per_gpu:5345.718515547554 mfu:25.216666311466618 +Step 6358 | loss:1.2526696920394897 lr:9.2770459127172e-05 tokens_per_second_per_gpu:4521.676657465615 mfu:21.329520270855316 +Step 6359 | loss:1.2327913045883179 lr:9.276821300802534e-05 tokens_per_second_per_gpu:4999.355607455237 mfu:23.58281337838929 +Step 6360 | loss:1.2690010070800781 lr:9.276596656721348e-05 tokens_per_second_per_gpu:5338.670894794514 mfu:25.183421481926434 +Step 6361 | loss:1.1696134805679321 lr:9.276371980475335e-05 tokens_per_second_per_gpu:5322.296202289689 mfu:25.106179263570542 +Step 6362 | loss:1.1806330680847168 lr:9.276147272066182e-05 tokens_per_second_per_gpu:4154.507917761805 mfu:19.597522680225662 +Step 6363 | loss:1.4917101860046387 lr:9.27592253149558e-05 tokens_per_second_per_gpu:5343.153895030281 mfu:25.204568558917263 +Step 6364 | loss:1.2901654243469238 lr:9.275697758765218e-05 tokens_per_second_per_gpu:4990.111447553947 mfu:23.53920709891859 +Step 6365 | loss:1.1501699686050415 lr:9.27547295387679e-05 tokens_per_second_per_gpu:5357.41242861931 mfu:25.2718284946134 +Step 6366 | loss:1.2068332433700562 lr:9.275248116831982e-05 tokens_per_second_per_gpu:3347.6082681847397 mfu:15.791239361894212 +Step 6367 | loss:1.238534927368164 lr:9.275023247632488e-05 tokens_per_second_per_gpu:5381.585856675545 mfu:25.385858679166784 +Step 6368 | loss:1.1233525276184082 lr:9.274798346279998e-05 tokens_per_second_per_gpu:5328.3607537486405 mfu:25.134786787521595 +Step 6369 | loss:1.4095934629440308 lr:9.274573412776206e-05 tokens_per_second_per_gpu:4798.281504110971 mfu:22.634312526134753 +Step 6370 | loss:1.3033674955368042 lr:9.2743484471228e-05 tokens_per_second_per_gpu:4110.433432048838 mfu:19.389615811234904 +Step 6371 | loss:1.397584080696106 lr:9.274123449321475e-05 tokens_per_second_per_gpu:4286.736527925578 mfu:20.221267594895437 +Step 6372 | loss:1.18070650100708 lr:9.273898419373922e-05 tokens_per_second_per_gpu:5335.861165898534 mfu:25.170167511334913 +Step 6373 | loss:1.1311603784561157 lr:9.273673357281834e-05 tokens_per_second_per_gpu:4185.371924031056 mfu:19.74311346374019 +Step 6374 | loss:1.118088722229004 lr:9.273448263046902e-05 tokens_per_second_per_gpu:4388.065070486648 mfu:20.699251618588093 +Step 6375 | loss:1.1727243661880493 lr:9.273223136670823e-05 tokens_per_second_per_gpu:5256.4366685284085 mfu:24.795508606023425 +Step 6376 | loss:1.2833213806152344 lr:9.272997978155285e-05 tokens_per_second_per_gpu:4945.4782846579055 mfu:23.32866485433512 +Step 6377 | loss:1.4154342412948608 lr:9.272772787501984e-05 tokens_per_second_per_gpu:4740.768700910245 mfu:22.36301482074085 +Step 6378 | loss:1.0842134952545166 lr:9.272547564712614e-05 tokens_per_second_per_gpu:5324.461497912981 mfu:25.116393332463144 +Step 6379 | loss:1.1827874183654785 lr:9.27232230978887e-05 tokens_per_second_per_gpu:3917.557631526744 mfu:18.47978775216725 +Step 6380 | loss:1.3073103427886963 lr:9.272097022732443e-05 tokens_per_second_per_gpu:4930.122174183375 mfu:23.25622746120414 +Step 6381 | loss:1.4351493120193481 lr:9.271871703545029e-05 tokens_per_second_per_gpu:5309.786307371137 mfu:25.04716795483157 +Step 6382 | loss:1.276058316230774 lr:9.271646352228324e-05 tokens_per_second_per_gpu:3388.181549152615 mfu:15.98263045073524 +Step 6383 | loss:1.185865879058838 lr:9.271420968784022e-05 tokens_per_second_per_gpu:4526.001027018608 mfu:21.349919059850308 +Step 6384 | loss:1.4825886487960815 lr:9.271195553213818e-05 tokens_per_second_per_gpu:5348.853844144787 mfu:25.231456191401985 +Step 6385 | loss:1.2760446071624756 lr:9.270970105519406e-05 tokens_per_second_per_gpu:3690.93863534565 mfu:17.410787281993255 +Step 6386 | loss:1.188319206237793 lr:9.270744625702483e-05 tokens_per_second_per_gpu:3600.5837027861785 mfu:16.984567648969072 +Step 6387 | loss:1.2765934467315674 lr:9.270519113764746e-05 tokens_per_second_per_gpu:4660.974109073989 mfu:21.986610116687995 +Step 6388 | loss:1.373928427696228 lr:9.270293569707888e-05 tokens_per_second_per_gpu:4473.778594830268 mfu:21.103576937152205 +Step 6389 | loss:1.1870970726013184 lr:9.270067993533609e-05 tokens_per_second_per_gpu:5311.33641993953 mfu:25.054480100293155 +Step 6390 | loss:1.2209887504577637 lr:9.269842385243602e-05 tokens_per_second_per_gpu:4430.884737703582 mfu:20.901239294638945 +Step 6391 | loss:1.3460460901260376 lr:9.269616744839567e-05 tokens_per_second_per_gpu:4913.652682281127 mfu:23.178538057916168 +Step 6392 | loss:1.1310434341430664 lr:9.269391072323199e-05 tokens_per_second_per_gpu:4636.560277611959 mfu:21.871445908252426 +Step 6393 | loss:1.2104867696762085 lr:9.269165367696195e-05 tokens_per_second_per_gpu:5097.806776427467 mfu:24.047224339932832 +Step 6394 | loss:1.2549139261245728 lr:9.268939630960256e-05 tokens_per_second_per_gpu:5344.889273846477 mfu:25.212754636879417 +Step 6395 | loss:1.080922245979309 lr:9.268713862117075e-05 tokens_per_second_per_gpu:4257.3927404765545 mfu:20.082847942931995 +Step 6396 | loss:1.384130597114563 lr:9.268488061168353e-05 tokens_per_second_per_gpu:4537.345324945442 mfu:21.403432048708005 +Step 6397 | loss:1.332559585571289 lr:9.268262228115788e-05 tokens_per_second_per_gpu:5334.2303030590365 mfu:25.16247445306711 +Step 6398 | loss:1.3058631420135498 lr:9.268036362961078e-05 tokens_per_second_per_gpu:5325.629763118514 mfu:25.12190423876424 +Step 6399 | loss:1.2423944473266602 lr:9.267810465705922e-05 tokens_per_second_per_gpu:4745.759685811134 mfu:22.38655814806818 +Step 6400 | loss:1.1505895853042603 lr:9.267584536352018e-05 tokens_per_second_per_gpu:5063.354763135222 mfu:23.88470831511371 +Step 6401 | loss:1.3010780811309814 lr:9.267358574901068e-05 tokens_per_second_per_gpu:5393.806064399658 mfu:25.443503483983854 +Step 6402 | loss:1.1739439964294434 lr:9.26713258135477e-05 tokens_per_second_per_gpu:3570.8035766049647 mfu:16.844089712758823 +Step 6403 | loss:1.130963683128357 lr:9.26690655571482e-05 tokens_per_second_per_gpu:4286.408592960111 mfu:20.219720669711855 +Step 6404 | loss:1.3241188526153564 lr:9.266680497982925e-05 tokens_per_second_per_gpu:4182.045645439945 mfu:19.72742284010427 +Step 6405 | loss:1.192813754081726 lr:9.266454408160779e-05 tokens_per_second_per_gpu:5301.095050382049 mfu:25.006169812732686 +Step 6406 | loss:1.1979060173034668 lr:9.266228286250085e-05 tokens_per_second_per_gpu:5329.756734090202 mfu:25.14137186497155 +Step 6407 | loss:1.423798680305481 lr:9.266002132252545e-05 tokens_per_second_per_gpu:4126.731540681594 mfu:19.46649677039969 +Step 6408 | loss:1.110102891921997 lr:9.265775946169858e-05 tokens_per_second_per_gpu:3913.2649375112 mfu:18.459538382086812 +Step 6409 | loss:1.410583734512329 lr:9.265549728003726e-05 tokens_per_second_per_gpu:4178.528442867302 mfu:19.710831595472477 +Step 6410 | loss:1.3175249099731445 lr:9.26532347775585e-05 tokens_per_second_per_gpu:5334.307236911446 mfu:25.162837363175196 +Step 6411 | loss:1.2572827339172363 lr:9.265097195427932e-05 tokens_per_second_per_gpu:3754.713235714044 mfu:17.711622952999814 +Step 6412 | loss:1.3746001720428467 lr:9.264870881021675e-05 tokens_per_second_per_gpu:5377.463414802491 mfu:25.366412417490473 +Step 6413 | loss:1.3237576484680176 lr:9.264644534538777e-05 tokens_per_second_per_gpu:5111.8167504904395 mfu:24.11331177793593 +Step 6414 | loss:1.4286895990371704 lr:9.264418155980947e-05 tokens_per_second_per_gpu:4025.754168732308 mfu:18.990169278398703 +Step 6415 | loss:1.271845817565918 lr:9.26419174534988e-05 tokens_per_second_per_gpu:5030.641955843451 mfu:23.730396421737968 +Step 6416 | loss:1.3018240928649902 lr:9.263965302647287e-05 tokens_per_second_per_gpu:4855.721728862691 mfu:22.905267866601537 +Step 6417 | loss:1.1879757642745972 lr:9.263738827874863e-05 tokens_per_second_per_gpu:5334.937544259325 mfu:25.165810630476663 +Step 6418 | loss:1.1758942604064941 lr:9.263512321034317e-05 tokens_per_second_per_gpu:5306.366090658575 mfu:25.031034209049267 +Step 6419 | loss:1.332468867301941 lr:9.263285782127352e-05 tokens_per_second_per_gpu:3500.3038387666797 mfu:16.511530420880018 +Step 6420 | loss:1.5042402744293213 lr:9.263059211155667e-05 tokens_per_second_per_gpu:5291.158035051075 mfu:24.95929521598684 +Step 6421 | loss:1.3288315534591675 lr:9.262832608120973e-05 tokens_per_second_per_gpu:5025.257097811982 mfu:23.704995127652012 +Step 6422 | loss:1.13582444190979 lr:9.26260597302497e-05 tokens_per_second_per_gpu:5342.940951278721 mfu:25.2035640669092 +Step 6423 | loss:1.1931006908416748 lr:9.262379305869362e-05 tokens_per_second_per_gpu:5366.7026515799635 mfu:25.315652061394285 +Step 6424 | loss:1.2010546922683716 lr:9.262152606655857e-05 tokens_per_second_per_gpu:5359.456421167739 mfu:25.28147035620543 +Step 6425 | loss:1.3005828857421875 lr:9.261925875386159e-05 tokens_per_second_per_gpu:3803.515433660025 mfu:17.941831247224577 +Step 6426 | loss:1.3452600240707397 lr:9.261699112061971e-05 tokens_per_second_per_gpu:4353.029997644737 mfu:20.53398520239298 +Step 6427 | loss:1.216253638267517 lr:9.261472316685002e-05 tokens_per_second_per_gpu:5354.293848854535 mfu:25.257117621778377 +Step 6428 | loss:1.2445412874221802 lr:9.261245489256955e-05 tokens_per_second_per_gpu:5246.921156926759 mfu:24.75062231428475 +Step 6429 | loss:1.3897640705108643 lr:9.261018629779537e-05 tokens_per_second_per_gpu:4339.697874019381 mfu:20.471095300557327 +Step 6430 | loss:1.3180617094039917 lr:9.260791738254453e-05 tokens_per_second_per_gpu:4139.885230762913 mfu:19.52854497076422 +Step 6431 | loss:1.4113154411315918 lr:9.260564814683412e-05 tokens_per_second_per_gpu:5348.737079713652 mfu:25.230905393658258 +Step 6432 | loss:1.013993740081787 lr:9.26033785906812e-05 tokens_per_second_per_gpu:5156.81283341464 mfu:24.325565978212776 +Step 6433 | loss:1.1365916728973389 lr:9.260110871410281e-05 tokens_per_second_per_gpu:5319.60283503723 mfu:25.093474190704857 +Step 6434 | loss:1.1341552734375 lr:9.259883851711608e-05 tokens_per_second_per_gpu:5106.088365800821 mfu:24.08629000999179 +Step 6435 | loss:1.1775181293487549 lr:9.259656799973802e-05 tokens_per_second_per_gpu:5343.11884030528 mfu:25.20440319979848 +Step 6436 | loss:1.137076497077942 lr:9.259429716198575e-05 tokens_per_second_per_gpu:3606.560151184186 mfu:17.012759575749715 +Step 6437 | loss:1.2601121664047241 lr:9.259202600387634e-05 tokens_per_second_per_gpu:5323.169879286447 mfu:25.11030054702892 +Step 6438 | loss:1.4771102666854858 lr:9.258975452542686e-05 tokens_per_second_per_gpu:5338.202443174867 mfu:25.18121171571047 +Step 6439 | loss:1.227283239364624 lr:9.258748272665438e-05 tokens_per_second_per_gpu:3701.4044375647613 mfu:17.460156256710793 +Step 6440 | loss:1.3774683475494385 lr:9.258521060757604e-05 tokens_per_second_per_gpu:5376.672471070208 mfu:25.36268140095638 +Step 6441 | loss:1.3302873373031616 lr:9.258293816820889e-05 tokens_per_second_per_gpu:3255.924827014574 mfu:15.35875292708644 +Step 6442 | loss:1.3332550525665283 lr:9.258066540857003e-05 tokens_per_second_per_gpu:5321.566782831351 mfu:25.102738467533772 +Step 6443 | loss:1.1806206703186035 lr:9.257839232867654e-05 tokens_per_second_per_gpu:4492.96818015307 mfu:21.19409748520073 +Step 6444 | loss:1.248720645904541 lr:9.257611892854554e-05 tokens_per_second_per_gpu:3560.049481831448 mfu:16.793360812874372 +Step 6445 | loss:1.1330006122589111 lr:9.257384520819412e-05 tokens_per_second_per_gpu:4827.7579503903435 mfu:22.773357952434885 +Step 6446 | loss:1.3299442529678345 lr:9.257157116763937e-05 tokens_per_second_per_gpu:3409.88894320067 mfu:16.08502793212325 +Step 6447 | loss:1.2194560766220093 lr:9.25692968068984e-05 tokens_per_second_per_gpu:4175.699479323348 mfu:19.697486891764697 +Step 6448 | loss:1.2421525716781616 lr:9.256702212598831e-05 tokens_per_second_per_gpu:4401.3658221234655 mfu:20.761993533401196 +Step 6449 | loss:1.2874120473861694 lr:9.256474712492622e-05 tokens_per_second_per_gpu:4826.548270787298 mfu:22.767651687354697 +Step 6450 | loss:1.2170404195785522 lr:9.256247180372925e-05 tokens_per_second_per_gpu:5344.78748525169 mfu:25.212274482710875 +Step 6451 | loss:1.2758592367172241 lr:9.256019616241449e-05 tokens_per_second_per_gpu:4391.0287165049 mfu:20.71323164250634 +Step 6452 | loss:1.2933871746063232 lr:9.255792020099906e-05 tokens_per_second_per_gpu:5347.29013064554 mfu:25.224079888029003 +Step 6453 | loss:1.3423879146575928 lr:9.255564391950007e-05 tokens_per_second_per_gpu:4117.233363763566 mfu:19.42169224932126 +Step 6454 | loss:1.4227851629257202 lr:9.255336731793467e-05 tokens_per_second_per_gpu:4618.411707405567 mfu:21.785836006123503 +Step 6455 | loss:1.1606762409210205 lr:9.255109039631998e-05 tokens_per_second_per_gpu:5304.099791093401 mfu:25.020343687329767 +Step 6456 | loss:1.3908579349517822 lr:9.254881315467308e-05 tokens_per_second_per_gpu:3007.9061270576353 mfu:14.188806403035825 +Step 6457 | loss:1.2907005548477173 lr:9.254653559301114e-05 tokens_per_second_per_gpu:4107.024058952652 mfu:19.373533216641164 +Step 6458 | loss:1.126332402229309 lr:9.254425771135127e-05 tokens_per_second_per_gpu:4662.247160807296 mfu:21.992615318919025 +Step 6459 | loss:1.2718907594680786 lr:9.254197950971061e-05 tokens_per_second_per_gpu:5360.610367833581 mfu:25.28691371951181 +Step 6460 | loss:1.2954182624816895 lr:9.25397009881063e-05 tokens_per_second_per_gpu:3991.08639472898 mfu:18.826635473492832 +Step 6461 | loss:1.3833017349243164 lr:9.253742214655546e-05 tokens_per_second_per_gpu:3843.1487848830143 mfu:18.12878852709043 +Step 6462 | loss:1.3057506084442139 lr:9.253514298507524e-05 tokens_per_second_per_gpu:3418.7150397628184 mfu:16.126662135494257 +Step 6463 | loss:1.2103981971740723 lr:9.25328635036828e-05 tokens_per_second_per_gpu:5314.96782234047 mfu:25.071610044999527 +Step 6464 | loss:1.2057178020477295 lr:9.253058370239525e-05 tokens_per_second_per_gpu:5135.008665263063 mfu:24.222711996867087 +Step 6465 | loss:1.1953274011611938 lr:9.252830358122977e-05 tokens_per_second_per_gpu:5364.662776278069 mfu:25.306029621556636 +Step 6466 | loss:1.2162784337997437 lr:9.252602314020346e-05 tokens_per_second_per_gpu:4871.007552769297 mfu:22.977373705999486 +Step 6467 | loss:1.423214077949524 lr:9.252374237933354e-05 tokens_per_second_per_gpu:5349.748662517984 mfu:25.235677202340113 +Step 6468 | loss:1.1241741180419922 lr:9.25214612986371e-05 tokens_per_second_per_gpu:5017.375020176768 mfu:23.66781402262559 +Step 6469 | loss:1.1936765909194946 lr:9.251917989813134e-05 tokens_per_second_per_gpu:4535.465572453951 mfu:21.39456493548679 +Step 6470 | loss:1.243556261062622 lr:9.25168981778334e-05 tokens_per_second_per_gpu:4592.315961837694 mfu:21.662737921886535 +Step 6471 | loss:1.3819243907928467 lr:9.251461613776044e-05 tokens_per_second_per_gpu:5052.789230431375 mfu:23.8348688946839 +Step 6472 | loss:1.2221571207046509 lr:9.251233377792962e-05 tokens_per_second_per_gpu:5338.92501406921 mfu:25.184620205902394 +Step 6473 | loss:1.186924934387207 lr:9.251005109835814e-05 tokens_per_second_per_gpu:4955.77560551546 mfu:23.377239073724304 +Step 6474 | loss:1.2979247570037842 lr:9.250776809906311e-05 tokens_per_second_per_gpu:2759.348967959506 mfu:13.01631987534561 +Step 6475 | loss:1.2703403234481812 lr:9.250548478006175e-05 tokens_per_second_per_gpu:5285.6708532503135 mfu:24.933411243222896 +Step 6476 | loss:1.1640342473983765 lr:9.250320114137121e-05 tokens_per_second_per_gpu:4567.0843995535615 mfu:21.543716337643847 +Step 6477 | loss:1.2427403926849365 lr:9.250091718300869e-05 tokens_per_second_per_gpu:5247.593086725686 mfu:24.753791921789336 +Step 6478 | loss:1.2634069919586182 lr:9.249863290499132e-05 tokens_per_second_per_gpu:5334.246757157785 mfu:25.16255206985827 +Step 6479 | loss:1.412485957145691 lr:9.24963483073363e-05 tokens_per_second_per_gpu:5333.594684515711 mfu:25.15947613195054 +Step 6480 | loss:1.1516475677490234 lr:9.249406339006086e-05 tokens_per_second_per_gpu:5342.889649791437 mfu:25.203322069040745 +Step 6481 | loss:1.1412230730056763 lr:9.249177815318213e-05 tokens_per_second_per_gpu:5046.58986851223 mfu:23.805625446791083 +Step 6482 | loss:1.320207118988037 lr:9.248949259671731e-05 tokens_per_second_per_gpu:5000.995636797079 mfu:23.590549676612618 +Step 6483 | loss:1.2001724243164062 lr:9.24872067206836e-05 tokens_per_second_per_gpu:4102.071834729752 mfu:19.35017273004516 +Step 6484 | loss:1.2199900150299072 lr:9.248492052509818e-05 tokens_per_second_per_gpu:4251.292427001734 mfu:20.05407172345046 +Step 6485 | loss:1.2812575101852417 lr:9.248263400997825e-05 tokens_per_second_per_gpu:4980.370054526382 mfu:23.49325528595453 +Step 6486 | loss:1.313138723373413 lr:9.248034717534102e-05 tokens_per_second_per_gpu:4942.999241951473 mfu:23.316970786920574 +Step 6487 | loss:1.4586914777755737 lr:9.247806002120367e-05 tokens_per_second_per_gpu:4777.428153648562 mfu:22.5359437140551 +Step 6488 | loss:1.273442029953003 lr:9.247577254758341e-05 tokens_per_second_per_gpu:4686.831364183884 mfu:22.10858320074401 +Step 6489 | loss:1.3295916318893433 lr:9.247348475449746e-05 tokens_per_second_per_gpu:4849.6548416871665 mfu:22.876649324676897 +Step 6490 | loss:1.319275140762329 lr:9.2471196641963e-05 tokens_per_second_per_gpu:4703.458661013034 mfu:22.187016996796487 +Step 6491 | loss:1.3155323266983032 lr:9.246890820999724e-05 tokens_per_second_per_gpu:4879.816167947453 mfu:23.018925446699576 +Step 6492 | loss:1.266255497932434 lr:9.246661945861741e-05 tokens_per_second_per_gpu:4645.2294143879 mfu:21.912339705531792 +Step 6493 | loss:1.0911993980407715 lr:9.246433038784073e-05 tokens_per_second_per_gpu:5293.47407093064 mfu:24.970220352387216 +Step 6494 | loss:1.1654980182647705 lr:9.246204099768438e-05 tokens_per_second_per_gpu:5339.5373995419095 mfu:25.187508932660855 +Step 6495 | loss:1.179142951965332 lr:9.245975128816562e-05 tokens_per_second_per_gpu:5140.212322046115 mfu:24.24725853374814 +Step 6496 | loss:1.3066658973693848 lr:9.245746125930164e-05 tokens_per_second_per_gpu:3926.958674116105 mfu:18.524134074044532 +Step 6497 | loss:1.2565193176269531 lr:9.24551709111097e-05 tokens_per_second_per_gpu:4603.594173203077 mfu:21.71593916915833 +Step 6498 | loss:1.3924986124038696 lr:9.245288024360697e-05 tokens_per_second_per_gpu:3613.9131639472052 mfu:17.04744499150622 +Step 6499 | loss:1.248461365699768 lr:9.245058925681072e-05 tokens_per_second_per_gpu:5331.749984928994 mfu:25.150774369261793 +Step 6500 | loss:1.3029110431671143 lr:9.244829795073817e-05 tokens_per_second_per_gpu:5073.430952735426 mfu:23.932239420633096 +Step 6501 | loss:1.3392655849456787 lr:9.244600632540656e-05 tokens_per_second_per_gpu:3252.2120763463968 mfu:15.341239248724065 +Step 6502 | loss:1.2122857570648193 lr:9.244371438083313e-05 tokens_per_second_per_gpu:5303.020044332491 mfu:25.015250337634757 +Step 6503 | loss:1.1333776712417603 lr:9.24414221170351e-05 tokens_per_second_per_gpu:4699.519652590648 mfu:22.168436022003952 +Step 6504 | loss:1.2844173908233643 lr:9.243912953402972e-05 tokens_per_second_per_gpu:3023.7979686640633 mfu:14.263770931321119 +Step 6505 | loss:1.1883481740951538 lr:9.243683663183423e-05 tokens_per_second_per_gpu:5037.351361944649 mfu:23.762045835060043 +Step 6506 | loss:1.3386703729629517 lr:9.243454341046587e-05 tokens_per_second_per_gpu:3983.129507361385 mfu:18.78910147819491 +Step 6507 | loss:1.2029011249542236 lr:9.24322498699419e-05 tokens_per_second_per_gpu:4316.286168999655 mfu:20.360658293530513 +Step 6508 | loss:1.1290245056152344 lr:9.242995601027957e-05 tokens_per_second_per_gpu:3828.905435506977 mfu:18.061600217917665 +Step 6509 | loss:1.28839910030365 lr:9.242766183149612e-05 tokens_per_second_per_gpu:4133.217517695234 mfu:19.49709223059482 +Step 6510 | loss:1.3521100282669067 lr:9.242536733360881e-05 tokens_per_second_per_gpu:4031.8856736859043 mfu:19.01909263340744 +Step 6511 | loss:1.1121288537979126 lr:9.242307251663491e-05 tokens_per_second_per_gpu:5314.774381496364 mfu:25.070697551533467 +Step 6512 | loss:1.1717535257339478 lr:9.242077738059165e-05 tokens_per_second_per_gpu:4345.0393269887845 mfu:20.496291845559856 +Step 6513 | loss:1.1728671789169312 lr:9.241848192549633e-05 tokens_per_second_per_gpu:4390.650803757655 mfu:20.71144896359902 +Step 6514 | loss:1.0945606231689453 lr:9.241618615136618e-05 tokens_per_second_per_gpu:5271.176881437871 mfu:24.865040705257016 +Step 6515 | loss:1.2428616285324097 lr:9.241389005821848e-05 tokens_per_second_per_gpu:4441.336371666081 mfu:20.950541435272687 +Step 6516 | loss:1.130885124206543 lr:9.241159364607051e-05 tokens_per_second_per_gpu:3305.527865148417 mfu:15.592738921114776 +Step 6517 | loss:1.2905733585357666 lr:9.240929691493954e-05 tokens_per_second_per_gpu:5274.51081240975 mfu:24.880767426478645 +Step 6518 | loss:1.3924058675765991 lr:9.240699986484283e-05 tokens_per_second_per_gpu:5370.296568062217 mfu:25.33260517862646 +Step 6519 | loss:1.3131892681121826 lr:9.240470249579765e-05 tokens_per_second_per_gpu:3551.971527730372 mfu:16.75525572513851 +Step 6520 | loss:1.2311981916427612 lr:9.24024048078213e-05 tokens_per_second_per_gpu:3517.495078371526 mfu:16.592624431224007 +Step 6521 | loss:1.2184053659439087 lr:9.240010680093105e-05 tokens_per_second_per_gpu:3934.0433440130128 mfu:18.557553670717574 +Step 6522 | loss:1.322561502456665 lr:9.23978084751442e-05 tokens_per_second_per_gpu:4767.000925909941 mfu:22.48675670174352 +Step 6523 | loss:1.2139698266983032 lr:9.2395509830478e-05 tokens_per_second_per_gpu:4573.619882295152 mfu:21.574545324804262 +Step 6524 | loss:1.1737275123596191 lr:9.239321086694979e-05 tokens_per_second_per_gpu:5345.998420423077 mfu:25.217986670521157 +Step 6525 | loss:1.2525626420974731 lr:9.239091158457681e-05 tokens_per_second_per_gpu:4076.9845218262844 mfu:19.23183159474255 +Step 6526 | loss:1.3630344867706299 lr:9.238861198337638e-05 tokens_per_second_per_gpu:3266.749084925695 mfu:15.409812798462037 +Step 6527 | loss:1.214999794960022 lr:9.238631206336578e-05 tokens_per_second_per_gpu:4540.875470424165 mfu:21.42008434723606 +Step 6528 | loss:1.2354954481124878 lr:9.238401182456234e-05 tokens_per_second_per_gpu:5306.273007114216 mfu:25.030595118088836 +Step 6529 | loss:1.2063134908676147 lr:9.238171126698334e-05 tokens_per_second_per_gpu:4936.526602468415 mfu:23.286438242132817 +Step 6530 | loss:1.2339922189712524 lr:9.237941039064606e-05 tokens_per_second_per_gpu:4449.977970621486 mfu:20.991305331954152 +Step 6531 | loss:1.275175929069519 lr:9.237710919556783e-05 tokens_per_second_per_gpu:3733.287385693727 mfu:17.610553562827008 +Step 6532 | loss:1.3768088817596436 lr:9.237480768176596e-05 tokens_per_second_per_gpu:3781.257508082641 mfu:17.83683681468236 +Step 6533 | loss:1.2947837114334106 lr:9.237250584925776e-05 tokens_per_second_per_gpu:3648.2986346516436 mfu:17.209647123585324 +Step 6534 | loss:1.0667283535003662 lr:9.237020369806053e-05 tokens_per_second_per_gpu:5203.828027741583 mfu:24.547344671472278 +Step 6535 | loss:1.4298510551452637 lr:9.23679012281916e-05 tokens_per_second_per_gpu:5359.9658556285 mfu:25.283873445475322 +Step 6536 | loss:1.151860237121582 lr:9.236559843966828e-05 tokens_per_second_per_gpu:5371.7933142789725 mfu:25.339665585902132 +Step 6537 | loss:1.3162652254104614 lr:9.236329533250788e-05 tokens_per_second_per_gpu:3127.4367938116357 mfu:14.752652952149159 +Step 6538 | loss:1.1437716484069824 lr:9.236099190672772e-05 tokens_per_second_per_gpu:5062.8382435544945 mfu:23.882271804124095 +Step 6539 | loss:1.224179744720459 lr:9.235868816234514e-05 tokens_per_second_per_gpu:4901.907784386023 mfu:23.12313537065874 +Step 6540 | loss:1.2144941091537476 lr:9.235638409937747e-05 tokens_per_second_per_gpu:4631.925355623525 mfu:21.84958219043375 +Step 6541 | loss:1.3559571504592896 lr:9.235407971784202e-05 tokens_per_second_per_gpu:4355.748228069173 mfu:20.546807559082982 +Step 6542 | loss:1.2782697677612305 lr:9.235177501775615e-05 tokens_per_second_per_gpu:4114.984465475882 mfu:19.411083812396644 +Step 6543 | loss:1.3068244457244873 lr:9.234946999913716e-05 tokens_per_second_per_gpu:4307.398015674577 mfu:20.318731357820788 +Step 6544 | loss:1.2094212770462036 lr:9.234716466200242e-05 tokens_per_second_per_gpu:3256.6841402309537 mfu:15.3623347370803 +Step 6545 | loss:1.2578035593032837 lr:9.234485900636923e-05 tokens_per_second_per_gpu:5125.566243640374 mfu:24.178170483029607 +Step 6546 | loss:1.3682752847671509 lr:9.234255303225496e-05 tokens_per_second_per_gpu:5340.743219771254 mfu:25.193196992416816 +Step 6547 | loss:1.1065739393234253 lr:9.234024673967695e-05 tokens_per_second_per_gpu:5351.419993074146 mfu:25.243561153731928 +Step 6548 | loss:1.220942497253418 lr:9.233794012865253e-05 tokens_per_second_per_gpu:5307.237747257118 mfu:25.035145961944547 +Step 6549 | loss:1.094349980354309 lr:9.233563319919906e-05 tokens_per_second_per_gpu:5329.879945892305 mfu:25.14195307606417 +Step 6550 | loss:1.308008074760437 lr:9.23333259513339e-05 tokens_per_second_per_gpu:5384.6016069893785 mfu:25.400084487937193 +Step 6551 | loss:1.2668620347976685 lr:9.233101838507441e-05 tokens_per_second_per_gpu:4081.5029882231242 mfu:19.25314597166631 +Step 6552 | loss:1.3701355457305908 lr:9.23287105004379e-05 tokens_per_second_per_gpu:5363.839446851602 mfu:25.302145836177502 +Step 6553 | loss:1.3807377815246582 lr:9.232640229744178e-05 tokens_per_second_per_gpu:4602.716774542967 mfu:21.711800329978946 +Step 6554 | loss:1.242123007774353 lr:9.232409377610337e-05 tokens_per_second_per_gpu:5350.827006573559 mfu:25.240763935234725 +Step 6555 | loss:1.1323060989379883 lr:9.232178493644006e-05 tokens_per_second_per_gpu:3797.142126791466 mfu:17.911767271327307 +Step 6556 | loss:1.2678720951080322 lr:9.23194757784692e-05 tokens_per_second_per_gpu:5227.850249615014 mfu:24.66066159066294 +Step 6557 | loss:1.442603588104248 lr:9.231716630220817e-05 tokens_per_second_per_gpu:3481.9733523682085 mfu:16.425062389034952 +Step 6558 | loss:1.2627975940704346 lr:9.231485650767433e-05 tokens_per_second_per_gpu:4028.1354973626653 mfu:19.00140241184466 +Step 6559 | loss:1.0989333391189575 lr:9.231254639488505e-05 tokens_per_second_per_gpu:5321.451552772812 mfu:25.10219490768737 +Step 6560 | loss:1.1876813173294067 lr:9.231023596385771e-05 tokens_per_second_per_gpu:5323.808736451452 mfu:25.113314145276522 +Step 6561 | loss:1.1959776878356934 lr:9.230792521460967e-05 tokens_per_second_per_gpu:5332.5831281039045 mfu:25.154704447766726 +Step 6562 | loss:1.2232941389083862 lr:9.230561414715835e-05 tokens_per_second_per_gpu:5091.000350729603 mfu:24.015117268620035 +Step 6563 | loss:1.4071050882339478 lr:9.230330276152111e-05 tokens_per_second_per_gpu:4721.302712487637 mfu:22.271190432111307 +Step 6564 | loss:1.2135355472564697 lr:9.230099105771533e-05 tokens_per_second_per_gpu:5359.293388763126 mfu:25.28070130453667 +Step 6565 | loss:1.1027746200561523 lr:9.229867903575837e-05 tokens_per_second_per_gpu:5335.489638238251 mfu:25.168414951972206 +Step 6566 | loss:1.1835418939590454 lr:9.229636669566769e-05 tokens_per_second_per_gpu:5223.354677557798 mfu:24.639455210245366 +Step 6567 | loss:1.2780300378799438 lr:9.22940540374606e-05 tokens_per_second_per_gpu:4959.915486310792 mfu:23.39676759777239 +Step 6568 | loss:1.1365643739700317 lr:9.229174106115455e-05 tokens_per_second_per_gpu:4692.478919032587 mfu:22.135223680537564 +Step 6569 | loss:1.2125319242477417 lr:9.228942776676692e-05 tokens_per_second_per_gpu:5035.384197502583 mfu:23.752766384753958 +Step 6570 | loss:1.3842891454696655 lr:9.228711415431512e-05 tokens_per_second_per_gpu:4045.103269847367 mfu:19.081442289655506 +Step 6571 | loss:1.2475841045379639 lr:9.228480022381652e-05 tokens_per_second_per_gpu:3835.2561583406214 mfu:18.091557661069015 +Step 6572 | loss:1.360610842704773 lr:9.228248597528854e-05 tokens_per_second_per_gpu:5336.785857224376 mfu:25.174529438087518 +Step 6573 | loss:1.1851377487182617 lr:9.228017140874861e-05 tokens_per_second_per_gpu:3753.6018266696524 mfu:17.706380246911483 +Step 6574 | loss:1.324029564857483 lr:9.22778565242141e-05 tokens_per_second_per_gpu:5353.4396880024 mfu:25.253088399303987 +Step 6575 | loss:1.320926547050476 lr:9.227554132170244e-05 tokens_per_second_per_gpu:4502.372338711482 mfu:21.238458505635663 +Step 6576 | loss:1.2977781295776367 lr:9.227322580123103e-05 tokens_per_second_per_gpu:4076.4702041828386 mfu:19.22940547066678 +Step 6577 | loss:1.2882720232009888 lr:9.22709099628173e-05 tokens_per_second_per_gpu:5384.581188577984 mfu:25.399988170806907 +Step 6578 | loss:1.2725980281829834 lr:9.226859380647867e-05 tokens_per_second_per_gpu:4156.178577429664 mfu:19.605403466922883 +Step 6579 | loss:1.0911712646484375 lr:9.226627733223254e-05 tokens_per_second_per_gpu:5000.840904568844 mfu:23.589819778291996 +Step 6580 | loss:1.2621667385101318 lr:9.226396054009636e-05 tokens_per_second_per_gpu:4259.882598084143 mfu:20.094593026080503 +Step 6581 | loss:1.2240753173828125 lr:9.226164343008753e-05 tokens_per_second_per_gpu:5058.406036726549 mfu:23.86136432830421 +Step 6582 | loss:1.3768420219421387 lr:9.225932600222349e-05 tokens_per_second_per_gpu:3478.663462598327 mfu:16.409449074264995 +Step 6583 | loss:1.16383957862854 lr:9.225700825652165e-05 tokens_per_second_per_gpu:5367.852195014074 mfu:25.321074653905317 +Step 6584 | loss:1.4106959104537964 lr:9.225469019299949e-05 tokens_per_second_per_gpu:5344.862195555018 mfu:25.21262690395857 +Step 6585 | loss:1.246080994606018 lr:9.22523718116744e-05 tokens_per_second_per_gpu:4387.494759361943 mfu:20.696561363707744 +Step 6586 | loss:1.2709242105484009 lr:9.225005311256383e-05 tokens_per_second_per_gpu:4999.008631557282 mfu:23.581176633878442 +Step 6587 | loss:1.3302185535430908 lr:9.224773409568521e-05 tokens_per_second_per_gpu:4233.436189052399 mfu:19.9698408024551 +Step 6588 | loss:1.1953336000442505 lr:9.224541476105601e-05 tokens_per_second_per_gpu:4948.329947675024 mfu:23.34211663533075 +Step 6589 | loss:1.3731811046600342 lr:9.224309510869363e-05 tokens_per_second_per_gpu:4218.33511857065 mfu:19.898606476484385 +Step 6590 | loss:1.357744812965393 lr:9.224077513861555e-05 tokens_per_second_per_gpu:5361.112385465484 mfu:25.289281822334537 +Step 6591 | loss:1.132334589958191 lr:9.223845485083922e-05 tokens_per_second_per_gpu:5024.535175648722 mfu:23.70158969763529 +Step 6592 | loss:1.3891247510910034 lr:9.223613424538209e-05 tokens_per_second_per_gpu:5359.266456569903 mfu:25.280574260786135 +Step 6593 | loss:1.1922998428344727 lr:9.223381332226159e-05 tokens_per_second_per_gpu:4729.791128649212 mfu:22.311231739418595 +Step 6594 | loss:1.4079869985580444 lr:9.223149208149519e-05 tokens_per_second_per_gpu:4946.316723210725 mfu:23.332619912041135 +Step 6595 | loss:1.2692975997924805 lr:9.222917052310035e-05 tokens_per_second_per_gpu:4156.710619346886 mfu:19.607913199422132 +Step 6596 | loss:1.314997673034668 lr:9.222684864709453e-05 tokens_per_second_per_gpu:3651.824445708416 mfu:17.226278975905515 +Step 6597 | loss:1.4117722511291504 lr:9.22245264534952e-05 tokens_per_second_per_gpu:3956.6947914473753 mfu:18.664404413000913 +Step 6598 | loss:1.2848408222198486 lr:9.222220394231982e-05 tokens_per_second_per_gpu:3799.584778137973 mfu:17.92328967448807 +Step 6599 | loss:1.077891230583191 lr:9.221988111358584e-05 tokens_per_second_per_gpu:4447.031204608513 mfu:20.977404933900832 +Step 6600 | loss:1.46079421043396 lr:9.221755796731076e-05 tokens_per_second_per_gpu:4263.5704123533515 mfu:20.111989075194355 +Step 6601 | loss:1.1886870861053467 lr:9.221523450351204e-05 tokens_per_second_per_gpu:5340.982002298739 mfu:25.19432337033946 +Step 6602 | loss:1.1774048805236816 lr:9.221291072220715e-05 tokens_per_second_per_gpu:4784.109946618835 mfu:22.567462871527315 +Step 6603 | loss:1.349856972694397 lr:9.221058662341357e-05 tokens_per_second_per_gpu:4876.174761280919 mfu:23.00174831016549 +Step 6604 | loss:1.260221242904663 lr:9.220826220714877e-05 tokens_per_second_per_gpu:4125.092539332904 mfu:19.458765321395454 +Step 6605 | loss:1.197527527809143 lr:9.220593747343027e-05 tokens_per_second_per_gpu:3450.048954020827 mfu:16.27446955516633 +Step 6606 | loss:1.2630668878555298 lr:9.220361242227553e-05 tokens_per_second_per_gpu:4428.205483316834 mfu:20.88860079456872 +Step 6607 | loss:1.246908187866211 lr:9.220128705370201e-05 tokens_per_second_per_gpu:4800.838121703095 mfu:22.646372527520743 +Step 6608 | loss:1.1667054891586304 lr:9.219896136772725e-05 tokens_per_second_per_gpu:4907.451776860985 mfu:23.149287329065793 +Step 6609 | loss:1.2435535192489624 lr:9.21966353643687e-05 tokens_per_second_per_gpu:3666.683171593637 mfu:17.296370121066154 +Step 6610 | loss:1.2445358037948608 lr:9.219430904364388e-05 tokens_per_second_per_gpu:5360.469300984464 mfu:25.286248283116038 +Step 6611 | loss:1.3329284191131592 lr:9.219198240557028e-05 tokens_per_second_per_gpu:3857.031085419769 mfu:18.19427370728701 +Step 6612 | loss:1.1168256998062134 lr:9.218965545016538e-05 tokens_per_second_per_gpu:5065.957458446314 mfu:23.896985672961744 +Step 6613 | loss:1.122235655784607 lr:9.21873281774467e-05 tokens_per_second_per_gpu:4157.63632575414 mfu:19.612279914486944 +Step 6614 | loss:1.2963415384292603 lr:9.218500058743175e-05 tokens_per_second_per_gpu:5318.131160454804 mfu:25.086532050605978 +Step 6615 | loss:1.2941410541534424 lr:9.218267268013805e-05 tokens_per_second_per_gpu:4516.916100050375 mfu:21.30706390929278 +Step 6616 | loss:1.1398495435714722 lr:9.218034445558305e-05 tokens_per_second_per_gpu:4940.810108001137 mfu:23.306644268572075 +Step 6617 | loss:1.176129698753357 lr:9.217801591378433e-05 tokens_per_second_per_gpu:5324.774543044679 mfu:25.11787001976037 +Step 6618 | loss:1.3392503261566162 lr:9.217568705475934e-05 tokens_per_second_per_gpu:4636.621374690542 mfu:21.871734113596343 +Step 6619 | loss:1.3292309045791626 lr:9.217335787852563e-05 tokens_per_second_per_gpu:5334.6156917243125 mfu:25.164292397155304 +Step 6620 | loss:1.3655835390090942 lr:9.217102838510073e-05 tokens_per_second_per_gpu:5302.018168241029 mfu:25.010524317173648 +Step 6621 | loss:1.2576959133148193 lr:9.216869857450215e-05 tokens_per_second_per_gpu:5373.571474075599 mfu:25.348053469048168 +Step 6622 | loss:1.126987338066101 lr:9.216636844674738e-05 tokens_per_second_per_gpu:4859.310822148079 mfu:22.92219823199117 +Step 6623 | loss:1.2198214530944824 lr:9.216403800185399e-05 tokens_per_second_per_gpu:5247.350084322239 mfu:24.752645637991794 +Step 6624 | loss:1.2914422750473022 lr:9.216170723983949e-05 tokens_per_second_per_gpu:3920.0345389474087 mfu:18.491471747074513 +Step 6625 | loss:1.1224766969680786 lr:9.21593761607214e-05 tokens_per_second_per_gpu:4380.255512373647 mfu:20.662412600521055 +Step 6626 | loss:1.3855682611465454 lr:9.215704476451728e-05 tokens_per_second_per_gpu:5354.149454332358 mfu:25.256436488181023 +Step 6627 | loss:1.3189185857772827 lr:9.215471305124462e-05 tokens_per_second_per_gpu:4464.679060983935 mfu:21.060652883458488 +Step 6628 | loss:1.4732636213302612 lr:9.215238102092101e-05 tokens_per_second_per_gpu:4344.507777721284 mfu:20.493784436973396 +Step 6629 | loss:1.369712233543396 lr:9.215004867356395e-05 tokens_per_second_per_gpu:4127.354882699667 mfu:19.469437181052054 +Step 6630 | loss:1.2461944818496704 lr:9.2147716009191e-05 tokens_per_second_per_gpu:4680.482045200032 mfu:22.07863237979166 +Step 6631 | loss:1.338451862335205 lr:9.21453830278197e-05 tokens_per_second_per_gpu:4671.334413432508 mfu:22.035481439997326 +Step 6632 | loss:1.4077885150909424 lr:9.214304972946759e-05 tokens_per_second_per_gpu:3178.657815180049 mfu:14.99427124915137 +Step 6633 | loss:1.3886401653289795 lr:9.214071611415224e-05 tokens_per_second_per_gpu:3813.205381022059 mfu:17.987540382206795 +Step 6634 | loss:1.177572250366211 lr:9.213838218189117e-05 tokens_per_second_per_gpu:5347.3107759586055 mfu:25.224177275493187 +Step 6635 | loss:1.3807413578033447 lr:9.213604793270196e-05 tokens_per_second_per_gpu:3958.2657651254917 mfu:18.671814963876336 +Step 6636 | loss:1.1764949560165405 lr:9.213371336660217e-05 tokens_per_second_per_gpu:3966.126187275358 mfu:18.708893916283568 +Step 6637 | loss:1.1910263299942017 lr:9.213137848360932e-05 tokens_per_second_per_gpu:5335.012119964611 mfu:25.166162416802425 +Step 6638 | loss:1.139305591583252 lr:9.212904328374101e-05 tokens_per_second_per_gpu:5077.459858041311 mfu:23.95124445436308 +Step 6639 | loss:1.2881865501403809 lr:9.212670776701477e-05 tokens_per_second_per_gpu:4399.657846313616 mfu:20.753936720095414 +Step 6640 | loss:1.2606109380722046 lr:9.212437193344822e-05 tokens_per_second_per_gpu:4575.732030840171 mfu:21.584508689860822 +Step 6641 | loss:1.2259514331817627 lr:9.212203578305887e-05 tokens_per_second_per_gpu:4084.663190752395 mfu:19.268053186183028 +Step 6642 | loss:1.1603543758392334 lr:9.211969931586432e-05 tokens_per_second_per_gpu:4427.872357518869 mfu:20.887029383342597 +Step 6643 | loss:1.2786617279052734 lr:9.211736253188214e-05 tokens_per_second_per_gpu:4327.089481797003 mfu:20.41161936323083 +Step 6644 | loss:1.2633615732192993 lr:9.21150254311299e-05 tokens_per_second_per_gpu:4456.217217799533 mfu:21.020736925418337 +Step 6645 | loss:1.0897352695465088 lr:9.211268801362519e-05 tokens_per_second_per_gpu:5351.189155560705 mfu:25.242472253796084 +Step 6646 | loss:1.1422245502471924 lr:9.211035027938556e-05 tokens_per_second_per_gpu:4338.524115011339 mfu:20.465558479052458 +Step 6647 | loss:1.2441898584365845 lr:9.210801222842862e-05 tokens_per_second_per_gpu:4897.05250482381 mfu:23.10023218856767 +Step 6648 | loss:1.1299285888671875 lr:9.210567386077195e-05 tokens_per_second_per_gpu:5007.215733730685 mfu:23.61989093510641 +Step 6649 | loss:1.3462541103363037 lr:9.210333517643313e-05 tokens_per_second_per_gpu:3877.0877648149913 mfu:18.28888448601665 +Step 6650 | loss:1.2687102556228638 lr:9.210099617542976e-05 tokens_per_second_per_gpu:4980.7912961555585 mfu:23.495242354590587 +Step 6651 | loss:1.3142623901367188 lr:9.209865685777942e-05 tokens_per_second_per_gpu:4840.044886935139 mfu:22.831317528484234 +Step 6652 | loss:1.1603507995605469 lr:9.209631722349971e-05 tokens_per_second_per_gpu:4956.795316868745 mfu:23.38204922615845 +Step 6653 | loss:1.2454036474227905 lr:9.209397727260822e-05 tokens_per_second_per_gpu:3892.4881771974624 mfu:18.361530858806063 +Step 6654 | loss:1.226844310760498 lr:9.209163700512258e-05 tokens_per_second_per_gpu:4817.626483190018 mfu:22.725566093044396 +Step 6655 | loss:1.2646853923797607 lr:9.208929642106035e-05 tokens_per_second_per_gpu:3834.768079522811 mfu:18.08925531001057 +Step 6656 | loss:1.387061595916748 lr:9.208695552043914e-05 tokens_per_second_per_gpu:5339.914078217412 mfu:25.18928579024121 +Step 6657 | loss:1.2342486381530762 lr:9.208461430327659e-05 tokens_per_second_per_gpu:5335.762896080957 mfu:25.169703955838845 +Step 6658 | loss:1.2860662937164307 lr:9.208227276959027e-05 tokens_per_second_per_gpu:4726.0932123378 mfu:22.293788037248085 +Step 6659 | loss:1.1874866485595703 lr:9.207993091939781e-05 tokens_per_second_per_gpu:4469.675212680963 mfu:21.084220583444747 +Step 6660 | loss:1.3453837633132935 lr:9.207758875271682e-05 tokens_per_second_per_gpu:4764.935836248058 mfu:22.47701532146779 +Step 6661 | loss:1.2904349565505981 lr:9.207524626956492e-05 tokens_per_second_per_gpu:3974.352720110798 mfu:18.747699875259574 +Step 6662 | loss:1.254504919052124 lr:9.207290346995973e-05 tokens_per_second_per_gpu:5316.884266224552 mfu:25.08065023777979 +Step 6663 | loss:1.315270185470581 lr:9.207056035391886e-05 tokens_per_second_per_gpu:4368.096210809109 mfu:20.605055100450357 +Step 6664 | loss:1.209720492362976 lr:9.206821692145992e-05 tokens_per_second_per_gpu:4773.02565361338 mfu:22.515176370245186 +Step 6665 | loss:1.3099726438522339 lr:9.206587317260057e-05 tokens_per_second_per_gpu:4697.054310243136 mfu:22.15680658151996 +Step 6666 | loss:1.1688663959503174 lr:9.206352910735842e-05 tokens_per_second_per_gpu:4162.732043710776 mfu:19.636317285507534 +Step 6667 | loss:1.1543997526168823 lr:9.206118472575109e-05 tokens_per_second_per_gpu:4003.489467692957 mfu:18.885142884846374 +Step 6668 | loss:1.2686136960983276 lr:9.205884002779623e-05 tokens_per_second_per_gpu:5236.125058536396 mfu:24.699695276173085 +Step 6669 | loss:1.2859166860580444 lr:9.205649501351148e-05 tokens_per_second_per_gpu:5353.331948440923 mfu:25.252580173410237 +Step 6670 | loss:1.2158476114273071 lr:9.205414968291444e-05 tokens_per_second_per_gpu:4887.767096440813 mfu:23.056431332971037 +Step 6671 | loss:1.1969542503356934 lr:9.20518040360228e-05 tokens_per_second_per_gpu:4083.14970611309 mfu:19.260913820912318 +Step 6672 | loss:1.1270474195480347 lr:9.204945807285414e-05 tokens_per_second_per_gpu:4292.945198692511 mfu:20.250554954211527 +Step 6673 | loss:1.3863582611083984 lr:9.204711179342618e-05 tokens_per_second_per_gpu:4005.6531882609274 mfu:18.895349523934286 +Step 6674 | loss:1.1569105386734009 lr:9.20447651977565e-05 tokens_per_second_per_gpu:4852.273888018769 mfu:22.88900381307845 +Step 6675 | loss:1.2799080610275269 lr:9.204241828586279e-05 tokens_per_second_per_gpu:3768.8802499790895 mfu:17.778451176429915 +Step 6676 | loss:1.3628226518630981 lr:9.204007105776268e-05 tokens_per_second_per_gpu:4229.332797122717 mfu:19.950484402611934 +Step 6677 | loss:1.3016360998153687 lr:9.203772351347384e-05 tokens_per_second_per_gpu:3213.9647141345386 mfu:15.16081991549783 +Step 6678 | loss:1.226167917251587 lr:9.20353756530139e-05 tokens_per_second_per_gpu:4219.853821047354 mfu:19.90577045518436 +Step 6679 | loss:1.3776979446411133 lr:9.203302747640054e-05 tokens_per_second_per_gpu:3762.8855604660243 mfu:17.75017320319729 +Step 6680 | loss:1.28085458278656 lr:9.203067898365143e-05 tokens_per_second_per_gpu:4546.779210711771 mfu:21.44793329745466 +Step 6681 | loss:1.0766888856887817 lr:9.202833017478422e-05 tokens_per_second_per_gpu:5338.544249844365 mfu:25.18282407608873 +Step 6682 | loss:1.2115801572799683 lr:9.202598104981657e-05 tokens_per_second_per_gpu:4632.271184605679 mfu:21.851213524747077 +Step 6683 | loss:1.251393437385559 lr:9.202363160876615e-05 tokens_per_second_per_gpu:4337.595513580697 mfu:20.461178107668378 +Step 6684 | loss:1.2576849460601807 lr:9.202128185165064e-05 tokens_per_second_per_gpu:5344.012070253163 mfu:25.208616717863386 +Step 6685 | loss:1.2294504642486572 lr:9.20189317784877e-05 tokens_per_second_per_gpu:4639.069864319335 mfu:21.883284056930762 +Step 6686 | loss:1.3192942142486572 lr:9.201658138929503e-05 tokens_per_second_per_gpu:4787.937384700752 mfu:22.58551754998799 +Step 6687 | loss:1.4293843507766724 lr:9.201423068409027e-05 tokens_per_second_per_gpu:4962.911621469767 mfu:23.410900878510642 +Step 6688 | loss:1.1186041831970215 lr:9.201187966289114e-05 tokens_per_second_per_gpu:4166.849210898822 mfu:19.655738665594704 +Step 6689 | loss:1.246148705482483 lr:9.200952832571528e-05 tokens_per_second_per_gpu:5371.456739458797 mfu:25.338077905049776 +Step 6690 | loss:1.3019369840621948 lr:9.20071766725804e-05 tokens_per_second_per_gpu:4069.302966441745 mfu:19.195596387385184 +Step 6691 | loss:1.3316583633422852 lr:9.20048247035042e-05 tokens_per_second_per_gpu:3998.491392584621 mfu:18.861566111800805 +Step 6692 | loss:1.2921642065048218 lr:9.200247241850434e-05 tokens_per_second_per_gpu:5353.3034731836415 mfu:25.25244585076326 +Step 6693 | loss:1.222656488418579 lr:9.200011981759852e-05 tokens_per_second_per_gpu:3321.1387499531006 mfu:15.666378128230068 +Step 6694 | loss:1.4088739156723022 lr:9.199776690080446e-05 tokens_per_second_per_gpu:5352.810702106426 mfu:25.25012136551662 +Step 6695 | loss:1.4139894247055054 lr:9.199541366813982e-05 tokens_per_second_per_gpu:4236.397550827906 mfu:19.98381005121059 +Step 6696 | loss:1.2488552331924438 lr:9.199306011962233e-05 tokens_per_second_per_gpu:5324.79624781873 mfu:25.11797240488278 +Step 6697 | loss:1.3743538856506348 lr:9.199070625526967e-05 tokens_per_second_per_gpu:3772.115673519005 mfu:17.793713221287877 +Step 6698 | loss:1.2418015003204346 lr:9.198835207509954e-05 tokens_per_second_per_gpu:5265.089472154711 mfu:24.836325357049226 +Step 6699 | loss:1.18913996219635 lr:9.198599757912966e-05 tokens_per_second_per_gpu:4850.810823842938 mfu:22.882102289736565 +Step 6700 | loss:1.1723827123641968 lr:9.198364276737774e-05 tokens_per_second_per_gpu:4677.509136248084 mfu:22.06460866103475 +Step 6701 | loss:1.4135394096374512 lr:9.198128763986149e-05 tokens_per_second_per_gpu:3594.6897559453287 mfu:16.956764896109757 +Step 6702 | loss:1.2515063285827637 lr:9.19789321965986e-05 tokens_per_second_per_gpu:4271.990897757281 mfu:20.15170994152765 +Step 6703 | loss:1.2181421518325806 lr:9.197657643760684e-05 tokens_per_second_per_gpu:4916.95013064162 mfu:23.19409268443531 +Step 6704 | loss:1.2031474113464355 lr:9.197422036290387e-05 tokens_per_second_per_gpu:4818.02319542266 mfu:22.72743745233186 +Step 6705 | loss:1.2307013273239136 lr:9.197186397250743e-05 tokens_per_second_per_gpu:3988.2753093574497 mfu:18.813375104174703 +Step 6706 | loss:1.2006025314331055 lr:9.196950726643525e-05 tokens_per_second_per_gpu:3793.269963776167 mfu:17.893501617725004 +Step 6707 | loss:1.4428364038467407 lr:9.196715024470505e-05 tokens_per_second_per_gpu:3866.6737403016564 mfu:18.239759755571445 +Step 6708 | loss:1.1962403059005737 lr:9.196479290733455e-05 tokens_per_second_per_gpu:4517.483754949164 mfu:21.30974163430228 +Step 6709 | loss:1.2451013326644897 lr:9.196243525434149e-05 tokens_per_second_per_gpu:5067.583301473765 mfu:23.90465505191965 +Step 6710 | loss:1.1855392456054688 lr:9.196007728574361e-05 tokens_per_second_per_gpu:5343.763374837704 mfu:25.20744357915685 +Step 6711 | loss:1.2817076444625854 lr:9.195771900155862e-05 tokens_per_second_per_gpu:4488.365756095446 mfu:21.172387065666403 +Step 6712 | loss:1.0589243173599243 lr:9.195536040180429e-05 tokens_per_second_per_gpu:4534.861364681744 mfu:21.391714784336813 +Step 6713 | loss:1.2761383056640625 lr:9.195300148649833e-05 tokens_per_second_per_gpu:4954.583591502287 mfu:23.37161614024561 +Step 6714 | loss:1.3283635377883911 lr:9.19506422556585e-05 tokens_per_second_per_gpu:4732.386239951512 mfu:22.323473322203192 +Step 6715 | loss:1.223610758781433 lr:9.194828270930254e-05 tokens_per_second_per_gpu:4662.813621952295 mfu:21.995287412788226 +Step 6716 | loss:1.2710615396499634 lr:9.194592284744817e-05 tokens_per_second_per_gpu:4410.102607971739 mfu:20.803206442896883 +Step 6717 | loss:1.2763108015060425 lr:9.194356267011318e-05 tokens_per_second_per_gpu:4474.729584659302 mfu:21.108062918431358 +Step 6718 | loss:1.391212821006775 lr:9.19412021773153e-05 tokens_per_second_per_gpu:4943.487602308517 mfu:23.31927446604762 +Step 6719 | loss:1.2928874492645264 lr:9.193884136907229e-05 tokens_per_second_per_gpu:5341.583986625243 mfu:25.1971630331167 +Step 6720 | loss:1.1683356761932373 lr:9.193648024540189e-05 tokens_per_second_per_gpu:5362.448142678009 mfu:25.2955828170097 +Step 6721 | loss:1.1878571510314941 lr:9.193411880632188e-05 tokens_per_second_per_gpu:4299.609712753122 mfu:20.281992604118777 +Step 6722 | loss:1.2000194787979126 lr:9.193175705185003e-05 tokens_per_second_per_gpu:3810.089110859401 mfu:17.97284040415902 +Step 6723 | loss:1.327053189277649 lr:9.192939498200406e-05 tokens_per_second_per_gpu:4589.107136394622 mfu:21.64760134478949 +Step 6724 | loss:1.2631909847259521 lr:9.192703259680175e-05 tokens_per_second_per_gpu:3156.675367510848 mfu:14.890576292903425 +Step 6725 | loss:1.2771495580673218 lr:9.19246698962609e-05 tokens_per_second_per_gpu:4193.302932763119 mfu:19.78052538509896 +Step 6726 | loss:1.2188456058502197 lr:9.192230688039924e-05 tokens_per_second_per_gpu:5360.876628608914 mfu:25.288169717017823 +Step 6727 | loss:1.2054685354232788 lr:9.191994354923459e-05 tokens_per_second_per_gpu:4371.130580638051 mfu:20.619368740650355 +Step 6728 | loss:1.2563740015029907 lr:9.191757990278467e-05 tokens_per_second_per_gpu:5351.6086594590715 mfu:25.244451125259424 +Step 6729 | loss:1.1607292890548706 lr:9.191521594106729e-05 tokens_per_second_per_gpu:3180.077248436265 mfu:15.000966958001225 +Step 6730 | loss:1.1318060159683228 lr:9.191285166410022e-05 tokens_per_second_per_gpu:5329.98065555485 mfu:25.142428140725233 +Step 6731 | loss:1.218291997909546 lr:9.191048707190125e-05 tokens_per_second_per_gpu:5324.3659003369385 mfu:25.115942382386322 +Step 6732 | loss:1.3089781999588013 lr:9.190812216448816e-05 tokens_per_second_per_gpu:4006.6512307597563 mfu:18.900057460684874 +Step 6733 | loss:1.3064639568328857 lr:9.190575694187873e-05 tokens_per_second_per_gpu:4112.478323347183 mfu:19.39926191238363 +Step 6734 | loss:1.2358356714248657 lr:9.190339140409076e-05 tokens_per_second_per_gpu:4405.835055645168 mfu:20.783075670452934 +Step 6735 | loss:1.1450111865997314 lr:9.190102555114203e-05 tokens_per_second_per_gpu:5336.096040263625 mfu:25.171275453788745 +Step 6736 | loss:1.2194918394088745 lr:9.189865938305035e-05 tokens_per_second_per_gpu:3093.74979402927 mfu:14.593745626580905 +Step 6737 | loss:1.1980825662612915 lr:9.189629289983349e-05 tokens_per_second_per_gpu:5232.542401940802 mfu:24.682795273747416 +Step 6738 | loss:1.2645633220672607 lr:9.189392610150929e-05 tokens_per_second_per_gpu:3407.015436948615 mfu:16.071473112861582 +Step 6739 | loss:1.319176197052002 lr:9.18915589880955e-05 tokens_per_second_per_gpu:4491.259763145786 mfu:21.186038590691375 +Step 6740 | loss:1.4955954551696777 lr:9.188919155960996e-05 tokens_per_second_per_gpu:4811.040744135712 mfu:22.694500038282285 +Step 6741 | loss:1.2544628381729126 lr:9.188682381607046e-05 tokens_per_second_per_gpu:5340.6253629710345 mfu:25.19264104178174 +Step 6742 | loss:1.262835144996643 lr:9.188445575749483e-05 tokens_per_second_per_gpu:5327.781793872664 mfu:25.132055735005874 +Step 6743 | loss:1.2357532978057861 lr:9.188208738390085e-05 tokens_per_second_per_gpu:4191.0554310931875 mfu:19.769923536258563 +Step 6744 | loss:1.3149471282958984 lr:9.187971869530634e-05 tokens_per_second_per_gpu:5318.875339385197 mfu:25.090042469590657 +Step 6745 | loss:1.1939257383346558 lr:9.187734969172915e-05 tokens_per_second_per_gpu:3970.8626198125576 mfu:18.73123647668026 +Step 6746 | loss:1.2377039194107056 lr:9.187498037318705e-05 tokens_per_second_per_gpu:4400.1216589552305 mfu:20.756124603460506 +Step 6747 | loss:1.2695643901824951 lr:9.187261073969787e-05 tokens_per_second_per_gpu:3264.0585806822796 mfu:15.397121230898485 +Step 6748 | loss:1.2703033685684204 lr:9.187024079127947e-05 tokens_per_second_per_gpu:4212.601294613453 mfu:19.87155905058704 +Step 6749 | loss:1.3994568586349487 lr:9.18678705279496e-05 tokens_per_second_per_gpu:3443.036069193932 mfu:16.241388580916336 +Step 6750 | loss:1.2299046516418457 lr:9.186549994972618e-05 tokens_per_second_per_gpu:5177.777700084094 mfu:24.424460831267687 +Step 6751 | loss:1.3405091762542725 lr:9.186312905662697e-05 tokens_per_second_per_gpu:4069.6396939233105 mfu:19.197184788367363 +Step 6752 | loss:1.3377913236618042 lr:9.186075784866984e-05 tokens_per_second_per_gpu:5282.368422848375 mfu:24.917833115564225 +Step 6753 | loss:1.2598004341125488 lr:9.185838632587258e-05 tokens_per_second_per_gpu:4703.821296803242 mfu:22.188727611690943 +Step 6754 | loss:1.245643973350525 lr:9.185601448825307e-05 tokens_per_second_per_gpu:5322.257863840695 mfu:25.105998414565427 +Step 6755 | loss:1.228530764579773 lr:9.185364233582914e-05 tokens_per_second_per_gpu:4313.857891780256 mfu:20.34920369557979 +Step 6756 | loss:1.24537193775177 lr:9.185126986861863e-05 tokens_per_second_per_gpu:3321.297369151676 mfu:15.667126362052972 +Step 6757 | loss:1.1056464910507202 lr:9.184889708663937e-05 tokens_per_second_per_gpu:4417.58171239391 mfu:20.83848665452306 +Step 6758 | loss:1.3310524225234985 lr:9.184652398990922e-05 tokens_per_second_per_gpu:3489.556706272404 mfu:16.460834363254637 +Step 6759 | loss:1.3491148948669434 lr:9.184415057844602e-05 tokens_per_second_per_gpu:4182.840867670173 mfu:19.731174039043164 +Step 6760 | loss:1.191598892211914 lr:9.184177685226763e-05 tokens_per_second_per_gpu:4519.439039313406 mfu:21.318965044254327 +Step 6761 | loss:1.3875634670257568 lr:9.183940281139188e-05 tokens_per_second_per_gpu:4300.586661470835 mfu:20.28660103790487 +Step 6762 | loss:1.208979606628418 lr:9.183702845583665e-05 tokens_per_second_per_gpu:5355.163305672306 mfu:25.26121899793243 +Step 6763 | loss:1.1724424362182617 lr:9.18346537856198e-05 tokens_per_second_per_gpu:3733.0938035815107 mfu:17.60964040297517 +Step 6764 | loss:1.2110966444015503 lr:9.183227880075917e-05 tokens_per_second_per_gpu:5370.600998117227 mfu:25.334041227137064 +Step 6765 | loss:1.2327417135238647 lr:9.182990350127265e-05 tokens_per_second_per_gpu:4853.921867433754 mfu:22.89677761315322 +Step 6766 | loss:1.2740142345428467 lr:9.182752788717808e-05 tokens_per_second_per_gpu:4805.174638305078 mfu:22.666828616218325 +Step 6767 | loss:1.1596999168395996 lr:9.182515195849332e-05 tokens_per_second_per_gpu:5329.900486052638 mfu:25.142049967504974 +Step 6768 | loss:1.4074428081512451 lr:9.182277571523627e-05 tokens_per_second_per_gpu:3873.4713190250773 mfu:18.271825094196345 +Step 6769 | loss:1.2173986434936523 lr:9.182039915742479e-05 tokens_per_second_per_gpu:5086.080197628837 mfu:23.991908066979036 +Step 6770 | loss:1.2562522888183594 lr:9.181802228507675e-05 tokens_per_second_per_gpu:5362.187756788397 mfu:25.29435453234198 +Step 6771 | loss:1.130122423171997 lr:9.181564509821001e-05 tokens_per_second_per_gpu:4037.4229469220136 mfu:19.0452128960184 +Step 6772 | loss:1.2868176698684692 lr:9.18132675968425e-05 tokens_per_second_per_gpu:5354.803759453024 mfu:25.25952295707117 +Step 6773 | loss:1.4365043640136719 lr:9.181088978099204e-05 tokens_per_second_per_gpu:3627.356839218392 mfu:17.110861101487934 +Step 6774 | loss:1.1926478147506714 lr:9.180851165067657e-05 tokens_per_second_per_gpu:4210.6124190229 mfu:19.862177185089237 +Step 6775 | loss:1.277727484703064 lr:9.180613320591392e-05 tokens_per_second_per_gpu:4755.212824455044 mfu:22.431150215923267 +Step 6776 | loss:1.2069640159606934 lr:9.180375444672203e-05 tokens_per_second_per_gpu:4327.73777389559 mfu:20.414677467669904 +Step 6777 | loss:1.2528455257415771 lr:9.180137537311875e-05 tokens_per_second_per_gpu:4383.078168717424 mfu:20.675727552089484 +Step 6778 | loss:1.4496992826461792 lr:9.1798995985122e-05 tokens_per_second_per_gpu:4210.125716628538 mfu:19.85988132685504 +Step 6779 | loss:1.2780166864395142 lr:9.179661628274967e-05 tokens_per_second_per_gpu:4619.559028929432 mfu:21.791248117504814 +Step 6780 | loss:1.306435227394104 lr:9.179423626601966e-05 tokens_per_second_per_gpu:4544.886556161641 mfu:21.439005323021536 +Step 6781 | loss:1.2364790439605713 lr:9.179185593494986e-05 tokens_per_second_per_gpu:4081.7153428503298 mfu:19.254147684674273 +Step 6782 | loss:1.1373579502105713 lr:9.178947528955818e-05 tokens_per_second_per_gpu:5295.729199837682 mfu:24.980858180206404 +Step 6783 | loss:1.1602801084518433 lr:9.178709432986252e-05 tokens_per_second_per_gpu:4489.809896994602 mfu:21.179199324683566 +Step 6784 | loss:1.3808866739273071 lr:9.17847130558808e-05 tokens_per_second_per_gpu:5326.076288140662 mfu:25.124010573478735 +Step 6785 | loss:1.4119993448257446 lr:9.178233146763091e-05 tokens_per_second_per_gpu:5118.143339046861 mfu:24.143155375583888 +Step 6786 | loss:1.240875482559204 lr:9.177994956513078e-05 tokens_per_second_per_gpu:5102.365138584564 mfu:24.06872691196464 +Step 6787 | loss:1.1696966886520386 lr:9.177756734839834e-05 tokens_per_second_per_gpu:4748.212887732251 mfu:22.398130320089496 +Step 6788 | loss:1.2271925210952759 lr:9.177518481745144e-05 tokens_per_second_per_gpu:4734.570514563056 mfu:22.333776918221524 +Step 6789 | loss:1.1960656642913818 lr:9.177280197230808e-05 tokens_per_second_per_gpu:4940.075580937501 mfu:23.303179379089663 +Step 6790 | loss:1.2923556566238403 lr:9.177041881298613e-05 tokens_per_second_per_gpu:3363.494269248585 mfu:15.866176339343644 +Step 6791 | loss:1.161501169204712 lr:9.176803533950353e-05 tokens_per_second_per_gpu:3986.616726379657 mfu:18.80555128528502 +Step 6792 | loss:1.2412420511245728 lr:9.176565155187822e-05 tokens_per_second_per_gpu:5062.383288364354 mfu:23.88012570286884 +Step 6793 | loss:1.1348716020584106 lr:9.176326745012809e-05 tokens_per_second_per_gpu:5288.17745000511 mfu:24.945235287784776 +Step 6794 | loss:1.163219690322876 lr:9.176088303427113e-05 tokens_per_second_per_gpu:4043.659181259003 mfu:19.074630277397276 +Step 6795 | loss:1.2338660955429077 lr:9.175849830432521e-05 tokens_per_second_per_gpu:5337.0210607293575 mfu:25.175638933150374 +Step 6796 | loss:1.414989709854126 lr:9.175611326030831e-05 tokens_per_second_per_gpu:4457.9123775991875 mfu:21.028733283417395 +Step 6797 | loss:1.470660924911499 lr:9.175372790223835e-05 tokens_per_second_per_gpu:4531.08290381912 mfu:21.373891139775914 +Step 6798 | loss:1.1064778566360474 lr:9.175134223013328e-05 tokens_per_second_per_gpu:5327.267917888401 mfu:25.129631694311822 +Step 6799 | loss:1.399492859840393 lr:9.174895624401103e-05 tokens_per_second_per_gpu:5336.931485225261 mfu:25.175216390214395 +Step 6800 | loss:1.354186773300171 lr:9.174656994388957e-05 tokens_per_second_per_gpu:5205.954500508799 mfu:24.5573756063289 +Step 6801 | loss:1.08116614818573 lr:9.174418332978681e-05 tokens_per_second_per_gpu:3948.6975161231226 mfu:18.62667990082054 +Step 6802 | loss:1.1031197309494019 lr:9.174179640172074e-05 tokens_per_second_per_gpu:4645.072500101422 mfu:21.91159951407014 +Step 6803 | loss:1.1598179340362549 lr:9.173940915970927e-05 tokens_per_second_per_gpu:5358.992909983577 mfu:25.279283894866573 +Step 6804 | loss:1.3215725421905518 lr:9.173702160377039e-05 tokens_per_second_per_gpu:3346.324876034654 mfu:15.785185382152978 +Step 6805 | loss:1.2526906728744507 lr:9.173463373392204e-05 tokens_per_second_per_gpu:4202.649395750034 mfu:19.82461424567926 +Step 6806 | loss:1.3032816648483276 lr:9.173224555018219e-05 tokens_per_second_per_gpu:4757.639515906368 mfu:22.442597333535417 +Step 6807 | loss:1.2337079048156738 lr:9.17298570525688e-05 tokens_per_second_per_gpu:5209.676786173067 mfu:24.57493426289468 +Step 6808 | loss:1.2714893817901611 lr:9.172746824109984e-05 tokens_per_second_per_gpu:5098.727648052499 mfu:24.051568248505006 +Step 6809 | loss:1.127336859703064 lr:9.172507911579323e-05 tokens_per_second_per_gpu:3934.806582487292 mfu:18.56115399682266 +Step 6810 | loss:1.250697135925293 lr:9.1722689676667e-05 tokens_per_second_per_gpu:4073.3931495927213 mfu:19.214890479165543 +Step 6811 | loss:1.2961281538009644 lr:9.172029992373909e-05 tokens_per_second_per_gpu:5337.122066349052 mfu:25.17611539389181 +Step 6812 | loss:1.1363298892974854 lr:9.171790985702749e-05 tokens_per_second_per_gpu:4405.224967928956 mfu:20.780197782603867 +Step 6813 | loss:1.3634594678878784 lr:9.171551947655017e-05 tokens_per_second_per_gpu:4672.459292074289 mfu:22.040787684474697 +Step 6814 | loss:1.35155189037323 lr:9.17131287823251e-05 tokens_per_second_per_gpu:3871.280702709632 mfu:18.261491583278705 +Step 6815 | loss:1.184105396270752 lr:9.171073777437026e-05 tokens_per_second_per_gpu:4263.09872777318 mfu:20.109764058552 +Step 6816 | loss:1.1717146635055542 lr:9.170834645270364e-05 tokens_per_second_per_gpu:3898.9211146590874 mfu:18.391876122384858 +Step 6817 | loss:1.2402631044387817 lr:9.170595481734324e-05 tokens_per_second_per_gpu:4955.473459398883 mfu:23.37581379894097 +Step 6818 | loss:1.2974982261657715 lr:9.1703562868307e-05 tokens_per_second_per_gpu:4362.753782543166 mfu:20.579853954807653 +Step 6819 | loss:1.301180362701416 lr:9.170117060561297e-05 tokens_per_second_per_gpu:4917.221042764335 mfu:23.195370623140533 +Step 6820 | loss:1.3342400789260864 lr:9.169877802927911e-05 tokens_per_second_per_gpu:4739.9044841303685 mfu:22.358938162739587 +Step 6821 | loss:1.1964975595474243 lr:9.169638513932341e-05 tokens_per_second_per_gpu:5319.095336168326 mfu:25.091080231952 +Step 6822 | loss:1.1928606033325195 lr:9.169399193576388e-05 tokens_per_second_per_gpu:3631.239347837145 mfu:17.12917555706687 +Step 6823 | loss:1.133927345275879 lr:9.169159841861852e-05 tokens_per_second_per_gpu:5315.293074069727 mfu:25.073144312900098 +Step 6824 | loss:1.4115463495254517 lr:9.168920458790533e-05 tokens_per_second_per_gpu:5303.285675526196 mfu:25.01650336529682 +Step 6825 | loss:1.3580665588378906 lr:9.168681044364231e-05 tokens_per_second_per_gpu:5161.251272056112 mfu:24.346502850560316 +Step 6826 | loss:1.3536748886108398 lr:9.168441598584746e-05 tokens_per_second_per_gpu:5359.083308182535 mfu:25.27971031859458 +Step 6827 | loss:1.3504317998886108 lr:9.168202121453881e-05 tokens_per_second_per_gpu:4683.746154008604 mfu:22.09402973795608 +Step 6828 | loss:1.3150100708007812 lr:9.167962612973435e-05 tokens_per_second_per_gpu:4961.485610931462 mfu:23.404174143498974 +Step 6829 | loss:1.2987780570983887 lr:9.167723073145212e-05 tokens_per_second_per_gpu:4125.497186022369 mfu:19.4606741088695 +Step 6830 | loss:1.1723392009735107 lr:9.16748350197101e-05 tokens_per_second_per_gpu:5135.77458536943 mfu:24.226324972688264 +Step 6831 | loss:1.187856674194336 lr:9.167243899452634e-05 tokens_per_second_per_gpu:5331.7967015734075 mfu:25.15099473964417 +Step 6832 | loss:1.2887446880340576 lr:9.167004265591886e-05 tokens_per_second_per_gpu:5094.555954419235 mfu:24.031889657872572 +Step 6833 | loss:1.196120023727417 lr:9.166764600390564e-05 tokens_per_second_per_gpu:5312.624737099945 mfu:25.060557312148408 +Step 6834 | loss:1.251590609550476 lr:9.166524903850478e-05 tokens_per_second_per_gpu:3394.77057228164 mfu:16.013711997038122 +Step 6835 | loss:1.179814100265503 lr:9.166285175973424e-05 tokens_per_second_per_gpu:5328.883767013091 mfu:25.13725393032527 +Step 6836 | loss:1.1998786926269531 lr:9.166045416761206e-05 tokens_per_second_per_gpu:3551.224838299379 mfu:16.75173346369416 +Step 6837 | loss:1.2402312755584717 lr:9.16580562621563e-05 tokens_per_second_per_gpu:4253.4905079327 mfu:20.064440446233107 +Step 6838 | loss:1.3238128423690796 lr:9.1655658043385e-05 tokens_per_second_per_gpu:5360.339427977348 mfu:25.28563564998285 +Step 6839 | loss:1.2344236373901367 lr:9.165325951131618e-05 tokens_per_second_per_gpu:5298.674016318561 mfu:24.99474938198421 +Step 6840 | loss:1.330690622329712 lr:9.165086066596786e-05 tokens_per_second_per_gpu:4170.205674633842 mfu:19.67157167770456 +Step 6841 | loss:1.2978054285049438 lr:9.164846150735811e-05 tokens_per_second_per_gpu:5338.529539916728 mfu:25.182754686850288 +Step 6842 | loss:1.2026346921920776 lr:9.164606203550497e-05 tokens_per_second_per_gpu:5338.10281719915 mfu:25.180741762985072 +Step 6843 | loss:1.2229639291763306 lr:9.164366225042648e-05 tokens_per_second_per_gpu:5342.011796550741 mfu:25.199181085527176 +Step 6844 | loss:1.0914874076843262 lr:9.16412621521407e-05 tokens_per_second_per_gpu:3912.6087145395463 mfu:18.456442866366334 +Step 6845 | loss:1.3152036666870117 lr:9.163886174066567e-05 tokens_per_second_per_gpu:3812.005028380206 mfu:17.981878113993893 +Step 6846 | loss:1.2899248600006104 lr:9.163646101601946e-05 tokens_per_second_per_gpu:4864.394049048697 mfu:22.946176680569327 +Step 6847 | loss:1.2038360834121704 lr:9.16340599782201e-05 tokens_per_second_per_gpu:2724.514586128176 mfu:12.852000152888825 +Step 6848 | loss:1.392948865890503 lr:9.163165862728566e-05 tokens_per_second_per_gpu:5226.1189499421425 mfu:24.65249475471582 +Step 6849 | loss:1.2288118600845337 lr:9.16292569632342e-05 tokens_per_second_per_gpu:4607.405710373204 mfu:21.733918840305115 +Step 6850 | loss:1.2569913864135742 lr:9.162685498608378e-05 tokens_per_second_per_gpu:5313.064957261513 mfu:25.062633905759824 +Step 6851 | loss:1.227471947669983 lr:9.162445269585249e-05 tokens_per_second_per_gpu:3234.1554222701607 mfu:15.256062930663594 +Step 6852 | loss:1.0812562704086304 lr:9.162205009255838e-05 tokens_per_second_per_gpu:5292.862475034415 mfu:24.967335350195704 +Step 6853 | loss:1.2079589366912842 lr:9.161964717621952e-05 tokens_per_second_per_gpu:4976.468007014486 mfu:23.47484866208296 +Step 6854 | loss:1.212632656097412 lr:9.161724394685398e-05 tokens_per_second_per_gpu:4820.619797999442 mfu:22.739686069712622 +Step 6855 | loss:1.172432780265808 lr:9.161484040447983e-05 tokens_per_second_per_gpu:4188.885143591787 mfu:19.759685919825227 +Step 6856 | loss:1.2563143968582153 lr:9.161243654911517e-05 tokens_per_second_per_gpu:3164.6460614014845 mfu:14.92817541592617 +Step 6857 | loss:1.3675792217254639 lr:9.161003238077806e-05 tokens_per_second_per_gpu:5084.865927507149 mfu:23.986180147639185 +Step 6858 | loss:1.3473891019821167 lr:9.160762789948657e-05 tokens_per_second_per_gpu:5323.441225839063 mfu:25.111580535014102 +Step 6859 | loss:1.3899034261703491 lr:9.160522310525881e-05 tokens_per_second_per_gpu:3181.140223422959 mfu:15.005981192375696 +Step 6860 | loss:1.432554841041565 lr:9.160281799811287e-05 tokens_per_second_per_gpu:5343.723444177499 mfu:25.207255219419793 +Step 6861 | loss:1.3099958896636963 lr:9.160041257806682e-05 tokens_per_second_per_gpu:3139.073853474074 mfu:14.807546948063996 +Step 6862 | loss:1.3096832036972046 lr:9.159800684513875e-05 tokens_per_second_per_gpu:4802.983291995139 mfu:22.65649166179211 +Step 6863 | loss:1.0325671434402466 lr:9.159560079934678e-05 tokens_per_second_per_gpu:5349.488727631377 mfu:25.23445104513052 +Step 6864 | loss:1.2343531847000122 lr:9.159319444070898e-05 tokens_per_second_per_gpu:4795.296944458631 mfu:22.62023384903628 +Step 6865 | loss:1.274787187576294 lr:9.159078776924346e-05 tokens_per_second_per_gpu:4475.174850302956 mfu:21.11016331244262 +Step 6866 | loss:1.213053822517395 lr:9.158838078496831e-05 tokens_per_second_per_gpu:5236.729325911574 mfu:24.702545708480447 +Step 6867 | loss:1.3778642416000366 lr:9.158597348790164e-05 tokens_per_second_per_gpu:5084.0106048374355 mfu:23.982145444673055 +Step 6868 | loss:1.3924919366836548 lr:9.158356587806157e-05 tokens_per_second_per_gpu:2471.1824998517172 mfu:11.656989479011846 +Step 6869 | loss:1.3145173788070679 lr:9.158115795546617e-05 tokens_per_second_per_gpu:4469.857477342108 mfu:21.08508035694892 +Step 6870 | loss:1.2423810958862305 lr:9.15787497201336e-05 tokens_per_second_per_gpu:5337.339294329504 mfu:25.177140095338515 +Step 6871 | loss:1.2550277709960938 lr:9.157634117208196e-05 tokens_per_second_per_gpu:4882.093800632799 mfu:23.02966942868056 +Step 6872 | loss:1.3381826877593994 lr:9.157393231132933e-05 tokens_per_second_per_gpu:3666.6059022086747 mfu:17.29600562819377 +Step 6873 | loss:1.1392003297805786 lr:9.157152313789385e-05 tokens_per_second_per_gpu:4843.253074005671 mfu:22.84645109426909 +Step 6874 | loss:1.3366912603378296 lr:9.156911365179364e-05 tokens_per_second_per_gpu:4160.554525322986 mfu:19.62604555974938 +Step 6875 | loss:1.15023934841156 lr:9.156670385304683e-05 tokens_per_second_per_gpu:5357.020651223805 mfu:25.26998041379468 +Step 6876 | loss:1.233681321144104 lr:9.156429374167154e-05 tokens_per_second_per_gpu:3981.9260261034246 mfu:18.783424451766038 +Step 6877 | loss:1.3319092988967896 lr:9.156188331768588e-05 tokens_per_second_per_gpu:4678.196291660787 mfu:22.06785009035727 +Step 6878 | loss:1.2120789289474487 lr:9.155947258110799e-05 tokens_per_second_per_gpu:5290.968193530393 mfu:24.95839970114348 +Step 6879 | loss:1.4608101844787598 lr:9.155706153195602e-05 tokens_per_second_per_gpu:5325.7219896873785 mfu:25.12233928722505 +Step 6880 | loss:1.1132491827011108 lr:9.155465017024808e-05 tokens_per_second_per_gpu:5145.399233290574 mfu:24.271726079066564 +Step 6881 | loss:1.3463997840881348 lr:9.155223849600232e-05 tokens_per_second_per_gpu:4552.854485815698 mfu:21.476591406668707 +Step 6882 | loss:1.3766348361968994 lr:9.154982650923685e-05 tokens_per_second_per_gpu:5322.3777469964 mfu:25.10656392386506 +Step 6883 | loss:1.2939084768295288 lr:9.154741420996987e-05 tokens_per_second_per_gpu:5356.48777690809 mfu:25.267466754729725 +Step 6884 | loss:1.127845287322998 lr:9.154500159821946e-05 tokens_per_second_per_gpu:4821.399406604679 mfu:22.74336361651847 +Step 6885 | loss:1.2753242254257202 lr:9.154258867400379e-05 tokens_per_second_per_gpu:4532.86567334339 mfu:21.382300767793748 +Step 6886 | loss:1.1704509258270264 lr:9.154017543734101e-05 tokens_per_second_per_gpu:4820.060958192662 mfu:22.737049926996132 +Step 6887 | loss:1.3190343379974365 lr:9.153776188824927e-05 tokens_per_second_per_gpu:3516.1886903271675 mfu:16.586461975925985 +Step 6888 | loss:1.2033604383468628 lr:9.153534802674674e-05 tokens_per_second_per_gpu:4217.289049006352 mfu:19.893671987871716 +Step 6889 | loss:1.167346477508545 lr:9.153293385285154e-05 tokens_per_second_per_gpu:5339.614065177189 mfu:25.187870577542977 +Step 6890 | loss:1.343651294708252 lr:9.153051936658186e-05 tokens_per_second_per_gpu:3911.7268058996733 mfu:18.452282752842923 +Step 6891 | loss:1.27097749710083 lr:9.152810456795582e-05 tokens_per_second_per_gpu:5345.685324435549 mfu:25.216509743328302 +Step 6892 | loss:1.324217677116394 lr:9.152568945699162e-05 tokens_per_second_per_gpu:4685.6309848535475 mfu:22.102920806636984 +Step 6893 | loss:1.3451951742172241 lr:9.152327403370743e-05 tokens_per_second_per_gpu:3653.2100487114976 mfu:17.2328151016793 +Step 6894 | loss:1.2140225172042847 lr:9.152085829812138e-05 tokens_per_second_per_gpu:4128.819279672184 mfu:19.476344991423304 +Step 6895 | loss:1.3326711654663086 lr:9.151844225025165e-05 tokens_per_second_per_gpu:4033.472788383279 mfu:19.02657932422464 +Step 6896 | loss:1.0724806785583496 lr:9.151602589011644e-05 tokens_per_second_per_gpu:4460.820795817934 mfu:21.04245278838254 +Step 6897 | loss:1.2446547746658325 lr:9.151360921773388e-05 tokens_per_second_per_gpu:5344.700441159159 mfu:25.211863880875153 +Step 6898 | loss:1.2783117294311523 lr:9.151119223312217e-05 tokens_per_second_per_gpu:5284.604391141328 mfu:24.928380559497615 +Step 6899 | loss:1.2260128259658813 lr:9.15087749362995e-05 tokens_per_second_per_gpu:5332.511314507002 mfu:25.15436569077747 +Step 6900 | loss:1.1968648433685303 lr:9.150635732728401e-05 tokens_per_second_per_gpu:4504.70917097576 mfu:21.249481742130957 +Step 6901 | loss:1.393951416015625 lr:9.150393940609394e-05 tokens_per_second_per_gpu:5346.214019013056 mfu:25.219003685854997 +Step 6902 | loss:1.1815569400787354 lr:9.150152117274745e-05 tokens_per_second_per_gpu:4904.5425260397715 mfu:23.135563896572755 +Step 6903 | loss:1.1769778728485107 lr:9.149910262726269e-05 tokens_per_second_per_gpu:5313.230331825711 mfu:25.06341400579496 +Step 6904 | loss:1.2517554759979248 lr:9.14966837696579e-05 tokens_per_second_per_gpu:4213.822392707731 mfu:19.87731917863857 +Step 6905 | loss:1.2985758781433105 lr:9.149426459995126e-05 tokens_per_second_per_gpu:4872.803077179606 mfu:22.985843500990967 +Step 6906 | loss:1.2506309747695923 lr:9.149184511816095e-05 tokens_per_second_per_gpu:5338.672440250775 mfu:25.183428772107302 +Step 6907 | loss:1.1014424562454224 lr:9.14894253243052e-05 tokens_per_second_per_gpu:4108.6716420886 mfu:19.3813051474001 +Step 6908 | loss:1.2670294046401978 lr:9.148700521840218e-05 tokens_per_second_per_gpu:5218.106084969717 mfu:24.61469670350515 +Step 6909 | loss:1.2868515253067017 lr:9.148458480047009e-05 tokens_per_second_per_gpu:3195.3159907056056 mfu:15.072850705283239 +Step 6910 | loss:1.3244379758834839 lr:9.148216407052716e-05 tokens_per_second_per_gpu:3903.195887107862 mfu:18.41204095337711 +Step 6911 | loss:1.282405138015747 lr:9.147974302859157e-05 tokens_per_second_per_gpu:3882.765826342481 mfu:18.31566887101901 +Step 6912 | loss:1.2289557456970215 lr:9.147732167468153e-05 tokens_per_second_per_gpu:5354.4353388905 mfu:25.25778505441872 +Step 6913 | loss:1.237489938735962 lr:9.147490000881528e-05 tokens_per_second_per_gpu:5189.45378157287 mfu:24.479538899795166 +Step 6914 | loss:1.187572717666626 lr:9.1472478031011e-05 tokens_per_second_per_gpu:5324.965098855486 mfu:25.118768904032116 +Step 6915 | loss:1.3867745399475098 lr:9.147005574128693e-05 tokens_per_second_per_gpu:3805.1567004972435 mfu:17.949573382924736 +Step 6916 | loss:1.28312349319458 lr:9.146763313966127e-05 tokens_per_second_per_gpu:5336.790370926355 mfu:25.174550729989868 +Step 6917 | loss:1.2056885957717896 lr:9.146521022615226e-05 tokens_per_second_per_gpu:3743.4039330795026 mfu:17.65827504290668 +Step 6918 | loss:1.408344030380249 lr:9.146278700077812e-05 tokens_per_second_per_gpu:5383.420799220332 mfu:25.39451441622402 +Step 6919 | loss:1.288079857826233 lr:9.146036346355706e-05 tokens_per_second_per_gpu:4140.234173558519 mfu:19.53019099346687 +Step 6920 | loss:1.137156367301941 lr:9.145793961450731e-05 tokens_per_second_per_gpu:5315.113408021096 mfu:25.07229679749452 +Step 6921 | loss:1.1968417167663574 lr:9.14555154536471e-05 tokens_per_second_per_gpu:4973.342291518434 mfu:23.460104128786025 +Step 6922 | loss:1.3040884733200073 lr:9.145309098099468e-05 tokens_per_second_per_gpu:5336.285880860798 mfu:25.172170964275757 +Step 6923 | loss:1.1932355165481567 lr:9.145066619656828e-05 tokens_per_second_per_gpu:4246.724515413922 mfu:20.03252410512467 +Step 6924 | loss:1.369738221168518 lr:9.144824110038613e-05 tokens_per_second_per_gpu:5391.276062467737 mfu:25.43156903320814 +Step 6925 | loss:1.200742244720459 lr:9.144581569246646e-05 tokens_per_second_per_gpu:5357.877508729867 mfu:25.274022356845748 +Step 6926 | loss:1.2377862930297852 lr:9.144338997282751e-05 tokens_per_second_per_gpu:4088.0585664326595 mfu:19.284069752577665 +Step 6927 | loss:1.298059105873108 lr:9.144096394148756e-05 tokens_per_second_per_gpu:4847.033970313643 mfu:22.864286227241223 +Step 6928 | loss:1.316067099571228 lr:9.143853759846481e-05 tokens_per_second_per_gpu:4487.766145455588 mfu:21.169558599974067 +Step 6929 | loss:1.2144569158554077 lr:9.143611094377756e-05 tokens_per_second_per_gpu:4158.835440574666 mfu:19.617936343685454 +Step 6930 | loss:1.1175066232681274 lr:9.143368397744401e-05 tokens_per_second_per_gpu:4936.070123002544 mfu:23.28428495060888 +Step 6931 | loss:1.1869679689407349 lr:9.143125669948244e-05 tokens_per_second_per_gpu:4063.2350181314846 mfu:19.166972839906947 +Step 6932 | loss:1.1559910774230957 lr:9.142882910991109e-05 tokens_per_second_per_gpu:5362.917888798466 mfu:25.297798689606843 +Step 6933 | loss:1.2897928953170776 lr:9.142640120874826e-05 tokens_per_second_per_gpu:5358.992248831259 mfu:25.279280776098283 +Step 6934 | loss:1.241062045097351 lr:9.142397299601215e-05 tokens_per_second_per_gpu:5188.270894707515 mfu:24.47395902062998 +Step 6935 | loss:1.1471792459487915 lr:9.142154447172106e-05 tokens_per_second_per_gpu:5318.043553632907 mfu:25.086118794279823 +Step 6936 | loss:1.1898887157440186 lr:9.141911563589328e-05 tokens_per_second_per_gpu:5164.104125854878 mfu:24.359960248676142 +Step 6937 | loss:1.1763172149658203 lr:9.1416686488547e-05 tokens_per_second_per_gpu:5309.309646663415 mfu:25.044919464946545 +Step 6938 | loss:1.3069075345993042 lr:9.141425702970057e-05 tokens_per_second_per_gpu:4666.984451204133 mfu:22.014961925985993 +Step 6939 | loss:1.245320439338684 lr:9.141182725937221e-05 tokens_per_second_per_gpu:3985.1162096939393 mfu:18.798473092063944 +Step 6940 | loss:1.2920317649841309 lr:9.140939717758022e-05 tokens_per_second_per_gpu:4469.814365207026 mfu:21.084876989652052 +Step 6941 | loss:1.4040286540985107 lr:9.140696678434286e-05 tokens_per_second_per_gpu:3833.0509385456935 mfu:18.081155263047044 +Step 6942 | loss:1.2903002500534058 lr:9.140453607967843e-05 tokens_per_second_per_gpu:4298.477007718848 mfu:20.2766494412127 +Step 6943 | loss:1.1928105354309082 lr:9.14021050636052e-05 tokens_per_second_per_gpu:4117.034345868727 mfu:19.420753447955352 +Step 6944 | loss:1.265821099281311 lr:9.139967373614147e-05 tokens_per_second_per_gpu:5045.535487349481 mfu:23.800651750950415 +Step 6945 | loss:1.2058937549591064 lr:9.139724209730549e-05 tokens_per_second_per_gpu:5018.805694084274 mfu:23.674562755545434 +Step 6946 | loss:1.1731535196304321 lr:9.139481014711557e-05 tokens_per_second_per_gpu:5326.840425732729 mfu:25.127615140875836 +Step 6947 | loss:1.3018735647201538 lr:9.139237788559e-05 tokens_per_second_per_gpu:4562.1563182661175 mfu:21.5204697373936 +Step 6948 | loss:1.1105456352233887 lr:9.138994531274708e-05 tokens_per_second_per_gpu:5354.245176003105 mfu:25.25688802363374 +Step 6949 | loss:1.3686871528625488 lr:9.138751242860511e-05 tokens_per_second_per_gpu:4957.34729239683 mfu:23.384652988901166 +Step 6950 | loss:1.299113392829895 lr:9.138507923318236e-05 tokens_per_second_per_gpu:4445.495711119722 mfu:20.970161749132 +Step 6951 | loss:1.2235745191574097 lr:9.138264572649715e-05 tokens_per_second_per_gpu:5366.703622425257 mfu:25.315656641037176 +Step 6952 | loss:1.2834711074829102 lr:9.13802119085678e-05 tokens_per_second_per_gpu:5317.680640025476 mfu:25.084406868874126 +Step 6953 | loss:1.29558265209198 lr:9.137777777941259e-05 tokens_per_second_per_gpu:4350.836944219897 mfu:20.523640195214515 +Step 6954 | loss:1.0394924879074097 lr:9.137534333904983e-05 tokens_per_second_per_gpu:4671.461948404957 mfu:22.03608304422208 +Step 6955 | loss:1.172853708267212 lr:9.137290858749782e-05 tokens_per_second_per_gpu:4449.507772918056 mfu:20.989087329163073 +Step 6956 | loss:1.3155322074890137 lr:9.13704735247749e-05 tokens_per_second_per_gpu:4387.5906186738675 mfu:20.69701354843738 +Step 6957 | loss:1.2406933307647705 lr:9.136803815089936e-05 tokens_per_second_per_gpu:5293.16505038579 mfu:24.96876264975282 +Step 6958 | loss:1.3102582693099976 lr:9.136560246588952e-05 tokens_per_second_per_gpu:5367.3829711660865 mfu:25.31886124495652 +Step 6959 | loss:1.2616000175476074 lr:9.136316646976373e-05 tokens_per_second_per_gpu:4260.258210841716 mfu:20.09636485554412 +Step 6960 | loss:1.3344836235046387 lr:9.136073016254027e-05 tokens_per_second_per_gpu:3869.9401479221187 mfu:18.255167957623772 +Step 6961 | loss:1.168906331062317 lr:9.135829354423749e-05 tokens_per_second_per_gpu:4923.117083690963 mfu:23.223183254159753 +Step 6962 | loss:1.1929413080215454 lr:9.135585661487369e-05 tokens_per_second_per_gpu:3793.790554796086 mfu:17.895957334387948 +Step 6963 | loss:1.1892917156219482 lr:9.135341937446723e-05 tokens_per_second_per_gpu:5339.363835899776 mfu:25.186690203347897 +Step 6964 | loss:1.2959507703781128 lr:9.135098182303643e-05 tokens_per_second_per_gpu:5028.283639325674 mfu:23.719271840353326 +Step 6965 | loss:1.0608915090560913 lr:9.134854396059961e-05 tokens_per_second_per_gpu:3961.1318194282153 mfu:18.685334631021572 +Step 6966 | loss:1.2385362386703491 lr:9.134610578717511e-05 tokens_per_second_per_gpu:3747.978606031208 mfu:17.679854555739567 +Step 6967 | loss:1.2004197835922241 lr:9.134366730278128e-05 tokens_per_second_per_gpu:4620.568913894788 mfu:21.79601191719054 +Step 6968 | loss:1.3332455158233643 lr:9.134122850743645e-05 tokens_per_second_per_gpu:3454.8098630605987 mfu:16.296927575401778 +Step 6969 | loss:1.1966538429260254 lr:9.133878940115898e-05 tokens_per_second_per_gpu:4996.008692462803 mfu:23.567025409327396 +Step 6970 | loss:1.1627026796340942 lr:9.133634998396719e-05 tokens_per_second_per_gpu:5312.08510381493 mfu:25.05801176987206 +Step 6971 | loss:1.2456955909729004 lr:9.133391025587943e-05 tokens_per_second_per_gpu:4447.819319997707 mfu:20.98112260865801 +Step 6972 | loss:1.263100504875183 lr:9.133147021691407e-05 tokens_per_second_per_gpu:5301.213028358043 mfu:25.00672633497457 +Step 6973 | loss:1.2387616634368896 lr:9.132902986708944e-05 tokens_per_second_per_gpu:4411.869796148179 mfu:20.811542571038544 +Step 6974 | loss:1.3378299474716187 lr:9.13265892064239e-05 tokens_per_second_per_gpu:4307.352965603157 mfu:20.318518848947562 +Step 6975 | loss:1.2945518493652344 lr:9.13241482349358e-05 tokens_per_second_per_gpu:5115.23520050428 mfu:24.129437189898933 +Step 6976 | loss:1.2982715368270874 lr:9.132170695264352e-05 tokens_per_second_per_gpu:5226.768023441608 mfu:24.655556545156163 +Step 6977 | loss:1.2804192304611206 lr:9.131926535956541e-05 tokens_per_second_per_gpu:5171.961161324416 mfu:24.39702322553444 +Step 6978 | loss:1.337809681892395 lr:9.131682345571984e-05 tokens_per_second_per_gpu:5359.0820641948585 mfu:25.279704450492464 +Step 6979 | loss:1.295696496963501 lr:9.131438124112516e-05 tokens_per_second_per_gpu:5117.140504891452 mfu:24.138424835771673 +Step 6980 | loss:1.1918781995773315 lr:9.131193871579975e-05 tokens_per_second_per_gpu:5344.937805976414 mfu:25.21298357121676 +Step 6981 | loss:1.2676918506622314 lr:9.130949587976198e-05 tokens_per_second_per_gpu:5337.231831030513 mfu:25.176633172622044 +Step 6982 | loss:1.2427141666412354 lr:9.130705273303022e-05 tokens_per_second_per_gpu:3165.00851066551 mfu:14.929885150944799 +Step 6983 | loss:1.2158615589141846 lr:9.130460927562284e-05 tokens_per_second_per_gpu:3792.147532890277 mfu:17.888206919729804 +Step 6984 | loss:1.2633453607559204 lr:9.130216550755821e-05 tokens_per_second_per_gpu:4247.40446182995 mfu:20.0357315283799 +Step 6985 | loss:1.223697304725647 lr:9.129972142885475e-05 tokens_per_second_per_gpu:5332.847308618985 mfu:25.155950632330203 +Step 6986 | loss:1.343097448348999 lr:9.129727703953079e-05 tokens_per_second_per_gpu:5150.183309632651 mfu:24.29429338341974 +Step 6987 | loss:1.2977964878082275 lr:9.129483233960476e-05 tokens_per_second_per_gpu:4737.864579208362 mfu:22.34931558317881 +Step 6988 | loss:1.2575790882110596 lr:9.129238732909502e-05 tokens_per_second_per_gpu:4199.376970500572 mfu:19.809177657444224 +Step 6989 | loss:1.5803457498550415 lr:9.128994200801997e-05 tokens_per_second_per_gpu:3301.6182498513604 mfu:15.574296598709408 +Step 6990 | loss:1.2956472635269165 lr:9.1287496376398e-05 tokens_per_second_per_gpu:4848.318229633832 mfu:22.870344297571787 +Step 6991 | loss:1.2829976081848145 lr:9.12850504342475e-05 tokens_per_second_per_gpu:5358.18186731613 mfu:25.275458068226076 +Step 6992 | loss:1.3428864479064941 lr:9.128260418158685e-05 tokens_per_second_per_gpu:3822.988503895868 mfu:18.033689041975432 +Step 6993 | loss:1.298033356666565 lr:9.128015761843447e-05 tokens_per_second_per_gpu:5342.937942775886 mfu:25.203549875288026 +Step 6994 | loss:1.1356089115142822 lr:9.127771074480878e-05 tokens_per_second_per_gpu:3988.6332714928803 mfu:18.815063672642168 +Step 6995 | loss:1.25045645236969 lr:9.127526356072813e-05 tokens_per_second_per_gpu:4526.6255426079 mfu:21.35286500643835 +Step 6996 | loss:1.194114089012146 lr:9.127281606621099e-05 tokens_per_second_per_gpu:5366.7335387121775 mfu:25.315797761266825 +Step 6997 | loss:1.2738614082336426 lr:9.127036826127571e-05 tokens_per_second_per_gpu:5343.776592209997 mfu:25.207505927757076 +Step 6998 | loss:1.3046183586120605 lr:9.126792014594072e-05 tokens_per_second_per_gpu:4628.449874614953 mfu:21.83318775353811 +Step 6999 | loss:1.1098310947418213 lr:9.126547172022446e-05 tokens_per_second_per_gpu:5344.9906960565 mfu:25.213233062748504 +Step 7000 | loss:1.4029467105865479 lr:9.12630229841453e-05 tokens_per_second_per_gpu:4338.167185368515 mfu:20.463874781028778 +Step 7001 | loss:1.2915127277374268 lr:9.126057393772169e-05 tokens_per_second_per_gpu:4735.405493118581 mfu:22.337715654529966 +Step 7002 | loss:1.3644903898239136 lr:9.125812458097204e-05 tokens_per_second_per_gpu:5230.363824060251 mfu:24.67251855018115 +Step 7003 | loss:1.0762344598770142 lr:9.125567491391476e-05 tokens_per_second_per_gpu:4918.390129261513 mfu:23.200885403614446 +Step 7004 | loss:1.3773376941680908 lr:9.125322493656829e-05 tokens_per_second_per_gpu:5174.254266003276 mfu:24.407840191548622 +Step 7005 | loss:1.2687933444976807 lr:9.125077464895106e-05 tokens_per_second_per_gpu:5337.074924412322 mfu:25.175893017333006 +Step 7006 | loss:1.3802878856658936 lr:9.124832405108148e-05 tokens_per_second_per_gpu:5349.327730853609 mfu:25.23369159586091 +Step 7007 | loss:1.3887826204299927 lr:9.124587314297799e-05 tokens_per_second_per_gpu:3556.759259721025 mfu:16.77784027380973 +Step 7008 | loss:1.1956855058670044 lr:9.124342192465904e-05 tokens_per_second_per_gpu:5368.903649365106 mfu:25.326034543475746 +Step 7009 | loss:1.1834114789962769 lr:9.124097039614304e-05 tokens_per_second_per_gpu:5345.63662928975 mfu:25.216280040017317 +Step 7010 | loss:1.1598689556121826 lr:9.123851855744843e-05 tokens_per_second_per_gpu:5306.317238517951 mfu:25.03080376516691 +Step 7011 | loss:1.2398158311843872 lr:9.123606640859368e-05 tokens_per_second_per_gpu:5340.677254926811 mfu:25.192885824990814 +Step 7012 | loss:1.1420570611953735 lr:9.12336139495972e-05 tokens_per_second_per_gpu:5077.867015087511 mfu:23.953165083618163 +Step 7013 | loss:1.5186601877212524 lr:9.123116118047745e-05 tokens_per_second_per_gpu:3154.8761396922973 mfu:14.882089028303401 +Step 7014 | loss:1.1604735851287842 lr:9.122870810125287e-05 tokens_per_second_per_gpu:3597.930596178197 mfu:16.972052492432557 +Step 7015 | loss:1.1156855821609497 lr:9.122625471194192e-05 tokens_per_second_per_gpu:4446.063954442899 mfu:20.972842249839736 +Step 7016 | loss:1.2302898168563843 lr:9.122380101256305e-05 tokens_per_second_per_gpu:4128.600701688319 mfu:19.475313921780163 +Step 7017 | loss:1.2587599754333496 lr:9.122134700313472e-05 tokens_per_second_per_gpu:5160.9043337426265 mfu:24.344866283341823 +Step 7018 | loss:1.199730634689331 lr:9.121889268367536e-05 tokens_per_second_per_gpu:5329.193601567128 mfu:25.13871547277989 +Step 7019 | loss:1.3695968389511108 lr:9.121643805420345e-05 tokens_per_second_per_gpu:3399.921409133132 mfu:16.038009373289995 +Step 7020 | loss:1.3091028928756714 lr:9.121398311473748e-05 tokens_per_second_per_gpu:4222.321490240799 mfu:19.917410871797617 +Step 7021 | loss:1.1406261920928955 lr:9.121152786529586e-05 tokens_per_second_per_gpu:5212.910797820628 mfu:24.59018964761551 +Step 7022 | loss:1.2897001504898071 lr:9.120907230589708e-05 tokens_per_second_per_gpu:3971.268251688972 mfu:18.733149911449242 +Step 7023 | loss:1.0697020292282104 lr:9.120661643655961e-05 tokens_per_second_per_gpu:4953.374797510461 mfu:23.36591405274464 +Step 7024 | loss:1.2335689067840576 lr:9.120416025730193e-05 tokens_per_second_per_gpu:4422.90418692751 mfu:20.863593675005845 +Step 7025 | loss:1.1656500101089478 lr:9.12017037681425e-05 tokens_per_second_per_gpu:4093.724751765304 mfu:19.310798115542653 +Step 7026 | loss:1.2260574102401733 lr:9.119924696909979e-05 tokens_per_second_per_gpu:4679.130586906652 mfu:22.072257320439224 +Step 7027 | loss:1.3580656051635742 lr:9.119678986019229e-05 tokens_per_second_per_gpu:3567.766369239498 mfu:16.829762687414906 +Step 7028 | loss:1.3049378395080566 lr:9.119433244143848e-05 tokens_per_second_per_gpu:4102.448648263308 mfu:19.351950223773965 +Step 7029 | loss:1.383090853691101 lr:9.119187471285684e-05 tokens_per_second_per_gpu:5267.059193518942 mfu:24.845616868793424 +Step 7030 | loss:1.453394889831543 lr:9.118941667446584e-05 tokens_per_second_per_gpu:5354.809224593213 mfu:25.259548737069892 +Step 7031 | loss:1.274945855140686 lr:9.118695832628399e-05 tokens_per_second_per_gpu:4314.55536627147 mfu:20.352493801756264 +Step 7032 | loss:1.2296901941299438 lr:9.118449966832978e-05 tokens_per_second_per_gpu:4576.850985693091 mfu:21.58978699081543 +Step 7033 | loss:1.237290382385254 lr:9.118204070062169e-05 tokens_per_second_per_gpu:3986.5929686117806 mfu:18.805439215840455 +Step 7034 | loss:1.3156217336654663 lr:9.117958142317821e-05 tokens_per_second_per_gpu:4077.85780766868 mfu:19.235951033059063 +Step 7035 | loss:1.633359670639038 lr:9.117712183601785e-05 tokens_per_second_per_gpu:3781.5107722519488 mfu:17.838031504980307 +Step 7036 | loss:1.1648173332214355 lr:9.11746619391591e-05 tokens_per_second_per_gpu:5112.24582382265 mfu:24.11533579005143 +Step 7037 | loss:1.2422020435333252 lr:9.117220173262046e-05 tokens_per_second_per_gpu:5383.9257589266535 mfu:25.39689639731375 +Step 7038 | loss:1.202294945716858 lr:9.116974121642043e-05 tokens_per_second_per_gpu:4522.225457068197 mfu:21.332109052215568 +Step 7039 | loss:1.2518718242645264 lr:9.116728039057754e-05 tokens_per_second_per_gpu:3894.7099002958626 mfu:18.372011105726266 +Step 7040 | loss:1.2317149639129639 lr:9.116481925511028e-05 tokens_per_second_per_gpu:4489.5662955893695 mfu:21.178050215292398 +Step 7041 | loss:1.153455376625061 lr:9.116235781003715e-05 tokens_per_second_per_gpu:4184.616747869193 mfu:19.739551168938345 +Step 7042 | loss:1.3070521354675293 lr:9.11598960553767e-05 tokens_per_second_per_gpu:3262.733964358919 mfu:15.390872789698344 +Step 7043 | loss:1.2606356143951416 lr:9.115743399114738e-05 tokens_per_second_per_gpu:3761.5901107196396 mfu:17.744062345717072 +Step 7044 | loss:1.383632779121399 lr:9.115497161736777e-05 tokens_per_second_per_gpu:4378.502263723241 mfu:20.654142227502025 +Step 7045 | loss:1.3017685413360596 lr:9.115250893405637e-05 tokens_per_second_per_gpu:5328.429206748159 mfu:25.135109691999613 +Step 7046 | loss:1.3770031929016113 lr:9.115004594123169e-05 tokens_per_second_per_gpu:5324.117297929853 mfu:25.114769682416153 +Step 7047 | loss:1.1661704778671265 lr:9.114758263891227e-05 tokens_per_second_per_gpu:3558.729381890017 mfu:16.787133676218367 +Step 7048 | loss:1.29051673412323 lr:9.114511902711662e-05 tokens_per_second_per_gpu:5051.282791965765 mfu:23.82776276741671 +Step 7049 | loss:1.251534342765808 lr:9.114265510586329e-05 tokens_per_second_per_gpu:5330.11837813401 mfu:25.143077801629023 +Step 7050 | loss:1.2720516920089722 lr:9.114019087517081e-05 tokens_per_second_per_gpu:3599.8672761053963 mfu:16.98118814208083 +Step 7051 | loss:1.2826672792434692 lr:9.113772633505768e-05 tokens_per_second_per_gpu:4159.3571063000445 mfu:19.620397129917414 +Step 7052 | loss:1.1847351789474487 lr:9.113526148554248e-05 tokens_per_second_per_gpu:4201.6970881332845 mfu:19.820122048170617 +Step 7053 | loss:1.2344965934753418 lr:9.113279632664373e-05 tokens_per_second_per_gpu:5169.720187395887 mfu:24.386452169163988 +Step 7054 | loss:1.1551495790481567 lr:9.113033085837996e-05 tokens_per_second_per_gpu:5324.832886155511 mfu:25.118145234168527 +Step 7055 | loss:1.3191862106323242 lr:9.112786508076972e-05 tokens_per_second_per_gpu:4335.82763666294 mfu:20.45283873061641 +Step 7056 | loss:1.2779496908187866 lr:9.112539899383156e-05 tokens_per_second_per_gpu:4337.542278007933 mfu:20.460926986388724 +Step 7057 | loss:1.2721456289291382 lr:9.112293259758404e-05 tokens_per_second_per_gpu:4155.10655748437 mfu:19.600346565935467 +Step 7058 | loss:1.2194279432296753 lr:9.11204658920457e-05 tokens_per_second_per_gpu:5197.058142276293 mfu:24.51540996667094 +Step 7059 | loss:1.2157964706420898 lr:9.111799887723506e-05 tokens_per_second_per_gpu:5363.146644272364 mfu:25.29887776820693 +Step 7060 | loss:1.209162712097168 lr:9.111553155317074e-05 tokens_per_second_per_gpu:4740.781596366359 mfu:22.36307565080753 +Step 7061 | loss:1.1806039810180664 lr:9.111306391987125e-05 tokens_per_second_per_gpu:4515.04857510133 mfu:21.298254475475524 +Step 7062 | loss:1.210463285446167 lr:9.111059597735514e-05 tokens_per_second_per_gpu:3815.130196304119 mfu:17.996620064299623 +Step 7063 | loss:1.2569451332092285 lr:9.110812772564101e-05 tokens_per_second_per_gpu:3040.287185583217 mfu:14.341553380879274 +Step 7064 | loss:1.216559886932373 lr:9.11056591647474e-05 tokens_per_second_per_gpu:5225.3837049151 mfu:24.649026478477523 +Step 7065 | loss:1.329345464706421 lr:9.110319029469288e-05 tokens_per_second_per_gpu:4412.2813975486015 mfu:20.813484165070776 +Step 7066 | loss:1.202030897140503 lr:9.110072111549603e-05 tokens_per_second_per_gpu:4252.2260043715505 mfu:20.058475567188612 +Step 7067 | loss:1.2683777809143066 lr:9.10982516271754e-05 tokens_per_second_per_gpu:5323.42373488383 mfu:25.111498027194006 +Step 7068 | loss:1.227781057357788 lr:9.109578182974958e-05 tokens_per_second_per_gpu:4974.7089276294755 mfu:23.466550784493975 +Step 7069 | loss:1.2470875978469849 lr:9.109331172323715e-05 tokens_per_second_per_gpu:5157.432321655048 mfu:24.32848820993837 +Step 7070 | loss:1.1244860887527466 lr:9.109084130765668e-05 tokens_per_second_per_gpu:4321.257860249086 mfu:20.38411060941207 +Step 7071 | loss:1.3608503341674805 lr:9.108837058302673e-05 tokens_per_second_per_gpu:5348.345761948646 mfu:25.22905948099503 +Step 7072 | loss:1.357700228691101 lr:9.108589954936592e-05 tokens_per_second_per_gpu:5359.896847967543 mfu:25.28354792456466 +Step 7073 | loss:1.3792228698730469 lr:9.10834282066928e-05 tokens_per_second_per_gpu:3442.832074625793 mfu:16.24042630373336 +Step 7074 | loss:1.3605704307556152 lr:9.1080956555026e-05 tokens_per_second_per_gpu:3747.0074578669305 mfu:17.675273484153706 +Step 7075 | loss:1.0378620624542236 lr:9.107848459438407e-05 tokens_per_second_per_gpu:5335.1590889709905 mfu:25.166855694681015 +Step 7076 | loss:1.312902569770813 lr:9.10760123247856e-05 tokens_per_second_per_gpu:5378.739706272791 mfu:25.37243290211366 +Step 7077 | loss:1.3561947345733643 lr:9.107353974624921e-05 tokens_per_second_per_gpu:4317.130435710019 mfu:20.36464084365005 +Step 7078 | loss:1.167182207107544 lr:9.10710668587935e-05 tokens_per_second_per_gpu:5348.307593619954 mfu:25.228879434476397 +Step 7079 | loss:1.2865217924118042 lr:9.106859366243704e-05 tokens_per_second_per_gpu:4268.317504574131 mfu:20.134381919137223 +Step 7080 | loss:1.300189733505249 lr:9.106612015719845e-05 tokens_per_second_per_gpu:4577.281588089751 mfu:21.591818215788816 +Step 7081 | loss:1.163472056388855 lr:9.106364634309635e-05 tokens_per_second_per_gpu:5358.879385175276 mfu:25.278748378976488 +Step 7082 | loss:1.1596990823745728 lr:9.106117222014929e-05 tokens_per_second_per_gpu:4284.533073711678 mfu:20.210873525420418 +Step 7083 | loss:1.1052045822143555 lr:9.105869778837594e-05 tokens_per_second_per_gpu:4233.395831269486 mfu:19.96965042790735 +Step 7084 | loss:1.2766892910003662 lr:9.105622304779488e-05 tokens_per_second_per_gpu:4158.920563587287 mfu:19.61833788345947 +Step 7085 | loss:1.130009412765503 lr:9.105374799842473e-05 tokens_per_second_per_gpu:5313.6045612982825 mfu:25.06517931006721 +Step 7086 | loss:1.3065317869186401 lr:9.10512726402841e-05 tokens_per_second_per_gpu:5309.396038553685 mfu:25.045326990232923 +Step 7087 | loss:1.2387549877166748 lr:9.104879697339159e-05 tokens_per_second_per_gpu:4409.527944863477 mfu:20.800495658967574 +Step 7088 | loss:1.306113839149475 lr:9.104632099776587e-05 tokens_per_second_per_gpu:4173.595480352946 mfu:19.68756196964207 +Step 7089 | loss:1.2287068367004395 lr:9.104384471342553e-05 tokens_per_second_per_gpu:4540.387428381501 mfu:21.41778216965226 +Step 7090 | loss:1.26667320728302 lr:9.104136812038916e-05 tokens_per_second_per_gpu:5202.547476536866 mfu:24.541304092955066 +Step 7091 | loss:1.3249212503433228 lr:9.103889121867546e-05 tokens_per_second_per_gpu:4801.392330781842 mfu:22.648986826303016 +Step 7092 | loss:1.3902740478515625 lr:9.103641400830302e-05 tokens_per_second_per_gpu:4779.365678146543 mfu:22.545083347687846 +Step 7093 | loss:1.2635161876678467 lr:9.103393648929046e-05 tokens_per_second_per_gpu:4736.833876322372 mfu:22.344453581809553 +Step 7094 | loss:1.2495578527450562 lr:9.103145866165644e-05 tokens_per_second_per_gpu:5325.112145548784 mfu:25.119462548373157 +Step 7095 | loss:1.1911258697509766 lr:9.102898052541958e-05 tokens_per_second_per_gpu:3780.415997398678 mfu:17.83286726520958 +Step 7096 | loss:1.3599491119384766 lr:9.102650208059852e-05 tokens_per_second_per_gpu:5355.821220805631 mfu:25.264322495868228 +Step 7097 | loss:1.26059889793396 lr:9.102402332721191e-05 tokens_per_second_per_gpu:3814.60724447556 mfu:17.994153211299526 +Step 7098 | loss:1.1289904117584229 lr:9.102154426527838e-05 tokens_per_second_per_gpu:3893.42933707249 mfu:18.365970470515002 +Step 7099 | loss:1.2007352113723755 lr:9.101906489481658e-05 tokens_per_second_per_gpu:5050.870069835907 mfu:23.82581588671252 +Step 7100 | loss:1.3565278053283691 lr:9.101658521584515e-05 tokens_per_second_per_gpu:4448.253307304538 mfu:20.983169800830204 +Step 7101 | loss:1.2137393951416016 lr:9.101410522838276e-05 tokens_per_second_per_gpu:5327.013050316007 mfu:25.12842944048795 +Step 7102 | loss:1.1738563776016235 lr:9.101162493244806e-05 tokens_per_second_per_gpu:4969.609919608379 mfu:23.44249789367798 +Step 7103 | loss:1.168424129486084 lr:9.100914432805969e-05 tokens_per_second_per_gpu:3622.3248671439283 mfu:17.08712443066972 +Step 7104 | loss:1.3492465019226074 lr:9.10066634152363e-05 tokens_per_second_per_gpu:5413.527702300308 mfu:25.53653381481962 +Step 7105 | loss:1.184008240699768 lr:9.100418219399658e-05 tokens_per_second_per_gpu:5345.481469581872 mfu:25.215548125202286 +Step 7106 | loss:1.3047586679458618 lr:9.100170066435915e-05 tokens_per_second_per_gpu:5313.162185336092 mfu:25.06309254717644 +Step 7107 | loss:1.18309485912323 lr:9.099921882634271e-05 tokens_per_second_per_gpu:3331.65036321517 mfu:15.715963201453407 +Step 7108 | loss:1.4229235649108887 lr:9.099673667996592e-05 tokens_per_second_per_gpu:5165.726467155817 mfu:24.367613109392643 +Step 7109 | loss:1.2604360580444336 lr:9.099425422524744e-05 tokens_per_second_per_gpu:4256.375258635372 mfu:20.078048307486835 +Step 7110 | loss:1.1890254020690918 lr:9.099177146220595e-05 tokens_per_second_per_gpu:3713.579437583485 mfu:17.517587809068317 +Step 7111 | loss:1.2749837636947632 lr:9.09892883908601e-05 tokens_per_second_per_gpu:4792.413834346356 mfu:22.606633726727374 +Step 7112 | loss:1.3536254167556763 lr:9.098680501122859e-05 tokens_per_second_per_gpu:3924.041103305481 mfu:18.51037139474191 +Step 7113 | loss:1.2750450372695923 lr:9.098432132333008e-05 tokens_per_second_per_gpu:4406.395101964904 mfu:20.785717504496887 +Step 7114 | loss:1.2562752962112427 lr:9.098183732718327e-05 tokens_per_second_per_gpu:4830.368168191641 mfu:22.785670795152686 +Step 7115 | loss:1.4150044918060303 lr:9.097935302280683e-05 tokens_per_second_per_gpu:4575.2389262911565 mfu:21.582182631570806 +Step 7116 | loss:1.2312184572219849 lr:9.097686841021944e-05 tokens_per_second_per_gpu:5318.669251068764 mfu:25.089070315839752 +Step 7117 | loss:1.2133698463439941 lr:9.09743834894398e-05 tokens_per_second_per_gpu:5350.52570423198 mfu:25.23934264068202 +Step 7118 | loss:1.2547810077667236 lr:9.09718982604866e-05 tokens_per_second_per_gpu:4501.029259094814 mfu:21.232122969930828 +Step 7119 | loss:1.227682113647461 lr:9.09694127233785e-05 tokens_per_second_per_gpu:4407.759655191567 mfu:20.792154334883843 +Step 7120 | loss:1.236759066581726 lr:9.096692687813426e-05 tokens_per_second_per_gpu:5343.141918501977 mfu:25.20451206358984 +Step 7121 | loss:1.1842573881149292 lr:9.096444072477252e-05 tokens_per_second_per_gpu:5349.456716194306 mfu:25.234300041720147 +Step 7122 | loss:1.3420037031173706 lr:9.096195426331198e-05 tokens_per_second_per_gpu:4259.967454544549 mfu:20.09499330848224 +Step 7123 | loss:1.2701383829116821 lr:9.095946749377135e-05 tokens_per_second_per_gpu:3394.275814052754 mfu:16.011378138058006 +Step 7124 | loss:1.4033859968185425 lr:9.095698041616935e-05 tokens_per_second_per_gpu:5184.98495385703 mfu:24.45845867699892 +Step 7125 | loss:1.2658114433288574 lr:9.095449303052469e-05 tokens_per_second_per_gpu:4554.957993401209 mfu:21.486514010844964 +Step 7126 | loss:1.464775562286377 lr:9.095200533685604e-05 tokens_per_second_per_gpu:5374.463415679227 mfu:25.352260909773406 +Step 7127 | loss:1.1870121955871582 lr:9.094951733518215e-05 tokens_per_second_per_gpu:5102.142204616747 mfu:24.06767529440199 +Step 7128 | loss:1.2107709646224976 lr:9.09470290255217e-05 tokens_per_second_per_gpu:4751.591241090797 mfu:22.414066589288275 +Step 7129 | loss:1.1666386127471924 lr:9.094454040789343e-05 tokens_per_second_per_gpu:3282.417659667553 mfu:15.483724138853896 +Step 7130 | loss:1.2564527988433838 lr:9.094205148231605e-05 tokens_per_second_per_gpu:5366.773154198051 mfu:25.31598463427279 +Step 7131 | loss:1.1669566631317139 lr:9.093956224880827e-05 tokens_per_second_per_gpu:4604.431873629756 mfu:21.719890744997407 +Step 7132 | loss:1.2241052389144897 lr:9.09370727073888e-05 tokens_per_second_per_gpu:5343.101070505554 mfu:25.20431937662133 +Step 7133 | loss:1.3713226318359375 lr:9.093458285807641e-05 tokens_per_second_per_gpu:4199.4008521045025 mfu:19.809290311044588 +Step 7134 | loss:1.2661933898925781 lr:9.093209270088979e-05 tokens_per_second_per_gpu:3656.594487020312 mfu:17.248780074626698 +Step 7135 | loss:1.5419178009033203 lr:9.092960223584768e-05 tokens_per_second_per_gpu:5345.098511006327 mfu:25.213741644261965 +Step 7136 | loss:1.3849163055419922 lr:9.09271114629688e-05 tokens_per_second_per_gpu:4106.810928238529 mfu:19.372527842698798 +Step 7137 | loss:1.3045545816421509 lr:9.092462038227189e-05 tokens_per_second_per_gpu:3634.590024019123 mfu:17.144981268301564 +Step 7138 | loss:1.3394641876220703 lr:9.09221289937757e-05 tokens_per_second_per_gpu:4839.62362718363 mfu:22.829330374362033 +Step 7139 | loss:1.3922019004821777 lr:9.091963729749895e-05 tokens_per_second_per_gpu:4209.419986481536 mfu:19.856552277341653 +Step 7140 | loss:1.2827917337417603 lr:9.091714529346039e-05 tokens_per_second_per_gpu:5204.505592369364 mfu:24.55054085942597 +Step 7141 | loss:1.0650641918182373 lr:9.091465298167876e-05 tokens_per_second_per_gpu:4782.4419678526265 mfu:22.559594731100585 +Step 7142 | loss:1.131142258644104 lr:9.09121603621728e-05 tokens_per_second_per_gpu:5164.994694562989 mfu:24.364161213218992 +Step 7143 | loss:1.182483434677124 lr:9.090966743496127e-05 tokens_per_second_per_gpu:4020.3059098588287 mfu:18.964468911723685 +Step 7144 | loss:1.1423126459121704 lr:9.090717420006289e-05 tokens_per_second_per_gpu:3845.5162334157176 mfu:18.13995618574756 +Step 7145 | loss:1.1934949159622192 lr:9.090468065749645e-05 tokens_per_second_per_gpu:5358.149195256244 mfu:25.275303948545165 +Step 7146 | loss:1.3162349462509155 lr:9.090218680728068e-05 tokens_per_second_per_gpu:4333.016974982197 mfu:20.439580359920015 +Step 7147 | loss:1.204891562461853 lr:9.089969264943436e-05 tokens_per_second_per_gpu:5349.608472574805 mfu:25.235015902459267 +Step 7148 | loss:1.2766205072402954 lr:9.089719818397622e-05 tokens_per_second_per_gpu:4967.150799543245 mfu:23.430897804761596 +Step 7149 | loss:1.2320027351379395 lr:9.089470341092503e-05 tokens_per_second_per_gpu:5024.018071661553 mfu:23.69915042990078 +Step 7150 | loss:1.2610487937927246 lr:9.089220833029957e-05 tokens_per_second_per_gpu:5357.220094907771 mfu:25.270921223680553 +Step 7151 | loss:1.1458520889282227 lr:9.08897129421186e-05 tokens_per_second_per_gpu:3398.102338131565 mfu:16.029428505010088 +Step 7152 | loss:1.1516611576080322 lr:9.088721724640085e-05 tokens_per_second_per_gpu:5361.971129429525 mfu:25.293332664129743 +Step 7153 | loss:1.3033125400543213 lr:9.088472124316513e-05 tokens_per_second_per_gpu:4283.4777862318815 mfu:20.205895554327608 +Step 7154 | loss:1.313075304031372 lr:9.088222493243023e-05 tokens_per_second_per_gpu:3319.8273509532273 mfu:15.660192035399923 +Step 7155 | loss:1.1362879276275635 lr:9.087972831421487e-05 tokens_per_second_per_gpu:5331.267739576932 mfu:25.148499535656573 +Step 7156 | loss:1.4620364904403687 lr:9.087723138853788e-05 tokens_per_second_per_gpu:4220.109941292627 mfu:19.906978618079933 +Step 7157 | loss:1.4192569255828857 lr:9.0874734155418e-05 tokens_per_second_per_gpu:3900.0765133403283 mfu:18.397326335095613 +Step 7158 | loss:1.2997612953186035 lr:9.087223661487405e-05 tokens_per_second_per_gpu:5364.637071013688 mfu:25.30590836543878 +Step 7159 | loss:1.3250679969787598 lr:9.086973876692478e-05 tokens_per_second_per_gpu:5287.416870060765 mfu:24.94164750241898 +Step 7160 | loss:1.137679934501648 lr:9.086724061158897e-05 tokens_per_second_per_gpu:3956.897862937501 mfu:18.665362335867407 +Step 7161 | loss:1.1434531211853027 lr:9.086474214888547e-05 tokens_per_second_per_gpu:4783.565330483188 mfu:22.56489382429437 +Step 7162 | loss:1.167588472366333 lr:9.0862243378833e-05 tokens_per_second_per_gpu:5341.608861605993 mfu:25.19728037264501 +Step 7163 | loss:1.188344955444336 lr:9.08597443014504e-05 tokens_per_second_per_gpu:5346.4401306341415 mfu:25.220070292950282 +Step 7164 | loss:1.3431636095046997 lr:9.085724491675642e-05 tokens_per_second_per_gpu:4987.912527744304 mfu:23.528834419002095 +Step 7165 | loss:1.295849323272705 lr:9.085474522476992e-05 tokens_per_second_per_gpu:5276.580911090108 mfu:24.890532435149154 +Step 7166 | loss:1.1644986867904663 lr:9.085224522550966e-05 tokens_per_second_per_gpu:5327.597996634288 mfu:25.13118873207341 +Step 7167 | loss:1.3215036392211914 lr:9.084974491899445e-05 tokens_per_second_per_gpu:4178.093496811276 mfu:19.70877988071675 +Step 7168 | loss:1.1693358421325684 lr:9.084724430524309e-05 tokens_per_second_per_gpu:5388.204902607369 mfu:25.417081848152883 +Step 7169 | loss:1.1677879095077515 lr:9.084474338427438e-05 tokens_per_second_per_gpu:4656.399546668939 mfu:21.965031125322547 +Step 7170 | loss:1.2328685522079468 lr:9.084224215610717e-05 tokens_per_second_per_gpu:5058.38942417434 mfu:23.861285964061047 +Step 7171 | loss:1.2410874366760254 lr:9.083974062076024e-05 tokens_per_second_per_gpu:5313.841632666641 mfu:25.066297616160302 +Step 7172 | loss:1.357066035270691 lr:9.08372387782524e-05 tokens_per_second_per_gpu:3619.86598059685 mfu:17.07552544329785 +Step 7173 | loss:1.3080984354019165 lr:9.083473662860248e-05 tokens_per_second_per_gpu:4075.880758063462 mfu:19.226624952751802 +Step 7174 | loss:1.331131935119629 lr:9.08322341718293e-05 tokens_per_second_per_gpu:4278.470652707402 mfu:20.182276051187564 +Step 7175 | loss:1.2939023971557617 lr:9.082973140795167e-05 tokens_per_second_per_gpu:4325.646254243857 mfu:20.404811412622628 +Step 7176 | loss:1.3841720819473267 lr:9.082722833698843e-05 tokens_per_second_per_gpu:4340.597905790726 mfu:20.475340903984016 +Step 7177 | loss:1.199861764907837 lr:9.082472495895838e-05 tokens_per_second_per_gpu:5342.60305637899 mfu:25.201970159016355 +Step 7178 | loss:1.2270958423614502 lr:9.082222127388038e-05 tokens_per_second_per_gpu:4327.073058620837 mfu:20.411541892306776 +Step 7179 | loss:1.142789363861084 lr:9.081971728177323e-05 tokens_per_second_per_gpu:5338.470046560713 mfu:25.182474046539355 +Step 7180 | loss:1.2891159057617188 lr:9.081721298265578e-05 tokens_per_second_per_gpu:3923.1616987125203 mfu:18.50622309323502 +Step 7181 | loss:1.4294098615646362 lr:9.081470837654688e-05 tokens_per_second_per_gpu:4256.615476065348 mfu:20.07918145408935 +Step 7182 | loss:1.1567686796188354 lr:9.081220346346532e-05 tokens_per_second_per_gpu:5343.454577037951 mfu:25.20598692724908 +Step 7183 | loss:1.2569245100021362 lr:9.080969824343001e-05 tokens_per_second_per_gpu:4598.329632777428 mfu:21.691105433747325 +Step 7184 | loss:1.3219255208969116 lr:9.080719271645972e-05 tokens_per_second_per_gpu:4665.712526952148 mfu:22.00896204227673 +Step 7185 | loss:1.3296140432357788 lr:9.080468688257334e-05 tokens_per_second_per_gpu:4370.347158735025 mfu:20.615673205868223 +Step 7186 | loss:1.2278289794921875 lr:9.08021807417897e-05 tokens_per_second_per_gpu:5236.632207614779 mfu:24.702087584904373 +Step 7187 | loss:1.1779003143310547 lr:9.079967429412765e-05 tokens_per_second_per_gpu:5332.619421933138 mfu:25.15487565195159 +Step 7188 | loss:1.4218995571136475 lr:9.079716753960605e-05 tokens_per_second_per_gpu:5255.568715155351 mfu:24.791414321874598 +Step 7189 | loss:1.2047792673110962 lr:9.079466047824374e-05 tokens_per_second_per_gpu:4068.1083388722864 mfu:19.189961125315566 +Step 7190 | loss:1.3024107217788696 lr:9.079215311005959e-05 tokens_per_second_per_gpu:5302.702336126038 mfu:25.01375165381831 +Step 7191 | loss:1.1941723823547363 lr:9.078964543507245e-05 tokens_per_second_per_gpu:5327.888302523912 mfu:25.13255815448606 +Step 7192 | loss:1.3302273750305176 lr:9.078713745330118e-05 tokens_per_second_per_gpu:3928.9129473104554 mfu:18.533352714135674 +Step 7193 | loss:1.2275562286376953 lr:9.078462916476464e-05 tokens_per_second_per_gpu:4655.796775329939 mfu:21.962187750073088 +Step 7194 | loss:1.2979187965393066 lr:9.078212056948173e-05 tokens_per_second_per_gpu:4918.895008568654 mfu:23.203267005447522 +Step 7195 | loss:1.2060291767120361 lr:9.077961166747124e-05 tokens_per_second_per_gpu:5338.176260320998 mfu:25.181088206721515 +Step 7196 | loss:1.29954195022583 lr:9.077710245875212e-05 tokens_per_second_per_gpu:5332.428794730586 mfu:25.153976430912792 +Step 7197 | loss:1.247435212135315 lr:9.07745929433432e-05 tokens_per_second_per_gpu:4811.851697537855 mfu:22.698325443843114 +Step 7198 | loss:1.2482680082321167 lr:9.077208312126335e-05 tokens_per_second_per_gpu:5171.750967605715 mfu:24.396031705900583 +Step 7199 | loss:1.328840970993042 lr:9.076957299253149e-05 tokens_per_second_per_gpu:5336.990516276643 mfu:25.1754948497555 +Step 7200 | loss:1.1653908491134644 lr:9.076706255716644e-05 tokens_per_second_per_gpu:3907.0781220615977 mfu:18.43035411803147 +Step 7201 | loss:1.3083430528640747 lr:9.076455181518713e-05 tokens_per_second_per_gpu:4918.406463882568 mfu:23.200962456808934 +Step 7202 | loss:1.3633663654327393 lr:9.076204076661241e-05 tokens_per_second_per_gpu:4488.664554825571 mfu:21.17379655025672 +Step 7203 | loss:1.2287517786026 lr:9.075952941146119e-05 tokens_per_second_per_gpu:3235.6654102328225 mfu:15.263185801513982 +Step 7204 | loss:1.2055493593215942 lr:9.075701774975234e-05 tokens_per_second_per_gpu:4221.358320050828 mfu:19.912867433677746 +Step 7205 | loss:1.1549233198165894 lr:9.075450578150478e-05 tokens_per_second_per_gpu:4832.752859230026 mfu:22.796919789650993 +Step 7206 | loss:1.2526147365570068 lr:9.075199350673735e-05 tokens_per_second_per_gpu:5213.771099573812 mfu:24.59424783776789 +Step 7207 | loss:1.1739048957824707 lr:9.0749480925469e-05 tokens_per_second_per_gpu:4305.268822480892 mfu:20.30868759024835 +Step 7208 | loss:1.2020962238311768 lr:9.074696803771858e-05 tokens_per_second_per_gpu:5347.563351810307 mfu:25.225368718879547 +Step 7209 | loss:1.3872380256652832 lr:9.074445484350505e-05 tokens_per_second_per_gpu:5058.85782435788 mfu:23.863495487644116 +Step 7210 | loss:1.4020296335220337 lr:9.074194134284726e-05 tokens_per_second_per_gpu:4093.1878342724813 mfu:19.30826538412144 +Step 7211 | loss:1.2429536581039429 lr:9.073942753576412e-05 tokens_per_second_per_gpu:4334.184027976769 mfu:20.445085548015186 +Step 7212 | loss:1.2985543012619019 lr:9.073691342227455e-05 tokens_per_second_per_gpu:5273.285355780696 mfu:24.874986738474938 +Step 7213 | loss:1.1665111780166626 lr:9.073439900239746e-05 tokens_per_second_per_gpu:4522.9434233366355 mfu:21.33549582160147 +Step 7214 | loss:1.264078140258789 lr:9.073188427615176e-05 tokens_per_second_per_gpu:4087.474679952252 mfu:19.281315460428274 +Step 7215 | loss:1.2166333198547363 lr:9.072936924355636e-05 tokens_per_second_per_gpu:5157.565946307977 mfu:24.329118540224954 +Step 7216 | loss:1.428252935409546 lr:9.072685390463018e-05 tokens_per_second_per_gpu:5355.047771799556 mfu:25.260674004942565 +Step 7217 | loss:1.3491744995117188 lr:9.072433825939213e-05 tokens_per_second_per_gpu:4771.296741563957 mfu:22.507020797125243 +Step 7218 | loss:1.2320750951766968 lr:9.072182230786113e-05 tokens_per_second_per_gpu:4392.976810382803 mfu:20.722421133707495 +Step 7219 | loss:1.0805426836013794 lr:9.071930605005612e-05 tokens_per_second_per_gpu:4859.203052136589 mfu:22.92168986245951 +Step 7220 | loss:1.3434251546859741 lr:9.0716789485996e-05 tokens_per_second_per_gpu:5351.984202034561 mfu:25.24622262366218 +Step 7221 | loss:1.3084263801574707 lr:9.071427261569972e-05 tokens_per_second_per_gpu:5004.330128316219 mfu:23.60627904203132 +Step 7222 | loss:1.1533806324005127 lr:9.07117554391862e-05 tokens_per_second_per_gpu:4385.296785159646 mfu:20.686193144382443 +Step 7223 | loss:1.1752426624298096 lr:9.070923795647436e-05 tokens_per_second_per_gpu:4705.607095295918 mfu:22.197151527869536 +Step 7224 | loss:1.1640565395355225 lr:9.070672016758317e-05 tokens_per_second_per_gpu:5359.92889139195 mfu:25.283699078864764 +Step 7225 | loss:1.162947177886963 lr:9.070420207253154e-05 tokens_per_second_per_gpu:4442.853304217948 mfu:20.95769706493523 +Step 7226 | loss:1.2154138088226318 lr:9.07016836713384e-05 tokens_per_second_per_gpu:4730.5198950137365 mfu:22.31466945470042 +Step 7227 | loss:1.2244112491607666 lr:9.069916496402271e-05 tokens_per_second_per_gpu:5327.828949270515 mfu:25.132278175063078 +Step 7228 | loss:1.4281282424926758 lr:9.06966459506034e-05 tokens_per_second_per_gpu:4648.671967379065 mfu:21.928578815351226 +Step 7229 | loss:1.1600730419158936 lr:9.069412663109943e-05 tokens_per_second_per_gpu:4723.806091421864 mfu:22.282999297664546 +Step 7230 | loss:1.3924839496612549 lr:9.069160700552975e-05 tokens_per_second_per_gpu:3863.9749188332894 mfu:18.227028954238733 +Step 7231 | loss:1.257385015487671 lr:9.068908707391328e-05 tokens_per_second_per_gpu:4782.6293741062545 mfu:22.56047875837375 +Step 7232 | loss:1.3511050939559937 lr:9.0686566836269e-05 tokens_per_second_per_gpu:5370.812689396042 mfu:25.335039811017488 +Step 7233 | loss:1.1817234754562378 lr:9.068404629261586e-05 tokens_per_second_per_gpu:4144.64571771651 mfu:19.551000999947167 +Step 7234 | loss:1.235736608505249 lr:9.068152544297282e-05 tokens_per_second_per_gpu:4300.380157266464 mfu:20.285626922339603 +Step 7235 | loss:1.273750901222229 lr:9.067900428735884e-05 tokens_per_second_per_gpu:5312.97874233123 mfu:25.0622272152239 +Step 7236 | loss:1.2766337394714355 lr:9.067648282579288e-05 tokens_per_second_per_gpu:3612.485332603618 mfu:17.040709667445373 +Step 7237 | loss:1.1295021772384644 lr:9.067396105829389e-05 tokens_per_second_per_gpu:4460.965811703659 mfu:21.043136853057767 +Step 7238 | loss:1.0506523847579956 lr:9.067143898488086e-05 tokens_per_second_per_gpu:4993.892751458651 mfu:23.557044154594056 +Step 7239 | loss:1.2755991220474243 lr:9.066891660557274e-05 tokens_per_second_per_gpu:3378.9117784667496 mfu:15.938903360824046 +Step 7240 | loss:1.2869855165481567 lr:9.066639392038851e-05 tokens_per_second_per_gpu:5189.5157211023925 mfu:24.479831079123898 +Step 7241 | loss:1.1639597415924072 lr:9.066387092934714e-05 tokens_per_second_per_gpu:3334.647566577877 mfu:15.730101521090923 +Step 7242 | loss:1.1577836275100708 lr:9.066134763246762e-05 tokens_per_second_per_gpu:5334.92707436163 mfu:25.16576124218289 +Step 7243 | loss:1.090371012687683 lr:9.065882402976891e-05 tokens_per_second_per_gpu:5355.9221337702365 mfu:25.2647985195398 +Step 7244 | loss:1.2546132802963257 lr:9.065630012126999e-05 tokens_per_second_per_gpu:5148.9229397761455 mfu:24.288348003765325 +Step 7245 | loss:1.231096625328064 lr:9.065377590698985e-05 tokens_per_second_per_gpu:4456.867615180631 mfu:21.023804960834216 +Step 7246 | loss:1.195662021636963 lr:9.065125138694748e-05 tokens_per_second_per_gpu:5355.0140960048175 mfu:25.260515150473097 +Step 7247 | loss:1.2083064317703247 lr:9.064872656116185e-05 tokens_per_second_per_gpu:5383.248434861397 mfu:25.39370134413453 +Step 7248 | loss:1.3382749557495117 lr:9.064620142965198e-05 tokens_per_second_per_gpu:4844.808117781632 mfu:22.853786501078122 +Step 7249 | loss:1.2643868923187256 lr:9.064367599243682e-05 tokens_per_second_per_gpu:4319.189390567973 mfu:20.374353285008944 +Step 7250 | loss:1.2211147546768188 lr:9.064115024953542e-05 tokens_per_second_per_gpu:4553.121068575733 mfu:21.477848923031498 +Step 7251 | loss:1.3089091777801514 lr:9.063862420096672e-05 tokens_per_second_per_gpu:4105.777649257954 mfu:19.36765368945185 +Step 7252 | loss:1.1423288583755493 lr:9.063609784674974e-05 tokens_per_second_per_gpu:4644.5879394968615 mfu:21.909313759023867 +Step 7253 | loss:1.0320019721984863 lr:9.063357118690348e-05 tokens_per_second_per_gpu:5344.76924643379 mfu:25.21218844709498 +Step 7254 | loss:1.2311509847640991 lr:9.063104422144695e-05 tokens_per_second_per_gpu:4321.766691154523 mfu:20.38651085160852 +Step 7255 | loss:1.385097622871399 lr:9.062851695039915e-05 tokens_per_second_per_gpu:4661.541201937624 mfu:21.98928519048178 +Step 7256 | loss:1.0893174409866333 lr:9.06259893737791e-05 tokens_per_second_per_gpu:4175.32737424817 mfu:19.69573160863741 +Step 7257 | loss:1.2509651184082031 lr:9.06234614916058e-05 tokens_per_second_per_gpu:4913.991544860192 mfu:23.180136530517533 +Step 7258 | loss:1.2930488586425781 lr:9.062093330389825e-05 tokens_per_second_per_gpu:3581.74043743023 mfu:16.895680751292474 +Step 7259 | loss:1.2005809545516968 lr:9.061840481067549e-05 tokens_per_second_per_gpu:4413.024578050017 mfu:20.81698987429562 +Step 7260 | loss:1.2186822891235352 lr:9.061587601195651e-05 tokens_per_second_per_gpu:5285.970946183873 mfu:24.93482683279163 +Step 7261 | loss:1.3088637590408325 lr:9.061334690776035e-05 tokens_per_second_per_gpu:4310.788188991093 mfu:20.334723383778005 +Step 7262 | loss:1.0637407302856445 lr:9.061081749810604e-05 tokens_per_second_per_gpu:203.66235744071696 mfu:0.9607100884291808 +Step 7263 | loss:1.270336389541626 lr:9.060828778301257e-05 tokens_per_second_per_gpu:3369.1756651098713 mfu:15.89297645296742 +Step 7264 | loss:1.076271414756775 lr:9.060575776249899e-05 tokens_per_second_per_gpu:5344.282198155142 mfu:25.209890957265593 +Step 7265 | loss:1.1114356517791748 lr:9.060322743658434e-05 tokens_per_second_per_gpu:4057.9725017230658 mfu:19.14214865213055 +Step 7266 | loss:1.3238672018051147 lr:9.060069680528762e-05 tokens_per_second_per_gpu:4261.184709994529 mfu:20.100735310124996 +Step 7267 | loss:1.0934031009674072 lr:9.059816586862788e-05 tokens_per_second_per_gpu:4317.071981223091 mfu:20.36436510386185 +Step 7268 | loss:1.1482107639312744 lr:9.059563462662416e-05 tokens_per_second_per_gpu:4801.742481109375 mfu:22.650638545139795 +Step 7269 | loss:1.0639268159866333 lr:9.05931030792955e-05 tokens_per_second_per_gpu:4391.328764675761 mfu:20.714647020921788 +Step 7270 | loss:1.0467530488967896 lr:9.05905712266609e-05 tokens_per_second_per_gpu:5368.8567144062945 mfu:25.3258131432671 +Step 7271 | loss:1.251394510269165 lr:9.058803906873946e-05 tokens_per_second_per_gpu:3994.2862420390215 mfu:18.841729699204546 +Step 7272 | loss:1.012963891029358 lr:9.05855066055502e-05 tokens_per_second_per_gpu:5090.248259834069 mfu:24.01156952754586 +Step 7273 | loss:1.102392554283142 lr:9.058297383711217e-05 tokens_per_second_per_gpu:4843.799871194943 mfu:22.849030429904104 +Step 7274 | loss:0.8678928017616272 lr:9.058044076344438e-05 tokens_per_second_per_gpu:5315.166822375711 mfu:25.072548762118664 +Step 7275 | loss:0.9866202473640442 lr:9.057790738456595e-05 tokens_per_second_per_gpu:4017.0336412746483 mfu:18.949033062505514 +Step 7276 | loss:0.8706361055374146 lr:9.057537370049589e-05 tokens_per_second_per_gpu:5307.116580053923 mfu:25.03457439557382 +Step 7277 | loss:1.1293072700500488 lr:9.057283971125328e-05 tokens_per_second_per_gpu:5333.937180589716 mfu:25.16109174436748 +Step 7278 | loss:0.9579396843910217 lr:9.057030541685716e-05 tokens_per_second_per_gpu:5011.2855564596875 mfu:23.639088983302475 +Step 7279 | loss:1.2683560848236084 lr:9.05677708173266e-05 tokens_per_second_per_gpu:5333.7457427758745 mfu:25.160188699536928 +Step 7280 | loss:1.067590594291687 lr:9.056523591268064e-05 tokens_per_second_per_gpu:3826.42714059463 mfu:18.04990967797541 +Step 7281 | loss:1.1044514179229736 lr:9.056270070293837e-05 tokens_per_second_per_gpu:4594.006645992499 mfu:21.670713167505212 +Step 7282 | loss:1.0610108375549316 lr:9.056016518811887e-05 tokens_per_second_per_gpu:5136.246035233706 mfu:24.228548882135694 +Step 7283 | loss:1.0701020956039429 lr:9.055762936824116e-05 tokens_per_second_per_gpu:3796.8568412123636 mfu:17.910421530575938 +Step 7284 | loss:1.177464246749878 lr:9.055509324332437e-05 tokens_per_second_per_gpu:3422.063069097312 mfu:16.142455361097774 +Step 7285 | loss:1.130411982536316 lr:9.055255681338755e-05 tokens_per_second_per_gpu:4745.349461988727 mfu:22.384623052306484 +Step 7286 | loss:1.0514124631881714 lr:9.055002007844977e-05 tokens_per_second_per_gpu:5331.153707195743 mfu:25.14796162545857 +Step 7287 | loss:0.951205313205719 lr:9.054748303853011e-05 tokens_per_second_per_gpu:5352.241573823475 mfu:25.247436690302642 +Step 7288 | loss:1.152618169784546 lr:9.054494569364766e-05 tokens_per_second_per_gpu:4462.571814468546 mfu:21.050712642112064 +Step 7289 | loss:0.9873129725456238 lr:9.05424080438215e-05 tokens_per_second_per_gpu:4905.4252021419325 mfu:23.139727630346723 +Step 7290 | loss:1.0793946981430054 lr:9.05398700890707e-05 tokens_per_second_per_gpu:5340.558348837888 mfu:25.1923249246819 +Step 7291 | loss:1.0110760927200317 lr:9.05373318294144e-05 tokens_per_second_per_gpu:5310.989596900002 mfu:25.05284407684159 +Step 7292 | loss:0.9782418608665466 lr:9.053479326487162e-05 tokens_per_second_per_gpu:3901.5253349641016 mfu:18.40416067388941 +Step 7293 | loss:1.0900591611862183 lr:9.053225439546152e-05 tokens_per_second_per_gpu:4233.723122114924 mfu:19.97119431466686 +Step 7294 | loss:1.0945069789886475 lr:9.052971522120313e-05 tokens_per_second_per_gpu:4648.576794206891 mfu:21.928129867259862 +Step 7295 | loss:1.0966514348983765 lr:9.05271757421156e-05 tokens_per_second_per_gpu:4190.056268719066 mfu:19.765210316865154 +Step 7296 | loss:1.0680525302886963 lr:9.0524635958218e-05 tokens_per_second_per_gpu:3505.586023200667 mfu:16.536447386088735 +Step 7297 | loss:1.0934010744094849 lr:9.052209586952945e-05 tokens_per_second_per_gpu:4198.690959799465 mfu:19.80594162792161 +Step 7298 | loss:1.1050633192062378 lr:9.051955547606904e-05 tokens_per_second_per_gpu:5330.282911915271 mfu:25.143853935547593 +Step 7299 | loss:1.3398637771606445 lr:9.051701477785587e-05 tokens_per_second_per_gpu:3745.8569256685037 mfu:17.669846227472693 +Step 7300 | loss:1.0389333963394165 lr:9.05144737749091e-05 tokens_per_second_per_gpu:5336.447550443154 mfu:25.172933587279406 +Step 7301 | loss:1.0629026889801025 lr:9.051193246724777e-05 tokens_per_second_per_gpu:5345.85801971854 mfu:25.217324376443024 +Step 7302 | loss:0.9993621706962585 lr:9.050939085489104e-05 tokens_per_second_per_gpu:5323.1992944712565 mfu:25.110439303474415 +Step 7303 | loss:1.2277791500091553 lr:9.050684893785802e-05 tokens_per_second_per_gpu:3407.820757003664 mfu:16.07527194496302 +Step 7304 | loss:1.3177061080932617 lr:9.05043067161678e-05 tokens_per_second_per_gpu:2955.629622723119 mfu:13.942209212798543 +Step 7305 | loss:1.0970361232757568 lr:9.050176418983953e-05 tokens_per_second_per_gpu:4517.98961342636 mfu:21.31212785504757 +Step 7306 | loss:1.0755738019943237 lr:9.049922135889233e-05 tokens_per_second_per_gpu:5078.7274892256555 mfu:23.957224086939817 +Step 7307 | loss:1.0001492500305176 lr:9.04966782233453e-05 tokens_per_second_per_gpu:4058.3850388438123 mfu:19.144094660115204 +Step 7308 | loss:1.1770870685577393 lr:9.04941347832176e-05 tokens_per_second_per_gpu:3546.2517900513494 mfu:16.72827474661868 +Step 7309 | loss:1.097977876663208 lr:9.049159103852835e-05 tokens_per_second_per_gpu:3955.4629543762126 mfu:18.658593627363164 +Step 7310 | loss:1.0632442235946655 lr:9.048904698929668e-05 tokens_per_second_per_gpu:4048.494095099397 mfu:19.097437390903778 +Step 7311 | loss:1.0299431085586548 lr:9.048650263554171e-05 tokens_per_second_per_gpu:5377.300121660495 mfu:25.365642135878996 +Step 7312 | loss:0.9073380827903748 lr:9.048395797728258e-05 tokens_per_second_per_gpu:3521.970772056098 mfu:16.61373704196622 +Step 7313 | loss:1.0788099765777588 lr:9.048141301453845e-05 tokens_per_second_per_gpu:5301.808571738898 mfu:25.00953561471224 +Step 7314 | loss:1.1006278991699219 lr:9.047886774732843e-05 tokens_per_second_per_gpu:4138.056230708007 mfu:19.519917265445642 +Step 7315 | loss:1.0889933109283447 lr:9.04763221756717e-05 tokens_per_second_per_gpu:5278.004161582726 mfu:24.897246150554402 +Step 7316 | loss:1.2007497549057007 lr:9.047377629958737e-05 tokens_per_second_per_gpu:4388.334726355003 mfu:20.700523631327158 +Step 7317 | loss:1.2291743755340576 lr:9.047123011909463e-05 tokens_per_second_per_gpu:3859.047882618038 mfu:18.20378728377233 +Step 7318 | loss:1.0145578384399414 lr:9.04686836342126e-05 tokens_per_second_per_gpu:4286.732968975362 mfu:20.221250806720054 +Step 7319 | loss:1.1057769060134888 lr:9.046613684496041e-05 tokens_per_second_per_gpu:4280.556139800486 mfu:20.192113649626428 +Step 7320 | loss:1.13406503200531 lr:9.046358975135725e-05 tokens_per_second_per_gpu:4770.9672491440915 mfu:22.505466525163534 +Step 7321 | loss:1.0837628841400146 lr:9.046104235342227e-05 tokens_per_second_per_gpu:4514.2632806460315 mfu:21.29455010754483 +Step 7322 | loss:1.1909680366516113 lr:9.045849465117464e-05 tokens_per_second_per_gpu:3648.108571576984 mfu:17.208750563634048 +Step 7323 | loss:1.060235857963562 lr:9.04559466446335e-05 tokens_per_second_per_gpu:4344.941342315334 mfu:20.495829635137277 +Step 7324 | loss:1.119115948677063 lr:9.045339833381803e-05 tokens_per_second_per_gpu:4941.927164426561 mfu:23.311913614320275 +Step 7325 | loss:1.0644665956497192 lr:9.045084971874738e-05 tokens_per_second_per_gpu:5329.077713279289 mfu:25.138168807953573 +Step 7326 | loss:1.1037777662277222 lr:9.044830079944073e-05 tokens_per_second_per_gpu:3805.1072909097147 mfu:17.94934030947022 +Step 7327 | loss:1.0201407670974731 lr:9.044575157591724e-05 tokens_per_second_per_gpu:5101.105391618053 mfu:24.06278447058861 +Step 7328 | loss:0.9486326575279236 lr:9.04432020481961e-05 tokens_per_second_per_gpu:4788.363886315704 mfu:22.587529430874397 +Step 7329 | loss:0.9584658741950989 lr:9.044065221629648e-05 tokens_per_second_per_gpu:5009.847935221819 mfu:23.632307478639976 +Step 7330 | loss:1.000154972076416 lr:9.043810208023755e-05 tokens_per_second_per_gpu:4999.263386903272 mfu:23.582378358311225 +Step 7331 | loss:1.0125982761383057 lr:9.04355516400385e-05 tokens_per_second_per_gpu:4473.360590539311 mfu:21.101605139592873 +Step 7332 | loss:1.2185156345367432 lr:9.04330008957185e-05 tokens_per_second_per_gpu:5317.311896876365 mfu:25.082667444525697 +Step 7333 | loss:1.2314772605895996 lr:9.043044984729674e-05 tokens_per_second_per_gpu:3617.3335055216417 mfu:17.063579326283286 +Step 7334 | loss:0.9559177756309509 lr:9.042789849479242e-05 tokens_per_second_per_gpu:4015.3368111655527 mfu:18.94102882537184 +Step 7335 | loss:1.0581319332122803 lr:9.04253468382247e-05 tokens_per_second_per_gpu:4923.15691152911 mfu:23.223371128867623 +Step 7336 | loss:1.085597276687622 lr:9.042279487761281e-05 tokens_per_second_per_gpu:4902.59901060839 mfu:23.126396002685002 +Step 7337 | loss:1.08017098903656 lr:9.04202426129759e-05 tokens_per_second_per_gpu:4513.400685376989 mfu:21.290481098486826 +Step 7338 | loss:0.8485097289085388 lr:9.041769004433321e-05 tokens_per_second_per_gpu:5333.277033097979 mfu:25.157977716016177 +Step 7339 | loss:1.2138817310333252 lr:9.04151371717039e-05 tokens_per_second_per_gpu:5213.242271692963 mfu:24.591753266426583 +Step 7340 | loss:1.0723823308944702 lr:9.041258399510717e-05 tokens_per_second_per_gpu:5060.300935637519 mfu:23.870302889770347 +Step 7341 | loss:1.3349099159240723 lr:9.041003051456225e-05 tokens_per_second_per_gpu:3961.048682150956 mfu:18.68494245830006 +Step 7342 | loss:1.0400654077529907 lr:9.040747673008835e-05 tokens_per_second_per_gpu:3857.7244693021416 mfu:18.197544517363653 +Step 7343 | loss:1.0693423748016357 lr:9.040492264170467e-05 tokens_per_second_per_gpu:3691.2896842804566 mfu:17.412443239713866 +Step 7344 | loss:1.1590933799743652 lr:9.040236824943039e-05 tokens_per_second_per_gpu:4053.3883017299954 mfu:19.120524198618067 +Step 7345 | loss:1.004236102104187 lr:9.039981355328473e-05 tokens_per_second_per_gpu:4808.475586743206 mfu:22.682399753200855 +Step 7346 | loss:1.218082308769226 lr:9.039725855328693e-05 tokens_per_second_per_gpu:3702.988779009352 mfu:17.46762986562145 +Step 7347 | loss:1.253673791885376 lr:9.03947032494562e-05 tokens_per_second_per_gpu:3565.5693930943607 mfu:16.81939917609551 +Step 7348 | loss:1.0352901220321655 lr:9.039214764181175e-05 tokens_per_second_per_gpu:5138.138265702218 mfu:24.237474856103475 +Step 7349 | loss:1.1885703802108765 lr:9.038959173037281e-05 tokens_per_second_per_gpu:5178.233477376215 mfu:24.426610810518245 +Step 7350 | loss:1.128467082977295 lr:9.038703551515859e-05 tokens_per_second_per_gpu:5349.612181596169 mfu:25.235033398545866 +Step 7351 | loss:0.9353082776069641 lr:9.038447899618832e-05 tokens_per_second_per_gpu:4390.443505296583 mfu:20.71047110138938 +Step 7352 | loss:1.0407627820968628 lr:9.038192217348124e-05 tokens_per_second_per_gpu:5273.8194061364175 mfu:24.87750594512126 +Step 7353 | loss:1.1394827365875244 lr:9.037936504705656e-05 tokens_per_second_per_gpu:3958.2888068335446 mfu:18.671923655544546 +Step 7354 | loss:1.162331461906433 lr:9.037680761693352e-05 tokens_per_second_per_gpu:5135.746942329003 mfu:24.22619457575018 +Step 7355 | loss:1.0989634990692139 lr:9.037424988313137e-05 tokens_per_second_per_gpu:5078.283797022346 mfu:23.955131115115112 +Step 7356 | loss:0.9952713847160339 lr:9.037169184566934e-05 tokens_per_second_per_gpu:4576.650517261487 mfu:21.588841347020114 +Step 7357 | loss:1.1888213157653809 lr:9.036913350456664e-05 tokens_per_second_per_gpu:5304.134309670892 mfu:25.02050651734927 +Step 7358 | loss:1.049747347831726 lr:9.036657485984254e-05 tokens_per_second_per_gpu:5359.228808116228 mfu:25.280396666606656 +Step 7359 | loss:1.1562391519546509 lr:9.03640159115163e-05 tokens_per_second_per_gpu:4352.3081819113295 mfu:20.530580274424036 +Step 7360 | loss:1.0345338582992554 lr:9.036145665960714e-05 tokens_per_second_per_gpu:5373.1595581225065 mfu:25.346110391217316 +Step 7361 | loss:1.2108725309371948 lr:9.03588971041343e-05 tokens_per_second_per_gpu:4815.984117238986 mfu:22.717818772636853 +Step 7362 | loss:1.234182357788086 lr:9.035633724511706e-05 tokens_per_second_per_gpu:4383.022498455227 mfu:20.675464945964375 +Step 7363 | loss:0.9141492247581482 lr:9.035377708257466e-05 tokens_per_second_per_gpu:5167.395502929353 mfu:24.37548623590328 +Step 7364 | loss:1.0115373134613037 lr:9.035121661652634e-05 tokens_per_second_per_gpu:3702.1467742121345 mfu:17.463657985332144 +Step 7365 | loss:1.0424748659133911 lr:9.034865584699138e-05 tokens_per_second_per_gpu:4160.3791625622025 mfu:19.625218343686793 +Step 7366 | loss:1.1233572959899902 lr:9.034609477398902e-05 tokens_per_second_per_gpu:3883.3769345483815 mfu:18.31855157269205 +Step 7367 | loss:1.0347970724105835 lr:9.034353339753854e-05 tokens_per_second_per_gpu:4670.638832140611 mfu:22.032200264365045 +Step 7368 | loss:1.0554202795028687 lr:9.034097171765921e-05 tokens_per_second_per_gpu:5369.9423385328255 mfu:25.330934217498527 +Step 7369 | loss:0.9155645370483398 lr:9.033840973437027e-05 tokens_per_second_per_gpu:5203.287155468015 mfu:24.54479328467553 +Step 7370 | loss:1.057682752609253 lr:9.0335847447691e-05 tokens_per_second_per_gpu:4285.689253067474 mfu:20.216327420706246 +Step 7371 | loss:1.0635603666305542 lr:9.033328485764068e-05 tokens_per_second_per_gpu:4616.627049131603 mfu:21.777417468551114 +Step 7372 | loss:0.9474616646766663 lr:9.033072196423858e-05 tokens_per_second_per_gpu:5326.912326953516 mfu:25.127954311201897 +Step 7373 | loss:0.9572389125823975 lr:9.032815876750397e-05 tokens_per_second_per_gpu:5275.121910872216 mfu:24.88365008219025 +Step 7374 | loss:1.1547831296920776 lr:9.032559526745614e-05 tokens_per_second_per_gpu:4383.216407346184 mfu:20.676379647287284 +Step 7375 | loss:1.0634710788726807 lr:9.032303146411434e-05 tokens_per_second_per_gpu:3770.492033122478 mfu:17.78605423251584 +Step 7376 | loss:0.9824548959732056 lr:9.032046735749789e-05 tokens_per_second_per_gpu:5300.393805217358 mfu:25.00286191964608 +Step 7377 | loss:1.025357723236084 lr:9.031790294762606e-05 tokens_per_second_per_gpu:5236.910791183324 mfu:24.703401711128457 +Step 7378 | loss:1.2079555988311768 lr:9.031533823451814e-05 tokens_per_second_per_gpu:3774.3351515187237 mfu:17.804182877694544 +Step 7379 | loss:1.017706036567688 lr:9.031277321819342e-05 tokens_per_second_per_gpu:5343.334690330995 mfu:25.205421401197587 +Step 7380 | loss:1.147085428237915 lr:9.03102078986712e-05 tokens_per_second_per_gpu:3048.1338880017415 mfu:14.378567614972939 +Step 7381 | loss:1.1163541078567505 lr:9.030764227597073e-05 tokens_per_second_per_gpu:5028.662788860044 mfu:23.72106035339653 +Step 7382 | loss:0.9102001786231995 lr:9.030507635011136e-05 tokens_per_second_per_gpu:4421.188721912459 mfu:20.855501533841338 +Step 7383 | loss:1.0330193042755127 lr:9.030251012111236e-05 tokens_per_second_per_gpu:5388.243723385281 mfu:25.4172649723858 +Step 7384 | loss:0.9309800267219543 lr:9.029994358899306e-05 tokens_per_second_per_gpu:5316.381345155367 mfu:25.07827787328926 +Step 7385 | loss:1.2708539962768555 lr:9.029737675377272e-05 tokens_per_second_per_gpu:3389.3523348548365 mfu:15.988153246649258 +Step 7386 | loss:1.0880035161972046 lr:9.029480961547067e-05 tokens_per_second_per_gpu:5052.649877923658 mfu:23.834211545129012 +Step 7387 | loss:0.9999893307685852 lr:9.029224217410622e-05 tokens_per_second_per_gpu:4895.405168347489 mfu:23.09246142134428 +Step 7388 | loss:0.9300323128700256 lr:9.028967442969867e-05 tokens_per_second_per_gpu:3801.3168967803053 mfu:17.93146037365387 +Step 7389 | loss:1.1065983772277832 lr:9.028710638226733e-05 tokens_per_second_per_gpu:3810.2400262517835 mfu:17.973552297813708 +Step 7390 | loss:1.2057344913482666 lr:9.028453803183154e-05 tokens_per_second_per_gpu:4030.547069899666 mfu:19.012778210958395 +Step 7391 | loss:1.0149592161178589 lr:9.028196937841058e-05 tokens_per_second_per_gpu:3438.5467398179076 mfu:16.220211648291052 +Step 7392 | loss:1.1439599990844727 lr:9.02794004220238e-05 tokens_per_second_per_gpu:4083.9126650127996 mfu:19.26451282821646 +Step 7393 | loss:0.9391782879829407 lr:9.027683116269053e-05 tokens_per_second_per_gpu:4489.068783199495 mfu:21.17570336446499 +Step 7394 | loss:1.0299538373947144 lr:9.027426160043005e-05 tokens_per_second_per_gpu:5345.400808339935 mfu:25.21516763236184 +Step 7395 | loss:1.1878583431243896 lr:9.027169173526171e-05 tokens_per_second_per_gpu:4515.9449178193545 mfu:21.302482677008282 +Step 7396 | loss:1.085407018661499 lr:9.026912156720484e-05 tokens_per_second_per_gpu:5108.530335869847 mfu:24.09780919945044 +Step 7397 | loss:1.100562334060669 lr:9.026655109627877e-05 tokens_per_second_per_gpu:5364.118432863479 mfu:25.30346186079545 +Step 7398 | loss:1.000531792640686 lr:9.026398032250282e-05 tokens_per_second_per_gpu:5256.212871711416 mfu:24.794452918253995 +Step 7399 | loss:1.1659764051437378 lr:9.026140924589635e-05 tokens_per_second_per_gpu:3542.706975084433 mfu:16.71155324961194 +Step 7400 | loss:0.9524544477462769 lr:9.025883786647869e-05 tokens_per_second_per_gpu:5370.340913361054 mfu:25.33281436296716 +Step 7401 | loss:1.0022200345993042 lr:9.025626618426915e-05 tokens_per_second_per_gpu:5332.881887591899 mfu:25.15611374724522 +Step 7402 | loss:1.1438593864440918 lr:9.02536941992871e-05 tokens_per_second_per_gpu:4591.221432071887 mfu:21.65757483823519 +Step 7403 | loss:1.0744179487228394 lr:9.02511219115519e-05 tokens_per_second_per_gpu:5330.272684387888 mfu:25.143805690555933 +Step 7404 | loss:1.0862964391708374 lr:9.024854932108286e-05 tokens_per_second_per_gpu:5352.061088740597 mfu:25.246585311372776 +Step 7405 | loss:1.1477464437484741 lr:9.024597642789935e-05 tokens_per_second_per_gpu:5361.073964451663 mfu:25.28910058385704 +Step 7406 | loss:1.0326889753341675 lr:9.024340323202071e-05 tokens_per_second_per_gpu:3973.8826530315923 mfu:18.745482488645244 +Step 7407 | loss:1.2371257543563843 lr:9.02408297334663e-05 tokens_per_second_per_gpu:4407.61906711469 mfu:20.791491156938413 +Step 7408 | loss:0.946965754032135 lr:9.023825593225548e-05 tokens_per_second_per_gpu:4024.6155111072467 mfu:18.984798036106266 +Step 7409 | loss:1.0944033861160278 lr:9.023568182840761e-05 tokens_per_second_per_gpu:5175.311752008933 mfu:24.412828533463024 +Step 7410 | loss:0.966607391834259 lr:9.023310742194202e-05 tokens_per_second_per_gpu:5331.744684718038 mfu:25.150749367262403 +Step 7411 | loss:1.2017115354537964 lr:9.023053271287812e-05 tokens_per_second_per_gpu:3405.188697922537 mfu:16.062856073201758 +Step 7412 | loss:1.047426462173462 lr:9.022795770123524e-05 tokens_per_second_per_gpu:4286.380724904393 mfu:20.219589211338253 +Step 7413 | loss:1.0695338249206543 lr:9.022538238703275e-05 tokens_per_second_per_gpu:4423.617711712143 mfu:20.866959493154845 +Step 7414 | loss:1.1172646284103394 lr:9.022280677029005e-05 tokens_per_second_per_gpu:3377.0189905199763 mfu:15.929974757136192 +Step 7415 | loss:1.0898853540420532 lr:9.022023085102646e-05 tokens_per_second_per_gpu:3863.7312490902877 mfu:18.22587952248665 +Step 7416 | loss:1.0723388195037842 lr:9.02176546292614e-05 tokens_per_second_per_gpu:5356.116500755238 mfu:25.265715381770477 +Step 7417 | loss:1.2084641456604004 lr:9.021507810501421e-05 tokens_per_second_per_gpu:4755.041411442192 mfu:22.430341631495573 +Step 7418 | loss:1.03225839138031 lr:9.021250127830429e-05 tokens_per_second_per_gpu:3313.915301356814 mfu:15.632303888753968 +Step 7419 | loss:1.1091488599777222 lr:9.020992414915102e-05 tokens_per_second_per_gpu:2633.7613446845035 mfu:12.423901628899364 +Step 7420 | loss:1.1320639848709106 lr:9.02073467175738e-05 tokens_per_second_per_gpu:3728.833249054013 mfu:17.589542640343403 +Step 7421 | loss:1.2487314939498901 lr:9.020476898359196e-05 tokens_per_second_per_gpu:3692.1799855152804 mfu:17.416642942539582 +Step 7422 | loss:1.0833005905151367 lr:9.020219094722495e-05 tokens_per_second_per_gpu:4234.74798809805 mfu:19.976028782369433 +Step 7423 | loss:1.0216530561447144 lr:9.019961260849213e-05 tokens_per_second_per_gpu:4728.366544932057 mfu:22.304511734965732 +Step 7424 | loss:1.0638415813446045 lr:9.019703396741288e-05 tokens_per_second_per_gpu:4976.202619092195 mfu:23.473596781973665 +Step 7425 | loss:1.2234811782836914 lr:9.01944550240066e-05 tokens_per_second_per_gpu:4713.387866294741 mfu:22.23385475220741 +Step 7426 | loss:1.0696156024932861 lr:9.019187577829272e-05 tokens_per_second_per_gpu:3832.010065092378 mfu:18.076245285376434 +Step 7427 | loss:1.1134191751480103 lr:9.01892962302906e-05 tokens_per_second_per_gpu:4400.552075494247 mfu:20.758154951711685 +Step 7428 | loss:1.099196434020996 lr:9.018671638001966e-05 tokens_per_second_per_gpu:4370.25513306733 mfu:20.615239105092506 +Step 7429 | loss:1.3401960134506226 lr:9.01841362274993e-05 tokens_per_second_per_gpu:4812.147840834822 mfu:22.699722402302164 +Step 7430 | loss:1.0422712564468384 lr:9.018155577274892e-05 tokens_per_second_per_gpu:5354.795841294754 mfu:25.259485605767633 +Step 7431 | loss:1.1760979890823364 lr:9.017897501578793e-05 tokens_per_second_per_gpu:5229.715901112335 mfu:24.669462187088772 +Step 7432 | loss:1.1138412952423096 lr:9.017639395663574e-05 tokens_per_second_per_gpu:4806.952468426427 mfu:22.67521494422983 +Step 7433 | loss:0.9739525318145752 lr:9.017381259531176e-05 tokens_per_second_per_gpu:4921.178088310853 mfu:23.214036682125048 +Step 7434 | loss:1.047301173210144 lr:9.017123093183542e-05 tokens_per_second_per_gpu:4109.682472774247 mfu:19.386073408209796 +Step 7435 | loss:1.0218290090560913 lr:9.016864896622612e-05 tokens_per_second_per_gpu:4448.915975832039 mfu:20.98629572133734 +Step 7436 | loss:1.1961843967437744 lr:9.016606669850329e-05 tokens_per_second_per_gpu:4393.177848442954 mfu:20.72336946453786 +Step 7437 | loss:1.012744426727295 lr:9.016348412868636e-05 tokens_per_second_per_gpu:4363.566574728323 mfu:20.583688034221684 +Step 7438 | loss:1.1791112422943115 lr:9.016090125679473e-05 tokens_per_second_per_gpu:5298.901637395214 mfu:24.99582310943894 +Step 7439 | loss:0.987127423286438 lr:9.015831808284782e-05 tokens_per_second_per_gpu:3132.294223811226 mfu:14.775566278220243 +Step 7440 | loss:0.8788605332374573 lr:9.015573460686509e-05 tokens_per_second_per_gpu:4000.223469906671 mfu:18.86973661605178 +Step 7441 | loss:1.2560256719589233 lr:9.015315082886596e-05 tokens_per_second_per_gpu:3455.0589225438493 mfu:16.298102431478203 +Step 7442 | loss:1.050790786743164 lr:9.015056674886986e-05 tokens_per_second_per_gpu:4392.610766378581 mfu:20.720694441686035 +Step 7443 | loss:0.9990808367729187 lr:9.01479823668962e-05 tokens_per_second_per_gpu:4708.4046421113635 mfu:22.2103480335087 +Step 7444 | loss:1.1010059118270874 lr:9.014539768296448e-05 tokens_per_second_per_gpu:4335.356196889589 mfu:20.4506148687696 +Step 7445 | loss:1.130815863609314 lr:9.014281269709407e-05 tokens_per_second_per_gpu:3888.0847147382797 mfu:18.34075897507793 +Step 7446 | loss:1.0308560132980347 lr:9.014022740930446e-05 tokens_per_second_per_gpu:4633.077841749896 mfu:21.85501866412625 +Step 7447 | loss:1.0449438095092773 lr:9.013764181961508e-05 tokens_per_second_per_gpu:5053.522363349319 mfu:23.838327207743216 +Step 7448 | loss:1.0641965866088867 lr:9.013505592804536e-05 tokens_per_second_per_gpu:4293.428111151556 mfu:20.2528329346742 +Step 7449 | loss:1.1223734617233276 lr:9.013246973461478e-05 tokens_per_second_per_gpu:3988.1355571895224 mfu:18.81271586935441 +Step 7450 | loss:1.0208994150161743 lr:9.012988323934277e-05 tokens_per_second_per_gpu:4113.8110318851295 mfu:19.40554852594065 +Step 7451 | loss:1.081407904624939 lr:9.012729644224877e-05 tokens_per_second_per_gpu:4116.794885526409 mfu:19.419623872664495 +Step 7452 | loss:1.1928505897521973 lr:9.012470934335227e-05 tokens_per_second_per_gpu:4003.9759110427835 mfu:18.8874375211233 +Step 7453 | loss:1.1830668449401855 lr:9.01221219426727e-05 tokens_per_second_per_gpu:3545.178912471676 mfu:16.723213800023572 +Step 7454 | loss:1.0252560377120972 lr:9.011953424022954e-05 tokens_per_second_per_gpu:5338.384491086721 mfu:25.182070466771265 +Step 7455 | loss:1.0955404043197632 lr:9.011694623604224e-05 tokens_per_second_per_gpu:5369.154703474751 mfu:25.32721880854519 +Step 7456 | loss:1.2544119358062744 lr:9.011435793013028e-05 tokens_per_second_per_gpu:3767.1766917007285 mfu:17.770415201373872 +Step 7457 | loss:0.9440094828605652 lr:9.01117693225131e-05 tokens_per_second_per_gpu:5329.115688347568 mfu:25.13834794282986 +Step 7458 | loss:1.127221703529358 lr:9.01091804132102e-05 tokens_per_second_per_gpu:5331.628087590509 mfu:25.150199358718325 +Step 7459 | loss:0.9452675580978394 lr:9.010659120224103e-05 tokens_per_second_per_gpu:5331.813151973768 mfu:25.1510723389894 +Step 7460 | loss:1.0000072717666626 lr:9.010400168962505e-05 tokens_per_second_per_gpu:3714.1155185081966 mfu:17.52011659425502 +Step 7461 | loss:1.134060025215149 lr:9.010141187538178e-05 tokens_per_second_per_gpu:4497.170316856041 mfu:21.213919680987093 +Step 7462 | loss:1.082453727722168 lr:9.009882175953067e-05 tokens_per_second_per_gpu:5318.261721725812 mfu:25.08714793039737 +Step 7463 | loss:1.0836055278778076 lr:9.00962313420912e-05 tokens_per_second_per_gpu:4252.94119161807 mfu:20.06184922745397 +Step 7464 | loss:1.0059758424758911 lr:9.009364062308285e-05 tokens_per_second_per_gpu:5344.641909437442 mfu:25.211587776757277 +Step 7465 | loss:1.144012212753296 lr:9.009104960252512e-05 tokens_per_second_per_gpu:3235.5582311579774 mfu:15.262680219531303 +Step 7466 | loss:1.1641881465911865 lr:9.00884582804375e-05 tokens_per_second_per_gpu:4065.772095086886 mfu:19.178940664775542 +Step 7467 | loss:0.9595193266868591 lr:9.008586665683946e-05 tokens_per_second_per_gpu:3813.565857845774 mfu:17.989240812887946 +Step 7468 | loss:1.0775063037872314 lr:9.008327473175052e-05 tokens_per_second_per_gpu:5330.047690323384 mfu:25.142744355165718 +Step 7469 | loss:1.069746732711792 lr:9.008068250519013e-05 tokens_per_second_per_gpu:5251.74601350968 mfu:24.77338198599181 +Step 7470 | loss:1.1263279914855957 lr:9.007808997717781e-05 tokens_per_second_per_gpu:3515.93280915163 mfu:16.585254940763143 +Step 7471 | loss:1.1653404235839844 lr:9.007549714773309e-05 tokens_per_second_per_gpu:4751.9308129749115 mfu:22.415668407802585 +Step 7472 | loss:1.13174307346344 lr:9.007290401687542e-05 tokens_per_second_per_gpu:3685.7830896860996 mfu:17.386467693490363 +Step 7473 | loss:1.028663992881775 lr:9.007031058462434e-05 tokens_per_second_per_gpu:5045.116959622783 mfu:23.79867748425539 +Step 7474 | loss:1.2260693311691284 lr:9.006771685099933e-05 tokens_per_second_per_gpu:4776.9382261457795 mfu:22.533632642434224 +Step 7475 | loss:1.2527683973312378 lr:9.006512281601991e-05 tokens_per_second_per_gpu:3486.7098999850814 mfu:16.447405492282098 +Step 7476 | loss:0.9619235396385193 lr:9.00625284797056e-05 tokens_per_second_per_gpu:3899.468852241498 mfu:18.39445989401476 +Step 7477 | loss:1.1240363121032715 lr:9.005993384207588e-05 tokens_per_second_per_gpu:5341.243397695388 mfu:25.195556417024115 +Step 7478 | loss:1.0569133758544922 lr:9.00573389031503e-05 tokens_per_second_per_gpu:4267.371554441966 mfu:20.129919710968064 +Step 7479 | loss:1.2639766931533813 lr:9.005474366294838e-05 tokens_per_second_per_gpu:5350.976399318311 mfu:25.241468646300756 +Step 7480 | loss:1.1615815162658691 lr:9.005214812148959e-05 tokens_per_second_per_gpu:5300.9929710429315 mfu:25.00568828707354 +Step 7481 | loss:1.0758447647094727 lr:9.004955227879349e-05 tokens_per_second_per_gpu:5215.786747907554 mfu:24.60375599486322 +Step 7482 | loss:1.066160798072815 lr:9.00469561348796e-05 tokens_per_second_per_gpu:4787.752178313083 mfu:22.584643899857888 +Step 7483 | loss:0.9289735555648804 lr:9.004435968976745e-05 tokens_per_second_per_gpu:4771.277267358384 mfu:22.506928933974912 +Step 7484 | loss:1.0040736198425293 lr:9.004176294347656e-05 tokens_per_second_per_gpu:4136.3083623669645 mfu:19.51167227226385 +Step 7485 | loss:1.057110071182251 lr:9.003916589602646e-05 tokens_per_second_per_gpu:5185.994729323064 mfu:24.463221960157668 +Step 7486 | loss:0.9416449069976807 lr:9.003656854743667e-05 tokens_per_second_per_gpu:4106.123810202932 mfu:19.36928658969046 +Step 7487 | loss:1.0444008111953735 lr:9.003397089772675e-05 tokens_per_second_per_gpu:5339.892295870118 mfu:25.18918303919257 +Step 7488 | loss:1.1180570125579834 lr:9.003137294691623e-05 tokens_per_second_per_gpu:4347.478468833175 mfu:20.507797693800356 +Step 7489 | loss:1.0389102697372437 lr:9.002877469502466e-05 tokens_per_second_per_gpu:5023.664313580333 mfu:23.697481692674312 +Step 7490 | loss:1.0704928636550903 lr:9.002617614207155e-05 tokens_per_second_per_gpu:4263.279387843879 mfu:20.11061626292927 +Step 7491 | loss:1.088767647743225 lr:9.002357728807646e-05 tokens_per_second_per_gpu:4619.220663597963 mfu:21.78965199050519 +Step 7492 | loss:1.1935009956359863 lr:9.002097813305895e-05 tokens_per_second_per_gpu:4368.461706792835 mfu:20.606779207365637 +Step 7493 | loss:1.1274901628494263 lr:9.001837867703856e-05 tokens_per_second_per_gpu:4275.816193171323 mfu:20.169754512658383 +Step 7494 | loss:1.288397192955017 lr:9.00157789200348e-05 tokens_per_second_per_gpu:3110.8775463409465 mfu:14.674540156530965 +Step 7495 | loss:1.12581467628479 lr:9.001317886206731e-05 tokens_per_second_per_gpu:4087.6760754831803 mfu:19.282265477509288 +Step 7496 | loss:1.108078122138977 lr:9.001057850315556e-05 tokens_per_second_per_gpu:5385.222813395974 mfu:25.40301482454585 +Step 7497 | loss:1.0761029720306396 lr:9.000797784331915e-05 tokens_per_second_per_gpu:5338.728735096685 mfu:25.18369432450035 +Step 7498 | loss:0.9665701389312744 lr:9.000537688257764e-05 tokens_per_second_per_gpu:5183.865966944638 mfu:24.453180224814314 +Step 7499 | loss:1.11775541305542 lr:9.00027756209506e-05 tokens_per_second_per_gpu:4938.019740493269 mfu:23.293481628951668 +Step 7500 | loss:1.0422084331512451 lr:9.000017405845755e-05 tokens_per_second_per_gpu:5216.989817682329 mfu:24.609431080255703 +Step 7501 | loss:1.1085313558578491 lr:8.999757219511812e-05 tokens_per_second_per_gpu:5327.83609243814 mfu:25.132311870603516 +Step 7502 | loss:1.0523591041564941 lr:8.999497003095184e-05 tokens_per_second_per_gpu:5025.415476291032 mfu:23.705742225959845 +Step 7503 | loss:1.0560503005981445 lr:8.999236756597827e-05 tokens_per_second_per_gpu:5189.684277748657 mfu:24.480626189583063 +Step 7504 | loss:0.9160750508308411 lr:8.9989764800217e-05 tokens_per_second_per_gpu:5351.351690377465 mfu:25.243238958258043 +Step 7505 | loss:1.0520801544189453 lr:8.998716173368762e-05 tokens_per_second_per_gpu:4580.269987837824 mfu:21.605915007274053 +Step 7506 | loss:1.1969929933547974 lr:8.998455836640968e-05 tokens_per_second_per_gpu:4290.521729575359 mfu:20.239123036899453 +Step 7507 | loss:0.8629629611968994 lr:8.998195469840278e-05 tokens_per_second_per_gpu:3581.855315327788 mfu:16.896222649935446 +Step 7508 | loss:0.9394901394844055 lr:8.99793507296865e-05 tokens_per_second_per_gpu:5152.624034271359 mfu:24.305806697971086 +Step 7509 | loss:1.2158482074737549 lr:8.997674646028042e-05 tokens_per_second_per_gpu:3996.096694651675 mfu:18.8502699130735 +Step 7510 | loss:1.0637760162353516 lr:8.997414189020414e-05 tokens_per_second_per_gpu:3936.783454020018 mfu:18.570479237131593 +Step 7511 | loss:1.2478137016296387 lr:8.997153701947722e-05 tokens_per_second_per_gpu:3937.5424961433255 mfu:18.57405976833318 +Step 7512 | loss:1.0981062650680542 lr:8.996893184811927e-05 tokens_per_second_per_gpu:4302.913254698281 mfu:20.297575975116473 +Step 7513 | loss:1.075429916381836 lr:8.99663263761499e-05 tokens_per_second_per_gpu:4027.4124747842243 mfu:18.99799179098178 +Step 7514 | loss:1.1106433868408203 lr:8.996372060358867e-05 tokens_per_second_per_gpu:3481.1634900036597 mfu:16.421242130084607 +Step 7515 | loss:1.1118321418762207 lr:8.996111453045522e-05 tokens_per_second_per_gpu:5279.056240361692 mfu:24.902208985657936 +Step 7516 | loss:0.9247779250144958 lr:8.995850815676911e-05 tokens_per_second_per_gpu:4168.252232262699 mfu:19.662356956748244 +Step 7517 | loss:1.0614186525344849 lr:8.995590148254997e-05 tokens_per_second_per_gpu:4790.931999880717 mfu:22.59964366072592 +Step 7518 | loss:1.0283774137496948 lr:8.99532945078174e-05 tokens_per_second_per_gpu:4706.317913397777 mfu:22.200504578134083 +Step 7519 | loss:1.095902681350708 lr:8.995068723259099e-05 tokens_per_second_per_gpu:4204.970408967215 mfu:19.83556286103039 +Step 7520 | loss:1.152601718902588 lr:8.994807965689038e-05 tokens_per_second_per_gpu:3898.46934807482 mfu:18.38974506232754 +Step 7521 | loss:1.132778525352478 lr:8.994547178073516e-05 tokens_per_second_per_gpu:5360.969160241497 mfu:25.288606204516242 +Step 7522 | loss:1.041934847831726 lr:8.994286360414494e-05 tokens_per_second_per_gpu:4343.907992767854 mfu:20.490955149014205 +Step 7523 | loss:1.129260540008545 lr:8.994025512713935e-05 tokens_per_second_per_gpu:4096.421522773681 mfu:19.32351924450575 +Step 7524 | loss:1.1349722146987915 lr:8.9937646349738e-05 tokens_per_second_per_gpu:5283.3770826098435 mfu:24.92259113575417 +Step 7525 | loss:1.0112991333007812 lr:8.993503727196053e-05 tokens_per_second_per_gpu:4029.982642510137 mfu:19.010115710660745 +Step 7526 | loss:1.1010814905166626 lr:8.993242789382654e-05 tokens_per_second_per_gpu:3934.730059969028 mfu:18.56079302704792 +Step 7527 | loss:1.2567427158355713 lr:8.992981821535567e-05 tokens_per_second_per_gpu:4372.05271176008 mfu:20.623718590484756 +Step 7528 | loss:1.0129599571228027 lr:8.992720823656755e-05 tokens_per_second_per_gpu:4250.854516885885 mfu:20.052006026718832 +Step 7529 | loss:1.1791589260101318 lr:8.992459795748178e-05 tokens_per_second_per_gpu:3895.171367559029 mfu:18.374187925541055 +Step 7530 | loss:1.060234546661377 lr:8.992198737811804e-05 tokens_per_second_per_gpu:5358.4734888113735 mfu:25.276833696574897 +Step 7531 | loss:1.1542141437530518 lr:8.991937649849592e-05 tokens_per_second_per_gpu:3659.104143120981 mfu:17.260618550645038 +Step 7532 | loss:1.1761515140533447 lr:8.991676531863508e-05 tokens_per_second_per_gpu:4777.295102581988 mfu:22.535316089473216 +Step 7533 | loss:0.9810968637466431 lr:8.991415383855514e-05 tokens_per_second_per_gpu:4830.055890112223 mfu:22.784197726171936 +Step 7534 | loss:0.9663745164871216 lr:8.991154205827579e-05 tokens_per_second_per_gpu:3925.282563121336 mfu:18.516227572508182 +Step 7535 | loss:1.1875288486480713 lr:8.990892997781662e-05 tokens_per_second_per_gpu:4209.753584963582 mfu:19.858125918299237 +Step 7536 | loss:1.0687305927276611 lr:8.990631759719728e-05 tokens_per_second_per_gpu:5221.652857864076 mfu:24.6314274363897 +Step 7537 | loss:1.1826977729797363 lr:8.990370491643746e-05 tokens_per_second_per_gpu:5335.203562941303 mfu:25.16706548598702 +Step 7538 | loss:1.0438543558120728 lr:8.990109193555677e-05 tokens_per_second_per_gpu:3731.2182613464074 mfu:17.600793150251686 +Step 7539 | loss:0.9119564890861511 lr:8.989847865457488e-05 tokens_per_second_per_gpu:4745.34132543624 mfu:22.384584670800166 +Step 7540 | loss:1.0816855430603027 lr:8.989586507351145e-05 tokens_per_second_per_gpu:4876.908384288892 mfu:23.005208935063095 +Step 7541 | loss:1.0524178743362427 lr:8.989325119238611e-05 tokens_per_second_per_gpu:4111.9336989196545 mfu:19.396692826036467 +Step 7542 | loss:1.1289359331130981 lr:8.989063701121854e-05 tokens_per_second_per_gpu:4196.4066514640635 mfu:19.795166155760327 +Step 7543 | loss:1.1038269996643066 lr:8.98880225300284e-05 tokens_per_second_per_gpu:3792.7101725073903 mfu:17.890860986799947 +Step 7544 | loss:0.9828106760978699 lr:8.988540774883536e-05 tokens_per_second_per_gpu:4670.533608093349 mfu:22.031703904581036 +Step 7545 | loss:1.1395796537399292 lr:8.988279266765908e-05 tokens_per_second_per_gpu:4581.439319218896 mfu:21.611430942907077 +Step 7546 | loss:1.1493210792541504 lr:8.988017728651922e-05 tokens_per_second_per_gpu:5077.2045869888625 mfu:23.950040297254947 +Step 7547 | loss:1.1382659673690796 lr:8.987756160543546e-05 tokens_per_second_per_gpu:5327.417119231421 mfu:25.13033550250322 +Step 7548 | loss:1.07377028465271 lr:8.987494562442748e-05 tokens_per_second_per_gpu:4660.899536303717 mfu:21.986258344207194 +Step 7549 | loss:1.1071584224700928 lr:8.987232934351493e-05 tokens_per_second_per_gpu:5338.190249099116 mfu:25.181154194174628 +Step 7550 | loss:1.3030434846878052 lr:8.986971276271752e-05 tokens_per_second_per_gpu:3036.741259299733 mfu:14.324826641602671 +Step 7551 | loss:1.2081060409545898 lr:8.986709588205488e-05 tokens_per_second_per_gpu:4500.752307696552 mfu:21.230816542932445 +Step 7552 | loss:1.0823131799697876 lr:8.986447870154676e-05 tokens_per_second_per_gpu:4578.393642969048 mfu:21.597063968390593 +Step 7553 | loss:1.1970984935760498 lr:8.986186122121279e-05 tokens_per_second_per_gpu:4370.979835676739 mfu:20.6186576509481 +Step 7554 | loss:1.0295149087905884 lr:8.985924344107269e-05 tokens_per_second_per_gpu:5306.6802626897925 mfu:25.0325162121238 +Step 7555 | loss:1.0794976949691772 lr:8.985662536114613e-05 tokens_per_second_per_gpu:4902.873668030758 mfu:23.1276916086077 +Step 7556 | loss:1.0590978860855103 lr:8.985400698145281e-05 tokens_per_second_per_gpu:5296.445728888397 mfu:24.984238169991087 +Step 7557 | loss:1.0852946043014526 lr:8.98513883020124e-05 tokens_per_second_per_gpu:4987.00391796018 mfu:23.52454835162557 +Step 7558 | loss:1.296587347984314 lr:8.984876932284463e-05 tokens_per_second_per_gpu:5333.510669398962 mfu:25.159079818310378 +Step 7559 | loss:1.0295449495315552 lr:8.984615004396917e-05 tokens_per_second_per_gpu:4779.105713773381 mfu:22.54385705138514 +Step 7560 | loss:1.1376945972442627 lr:8.984353046540573e-05 tokens_per_second_per_gpu:4358.010297788612 mfu:20.557478127898463 +Step 7561 | loss:1.0824942588806152 lr:8.984091058717402e-05 tokens_per_second_per_gpu:3614.7167036577016 mfu:17.051235425418646 +Step 7562 | loss:1.2665798664093018 lr:8.983829040929373e-05 tokens_per_second_per_gpu:4162.4679415146875 mfu:19.6350714703881 +Step 7563 | loss:1.1837059259414673 lr:8.983566993178458e-05 tokens_per_second_per_gpu:4307.710020811332 mfu:20.32020313928442 +Step 7564 | loss:1.17813241481781 lr:8.983304915466628e-05 tokens_per_second_per_gpu:3448.95090481277 mfu:16.269289869705414 +Step 7565 | loss:1.0957235097885132 lr:8.983042807795854e-05 tokens_per_second_per_gpu:4193.677965098399 mfu:19.782294476611586 +Step 7566 | loss:1.1150388717651367 lr:8.982780670168105e-05 tokens_per_second_per_gpu:3567.930800427439 mfu:16.830538337383278 +Step 7567 | loss:1.0822079181671143 lr:8.982518502585355e-05 tokens_per_second_per_gpu:4291.247458231896 mfu:20.242546422794415 +Step 7568 | loss:1.1378111839294434 lr:8.982256305049575e-05 tokens_per_second_per_gpu:4865.44795087666 mfu:22.951148115306022 +Step 7569 | loss:1.0636770725250244 lr:8.981994077562738e-05 tokens_per_second_per_gpu:5341.457967437679 mfu:25.196568579107872 +Step 7570 | loss:1.0686982870101929 lr:8.981731820126815e-05 tokens_per_second_per_gpu:5353.120844516454 mfu:25.251584360180473 +Step 7571 | loss:1.117903232574463 lr:8.981469532743778e-05 tokens_per_second_per_gpu:5381.810733214591 mfu:25.386919460169356 +Step 7572 | loss:1.3652158975601196 lr:8.981207215415601e-05 tokens_per_second_per_gpu:3347.3119003646375 mfu:15.789841344321225 +Step 7573 | loss:0.9857797026634216 lr:8.980944868144258e-05 tokens_per_second_per_gpu:4651.825691868857 mfu:21.94345547185046 +Step 7574 | loss:1.1699963808059692 lr:8.98068249093172e-05 tokens_per_second_per_gpu:3549.3381447012766 mfu:16.74283360808879 +Step 7575 | loss:1.1423028707504272 lr:8.980420083779961e-05 tokens_per_second_per_gpu:3686.1839180326424 mfu:17.388358469189374 +Step 7576 | loss:1.0721871852874756 lr:8.980157646690953e-05 tokens_per_second_per_gpu:5348.988309157373 mfu:25.23209048580845 +Step 7577 | loss:1.2387290000915527 lr:8.979895179666673e-05 tokens_per_second_per_gpu:5319.136144421703 mfu:25.091272731445784 +Step 7578 | loss:1.0253667831420898 lr:8.979632682709092e-05 tokens_per_second_per_gpu:5320.3410403720945 mfu:25.09695643122023 +Step 7579 | loss:1.207956075668335 lr:8.979370155820188e-05 tokens_per_second_per_gpu:5327.913519666359 mfu:25.13267710804947 +Step 7580 | loss:0.9982014894485474 lr:8.979107599001933e-05 tokens_per_second_per_gpu:5307.948607242325 mfu:25.038499209779992 +Step 7581 | loss:1.0725375413894653 lr:8.9788450122563e-05 tokens_per_second_per_gpu:5236.749614878467 mfu:24.702641414998933 +Step 7582 | loss:1.2130495309829712 lr:8.978582395585268e-05 tokens_per_second_per_gpu:3997.355835561753 mfu:18.85620949557793 +Step 7583 | loss:1.0603229999542236 lr:8.978319748990808e-05 tokens_per_second_per_gpu:4242.070671158549 mfu:20.010571127744367 +Step 7584 | loss:1.068774938583374 lr:8.978057072474902e-05 tokens_per_second_per_gpu:5346.011759782628 mfu:25.21804959455609 +Step 7585 | loss:1.125206470489502 lr:8.977794366039516e-05 tokens_per_second_per_gpu:3783.8156153453697 mfu:17.848903843098757 +Step 7586 | loss:1.0124150514602661 lr:8.977531629686633e-05 tokens_per_second_per_gpu:4077.4647897852724 mfu:19.23409710064726 +Step 7587 | loss:1.1560131311416626 lr:8.977268863418228e-05 tokens_per_second_per_gpu:4350.117822955819 mfu:20.520247977517116 +Step 7588 | loss:1.138476014137268 lr:8.977006067236275e-05 tokens_per_second_per_gpu:4806.363586492993 mfu:22.672437087676197 +Step 7589 | loss:1.1048388481140137 lr:8.976743241142753e-05 tokens_per_second_per_gpu:4738.631055908143 mfu:22.35293118454609 +Step 7590 | loss:1.1746587753295898 lr:8.976480385139637e-05 tokens_per_second_per_gpu:4294.15409834689 mfu:20.256257540140783 +Step 7591 | loss:1.0521652698516846 lr:8.976217499228905e-05 tokens_per_second_per_gpu:4439.25404213445 mfu:20.940718731589932 +Step 7592 | loss:1.1733835935592651 lr:8.975954583412534e-05 tokens_per_second_per_gpu:4252.817015315884 mfu:20.061263466649898 +Step 7593 | loss:1.245418906211853 lr:8.9756916376925e-05 tokens_per_second_per_gpu:4194.875697469186 mfu:19.787944384558806 +Step 7594 | loss:1.1005254983901978 lr:8.975428662070783e-05 tokens_per_second_per_gpu:4100.615180138064 mfu:19.34330144180527 +Step 7595 | loss:0.9266560673713684 lr:8.97516565654936e-05 tokens_per_second_per_gpu:5351.170313092752 mfu:25.24238337065989 +Step 7596 | loss:1.0205585956573486 lr:8.974902621130207e-05 tokens_per_second_per_gpu:4481.923485898366 mfu:21.14199777798214 +Step 7597 | loss:1.1213502883911133 lr:8.974639555815306e-05 tokens_per_second_per_gpu:5111.216071966787 mfu:24.11047827485354 +Step 7598 | loss:1.019334316253662 lr:8.974376460606633e-05 tokens_per_second_per_gpu:4374.541105464973 mfu:20.635456768154484 +Step 7599 | loss:1.094190239906311 lr:8.974113335506167e-05 tokens_per_second_per_gpu:5337.649628713231 mfu:25.178603995575063 +Step 7600 | loss:1.073276400566101 lr:8.97385018051589e-05 tokens_per_second_per_gpu:4697.696186424411 mfu:22.159834420982406 +Step 7601 | loss:1.0009591579437256 lr:8.973586995637778e-05 tokens_per_second_per_gpu:4722.347062429734 mfu:22.276116808972656 +Step 7602 | loss:1.3635505437850952 lr:8.973323780873808e-05 tokens_per_second_per_gpu:5352.53853557839 mfu:25.24883750956748 +Step 7603 | loss:1.0000871419906616 lr:8.973060536225965e-05 tokens_per_second_per_gpu:4667.206087569507 mfu:22.01600742253653 +Step 7604 | loss:0.9523777365684509 lr:8.972797261696227e-05 tokens_per_second_per_gpu:5333.630440698318 mfu:25.15964479996453 +Step 7605 | loss:1.0434272289276123 lr:8.972533957286573e-05 tokens_per_second_per_gpu:5204.104113500008 mfu:24.548647015099853 +Step 7606 | loss:1.0755387544631958 lr:8.972270622998986e-05 tokens_per_second_per_gpu:3520.9410981770375 mfu:16.608879894598214 +Step 7607 | loss:1.0576366186141968 lr:8.972007258835444e-05 tokens_per_second_per_gpu:4671.643526291726 mfu:22.03693957809435 +Step 7608 | loss:1.1691159009933472 lr:8.971743864797928e-05 tokens_per_second_per_gpu:4587.730228533188 mfu:21.641106235482397 +Step 7609 | loss:1.0698944330215454 lr:8.971480440888419e-05 tokens_per_second_per_gpu:5346.765932566062 mfu:25.22160715625069 +Step 7610 | loss:0.9794967174530029 lr:8.9712169871089e-05 tokens_per_second_per_gpu:4640.270346890834 mfu:21.888946938043972 +Step 7611 | loss:1.2382633686065674 lr:8.97095350346135e-05 tokens_per_second_per_gpu:4741.245219480441 mfu:22.365262640138916 +Step 7612 | loss:0.9549009203910828 lr:8.970689989947754e-05 tokens_per_second_per_gpu:4295.295735342615 mfu:20.261642836632976 +Step 7613 | loss:0.9709518551826477 lr:8.97042644657009e-05 tokens_per_second_per_gpu:4725.300122602853 mfu:22.290046897653276 +Step 7614 | loss:1.1872512102127075 lr:8.970162873330343e-05 tokens_per_second_per_gpu:4050.8565250771344 mfu:19.108581376180883 +Step 7615 | loss:1.223156452178955 lr:8.969899270230496e-05 tokens_per_second_per_gpu:4966.8077735107145 mfu:23.429279692439877 +Step 7616 | loss:1.041210651397705 lr:8.969635637272526e-05 tokens_per_second_per_gpu:5340.860956921896 mfu:25.193752378644426 +Step 7617 | loss:0.9983565807342529 lr:8.969371974458423e-05 tokens_per_second_per_gpu:4783.9023818062615 mfu:22.566483752892193 +Step 7618 | loss:1.0651615858078003 lr:8.969108281790166e-05 tokens_per_second_per_gpu:4688.566258949748 mfu:22.116766995358162 +Step 7619 | loss:1.046679973602295 lr:8.96884455926974e-05 tokens_per_second_per_gpu:4671.0695785309945 mfu:22.034232168581397 +Step 7620 | loss:1.2049366235733032 lr:8.968580806899126e-05 tokens_per_second_per_gpu:5366.510855129867 mfu:25.314747324815333 +Step 7621 | loss:1.1401722431182861 lr:8.968317024680311e-05 tokens_per_second_per_gpu:4398.945967745609 mfu:20.750578667431117 +Step 7622 | loss:1.0569616556167603 lr:8.968053212615276e-05 tokens_per_second_per_gpu:4617.5006762918465 mfu:21.781538516922147 +Step 7623 | loss:1.0029584169387817 lr:8.967789370706007e-05 tokens_per_second_per_gpu:4168.654514243583 mfu:19.664254589488458 +Step 7624 | loss:1.0297513008117676 lr:8.967525498954487e-05 tokens_per_second_per_gpu:4716.182601459733 mfu:22.247037994811663 +Step 7625 | loss:1.1246689558029175 lr:8.967261597362702e-05 tokens_per_second_per_gpu:4662.46822943685 mfu:21.993658137362836 +Step 7626 | loss:1.0966308116912842 lr:8.966997665932637e-05 tokens_per_second_per_gpu:5318.944518002847 mfu:25.090368796933205 +Step 7627 | loss:1.0623993873596191 lr:8.966733704666276e-05 tokens_per_second_per_gpu:4483.700968363698 mfu:21.150382466040888 +Step 7628 | loss:1.0790055990219116 lr:8.966469713565606e-05 tokens_per_second_per_gpu:5026.345073447084 mfu:23.71012729435057 +Step 7629 | loss:1.1068286895751953 lr:8.96620569263261e-05 tokens_per_second_per_gpu:4989.588759607059 mfu:23.53674149069336 +Step 7630 | loss:1.109859824180603 lr:8.965941641869273e-05 tokens_per_second_per_gpu:4931.168977521199 mfu:23.261165411151726 +Step 7631 | loss:1.0723117589950562 lr:8.965677561277585e-05 tokens_per_second_per_gpu:4423.472820043112 mfu:20.86627601443095 +Step 7632 | loss:1.0320981740951538 lr:8.96541345085953e-05 tokens_per_second_per_gpu:4348.508801115169 mfu:20.512657946967337 +Step 7633 | loss:1.228503942489624 lr:8.965149310617094e-05 tokens_per_second_per_gpu:4760.7316131404905 mfu:22.457183283754965 +Step 7634 | loss:0.987400233745575 lr:8.964885140552264e-05 tokens_per_second_per_gpu:4510.682535487349 mfu:21.2776591216932 +Step 7635 | loss:1.1062664985656738 lr:8.964620940667028e-05 tokens_per_second_per_gpu:3822.4555541085083 mfu:18.031175026897845 +Step 7636 | loss:0.9848648309707642 lr:8.96435671096337e-05 tokens_per_second_per_gpu:4957.895813203859 mfu:23.387240455135217 +Step 7637 | loss:0.9581143260002136 lr:8.96409245144328e-05 tokens_per_second_per_gpu:3669.4850937722545 mfu:17.309587266039912 +Step 7638 | loss:1.1284915208816528 lr:8.963828162108745e-05 tokens_per_second_per_gpu:4921.2709461704 mfu:23.214474708491718 +Step 7639 | loss:1.2252148389816284 lr:8.963563842961753e-05 tokens_per_second_per_gpu:3022.926702600574 mfu:14.25966102064632 +Step 7640 | loss:1.158568263053894 lr:8.963299494004291e-05 tokens_per_second_per_gpu:3430.0882111839455 mfu:16.180311325550857 +Step 7641 | loss:1.0635054111480713 lr:8.963035115238349e-05 tokens_per_second_per_gpu:4953.242149071683 mfu:23.36528832742677 +Step 7642 | loss:1.1767501831054688 lr:8.962770706665913e-05 tokens_per_second_per_gpu:4967.62056128297 mfu:23.4331137510375 +Step 7643 | loss:1.149670958518982 lr:8.962506268288973e-05 tokens_per_second_per_gpu:3440.7791998360008 mfu:16.230742543093356 +Step 7644 | loss:1.0810874700546265 lr:8.96224180010952e-05 tokens_per_second_per_gpu:5352.574265362438 mfu:25.24900605305496 +Step 7645 | loss:1.0118738412857056 lr:8.961977302129537e-05 tokens_per_second_per_gpu:4635.110724950867 mfu:21.864608121030866 +Step 7646 | loss:1.1822872161865234 lr:8.96171277435102e-05 tokens_per_second_per_gpu:3787.086894104113 mfu:17.864335023141457 +Step 7647 | loss:0.9620224833488464 lr:8.961448216775954e-05 tokens_per_second_per_gpu:5194.903453101776 mfu:24.505245930206858 +Step 7648 | loss:1.0663647651672363 lr:8.961183629406332e-05 tokens_per_second_per_gpu:5318.432083343047 mfu:25.087951555212562 +Step 7649 | loss:1.061337947845459 lr:8.96091901224414e-05 tokens_per_second_per_gpu:5093.804331653101 mfu:24.02834412504465 +Step 7650 | loss:0.9824526906013489 lr:8.960654365291374e-05 tokens_per_second_per_gpu:4889.33850809934 mfu:23.063843949874787 +Step 7651 | loss:0.9327400922775269 lr:8.96038968855002e-05 tokens_per_second_per_gpu:4865.39578068151 mfu:22.950902019594345 +Step 7652 | loss:1.128760576248169 lr:8.960124982022067e-05 tokens_per_second_per_gpu:4115.618001273355 mfu:19.414072308845675 +Step 7653 | loss:1.1430751085281372 lr:8.959860245709512e-05 tokens_per_second_per_gpu:5352.218835439284 mfu:25.247329429464845 +Step 7654 | loss:1.013258457183838 lr:8.959595479614342e-05 tokens_per_second_per_gpu:4542.25197131812 mfu:21.42657753680812 +Step 7655 | loss:1.1899417638778687 lr:8.959330683738549e-05 tokens_per_second_per_gpu:5222.44884371363 mfu:24.635182237451232 +Step 7656 | loss:0.9646562337875366 lr:8.959065858084123e-05 tokens_per_second_per_gpu:3876.6176568623823 mfu:18.286666906595503 +Step 7657 | loss:1.0713478326797485 lr:8.958801002653061e-05 tokens_per_second_per_gpu:4683.220620990746 mfu:22.091550709045666 +Step 7658 | loss:1.1120781898498535 lr:8.95853611744735e-05 tokens_per_second_per_gpu:4751.553698928221 mfu:22.41388949650205 +Step 7659 | loss:0.9747166633605957 lr:8.958271202468982e-05 tokens_per_second_per_gpu:5337.355757898098 mfu:25.17721775680055 +Step 7660 | loss:1.0983842611312866 lr:8.958006257719953e-05 tokens_per_second_per_gpu:4985.506614665275 mfu:23.51748531651722 +Step 7661 | loss:1.0935055017471313 lr:8.957741283202254e-05 tokens_per_second_per_gpu:4636.394928539815 mfu:21.870665928467663 +Step 7662 | loss:1.1545239686965942 lr:8.957476278917878e-05 tokens_per_second_per_gpu:3538.568302060521 mfu:16.692030422827713 +Step 7663 | loss:1.1950194835662842 lr:8.957211244868817e-05 tokens_per_second_per_gpu:5331.662140140149 mfu:25.150359990404944 +Step 7664 | loss:1.0887926816940308 lr:8.956946181057064e-05 tokens_per_second_per_gpu:4507.809060146002 mfu:21.26410444824262 +Step 7665 | loss:1.1394827365875244 lr:8.956681087484615e-05 tokens_per_second_per_gpu:3483.7593229247886 mfu:16.43348711686834 +Step 7666 | loss:1.127658724784851 lr:8.956415964153464e-05 tokens_per_second_per_gpu:4365.359098802206 mfu:20.592143676114333 +Step 7667 | loss:1.00689697265625 lr:8.956150811065603e-05 tokens_per_second_per_gpu:4839.372882781527 mfu:22.82814757023591 +Step 7668 | loss:0.9434032440185547 lr:8.955885628223026e-05 tokens_per_second_per_gpu:4750.186505141946 mfu:22.40744021014488 +Step 7669 | loss:1.072522521018982 lr:8.95562041562773e-05 tokens_per_second_per_gpu:3880.4723996387265 mfu:18.304850385958456 +Step 7670 | loss:1.075884461402893 lr:8.955355173281708e-05 tokens_per_second_per_gpu:4542.367176369966 mfu:21.427120978693686 +Step 7671 | loss:1.209609031677246 lr:8.955089901186953e-05 tokens_per_second_per_gpu:5022.907631507133 mfu:23.693912294231755 +Step 7672 | loss:1.1631672382354736 lr:8.954824599345464e-05 tokens_per_second_per_gpu:5340.208644194169 mfu:25.19067530821171 +Step 7673 | loss:1.0896589756011963 lr:8.954559267759233e-05 tokens_per_second_per_gpu:5346.049745886576 mfu:25.218228781489504 +Step 7674 | loss:1.024855136871338 lr:8.954293906430257e-05 tokens_per_second_per_gpu:3818.6032714448706 mfu:18.013003152306272 +Step 7675 | loss:1.170635461807251 lr:8.954028515360536e-05 tokens_per_second_per_gpu:5382.068595852852 mfu:25.388135842229932 +Step 7676 | loss:1.1530098915100098 lr:8.953763094552057e-05 tokens_per_second_per_gpu:3297.4187697619723 mfu:15.554486934622576 +Step 7677 | loss:1.1411244869232178 lr:8.953497644006823e-05 tokens_per_second_per_gpu:4743.10468236942 mfu:22.374034043843313 +Step 7678 | loss:0.9790664911270142 lr:8.95323216372683e-05 tokens_per_second_per_gpu:4465.419722219804 mfu:21.064146708877914 +Step 7679 | loss:1.24519944190979 lr:8.952966653714072e-05 tokens_per_second_per_gpu:4161.516899495477 mfu:19.63058524291899 +Step 7680 | loss:1.2312114238739014 lr:8.952701113970549e-05 tokens_per_second_per_gpu:4312.73329573937 mfu:20.343898784179057 +Step 7681 | loss:1.0163135528564453 lr:8.952435544498255e-05 tokens_per_second_per_gpu:4857.884067153519 mfu:22.915467985252327 +Step 7682 | loss:1.1674797534942627 lr:8.952169945299189e-05 tokens_per_second_per_gpu:5307.278048741696 mfu:25.035336070923442 +Step 7683 | loss:0.9658762812614441 lr:8.951904316375349e-05 tokens_per_second_per_gpu:5361.308299211566 mfu:25.29020598090113 +Step 7684 | loss:1.0770317316055298 lr:8.951638657728733e-05 tokens_per_second_per_gpu:5328.312865937597 mfu:25.1345608925464 +Step 7685 | loss:1.0061010122299194 lr:8.951372969361338e-05 tokens_per_second_per_gpu:4471.199868093194 mfu:21.091412643157273 +Step 7686 | loss:1.0560873746871948 lr:8.951107251275161e-05 tokens_per_second_per_gpu:4516.098436637223 mfu:21.30320685146518 +Step 7687 | loss:1.1477776765823364 lr:8.950841503472206e-05 tokens_per_second_per_gpu:5037.148397936878 mfu:23.761088419206157 +Step 7688 | loss:1.1222115755081177 lr:8.950575725954464e-05 tokens_per_second_per_gpu:4005.307444846222 mfu:18.893718593257212 +Step 7689 | loss:0.9384452104568481 lr:8.950309918723941e-05 tokens_per_second_per_gpu:5349.501645206293 mfu:25.234511979535366 +Step 7690 | loss:1.2146128416061401 lr:8.95004408178263e-05 tokens_per_second_per_gpu:3274.6437092247506 mfu:15.447053088242482 +Step 7691 | loss:1.1164205074310303 lr:8.949778215132534e-05 tokens_per_second_per_gpu:4971.634971248129 mfu:23.45205040777212 +Step 7692 | loss:1.0029700994491577 lr:8.949512318775654e-05 tokens_per_second_per_gpu:5320.773771415243 mfu:25.098997697382288 +Step 7693 | loss:1.1208540201187134 lr:8.949246392713987e-05 tokens_per_second_per_gpu:4658.270192936086 mfu:21.97385528292984 +Step 7694 | loss:1.1027964353561401 lr:8.948980436949534e-05 tokens_per_second_per_gpu:4905.558126223245 mfu:23.14035465591746 +Step 7695 | loss:1.0804041624069214 lr:8.948714451484294e-05 tokens_per_second_per_gpu:3635.7268618005883 mfu:17.150343926081263 +Step 7696 | loss:1.149263620376587 lr:8.94844843632027e-05 tokens_per_second_per_gpu:4574.872435105591 mfu:21.58045383011865 +Step 7697 | loss:1.1106668710708618 lr:8.948182391459462e-05 tokens_per_second_per_gpu:3746.5550373221095 mfu:17.67313933925288 +Step 7698 | loss:0.857421338558197 lr:8.947916316903871e-05 tokens_per_second_per_gpu:3810.4552252422113 mfu:17.974567428168292 +Step 7699 | loss:1.064816951751709 lr:8.947650212655496e-05 tokens_per_second_per_gpu:4545.7981025210165 mfu:21.443305242724556 +Step 7700 | loss:1.0983375310897827 lr:8.947384078716339e-05 tokens_per_second_per_gpu:4477.913588107518 mfu:21.123082405943595 +Step 7701 | loss:1.1720243692398071 lr:8.947117915088404e-05 tokens_per_second_per_gpu:5319.307967642768 mfu:25.092083250895612 +Step 7702 | loss:1.0519676208496094 lr:8.946851721773692e-05 tokens_per_second_per_gpu:4659.7095548879715 mfu:21.98064499883736 +Step 7703 | loss:1.1047359704971313 lr:8.946585498774203e-05 tokens_per_second_per_gpu:5350.983437046966 mfu:25.241501844467578 +Step 7704 | loss:1.2210910320281982 lr:8.946319246091943e-05 tokens_per_second_per_gpu:3185.1401572775517 mfu:15.024849562826898 +Step 7705 | loss:0.9752272963523865 lr:8.946052963728912e-05 tokens_per_second_per_gpu:5395.381748178434 mfu:25.45093625320771 +Step 7706 | loss:1.1260837316513062 lr:8.945786651687111e-05 tokens_per_second_per_gpu:3095.2292186034497 mfu:14.600724324713171 +Step 7707 | loss:1.0399986505508423 lr:8.945520309968548e-05 tokens_per_second_per_gpu:5393.564021938402 mfu:25.44236172839741 +Step 7708 | loss:0.9237821698188782 lr:8.945253938575222e-05 tokens_per_second_per_gpu:5322.516788721993 mfu:25.107219807407652 +Step 7709 | loss:0.9149113297462463 lr:8.944987537509137e-05 tokens_per_second_per_gpu:5351.0831684045115 mfu:25.241972294297046 +Step 7710 | loss:1.1295956373214722 lr:8.944721106772298e-05 tokens_per_second_per_gpu:4973.887235355446 mfu:23.46267472184222 +Step 7711 | loss:0.954516589641571 lr:8.944454646366709e-05 tokens_per_second_per_gpu:5318.130385945474 mfu:25.086528397113323 +Step 7712 | loss:0.99750155210495 lr:8.944188156294371e-05 tokens_per_second_per_gpu:3669.5806754843875 mfu:17.310038141283968 +Step 7713 | loss:0.9872419238090515 lr:8.943921636557293e-05 tokens_per_second_per_gpu:5312.641792966091 mfu:25.06063776757905 +Step 7714 | loss:1.025901198387146 lr:8.943655087157476e-05 tokens_per_second_per_gpu:3760.812960493746 mfu:17.740396395506032 +Step 7715 | loss:1.1666877269744873 lr:8.943388508096926e-05 tokens_per_second_per_gpu:4550.438634793188 mfu:21.465195425208837 +Step 7716 | loss:1.0315557718276978 lr:8.943121899377649e-05 tokens_per_second_per_gpu:4387.026011038236 mfu:20.69435019788811 +Step 7717 | loss:1.023603081703186 lr:8.942855261001647e-05 tokens_per_second_per_gpu:4833.3360618275 mfu:22.799670855807904 +Step 7718 | loss:1.3505096435546875 lr:8.942588592970929e-05 tokens_per_second_per_gpu:5365.031458718344 mfu:25.307768759531122 +Step 7719 | loss:1.3548029661178589 lr:8.942321895287498e-05 tokens_per_second_per_gpu:5354.826614342553 mfu:25.259630767484527 +Step 7720 | loss:1.0904593467712402 lr:8.942055167953362e-05 tokens_per_second_per_gpu:5359.087213197469 mfu:25.279728739216292 +Step 7721 | loss:1.039280652999878 lr:8.941788410970525e-05 tokens_per_second_per_gpu:5338.714477792606 mfu:25.183627070364462 +Step 7722 | loss:1.0053595304489136 lr:8.941521624340995e-05 tokens_per_second_per_gpu:3976.7362693292603 mfu:18.758943483598493 +Step 7723 | loss:1.0276620388031006 lr:8.94125480806678e-05 tokens_per_second_per_gpu:5341.066943753302 mfu:25.19472405367315 +Step 7724 | loss:1.1636748313903809 lr:8.94098796214988e-05 tokens_per_second_per_gpu:4495.263214561729 mfu:21.204923554079624 +Step 7725 | loss:1.1627877950668335 lr:8.940721086592309e-05 tokens_per_second_per_gpu:5313.438058862468 mfu:25.06439388966923 +Step 7726 | loss:0.9439690709114075 lr:8.940454181396074e-05 tokens_per_second_per_gpu:5290.129572253887 mfu:24.954443781498657 +Step 7727 | loss:1.0208159685134888 lr:8.940187246563178e-05 tokens_per_second_per_gpu:4023.0753952338114 mfu:18.97753304676041 +Step 7728 | loss:1.0248987674713135 lr:8.939920282095631e-05 tokens_per_second_per_gpu:4365.416920163809 mfu:20.592416429342354 +Step 7729 | loss:1.1523494720458984 lr:8.939653287995442e-05 tokens_per_second_per_gpu:4013.4217952006898 mfu:18.931995368330167 +Step 7730 | loss:1.0404084920883179 lr:8.939386264264616e-05 tokens_per_second_per_gpu:5322.848929879045 mfu:25.10878657391459 +Step 7731 | loss:1.018977403640747 lr:8.939119210905165e-05 tokens_per_second_per_gpu:5140.735475877926 mfu:24.249726339632037 +Step 7732 | loss:1.0446362495422363 lr:8.938852127919095e-05 tokens_per_second_per_gpu:4606.9638819325 mfu:21.731834660166612 +Step 7733 | loss:1.2128676176071167 lr:8.938585015308416e-05 tokens_per_second_per_gpu:3540.485775236817 mfu:16.70107547095496 +Step 7734 | loss:1.0981471538543701 lr:8.938317873075136e-05 tokens_per_second_per_gpu:4281.143784595066 mfu:20.194885670385002 +Step 7735 | loss:1.1787807941436768 lr:8.938050701221265e-05 tokens_per_second_per_gpu:2935.0105785550923 mfu:13.844945663486264 +Step 7736 | loss:1.0956745147705078 lr:8.937783499748811e-05 tokens_per_second_per_gpu:4328.6079659084 mfu:20.418782311864398 +Step 7737 | loss:1.1292273998260498 lr:8.937516268659787e-05 tokens_per_second_per_gpu:4232.548915978001 mfu:19.965655383978767 +Step 7738 | loss:1.0674999952316284 lr:8.9372490079562e-05 tokens_per_second_per_gpu:4628.036009082366 mfu:21.831235479208154 +Step 7739 | loss:1.0987002849578857 lr:8.936981717640061e-05 tokens_per_second_per_gpu:3321.5085820929476 mfu:15.668122689533595 +Step 7740 | loss:1.1445375680923462 lr:8.936714397713379e-05 tokens_per_second_per_gpu:3541.003674215249 mfu:16.7035184887989 +Step 7741 | loss:1.06948721408844 lr:8.936447048178166e-05 tokens_per_second_per_gpu:5069.773341498493 mfu:23.914985844375657 +Step 7742 | loss:1.110504150390625 lr:8.936179669036433e-05 tokens_per_second_per_gpu:4540.045735935182 mfu:21.41617034808423 +Step 7743 | loss:1.1135708093643188 lr:8.935912260290191e-05 tokens_per_second_per_gpu:4574.321540679194 mfu:21.577855167117054 +Step 7744 | loss:1.2219629287719727 lr:8.93564482194145e-05 tokens_per_second_per_gpu:4133.2487913270115 mfu:19.497239753651744 +Step 7745 | loss:1.171474575996399 lr:8.935377353992222e-05 tokens_per_second_per_gpu:5321.129407057343 mfu:25.100675291383773 +Step 7746 | loss:1.0941970348358154 lr:8.935109856444519e-05 tokens_per_second_per_gpu:4067.3192985061223 mfu:19.186239087284203 +Step 7747 | loss:1.2112513780593872 lr:8.934842329300351e-05 tokens_per_second_per_gpu:5396.057637081964 mfu:25.4541245364842 +Step 7748 | loss:1.03138267993927 lr:8.934574772561732e-05 tokens_per_second_per_gpu:4865.169008218632 mfu:22.949832295195584 +Step 7749 | loss:1.030954360961914 lr:8.934307186230677e-05 tokens_per_second_per_gpu:4013.611086850231 mfu:18.932888289337537 +Step 7750 | loss:0.8524751663208008 lr:8.934039570309194e-05 tokens_per_second_per_gpu:4786.117054350943 mfu:22.57693074114756 +Step 7751 | loss:1.099646806716919 lr:8.933771924799297e-05 tokens_per_second_per_gpu:5382.440886699454 mfu:25.389892001672724 +Step 7752 | loss:1.0282065868377686 lr:8.933504249703e-05 tokens_per_second_per_gpu:4541.145354446052 mfu:21.42135743621427 +Step 7753 | loss:1.2450743913650513 lr:8.933236545022315e-05 tokens_per_second_per_gpu:3018.529112794414 mfu:14.238916839224471 +Step 7754 | loss:0.9780746102333069 lr:8.932968810759256e-05 tokens_per_second_per_gpu:4749.129626144872 mfu:22.402454731593533 +Step 7755 | loss:1.0404691696166992 lr:8.932701046915837e-05 tokens_per_second_per_gpu:5349.86871519837 mfu:25.236243511316594 +Step 7756 | loss:1.152927041053772 lr:8.932433253494072e-05 tokens_per_second_per_gpu:5166.97547507943 mfu:24.373504892871054 +Step 7757 | loss:1.1485874652862549 lr:8.932165430495973e-05 tokens_per_second_per_gpu:5060.514939702386 mfu:23.871312383457184 +Step 7758 | loss:1.2332676649093628 lr:8.931897577923558e-05 tokens_per_second_per_gpu:3103.460134606719 mfu:14.639550959839672 +Step 7759 | loss:1.2792861461639404 lr:8.931629695778841e-05 tokens_per_second_per_gpu:3328.0192001050364 mfu:15.698834385522597 +Step 7760 | loss:0.9891197085380554 lr:8.931361784063831e-05 tokens_per_second_per_gpu:3528.7550171098924 mfu:16.645739483396532 +Step 7761 | loss:0.9164798259735107 lr:8.931093842780551e-05 tokens_per_second_per_gpu:4791.1230517360445 mfu:22.60054488492431 +Step 7762 | loss:1.268013834953308 lr:8.930825871931013e-05 tokens_per_second_per_gpu:4007.457265564271 mfu:18.90385966438149 +Step 7763 | loss:1.0865681171417236 lr:8.930557871517229e-05 tokens_per_second_per_gpu:3944.994089676532 mfu:18.60921020640227 +Step 7764 | loss:1.286546230316162 lr:8.93028984154122e-05 tokens_per_second_per_gpu:5005.815964442561 mfu:23.613287984549064 +Step 7765 | loss:1.082092046737671 lr:8.930021782004998e-05 tokens_per_second_per_gpu:5315.912535403367 mfu:25.07606641770188 +Step 7766 | loss:1.1038893461227417 lr:8.929753692910581e-05 tokens_per_second_per_gpu:5307.148140249958 mfu:25.03472327040324 +Step 7767 | loss:0.9793994426727295 lr:8.929485574259986e-05 tokens_per_second_per_gpu:4081.757033718769 mfu:19.2543443476139 +Step 7768 | loss:1.1864597797393799 lr:8.929217426055228e-05 tokens_per_second_per_gpu:3843.762711486204 mfu:18.131684523624354 +Step 7769 | loss:0.981901228427887 lr:8.928949248298323e-05 tokens_per_second_per_gpu:5343.176214034483 mfu:25.204673841468093 +Step 7770 | loss:1.0882761478424072 lr:8.928681040991292e-05 tokens_per_second_per_gpu:5337.146626146502 mfu:25.176231246646715 +Step 7771 | loss:1.133741021156311 lr:8.928412804136148e-05 tokens_per_second_per_gpu:5190.142750457383 mfu:24.48278888357363 +Step 7772 | loss:1.109126091003418 lr:8.928144537734911e-05 tokens_per_second_per_gpu:3947.1839530791935 mfu:18.619540165701846 +Step 7773 | loss:1.1363097429275513 lr:8.927876241789597e-05 tokens_per_second_per_gpu:4340.5174518827325 mfu:20.47496138917316 +Step 7774 | loss:1.1369014978408813 lr:8.927607916302225e-05 tokens_per_second_per_gpu:4687.998713340493 mfu:22.114089785886044 +Step 7775 | loss:1.1807794570922852 lr:8.927339561274812e-05 tokens_per_second_per_gpu:3448.0060267993986 mfu:16.264832718903527 +Step 7776 | loss:1.3239703178405762 lr:8.927071176709378e-05 tokens_per_second_per_gpu:3351.0925507219295 mfu:15.807675317102039 +Step 7777 | loss:1.0834580659866333 lr:8.92680276260794e-05 tokens_per_second_per_gpu:4960.385177323144 mfu:23.3989832104152 +Step 7778 | loss:1.2100540399551392 lr:8.926534318972517e-05 tokens_per_second_per_gpu:4166.091259992141 mfu:19.65216328184813 +Step 7779 | loss:1.1086498498916626 lr:8.926265845805128e-05 tokens_per_second_per_gpu:5365.015983998562 mfu:25.30769576263765 +Step 7780 | loss:0.9281880855560303 lr:8.925997343107795e-05 tokens_per_second_per_gpu:4415.9136375922135 mfu:20.83061806108055 +Step 7781 | loss:1.038864016532898 lr:8.925728810882532e-05 tokens_per_second_per_gpu:5111.95363959054 mfu:24.113957507176792 +Step 7782 | loss:1.1846742630004883 lr:8.925460249131363e-05 tokens_per_second_per_gpu:5344.299625128322 mfu:25.209973163271428 +Step 7783 | loss:1.182353138923645 lr:8.925191657856307e-05 tokens_per_second_per_gpu:4251.086732442426 mfu:20.053101427119095 +Step 7784 | loss:1.0304397344589233 lr:8.924923037059383e-05 tokens_per_second_per_gpu:4401.847842470713 mfu:20.764267305620315 +Step 7785 | loss:1.1946096420288086 lr:8.924654386742613e-05 tokens_per_second_per_gpu:3688.2014214207948 mfu:17.397875376892863 +Step 7786 | loss:1.0621881484985352 lr:8.924385706908016e-05 tokens_per_second_per_gpu:5310.625293564938 mfu:25.051125595853534 +Step 7787 | loss:1.204806923866272 lr:8.924116997557613e-05 tokens_per_second_per_gpu:4597.973139523892 mfu:21.689423794246373 +Step 7788 | loss:1.037915825843811 lr:8.923848258693426e-05 tokens_per_second_per_gpu:5339.294140710772 mfu:25.186361439250142 +Step 7789 | loss:0.8732230067253113 lr:8.923579490317475e-05 tokens_per_second_per_gpu:5337.750826560574 mfu:25.17908136308711 +Step 7790 | loss:1.1943050622940063 lr:8.923310692431782e-05 tokens_per_second_per_gpu:3219.493804524453 mfu:15.186901578227154 +Step 7791 | loss:1.0924513339996338 lr:8.923041865038369e-05 tokens_per_second_per_gpu:4964.310494001581 mfu:23.417499598915473 +Step 7792 | loss:1.137145757675171 lr:8.922773008139256e-05 tokens_per_second_per_gpu:5049.632957970642 mfu:23.819980218971008 +Step 7793 | loss:1.111496925354004 lr:8.922504121736469e-05 tokens_per_second_per_gpu:5331.917882644502 mfu:25.151566371432146 +Step 7794 | loss:0.953950822353363 lr:8.922235205832028e-05 tokens_per_second_per_gpu:4762.515351421506 mfu:22.46559748156365 +Step 7795 | loss:1.1921329498291016 lr:8.921966260427954e-05 tokens_per_second_per_gpu:3881.1875887662136 mfu:18.30822405509666 +Step 7796 | loss:1.0924334526062012 lr:8.921697285526272e-05 tokens_per_second_per_gpu:3158.1963975174613 mfu:14.8977512509591 +Step 7797 | loss:1.1257964372634888 lr:8.921428281129003e-05 tokens_per_second_per_gpu:4893.45352254917 mfu:23.083255175128496 +Step 7798 | loss:1.1315431594848633 lr:8.921159247238174e-05 tokens_per_second_per_gpu:4736.154810780816 mfu:22.341250313789505 +Step 7799 | loss:1.031562089920044 lr:8.920890183855803e-05 tokens_per_second_per_gpu:3924.053491429029 mfu:18.510429831634312 +Step 7800 | loss:1.2378133535385132 lr:8.920621090983918e-05 tokens_per_second_per_gpu:2941.9710487231882 mfu:13.877779388847845 +Step 7801 | loss:0.8979234099388123 lr:8.920351968624543e-05 tokens_per_second_per_gpu:4990.459834977926 mfu:23.540850501838523 +Step 7802 | loss:1.0316067934036255 lr:8.9200828167797e-05 tokens_per_second_per_gpu:5360.650975972765 mfu:25.2871052750328 +Step 7803 | loss:1.1447032690048218 lr:8.919813635451413e-05 tokens_per_second_per_gpu:5316.554446133624 mfu:25.079094420137768 +Step 7804 | loss:1.0156114101409912 lr:8.919544424641707e-05 tokens_per_second_per_gpu:4610.830811995892 mfu:21.750075629041934 +Step 7805 | loss:1.386386752128601 lr:8.919275184352609e-05 tokens_per_second_per_gpu:3638.180890237366 mfu:17.16191999691797 +Step 7806 | loss:1.0712995529174805 lr:8.91900591458614e-05 tokens_per_second_per_gpu:5390.883494645164 mfu:25.42971722381009 +Step 7807 | loss:1.0001814365386963 lr:8.91873661534433e-05 tokens_per_second_per_gpu:5362.8711320036455 mfu:25.297578129828143 +Step 7808 | loss:1.1040101051330566 lr:8.9184672866292e-05 tokens_per_second_per_gpu:5253.685859713407 mfu:24.78253257150668 +Step 7809 | loss:0.8429078459739685 lr:8.918197928442777e-05 tokens_per_second_per_gpu:4261.736000750095 mfu:20.1033358426789 +Step 7810 | loss:1.0880711078643799 lr:8.917928540787089e-05 tokens_per_second_per_gpu:4401.757997171264 mfu:20.763843490013628 +Step 7811 | loss:1.099806785583496 lr:8.91765912366416e-05 tokens_per_second_per_gpu:4536.591565518259 mfu:21.39987643688536 +Step 7812 | loss:1.1109251976013184 lr:8.917389677076017e-05 tokens_per_second_per_gpu:4502.299477050788 mfu:21.238114804750687 +Step 7813 | loss:1.100503921508789 lr:8.917120201024687e-05 tokens_per_second_per_gpu:5255.668578873952 mfu:24.79188539607375 +Step 7814 | loss:1.0701037645339966 lr:8.916850695512195e-05 tokens_per_second_per_gpu:4655.909183556087 mfu:21.96271799885465 +Step 7815 | loss:1.1469327211380005 lr:8.916581160540569e-05 tokens_per_second_per_gpu:3358.4036056813043 mfu:15.842162810740012 +Step 7816 | loss:1.1932566165924072 lr:8.916311596111837e-05 tokens_per_second_per_gpu:4300.388244227914 mfu:20.285665069916536 +Step 7817 | loss:1.0420907735824585 lr:8.916042002228024e-05 tokens_per_second_per_gpu:4252.080929051284 mfu:20.057791222150513 +Step 7818 | loss:1.1028153896331787 lr:8.915772378891163e-05 tokens_per_second_per_gpu:3869.0778072579465 mfu:18.251100149580385 +Step 7819 | loss:1.1400285959243774 lr:8.915502726103276e-05 tokens_per_second_per_gpu:5324.002783950689 mfu:25.114229500438338 +Step 7820 | loss:1.0494859218597412 lr:8.915233043866393e-05 tokens_per_second_per_gpu:3628.418323015007 mfu:17.115868301664378 +Step 7821 | loss:0.8832619190216064 lr:8.914963332182544e-05 tokens_per_second_per_gpu:5352.491902595095 mfu:25.248617533828867 +Step 7822 | loss:1.0939249992370605 lr:8.914693591053755e-05 tokens_per_second_per_gpu:3853.116236683391 mfu:18.175806697855446 +Step 7823 | loss:1.1235557794570923 lr:8.914423820482056e-05 tokens_per_second_per_gpu:5312.653517639567 mfu:25.060693074864272 +Step 7824 | loss:1.0935312509536743 lr:8.914154020469477e-05 tokens_per_second_per_gpu:5152.450526827293 mfu:24.304988233754006 +Step 7825 | loss:1.040705680847168 lr:8.913884191018046e-05 tokens_per_second_per_gpu:5261.330582361041 mfu:24.818594032559755 +Step 7826 | loss:1.1106467247009277 lr:8.913614332129791e-05 tokens_per_second_per_gpu:5331.101192994766 mfu:25.14771390701272 +Step 7827 | loss:1.1037908792495728 lr:8.913344443806746e-05 tokens_per_second_per_gpu:4869.949457189782 mfu:22.972382488621193 +Step 7828 | loss:1.11761474609375 lr:8.913074526050936e-05 tokens_per_second_per_gpu:3310.6168679009334 mfu:15.6167446153718 +Step 7829 | loss:1.1563177108764648 lr:8.912804578864393e-05 tokens_per_second_per_gpu:3826.626172325554 mfu:18.050848544608492 +Step 7830 | loss:1.1506121158599854 lr:8.91253460224915e-05 tokens_per_second_per_gpu:5136.505365905152 mfu:24.22977218915928 +Step 7831 | loss:1.0682047605514526 lr:8.912264596207233e-05 tokens_per_second_per_gpu:4883.989208445725 mfu:23.038610390724052 +Step 7832 | loss:1.1390297412872314 lr:8.911994560740675e-05 tokens_per_second_per_gpu:3795.3724704961305 mfu:17.903419500647708 +Step 7833 | loss:1.1485081911087036 lr:8.911724495851505e-05 tokens_per_second_per_gpu:5332.455045692685 mfu:25.154100261179895 +Step 7834 | loss:1.0358829498291016 lr:8.91145440154176e-05 tokens_per_second_per_gpu:5303.369502117542 mfu:25.016898789630492 +Step 7835 | loss:1.0843298435211182 lr:8.911184277813464e-05 tokens_per_second_per_gpu:3928.602155549418 mfu:18.531886656372315 +Step 7836 | loss:1.0911945104599 lr:8.910914124668652e-05 tokens_per_second_per_gpu:4785.892944799501 mfu:22.575873578156372 +Step 7837 | loss:1.1170153617858887 lr:8.910643942109358e-05 tokens_per_second_per_gpu:4685.633301593181 mfu:22.10293173509311 +Step 7838 | loss:0.950791597366333 lr:8.910373730137611e-05 tokens_per_second_per_gpu:5359.732606528536 mfu:25.28277316967439 +Step 7839 | loss:1.1559282541275024 lr:8.910103488755443e-05 tokens_per_second_per_gpu:4614.145665751896 mfu:21.765712359781375 +Step 7840 | loss:1.16227388381958 lr:8.909833217964889e-05 tokens_per_second_per_gpu:2699.755145481382 mfu:12.735205646961719 +Step 7841 | loss:1.195900797843933 lr:8.90956291776798e-05 tokens_per_second_per_gpu:5392.512373632348 mfu:25.437400923908047 +Step 7842 | loss:1.097204566001892 lr:8.90929258816675e-05 tokens_per_second_per_gpu:4020.392544314291 mfu:18.964877581231462 +Step 7843 | loss:1.1279114484786987 lr:8.90902222916323e-05 tokens_per_second_per_gpu:5320.201204862346 mfu:25.096296803262344 +Step 7844 | loss:1.1173609495162964 lr:8.908751840759457e-05 tokens_per_second_per_gpu:4546.096287706563 mfu:21.444711833120415 +Step 7845 | loss:1.1503881216049194 lr:8.908481422957462e-05 tokens_per_second_per_gpu:4571.863231686945 mfu:21.566258904169853 +Step 7846 | loss:1.2104371786117554 lr:8.90821097575928e-05 tokens_per_second_per_gpu:3421.060477334997 mfu:16.13772596469514 +Step 7847 | loss:1.059737205505371 lr:8.907940499166944e-05 tokens_per_second_per_gpu:4894.451883034127 mfu:23.0879646118739 +Step 7848 | loss:1.1059942245483398 lr:8.907669993182489e-05 tokens_per_second_per_gpu:4373.365228879195 mfu:20.629909957674137 +Step 7849 | loss:0.9993776679039001 lr:8.907399457807947e-05 tokens_per_second_per_gpu:3479.5202135770905 mfu:16.41349051480861 +Step 7850 | loss:0.9995758533477783 lr:8.907128893045359e-05 tokens_per_second_per_gpu:3934.913812791737 mfu:18.561659820464104 +Step 7851 | loss:0.9746866822242737 lr:8.906858298896755e-05 tokens_per_second_per_gpu:4188.644625202322 mfu:19.758551353545453 +Step 7852 | loss:1.132820725440979 lr:8.90658767536417e-05 tokens_per_second_per_gpu:3184.509421428655 mfu:15.021874274213019 +Step 7853 | loss:1.0075081586837769 lr:8.906317022449639e-05 tokens_per_second_per_gpu:4697.384959764591 mfu:22.158366311727292 +Step 7854 | loss:1.1221126317977905 lr:8.906046340155203e-05 tokens_per_second_per_gpu:5201.230310729829 mfu:24.53509079711161 +Step 7855 | loss:1.0943658351898193 lr:8.905775628482892e-05 tokens_per_second_per_gpu:3926.6769621457943 mfu:18.52280519061556 +Step 7856 | loss:1.0836118459701538 lr:8.905504887434745e-05 tokens_per_second_per_gpu:5344.590219687862 mfu:25.211343947389327 +Step 7857 | loss:1.0917822122573853 lr:8.905234117012796e-05 tokens_per_second_per_gpu:4136.282740503335 mfu:19.51155140956157 +Step 7858 | loss:1.0424394607543945 lr:8.904963317219083e-05 tokens_per_second_per_gpu:4250.878748702016 mfu:20.052120332329896 +Step 7859 | loss:1.0387797355651855 lr:8.904692488055645e-05 tokens_per_second_per_gpu:4567.987114345134 mfu:21.54797459733472 +Step 7860 | loss:1.0410526990890503 lr:8.904421629524515e-05 tokens_per_second_per_gpu:3395.697218326168 mfu:16.018083144532277 +Step 7861 | loss:1.1010462045669556 lr:8.904150741627732e-05 tokens_per_second_per_gpu:4845.371176694051 mfu:22.856442546035762 +Step 7862 | loss:1.257897973060608 lr:8.903879824367332e-05 tokens_per_second_per_gpu:3746.4357346082793 mfu:17.672576567996604 +Step 7863 | loss:1.116545557975769 lr:8.903608877745355e-05 tokens_per_second_per_gpu:4680.7649806384325 mfu:22.079967034528355 +Step 7864 | loss:1.1704463958740234 lr:8.903337901763836e-05 tokens_per_second_per_gpu:3568.823448396196 mfu:16.83474911015335 +Step 7865 | loss:1.2117916345596313 lr:8.903066896424817e-05 tokens_per_second_per_gpu:4099.624082658505 mfu:19.338626265895527 +Step 7866 | loss:1.1432445049285889 lr:8.902795861730331e-05 tokens_per_second_per_gpu:4524.094134429899 mfu:21.34092392215995 +Step 7867 | loss:0.9985201358795166 lr:8.902524797682421e-05 tokens_per_second_per_gpu:4976.272318796246 mfu:23.473925567369665 +Step 7868 | loss:1.1211521625518799 lr:8.902253704283123e-05 tokens_per_second_per_gpu:3637.5266934295387 mfu:17.158834038957963 +Step 7869 | loss:1.0179927349090576 lr:8.901982581534479e-05 tokens_per_second_per_gpu:5357.645557398617 mfu:25.272928202841257 +Step 7870 | loss:1.1370817422866821 lr:8.901711429438524e-05 tokens_per_second_per_gpu:5374.149729895149 mfu:25.350781200409596 +Step 7871 | loss:1.040130853652954 lr:8.901440247997301e-05 tokens_per_second_per_gpu:5314.881120577008 mfu:25.071201057988432 +Step 7872 | loss:1.0947908163070679 lr:8.901169037212846e-05 tokens_per_second_per_gpu:4776.904000082653 mfu:22.533471192254883 +Step 7873 | loss:1.175646185874939 lr:8.900897797087204e-05 tokens_per_second_per_gpu:5346.688767300418 mfu:25.221243154526896 +Step 7874 | loss:1.1415964365005493 lr:8.900626527622409e-05 tokens_per_second_per_gpu:5313.935472382111 mfu:25.06674027410888 +Step 7875 | loss:0.9545013308525085 lr:8.900355228820507e-05 tokens_per_second_per_gpu:4616.115259243261 mfu:21.775003268327445 +Step 7876 | loss:1.1937549114227295 lr:8.900083900683533e-05 tokens_per_second_per_gpu:4236.097118916859 mfu:19.982392862627208 +Step 7877 | loss:1.1793532371520996 lr:8.899812543213532e-05 tokens_per_second_per_gpu:5327.028330778364 mfu:25.128501521036 +Step 7878 | loss:1.122210144996643 lr:8.89954115641254e-05 tokens_per_second_per_gpu:4850.2310049110065 mfu:22.879367184907334 +Step 7879 | loss:1.1129270792007446 lr:8.899269740282606e-05 tokens_per_second_per_gpu:4914.916723595305 mfu:23.18450075646283 +Step 7880 | loss:1.059735894203186 lr:8.898998294825764e-05 tokens_per_second_per_gpu:5215.465596488887 mfu:24.602241068827283 +Step 7881 | loss:1.1663329601287842 lr:8.898726820044058e-05 tokens_per_second_per_gpu:5323.18679515248 mfu:25.11038034205514 +Step 7882 | loss:1.150267481803894 lr:8.898455315939531e-05 tokens_per_second_per_gpu:4727.072997494136 mfu:22.298409850999107 +Step 7883 | loss:1.0057700872421265 lr:8.898183782514224e-05 tokens_per_second_per_gpu:5350.412907260911 mfu:25.23881055812434 +Step 7884 | loss:0.9684438705444336 lr:8.897912219770182e-05 tokens_per_second_per_gpu:5337.428246091957 mfu:25.177559695978744 +Step 7885 | loss:1.0189130306243896 lr:8.897640627709442e-05 tokens_per_second_per_gpu:5363.4157488109895 mfu:25.300147180229594 +Step 7886 | loss:1.0802704095840454 lr:8.897369006334048e-05 tokens_per_second_per_gpu:5091.448254272437 mfu:24.017230106051805 +Step 7887 | loss:1.1144990921020508 lr:8.897097355646048e-05 tokens_per_second_per_gpu:5330.107374018705 mfu:25.143025893339637 +Step 7888 | loss:1.2395167350769043 lr:8.89682567564748e-05 tokens_per_second_per_gpu:3240.8012865184774 mfu:15.287412606224269 +Step 7889 | loss:1.1359360218048096 lr:8.896553966340388e-05 tokens_per_second_per_gpu:4366.487172645486 mfu:20.597464992902324 +Step 7890 | loss:1.0066637992858887 lr:8.896282227726816e-05 tokens_per_second_per_gpu:3885.4874296316257 mfu:18.328507138087307 +Step 7891 | loss:1.0598660707473755 lr:8.89601045980881e-05 tokens_per_second_per_gpu:4059.4791937265404 mfu:19.149255975379106 +Step 7892 | loss:1.1414010524749756 lr:8.895738662588411e-05 tokens_per_second_per_gpu:3535.803440761491 mfu:16.678988100346206 +Step 7893 | loss:1.0474421977996826 lr:8.895466836067665e-05 tokens_per_second_per_gpu:5357.775890587572 mfu:25.273543006730694 +Step 7894 | loss:1.1845695972442627 lr:8.895194980248615e-05 tokens_per_second_per_gpu:4014.581435855783 mfu:18.93746559115534 +Step 7895 | loss:1.1297407150268555 lr:8.894923095133308e-05 tokens_per_second_per_gpu:5343.36077774503 mfu:25.205544459980562 +Step 7896 | loss:1.0499329566955566 lr:8.894651180723788e-05 tokens_per_second_per_gpu:5240.7927744951385 mfu:24.721713688744934 +Step 7897 | loss:1.0588771104812622 lr:8.894379237022097e-05 tokens_per_second_per_gpu:4223.688908168144 mfu:19.92386121546646 +Step 7898 | loss:1.0866867303848267 lr:8.894107264030283e-05 tokens_per_second_per_gpu:4587.719400046009 mfu:21.641055155660844 +Step 7899 | loss:1.0370467901229858 lr:8.893835261750394e-05 tokens_per_second_per_gpu:4469.330203706201 mfu:21.08259311724619 +Step 7900 | loss:1.0560824871063232 lr:8.89356323018447e-05 tokens_per_second_per_gpu:4269.533670826159 mfu:20.140118782847402 +Step 7901 | loss:1.196705937385559 lr:8.893291169334564e-05 tokens_per_second_per_gpu:4252.473783176104 mfu:20.05964438208511 +Step 7902 | loss:1.181380033493042 lr:8.893019079202715e-05 tokens_per_second_per_gpu:5328.4635751261885 mfu:25.135271813502627 +Step 7903 | loss:1.0922783613204956 lr:8.892746959790974e-05 tokens_per_second_per_gpu:4826.04350819516 mfu:22.765270636086203 +Step 7904 | loss:1.1693793535232544 lr:8.892474811101386e-05 tokens_per_second_per_gpu:3907.1877392195615 mfu:18.430871200868204 +Step 7905 | loss:1.0790644884109497 lr:8.892202633135997e-05 tokens_per_second_per_gpu:5313.536326568779 mfu:25.0648574352061 +Step 7906 | loss:1.2779273986816406 lr:8.891930425896857e-05 tokens_per_second_per_gpu:4026.0346087948733 mfu:18.99149216202183 +Step 7907 | loss:0.9862499833106995 lr:8.891658189386012e-05 tokens_per_second_per_gpu:3939.30176851212 mfu:18.582358556259596 +Step 7908 | loss:1.0170961618423462 lr:8.891385923605509e-05 tokens_per_second_per_gpu:5393.354222307073 mfu:25.44137206774092 +Step 7909 | loss:1.016536831855774 lr:8.891113628557395e-05 tokens_per_second_per_gpu:4505.19965612813 mfu:21.251795444280084 +Step 7910 | loss:1.0797713994979858 lr:8.89084130424372e-05 tokens_per_second_per_gpu:3924.1299594930497 mfu:18.5107905445391 +Step 7911 | loss:1.1186023950576782 lr:8.89056895066653e-05 tokens_per_second_per_gpu:4954.274350613775 mfu:23.37015739821256 +Step 7912 | loss:1.0441442728042603 lr:8.890296567827874e-05 tokens_per_second_per_gpu:4111.655744899852 mfu:19.395381669501212 +Step 7913 | loss:1.1894798278808594 lr:8.890024155729803e-05 tokens_per_second_per_gpu:3954.3148667480204 mfu:18.65317790213703 +Step 7914 | loss:1.0612306594848633 lr:8.889751714374363e-05 tokens_per_second_per_gpu:4022.886256270958 mfu:18.97664084600219 +Step 7915 | loss:1.0249965190887451 lr:8.889479243763604e-05 tokens_per_second_per_gpu:4696.017513784589 mfu:22.151915835729277 +Step 7916 | loss:1.1811093091964722 lr:8.889206743899574e-05 tokens_per_second_per_gpu:4995.5014706243155 mfu:23.564632757374284 +Step 7917 | loss:1.0011738538742065 lr:8.888934214784326e-05 tokens_per_second_per_gpu:4529.733305102327 mfu:21.36752485236352 +Step 7918 | loss:1.137878179550171 lr:8.888661656419907e-05 tokens_per_second_per_gpu:4355.780980391861 mfu:20.546962057377222 +Step 7919 | loss:1.0338268280029297 lr:8.888389068808367e-05 tokens_per_second_per_gpu:5296.692182845541 mfu:24.985400735356315 +Step 7920 | loss:1.0196770429611206 lr:8.888116451951756e-05 tokens_per_second_per_gpu:3665.928497873966 mfu:17.292810196369853 +Step 7921 | loss:1.1152983903884888 lr:8.887843805852126e-05 tokens_per_second_per_gpu:3965.4137889778262 mfu:18.705533411966428 +Step 7922 | loss:1.123892903327942 lr:8.887571130511528e-05 tokens_per_second_per_gpu:4288.0620718029995 mfu:20.227520411526342 +Step 7923 | loss:1.0419427156448364 lr:8.88729842593201e-05 tokens_per_second_per_gpu:4894.616086678539 mfu:23.088739188480627 +Step 7924 | loss:1.1593914031982422 lr:8.887025692115622e-05 tokens_per_second_per_gpu:5111.537429476352 mfu:24.111994173055404 +Step 7925 | loss:1.1195333003997803 lr:8.886752929064422e-05 tokens_per_second_per_gpu:4428.66092332227 mfu:20.89074918278044 +Step 7926 | loss:1.036163091659546 lr:8.886480136780455e-05 tokens_per_second_per_gpu:4912.728108297808 mfu:23.17417668468826 +Step 7927 | loss:1.179380178451538 lr:8.886207315265775e-05 tokens_per_second_per_gpu:3593.2182706324784 mfu:16.94982364882776 +Step 7928 | loss:0.9931125044822693 lr:8.885934464522436e-05 tokens_per_second_per_gpu:5218.548099444816 mfu:24.61678176119964 +Step 7929 | loss:1.1514455080032349 lr:8.885661584552485e-05 tokens_per_second_per_gpu:5354.510064917431 mfu:25.258137549837322 +Step 7930 | loss:0.9750998616218567 lr:8.88538867535798e-05 tokens_per_second_per_gpu:5349.9588074607145 mfu:25.2366684918894 +Step 7931 | loss:1.002795696258545 lr:8.885115736940969e-05 tokens_per_second_per_gpu:5316.564841644861 mfu:25.079143457537935 +Step 7932 | loss:1.1187849044799805 lr:8.884842769303507e-05 tokens_per_second_per_gpu:5332.208771044889 mfu:25.15293854164745 +Step 7933 | loss:1.122623324394226 lr:8.884569772447648e-05 tokens_per_second_per_gpu:3967.238089456656 mfu:18.71413894858282 +Step 7934 | loss:1.0576441287994385 lr:8.884296746375443e-05 tokens_per_second_per_gpu:5304.076917793242 mfu:25.020235790070327 +Step 7935 | loss:1.2186822891235352 lr:8.884023691088948e-05 tokens_per_second_per_gpu:5009.528554440145 mfu:23.630800905000836 +Step 7936 | loss:1.1386749744415283 lr:8.883750606590213e-05 tokens_per_second_per_gpu:4905.018295715919 mfu:23.137808183309815 +Step 7937 | loss:1.0704362392425537 lr:8.883477492881296e-05 tokens_per_second_per_gpu:4899.291778703727 mfu:23.110795225518757 +Step 7938 | loss:1.079153060913086 lr:8.883204349964249e-05 tokens_per_second_per_gpu:5321.705638912893 mfu:25.103393475361695 +Step 7939 | loss:1.1857942342758179 lr:8.882931177841126e-05 tokens_per_second_per_gpu:4701.84961119955 mfu:22.179426834294073 +Step 7940 | loss:1.0445729494094849 lr:8.882657976513982e-05 tokens_per_second_per_gpu:4585.538642232695 mfu:21.630768148979794 +Step 7941 | loss:1.0982354879379272 lr:8.882384745984873e-05 tokens_per_second_per_gpu:4208.664801974185 mfu:19.852989943172712 +Step 7942 | loss:1.2767707109451294 lr:8.882111486255852e-05 tokens_per_second_per_gpu:4832.103829571166 mfu:22.793858206014274 +Step 7943 | loss:1.0879571437835693 lr:8.881838197328975e-05 tokens_per_second_per_gpu:4074.8542199122935 mfu:19.22178259714727 +Step 7944 | loss:1.1330963373184204 lr:8.881564879206298e-05 tokens_per_second_per_gpu:3466.7002768663015 mfu:16.35301668603707 +Step 7945 | loss:0.9543188214302063 lr:8.881291531889878e-05 tokens_per_second_per_gpu:4603.357556771797 mfu:21.714823009081055 +Step 7946 | loss:1.1018801927566528 lr:8.881018155381766e-05 tokens_per_second_per_gpu:4701.970683690112 mfu:22.179997953889153 +Step 7947 | loss:1.1237844228744507 lr:8.880744749684022e-05 tokens_per_second_per_gpu:4657.227564994584 mfu:21.96893702903911 +Step 7948 | loss:1.052039384841919 lr:8.880471314798702e-05 tokens_per_second_per_gpu:5356.003919630304 mfu:25.265184317395843 +Step 7949 | loss:1.2272675037384033 lr:8.880197850727863e-05 tokens_per_second_per_gpu:4886.146878825112 mfu:23.04878849004015 +Step 7950 | loss:1.1059337854385376 lr:8.879924357473559e-05 tokens_per_second_per_gpu:4366.503352609637 mfu:20.597541316553816 +Step 7951 | loss:1.069696068763733 lr:8.879650835037851e-05 tokens_per_second_per_gpu:4663.068218278218 mfu:21.996488387096218 +Step 7952 | loss:1.0143173933029175 lr:8.879377283422792e-05 tokens_per_second_per_gpu:4303.3159331487 mfu:20.299475478071127 +Step 7953 | loss:0.9716662764549255 lr:8.879103702630442e-05 tokens_per_second_per_gpu:5330.858508367407 mfu:25.146569122219198 +Step 7954 | loss:0.9844120144844055 lr:8.87883009266286e-05 tokens_per_second_per_gpu:5129.724760472395 mfu:24.197786916442386 +Step 7955 | loss:1.1532984972000122 lr:8.8785564535221e-05 tokens_per_second_per_gpu:5334.322509107198 mfu:25.162909404728264 +Step 7956 | loss:0.9750901460647583 lr:8.878282785210221e-05 tokens_per_second_per_gpu:4967.089219772692 mfu:23.430607322477307 +Step 7957 | loss:1.196700930595398 lr:8.878009087729285e-05 tokens_per_second_per_gpu:5098.857602742797 mfu:24.0521812669517 +Step 7958 | loss:1.103031873703003 lr:8.877735361081345e-05 tokens_per_second_per_gpu:3730.495929556031 mfu:17.597385787953968 +Step 7959 | loss:1.1095956563949585 lr:8.877461605268464e-05 tokens_per_second_per_gpu:4152.239342672649 mfu:19.58682142447129 +Step 7960 | loss:1.1190252304077148 lr:8.8771878202927e-05 tokens_per_second_per_gpu:4141.8475405859135 mfu:19.537801521003892 +Step 7961 | loss:1.131853699684143 lr:8.876914006156112e-05 tokens_per_second_per_gpu:4897.930332847385 mfu:23.104373053127777 +Step 7962 | loss:1.1945083141326904 lr:8.876640162860759e-05 tokens_per_second_per_gpu:4743.401141131478 mfu:22.375432490405025 +Step 7963 | loss:1.0601426362991333 lr:8.8763662904087e-05 tokens_per_second_per_gpu:4531.312165896292 mfu:21.374972608110085 +Step 7964 | loss:1.1576217412948608 lr:8.876092388801996e-05 tokens_per_second_per_gpu:3529.303414977644 mfu:16.648326369705043 +Step 7965 | loss:1.1549651622772217 lr:8.875818458042704e-05 tokens_per_second_per_gpu:3904.7275358337392 mfu:18.419266001743296 +Step 7966 | loss:1.238975167274475 lr:8.87554449813289e-05 tokens_per_second_per_gpu:4894.010821090284 mfu:23.08588404743187 +Step 7967 | loss:1.1966667175292969 lr:8.875270509074609e-05 tokens_per_second_per_gpu:3652.0163440023657 mfu:17.2271841928993 +Step 7968 | loss:1.1298980712890625 lr:8.874996490869925e-05 tokens_per_second_per_gpu:4756.31355770299 mfu:22.436342562458286 +Step 7969 | loss:0.99672931432724 lr:8.874722443520899e-05 tokens_per_second_per_gpu:5334.513006446166 mfu:25.163808013178375 +Step 7970 | loss:0.9768001437187195 lr:8.87444836702959e-05 tokens_per_second_per_gpu:5346.982976083792 mfu:25.222630987555263 +Step 7971 | loss:1.039735198020935 lr:8.87417426139806e-05 tokens_per_second_per_gpu:4772.659465851557 mfu:22.51344900009459 +Step 7972 | loss:1.0370560884475708 lr:8.873900126628371e-05 tokens_per_second_per_gpu:4629.913655862101 mfu:21.840092659427743 +Step 7973 | loss:1.2644559144973755 lr:8.873625962722587e-05 tokens_per_second_per_gpu:5195.800832109274 mfu:24.509479020093874 +Step 7974 | loss:1.1491615772247314 lr:8.873351769682765e-05 tokens_per_second_per_gpu:5176.613859860158 mfu:24.418970798359766 +Step 7975 | loss:0.9957515001296997 lr:8.873077547510973e-05 tokens_per_second_per_gpu:4670.628878470535 mfu:22.032153311205022 +Step 7976 | loss:1.0156828165054321 lr:8.872803296209267e-05 tokens_per_second_per_gpu:3750.775396815617 mfu:17.693047495051335 +Step 7977 | loss:1.1167805194854736 lr:8.872529015779716e-05 tokens_per_second_per_gpu:5254.232921023603 mfu:24.78511315304516 +Step 7978 | loss:1.1412084102630615 lr:8.87225470622438e-05 tokens_per_second_per_gpu:5344.347680587554 mfu:25.21019984907174 +Step 7979 | loss:1.1123095750808716 lr:8.87198036754532e-05 tokens_per_second_per_gpu:4395.72379214392 mfu:20.735379115358104 +Step 7980 | loss:1.155502200126648 lr:8.871705999744604e-05 tokens_per_second_per_gpu:5365.471131528174 mfu:25.309842771191004 +Step 7981 | loss:1.1205816268920898 lr:8.871431602824293e-05 tokens_per_second_per_gpu:5363.109644326162 mfu:25.298703233147855 +Step 7982 | loss:1.068301796913147 lr:8.87115717678645e-05 tokens_per_second_per_gpu:3797.8658299070626 mfu:17.91518110240006 +Step 7983 | loss:0.9046038389205933 lr:8.870882721633141e-05 tokens_per_second_per_gpu:4372.55158937701 mfu:20.626071881321117 +Step 7984 | loss:1.217273473739624 lr:8.870608237366427e-05 tokens_per_second_per_gpu:5367.689285245224 mfu:25.320306180730377 +Step 7985 | loss:0.9883871674537659 lr:8.870333723988376e-05 tokens_per_second_per_gpu:4650.962737656673 mfu:21.93938476959642 +Step 7986 | loss:1.0536863803863525 lr:8.870059181501052e-05 tokens_per_second_per_gpu:5114.215599698677 mfu:24.124627558937068 +Step 7987 | loss:0.999680757522583 lr:8.869784609906517e-05 tokens_per_second_per_gpu:4190.259181011167 mfu:19.7661674887674 +Step 7988 | loss:0.9939998388290405 lr:8.86951000920684e-05 tokens_per_second_per_gpu:4924.26913309968 mfu:23.228617667780572 +Step 7989 | loss:1.1381975412368774 lr:8.869235379404084e-05 tokens_per_second_per_gpu:3750.483803149814 mfu:17.691671997978734 +Step 7990 | loss:0.9962259531021118 lr:8.868960720500315e-05 tokens_per_second_per_gpu:3735.136664376129 mfu:17.619276925891377 +Step 7991 | loss:1.1230584383010864 lr:8.868686032497597e-05 tokens_per_second_per_gpu:5327.618048092606 mfu:25.13128331822343 +Step 7992 | loss:1.0780390501022339 lr:8.868411315398e-05 tokens_per_second_per_gpu:4817.872615999329 mfu:22.726727143500504 +Step 7993 | loss:1.1825058460235596 lr:8.868136569203587e-05 tokens_per_second_per_gpu:3153.909367032084 mfu:14.87752859671675 +Step 7994 | loss:1.2785009145736694 lr:8.867861793916425e-05 tokens_per_second_per_gpu:4416.954019530434 mfu:20.8355257201906 +Step 7995 | loss:1.126320242881775 lr:8.867586989538581e-05 tokens_per_second_per_gpu:4396.324829886694 mfu:20.738214312938382 +Step 7996 | loss:1.1546450853347778 lr:8.867312156072123e-05 tokens_per_second_per_gpu:4177.090343668183 mfu:19.70404783618521 +Step 7997 | loss:1.1216741800308228 lr:8.867037293519115e-05 tokens_per_second_per_gpu:5131.827681721698 mfu:24.207706754770708 +Step 7998 | loss:1.0541560649871826 lr:8.866762401881627e-05 tokens_per_second_per_gpu:4470.5767002410275 mfu:21.08847305407521 +Step 7999 | loss:1.0853689908981323 lr:8.866487481161725e-05 tokens_per_second_per_gpu:4361.867889061344 mfu:20.57567504410488 +Step 8000 | loss:1.0643881559371948 lr:8.866212531361479e-05 tokens_per_second_per_gpu:4056.895663010664 mfu:19.13706902019652 +Step 8001 | loss:1.069576621055603 lr:8.865937552482953e-05 tokens_per_second_per_gpu:5343.160306876647 mfu:25.204598804690544 +Step 8002 | loss:1.018193006515503 lr:8.86566254452822e-05 tokens_per_second_per_gpu:4470.158676698551 mfu:21.086501165703176 +Step 8003 | loss:1.1785906553268433 lr:8.865387507499343e-05 tokens_per_second_per_gpu:3837.2863753477086 mfu:18.101134541082185 +Step 8004 | loss:1.1502739191055298 lr:8.865112441398393e-05 tokens_per_second_per_gpu:5194.24003375042 mfu:24.502116467935497 +Step 8005 | loss:0.9935035109519958 lr:8.864837346227442e-05 tokens_per_second_per_gpu:3623.5932213353076 mfu:17.093107473794 +Step 8006 | loss:1.0990945100784302 lr:8.864562221988554e-05 tokens_per_second_per_gpu:4359.699640396259 mfu:20.56544704521031 +Step 8007 | loss:1.099465012550354 lr:8.864287068683801e-05 tokens_per_second_per_gpu:5326.668005615193 mfu:25.12680180576351 +Step 8008 | loss:1.1183459758758545 lr:8.864011886315252e-05 tokens_per_second_per_gpu:4489.972077307812 mfu:21.17996435689179 +Step 8009 | loss:1.2013038396835327 lr:8.863736674884977e-05 tokens_per_second_per_gpu:4130.935814317323 mfu:19.486329046462476 +Step 8010 | loss:1.1320128440856934 lr:8.863461434395045e-05 tokens_per_second_per_gpu:5118.60948070015 mfu:24.145354245294417 +Step 8011 | loss:1.1431318521499634 lr:8.863186164847525e-05 tokens_per_second_per_gpu:4779.253661420782 mfu:22.544554945680403 +Step 8012 | loss:1.098292350769043 lr:8.862910866244491e-05 tokens_per_second_per_gpu:5264.985203386817 mfu:24.835833503480984 +Step 8013 | loss:1.1542611122131348 lr:8.862635538588011e-05 tokens_per_second_per_gpu:4293.3404794734415 mfu:20.25241956109703 +Step 8014 | loss:1.1797362565994263 lr:8.862360181880156e-05 tokens_per_second_per_gpu:5360.101085226027 mfu:25.284511346559402 +Step 8015 | loss:0.9791663885116577 lr:8.862084796122998e-05 tokens_per_second_per_gpu:4049.477669644772 mfu:19.102077079849433 +Step 8016 | loss:0.9943180084228516 lr:8.861809381318606e-05 tokens_per_second_per_gpu:5368.194770664375 mfu:25.32269064169722 +Step 8017 | loss:1.1286605596542358 lr:8.861533937469056e-05 tokens_per_second_per_gpu:3757.3868776469963 mfu:17.724234978167903 +Step 8018 | loss:0.8969491720199585 lr:8.861258464576413e-05 tokens_per_second_per_gpu:4847.904874371666 mfu:22.86839443027447 +Step 8019 | loss:1.0258632898330688 lr:8.860982962642754e-05 tokens_per_second_per_gpu:4533.391279169028 mfu:21.384780140150884 +Step 8020 | loss:1.037096381187439 lr:8.860707431670149e-05 tokens_per_second_per_gpu:4861.028770048106 mfu:22.930302085347734 +Step 8021 | loss:1.1852918863296509 lr:8.86043187166067e-05 tokens_per_second_per_gpu:4407.49948560207 mfu:20.79092707054068 +Step 8022 | loss:1.1248432397842407 lr:8.860156282616392e-05 tokens_per_second_per_gpu:3708.7693436412314 mfu:17.494897775256053 +Step 8023 | loss:1.1551593542099 lr:8.859880664539384e-05 tokens_per_second_per_gpu:4168.6931461301965 mfu:19.66443682268889 +Step 8024 | loss:1.1260255575180054 lr:8.859605017431722e-05 tokens_per_second_per_gpu:3321.21952840882 mfu:15.666759173987924 +Step 8025 | loss:1.178269863128662 lr:8.859329341295479e-05 tokens_per_second_per_gpu:3604.353282439383 mfu:17.002349399349846 +Step 8026 | loss:1.0366013050079346 lr:8.859053636132728e-05 tokens_per_second_per_gpu:4041.4796239634898 mfu:19.06434893376352 +Step 8027 | loss:1.1588072776794434 lr:8.858777901945541e-05 tokens_per_second_per_gpu:5321.0644198183 mfu:25.10036873548924 +Step 8028 | loss:1.0007541179656982 lr:8.858502138735992e-05 tokens_per_second_per_gpu:5326.852751887117 mfu:25.12767328544914 +Step 8029 | loss:1.1286392211914062 lr:8.858226346506158e-05 tokens_per_second_per_gpu:5316.7334265216605 mfu:25.07993870116534 +Step 8030 | loss:1.1250885725021362 lr:8.85795052525811e-05 tokens_per_second_per_gpu:5076.4300477496445 mfu:23.946386663513536 +Step 8031 | loss:1.1123095750808716 lr:8.857674674993926e-05 tokens_per_second_per_gpu:5325.001777446031 mfu:25.118941923201735 +Step 8032 | loss:1.2167129516601562 lr:8.857398795715675e-05 tokens_per_second_per_gpu:5312.039104078979 mfu:25.057794781269212 +Step 8033 | loss:1.1106570959091187 lr:8.857122887425438e-05 tokens_per_second_per_gpu:3524.576381393733 mfu:16.62602814577464 +Step 8034 | loss:1.1677249670028687 lr:8.856846950125287e-05 tokens_per_second_per_gpu:3720.3236965335823 mfu:17.54940162922516 +Step 8035 | loss:1.0244382619857788 lr:8.856570983817299e-05 tokens_per_second_per_gpu:4836.35020538893 mfu:22.813889085253088 +Step 8036 | loss:1.1370315551757812 lr:8.856294988503547e-05 tokens_per_second_per_gpu:4690.917951776355 mfu:22.12786033166144 +Step 8037 | loss:1.1466950178146362 lr:8.856018964186109e-05 tokens_per_second_per_gpu:4479.923943037353 mfu:21.132565593149543 +Step 8038 | loss:1.0377755165100098 lr:8.855742910867059e-05 tokens_per_second_per_gpu:5358.350510113838 mfu:25.276253585076294 +Step 8039 | loss:1.189481258392334 lr:8.855466828548475e-05 tokens_per_second_per_gpu:3908.585602696787 mfu:18.437465161390378 +Step 8040 | loss:1.0133929252624512 lr:8.855190717232435e-05 tokens_per_second_per_gpu:4892.174401788104 mfu:23.077221344258042 +Step 8041 | loss:1.0677974224090576 lr:8.854914576921012e-05 tokens_per_second_per_gpu:4307.556982139719 mfu:20.319481229759337 +Step 8042 | loss:1.0634419918060303 lr:8.854638407616284e-05 tokens_per_second_per_gpu:4042.8807897692914 mfu:19.07095847193302 +Step 8043 | loss:1.1625854969024658 lr:8.854362209320331e-05 tokens_per_second_per_gpu:5386.991046602612 mfu:25.411355882273252 +Step 8044 | loss:1.1292353868484497 lr:8.854085982035226e-05 tokens_per_second_per_gpu:4387.946296501588 mfu:20.698691341435637 +Step 8045 | loss:1.1701302528381348 lr:8.85380972576305e-05 tokens_per_second_per_gpu:3568.4796964897287 mfu:16.83312757376042 +Step 8046 | loss:1.230594277381897 lr:8.853533440505877e-05 tokens_per_second_per_gpu:4252.7026934345895 mfu:20.06072419083051 +Step 8047 | loss:1.1254914999008179 lr:8.853257126265789e-05 tokens_per_second_per_gpu:5214.874957422163 mfu:24.59945492357613 +Step 8048 | loss:1.015438199043274 lr:8.852980783044862e-05 tokens_per_second_per_gpu:5332.567447214679 mfu:25.154630478336816 +Step 8049 | loss:1.092326283454895 lr:8.852704410845176e-05 tokens_per_second_per_gpu:3877.219749861883 mfu:18.289507082001805 +Step 8050 | loss:1.0437884330749512 lr:8.852428009668807e-05 tokens_per_second_per_gpu:4602.457319981201 mfu:21.71057643854358 +Step 8051 | loss:1.198799967765808 lr:8.852151579517835e-05 tokens_per_second_per_gpu:3905.720477551869 mfu:18.423949877241043 +Step 8052 | loss:0.9820043444633484 lr:8.851875120394342e-05 tokens_per_second_per_gpu:5161.529872322514 mfu:24.34781705555104 +Step 8053 | loss:1.0721112489700317 lr:8.851598632300403e-05 tokens_per_second_per_gpu:5308.64975922711 mfu:25.041806663321083 +Step 8054 | loss:1.0617916584014893 lr:8.851322115238099e-05 tokens_per_second_per_gpu:5309.264180485042 mfu:25.044704993225825 +Step 8055 | loss:0.9257218241691589 lr:8.85104556920951e-05 tokens_per_second_per_gpu:4891.222079515393 mfu:23.07272907761473 +Step 8056 | loss:1.140078067779541 lr:8.850768994216716e-05 tokens_per_second_per_gpu:4596.221525772873 mfu:21.681161133326693 +Step 8057 | loss:1.2120248079299927 lr:8.850492390261798e-05 tokens_per_second_per_gpu:4558.0378507919695 mfu:21.501042223635036 +Step 8058 | loss:1.1477582454681396 lr:8.850215757346835e-05 tokens_per_second_per_gpu:5331.492663771084 mfu:25.149560541456413 +Step 8059 | loss:1.2283961772918701 lr:8.849939095473908e-05 tokens_per_second_per_gpu:4359.41349145426 mfu:20.564097231828853 +Step 8060 | loss:1.1917153596878052 lr:8.849662404645096e-05 tokens_per_second_per_gpu:5373.224824515447 mfu:25.346418263928413 +Step 8061 | loss:1.0025335550308228 lr:8.849385684862483e-05 tokens_per_second_per_gpu:4050.4383635167005 mfu:19.106608836754596 +Step 8062 | loss:1.0967596769332886 lr:8.84910893612815e-05 tokens_per_second_per_gpu:4111.998714207845 mfu:19.39699951424362 +Step 8063 | loss:0.9801437258720398 lr:8.848832158444179e-05 tokens_per_second_per_gpu:3960.9815294989567 mfu:18.684625687783154 +Step 8064 | loss:1.2012310028076172 lr:8.848555351812647e-05 tokens_per_second_per_gpu:4336.409879193552 mfu:20.455585267974126 +Step 8065 | loss:1.068786859512329 lr:8.84827851623564e-05 tokens_per_second_per_gpu:5345.077416681899 mfu:25.213642138734578 +Step 8066 | loss:1.2467325925827026 lr:8.848001651715239e-05 tokens_per_second_per_gpu:4413.805123899737 mfu:20.820671842243602 +Step 8067 | loss:1.1823583841323853 lr:8.847724758253526e-05 tokens_per_second_per_gpu:5380.808133435405 mfu:25.38219002594977 +Step 8068 | loss:0.9945719242095947 lr:8.847447835852585e-05 tokens_per_second_per_gpu:5016.505443451771 mfu:23.663712080848686 +Step 8069 | loss:1.1715551614761353 lr:8.847170884514499e-05 tokens_per_second_per_gpu:4625.983110342441 mfu:21.82155160559129 +Step 8070 | loss:1.0919209718704224 lr:8.846893904241348e-05 tokens_per_second_per_gpu:4573.219307470865 mfu:21.57265574501299 +Step 8071 | loss:1.0839931964874268 lr:8.846616895035219e-05 tokens_per_second_per_gpu:5055.592310457522 mfu:23.848091501422005 +Step 8072 | loss:1.0400577783584595 lr:8.846339856898192e-05 tokens_per_second_per_gpu:4689.29210751134 mfu:22.120190946865726 +Step 8073 | loss:1.0836197137832642 lr:8.846062789832354e-05 tokens_per_second_per_gpu:5320.518866215964 mfu:25.097795266065983 +Step 8074 | loss:1.1873353719711304 lr:8.845785693839786e-05 tokens_per_second_per_gpu:5350.01356714241 mfu:25.23692680265092 +Step 8075 | loss:1.0236765146255493 lr:8.845508568922571e-05 tokens_per_second_per_gpu:5336.2433247424015 mfu:25.17197021980446 +Step 8076 | loss:0.9599260687828064 lr:8.845231415082798e-05 tokens_per_second_per_gpu:5332.611417301427 mfu:25.154837892738076 +Step 8077 | loss:0.9695792198181152 lr:8.844954232322546e-05 tokens_per_second_per_gpu:3658.2174462007083 mfu:17.256435849986016 +Step 8078 | loss:1.1351754665374756 lr:8.844677020643906e-05 tokens_per_second_per_gpu:5260.429413041063 mfu:24.814343063122074 +Step 8079 | loss:1.1694772243499756 lr:8.844399780048958e-05 tokens_per_second_per_gpu:4426.280897928624 mfu:20.879522196924405 +Step 8080 | loss:0.9885969758033752 lr:8.844122510539789e-05 tokens_per_second_per_gpu:5340.728267488938 mfu:25.193126459950015 +Step 8081 | loss:1.0620821714401245 lr:8.843845212118483e-05 tokens_per_second_per_gpu:4833.64119749666 mfu:22.801110232820847 +Step 8082 | loss:1.1413534879684448 lr:8.843567884787128e-05 tokens_per_second_per_gpu:5332.373450512912 mfu:25.15371536279533 +Step 8083 | loss:1.0473980903625488 lr:8.843290528547809e-05 tokens_per_second_per_gpu:4613.137689272463 mfu:21.76095756274937 +Step 8084 | loss:1.046097755432129 lr:8.84301314340261e-05 tokens_per_second_per_gpu:5314.445440273609 mfu:25.06914587966257 +Step 8085 | loss:0.9677844643592834 lr:8.84273572935362e-05 tokens_per_second_per_gpu:5185.191983339572 mfu:24.459435270390944 +Step 8086 | loss:1.070115566253662 lr:8.842458286402922e-05 tokens_per_second_per_gpu:4580.049724768161 mfu:21.604875988793694 +Step 8087 | loss:1.059718370437622 lr:8.842180814552607e-05 tokens_per_second_per_gpu:4247.016313023407 mfu:20.033900564234415 +Step 8088 | loss:1.1022464036941528 lr:8.84190331380476e-05 tokens_per_second_per_gpu:5306.69978028613 mfu:25.032608279955767 +Step 8089 | loss:1.0934662818908691 lr:8.841625784161467e-05 tokens_per_second_per_gpu:4174.577575321954 mfu:19.692194679173557 +Step 8090 | loss:1.1388390064239502 lr:8.841348225624817e-05 tokens_per_second_per_gpu:5352.911957711279 mfu:25.2505990054807 +Step 8091 | loss:1.2092148065567017 lr:8.841070638196895e-05 tokens_per_second_per_gpu:5334.326463846999 mfu:25.16292805991065 +Step 8092 | loss:1.020453691482544 lr:8.840793021879794e-05 tokens_per_second_per_gpu:4472.796809010155 mfu:21.098945685928797 +Step 8093 | loss:1.1741821765899658 lr:8.840515376675595e-05 tokens_per_second_per_gpu:5371.234072095004 mfu:25.337027545103794 +Step 8094 | loss:1.1945141553878784 lr:8.840237702586391e-05 tokens_per_second_per_gpu:4387.720421528816 mfu:20.697625850650823 +Step 8095 | loss:1.3464102745056152 lr:8.839959999614271e-05 tokens_per_second_per_gpu:5295.078758474954 mfu:24.977789937321674 +Step 8096 | loss:1.1789438724517822 lr:8.83968226776132e-05 tokens_per_second_per_gpu:3850.030278445845 mfu:18.16124970633017 +Step 8097 | loss:1.0360251665115356 lr:8.83940450702963e-05 tokens_per_second_per_gpu:4155.933197589374 mfu:19.604245968350114 +Step 8098 | loss:1.224610686302185 lr:8.839126717421288e-05 tokens_per_second_per_gpu:3153.5188743185436 mfu:14.875686575962055 +Step 8099 | loss:0.9902056455612183 lr:8.838848898938384e-05 tokens_per_second_per_gpu:3941.612861658341 mfu:18.593260376943473 +Step 8100 | loss:1.076939344406128 lr:8.838571051583007e-05 tokens_per_second_per_gpu:5248.894482136057 mfu:24.759930825981012 +Step 8101 | loss:1.048490285873413 lr:8.838293175357249e-05 tokens_per_second_per_gpu:4184.157113440285 mfu:19.737382994915468 +Step 8102 | loss:1.3100088834762573 lr:8.838015270263196e-05 tokens_per_second_per_gpu:2946.7982497936373 mfu:13.900550119902583 +Step 8103 | loss:1.2338331937789917 lr:8.837737336302943e-05 tokens_per_second_per_gpu:3896.1051899865524 mfu:18.37859292525822 +Step 8104 | loss:1.032575249671936 lr:8.837459373478576e-05 tokens_per_second_per_gpu:4823.224495167387 mfu:22.751972870660982 +Step 8105 | loss:1.0219452381134033 lr:8.837181381792188e-05 tokens_per_second_per_gpu:5326.55086061408 mfu:25.126249212806265 +Step 8106 | loss:1.1168217658996582 lr:8.836903361245869e-05 tokens_per_second_per_gpu:3954.2960859156224 mfu:18.653089309746267 +Step 8107 | loss:1.020424246788025 lr:8.836625311841711e-05 tokens_per_second_per_gpu:3458.1941759123265 mfu:16.312891956546693 +Step 8108 | loss:1.1854201555252075 lr:8.836347233581804e-05 tokens_per_second_per_gpu:3942.4467681183487 mfu:18.597194055994443 +Step 8109 | loss:1.2344727516174316 lr:8.836069126468238e-05 tokens_per_second_per_gpu:4102.5972302233195 mfu:19.35265111023004 +Step 8110 | loss:1.0375624895095825 lr:8.835790990503108e-05 tokens_per_second_per_gpu:4394.3222105651585 mfu:20.72876761591661 +Step 8111 | loss:1.051243782043457 lr:8.835512825688506e-05 tokens_per_second_per_gpu:3813.1498384537545 mfu:17.987278378435562 +Step 8112 | loss:1.161419153213501 lr:8.835234632026521e-05 tokens_per_second_per_gpu:3928.000969912463 mfu:18.529050761149954 +Step 8113 | loss:1.18804132938385 lr:8.834956409519248e-05 tokens_per_second_per_gpu:4543.717692296315 mfu:21.43349159274005 +Step 8114 | loss:1.025285243988037 lr:8.834678158168777e-05 tokens_per_second_per_gpu:5331.306353348629 mfu:25.148681683405858 +Step 8115 | loss:1.104897379875183 lr:8.834399877977204e-05 tokens_per_second_per_gpu:5287.631086634522 mfu:24.942657998546512 +Step 8116 | loss:1.1103929281234741 lr:8.834121568946618e-05 tokens_per_second_per_gpu:4320.104398760755 mfu:20.378669534770793 +Step 8117 | loss:1.033003568649292 lr:8.833843231079116e-05 tokens_per_second_per_gpu:5351.836819093177 mfu:25.24552739318155 +Step 8118 | loss:1.0419375896453857 lr:8.833564864376792e-05 tokens_per_second_per_gpu:4662.865602889799 mfu:21.995532615739297 +Step 8119 | loss:1.3650968074798584 lr:8.833286468841732e-05 tokens_per_second_per_gpu:3238.5549219290097 mfu:15.276816121186428 +Step 8120 | loss:1.1610685586929321 lr:8.83300804447604e-05 tokens_per_second_per_gpu:4049.946613056184 mfu:19.10428916593106 +Step 8121 | loss:1.1387665271759033 lr:8.832729591281804e-05 tokens_per_second_per_gpu:5361.662558908953 mfu:25.291877084336644 +Step 8122 | loss:1.175946831703186 lr:8.832451109261119e-05 tokens_per_second_per_gpu:5355.087734952343 mfu:25.260862517952674 +Step 8123 | loss:1.1676995754241943 lr:8.832172598416082e-05 tokens_per_second_per_gpu:5319.154295562741 mfu:25.09135835347497 +Step 8124 | loss:1.3151249885559082 lr:8.831894058748784e-05 tokens_per_second_per_gpu:3910.496350938634 mfu:18.446478486854435 +Step 8125 | loss:1.1976213455200195 lr:8.831615490261322e-05 tokens_per_second_per_gpu:5346.068514944737 mfu:25.21831731833919 +Step 8126 | loss:1.0477876663208008 lr:8.831336892955793e-05 tokens_per_second_per_gpu:4256.7625167626875 mfu:20.079875070146432 +Step 8127 | loss:1.0834354162216187 lr:8.83105826683429e-05 tokens_per_second_per_gpu:4092.407114334866 mfu:19.30458259497115 +Step 8128 | loss:1.0654324293136597 lr:8.830779611898904e-05 tokens_per_second_per_gpu:3899.129225104518 mfu:18.392857814863255 +Step 8129 | loss:1.1978180408477783 lr:8.830500928151741e-05 tokens_per_second_per_gpu:5355.317264037284 mfu:25.26194524580804 +Step 8130 | loss:1.0953329801559448 lr:8.83022221559489e-05 tokens_per_second_per_gpu:5330.713864771903 mfu:25.145886813700464 +Step 8131 | loss:1.268028736114502 lr:8.82994347423045e-05 tokens_per_second_per_gpu:3715.083986608307 mfu:17.524685023520938 +Step 8132 | loss:1.1843156814575195 lr:8.829664704060516e-05 tokens_per_second_per_gpu:5255.482399943304 mfu:24.791007158292476 +Step 8133 | loss:1.1197484731674194 lr:8.829385905087184e-05 tokens_per_second_per_gpu:5379.9654120678315 mfu:25.37821476547584 +Step 8134 | loss:1.1916791200637817 lr:8.829107077312553e-05 tokens_per_second_per_gpu:4332.706647481796 mfu:20.43811649215329 +Step 8135 | loss:1.1339727640151978 lr:8.828828220738718e-05 tokens_per_second_per_gpu:5036.573128242711 mfu:23.758374773907644 +Step 8136 | loss:1.1828252077102661 lr:8.828549335367779e-05 tokens_per_second_per_gpu:5352.5836303727465 mfu:25.24905022940642 +Step 8137 | loss:1.0393996238708496 lr:8.828270421201831e-05 tokens_per_second_per_gpu:4323.822122316379 mfu:20.39620667109172 +Step 8138 | loss:1.045430064201355 lr:8.827991478242972e-05 tokens_per_second_per_gpu:4577.174930068767 mfu:21.591315091706104 +Step 8139 | loss:1.0819202661514282 lr:8.827712506493301e-05 tokens_per_second_per_gpu:5312.436665025366 mfu:25.059670144084212 +Step 8140 | loss:1.0656317472457886 lr:8.827433505954917e-05 tokens_per_second_per_gpu:4478.356216587243 mfu:21.125170360002663 +Step 8141 | loss:1.002426266670227 lr:8.827154476629918e-05 tokens_per_second_per_gpu:4420.133204441746 mfu:20.850522477842137 +Step 8142 | loss:1.0627362728118896 lr:8.8268754185204e-05 tokens_per_second_per_gpu:5370.262600550461 mfu:25.332444948078045 +Step 8143 | loss:1.180201768875122 lr:8.826596331628463e-05 tokens_per_second_per_gpu:4321.86327201352 mfu:20.386966439998808 +Step 8144 | loss:1.0312137603759766 lr:8.826317215956208e-05 tokens_per_second_per_gpu:4755.765463213885 mfu:22.433757107238616 +Step 8145 | loss:1.1136534214019775 lr:8.826038071505734e-05 tokens_per_second_per_gpu:3615.235492751779 mfu:17.053682642089957 +Step 8146 | loss:1.057987928390503 lr:8.825758898279138e-05 tokens_per_second_per_gpu:4371.630891091067 mfu:20.621728790418995 +Step 8147 | loss:1.216765284538269 lr:8.825479696278521e-05 tokens_per_second_per_gpu:3941.3798968608453 mfu:18.592161442245793 +Step 8148 | loss:0.9724685549736023 lr:8.825200465505983e-05 tokens_per_second_per_gpu:4352.50357442544 mfu:20.531501974250496 +Step 8149 | loss:1.0027230978012085 lr:8.824921205963624e-05 tokens_per_second_per_gpu:3521.2923789579054 mfu:16.61053694597625 +Step 8150 | loss:1.207131266593933 lr:8.824641917653545e-05 tokens_per_second_per_gpu:5334.283352937287 mfu:25.16272469839406 +Step 8151 | loss:1.1254500150680542 lr:8.824362600577847e-05 tokens_per_second_per_gpu:4535.544840300613 mfu:21.394938855443982 +Step 8152 | loss:1.0318726301193237 lr:8.824083254738629e-05 tokens_per_second_per_gpu:5310.643643171301 mfu:25.05121215407751 +Step 8153 | loss:1.135325312614441 lr:8.823803880137993e-05 tokens_per_second_per_gpu:5317.216067294333 mfu:25.082215400037043 +Step 8154 | loss:0.9859396815299988 lr:8.82352447677804e-05 tokens_per_second_per_gpu:4709.155648428223 mfu:22.213890658440214 +Step 8155 | loss:1.1354011297225952 lr:8.82324504466087e-05 tokens_per_second_per_gpu:4290.72422945752 mfu:20.240078263393645 +Step 8156 | loss:1.1451199054718018 lr:8.82296558378859e-05 tokens_per_second_per_gpu:4931.868571955889 mfu:23.264465517463275 +Step 8157 | loss:1.06582510471344 lr:8.822686094163295e-05 tokens_per_second_per_gpu:3251.7528993328065 mfu:15.339073232407282 +Step 8158 | loss:1.0528305768966675 lr:8.82240657578709e-05 tokens_per_second_per_gpu:4867.035634366648 mfu:22.958637489215032 +Step 8159 | loss:1.0864784717559814 lr:8.82212702866208e-05 tokens_per_second_per_gpu:5340.564580980821 mfu:25.192354322763425 +Step 8160 | loss:1.043271780014038 lr:8.821847452790361e-05 tokens_per_second_per_gpu:2882.82896223995 mfu:13.598796076225938 +Step 8161 | loss:1.032391905784607 lr:8.821567848174044e-05 tokens_per_second_per_gpu:3895.3373291525218 mfu:18.374970794693528 +Step 8162 | loss:1.0529780387878418 lr:8.821288214815224e-05 tokens_per_second_per_gpu:5381.04856655316 mfu:25.38332418998834 +Step 8163 | loss:1.036288857460022 lr:8.82100855271601e-05 tokens_per_second_per_gpu:3862.5110960248758 mfu:18.220123852297522 +Step 8164 | loss:1.0086828470230103 lr:8.820728861878502e-05 tokens_per_second_per_gpu:5349.352087551491 mfu:25.2338064905605 +Step 8165 | loss:1.0102566480636597 lr:8.820449142304806e-05 tokens_per_second_per_gpu:5353.5556738292435 mfu:25.25363552423843 +Step 8166 | loss:1.2154936790466309 lr:8.820169393997022e-05 tokens_per_second_per_gpu:3258.645031507658 mfu:15.371584595795063 +Step 8167 | loss:1.1665822267532349 lr:8.819889616957258e-05 tokens_per_second_per_gpu:5325.253011006273 mfu:25.120127034770256 +Step 8168 | loss:1.063515543937683 lr:8.819609811187617e-05 tokens_per_second_per_gpu:5353.1011169875 mfu:25.251491302060533 +Step 8169 | loss:1.1270108222961426 lr:8.819329976690203e-05 tokens_per_second_per_gpu:4583.261581598801 mfu:21.62002686546311 +Step 8170 | loss:1.0769555568695068 lr:8.819050113467122e-05 tokens_per_second_per_gpu:5371.964304460092 mfu:25.3404721757604 +Step 8171 | loss:1.2019304037094116 lr:8.818770221520475e-05 tokens_per_second_per_gpu:4225.648735829685 mfu:19.933106056927034 +Step 8172 | loss:1.2151063680648804 lr:8.818490300852372e-05 tokens_per_second_per_gpu:5332.138355857794 mfu:25.15260638119585 +Step 8173 | loss:1.061168909072876 lr:8.818210351464915e-05 tokens_per_second_per_gpu:4607.920567797166 mfu:21.736347510617684 +Step 8174 | loss:1.0982646942138672 lr:8.817930373360212e-05 tokens_per_second_per_gpu:5317.013776919759 mfu:25.081261161825882 +Step 8175 | loss:1.2807245254516602 lr:8.817650366540366e-05 tokens_per_second_per_gpu:5328.878772221293 mfu:25.13723036904016 +Step 8176 | loss:1.241768479347229 lr:8.817370331007488e-05 tokens_per_second_per_gpu:4883.79064068269 mfu:23.037673712706656 +Step 8177 | loss:1.2435452938079834 lr:8.817090266763677e-05 tokens_per_second_per_gpu:3359.567466834095 mfu:15.847652942372914 +Step 8178 | loss:1.170387864112854 lr:8.816810173811044e-05 tokens_per_second_per_gpu:4173.018639623776 mfu:19.684840913503393 +Step 8179 | loss:0.8265235424041748 lr:8.816530052151695e-05 tokens_per_second_per_gpu:4434.3414737932735 mfu:20.917545308554015 +Step 8180 | loss:1.0100901126861572 lr:8.816249901787737e-05 tokens_per_second_per_gpu:4504.19141303868 mfu:21.24703938960314 +Step 8181 | loss:1.0798938274383545 lr:8.815969722721275e-05 tokens_per_second_per_gpu:5332.899169163436 mfu:25.156195267366705 +Step 8182 | loss:1.1557791233062744 lr:8.815689514954418e-05 tokens_per_second_per_gpu:4442.309944209476 mfu:20.95513394306849 +Step 8183 | loss:1.0845259428024292 lr:8.815409278489275e-05 tokens_per_second_per_gpu:5375.407460381405 mfu:25.356714129704216 +Step 8184 | loss:1.089927077293396 lr:8.81512901332795e-05 tokens_per_second_per_gpu:4809.658915348646 mfu:22.687981716129393 +Step 8185 | loss:1.1263072490692139 lr:8.814848719472555e-05 tokens_per_second_per_gpu:5347.949276260197 mfu:25.22718919035518 +Step 8186 | loss:1.1361374855041504 lr:8.814568396925194e-05 tokens_per_second_per_gpu:3975.439427203984 mfu:18.75282605802948 +Step 8187 | loss:1.0704267024993896 lr:8.814288045687978e-05 tokens_per_second_per_gpu:5055.729375483927 mfu:23.848738060537258 +Step 8188 | loss:1.1537952423095703 lr:8.814007665763016e-05 tokens_per_second_per_gpu:3852.820181814523 mfu:18.174410156526825 +Step 8189 | loss:1.079630970954895 lr:8.813727257152413e-05 tokens_per_second_per_gpu:4471.452288442622 mfu:21.092603353012883 +Step 8190 | loss:1.0731607675552368 lr:8.813446819858283e-05 tokens_per_second_per_gpu:4671.932390606206 mfu:22.03830220035109 +Step 8191 | loss:1.0373711585998535 lr:8.813166353882732e-05 tokens_per_second_per_gpu:5264.184256840131 mfu:24.83205530196469 +Step 8192 | loss:1.1236218214035034 lr:8.81288585922787e-05 tokens_per_second_per_gpu:4181.7898991389175 mfu:19.726216441167505 +Step 8193 | loss:0.9047697186470032 lr:8.812605335895808e-05 tokens_per_second_per_gpu:4638.727140569751 mfu:21.88166737052928 +Step 8194 | loss:1.1670057773590088 lr:8.812324783888655e-05 tokens_per_second_per_gpu:4085.130734822486 mfu:19.270258671332893 +Step 8195 | loss:1.0422849655151367 lr:8.812044203208519e-05 tokens_per_second_per_gpu:3908.465911031851 mfu:18.43690055538537 +Step 8196 | loss:0.9484177827835083 lr:8.811763593857514e-05 tokens_per_second_per_gpu:4350.349160390295 mfu:20.521339235665774 +Step 8197 | loss:1.0724912881851196 lr:8.811482955837748e-05 tokens_per_second_per_gpu:4057.6309235607887 mfu:19.140537369659846 +Step 8198 | loss:1.116102695465088 lr:8.811202289151331e-05 tokens_per_second_per_gpu:4208.74647280063 mfu:19.853375198395536 +Step 8199 | loss:1.0211752653121948 lr:8.810921593800376e-05 tokens_per_second_per_gpu:4893.351675742767 mfu:23.082774746365835 +Step 8200 | loss:1.3083381652832031 lr:8.810640869786995e-05 tokens_per_second_per_gpu:3733.107789654322 mfu:17.609706377666893 +Step 8201 | loss:1.0425435304641724 lr:8.810360117113297e-05 tokens_per_second_per_gpu:5349.805441738819 mfu:25.23594503961233 +Step 8202 | loss:1.1085139513015747 lr:8.810079335781393e-05 tokens_per_second_per_gpu:5320.414201519781 mfu:25.097301544836448 +Step 8203 | loss:1.4255366325378418 lr:8.809798525793398e-05 tokens_per_second_per_gpu:5357.4009623754 mfu:25.27177440638453 +Step 8204 | loss:0.9883335828781128 lr:8.809517687151421e-05 tokens_per_second_per_gpu:5036.864135047368 mfu:23.75974750265609 +Step 8205 | loss:1.1081384420394897 lr:8.809236819857576e-05 tokens_per_second_per_gpu:3181.5971968152985 mfu:15.008136813835062 +Step 8206 | loss:0.9767167568206787 lr:8.808955923913973e-05 tokens_per_second_per_gpu:5020.815409025741 mfu:23.684042923817064 +Step 8207 | loss:1.1859411001205444 lr:8.808674999322728e-05 tokens_per_second_per_gpu:5049.773563569977 mfu:23.820643479572865 +Step 8208 | loss:0.9527947306632996 lr:8.808394046085953e-05 tokens_per_second_per_gpu:5386.282514345665 mfu:25.408013614729164 +Step 8209 | loss:1.0066306591033936 lr:8.808113064205758e-05 tokens_per_second_per_gpu:4348.291647860314 mfu:20.51163359801402 +Step 8210 | loss:0.9877765774726868 lr:8.807832053684261e-05 tokens_per_second_per_gpu:5057.459999890859 mfu:23.85690171113973 +Step 8211 | loss:0.9243258237838745 lr:8.807551014523575e-05 tokens_per_second_per_gpu:4596.572349011622 mfu:21.682816026400488 +Step 8212 | loss:1.0836011171340942 lr:8.807269946725807e-05 tokens_per_second_per_gpu:5387.807437290425 mfu:25.41520693643828 +Step 8213 | loss:1.1316044330596924 lr:8.806988850293079e-05 tokens_per_second_per_gpu:5039.150102176844 mfu:23.770530799632038 +Step 8214 | loss:1.101801872253418 lr:8.806707725227502e-05 tokens_per_second_per_gpu:5346.760279077669 mfu:25.221580487781356 +Step 8215 | loss:1.0780644416809082 lr:8.80642657153119e-05 tokens_per_second_per_gpu:3882.8532709275887 mfu:18.31608136204655 +Step 8216 | loss:1.1374220848083496 lr:8.806145389206259e-05 tokens_per_second_per_gpu:4601.55946004747 mfu:21.70634108004547 +Step 8217 | loss:1.051131010055542 lr:8.805864178254821e-05 tokens_per_second_per_gpu:5323.630759403202 mfu:25.112474597173907 +Step 8218 | loss:1.2629762887954712 lr:8.805582938678993e-05 tokens_per_second_per_gpu:5341.598196005904 mfu:25.19723006118965 +Step 8219 | loss:1.0630723237991333 lr:8.805301670480893e-05 tokens_per_second_per_gpu:3910.5938771129263 mfu:18.44693853445826 +Step 8220 | loss:1.0294525623321533 lr:8.805020373662631e-05 tokens_per_second_per_gpu:3942.016875611891 mfu:18.595166179694257 +Step 8221 | loss:1.213547945022583 lr:8.804739048226326e-05 tokens_per_second_per_gpu:5055.026122404102 mfu:23.84542069577238 +Step 8222 | loss:1.1659276485443115 lr:8.804457694174094e-05 tokens_per_second_per_gpu:4781.358600763176 mfu:22.55448430369816 +Step 8223 | loss:1.1374033689498901 lr:8.804176311508048e-05 tokens_per_second_per_gpu:4909.752229848707 mfu:23.160138958306632 +Step 8224 | loss:1.2553542852401733 lr:8.803894900230308e-05 tokens_per_second_per_gpu:5379.439288972756 mfu:25.375732953070507 +Step 8225 | loss:1.0332642793655396 lr:8.803613460342988e-05 tokens_per_second_per_gpu:5225.258002535016 mfu:24.648433518903634 +Step 8226 | loss:1.0226479768753052 lr:8.803331991848208e-05 tokens_per_second_per_gpu:5363.284830423422 mfu:25.299529615858457 +Step 8227 | loss:1.1231715679168701 lr:8.80305049474808e-05 tokens_per_second_per_gpu:4216.58312273686 mfu:19.890342013215083 +Step 8228 | loss:1.1301873922348022 lr:8.802768969044725e-05 tokens_per_second_per_gpu:4155.954284682122 mfu:19.60434543976444 +Step 8229 | loss:1.195993423461914 lr:8.80248741474026e-05 tokens_per_second_per_gpu:4073.1919121702226 mfu:19.213941207908974 +Step 8230 | loss:1.1373811960220337 lr:8.802205831836803e-05 tokens_per_second_per_gpu:4476.969724884131 mfu:21.118630042079207 +Step 8231 | loss:1.1042377948760986 lr:8.801924220336468e-05 tokens_per_second_per_gpu:4161.528709449089 mfu:19.630640952485155 +Step 8232 | loss:1.0632377862930298 lr:8.801642580241379e-05 tokens_per_second_per_gpu:3789.8169438625137 mfu:17.877213133646354 +Step 8233 | loss:1.1855075359344482 lr:8.801360911553648e-05 tokens_per_second_per_gpu:5179.409783597417 mfu:24.432159647662193 +Step 8234 | loss:0.973938524723053 lr:8.801079214275397e-05 tokens_per_second_per_gpu:5310.7070325302775 mfu:25.051511172499147 +Step 8235 | loss:1.1045033931732178 lr:8.800797488408745e-05 tokens_per_second_per_gpu:4700.127852355928 mfu:22.171305004066358 +Step 8236 | loss:1.1070998907089233 lr:8.800515733955811e-05 tokens_per_second_per_gpu:5347.71591600355 mfu:25.22608839020921 +Step 8237 | loss:1.1699103116989136 lr:8.800233950918712e-05 tokens_per_second_per_gpu:4564.606810433152 mfu:21.532029127042577 +Step 8238 | loss:1.13466215133667 lr:8.79995213929957e-05 tokens_per_second_per_gpu:4060.046339830425 mfu:19.15193130031623 +Step 8239 | loss:1.2316147089004517 lr:8.799670299100502e-05 tokens_per_second_per_gpu:4710.935457857954 mfu:22.222286323187056 +Step 8240 | loss:1.0665345191955566 lr:8.799388430323628e-05 tokens_per_second_per_gpu:4724.516908309336 mfu:22.28635234220062 +Step 8241 | loss:1.0279009342193604 lr:8.79910653297107e-05 tokens_per_second_per_gpu:4511.608000994827 mfu:21.282024700392675 +Step 8242 | loss:0.9796028733253479 lr:8.798824607044947e-05 tokens_per_second_per_gpu:4580.299591505147 mfu:21.606054652823723 +Step 8243 | loss:1.161458969116211 lr:8.798542652547379e-05 tokens_per_second_per_gpu:3606.710625114921 mfu:17.013469386954494 +Step 8244 | loss:1.1241575479507446 lr:8.798260669480487e-05 tokens_per_second_per_gpu:5321.924436016845 mfu:25.104425578631673 +Step 8245 | loss:1.1159366369247437 lr:8.797978657846391e-05 tokens_per_second_per_gpu:4954.99963598674 mfu:23.37357869306328 +Step 8246 | loss:1.051717758178711 lr:8.797696617647213e-05 tokens_per_second_per_gpu:3890.965850509812 mfu:18.3543497840847 +Step 8247 | loss:1.0595628023147583 lr:8.797414548885075e-05 tokens_per_second_per_gpu:4953.016818374688 mfu:23.364225404083665 +Step 8248 | loss:1.059335470199585 lr:8.797132451562097e-05 tokens_per_second_per_gpu:3520.0639907752025 mfu:16.604742429333804 +Step 8249 | loss:1.2459204196929932 lr:8.796850325680404e-05 tokens_per_second_per_gpu:4572.197026984878 mfu:21.567833473541555 +Step 8250 | loss:1.0382410287857056 lr:8.796568171242112e-05 tokens_per_second_per_gpu:4781.253933577156 mfu:22.553990570723645 +Step 8251 | loss:1.0985431671142578 lr:8.796285988249348e-05 tokens_per_second_per_gpu:4207.040865408443 mfu:19.84532955731996 +Step 8252 | loss:1.039971947669983 lr:8.796003776704232e-05 tokens_per_second_per_gpu:5334.449600718062 mfu:25.163508917540778 +Step 8253 | loss:0.9786067008972168 lr:8.795721536608888e-05 tokens_per_second_per_gpu:4149.583612279907 mfu:19.5742938910944 +Step 8254 | loss:1.1610220670700073 lr:8.795439267965437e-05 tokens_per_second_per_gpu:4998.876713955227 mfu:23.58055435604209 +Step 8255 | loss:1.1217931509017944 lr:8.795156970776001e-05 tokens_per_second_per_gpu:5357.6445702479505 mfu:25.272923546283142 +Step 8256 | loss:1.0429706573486328 lr:8.794874645042708e-05 tokens_per_second_per_gpu:4724.271587523998 mfu:22.285195122200143 +Step 8257 | loss:1.1316543817520142 lr:8.79459229076768e-05 tokens_per_second_per_gpu:3901.361791496124 mfu:18.403389211447568 +Step 8258 | loss:1.0319122076034546 lr:8.794309907953036e-05 tokens_per_second_per_gpu:4998.36569708859 mfu:23.578143802293695 +Step 8259 | loss:1.0672807693481445 lr:8.794027496600906e-05 tokens_per_second_per_gpu:4808.431844265991 mfu:22.682193412472593 +Step 8260 | loss:1.150072693824768 lr:8.793745056713407e-05 tokens_per_second_per_gpu:5384.2881225754045 mfu:25.398605728470525 +Step 8261 | loss:1.0880271196365356 lr:8.79346258829267e-05 tokens_per_second_per_gpu:5322.687498598051 mfu:25.10802507501912 +Step 8262 | loss:1.0638600587844849 lr:8.793180091340817e-05 tokens_per_second_per_gpu:4514.153320357208 mfu:21.29403140610129 +Step 8263 | loss:1.1439753770828247 lr:8.792897565859972e-05 tokens_per_second_per_gpu:3782.619701176599 mfu:17.843262511920667 +Step 8264 | loss:1.2060133218765259 lr:8.79261501185226e-05 tokens_per_second_per_gpu:3700.163596665954 mfu:17.454302998481833 +Step 8265 | loss:1.1159943342208862 lr:8.792332429319807e-05 tokens_per_second_per_gpu:3377.881293903874 mfu:15.934042389322165 +Step 8266 | loss:1.038296103477478 lr:8.792049818264738e-05 tokens_per_second_per_gpu:5151.572776622027 mfu:24.300847736276136 +Step 8267 | loss:1.1006087064743042 lr:8.791767178689179e-05 tokens_per_second_per_gpu:5124.528315826759 mfu:24.17327440044404 +Step 8268 | loss:0.9879319667816162 lr:8.791484510595254e-05 tokens_per_second_per_gpu:4561.359118851941 mfu:21.516709211740686 +Step 8269 | loss:0.9908435344696045 lr:8.79120181398509e-05 tokens_per_second_per_gpu:5372.413581808356 mfu:25.342591493666788 +Step 8270 | loss:1.318074107170105 lr:8.790919088860814e-05 tokens_per_second_per_gpu:5357.7907181925675 mfu:25.27361295107331 +Step 8271 | loss:1.0905709266662598 lr:8.790636335224553e-05 tokens_per_second_per_gpu:5362.170781734275 mfu:25.294274458115446 +Step 8272 | loss:1.0778096914291382 lr:8.79035355307843e-05 tokens_per_second_per_gpu:3200.2806920594435 mfu:15.096270048634691 +Step 8273 | loss:1.2006661891937256 lr:8.790070742424576e-05 tokens_per_second_per_gpu:5267.12701344652 mfu:24.845936786963083 +Step 8274 | loss:0.98252934217453 lr:8.789787903265117e-05 tokens_per_second_per_gpu:5360.223010833883 mfu:25.285086490454308 +Step 8275 | loss:1.1264634132385254 lr:8.789505035602179e-05 tokens_per_second_per_gpu:5340.692286369371 mfu:25.19295673087 +Step 8276 | loss:1.1787383556365967 lr:8.789222139437888e-05 tokens_per_second_per_gpu:3538.4021055093353 mfu:16.691246445339715 +Step 8277 | loss:1.042150855064392 lr:8.788939214774376e-05 tokens_per_second_per_gpu:5343.777253316671 mfu:25.20750904631005 +Step 8278 | loss:1.11544668674469 lr:8.788656261613768e-05 tokens_per_second_per_gpu:3150.2879608803014 mfu:14.86044580602319 +Step 8279 | loss:1.0475809574127197 lr:8.788373279958193e-05 tokens_per_second_per_gpu:4680.900451974248 mfu:22.080606075933193 +Step 8280 | loss:1.1849571466445923 lr:8.78809026980978e-05 tokens_per_second_per_gpu:5186.17331391589 mfu:24.464064374151697 +Step 8281 | loss:1.2174947261810303 lr:8.787807231170655e-05 tokens_per_second_per_gpu:4523.510932111234 mfu:21.338172857318227 +Step 8282 | loss:1.1571447849273682 lr:8.787524164042949e-05 tokens_per_second_per_gpu:4865.910967623007 mfu:22.953332244297346 +Step 8283 | loss:1.0848575830459595 lr:8.78724106842879e-05 tokens_per_second_per_gpu:4912.880084371706 mfu:23.174893581758628 +Step 8284 | loss:1.1869040727615356 lr:8.78695794433031e-05 tokens_per_second_per_gpu:3705.4461879912674 mfu:17.479221883066355 +Step 8285 | loss:1.3015468120574951 lr:8.786674791749634e-05 tokens_per_second_per_gpu:4348.502309403694 mfu:20.512627324456712 +Step 8286 | loss:0.9712510704994202 lr:8.786391610688894e-05 tokens_per_second_per_gpu:5347.856775726428 mfu:25.226752849555172 +Step 8287 | loss:1.152396559715271 lr:8.786108401150219e-05 tokens_per_second_per_gpu:5346.963514788834 mfu:25.22253918530651 +Step 8288 | loss:1.095987319946289 lr:8.785825163135742e-05 tokens_per_second_per_gpu:5064.1761125755 mfu:23.88858276056017 +Step 8289 | loss:1.1205374002456665 lr:8.785541896647588e-05 tokens_per_second_per_gpu:5076.8835855426305 mfu:23.948526078664578 +Step 8290 | loss:1.2458304166793823 lr:8.785258601687892e-05 tokens_per_second_per_gpu:4298.2493532930985 mfu:20.27557555644459 +Step 8291 | loss:1.006859540939331 lr:8.784975278258783e-05 tokens_per_second_per_gpu:3432.3995867399412 mfu:16.191214478409822 +Step 8292 | loss:0.987496554851532 lr:8.784691926362392e-05 tokens_per_second_per_gpu:4581.15579913683 mfu:21.610093530306454 +Step 8293 | loss:1.1163830757141113 lr:8.784408546000849e-05 tokens_per_second_per_gpu:5017.053350284818 mfu:23.66629665086272 +Step 8294 | loss:1.0730903148651123 lr:8.784125137176287e-05 tokens_per_second_per_gpu:4340.594408053793 mfu:20.475324404562127 +Step 8295 | loss:1.0406476259231567 lr:8.783841699890838e-05 tokens_per_second_per_gpu:4808.897043533397 mfu:22.68438783678813 +Step 8296 | loss:1.0703400373458862 lr:8.783558234146631e-05 tokens_per_second_per_gpu:4179.427293619984 mfu:19.715071627832632 +Step 8297 | loss:1.2159606218338013 lr:8.783274739945802e-05 tokens_per_second_per_gpu:4528.291827505499 mfu:21.360725156597844 +Step 8298 | loss:1.1873518228530884 lr:8.782991217290478e-05 tokens_per_second_per_gpu:4413.542993685783 mfu:20.819435329300354 +Step 8299 | loss:1.1267485618591309 lr:8.782707666182797e-05 tokens_per_second_per_gpu:4768.195761072523 mfu:22.49239294306532 +Step 8300 | loss:1.1694048643112183 lr:8.782424086624888e-05 tokens_per_second_per_gpu:4612.9251078527595 mfu:21.759954779922495 +Step 8301 | loss:1.1870801448822021 lr:8.782140478618883e-05 tokens_per_second_per_gpu:5252.298508491624 mfu:24.775988199087003 +Step 8302 | loss:1.2061353921890259 lr:8.78185684216692e-05 tokens_per_second_per_gpu:5315.509567279113 mfu:25.074165548306325 +Step 8303 | loss:0.9084635376930237 lr:8.781573177271127e-05 tokens_per_second_per_gpu:4133.661693209187 mfu:19.499187482278295 +Step 8304 | loss:1.1314481496810913 lr:8.78128948393364e-05 tokens_per_second_per_gpu:4868.401763178558 mfu:22.96508175190642 +Step 8305 | loss:1.0596222877502441 lr:8.781005762156593e-05 tokens_per_second_per_gpu:4345.314679203962 mfu:20.497590728938782 +Step 8306 | loss:1.2093602418899536 lr:8.780722011942118e-05 tokens_per_second_per_gpu:2943.814196501833 mfu:13.88647383139315 +Step 8307 | loss:1.1080729961395264 lr:8.780438233292351e-05 tokens_per_second_per_gpu:5337.865608606645 mfu:25.179622809582266 +Step 8308 | loss:1.068053126335144 lr:8.780154426209426e-05 tokens_per_second_per_gpu:3761.0274916616504 mfu:17.741408375627827 +Step 8309 | loss:1.001102089881897 lr:8.779870590695477e-05 tokens_per_second_per_gpu:4573.97254290788 mfu:21.576208885084608 +Step 8310 | loss:1.1158324480056763 lr:8.77958672675264e-05 tokens_per_second_per_gpu:4588.029398252511 mfu:21.642517470092066 +Step 8311 | loss:1.1489264965057373 lr:8.779302834383047e-05 tokens_per_second_per_gpu:5345.090964191786 mfu:25.21370604465016 +Step 8312 | loss:0.8917993307113647 lr:8.779018913588837e-05 tokens_per_second_per_gpu:5307.48434825887 mfu:25.036309220944283 +Step 8313 | loss:1.2181460857391357 lr:8.778734964372141e-05 tokens_per_second_per_gpu:4283.271330715336 mfu:20.20492166843112 +Step 8314 | loss:1.2288683652877808 lr:8.778450986735099e-05 tokens_per_second_per_gpu:5363.449189897022 mfu:25.300304927538047 +Step 8315 | loss:1.0652921199798584 lr:8.778166980679845e-05 tokens_per_second_per_gpu:5348.772954647805 mfu:25.23107462184355 +Step 8316 | loss:1.2524038553237915 lr:8.777882946208514e-05 tokens_per_second_per_gpu:4553.636099015423 mfu:21.48027840948877 +Step 8317 | loss:1.1595149040222168 lr:8.777598883323245e-05 tokens_per_second_per_gpu:3726.669719603899 mfu:17.579336902791027 +Step 8318 | loss:1.0369329452514648 lr:8.777314792026172e-05 tokens_per_second_per_gpu:5322.105059774389 mfu:25.105277611713596 +Step 8319 | loss:1.0921995639801025 lr:8.777030672319432e-05 tokens_per_second_per_gpu:3928.259044602541 mfu:18.530268143495135 +Step 8320 | loss:1.1436848640441895 lr:8.776746524205163e-05 tokens_per_second_per_gpu:4863.467361930947 mfu:22.941805339325775 +Step 8321 | loss:1.1323142051696777 lr:8.776462347685502e-05 tokens_per_second_per_gpu:5202.945012874039 mfu:24.5431793396842 +Step 8322 | loss:1.0000991821289062 lr:8.776178142762583e-05 tokens_per_second_per_gpu:4299.799211152269 mfu:20.282886500399318 +Step 8323 | loss:1.0914692878723145 lr:8.775893909438549e-05 tokens_per_second_per_gpu:4371.6674255304 mfu:20.621901129602747 +Step 8324 | loss:0.9945231676101685 lr:8.775609647715534e-05 tokens_per_second_per_gpu:4079.184656417673 mfu:19.2422100049918 +Step 8325 | loss:1.2029380798339844 lr:8.775325357595677e-05 tokens_per_second_per_gpu:5221.733055486071 mfu:24.631805742255775 +Step 8326 | loss:1.0761092901229858 lr:8.775041039081118e-05 tokens_per_second_per_gpu:4028.221151287151 mfu:19.001806456020525 +Step 8327 | loss:0.9701605439186096 lr:8.774756692173991e-05 tokens_per_second_per_gpu:5325.4242833497765 mfu:25.12093495563664 +Step 8328 | loss:1.1907176971435547 lr:8.774472316876439e-05 tokens_per_second_per_gpu:5326.995203595416 mfu:25.128345254461955 +Step 8329 | loss:1.0655993223190308 lr:8.774187913190598e-05 tokens_per_second_per_gpu:5311.463824995942 mfu:25.055081091681792 +Step 8330 | loss:1.0372999906539917 lr:8.77390348111861e-05 tokens_per_second_per_gpu:5350.151661469366 mfu:25.237578217152496 +Step 8331 | loss:0.9187007546424866 lr:8.773619020662611e-05 tokens_per_second_per_gpu:4791.788630570975 mfu:22.60368453384814 +Step 8332 | loss:1.169076919555664 lr:8.773334531824742e-05 tokens_per_second_per_gpu:4857.430496938977 mfu:22.913328417163275 +Step 8333 | loss:1.0075255632400513 lr:8.773050014607142e-05 tokens_per_second_per_gpu:4881.382872250682 mfu:23.026315858205226 +Step 8334 | loss:1.1446177959442139 lr:8.772765469011952e-05 tokens_per_second_per_gpu:5324.563364684533 mfu:25.116873855404812 +Step 8335 | loss:1.0062049627304077 lr:8.772480895041311e-05 tokens_per_second_per_gpu:5330.158159019685 mfu:25.14326545485407 +Step 8336 | loss:1.1508841514587402 lr:8.772196292697361e-05 tokens_per_second_per_gpu:5341.372577292675 mfu:25.196165779224824 +Step 8337 | loss:1.0440645217895508 lr:8.77191166198224e-05 tokens_per_second_per_gpu:4243.420558760298 mfu:20.016938777876394 +Step 8338 | loss:0.9573418498039246 lr:8.77162700289809e-05 tokens_per_second_per_gpu:4855.480080626055 mfu:22.904127970639884 +Step 8339 | loss:1.1034356355667114 lr:8.771342315447053e-05 tokens_per_second_per_gpu:3499.304016728817 mfu:16.50681408974019 +Step 8340 | loss:1.113578200340271 lr:8.77105759963127e-05 tokens_per_second_per_gpu:5084.596043640241 mfu:23.98490705939205 +Step 8341 | loss:1.215135097503662 lr:8.77077285545288e-05 tokens_per_second_per_gpu:5238.344953607864 mfu:24.710166899978354 +Step 8342 | loss:1.2299532890319824 lr:8.770488082914027e-05 tokens_per_second_per_gpu:3569.8802715274223 mfu:16.839734325175964 +Step 8343 | loss:1.12523353099823 lr:8.770203282016851e-05 tokens_per_second_per_gpu:5027.503106878848 mfu:23.71558993562908 +Step 8344 | loss:1.0423437356948853 lr:8.769918452763497e-05 tokens_per_second_per_gpu:4344.266074520155 mfu:20.49264428173485 +Step 8345 | loss:1.1117119789123535 lr:8.769633595156104e-05 tokens_per_second_per_gpu:4647.754481054187 mfu:21.924250875817467 +Step 8346 | loss:1.0694750547409058 lr:8.769348709196815e-05 tokens_per_second_per_gpu:5015.805588741332 mfu:23.660410746772023 +Step 8347 | loss:1.1959118843078613 lr:8.769063794887775e-05 tokens_per_second_per_gpu:5357.718157208586 mfu:25.273270668531687 +Step 8348 | loss:1.0248252153396606 lr:8.768778852231124e-05 tokens_per_second_per_gpu:5320.381816252785 mfu:25.09714877800653 +Step 8349 | loss:1.1699525117874146 lr:8.768493881229007e-05 tokens_per_second_per_gpu:5337.065114007147 mfu:25.17584673997797 +Step 8350 | loss:1.1736544370651245 lr:8.768208881883566e-05 tokens_per_second_per_gpu:5297.64178253539 mfu:24.98988015911175 +Step 8351 | loss:1.0445518493652344 lr:8.767923854196946e-05 tokens_per_second_per_gpu:3747.269714389375 mfu:17.676510592915655 +Step 8352 | loss:1.2532944679260254 lr:8.76763879817129e-05 tokens_per_second_per_gpu:3162.604042321289 mfu:14.918542863520505 +Step 8353 | loss:1.188488483428955 lr:8.76735371380874e-05 tokens_per_second_per_gpu:5382.298372159552 mfu:25.38921973627252 +Step 8354 | loss:1.1137765645980835 lr:8.767068601111443e-05 tokens_per_second_per_gpu:5326.113364204295 mfu:25.124185467596725 +Step 8355 | loss:1.1666874885559082 lr:8.766783460081543e-05 tokens_per_second_per_gpu:3532.8485168037364 mfu:16.665049219875463 +Step 8356 | loss:1.1283881664276123 lr:8.766498290721183e-05 tokens_per_second_per_gpu:5339.299235554129 mfu:25.186385472495527 +Step 8357 | loss:1.1308704614639282 lr:8.766213093032511e-05 tokens_per_second_per_gpu:4982.754703960769 mfu:23.50450408419968 +Step 8358 | loss:1.012881875038147 lr:8.765927867017667e-05 tokens_per_second_per_gpu:3544.242779009993 mfu:16.71879789876397 +Step 8359 | loss:1.141052484512329 lr:8.7656426126788e-05 tokens_per_second_per_gpu:5301.121760500167 mfu:25.006295808917063 +Step 8360 | loss:0.9699453711509705 lr:8.765357330018056e-05 tokens_per_second_per_gpu:4287.411846506942 mfu:20.224453187864967 +Step 8361 | loss:1.0958269834518433 lr:8.765072019037576e-05 tokens_per_second_per_gpu:4258.970198725857 mfu:20.090289082636087 +Step 8362 | loss:0.9869994521141052 lr:8.764786679739511e-05 tokens_per_second_per_gpu:5308.622878680535 mfu:25.041679863196595 +Step 8363 | loss:1.3135069608688354 lr:8.764501312126005e-05 tokens_per_second_per_gpu:4741.452334799247 mfu:22.366239638435243 +Step 8364 | loss:0.974702775478363 lr:8.764215916199203e-05 tokens_per_second_per_gpu:5333.722820063222 mfu:25.160080569190324 +Step 8365 | loss:1.0896557569503784 lr:8.763930491961254e-05 tokens_per_second_per_gpu:5307.897197128623 mfu:25.03825669950219 +Step 8366 | loss:1.055256962776184 lr:8.763645039414303e-05 tokens_per_second_per_gpu:4266.357653216526 mfu:20.125136965898278 +Step 8367 | loss:1.0697695016860962 lr:8.763359558560498e-05 tokens_per_second_per_gpu:4690.374217455216 mfu:22.125295444097986 +Step 8368 | loss:1.0739144086837769 lr:8.763074049401985e-05 tokens_per_second_per_gpu:4719.824501046464 mfu:22.264217456534418 +Step 8369 | loss:1.112638235092163 lr:8.762788511940912e-05 tokens_per_second_per_gpu:4730.832985220568 mfu:22.316146354624745 +Step 8370 | loss:1.0388606786727905 lr:8.762502946179425e-05 tokens_per_second_per_gpu:5275.837807947844 mfu:24.887027090840622 +Step 8371 | loss:0.9568794369697571 lr:8.762217352119674e-05 tokens_per_second_per_gpu:3898.9483931961663 mfu:18.392004799898455 +Step 8372 | loss:1.2037932872772217 lr:8.761931729763808e-05 tokens_per_second_per_gpu:5313.015483832832 mfu:25.0624005311557 +Step 8373 | loss:1.127697467803955 lr:8.761646079113974e-05 tokens_per_second_per_gpu:4170.867412967056 mfu:19.67469321032653 +Step 8374 | loss:1.1107975244522095 lr:8.761360400172316e-05 tokens_per_second_per_gpu:5381.617297300037 mfu:25.386006989956833 +Step 8375 | loss:1.1390255689620972 lr:8.76107469294099e-05 tokens_per_second_per_gpu:5317.457564392389 mfu:25.083354583052227 +Step 8376 | loss:1.096557855606079 lr:8.760788957422139e-05 tokens_per_second_per_gpu:4164.060323187778 mfu:19.642583006428396 +Step 8377 | loss:1.1648874282836914 lr:8.760503193617915e-05 tokens_per_second_per_gpu:4502.047365679252 mfu:21.236925552396517 +Step 8378 | loss:1.069507360458374 lr:8.760217401530468e-05 tokens_per_second_per_gpu:5331.854953899676 mfu:25.15126952580605 +Step 8379 | loss:1.004311203956604 lr:8.759931581161944e-05 tokens_per_second_per_gpu:5305.170385613498 mfu:25.025393864343325 +Step 8380 | loss:1.1399608850479126 lr:8.759645732514496e-05 tokens_per_second_per_gpu:3468.4455692635324 mfu:16.361249528052397 +Step 8381 | loss:1.0326297283172607 lr:8.759359855590273e-05 tokens_per_second_per_gpu:4447.730893135391 mfu:20.98070548406129 +Step 8382 | loss:1.1810894012451172 lr:8.759073950391426e-05 tokens_per_second_per_gpu:5338.4640153020655 mfu:25.18244559606327 +Step 8383 | loss:1.005090355873108 lr:8.7587880169201e-05 tokens_per_second_per_gpu:4518.349762408207 mfu:21.313826739242696 +Step 8384 | loss:1.0667058229446411 lr:8.758502055178454e-05 tokens_per_second_per_gpu:4032.8004954100447 mfu:19.02340800356489 +Step 8385 | loss:1.0837035179138184 lr:8.758216065168632e-05 tokens_per_second_per_gpu:5337.8438783276 mfu:25.17952030414844 +Step 8386 | loss:1.0490612983703613 lr:8.757930046892788e-05 tokens_per_second_per_gpu:5370.935038857061 mfu:25.335616954300086 +Step 8387 | loss:1.098212480545044 lr:8.757644000353074e-05 tokens_per_second_per_gpu:5320.901674205307 mfu:25.099601036665494 +Step 8388 | loss:1.219868540763855 lr:8.757357925551638e-05 tokens_per_second_per_gpu:3955.8983575132675 mfu:18.660647498248085 +Step 8389 | loss:1.2984074354171753 lr:8.757071822490635e-05 tokens_per_second_per_gpu:3852.2744801469826 mfu:18.17183598865473 +Step 8390 | loss:0.9729951024055481 lr:8.756785691172213e-05 tokens_per_second_per_gpu:5374.846514662566 mfu:25.354068053041242 +Step 8391 | loss:1.0596294403076172 lr:8.75649953159853e-05 tokens_per_second_per_gpu:5200.00207174704 mfu:24.529296984270612 +Step 8392 | loss:1.1549654006958008 lr:8.756213343771732e-05 tokens_per_second_per_gpu:3402.517864020583 mfu:16.05025729399499 +Step 8393 | loss:1.0594463348388672 lr:8.755927127693975e-05 tokens_per_second_per_gpu:4007.188431833256 mfu:18.902591529804525 +Step 8394 | loss:1.1418761014938354 lr:8.755640883367411e-05 tokens_per_second_per_gpu:3424.5609597771736 mfu:16.15423833761878 +Step 8395 | loss:1.1885935068130493 lr:8.755354610794193e-05 tokens_per_second_per_gpu:4170.7046402485785 mfu:19.67392538364158 +Step 8396 | loss:1.0990163087844849 lr:8.755068309976473e-05 tokens_per_second_per_gpu:5359.627488994951 mfu:25.282277312333566 +Step 8397 | loss:1.0942217111587524 lr:8.754781980916406e-05 tokens_per_second_per_gpu:4925.820588094845 mfu:23.235936145698282 +Step 8398 | loss:1.1686478853225708 lr:8.754495623616143e-05 tokens_per_second_per_gpu:5359.381119451369 mfu:25.28111514516151 +Step 8399 | loss:1.0197856426239014 lr:8.754209238077841e-05 tokens_per_second_per_gpu:5349.151030253913 mfu:25.232858068983912 +Step 8400 | loss:1.0741835832595825 lr:8.753922824303652e-05 tokens_per_second_per_gpu:5331.2658185967875 mfu:25.148490474065476 +Step 8401 | loss:1.1530076265335083 lr:8.753636382295729e-05 tokens_per_second_per_gpu:5321.704064492834 mfu:25.103386048553656 +Step 8402 | loss:1.0891659259796143 lr:8.753349912056231e-05 tokens_per_second_per_gpu:4242.914167297703 mfu:20.01455004294804 +Step 8403 | loss:1.0683573484420776 lr:8.753063413587305e-05 tokens_per_second_per_gpu:5322.134601660511 mfu:25.105416965830802 +Step 8404 | loss:0.9869914650917053 lr:8.752776886891113e-05 tokens_per_second_per_gpu:4951.660365988154 mfu:23.35782678674174 +Step 8405 | loss:0.9617723822593689 lr:8.752490331969807e-05 tokens_per_second_per_gpu:5327.337915726169 mfu:25.129961886055433 +Step 8406 | loss:1.0030148029327393 lr:8.752203748825541e-05 tokens_per_second_per_gpu:5339.2379382726285 mfu:25.186096322760296 +Step 8407 | loss:1.0560623407363892 lr:8.751917137460473e-05 tokens_per_second_per_gpu:4514.836361856312 mfu:21.297253429391457 +Step 8408 | loss:0.989155113697052 lr:8.751630497876759e-05 tokens_per_second_per_gpu:4810.503165792326 mfu:22.691964189516213 +Step 8409 | loss:1.202552080154419 lr:8.75134383007655e-05 tokens_per_second_per_gpu:4837.160190612544 mfu:22.81770992375061 +Step 8410 | loss:1.118916630744934 lr:8.751057134062009e-05 tokens_per_second_per_gpu:3548.302681152643 mfu:16.73794914986159 +Step 8411 | loss:1.0886024236679077 lr:8.750770409835285e-05 tokens_per_second_per_gpu:4563.833863715645 mfu:21.528383005497787 +Step 8412 | loss:1.1259511709213257 lr:8.75048365739854e-05 tokens_per_second_per_gpu:3400.4978563081595 mfu:16.040728572966632 +Step 8413 | loss:1.0785808563232422 lr:8.75019687675393e-05 tokens_per_second_per_gpu:3215.7393795747703 mfu:15.16919131516855 +Step 8414 | loss:1.2884876728057861 lr:8.749910067903608e-05 tokens_per_second_per_gpu:5188.976015467232 mfu:24.47728519555957 +Step 8415 | loss:1.0753856897354126 lr:8.749623230849735e-05 tokens_per_second_per_gpu:4822.2466176179005 mfu:22.747360055416937 +Step 8416 | loss:1.080051302909851 lr:8.749336365594468e-05 tokens_per_second_per_gpu:4726.946093132152 mfu:22.29781122147151 +Step 8417 | loss:1.1862000226974487 lr:8.749049472139963e-05 tokens_per_second_per_gpu:3705.4933049833626 mfu:17.479444141957057 +Step 8418 | loss:1.1316237449645996 lr:8.748762550488378e-05 tokens_per_second_per_gpu:5324.485950997385 mfu:25.116508681834404 +Step 8419 | loss:1.1238908767700195 lr:8.748475600641873e-05 tokens_per_second_per_gpu:4686.010745026428 mfu:22.104712200166226 +Step 8420 | loss:1.1881729364395142 lr:8.748188622602605e-05 tokens_per_second_per_gpu:5329.855690298338 mfu:25.141838658288997 +Step 8421 | loss:1.0515646934509277 lr:8.74790161637273e-05 tokens_per_second_per_gpu:4387.079014428654 mfu:20.694600223924024 +Step 8422 | loss:1.1633397340774536 lr:8.74761458195441e-05 tokens_per_second_per_gpu:3692.4357780674577 mfu:17.41784955965091 +Step 8423 | loss:1.0137070417404175 lr:8.747327519349803e-05 tokens_per_second_per_gpu:5023.781399592877 mfu:23.698034007372417 +Step 8424 | loss:1.1041470766067505 lr:8.747040428561066e-05 tokens_per_second_per_gpu:4964.326636508016 mfu:23.417575745872526 +Step 8425 | loss:1.2239569425582886 lr:8.746753309590362e-05 tokens_per_second_per_gpu:3976.9599321701644 mfu:18.759998539379854 +Step 8426 | loss:0.9213553071022034 lr:8.746466162439847e-05 tokens_per_second_per_gpu:5342.07527673518 mfu:25.19948053238801 +Step 8427 | loss:1.0433658361434937 lr:8.746178987111682e-05 tokens_per_second_per_gpu:5326.234282230639 mfu:25.12475585855773 +Step 8428 | loss:1.2073014974594116 lr:8.745891783608027e-05 tokens_per_second_per_gpu:4805.980909246835 mfu:22.6706319338146 +Step 8429 | loss:1.106474757194519 lr:8.745604551931042e-05 tokens_per_second_per_gpu:4806.294513699639 mfu:22.672111259524915 +Step 8430 | loss:1.0730164051055908 lr:8.745317292082886e-05 tokens_per_second_per_gpu:5305.002030980372 mfu:25.02459970681461 +Step 8431 | loss:1.1590508222579956 lr:8.745030004065722e-05 tokens_per_second_per_gpu:4280.638248493023 mfu:20.192500970338312 +Step 8432 | loss:1.0665435791015625 lr:8.744742687881709e-05 tokens_per_second_per_gpu:5162.046644325839 mfu:24.35025475726112 +Step 8433 | loss:0.9486701488494873 lr:8.744455343533009e-05 tokens_per_second_per_gpu:5294.960447768351 mfu:24.97723184553274 +Step 8434 | loss:1.0612444877624512 lr:8.744167971021782e-05 tokens_per_second_per_gpu:4700.250185566235 mfu:22.171882070691574 +Step 8435 | loss:1.0553945302963257 lr:8.74388057035019e-05 tokens_per_second_per_gpu:4692.297576092735 mfu:22.134368254949987 +Step 8436 | loss:0.9805452227592468 lr:8.743593141520395e-05 tokens_per_second_per_gpu:4825.694168471495 mfu:22.763622741006174 +Step 8437 | loss:1.0319840908050537 lr:8.743305684534557e-05 tokens_per_second_per_gpu:5338.239864821563 mfu:25.18138824000273 +Step 8438 | loss:1.1607478857040405 lr:8.74301819939484e-05 tokens_per_second_per_gpu:5344.730955388919 mfu:25.2120078217032 +Step 8439 | loss:0.9617558717727661 lr:8.742730686103406e-05 tokens_per_second_per_gpu:5327.369927565109 mfu:25.130112891361495 +Step 8440 | loss:1.086778163909912 lr:8.742443144662416e-05 tokens_per_second_per_gpu:4400.789667760234 mfu:20.75927571496776 +Step 8441 | loss:1.013617753982544 lr:8.742155575074035e-05 tokens_per_second_per_gpu:5367.9220202574315 mfu:25.321404031491653 +Step 8442 | loss:1.0616014003753662 lr:8.741867977340423e-05 tokens_per_second_per_gpu:5224.194624393429 mfu:24.643417382778747 +Step 8443 | loss:1.1921864748001099 lr:8.741580351463745e-05 tokens_per_second_per_gpu:4997.536412792569 mfu:23.574231926778868 +Step 8444 | loss:1.2862588167190552 lr:8.741292697446164e-05 tokens_per_second_per_gpu:3475.345020006314 mfu:16.39379541437378 +Step 8445 | loss:1.034891128540039 lr:8.741005015289843e-05 tokens_per_second_per_gpu:4256.215861075052 mfu:20.077296401998776 +Step 8446 | loss:1.0248136520385742 lr:8.740717304996945e-05 tokens_per_second_per_gpu:4735.702419497557 mfu:22.339116306920456 +Step 8447 | loss:1.045821189880371 lr:8.740429566569637e-05 tokens_per_second_per_gpu:4183.524138304146 mfu:19.7343971432016 +Step 8448 | loss:1.1231565475463867 lr:8.74014180001008e-05 tokens_per_second_per_gpu:4654.55898074089 mfu:21.956348861807008 +Step 8449 | loss:1.1668566465377808 lr:8.739854005320438e-05 tokens_per_second_per_gpu:5319.705315468936 mfu:25.093957608386177 +Step 8450 | loss:0.9684146642684937 lr:8.739566182502878e-05 tokens_per_second_per_gpu:4507.407823976469 mfu:21.262211748772987 +Step 8451 | loss:1.1534247398376465 lr:8.739278331559563e-05 tokens_per_second_per_gpu:5302.703689418082 mfu:25.01375803752778 +Step 8452 | loss:1.2138569355010986 lr:8.73899045249266e-05 tokens_per_second_per_gpu:3925.0491152216982 mfu:18.515126358935273 +Step 8453 | loss:1.0769389867782593 lr:8.738702545304331e-05 tokens_per_second_per_gpu:5343.980140560592 mfu:25.208466100055762 +Step 8454 | loss:1.2299216985702515 lr:8.738414609996744e-05 tokens_per_second_per_gpu:3961.2609600669794 mfu:18.685943809448286 +Step 8455 | loss:1.1043461561203003 lr:8.738126646572063e-05 tokens_per_second_per_gpu:4516.408863021578 mfu:21.304671185684384 +Step 8456 | loss:1.0496658086776733 lr:8.737838655032455e-05 tokens_per_second_per_gpu:4573.668688294075 mfu:21.57477555102835 +Step 8457 | loss:1.1673555374145508 lr:8.737550635380084e-05 tokens_per_second_per_gpu:4197.98792510235 mfu:19.80262529330523 +Step 8458 | loss:1.1107325553894043 lr:8.73726258761712e-05 tokens_per_second_per_gpu:5300.57159466166 mfu:25.00370058278886 +Step 8459 | loss:1.2505853176116943 lr:8.736974511745727e-05 tokens_per_second_per_gpu:4969.932326435783 mfu:23.444018741691128 +Step 8460 | loss:1.1323347091674805 lr:8.736686407768072e-05 tokens_per_second_per_gpu:3916.329004071005 mfu:18.473992106833073 +Step 8461 | loss:1.2350105047225952 lr:8.736398275686322e-05 tokens_per_second_per_gpu:4322.879594401973 mfu:20.39176060610762 +Step 8462 | loss:1.1239300966262817 lr:8.736110115502643e-05 tokens_per_second_per_gpu:3898.059198611804 mfu:18.387810317331777 +Step 8463 | loss:1.1287752389907837 lr:8.735821927219202e-05 tokens_per_second_per_gpu:3811.1956509368692 mfu:17.978060142499643 +Step 8464 | loss:1.2001641988754272 lr:8.735533710838169e-05 tokens_per_second_per_gpu:3483.1184103602195 mfu:16.43046382295031 +Step 8465 | loss:1.1677513122558594 lr:8.735245466361709e-05 tokens_per_second_per_gpu:4939.945176909282 mfu:23.30256424103933 +Step 8466 | loss:1.1157002449035645 lr:8.734957193791991e-05 tokens_per_second_per_gpu:4793.974191944873 mfu:22.613994199744038 +Step 8467 | loss:0.9592921137809753 lr:8.734668893131186e-05 tokens_per_second_per_gpu:4495.689699489813 mfu:21.20693535625123 +Step 8468 | loss:1.0855638980865479 lr:8.734380564381457e-05 tokens_per_second_per_gpu:4977.283428584076 mfu:23.478695144751985 +Step 8469 | loss:0.9970905780792236 lr:8.734092207544977e-05 tokens_per_second_per_gpu:4617.254794802652 mfu:21.78037865198591 +Step 8470 | loss:1.0291107892990112 lr:8.733803822623913e-05 tokens_per_second_per_gpu:3484.1254172384947 mfu:16.43521404620822 +Step 8471 | loss:1.1613256931304932 lr:8.733515409620432e-05 tokens_per_second_per_gpu:5336.207527043757 mfu:25.171801355952237 +Step 8472 | loss:1.132411003112793 lr:8.733226968536708e-05 tokens_per_second_per_gpu:4753.514404073108 mfu:22.423138477201277 +Step 8473 | loss:1.2190157175064087 lr:8.732938499374906e-05 tokens_per_second_per_gpu:4091.1965595056186 mfu:19.29887220129964 +Step 8474 | loss:1.0142643451690674 lr:8.732650002137197e-05 tokens_per_second_per_gpu:3104.943243155391 mfu:14.64654703590817 +Step 8475 | loss:0.9957086443901062 lr:8.732361476825752e-05 tokens_per_second_per_gpu:3716.455789564738 mfu:17.531156052120142 +Step 8476 | loss:1.1772608757019043 lr:8.732072923442738e-05 tokens_per_second_per_gpu:3889.551534428923 mfu:18.34767821382394 +Step 8477 | loss:1.1723202466964722 lr:8.73178434199033e-05 tokens_per_second_per_gpu:4256.82916675393 mfu:20.080189469526783 +Step 8478 | loss:1.035927653312683 lr:8.731495732470695e-05 tokens_per_second_per_gpu:5147.180256593397 mfu:24.28012746985989 +Step 8479 | loss:1.0129786729812622 lr:8.731207094886003e-05 tokens_per_second_per_gpu:3189.183990182196 mfu:15.043925012587142 +Step 8480 | loss:1.0545953512191772 lr:8.730918429238428e-05 tokens_per_second_per_gpu:5368.701933987219 mfu:25.32508301762206 +Step 8481 | loss:1.1668249368667603 lr:8.730629735530139e-05 tokens_per_second_per_gpu:4618.9664141717985 mfu:21.788452652582553 +Step 8482 | loss:1.0528101921081543 lr:8.730341013763308e-05 tokens_per_second_per_gpu:4101.722963117037 mfu:19.348527043124353 +Step 8483 | loss:1.0946112871170044 lr:8.730052263940104e-05 tokens_per_second_per_gpu:5267.9237395033615 mfu:24.849695079708685 +Step 8484 | loss:0.946655809879303 lr:8.729763486062704e-05 tokens_per_second_per_gpu:5067.54332147817 mfu:23.90446645945914 +Step 8485 | loss:1.1170798540115356 lr:8.729474680133276e-05 tokens_per_second_per_gpu:4861.000575599338 mfu:22.930169087322426 +Step 8486 | loss:1.0350276231765747 lr:8.729185846153993e-05 tokens_per_second_per_gpu:4905.451957477328 mfu:23.13985383982871 +Step 8487 | loss:1.0724828243255615 lr:8.728896984127026e-05 tokens_per_second_per_gpu:5336.49762750931 mfu:25.173169809344493 +Step 8488 | loss:1.1838698387145996 lr:8.728608094054552e-05 tokens_per_second_per_gpu:3665.0614618113786 mfu:17.288720239331024 +Step 8489 | loss:1.2608360052108765 lr:8.728319175938738e-05 tokens_per_second_per_gpu:5341.177464279054 mfu:25.195245397849863 +Step 8490 | loss:1.207492709159851 lr:8.72803022978176e-05 tokens_per_second_per_gpu:5305.611980559917 mfu:25.027476943049454 +Step 8491 | loss:1.013198733329773 lr:8.727741255585792e-05 tokens_per_second_per_gpu:5321.55001247818 mfu:25.10265935891655 +Step 8492 | loss:1.0359007120132446 lr:8.727452253353005e-05 tokens_per_second_per_gpu:3893.5999992169936 mfu:18.36677551296859 +Step 8493 | loss:1.2621697187423706 lr:8.727163223085575e-05 tokens_per_second_per_gpu:4658.537709714263 mfu:21.97511720521557 +Step 8494 | loss:1.1921064853668213 lr:8.726874164785675e-05 tokens_per_second_per_gpu:3905.7660125974417 mfu:18.42416467382009 +Step 8495 | loss:1.280157208442688 lr:8.72658507845548e-05 tokens_per_second_per_gpu:5378.591937988695 mfu:25.371735853905708 +Step 8496 | loss:1.079879879951477 lr:8.72629596409716e-05 tokens_per_second_per_gpu:5178.729018935739 mfu:24.42894836460263 +Step 8497 | loss:1.187921404838562 lr:8.726006821712895e-05 tokens_per_second_per_gpu:4837.046360119613 mfu:22.817172965893967 +Step 8498 | loss:1.0777599811553955 lr:8.725717651304857e-05 tokens_per_second_per_gpu:5368.418830877745 mfu:25.3237475719523 +Step 8499 | loss:1.3241784572601318 lr:8.72542845287522e-05 tokens_per_second_per_gpu:4441.8994465374035 mfu:20.9531975555112 +Step 8500 | loss:1.172371745109558 lr:8.725139226426161e-05 tokens_per_second_per_gpu:5324.996543833499 mfu:25.118917235358474 +Step 8501 | loss:1.1168627738952637 lr:8.724849971959855e-05 tokens_per_second_per_gpu:4722.701137635475 mfu:22.277787042129592 +Step 8502 | loss:1.2116925716400146 lr:8.724560689478476e-05 tokens_per_second_per_gpu:5164.649579113353 mfu:24.362533244760296 +Step 8503 | loss:1.0392765998840332 lr:8.724271378984201e-05 tokens_per_second_per_gpu:4535.204301146456 mfu:21.393332474151855 +Step 8504 | loss:1.182425618171692 lr:8.723982040479206e-05 tokens_per_second_per_gpu:4397.368916366162 mfu:20.743139447001553 +Step 8505 | loss:1.0429956912994385 lr:8.723692673965667e-05 tokens_per_second_per_gpu:4062.546510127012 mfu:19.163725030178274 +Step 8506 | loss:1.0286552906036377 lr:8.72340327944576e-05 tokens_per_second_per_gpu:4498.440591555183 mfu:21.219911783474164 +Step 8507 | loss:1.108695387840271 lr:8.723113856921663e-05 tokens_per_second_per_gpu:4441.123871705823 mfu:20.949539036703925 +Step 8508 | loss:1.0924664735794067 lr:8.72282440639555e-05 tokens_per_second_per_gpu:4442.518442978639 mfu:20.956117467336185 +Step 8509 | loss:1.0766360759735107 lr:8.7225349278696e-05 tokens_per_second_per_gpu:3701.4819752414946 mfu:17.460522015161235 +Step 8510 | loss:1.136959433555603 lr:8.722245421345991e-05 tokens_per_second_per_gpu:5313.707078783667 mfu:25.06566290253706 +Step 8511 | loss:1.1771873235702515 lr:8.721955886826898e-05 tokens_per_second_per_gpu:4865.381183611301 mfu:22.950833162723658 +Step 8512 | loss:0.9984563589096069 lr:8.721666324314499e-05 tokens_per_second_per_gpu:4643.8748229627845 mfu:21.905949866662397 +Step 8513 | loss:1.0810256004333496 lr:8.721376733810975e-05 tokens_per_second_per_gpu:4074.416247343428 mfu:19.219716605814106 +Step 8514 | loss:1.0027779340744019 lr:8.7210871153185e-05 tokens_per_second_per_gpu:4157.887667200989 mfu:19.613465534975006 +Step 8515 | loss:1.1138699054718018 lr:8.720797468839254e-05 tokens_per_second_per_gpu:5357.822171849752 mfu:25.273761323340793 +Step 8516 | loss:1.1861786842346191 lr:8.720507794375417e-05 tokens_per_second_per_gpu:4918.966276560412 mfu:23.20360318872422 +Step 8517 | loss:1.1698429584503174 lr:8.720218091929164e-05 tokens_per_second_per_gpu:5228.291682161055 mfu:24.662743903298775 +Step 8518 | loss:1.1513651609420776 lr:8.719928361502678e-05 tokens_per_second_per_gpu:5152.331884986309 mfu:24.304428579947697 +Step 8519 | loss:1.1557782888412476 lr:8.719638603098135e-05 tokens_per_second_per_gpu:4975.46969265112 mfu:23.470139442900965 +Step 8520 | loss:1.0850622653961182 lr:8.719348816717716e-05 tokens_per_second_per_gpu:3568.600789565974 mfu:16.833698790461682 +Step 8521 | loss:0.9762612581253052 lr:8.719059002363598e-05 tokens_per_second_per_gpu:4285.528212873013 mfu:20.215567766632685 +Step 8522 | loss:1.159610390663147 lr:8.718769160037964e-05 tokens_per_second_per_gpu:4239.889504223865 mfu:20.000282191168008 +Step 8523 | loss:1.1281510591506958 lr:8.718479289742995e-05 tokens_per_second_per_gpu:4246.609178396942 mfu:20.03198004073705 +Step 8524 | loss:1.0889402627944946 lr:8.718189391480866e-05 tokens_per_second_per_gpu:4144.586829206532 mfu:19.55072321279818 +Step 8525 | loss:1.0715645551681519 lr:8.717899465253762e-05 tokens_per_second_per_gpu:5373.647104503575 mfu:25.34841023068081 +Step 8526 | loss:1.0942506790161133 lr:8.71760951106386e-05 tokens_per_second_per_gpu:3334.437487597948 mfu:15.729110542699425 +Step 8527 | loss:1.2093150615692139 lr:8.717319528913343e-05 tokens_per_second_per_gpu:5345.694974879834 mfu:25.21655526612053 +Step 8528 | loss:1.1950041055679321 lr:8.717029518804391e-05 tokens_per_second_per_gpu:3573.157449097382 mfu:16.855193330917615 +Step 8529 | loss:1.2336418628692627 lr:8.716739480739187e-05 tokens_per_second_per_gpu:3395.7796626979207 mfu:16.018472048700296 +Step 8530 | loss:1.161651849746704 lr:8.71644941471991e-05 tokens_per_second_per_gpu:3120.693992598475 mfu:14.720846008386161 +Step 8531 | loss:1.1616202592849731 lr:8.716159320748743e-05 tokens_per_second_per_gpu:5343.44816092329 mfu:25.20595666134152 +Step 8532 | loss:1.0286078453063965 lr:8.715869198827868e-05 tokens_per_second_per_gpu:4993.701518167337 mfu:23.55614207453116 +Step 8533 | loss:1.090820074081421 lr:8.715579048959465e-05 tokens_per_second_per_gpu:5207.167158321562 mfu:24.563095920132465 +Step 8534 | loss:1.0350055694580078 lr:8.715288871145722e-05 tokens_per_second_per_gpu:3276.9703178743907 mfu:15.4580280982032 +Step 8535 | loss:1.3416045904159546 lr:8.714998665388814e-05 tokens_per_second_per_gpu:5324.34354321765 mfu:25.11583692004012 +Step 8536 | loss:1.2263827323913574 lr:8.714708431690926e-05 tokens_per_second_per_gpu:5322.325947409922 mfu:25.10631957637773 +Step 8537 | loss:1.1199524402618408 lr:8.714418170054243e-05 tokens_per_second_per_gpu:5331.4717445617125 mfu:25.149461861976622 +Step 8538 | loss:1.1069802045822144 lr:8.714127880480948e-05 tokens_per_second_per_gpu:4840.563899634827 mfu:22.833765799941357 +Step 8539 | loss:1.10409677028656 lr:8.713837562973223e-05 tokens_per_second_per_gpu:5306.4141433992145 mfu:25.03126088202551 +Step 8540 | loss:1.1195987462997437 lr:8.713547217533251e-05 tokens_per_second_per_gpu:4245.711571533293 mfu:20.027745876013693 +Step 8541 | loss:1.0371814966201782 lr:8.713256844163218e-05 tokens_per_second_per_gpu:4117.729907758362 mfu:19.424034532063857 +Step 8542 | loss:1.0247265100479126 lr:8.712966442865305e-05 tokens_per_second_per_gpu:5172.251130100412 mfu:24.39839105772459 +Step 8543 | loss:1.29762864112854 lr:8.712676013641698e-05 tokens_per_second_per_gpu:5322.660196349391 mfu:25.107896285653954 +Step 8544 | loss:1.0281914472579956 lr:8.712385556494579e-05 tokens_per_second_per_gpu:5331.411384287967 mfu:25.14917713226684 +Step 8545 | loss:1.204944372177124 lr:8.712095071426138e-05 tokens_per_second_per_gpu:5326.259722382179 mfu:25.12487586409308 +Step 8546 | loss:1.0946663618087769 lr:8.711804558438555e-05 tokens_per_second_per_gpu:4504.380200417451 mfu:21.247929931879252 +Step 8547 | loss:1.0591776371002197 lr:8.711514017534016e-05 tokens_per_second_per_gpu:5215.0327992234 mfu:24.600199490283156 +Step 8548 | loss:1.1924833059310913 lr:8.711223448714707e-05 tokens_per_second_per_gpu:4686.87063596649 mfu:22.108768452443027 +Step 8549 | loss:1.0334776639938354 lr:8.710932851982813e-05 tokens_per_second_per_gpu:4657.844790358713 mfu:21.97184858639121 +Step 8550 | loss:1.0407973527908325 lr:8.710642227340518e-05 tokens_per_second_per_gpu:3450.873040759468 mfu:16.2783569129162 +Step 8551 | loss:1.1695914268493652 lr:8.71035157479001e-05 tokens_per_second_per_gpu:5081.375527890313 mfu:23.969715337122988 +Step 8552 | loss:0.945419192314148 lr:8.710060894333477e-05 tokens_per_second_per_gpu:5033.628033066037 mfu:23.744482257474218 +Step 8553 | loss:0.9391169548034668 lr:8.709770185973099e-05 tokens_per_second_per_gpu:4754.325181585045 mfu:22.426963053057992 +Step 8554 | loss:1.133467197418213 lr:8.709479449711067e-05 tokens_per_second_per_gpu:5250.215735390954 mfu:24.76616340301341 +Step 8555 | loss:1.1126347780227661 lr:8.709188685549568e-05 tokens_per_second_per_gpu:3992.0477281871526 mfu:18.831170247435434 +Step 8556 | loss:1.1934300661087036 lr:8.708897893490786e-05 tokens_per_second_per_gpu:5370.619372672016 mfu:25.334127903047023 +Step 8557 | loss:1.072581171989441 lr:8.70860707353691e-05 tokens_per_second_per_gpu:5328.156210559573 mfu:25.1338219224001 +Step 8558 | loss:1.0129196643829346 lr:8.708316225690128e-05 tokens_per_second_per_gpu:5345.783822005868 mfu:25.216974373172917 +Step 8559 | loss:1.041150689125061 lr:8.708025349952625e-05 tokens_per_second_per_gpu:4676.714026299995 mfu:22.060857991749863 +Step 8560 | loss:1.0002700090408325 lr:8.707734446326591e-05 tokens_per_second_per_gpu:4776.765654395617 mfu:22.53281859204493 +Step 8561 | loss:1.23012375831604 lr:8.707443514814213e-05 tokens_per_second_per_gpu:3871.080201757833 mfu:18.26054578608008 +Step 8562 | loss:1.0399174690246582 lr:8.707152555417679e-05 tokens_per_second_per_gpu:4044.730651616756 mfu:19.079684585886486 +Step 8563 | loss:1.235894799232483 lr:8.706861568139179e-05 tokens_per_second_per_gpu:5324.0311578098745 mfu:25.114363344773196 +Step 8564 | loss:0.8943656086921692 lr:8.7065705529809e-05 tokens_per_second_per_gpu:4717.043007449461 mfu:22.25109667666561 +Step 8565 | loss:1.151955008506775 lr:8.706279509945028e-05 tokens_per_second_per_gpu:5094.5349944288855 mfu:24.03179078602195 +Step 8566 | loss:1.056320309638977 lr:8.705988439033758e-05 tokens_per_second_per_gpu:5349.364212305344 mfu:25.23386368509312 +Step 8567 | loss:1.1527135372161865 lr:8.705697340249275e-05 tokens_per_second_per_gpu:3850.477341767496 mfu:18.163358580295267 +Step 8568 | loss:1.185470461845398 lr:8.70540621359377e-05 tokens_per_second_per_gpu:4347.766020218407 mfu:20.50915412274582 +Step 8569 | loss:1.0844117403030396 lr:8.705115059069431e-05 tokens_per_second_per_gpu:5343.6883316607755 mfu:25.207089587687594 +Step 8570 | loss:1.1868919134140015 lr:8.70482387667845e-05 tokens_per_second_per_gpu:4427.080319225548 mfu:20.883293203577335 +Step 8571 | loss:1.111674189567566 lr:8.704532666423013e-05 tokens_per_second_per_gpu:5354.050614026645 mfu:25.255970241595083 +Step 8572 | loss:1.001159906387329 lr:8.704241428305316e-05 tokens_per_second_per_gpu:4435.502331849111 mfu:20.92302127406621 +Step 8573 | loss:1.028596043586731 lr:8.703950162327546e-05 tokens_per_second_per_gpu:5328.926723989132 mfu:25.137456565710476 +Step 8574 | loss:1.047886610031128 lr:8.703658868491894e-05 tokens_per_second_per_gpu:5223.379159814061 mfu:24.639570697225256 +Step 8575 | loss:1.158403992652893 lr:8.703367546800552e-05 tokens_per_second_per_gpu:3639.1238266099895 mfu:17.166367988669172 +Step 8576 | loss:0.9955596327781677 lr:8.703076197255708e-05 tokens_per_second_per_gpu:5349.547767362665 mfu:25.234729545616222 +Step 8577 | loss:1.2694038152694702 lr:8.702784819859558e-05 tokens_per_second_per_gpu:4734.435441343814 mfu:22.333139754800488 +Step 8578 | loss:1.137729287147522 lr:8.702493414614288e-05 tokens_per_second_per_gpu:4690.425958577455 mfu:22.125539515799534 +Step 8579 | loss:1.4194597005844116 lr:8.702201981522095e-05 tokens_per_second_per_gpu:3190.7835257982583 mfu:15.051470294997028 +Step 8580 | loss:1.023520588874817 lr:8.701910520585168e-05 tokens_per_second_per_gpu:5300.8551394240385 mfu:25.005038111810723 +Step 8581 | loss:1.1668059825897217 lr:8.701619031805699e-05 tokens_per_second_per_gpu:2810.400704701619 mfu:13.257139627882625 +Step 8582 | loss:0.9629144668579102 lr:8.701327515185882e-05 tokens_per_second_per_gpu:5113.267151327857 mfu:24.120153566149934 +Step 8583 | loss:1.1678736209869385 lr:8.701035970727907e-05 tokens_per_second_per_gpu:4418.435188474619 mfu:20.8425126468138 +Step 8584 | loss:1.0836390256881714 lr:8.700744398433969e-05 tokens_per_second_per_gpu:4137.114927303111 mfu:19.51547697668182 +Step 8585 | loss:1.0326224565505981 lr:8.70045279830626e-05 tokens_per_second_per_gpu:5321.996067735708 mfu:25.104763477671177 +Step 8586 | loss:1.0734277963638306 lr:8.700161170346974e-05 tokens_per_second_per_gpu:4076.9753258922797 mfu:19.23178821595286 +Step 8587 | loss:1.1694170236587524 lr:8.699869514558303e-05 tokens_per_second_per_gpu:4175.291639912805 mfu:19.695563043680586 +Step 8588 | loss:1.0029900074005127 lr:8.69957783094244e-05 tokens_per_second_per_gpu:4729.996598105669 mfu:22.31220097390141 +Step 8589 | loss:1.2872705459594727 lr:8.699286119501582e-05 tokens_per_second_per_gpu:3446.2442187716592 mfu:16.256521969841234 +Step 8590 | loss:1.0573196411132812 lr:8.69899438023792e-05 tokens_per_second_per_gpu:5028.501029051379 mfu:23.720297304779987 +Step 8591 | loss:1.1155222654342651 lr:8.69870261315365e-05 tokens_per_second_per_gpu:5299.814353540034 mfu:25.00012854721949 +Step 8592 | loss:1.1100033521652222 lr:8.698410818250966e-05 tokens_per_second_per_gpu:4101.178607665611 mfu:19.345959225583357 +Step 8593 | loss:1.0342522859573364 lr:8.698118995532061e-05 tokens_per_second_per_gpu:5331.9835294721715 mfu:25.151876038718765 +Step 8594 | loss:1.1828519105911255 lr:8.697827144999132e-05 tokens_per_second_per_gpu:4889.5912817292565 mfu:23.06503632621473 +Step 8595 | loss:1.1303235292434692 lr:8.697535266654373e-05 tokens_per_second_per_gpu:4051.005000984777 mfu:19.109281762369832 +Step 8596 | loss:1.0676084756851196 lr:8.69724336049998e-05 tokens_per_second_per_gpu:4324.230286350939 mfu:20.3981320504827 +Step 8597 | loss:1.2045512199401855 lr:8.696951426538147e-05 tokens_per_second_per_gpu:3300.113910803813 mfu:15.567200374755418 +Step 8598 | loss:1.1500269174575806 lr:8.69665946477107e-05 tokens_per_second_per_gpu:4602.101557539317 mfu:21.70889824640588 +Step 8599 | loss:1.016028642654419 lr:8.696367475200949e-05 tokens_per_second_per_gpu:5309.736449404312 mfu:25.04693276629541 +Step 8600 | loss:1.0973623991012573 lr:8.696075457829973e-05 tokens_per_second_per_gpu:3570.301411255301 mfu:16.841720913126217 +Step 8601 | loss:1.23542320728302 lr:8.695783412660343e-05 tokens_per_second_per_gpu:3933.0539222222906 mfu:18.552886399317973 +Step 8602 | loss:1.1735541820526123 lr:8.695491339694255e-05 tokens_per_second_per_gpu:5367.069742912652 mfu:25.317383693842817 +Step 8603 | loss:1.2070852518081665 lr:8.695199238933904e-05 tokens_per_second_per_gpu:5304.4122250513365 mfu:25.021817491616076 +Step 8604 | loss:1.0874048471450806 lr:8.694907110381488e-05 tokens_per_second_per_gpu:4572.232284020135 mfu:21.56799978699233 +Step 8605 | loss:0.9752560257911682 lr:8.694614954039204e-05 tokens_per_second_per_gpu:4660.93633034104 mfu:21.98643190795896 +Step 8606 | loss:1.3033697605133057 lr:8.694322769909252e-05 tokens_per_second_per_gpu:3785.287485653014 mfu:17.855846906466883 +Step 8607 | loss:1.1492836475372314 lr:8.694030557993824e-05 tokens_per_second_per_gpu:4358.64580944488 mfu:20.560475944810552 +Step 8608 | loss:1.0384089946746826 lr:8.693738318295121e-05 tokens_per_second_per_gpu:4631.680222726858 mfu:21.848425856736185 +Step 8609 | loss:1.0646272897720337 lr:8.693446050815341e-05 tokens_per_second_per_gpu:5066.693044420346 mfu:23.90045555750516 +Step 8610 | loss:1.1021130084991455 lr:8.693153755556684e-05 tokens_per_second_per_gpu:3879.800440975003 mfu:18.3016806422937 +Step 8611 | loss:1.1443238258361816 lr:8.692861432521344e-05 tokens_per_second_per_gpu:4531.443032281776 mfu:21.37558992717852 +Step 8612 | loss:1.0580497980117798 lr:8.692569081711523e-05 tokens_per_second_per_gpu:4784.378169533101 mfu:22.568728124777454 +Step 8613 | loss:1.1591709852218628 lr:8.692276703129421e-05 tokens_per_second_per_gpu:3857.9880992323765 mfu:18.198788104724475 +Step 8614 | loss:1.0699907541275024 lr:8.691984296777231e-05 tokens_per_second_per_gpu:5340.317226009737 mfu:25.19118750716171 +Step 8615 | loss:1.1733927726745605 lr:8.691691862657157e-05 tokens_per_second_per_gpu:2690.3939566494623 mfu:12.691047322058042 +Step 8616 | loss:1.062640905380249 lr:8.691399400771399e-05 tokens_per_second_per_gpu:5359.830820647534 mfu:25.283236462430757 +Step 8617 | loss:1.0660641193389893 lr:8.691106911122153e-05 tokens_per_second_per_gpu:5300.912991356413 mfu:25.005311009246288 +Step 8618 | loss:1.0051438808441162 lr:8.690814393711622e-05 tokens_per_second_per_gpu:5315.99874232028 mfu:25.07647307043738 +Step 8619 | loss:1.0703896284103394 lr:8.690521848542006e-05 tokens_per_second_per_gpu:5044.908067441376 mfu:23.797692104195015 +Step 8620 | loss:1.146668791770935 lr:8.690229275615503e-05 tokens_per_second_per_gpu:3726.350519245843 mfu:17.577831180240878 +Step 8621 | loss:1.2427302598953247 lr:8.689936674934314e-05 tokens_per_second_per_gpu:5355.622738865425 mfu:25.263386222691764 +Step 8622 | loss:1.1245336532592773 lr:8.689644046500642e-05 tokens_per_second_per_gpu:4348.371221816794 mfu:20.512008961943454 +Step 8623 | loss:1.1353411674499512 lr:8.689351390316686e-05 tokens_per_second_per_gpu:3567.9849120881945 mfu:16.830793591319463 +Step 8624 | loss:1.1166256666183472 lr:8.689058706384647e-05 tokens_per_second_per_gpu:4380.984791544854 mfu:20.665852734799493 +Step 8625 | loss:1.07034170627594 lr:8.688765994706727e-05 tokens_per_second_per_gpu:4276.853742016949 mfu:20.174648807587285 +Step 8626 | loss:1.077646255493164 lr:8.688473255285127e-05 tokens_per_second_per_gpu:4426.663626875437 mfu:20.881327594666814 +Step 8627 | loss:1.2173676490783691 lr:8.688180488122048e-05 tokens_per_second_per_gpu:3903.7600980610237 mfu:18.414702432707493 +Step 8628 | loss:1.0384174585342407 lr:8.687887693219693e-05 tokens_per_second_per_gpu:5339.96975899409 mfu:25.189548445964917 +Step 8629 | loss:1.0971099138259888 lr:8.687594870580266e-05 tokens_per_second_per_gpu:4724.228895446935 mfu:22.284993736388614 +Step 8630 | loss:1.140141248703003 lr:8.687302020205968e-05 tokens_per_second_per_gpu:4088.55303406343 mfu:19.286402240756885 +Step 8631 | loss:1.097469687461853 lr:8.687009142099e-05 tokens_per_second_per_gpu:5364.74506488157 mfu:25.306417790939232 +Step 8632 | loss:1.0867823362350464 lr:8.686716236261566e-05 tokens_per_second_per_gpu:4663.419338570494 mfu:21.998144681422183 +Step 8633 | loss:1.0283434391021729 lr:8.686423302695868e-05 tokens_per_second_per_gpu:5371.079129974703 mfu:25.336296656686482 +Step 8634 | loss:1.059037685394287 lr:8.686130341404108e-05 tokens_per_second_per_gpu:3903.4483381155524 mfu:18.413231807852974 +Step 8635 | loss:1.1432414054870605 lr:8.685837352388495e-05 tokens_per_second_per_gpu:4982.137795340715 mfu:23.501594020983422 +Step 8636 | loss:1.0109184980392456 lr:8.685544335651225e-05 tokens_per_second_per_gpu:5291.506593826248 mfu:24.96093942719959 +Step 8637 | loss:1.2461282014846802 lr:8.68525129119451e-05 tokens_per_second_per_gpu:5300.406330884588 mfu:25.002921005355503 +Step 8638 | loss:1.0816632509231567 lr:8.684958219020546e-05 tokens_per_second_per_gpu:5341.354545031055 mfu:25.196080717970144 +Step 8639 | loss:1.1008964776992798 lr:8.684665119131546e-05 tokens_per_second_per_gpu:5342.404739976734 mfu:25.201034666710644 +Step 8640 | loss:1.2533862590789795 lr:8.684371991529705e-05 tokens_per_second_per_gpu:3749.5186816316022 mfu:17.687119355110617 +Step 8641 | loss:1.0792440176010132 lr:8.684078836217233e-05 tokens_per_second_per_gpu:3642.1703288567705 mfu:17.18073885955412 +Step 8642 | loss:1.0937210321426392 lr:8.683785653196335e-05 tokens_per_second_per_gpu:4082.697821855078 mfu:19.258782205767876 +Step 8643 | loss:0.9929685592651367 lr:8.683492442469214e-05 tokens_per_second_per_gpu:4620.106108348498 mfu:21.793828784465923 +Step 8644 | loss:1.1669974327087402 lr:8.683199204038077e-05 tokens_per_second_per_gpu:5266.767689413671 mfu:24.844241794193128 +Step 8645 | loss:1.081159234046936 lr:8.68290593790513e-05 tokens_per_second_per_gpu:5179.583814665834 mfu:24.43298058190468 +Step 8646 | loss:1.2630987167358398 lr:8.682612644072577e-05 tokens_per_second_per_gpu:4039.4777152633874 mfu:19.05490558886429 +Step 8647 | loss:1.1269659996032715 lr:8.682319322542622e-05 tokens_per_second_per_gpu:3521.461796231448 mfu:16.611336116161116 +Step 8648 | loss:1.2415375709533691 lr:8.682025973317476e-05 tokens_per_second_per_gpu:4519.984677239166 mfu:21.32153891144556 +Step 8649 | loss:1.1103615760803223 lr:8.68173259639934e-05 tokens_per_second_per_gpu:4878.273001239954 mfu:23.011646066048915 +Step 8650 | loss:1.1319235563278198 lr:8.681439191790427e-05 tokens_per_second_per_gpu:5359.1319188408415 mfu:25.279939623363646 +Step 8651 | loss:1.1743049621582031 lr:8.681145759492939e-05 tokens_per_second_per_gpu:5335.6046938267855 mfu:25.16895768881383 +Step 8652 | loss:1.0260319709777832 lr:8.680852299509082e-05 tokens_per_second_per_gpu:4101.793879031677 mfu:19.34886156559321 +Step 8653 | loss:1.0871672630310059 lr:8.680558811841068e-05 tokens_per_second_per_gpu:5142.755445983084 mfu:24.25925488326051 +Step 8654 | loss:1.22939932346344 lr:8.6802652964911e-05 tokens_per_second_per_gpu:3781.270808239854 mfu:17.836899553793174 +Step 8655 | loss:0.9632711410522461 lr:8.679971753461387e-05 tokens_per_second_per_gpu:5310.871731699247 mfu:25.05228808657975 +Step 8656 | loss:1.0802139043807983 lr:8.679678182754136e-05 tokens_per_second_per_gpu:3580.159538727868 mfu:16.8882233823843 +Step 8657 | loss:1.246041178703308 lr:8.679384584371557e-05 tokens_per_second_per_gpu:5321.04273504845 mfu:25.100266444730053 +Step 8658 | loss:1.1552196741104126 lr:8.679090958315859e-05 tokens_per_second_per_gpu:3406.987722870231 mfu:16.07134238082534 +Step 8659 | loss:1.146290898323059 lr:8.678797304589246e-05 tokens_per_second_per_gpu:5237.3334231707195 mfu:24.705395338321452 +Step 8660 | loss:1.0434770584106445 lr:8.678503623193929e-05 tokens_per_second_per_gpu:5307.59863313668 mfu:25.036848322211924 +Step 8661 | loss:0.9831801056861877 lr:8.678209914132117e-05 tokens_per_second_per_gpu:4459.997143991915 mfu:21.03856748218972 +Step 8662 | loss:1.1578729152679443 lr:8.677916177406019e-05 tokens_per_second_per_gpu:5299.45059611197 mfu:24.9984126413679 +Step 8663 | loss:1.229561448097229 lr:8.677622413017846e-05 tokens_per_second_per_gpu:4497.1512194559755 mfu:21.213829595292633 +Step 8664 | loss:1.0459908246994019 lr:8.677328620969802e-05 tokens_per_second_per_gpu:3284.0932374253375 mfu:15.49162812502041 +Step 8665 | loss:1.0268211364746094 lr:8.677034801264103e-05 tokens_per_second_per_gpu:5027.41666369676 mfu:23.71518216839044 +Step 8666 | loss:1.186358094215393 lr:8.676740953902955e-05 tokens_per_second_per_gpu:5183.646507572826 mfu:24.452144997513845 +Step 8667 | loss:1.0608866214752197 lr:8.67644707888857e-05 tokens_per_second_per_gpu:4732.813486348936 mfu:22.32548871635559 +Step 8668 | loss:1.0071669816970825 lr:8.676153176223156e-05 tokens_per_second_per_gpu:3853.5173345927815 mfu:18.1776987451284 +Step 8669 | loss:1.1717828512191772 lr:8.675859245908926e-05 tokens_per_second_per_gpu:5333.768089982556 mfu:25.160294115123673 +Step 8670 | loss:1.1330006122589111 lr:8.675565287948088e-05 tokens_per_second_per_gpu:4860.338625178965 mfu:22.92704655424914 +Step 8671 | loss:1.173568606376648 lr:8.675271302342857e-05 tokens_per_second_per_gpu:5040.347567772949 mfu:23.776179449157564 +Step 8672 | loss:1.0697271823883057 lr:8.674977289095441e-05 tokens_per_second_per_gpu:4192.208528344242 mfu:19.775362892730467 +Step 8673 | loss:1.2740806341171265 lr:8.67468324820805e-05 tokens_per_second_per_gpu:5348.754784567746 mfu:25.230988910475777 +Step 8674 | loss:1.1767408847808838 lr:8.674389179682899e-05 tokens_per_second_per_gpu:4718.905677518341 mfu:22.2598832091848 +Step 8675 | loss:1.18084716796875 lr:8.674095083522197e-05 tokens_per_second_per_gpu:5361.947760818122 mfu:25.29322243040266 +Step 8676 | loss:0.9926834106445312 lr:8.673800959728158e-05 tokens_per_second_per_gpu:3469.330327518653 mfu:16.365423083697124 +Step 8677 | loss:1.0783249139785767 lr:8.673506808302994e-05 tokens_per_second_per_gpu:5358.29710723514 mfu:25.276001674586006 +Step 8678 | loss:1.1351453065872192 lr:8.673212629248914e-05 tokens_per_second_per_gpu:3662.91965481241 mfu:17.278616970286933 +Step 8679 | loss:1.1526302099227905 lr:8.672918422568134e-05 tokens_per_second_per_gpu:5317.037420494242 mfu:25.081372692600727 +Step 8680 | loss:1.129271388053894 lr:8.672624188262868e-05 tokens_per_second_per_gpu:4029.7315534723375 mfu:19.008931280829238 +Step 8681 | loss:1.158353328704834 lr:8.672329926335325e-05 tokens_per_second_per_gpu:5297.807508629537 mfu:24.990661917373625 +Step 8682 | loss:0.9906163215637207 lr:8.672035636787721e-05 tokens_per_second_per_gpu:4722.7958937737685 mfu:22.27823402300099 +Step 8683 | loss:1.0495691299438477 lr:8.671741319622267e-05 tokens_per_second_per_gpu:5335.446544643893 mfu:25.16821167213509 +Step 8684 | loss:1.080491542816162 lr:8.67144697484118e-05 tokens_per_second_per_gpu:4718.626627845954 mfu:22.25856688426937 +Step 8685 | loss:1.1321913003921509 lr:8.671152602446669e-05 tokens_per_second_per_gpu:4307.254391766816 mfu:20.31805385934311 +Step 8686 | loss:1.041043996810913 lr:8.670858202440953e-05 tokens_per_second_per_gpu:4094.983388329255 mfu:19.31673531896035 +Step 8687 | loss:1.0611838102340698 lr:8.670563774826243e-05 tokens_per_second_per_gpu:5366.75533313284 mfu:25.315900569267605 +Step 8688 | loss:1.0879440307617188 lr:8.670269319604756e-05 tokens_per_second_per_gpu:4628.301935502857 mfu:21.83248989950534 +Step 8689 | loss:1.1157170534133911 lr:8.669974836778704e-05 tokens_per_second_per_gpu:4053.6829922457982 mfu:19.121914304060496 +Step 8690 | loss:1.1628265380859375 lr:8.669680326350302e-05 tokens_per_second_per_gpu:5323.8392349305095 mfu:25.11345801180581 +Step 8691 | loss:1.126304030418396 lr:8.669385788321768e-05 tokens_per_second_per_gpu:5180.739454322599 mfu:24.438431931338094 +Step 8692 | loss:1.1374413967132568 lr:8.669091222695314e-05 tokens_per_second_per_gpu:3715.73862843921 mfu:17.52777308073072 +Step 8693 | loss:1.127208948135376 lr:8.668796629473157e-05 tokens_per_second_per_gpu:3642.055525029392 mfu:17.18019731031282 +Step 8694 | loss:1.0636664628982544 lr:8.668502008657513e-05 tokens_per_second_per_gpu:3778.477857759326 mfu:17.823724729850035 +Step 8695 | loss:0.9651951193809509 lr:8.668207360250596e-05 tokens_per_second_per_gpu:5343.912930202336 mfu:25.208149057328594 +Step 8696 | loss:1.105912685394287 lr:8.667912684254625e-05 tokens_per_second_per_gpu:4797.807149896581 mfu:22.632074916372304 +Step 8697 | loss:1.2999756336212158 lr:8.667617980671813e-05 tokens_per_second_per_gpu:4773.101311834533 mfu:22.515533262982984 +Step 8698 | loss:1.0982869863510132 lr:8.66732324950438e-05 tokens_per_second_per_gpu:4317.845637526773 mfu:20.368014572645695 +Step 8699 | loss:0.9463199377059937 lr:8.66702849075454e-05 tokens_per_second_per_gpu:4149.143583226928 mfu:19.572218198974795 +Step 8700 | loss:1.2093572616577148 lr:8.66673370442451e-05 tokens_per_second_per_gpu:5217.227875431853 mfu:24.610554039277048 +Step 8701 | loss:1.0205905437469482 lr:8.666438890516508e-05 tokens_per_second_per_gpu:5338.967846293897 mfu:25.184822252814175 +Step 8702 | loss:1.0944992303848267 lr:8.666144049032752e-05 tokens_per_second_per_gpu:5332.385692378408 mfu:25.153773109763552 +Step 8703 | loss:1.060723900794983 lr:8.665849179975459e-05 tokens_per_second_per_gpu:5323.643219421541 mfu:25.112533373206315 +Step 8704 | loss:1.081768274307251 lr:8.665554283346846e-05 tokens_per_second_per_gpu:5014.005189644761 mfu:23.651917957053737 +Step 8705 | loss:1.1062160730361938 lr:8.665259359149132e-05 tokens_per_second_per_gpu:3741.9691826530875 mfu:17.651507080351617 +Step 8706 | loss:1.1729110479354858 lr:8.664964407384535e-05 tokens_per_second_per_gpu:4970.0212929961945 mfu:23.444438412135845 +Step 8707 | loss:1.1260532140731812 lr:8.664669428055272e-05 tokens_per_second_per_gpu:3558.8739793092427 mfu:16.78781576691572 +Step 8708 | loss:1.063171625137329 lr:8.664374421163564e-05 tokens_per_second_per_gpu:5352.2366317420165 mfu:25.247413377661196 +Step 8709 | loss:1.2331359386444092 lr:8.664079386711627e-05 tokens_per_second_per_gpu:4960.750954696316 mfu:23.400708644692507 +Step 8710 | loss:1.0014944076538086 lr:8.663784324701681e-05 tokens_per_second_per_gpu:4938.977090065228 mfu:23.297997610223284 +Step 8711 | loss:1.1991450786590576 lr:8.663489235135947e-05 tokens_per_second_per_gpu:4501.853041268213 mfu:21.236008890994484 +Step 8712 | loss:1.0850141048431396 lr:8.663194118016643e-05 tokens_per_second_per_gpu:5283.0288043021455 mfu:24.920948247554353 +Step 8713 | loss:1.0264586210250854 lr:8.662898973345987e-05 tokens_per_second_per_gpu:5333.238859652888 mfu:25.15779764536262 +Step 8714 | loss:1.1589921712875366 lr:8.662603801126202e-05 tokens_per_second_per_gpu:4422.214208794355 mfu:20.860338930429347 +Step 8715 | loss:1.1184580326080322 lr:8.662308601359505e-05 tokens_per_second_per_gpu:5381.833396093397 mfu:25.387026364835368 +Step 8716 | loss:1.1494166851043701 lr:8.66201337404812e-05 tokens_per_second_per_gpu:4249.020398090737 mfu:20.043354175429187 +Step 8717 | loss:1.0838499069213867 lr:8.661718119194264e-05 tokens_per_second_per_gpu:4156.612588772817 mfu:19.607450772478504 +Step 8718 | loss:1.068753719329834 lr:8.661422836800159e-05 tokens_per_second_per_gpu:4352.257879438638 mfu:20.530342989077866 +Step 8719 | loss:1.1133171319961548 lr:8.661127526868025e-05 tokens_per_second_per_gpu:5360.83610036444 mfu:25.28797853837381 +Step 8720 | loss:1.0069761276245117 lr:8.660832189400083e-05 tokens_per_second_per_gpu:5318.490770619335 mfu:25.088228393107894 +Step 8721 | loss:1.014335036277771 lr:8.660536824398555e-05 tokens_per_second_per_gpu:5332.73600907899 mfu:25.15542561341039 +Step 8722 | loss:1.1929455995559692 lr:8.660241431865665e-05 tokens_per_second_per_gpu:4926.238333724897 mfu:23.237906723109987 +Step 8723 | loss:1.1919562816619873 lr:8.65994601180363e-05 tokens_per_second_per_gpu:4263.221869023976 mfu:20.11034493684207 +Step 8724 | loss:1.1307307481765747 lr:8.659650564214675e-05 tokens_per_second_per_gpu:4677.942563242208 mfu:22.066653210116304 +Step 8725 | loss:0.996303379535675 lr:8.65935508910102e-05 tokens_per_second_per_gpu:5322.92128386493 mfu:25.10912788001087 +Step 8726 | loss:1.017573356628418 lr:8.659059586464889e-05 tokens_per_second_per_gpu:5116.925184265192 mfu:24.137409131640116 +Step 8727 | loss:1.098190188407898 lr:8.658764056308504e-05 tokens_per_second_per_gpu:5033.446939126458 mfu:23.743628006463414 +Step 8728 | loss:1.0671658515930176 lr:8.658468498634088e-05 tokens_per_second_per_gpu:5341.354074724574 mfu:25.19607849945423 +Step 8729 | loss:1.0899406671524048 lr:8.658172913443862e-05 tokens_per_second_per_gpu:5355.41778793417 mfu:25.262419434180796 +Step 8730 | loss:1.0004440546035767 lr:8.657877300740053e-05 tokens_per_second_per_gpu:5360.28564572855 mfu:25.28538194993907 +Step 8731 | loss:1.1180996894836426 lr:8.657581660524881e-05 tokens_per_second_per_gpu:3593.1404474955734 mfu:16.949456543812982 +Step 8732 | loss:1.1039295196533203 lr:8.657285992800571e-05 tokens_per_second_per_gpu:4440.851528331923 mfu:20.948254346542665 +Step 8733 | loss:1.1580703258514404 lr:8.656990297569346e-05 tokens_per_second_per_gpu:3896.0194469202056 mfu:18.378188460584987 +Step 8734 | loss:1.0125921964645386 lr:8.65669457483343e-05 tokens_per_second_per_gpu:4298.116272885692 mfu:20.274947793456995 +Step 8735 | loss:1.3060266971588135 lr:8.656398824595049e-05 tokens_per_second_per_gpu:3986.0194963665977 mfu:18.80273404941547 +Step 8736 | loss:1.008391261100769 lr:8.656103046856425e-05 tokens_per_second_per_gpu:4919.163896006246 mfu:23.204535393367575 +Step 8737 | loss:1.075060486793518 lr:8.655807241619783e-05 tokens_per_second_per_gpu:4366.020327494662 mfu:20.595262804673894 +Step 8738 | loss:1.0427988767623901 lr:8.65551140888735e-05 tokens_per_second_per_gpu:4875.4349617990965 mfu:22.99825854978693 +Step 8739 | loss:1.0934427976608276 lr:8.655215548661346e-05 tokens_per_second_per_gpu:4703.704294098637 mfu:22.1881756899706 +Step 8740 | loss:1.1383949518203735 lr:8.654919660944002e-05 tokens_per_second_per_gpu:5339.536892280584 mfu:25.187506539822632 +Step 8741 | loss:1.1748061180114746 lr:8.65462374573754e-05 tokens_per_second_per_gpu:3639.7272024089266 mfu:17.16921421525928 +Step 8742 | loss:1.1699274778366089 lr:8.654327803044186e-05 tokens_per_second_per_gpu:4510.316167064434 mfu:21.275930899333417 +Step 8743 | loss:1.0315724611282349 lr:8.654031832866167e-05 tokens_per_second_per_gpu:3937.9781054010473 mfu:18.576114611524453 +Step 8744 | loss:1.182169795036316 lr:8.653735835205708e-05 tokens_per_second_per_gpu:3925.8170548776657 mfu:18.518748861316485 +Step 8745 | loss:1.1988340616226196 lr:8.653439810065035e-05 tokens_per_second_per_gpu:5380.926295736672 mfu:25.382747417685415 +Step 8746 | loss:0.9419161081314087 lr:8.653143757446377e-05 tokens_per_second_per_gpu:4446.878420477314 mfu:20.976684225086306 +Step 8747 | loss:0.9366931915283203 lr:8.652847677351959e-05 tokens_per_second_per_gpu:5347.8222087448585 mfu:25.226589791205374 +Step 8748 | loss:1.0175882577896118 lr:8.652551569784004e-05 tokens_per_second_per_gpu:5309.1668334061005 mfu:25.044245790445334 +Step 8749 | loss:1.247621774673462 lr:8.652255434744744e-05 tokens_per_second_per_gpu:4461.606194523334 mfu:21.04615764807875 +Step 8750 | loss:1.2745015621185303 lr:8.651959272236406e-05 tokens_per_second_per_gpu:4234.724333255664 mfu:19.975917198441962 +Step 8751 | loss:1.1010569334030151 lr:8.651663082261216e-05 tokens_per_second_per_gpu:5317.369333493803 mfu:25.082938382849857 +Step 8752 | loss:1.072896957397461 lr:8.651366864821402e-05 tokens_per_second_per_gpu:4977.102455325818 mfu:23.477841463015555 +Step 8753 | loss:1.2702165842056274 lr:8.651070619919192e-05 tokens_per_second_per_gpu:5340.928604543939 mfu:25.194071484019478 +Step 8754 | loss:1.1347832679748535 lr:8.650774347556814e-05 tokens_per_second_per_gpu:2786.9681757952853 mfu:13.146604391029758 +Step 8755 | loss:1.209189534187317 lr:8.650478047736494e-05 tokens_per_second_per_gpu:3301.194780900497 mfu:15.57229902341693 +Step 8756 | loss:1.366791009902954 lr:8.650181720460466e-05 tokens_per_second_per_gpu:5070.789489147396 mfu:23.9197791862082 +Step 8757 | loss:1.2906150817871094 lr:8.649885365730954e-05 tokens_per_second_per_gpu:4232.570317433698 mfu:19.965756338296753 +Step 8758 | loss:1.0188772678375244 lr:8.649588983550189e-05 tokens_per_second_per_gpu:3543.919207261385 mfu:16.717271555618755 +Step 8759 | loss:1.1799532175064087 lr:8.649292573920397e-05 tokens_per_second_per_gpu:5336.479731750304 mfu:25.173085391995933 +Step 8760 | loss:1.091488242149353 lr:8.648996136843815e-05 tokens_per_second_per_gpu:5318.855446347824 mfu:25.08994863073927 +Step 8761 | loss:1.1990865468978882 lr:8.648699672322663e-05 tokens_per_second_per_gpu:4203.670999656162 mfu:19.829433325608065 +Step 8762 | loss:1.0456267595291138 lr:8.648403180359176e-05 tokens_per_second_per_gpu:4160.596317481076 mfu:19.62624270048957 +Step 8763 | loss:1.0452417135238647 lr:8.648106660955584e-05 tokens_per_second_per_gpu:5167.025661584911 mfu:24.37374163118011 +Step 8764 | loss:1.1784968376159668 lr:8.647810114114117e-05 tokens_per_second_per_gpu:5006.142171568603 mfu:23.614826759219277 +Step 8765 | loss:1.078969120979309 lr:8.647513539837005e-05 tokens_per_second_per_gpu:4792.429243050047 mfu:22.60670641221168 +Step 8766 | loss:1.219407320022583 lr:8.647216938126477e-05 tokens_per_second_per_gpu:4552.427160198774 mfu:21.474575638824405 +Step 8767 | loss:1.2405451536178589 lr:8.646920308984763e-05 tokens_per_second_per_gpu:3742.199554354385 mfu:17.6525937829721 +Step 8768 | loss:1.1122227907180786 lr:8.646623652414098e-05 tokens_per_second_per_gpu:4663.152902299936 mfu:21.996887856075244 +Step 8769 | loss:1.1773167848587036 lr:8.646326968416713e-05 tokens_per_second_per_gpu:4528.688441623121 mfu:21.362596053060017 +Step 8770 | loss:0.9911858439445496 lr:8.646030256994835e-05 tokens_per_second_per_gpu:4436.103223486736 mfu:20.92585578244357 +Step 8771 | loss:1.0917145013809204 lr:8.645733518150701e-05 tokens_per_second_per_gpu:3613.5533177664656 mfu:17.045747535675584 +Step 8772 | loss:1.119856357574463 lr:8.645436751886538e-05 tokens_per_second_per_gpu:4856.610037175195 mfu:22.909458168472646 +Step 8773 | loss:1.1836506128311157 lr:8.645139958204579e-05 tokens_per_second_per_gpu:5327.914295805641 mfu:25.132680769230888 +Step 8774 | loss:1.1490821838378906 lr:8.644843137107059e-05 tokens_per_second_per_gpu:4597.857169063186 mfu:21.688876741796623 +Step 8775 | loss:1.0697968006134033 lr:8.644546288596207e-05 tokens_per_second_per_gpu:5315.637678565171 mfu:25.0747698711007 +Step 8776 | loss:1.0578691959381104 lr:8.644249412674258e-05 tokens_per_second_per_gpu:3792.655610137799 mfu:17.890603606793285 +Step 8777 | loss:1.0854425430297852 lr:8.643952509343445e-05 tokens_per_second_per_gpu:5395.0014193687075 mfu:25.449142177321452 +Step 8778 | loss:1.1451154947280884 lr:8.643655578606001e-05 tokens_per_second_per_gpu:5360.174785412263 mfu:25.284859002912423 +Step 8779 | loss:1.1562694311141968 lr:8.643358620464155e-05 tokens_per_second_per_gpu:4685.054103843516 mfu:22.10019956048661 +Step 8780 | loss:1.048055648803711 lr:8.643061634920146e-05 tokens_per_second_per_gpu:5320.968823595872 mfu:25.09991779179765 +Step 8781 | loss:1.091342806816101 lr:8.642764621976207e-05 tokens_per_second_per_gpu:3739.2064327262246 mfu:17.638474717572862 +Step 8782 | loss:0.9627646803855896 lr:8.642467581634569e-05 tokens_per_second_per_gpu:4823.723600122621 mfu:22.754327233890905 +Step 8783 | loss:1.1591979265213013 lr:8.642170513897467e-05 tokens_per_second_per_gpu:4978.845904634211 mfu:23.486065610866373 +Step 8784 | loss:1.1497113704681396 lr:8.641873418767136e-05 tokens_per_second_per_gpu:4087.5140528670636 mfu:19.281501189185153 +Step 8785 | loss:1.0899875164031982 lr:8.641576296245812e-05 tokens_per_second_per_gpu:5055.8108198381615 mfu:23.849122247452023 +Step 8786 | loss:1.103175401687622 lr:8.641279146335727e-05 tokens_per_second_per_gpu:5382.444797561878 mfu:25.389910449878037 +Step 8787 | loss:1.1792330741882324 lr:8.640981969039117e-05 tokens_per_second_per_gpu:4864.5586490073665 mfu:22.946953126658023 +Step 8788 | loss:1.1064186096191406 lr:8.640684764358217e-05 tokens_per_second_per_gpu:4855.063713311636 mfu:22.90216389497818 +Step 8789 | loss:1.1119824647903442 lr:8.640387532295262e-05 tokens_per_second_per_gpu:4762.657934310788 mfu:22.46627006937954 +Step 8790 | loss:1.1094086170196533 lr:8.640090272852488e-05 tokens_per_second_per_gpu:3923.3804854334235 mfu:18.507255147526383 +Step 8791 | loss:1.1506469249725342 lr:8.63979298603213e-05 tokens_per_second_per_gpu:3769.2920622204347 mfu:17.780393765034756 +Step 8792 | loss:1.414788842201233 lr:8.639495671836427e-05 tokens_per_second_per_gpu:3136.0821049406645 mfu:14.79343435978697 +Step 8793 | loss:1.2073854207992554 lr:8.639198330267608e-05 tokens_per_second_per_gpu:5380.366769611967 mfu:25.380108037490185 +Step 8794 | loss:1.2051846981048584 lr:8.638900961327918e-05 tokens_per_second_per_gpu:5299.33324682667 mfu:24.997859084767285 +Step 8795 | loss:1.1007498502731323 lr:8.638603565019587e-05 tokens_per_second_per_gpu:5344.555337709371 mfu:25.211179403145305 +Step 8796 | loss:1.1600754261016846 lr:8.638306141344856e-05 tokens_per_second_per_gpu:5329.48564264642 mfu:25.14009308037826 +Step 8797 | loss:1.0304183959960938 lr:8.638008690305961e-05 tokens_per_second_per_gpu:4358.718258997564 mfu:20.56081770171173 +Step 8798 | loss:0.8867979049682617 lr:8.637711211905138e-05 tokens_per_second_per_gpu:4709.484010198977 mfu:22.215439596937156 +Step 8799 | loss:1.0735664367675781 lr:8.637413706144623e-05 tokens_per_second_per_gpu:3864.8859322991702 mfu:18.231326360192217 +Step 8800 | loss:1.0747965574264526 lr:8.637116173026655e-05 tokens_per_second_per_gpu:5125.359887282562 mfu:24.177197064881785 +Step 8801 | loss:0.9693857431411743 lr:8.636818612553474e-05 tokens_per_second_per_gpu:5311.48193022689 mfu:25.055166497145247 +Step 8802 | loss:1.1165393590927124 lr:8.636521024727315e-05 tokens_per_second_per_gpu:4371.057423150017 mfu:20.619023644297343 +Step 8803 | loss:1.0666345357894897 lr:8.63622340955042e-05 tokens_per_second_per_gpu:5348.906888192647 mfu:25.23170640922599 +Step 8804 | loss:1.113147497177124 lr:8.635925767025021e-05 tokens_per_second_per_gpu:4965.68110697389 mfu:23.423965014156543 +Step 8805 | loss:1.1482329368591309 lr:8.635628097153363e-05 tokens_per_second_per_gpu:4751.667307401644 mfu:22.414425407054793 +Step 8806 | loss:1.0365855693817139 lr:8.635330399937682e-05 tokens_per_second_per_gpu:5313.832119358704 mfu:25.06625274026363 +Step 8807 | loss:1.0892261266708374 lr:8.635032675380217e-05 tokens_per_second_per_gpu:5043.372971618827 mfu:23.790450795285693 +Step 8808 | loss:1.1709800958633423 lr:8.634734923483206e-05 tokens_per_second_per_gpu:5180.720717837293 mfu:24.438343548140132 +Step 8809 | loss:1.2937370538711548 lr:8.634437144248891e-05 tokens_per_second_per_gpu:3399.5180785147295 mfu:16.036106793947692 +Step 8810 | loss:1.016386866569519 lr:8.634139337679509e-05 tokens_per_second_per_gpu:4031.1470178812 mfu:19.01560826794897 +Step 8811 | loss:1.1349635124206543 lr:8.633841503777303e-05 tokens_per_second_per_gpu:5332.285653061119 mfu:25.15330120723543 +Step 8812 | loss:1.1242375373840332 lr:8.633543642544512e-05 tokens_per_second_per_gpu:3466.87828852444 mfu:16.353856397400428 +Step 8813 | loss:0.9833395481109619 lr:8.633245753983374e-05 tokens_per_second_per_gpu:5072.340442715152 mfu:23.927095298807263 +Step 8814 | loss:1.0921190977096558 lr:8.632947838096133e-05 tokens_per_second_per_gpu:4207.416217247961 mfu:19.847100155988596 +Step 8815 | loss:1.2283086776733398 lr:8.632649894885025e-05 tokens_per_second_per_gpu:5102.481901740349 mfu:24.069277703692325 +Step 8816 | loss:1.1064834594726562 lr:8.632351924352295e-05 tokens_per_second_per_gpu:3198.0912756480147 mfu:15.085942197869022 +Step 8817 | loss:1.0037014484405518 lr:8.632053926500183e-05 tokens_per_second_per_gpu:5061.681346732618 mfu:23.876814524428347 +Step 8818 | loss:1.2073438167572021 lr:8.63175590133093e-05 tokens_per_second_per_gpu:4695.877029715183 mfu:22.151253148404837 +Step 8819 | loss:1.1577017307281494 lr:8.631457848846778e-05 tokens_per_second_per_gpu:3907.183297252722 mfu:18.43085024735274 +Step 8820 | loss:1.0961101055145264 lr:8.631159769049965e-05 tokens_per_second_per_gpu:4581.013509324315 mfu:21.60942232498368 +Step 8821 | loss:1.0035507678985596 lr:8.63086166194274e-05 tokens_per_second_per_gpu:5350.866870948652 mfu:25.240951982293605 +Step 8822 | loss:1.095099925994873 lr:8.63056352752734e-05 tokens_per_second_per_gpu:5334.251803554622 mfu:25.16257587457315 +Step 8823 | loss:1.0693155527114868 lr:8.630265365806007e-05 tokens_per_second_per_gpu:4039.50683308871 mfu:19.05504294261435 +Step 8824 | loss:0.9876213073730469 lr:8.629967176780986e-05 tokens_per_second_per_gpu:4138.041953675759 mfu:19.519849918248614 +Step 8825 | loss:1.1934685707092285 lr:8.629668960454518e-05 tokens_per_second_per_gpu:4004.1982390943535 mfu:18.888486280475632 +Step 8826 | loss:1.0581270456314087 lr:8.629370716828847e-05 tokens_per_second_per_gpu:4626.020619886356 mfu:21.821728544509725 +Step 8827 | loss:1.1525533199310303 lr:8.629072445906217e-05 tokens_per_second_per_gpu:3547.3052759760676 mfu:16.733244219468812 +Step 8828 | loss:1.179598093032837 lr:8.628774147688869e-05 tokens_per_second_per_gpu:4077.334483045808 mfu:19.233482421524588 +Step 8829 | loss:1.1308636665344238 lr:8.628475822179048e-05 tokens_per_second_per_gpu:5129.685325244119 mfu:24.197600893742262 +Step 8830 | loss:1.22025728225708 lr:8.628177469378995e-05 tokens_per_second_per_gpu:4633.827553223582 mfu:21.858555181061593 +Step 8831 | loss:1.2219089269638062 lr:8.627879089290958e-05 tokens_per_second_per_gpu:5248.209927088282 mfu:24.75670166302352 +Step 8832 | loss:1.1287339925765991 lr:8.62758068191718e-05 tokens_per_second_per_gpu:4384.978374068633 mfu:20.6846911449396 +Step 8833 | loss:1.003024935722351 lr:8.627282247259905e-05 tokens_per_second_per_gpu:5361.031397610972 mfu:25.288899788806848 +Step 8834 | loss:1.0228550434112549 lr:8.626983785321377e-05 tokens_per_second_per_gpu:3517.62996957504 mfu:16.5932607360458 +Step 8835 | loss:0.9957488775253296 lr:8.62668529610384e-05 tokens_per_second_per_gpu:4875.9071759006565 mfu:23.00048606427202 +Step 8836 | loss:1.0563867092132568 lr:8.626386779609541e-05 tokens_per_second_per_gpu:3423.8647926219733 mfu:16.150954398369333 +Step 8837 | loss:1.0551923513412476 lr:8.626088235840726e-05 tokens_per_second_per_gpu:3326.892432851607 mfu:15.693519232141853 +Step 8838 | loss:1.1809989213943481 lr:8.625789664799635e-05 tokens_per_second_per_gpu:3252.189621014107 mfu:15.34113332309038 +Step 8839 | loss:1.0006762742996216 lr:8.625491066488521e-05 tokens_per_second_per_gpu:4806.545503846834 mfu:22.67329522287257 +Step 8840 | loss:1.155836820602417 lr:8.625192440909624e-05 tokens_per_second_per_gpu:5361.788323978504 mfu:25.292470339627453 +Step 8841 | loss:1.1647355556488037 lr:8.624893788065192e-05 tokens_per_second_per_gpu:5329.859945224792 mfu:25.141858729503042 +Step 8842 | loss:1.1316123008728027 lr:8.624595107957471e-05 tokens_per_second_per_gpu:4260.902761119281 mfu:20.099405309175012 +Step 8843 | loss:1.1637372970581055 lr:8.62429640058871e-05 tokens_per_second_per_gpu:3277.6936921907127 mfu:15.461440378273679 +Step 8844 | loss:1.0587265491485596 lr:8.623997665961151e-05 tokens_per_second_per_gpu:5333.127095927667 mfu:25.1572704368019 +Step 8845 | loss:1.021761178970337 lr:8.623698904077044e-05 tokens_per_second_per_gpu:4225.06453496324 mfu:19.930350281765058 +Step 8846 | loss:0.9881749749183655 lr:8.623400114938636e-05 tokens_per_second_per_gpu:5391.518971994234 mfu:25.432714878889264 +Step 8847 | loss:1.2550945281982422 lr:8.62310129854817e-05 tokens_per_second_per_gpu:4430.403373542572 mfu:20.898968617762268 +Step 8848 | loss:1.0543270111083984 lr:8.622802454907901e-05 tokens_per_second_per_gpu:5366.637576277022 mfu:25.31534509008732 +Step 8849 | loss:1.3747926950454712 lr:8.62250358402007e-05 tokens_per_second_per_gpu:3697.9401626518634 mfu:17.443814680880404 +Step 8850 | loss:1.0422914028167725 lr:8.622204685886927e-05 tokens_per_second_per_gpu:3337.623739118238 mfu:15.7441406347512 +Step 8851 | loss:1.3410298824310303 lr:8.621905760510722e-05 tokens_per_second_per_gpu:3430.544567793833 mfu:16.18246403754216 +Step 8852 | loss:1.0208470821380615 lr:8.621606807893702e-05 tokens_per_second_per_gpu:5358.585919152665 mfu:25.277364049677363 +Step 8853 | loss:1.1472009420394897 lr:8.621307828038114e-05 tokens_per_second_per_gpu:3993.6192691969777 mfu:18.838583474511324 +Step 8854 | loss:1.123818039894104 lr:8.621008820946208e-05 tokens_per_second_per_gpu:5182.501612980191 mfu:24.446744334381226 +Step 8855 | loss:1.2267614603042603 lr:8.620709786620231e-05 tokens_per_second_per_gpu:5360.663581609164 mfu:25.28716473797044 +Step 8856 | loss:1.1932990550994873 lr:8.620410725062436e-05 tokens_per_second_per_gpu:5330.425495902114 mfu:25.144526528540865 +Step 8857 | loss:1.1052559614181519 lr:8.62011163627507e-05 tokens_per_second_per_gpu:3391.123137655967 mfu:15.996506425592303 +Step 8858 | loss:1.1505424976348877 lr:8.619812520260381e-05 tokens_per_second_per_gpu:5355.768862941825 mfu:25.26407551489148 +Step 8859 | loss:1.050838828086853 lr:8.61951337702062e-05 tokens_per_second_per_gpu:5245.413619819053 mfu:24.74351100453535 +Step 8860 | loss:1.0940145254135132 lr:8.619214206558039e-05 tokens_per_second_per_gpu:4873.402530216207 mfu:22.98867122324185 +Step 8861 | loss:1.1112338304519653 lr:8.618915008874884e-05 tokens_per_second_per_gpu:5355.240777695743 mfu:25.26158444668508 +Step 8862 | loss:1.034326195716858 lr:8.618615783973409e-05 tokens_per_second_per_gpu:4608.480043052803 mfu:21.738986650854844 +Step 8863 | loss:1.1357700824737549 lr:8.618316531855862e-05 tokens_per_second_per_gpu:5331.993309892174 mfu:25.15192217462863 +Step 8864 | loss:1.0612305402755737 lr:8.618017252524495e-05 tokens_per_second_per_gpu:5344.015019901344 mfu:25.208630631857115 +Step 8865 | loss:1.0549869537353516 lr:8.617717945981558e-05 tokens_per_second_per_gpu:4337.013993667757 mfu:20.458434979021508 +Step 8866 | loss:1.1486002206802368 lr:8.617418612229304e-05 tokens_per_second_per_gpu:4976.372951840781 mfu:23.47440027061051 +Step 8867 | loss:1.2032634019851685 lr:8.617119251269982e-05 tokens_per_second_per_gpu:4638.723193717348 mfu:21.88164875255309 +Step 8868 | loss:0.9285176396369934 lr:8.616819863105845e-05 tokens_per_second_per_gpu:5329.287525678033 mfu:25.139158528836134 +Step 8869 | loss:1.0321232080459595 lr:8.616520447739144e-05 tokens_per_second_per_gpu:5322.962363444051 mfu:25.10932165939449 +Step 8870 | loss:1.2150696516036987 lr:8.616221005172132e-05 tokens_per_second_per_gpu:4651.932056084582 mfu:21.94395721000364 +Step 8871 | loss:1.0070087909698486 lr:8.61592153540706e-05 tokens_per_second_per_gpu:4003.4779272982328 mfu:18.885088446835468 +Step 8872 | loss:1.2123950719833374 lr:8.615622038446179e-05 tokens_per_second_per_gpu:4828.031797474267 mfu:22.774649735852883 +Step 8873 | loss:1.1444933414459229 lr:8.615322514291745e-05 tokens_per_second_per_gpu:3660.2011885590996 mfu:17.265793501151897 +Step 8874 | loss:1.1964240074157715 lr:8.61502296294601e-05 tokens_per_second_per_gpu:4208.757279631658 mfu:19.85342617606133 +Step 8875 | loss:1.0779755115509033 lr:8.614723384411223e-05 tokens_per_second_per_gpu:5342.217641802347 mfu:25.20015209269982 +Step 8876 | loss:1.0498337745666504 lr:8.614423778689642e-05 tokens_per_second_per_gpu:5339.467362736778 mfu:25.18717855710154 +Step 8877 | loss:1.0744413137435913 lr:8.61412414578352e-05 tokens_per_second_per_gpu:3598.451870102966 mfu:16.974511430474195 +Step 8878 | loss:1.1547271013259888 lr:8.613824485695106e-05 tokens_per_second_per_gpu:5319.659553643021 mfu:25.093741742045566 +Step 8879 | loss:1.099489688873291 lr:8.61352479842666e-05 tokens_per_second_per_gpu:4711.021286161102 mfu:22.222691189936825 +Step 8880 | loss:1.144225835800171 lr:8.613225083980431e-05 tokens_per_second_per_gpu:3666.259224079119 mfu:17.29437028830727 +Step 8881 | loss:1.094651460647583 lr:8.612925342358675e-05 tokens_per_second_per_gpu:4293.950825603015 mfu:20.255298667926017 +Step 8882 | loss:0.9684179425239563 lr:8.612625573563648e-05 tokens_per_second_per_gpu:3991.4590854786675 mfu:18.828393519346704 +Step 8883 | loss:1.0561524629592896 lr:8.612325777597602e-05 tokens_per_second_per_gpu:3821.3806094887436 mfu:18.026104329721953 +Step 8884 | loss:1.2053039073944092 lr:8.612025954462792e-05 tokens_per_second_per_gpu:5235.223250928953 mfu:24.695441295824534 +Step 8885 | loss:1.1645444631576538 lr:8.611726104161475e-05 tokens_per_second_per_gpu:4372.724581464684 mfu:20.626887914514167 +Step 8886 | loss:1.2563995122909546 lr:8.611426226695907e-05 tokens_per_second_per_gpu:4921.485343692308 mfu:23.215486058182545 +Step 8887 | loss:1.204188346862793 lr:8.61112632206834e-05 tokens_per_second_per_gpu:3939.25546167797 mfu:18.582140119022153 +Step 8888 | loss:1.1048860549926758 lr:8.61082639028103e-05 tokens_per_second_per_gpu:4134.64066177147 mfu:19.503805444015843 +Step 8889 | loss:1.0178134441375732 lr:8.610526431336234e-05 tokens_per_second_per_gpu:5354.419098983673 mfu:25.257708448007403 +Step 8890 | loss:1.0219786167144775 lr:8.610226445236209e-05 tokens_per_second_per_gpu:4513.637420221734 mfu:21.29159781714062 +Step 8891 | loss:1.0251926183700562 lr:8.60992643198321e-05 tokens_per_second_per_gpu:4441.329769055216 mfu:20.950510289630717 +Step 8892 | loss:1.126672387123108 lr:8.609626391579492e-05 tokens_per_second_per_gpu:5322.3013436449055 mfu:25.106203516220308 +Step 8893 | loss:1.11259925365448 lr:8.609326324027316e-05 tokens_per_second_per_gpu:4567.673920512509 mfu:21.546497208589845 +Step 8894 | loss:0.9881513714790344 lr:8.609026229328935e-05 tokens_per_second_per_gpu:4219.594799617039 mfu:19.90454860690395 +Step 8895 | loss:1.0889265537261963 lr:8.608726107486608e-05 tokens_per_second_per_gpu:5253.303813641203 mfu:24.780730395000315 +Step 8896 | loss:1.0720123052597046 lr:8.60842595850259e-05 tokens_per_second_per_gpu:5297.581933522891 mfu:24.989597841108495 +Step 8897 | loss:1.0928784608840942 lr:8.608125782379141e-05 tokens_per_second_per_gpu:5274.812420055028 mfu:24.882190161200995 +Step 8898 | loss:1.1267385482788086 lr:8.607825579118517e-05 tokens_per_second_per_gpu:4987.49949844852 mfu:23.526886089343822 +Step 8899 | loss:1.2316691875457764 lr:8.607525348722977e-05 tokens_per_second_per_gpu:4731.346556247447 mfu:22.318568956782784 +Step 8900 | loss:1.1591744422912598 lr:8.607225091194779e-05 tokens_per_second_per_gpu:4451.8544666550715 mfu:21.000157083907816 +Step 8901 | loss:1.025469422340393 lr:8.606924806536181e-05 tokens_per_second_per_gpu:4545.853750362619 mfu:21.443567743087893 +Step 8902 | loss:1.1116682291030884 lr:8.60662449474944e-05 tokens_per_second_per_gpu:5351.766978620457 mfu:25.245197943755713 +Step 8903 | loss:1.0411487817764282 lr:8.606324155836816e-05 tokens_per_second_per_gpu:5319.224475993657 mfu:25.091689406541615 +Step 8904 | loss:1.0775907039642334 lr:8.606023789800568e-05 tokens_per_second_per_gpu:3390.093668270076 mfu:15.991650242853463 +Step 8905 | loss:1.063226342201233 lr:8.605723396642956e-05 tokens_per_second_per_gpu:3806.123353071349 mfu:17.954133248044727 +Step 8906 | loss:1.0803172588348389 lr:8.605422976366237e-05 tokens_per_second_per_gpu:5226.383616257053 mfu:24.653743230879982 +Step 8907 | loss:0.9723978638648987 lr:8.605122528972671e-05 tokens_per_second_per_gpu:5340.785572054605 mfu:25.19339677536343 +Step 8908 | loss:1.0891695022583008 lr:8.604822054464521e-05 tokens_per_second_per_gpu:5321.270170192472 mfu:25.101339295108847 +Step 8909 | loss:1.004375696182251 lr:8.604521552844043e-05 tokens_per_second_per_gpu:5345.122618081527 mfu:25.213855361448775 +Step 8910 | loss:1.0655075311660767 lr:8.604221024113498e-05 tokens_per_second_per_gpu:4439.644968982167 mfu:20.94256280022961 +Step 8911 | loss:1.1605331897735596 lr:8.603920468275147e-05 tokens_per_second_per_gpu:5364.91192054467 mfu:25.307204877570584 +Step 8912 | loss:1.1235915422439575 lr:8.60361988533125e-05 tokens_per_second_per_gpu:4433.893433366417 mfu:20.915431825416956 +Step 8913 | loss:1.1711039543151855 lr:8.60331927528407e-05 tokens_per_second_per_gpu:4221.784756398534 mfu:19.914879006687418 +Step 8914 | loss:1.1355595588684082 lr:8.603018638135863e-05 tokens_per_second_per_gpu:5338.264810717679 mfu:25.181505914050916 +Step 8915 | loss:1.0854496955871582 lr:8.602717973888896e-05 tokens_per_second_per_gpu:5313.227828473997 mfu:25.06340219705781 +Step 8916 | loss:1.0981558561325073 lr:8.602417282545425e-05 tokens_per_second_per_gpu:5159.725665036117 mfu:24.3393063019516 +Step 8917 | loss:1.0777809619903564 lr:8.602116564107716e-05 tokens_per_second_per_gpu:3692.8849368576434 mfu:17.419968318298608 +Step 8918 | loss:1.1424371004104614 lr:8.601815818578026e-05 tokens_per_second_per_gpu:5367.890985135759 mfu:25.321257633527587 +Step 8919 | loss:1.0160554647445679 lr:8.601515045958622e-05 tokens_per_second_per_gpu:5318.871548432144 mfu:25.090024587018306 +Step 8920 | loss:1.0381015539169312 lr:8.601214246251763e-05 tokens_per_second_per_gpu:5292.710654060902 mfu:24.966619184760315 +Step 8921 | loss:1.1264266967773438 lr:8.600913419459712e-05 tokens_per_second_per_gpu:4298.887366770104 mfu:20.278585174871946 +Step 8922 | loss:1.065877914428711 lr:8.600612565584732e-05 tokens_per_second_per_gpu:3441.964008491974 mfu:16.236331487672782 +Step 8923 | loss:1.0969648361206055 lr:8.600311684629086e-05 tokens_per_second_per_gpu:5310.79576734923 mfu:25.051929749779866 +Step 8924 | loss:1.2784918546676636 lr:8.600010776595036e-05 tokens_per_second_per_gpu:3953.802547801095 mfu:18.650761206253758 +Step 8925 | loss:1.1499890089035034 lr:8.599709841484844e-05 tokens_per_second_per_gpu:5345.40764989783 mfu:25.21519990515772 +Step 8926 | loss:0.9904934167861938 lr:8.599408879300777e-05 tokens_per_second_per_gpu:5001.2320610731 mfu:23.59166493026075 +Step 8927 | loss:1.120187520980835 lr:8.599107890045095e-05 tokens_per_second_per_gpu:5312.157406077104 mfu:25.05835283197877 +Step 8928 | loss:0.9714019298553467 lr:8.598806873720064e-05 tokens_per_second_per_gpu:4800.808957740838 mfu:22.646234956134897 +Step 8929 | loss:1.1751452684402466 lr:8.598505830327948e-05 tokens_per_second_per_gpu:4292.975567898772 mfu:20.25069821093898 +Step 8930 | loss:1.0858113765716553 lr:8.59820475987101e-05 tokens_per_second_per_gpu:4777.362230339756 mfu:22.535632742559784 +Step 8931 | loss:0.9936785697937012 lr:8.597903662351515e-05 tokens_per_second_per_gpu:5344.815540343054 mfu:25.212406823363533 +Step 8932 | loss:1.0821139812469482 lr:8.597602537771727e-05 tokens_per_second_per_gpu:4301.508559099074 mfu:20.29094978630923 +Step 8933 | loss:1.0925503969192505 lr:8.597301386133912e-05 tokens_per_second_per_gpu:4392.654473633611 mfu:20.720900616265236 +Step 8934 | loss:1.1339905261993408 lr:8.597000207440332e-05 tokens_per_second_per_gpu:4290.120035351661 mfu:20.237228176710165 +Step 8935 | loss:0.9357724785804749 lr:8.596699001693255e-05 tokens_per_second_per_gpu:4826.559032682512 mfu:22.767702453050592 +Step 8936 | loss:1.1095088720321655 lr:8.596397768894948e-05 tokens_per_second_per_gpu:5096.853012003326 mfu:24.04272527041517 +Step 8937 | loss:1.1328884363174438 lr:8.596096509047673e-05 tokens_per_second_per_gpu:3316.626866253099 mfu:15.645094803010421 +Step 8938 | loss:1.0157259702682495 lr:8.595795222153698e-05 tokens_per_second_per_gpu:4273.061168786625 mfu:20.15675859258013 +Step 8939 | loss:1.080678939819336 lr:8.595493908215288e-05 tokens_per_second_per_gpu:4932.304712657331 mfu:23.266522867565254 +Step 8940 | loss:1.1460561752319336 lr:8.59519256723471e-05 tokens_per_second_per_gpu:4754.598185030988 mfu:22.428250856891697 +Step 8941 | loss:0.9671249389648438 lr:8.59489119921423e-05 tokens_per_second_per_gpu:5349.40907469784 mfu:25.234075308652837 +Step 8942 | loss:1.0567071437835693 lr:8.594589804156115e-05 tokens_per_second_per_gpu:4163.4869587788335 mfu:19.639878348685354 +Step 8943 | loss:1.1015719175338745 lr:8.59428838206263e-05 tokens_per_second_per_gpu:3953.5930017457786 mfu:18.64977274175859 +Step 8944 | loss:1.0319883823394775 lr:8.593986932936043e-05 tokens_per_second_per_gpu:5185.445452839697 mfu:24.460630929269392 +Step 8945 | loss:1.2270690202713013 lr:8.593685456778623e-05 tokens_per_second_per_gpu:4074.864775475767 mfu:19.22183238954114 +Step 8946 | loss:0.9979442358016968 lr:8.593383953592635e-05 tokens_per_second_per_gpu:4290.763908365506 mfu:20.24026543557254 +Step 8947 | loss:1.128464937210083 lr:8.59308242338035e-05 tokens_per_second_per_gpu:4059.134709975788 mfu:19.147630986751818 +Step 8948 | loss:1.1490474939346313 lr:8.592780866144032e-05 tokens_per_second_per_gpu:4106.764360900931 mfu:19.37230817662214 +Step 8949 | loss:0.9699739217758179 lr:8.59247928188595e-05 tokens_per_second_per_gpu:4946.910740198798 mfu:23.335421991523624 +Step 8950 | loss:1.144296646118164 lr:8.592177670608375e-05 tokens_per_second_per_gpu:5355.258457163439 mfu:25.26166784375063 +Step 8951 | loss:1.0258325338363647 lr:8.591876032313572e-05 tokens_per_second_per_gpu:5277.86022698887 mfu:24.89656718651741 +Step 8952 | loss:1.2613790035247803 lr:8.59157436700381e-05 tokens_per_second_per_gpu:5341.835921187052 mfu:25.19835145142922 +Step 8953 | loss:1.0452810525894165 lr:8.591272674681361e-05 tokens_per_second_per_gpu:4945.825346857803 mfu:23.33030200594695 +Step 8954 | loss:1.0682772397994995 lr:8.590970955348489e-05 tokens_per_second_per_gpu:5318.821667087087 mfu:25.089789288203036 +Step 8955 | loss:1.262666940689087 lr:8.590669209007469e-05 tokens_per_second_per_gpu:4694.730356033407 mfu:22.14584409300519 +Step 8956 | loss:1.0294893980026245 lr:8.590367435660568e-05 tokens_per_second_per_gpu:4767.706845533725 mfu:22.490086645051214 +Step 8957 | loss:1.105831265449524 lr:8.590065635310053e-05 tokens_per_second_per_gpu:4860.762360498115 mfu:22.929045386046273 +Step 8958 | loss:1.245742917060852 lr:8.589763807958199e-05 tokens_per_second_per_gpu:4036.112586350238 mfu:19.039031701630766 +Step 8959 | loss:1.0963226556777954 lr:8.589461953607271e-05 tokens_per_second_per_gpu:5095.3580370865275 mfu:24.035673218663604 +Step 8960 | loss:1.0021307468414307 lr:8.589160072259543e-05 tokens_per_second_per_gpu:5326.397726860883 mfu:25.12552685476512 +Step 8961 | loss:1.0647084712982178 lr:8.588858163917283e-05 tokens_per_second_per_gpu:4338.749400701683 mfu:20.466621190091512 +Step 8962 | loss:1.2985972166061401 lr:8.588556228582765e-05 tokens_per_second_per_gpu:3752.8246148071357 mfu:17.702714005949975 +Step 8963 | loss:1.0269813537597656 lr:8.588254266258256e-05 tokens_per_second_per_gpu:4339.164593684799 mfu:20.468579726232097 +Step 8964 | loss:1.225942611694336 lr:8.587952276946028e-05 tokens_per_second_per_gpu:3860.953933340077 mfu:18.212778450233976 +Step 8965 | loss:0.9723926782608032 lr:8.587650260648355e-05 tokens_per_second_per_gpu:5342.951192221803 mfu:25.203612375185003 +Step 8966 | loss:1.1662092208862305 lr:8.587348217367506e-05 tokens_per_second_per_gpu:3545.8141260402217 mfu:16.72621021080501 +Step 8967 | loss:1.054153561592102 lr:8.587046147105752e-05 tokens_per_second_per_gpu:4185.381141213699 mfu:19.74315694276333 +Step 8968 | loss:1.0263996124267578 lr:8.586744049865368e-05 tokens_per_second_per_gpu:4285.770258797536 mfu:20.21670953855595 +Step 8969 | loss:1.1126601696014404 lr:8.586441925648623e-05 tokens_per_second_per_gpu:5369.800435908842 mfu:25.330264838610113 +Step 8970 | loss:1.0579787492752075 lr:8.586139774457791e-05 tokens_per_second_per_gpu:5331.734108566387 mfu:25.150699477750585 +Step 8971 | loss:1.1679351329803467 lr:8.585837596295145e-05 tokens_per_second_per_gpu:4289.434303733582 mfu:20.233993463669808 +Step 8972 | loss:1.0217456817626953 lr:8.585535391162956e-05 tokens_per_second_per_gpu:4876.802264605682 mfu:23.0047083504115 +Step 8973 | loss:1.159969449043274 lr:8.585233159063499e-05 tokens_per_second_per_gpu:3799.295991801514 mfu:17.921927420066968 +Step 8974 | loss:1.2786203622817993 lr:8.584930899999044e-05 tokens_per_second_per_gpu:5344.327855312635 mfu:25.2101063298674 +Step 8975 | loss:1.0637651681900024 lr:8.584628613971868e-05 tokens_per_second_per_gpu:5289.421787203474 mfu:24.95110503865528 +Step 8976 | loss:1.157995581626892 lr:8.584326300984242e-05 tokens_per_second_per_gpu:4222.840192289109 mfu:19.91985767785919 +Step 8977 | loss:1.054795265197754 lr:8.584023961038441e-05 tokens_per_second_per_gpu:4734.587862759604 mfu:22.333858752624547 +Step 8978 | loss:1.0788131952285767 lr:8.583721594136738e-05 tokens_per_second_per_gpu:4873.369061591318 mfu:22.988513346027915 +Step 8979 | loss:1.1086870431900024 lr:8.583419200281406e-05 tokens_per_second_per_gpu:5359.110152010048 mfu:25.27983694550899 +Step 8980 | loss:1.0414388179779053 lr:8.583116779474724e-05 tokens_per_second_per_gpu:3468.046948964541 mfu:16.359369168090417 +Step 8981 | loss:1.055116057395935 lr:8.582814331718961e-05 tokens_per_second_per_gpu:5279.340131725605 mfu:24.903548149658373 +Step 8982 | loss:1.1876914501190186 lr:8.582511857016395e-05 tokens_per_second_per_gpu:3889.7995578456294 mfu:18.348848182597703 +Step 8983 | loss:1.0856249332427979 lr:8.582209355369302e-05 tokens_per_second_per_gpu:5349.489690631914 mfu:25.23445558776836 +Step 8984 | loss:1.0466045141220093 lr:8.581906826779953e-05 tokens_per_second_per_gpu:5355.812285443298 mfu:25.26428034623957 +Step 8985 | loss:0.9942601919174194 lr:8.581604271250627e-05 tokens_per_second_per_gpu:5064.815001581012 mfu:23.891596509004874 +Step 8986 | loss:1.312104344367981 lr:8.581301688783598e-05 tokens_per_second_per_gpu:5303.302864933348 mfu:25.016584450663174 +Step 8987 | loss:1.2892290353775024 lr:8.58099907938114e-05 tokens_per_second_per_gpu:4828.34127354935 mfu:22.77610958730111 +Step 8988 | loss:0.9775769114494324 lr:8.580696443045535e-05 tokens_per_second_per_gpu:5359.671627243068 mfu:25.282485519980696 +Step 8989 | loss:1.1527178287506104 lr:8.580393779779052e-05 tokens_per_second_per_gpu:3625.448249195197 mfu:17.101857956709487 +Step 8990 | loss:1.087906002998352 lr:8.580091089583971e-05 tokens_per_second_per_gpu:3278.420621843161 mfu:15.464869429471264 +Step 8991 | loss:1.0794464349746704 lr:8.57978837246257e-05 tokens_per_second_per_gpu:5352.22554425211 mfu:25.24736107607959 +Step 8992 | loss:1.1765153408050537 lr:8.579485628417123e-05 tokens_per_second_per_gpu:4376.050990943594 mfu:20.642579155570168 +Step 8993 | loss:1.165259838104248 lr:8.579182857449906e-05 tokens_per_second_per_gpu:2997.041666815088 mfu:14.137556890403665 +Step 8994 | loss:1.0480480194091797 lr:8.5788800595632e-05 tokens_per_second_per_gpu:4637.273904666751 mfu:21.874812208827215 +Step 8995 | loss:1.2015496492385864 lr:8.57857723475928e-05 tokens_per_second_per_gpu:3969.2757978437385 mfu:18.723751166713594 +Step 8996 | loss:1.0153452157974243 lr:8.578274383040424e-05 tokens_per_second_per_gpu:4557.234113636764 mfu:21.497250858342234 +Step 8997 | loss:1.2114421129226685 lr:8.577971504408908e-05 tokens_per_second_per_gpu:5321.087380429959 mfu:25.100477044611925 +Step 8998 | loss:1.1192148923873901 lr:8.577668598867014e-05 tokens_per_second_per_gpu:4881.041215287691 mfu:23.024704204018104 +Step 8999 | loss:1.2649704217910767 lr:8.577365666417016e-05 tokens_per_second_per_gpu:4450.130794959811 mfu:20.992026230432554 +Step 9000 | loss:1.1227377653121948 lr:8.577062707061196e-05 tokens_per_second_per_gpu:5306.981305754522 mfu:25.03393628362337 +Step 9001 | loss:1.0217821598052979 lr:8.576759720801832e-05 tokens_per_second_per_gpu:5328.7955081937325 mfu:25.136837598415852 +Step 9002 | loss:1.0274542570114136 lr:8.576456707641199e-05 tokens_per_second_per_gpu:5018.218954003896 mfu:23.67179500247811 +Step 9003 | loss:1.0723458528518677 lr:8.57615366758158e-05 tokens_per_second_per_gpu:5345.576243213281 mfu:25.215995188591684 +Step 9004 | loss:1.0729267597198486 lr:8.575850600625252e-05 tokens_per_second_per_gpu:4428.718070793026 mfu:20.89101875715036 +Step 9005 | loss:1.1091607809066772 lr:8.575547506774497e-05 tokens_per_second_per_gpu:5340.9300638835275 mfu:25.194078367973308 +Step 9006 | loss:1.200180172920227 lr:8.575244386031594e-05 tokens_per_second_per_gpu:4589.572588339209 mfu:21.649796961026393 +Step 9007 | loss:0.9951605200767517 lr:8.574941238398819e-05 tokens_per_second_per_gpu:5366.1653794366875 mfu:25.313117657026375 +Step 9008 | loss:1.0792598724365234 lr:8.574638063878455e-05 tokens_per_second_per_gpu:4895.086746588218 mfu:23.090959371577444 +Step 9009 | loss:0.9264690279960632 lr:8.574334862472784e-05 tokens_per_second_per_gpu:4682.570584136333 mfu:22.088484374295675 +Step 9010 | loss:1.20123291015625 lr:8.574031634184083e-05 tokens_per_second_per_gpu:4868.276257316735 mfu:22.964489719342694 +Step 9011 | loss:1.107080340385437 lr:8.573728379014634e-05 tokens_per_second_per_gpu:4412.101245229085 mfu:20.812634355842462 +Step 9012 | loss:1.0797470808029175 lr:8.573425096966718e-05 tokens_per_second_per_gpu:4651.834599666073 mfu:21.943497491449754 +Step 9013 | loss:0.8881632685661316 lr:8.573121788042615e-05 tokens_per_second_per_gpu:3944.2282846194676 mfu:18.605597773288277 +Step 9014 | loss:1.0589238405227661 lr:8.572818452244609e-05 tokens_per_second_per_gpu:4998.563352434306 mfu:23.57907617628291 +Step 9015 | loss:1.1859476566314697 lr:8.572515089574979e-05 tokens_per_second_per_gpu:4831.642938367718 mfu:22.79168410357146 +Step 9016 | loss:1.0220115184783936 lr:8.572211700036006e-05 tokens_per_second_per_gpu:3508.2887903193364 mfu:16.549196799727117 +Step 9017 | loss:1.156931757926941 lr:8.571908283629973e-05 tokens_per_second_per_gpu:5257.617509988416 mfu:24.801078836662604 +Step 9018 | loss:1.1493338346481323 lr:8.571604840359165e-05 tokens_per_second_per_gpu:3818.561844138955 mfu:18.01280773263604 +Step 9019 | loss:1.1793981790542603 lr:8.571301370225858e-05 tokens_per_second_per_gpu:5301.660878092738 mfu:25.00883891858417 +Step 9020 | loss:1.1105598211288452 lr:8.570997873232338e-05 tokens_per_second_per_gpu:3701.4707225426937 mfu:17.460468934261083 +Step 9021 | loss:1.207761526107788 lr:8.570694349380887e-05 tokens_per_second_per_gpu:5323.666593659589 mfu:25.112643633475244 +Step 9022 | loss:1.185595989227295 lr:8.570390798673791e-05 tokens_per_second_per_gpu:5344.7646266674965 mfu:25.21216665486916 +Step 9023 | loss:1.1299678087234497 lr:8.570087221113327e-05 tokens_per_second_per_gpu:5281.583657568366 mfu:24.91413124384366 +Step 9024 | loss:1.0300648212432861 lr:8.569783616701783e-05 tokens_per_second_per_gpu:5120.277542272443 mfu:24.153222776331575 +Step 9025 | loss:1.1916667222976685 lr:8.569479985441441e-05 tokens_per_second_per_gpu:3934.877367482449 mfu:18.561487901721982 +Step 9026 | loss:1.2324587106704712 lr:8.569176327334584e-05 tokens_per_second_per_gpu:4538.671977342243 mfu:21.40969009441395 +Step 9027 | loss:1.136935830116272 lr:8.568872642383498e-05 tokens_per_second_per_gpu:4143.418794202848 mfu:19.545213392398484 +Step 9028 | loss:1.0695664882659912 lr:8.568568930590464e-05 tokens_per_second_per_gpu:4780.616912036792 mfu:22.550985631430667 +Step 9029 | loss:1.0840526819229126 lr:8.568265191957769e-05 tokens_per_second_per_gpu:5344.065375459713 mfu:25.20886816761749 +Step 9030 | loss:1.1150755882263184 lr:8.567961426487694e-05 tokens_per_second_per_gpu:3516.7043074596845 mfu:16.588894229913425 +Step 9031 | loss:1.1510956287384033 lr:8.567657634182527e-05 tokens_per_second_per_gpu:4137.9545982751 mfu:19.51943784791933 +Step 9032 | loss:1.190841794013977 lr:8.567353815044552e-05 tokens_per_second_per_gpu:4446.269690776822 mfu:20.973812743229093 +Step 9033 | loss:1.178733229637146 lr:8.567049969076053e-05 tokens_per_second_per_gpu:3827.257632251652 mfu:18.053827248818074 +Step 9034 | loss:1.122822642326355 lr:8.566746096279317e-05 tokens_per_second_per_gpu:3810.099204180879 mfu:17.97288801607853 +Step 9035 | loss:1.1171878576278687 lr:8.566442196656627e-05 tokens_per_second_per_gpu:4661.108332932243 mfu:21.987243273528332 +Step 9036 | loss:1.3020037412643433 lr:8.56613827021027e-05 tokens_per_second_per_gpu:3945.081921565137 mfu:18.609624524406485 +Step 9037 | loss:1.045275092124939 lr:8.565834316942534e-05 tokens_per_second_per_gpu:4953.468577842474 mfu:23.366356430571322 +Step 9038 | loss:1.1662185192108154 lr:8.565530336855703e-05 tokens_per_second_per_gpu:5289.459966228216 mfu:24.951285135629 +Step 9039 | loss:1.0794689655303955 lr:8.565226329952063e-05 tokens_per_second_per_gpu:5351.428523097961 mfu:25.243601391309635 +Step 9040 | loss:1.1254950761795044 lr:8.564922296233899e-05 tokens_per_second_per_gpu:4236.905953544907 mfu:19.98620827356913 +Step 9041 | loss:1.0190179347991943 lr:8.5646182357035e-05 tokens_per_second_per_gpu:5349.551471419669 mfu:25.234747018285088 +Step 9042 | loss:1.2626031637191772 lr:8.564314148363155e-05 tokens_per_second_per_gpu:4178.6545051620315 mfu:19.71142625282516 +Step 9043 | loss:1.10406494140625 lr:8.564010034215146e-05 tokens_per_second_per_gpu:4986.903222479379 mfu:23.524073353862214 +Step 9044 | loss:1.1372404098510742 lr:8.563705893261764e-05 tokens_per_second_per_gpu:4399.395703008045 mfu:20.752700145396776 +Step 9045 | loss:1.2494231462478638 lr:8.563401725505295e-05 tokens_per_second_per_gpu:3817.7878439003307 mfu:18.00915664145239 +Step 9046 | loss:1.0447815656661987 lr:8.563097530948028e-05 tokens_per_second_per_gpu:5360.45879350579 mfu:25.28619871754637 +Step 9047 | loss:1.1568772792816162 lr:8.562793309592248e-05 tokens_per_second_per_gpu:4049.5721320017806 mfu:19.102522674904385 +Step 9048 | loss:1.0098689794540405 lr:8.562489061440245e-05 tokens_per_second_per_gpu:5386.826534323823 mfu:25.410579849785517 +Step 9049 | loss:1.028308629989624 lr:8.562184786494309e-05 tokens_per_second_per_gpu:4319.359689553893 mfu:20.375156614381527 +Step 9050 | loss:1.0845540761947632 lr:8.561880484756725e-05 tokens_per_second_per_gpu:4347.147255064871 mfu:20.506235301944372 +Step 9051 | loss:1.122410774230957 lr:8.561576156229783e-05 tokens_per_second_per_gpu:3988.7369111012554 mfu:18.815552558357613 +Step 9052 | loss:1.0233944654464722 lr:8.561271800915774e-05 tokens_per_second_per_gpu:4204.030304115753 mfu:19.83112822604762 +Step 9053 | loss:1.0060776472091675 lr:8.560967418816984e-05 tokens_per_second_per_gpu:5354.631027327235 mfu:25.258708150162686 +Step 9054 | loss:1.205386996269226 lr:8.560663009935703e-05 tokens_per_second_per_gpu:3361.963574865368 mfu:15.858955792774617 +Step 9055 | loss:1.0365815162658691 lr:8.560358574274223e-05 tokens_per_second_per_gpu:4255.670328349649 mfu:20.074723031055573 +Step 9056 | loss:1.3152165412902832 lr:8.560054111834832e-05 tokens_per_second_per_gpu:5356.477351901257 mfu:25.267417578193804 +Step 9057 | loss:1.0002906322479248 lr:8.559749622619817e-05 tokens_per_second_per_gpu:4166.88652569901 mfu:19.655914685873757 +Step 9058 | loss:1.222550392150879 lr:8.559445106631475e-05 tokens_per_second_per_gpu:3325.213271561925 mfu:15.685598341843964 +Step 9059 | loss:1.0505428314208984 lr:8.559140563872088e-05 tokens_per_second_per_gpu:4670.825970108096 mfu:22.0330830260867 +Step 9060 | loss:1.1294721364974976 lr:8.558835994343952e-05 tokens_per_second_per_gpu:5135.8166365580155 mfu:24.226523335319495 +Step 9061 | loss:1.150341510772705 lr:8.558531398049357e-05 tokens_per_second_per_gpu:3799.0688913449835 mfu:17.92085614846618 +Step 9062 | loss:1.2116345167160034 lr:8.558226774990594e-05 tokens_per_second_per_gpu:3743.955642065078 mfu:17.6608775483229 +Step 9063 | loss:1.0889164209365845 lr:8.557922125169953e-05 tokens_per_second_per_gpu:5070.836666978063 mfu:23.920001732084817 +Step 9064 | loss:0.98003751039505 lr:8.557617448589726e-05 tokens_per_second_per_gpu:5312.610115108062 mfu:25.060488337717924 +Step 9065 | loss:1.1826354265213013 lr:8.557312745252202e-05 tokens_per_second_per_gpu:4497.146465421215 mfu:21.213807169699546 +Step 9066 | loss:1.0171540975570679 lr:8.557008015159677e-05 tokens_per_second_per_gpu:5192.647226403284 mfu:24.4946029239181 +Step 9067 | loss:1.0418709516525269 lr:8.55670325831444e-05 tokens_per_second_per_gpu:3765.77609412079 mfu:17.763808343622713 +Step 9068 | loss:1.01344633102417 lr:8.556398474718784e-05 tokens_per_second_per_gpu:4046.0935035035213 mfu:19.086113390762872 +Step 9069 | loss:1.090976595878601 lr:8.556093664375003e-05 tokens_per_second_per_gpu:5360.309257912176 mfu:25.285493332637888 +Step 9070 | loss:1.0863523483276367 lr:8.555788827285385e-05 tokens_per_second_per_gpu:4613.296174765272 mfu:21.76170516585936 +Step 9071 | loss:1.0122705698013306 lr:8.555483963452227e-05 tokens_per_second_per_gpu:4758.381836263712 mfu:22.446098985313924 +Step 9072 | loss:1.108883261680603 lr:8.555179072877819e-05 tokens_per_second_per_gpu:4873.936946516008 mfu:22.991192156108838 +Step 9073 | loss:1.022379755973816 lr:8.554874155564458e-05 tokens_per_second_per_gpu:5343.635061568926 mfu:25.206838303575527 +Step 9074 | loss:1.111969232559204 lr:8.554569211514434e-05 tokens_per_second_per_gpu:4487.339078023005 mfu:21.167544050029356 +Step 9075 | loss:1.4226014614105225 lr:8.554264240730041e-05 tokens_per_second_per_gpu:3223.3919858490044 mfu:15.20528996463338 +Step 9076 | loss:1.1766606569290161 lr:8.553959243213572e-05 tokens_per_second_per_gpu:3428.7819782775296 mfu:16.174149602064976 +Step 9077 | loss:1.2019165754318237 lr:8.553654218967325e-05 tokens_per_second_per_gpu:4455.485514757357 mfu:21.017285357326983 +Step 9078 | loss:1.2110872268676758 lr:8.553349167993587e-05 tokens_per_second_per_gpu:4866.500901229504 mfu:22.95611506177224 +Step 9079 | loss:1.103087306022644 lr:8.55304409029466e-05 tokens_per_second_per_gpu:3677.518233067872 mfu:17.347480954692237 +Step 9080 | loss:1.0890884399414062 lr:8.552738985872833e-05 tokens_per_second_per_gpu:4787.876692848236 mfu:22.58523125616458 +Step 9081 | loss:1.0843020677566528 lr:8.552433854730404e-05 tokens_per_second_per_gpu:5340.042778558656 mfu:25.1898928917094 +Step 9082 | loss:0.9576568007469177 lr:8.552128696869667e-05 tokens_per_second_per_gpu:5344.359242016742 mfu:25.2102543863058 +Step 9083 | loss:1.0906814336776733 lr:8.551823512292916e-05 tokens_per_second_per_gpu:4755.516036833782 mfu:22.432580520447143 +Step 9084 | loss:0.9667946100234985 lr:8.551518301002446e-05 tokens_per_second_per_gpu:5349.265302985372 mfu:25.233397112955142 +Step 9085 | loss:1.0599822998046875 lr:8.551213063000556e-05 tokens_per_second_per_gpu:4780.226621048507 mfu:22.54914456224841 +Step 9086 | loss:1.163902997970581 lr:8.550907798289538e-05 tokens_per_second_per_gpu:3792.403394355637 mfu:17.889413861916374 +Step 9087 | loss:1.0885475873947144 lr:8.55060250687169e-05 tokens_per_second_per_gpu:5333.140506939341 mfu:25.15733369883212 +Step 9088 | loss:1.0448811054229736 lr:8.550297188749307e-05 tokens_per_second_per_gpu:4702.66536143674 mfu:22.183274867344238 +Step 9089 | loss:1.046190619468689 lr:8.549991843924686e-05 tokens_per_second_per_gpu:5209.9382225444 mfu:24.5761675028639 +Step 9090 | loss:1.1731654405593872 lr:8.549686472400123e-05 tokens_per_second_per_gpu:4105.072298561009 mfu:19.36432642986912 +Step 9091 | loss:1.0932273864746094 lr:8.549381074177916e-05 tokens_per_second_per_gpu:3765.2507092656174 mfu:17.761330013620963 +Step 9092 | loss:0.9331742525100708 lr:8.549075649260361e-05 tokens_per_second_per_gpu:3436.3028687962706 mfu:16.20962692583812 +Step 9093 | loss:1.259427785873413 lr:8.548770197649755e-05 tokens_per_second_per_gpu:4547.327852979283 mfu:21.45052133663854 +Step 9094 | loss:1.0291712284088135 lr:8.548464719348394e-05 tokens_per_second_per_gpu:3617.4027310069728 mfu:17.06390587470868 +Step 9095 | loss:1.1376162767410278 lr:8.548159214358578e-05 tokens_per_second_per_gpu:4612.693866509457 mfu:21.758863975052297 +Step 9096 | loss:1.0525922775268555 lr:8.547853682682604e-05 tokens_per_second_per_gpu:4583.540116979639 mfu:21.621340764377557 +Step 9097 | loss:1.095644474029541 lr:8.547548124322771e-05 tokens_per_second_per_gpu:3590.8989143886333 mfu:16.938882849702562 +Step 9098 | loss:1.1669584512710571 lr:8.547242539281375e-05 tokens_per_second_per_gpu:5219.034078824047 mfu:24.619074208848104 +Step 9099 | loss:0.932542622089386 lr:8.546936927560715e-05 tokens_per_second_per_gpu:5327.671280057867 mfu:25.131534422485878 +Step 9100 | loss:1.0181121826171875 lr:8.54663128916309e-05 tokens_per_second_per_gpu:4054.707281631944 mfu:19.126746051856735 +Step 9101 | loss:1.0546019077301025 lr:8.546325624090799e-05 tokens_per_second_per_gpu:4833.401419460673 mfu:22.79997915891474 +Step 9102 | loss:1.0618196725845337 lr:8.54601993234614e-05 tokens_per_second_per_gpu:5155.267438594121 mfu:24.318276087172837 +Step 9103 | loss:1.0578125715255737 lr:8.54571421393141e-05 tokens_per_second_per_gpu:5237.234208254824 mfu:24.704927324635364 +Step 9104 | loss:1.2355420589447021 lr:8.545408468848914e-05 tokens_per_second_per_gpu:3669.0467002316127 mfu:17.307519288911056 +Step 9105 | loss:1.0092964172363281 lr:8.545102697100948e-05 tokens_per_second_per_gpu:5110.694772625526 mfu:24.10801921691797 +Step 9106 | loss:0.981086790561676 lr:8.544796898689811e-05 tokens_per_second_per_gpu:5346.633508549786 mfu:25.220982489572048 +Step 9107 | loss:1.0069295167922974 lr:8.544491073617806e-05 tokens_per_second_per_gpu:4478.43411003389 mfu:21.12553779668049 +Step 9108 | loss:1.0647865533828735 lr:8.54418522188723e-05 tokens_per_second_per_gpu:5305.182363706417 mfu:25.025450367051295 +Step 9109 | loss:1.2067819833755493 lr:8.543879343500386e-05 tokens_per_second_per_gpu:5314.442606841728 mfu:25.06913251388096 +Step 9110 | loss:1.101110577583313 lr:8.543573438459572e-05 tokens_per_second_per_gpu:3828.0281143643215 mfu:18.057461744401337 +Step 9111 | loss:1.1117697954177856 lr:8.54326750676709e-05 tokens_per_second_per_gpu:5341.186454661345 mfu:25.19528780701705 +Step 9112 | loss:1.0292664766311646 lr:8.542961548425243e-05 tokens_per_second_per_gpu:4190.1602207709075 mfu:19.765700676429738 +Step 9113 | loss:1.0166057348251343 lr:8.542655563436329e-05 tokens_per_second_per_gpu:4395.497381443077 mfu:20.73431109745258 +Step 9114 | loss:0.9628176689147949 lr:8.542349551802648e-05 tokens_per_second_per_gpu:4789.694883117624 mfu:22.59380796988053 +Step 9115 | loss:1.1322972774505615 lr:8.542043513526508e-05 tokens_per_second_per_gpu:5024.617204542675 mfu:23.70197664192364 +Step 9116 | loss:1.035550832748413 lr:8.541737448610204e-05 tokens_per_second_per_gpu:5046.783953471013 mfu:23.806540978655967 +Step 9117 | loss:1.3027476072311401 lr:8.541431357056042e-05 tokens_per_second_per_gpu:5157.9487419182815 mfu:24.330924252430428 +Step 9118 | loss:1.0447255373001099 lr:8.541125238866323e-05 tokens_per_second_per_gpu:4130.192939186762 mfu:19.482824777724048 +Step 9119 | loss:1.2198609113693237 lr:8.540819094043349e-05 tokens_per_second_per_gpu:5116.417989029098 mfu:24.13501660517503 +Step 9120 | loss:1.2483351230621338 lr:8.540512922589421e-05 tokens_per_second_per_gpu:4632.292472959573 mfu:21.851313945544565 +Step 9121 | loss:1.1329231262207031 lr:8.540206724506845e-05 tokens_per_second_per_gpu:4832.558491449408 mfu:22.796002923667256 +Step 9122 | loss:1.023567795753479 lr:8.539900499797923e-05 tokens_per_second_per_gpu:3728.0981179951773 mfu:17.586074901712582 +Step 9123 | loss:1.0246591567993164 lr:8.539594248464957e-05 tokens_per_second_per_gpu:3849.480967511791 mfu:18.158658512932128 +Step 9124 | loss:1.074845790863037 lr:8.539287970510252e-05 tokens_per_second_per_gpu:5257.513914509779 mfu:24.800590159114666 +Step 9125 | loss:1.0875211954116821 lr:8.53898166593611e-05 tokens_per_second_per_gpu:4806.213455250634 mfu:22.67172889299096 +Step 9126 | loss:1.0242825746536255 lr:8.538675334744835e-05 tokens_per_second_per_gpu:5324.9593140773295 mfu:25.118741616246567 +Step 9127 | loss:1.0015243291854858 lr:8.538368976938731e-05 tokens_per_second_per_gpu:4677.970282333496 mfu:22.066783965799267 +Step 9128 | loss:1.124880075454712 lr:8.538062592520102e-05 tokens_per_second_per_gpu:4449.40303285478 mfu:20.988593252414102 +Step 9129 | loss:1.2600173950195312 lr:8.537756181491254e-05 tokens_per_second_per_gpu:4507.963209017025 mfu:21.26483159476775 +Step 9130 | loss:1.1637340784072876 lr:8.53744974385449e-05 tokens_per_second_per_gpu:4599.374555594218 mfu:21.69603451295634 +Step 9131 | loss:1.0157146453857422 lr:8.537143279612115e-05 tokens_per_second_per_gpu:5320.015568844253 mfu:25.095421126492234 +Step 9132 | loss:1.1354953050613403 lr:8.536836788766434e-05 tokens_per_second_per_gpu:4764.458690708135 mfu:22.47476454454655 +Step 9133 | loss:1.017547607421875 lr:8.536530271319752e-05 tokens_per_second_per_gpu:5327.74512053908 mfu:25.13188274063424 +Step 9134 | loss:1.1087169647216797 lr:8.536223727274374e-05 tokens_per_second_per_gpu:5366.791586068277 mfu:25.316071580549448 +Step 9135 | loss:1.0858049392700195 lr:8.535917156632607e-05 tokens_per_second_per_gpu:5360.437602269635 mfu:25.28609875486983 +Step 9136 | loss:1.2331827878952026 lr:8.535610559396756e-05 tokens_per_second_per_gpu:4262.801712057099 mfu:20.108362984743138 +Step 9137 | loss:1.1471501588821411 lr:8.535303935569127e-05 tokens_per_second_per_gpu:3698.3706829435773 mfu:17.445845518550755 +Step 9138 | loss:1.2064582109451294 lr:8.534997285152026e-05 tokens_per_second_per_gpu:3863.9150163037334 mfu:18.22674638378639 +Step 9139 | loss:1.1614294052124023 lr:8.534690608147758e-05 tokens_per_second_per_gpu:4187.347217289687 mfu:19.752431259060938 +Step 9140 | loss:1.0869622230529785 lr:8.534383904558632e-05 tokens_per_second_per_gpu:3251.951397315825 mfu:15.340009581260428 +Step 9141 | loss:1.1514129638671875 lr:8.534077174386954e-05 tokens_per_second_per_gpu:4505.634311361625 mfu:21.25384578717631 +Step 9142 | loss:1.05171537399292 lr:8.53377041763503e-05 tokens_per_second_per_gpu:4627.771862304599 mfu:21.82998945378934 +Step 9143 | loss:1.1731334924697876 lr:8.533463634305168e-05 tokens_per_second_per_gpu:4913.763069480345 mfu:23.17905877316887 +Step 9144 | loss:1.0955349206924438 lr:8.533156824399675e-05 tokens_per_second_per_gpu:5312.732623637676 mfu:25.061066231354225 +Step 9145 | loss:1.2332038879394531 lr:8.532849987920859e-05 tokens_per_second_per_gpu:5391.193522988081 mfu:25.431179680399918 +Step 9146 | loss:1.047968864440918 lr:8.532543124871027e-05 tokens_per_second_per_gpu:5355.247495716698 mfu:25.261616136736134 +Step 9147 | loss:1.0201373100280762 lr:8.532236235252488e-05 tokens_per_second_per_gpu:4839.688405783695 mfu:22.829635946070912 +Step 9148 | loss:1.166533350944519 lr:8.531929319067549e-05 tokens_per_second_per_gpu:4409.550631487562 mfu:20.800602675644118 +Step 9149 | loss:1.0074859857559204 lr:8.53162237631852e-05 tokens_per_second_per_gpu:5369.635937690199 mfu:25.329488872446483 +Step 9150 | loss:0.9992457032203674 lr:8.531315407007706e-05 tokens_per_second_per_gpu:4832.072639595515 mfu:22.79371107757757 +Step 9151 | loss:1.158376693725586 lr:8.53100841113742e-05 tokens_per_second_per_gpu:5339.04085466365 mfu:25.18516664575081 +Step 9152 | loss:1.2311533689498901 lr:8.53070138870997e-05 tokens_per_second_per_gpu:4611.893116329832 mfu:21.755086699832937 +Step 9153 | loss:1.0366253852844238 lr:8.530394339727663e-05 tokens_per_second_per_gpu:5282.8949586863555 mfu:24.920316874948256 +Step 9154 | loss:1.0714850425720215 lr:8.530087264192809e-05 tokens_per_second_per_gpu:4229.630252177616 mfu:19.95188754885696 +Step 9155 | loss:1.1395149230957031 lr:8.529780162107718e-05 tokens_per_second_per_gpu:4927.534111195428 mfu:23.244019126520914 +Step 9156 | loss:1.0757832527160645 lr:8.529473033474702e-05 tokens_per_second_per_gpu:4363.05241921471 mfu:20.581262674939996 +Step 9157 | loss:1.1986355781555176 lr:8.52916587829607e-05 tokens_per_second_per_gpu:4027.942761909204 mfu:19.000493245826625 +Step 9158 | loss:1.1692087650299072 lr:8.528858696574128e-05 tokens_per_second_per_gpu:5064.929348029913 mfu:23.892135900713853 +Step 9159 | loss:1.1177181005477905 lr:8.52855148831119e-05 tokens_per_second_per_gpu:5361.871585037106 mfu:25.29286309624668 +Step 9160 | loss:1.1506140232086182 lr:8.528244253509567e-05 tokens_per_second_per_gpu:3906.229259756094 mfu:18.426349889705566 +Step 9161 | loss:1.1076810359954834 lr:8.527936992171569e-05 tokens_per_second_per_gpu:3933.6531724876004 mfu:18.55571316506208 +Step 9162 | loss:1.0999799966812134 lr:8.527629704299507e-05 tokens_per_second_per_gpu:4725.683511964164 mfu:22.29185541068355 +Step 9163 | loss:1.1688576936721802 lr:8.52732238989569e-05 tokens_per_second_per_gpu:3407.9650102045866 mfu:16.075952411923907 +Step 9164 | loss:0.9745709896087646 lr:8.527015048962432e-05 tokens_per_second_per_gpu:4868.800351352185 mfu:22.96696196032758 +Step 9165 | loss:1.0470302104949951 lr:8.526707681502044e-05 tokens_per_second_per_gpu:5233.323551318319 mfu:24.686480088638994 +Step 9166 | loss:1.19441556930542 lr:8.526400287516838e-05 tokens_per_second_per_gpu:3897.0568400641205 mfu:18.383082021042355 +Step 9167 | loss:1.040051817893982 lr:8.526092867009126e-05 tokens_per_second_per_gpu:4451.390789781181 mfu:20.99796984098224 +Step 9168 | loss:1.0595650672912598 lr:8.525785419981218e-05 tokens_per_second_per_gpu:5336.2034873226185 mfu:25.17178229989838 +Step 9169 | loss:0.9785060286521912 lr:8.525477946435429e-05 tokens_per_second_per_gpu:4575.3664238435385 mfu:21.582784059278723 +Step 9170 | loss:1.1379456520080566 lr:8.525170446374069e-05 tokens_per_second_per_gpu:3503.6210395512767 mfu:16.52717822295378 +Step 9171 | loss:1.2894532680511475 lr:8.524862919799455e-05 tokens_per_second_per_gpu:3986.4438815445856 mfu:18.804735946707616 +Step 9172 | loss:0.9336310029029846 lr:8.524555366713895e-05 tokens_per_second_per_gpu:5357.287412012083 mfu:25.27123876994708 +Step 9173 | loss:1.0501078367233276 lr:8.524247787119705e-05 tokens_per_second_per_gpu:4739.249570659025 mfu:22.355848824155498 +Step 9174 | loss:1.0161577463150024 lr:8.523940181019196e-05 tokens_per_second_per_gpu:5197.976705283441 mfu:24.519742985098667 +Step 9175 | loss:1.2084413766860962 lr:8.523632548414686e-05 tokens_per_second_per_gpu:3489.0799256411324 mfu:16.458585307669765 +Step 9176 | loss:0.9603943228721619 lr:8.523324889308486e-05 tokens_per_second_per_gpu:5335.491418244727 mfu:25.168423348566456 +Step 9177 | loss:1.0086666345596313 lr:8.523017203702908e-05 tokens_per_second_per_gpu:4243.27637518013 mfu:20.016258639328484 +Step 9178 | loss:1.3645471334457397 lr:8.522709491600269e-05 tokens_per_second_per_gpu:4341.775790089202 mfu:20.48089718518771 +Step 9179 | loss:1.0331007242202759 lr:8.522401753002882e-05 tokens_per_second_per_gpu:3160.6521581490347 mfu:14.909335492854202 +Step 9180 | loss:1.188092827796936 lr:8.522093987913062e-05 tokens_per_second_per_gpu:5357.350909891405 mfu:25.27153830027769 +Step 9181 | loss:1.0986955165863037 lr:8.521786196333123e-05 tokens_per_second_per_gpu:3772.9776903647576 mfu:17.79777950182453 +Step 9182 | loss:1.2440868616104126 lr:8.521478378265382e-05 tokens_per_second_per_gpu:5335.7035824873765 mfu:25.169424163497983 +Step 9183 | loss:1.1409567594528198 lr:8.521170533712152e-05 tokens_per_second_per_gpu:3363.0978531567325 mfu:15.864306377002816 +Step 9184 | loss:1.2442262172698975 lr:8.520862662675751e-05 tokens_per_second_per_gpu:4107.949007339835 mfu:19.377896355995972 +Step 9185 | loss:1.270661473274231 lr:8.520554765158491e-05 tokens_per_second_per_gpu:5378.359960611989 mfu:25.370641577040363 +Step 9186 | loss:0.9581702351570129 lr:8.520246841162688e-05 tokens_per_second_per_gpu:4690.476049615825 mfu:22.125775803773937 +Step 9187 | loss:1.0871832370758057 lr:8.519938890690662e-05 tokens_per_second_per_gpu:4702.490419275095 mfu:22.18244963531955 +Step 9188 | loss:1.0045157670974731 lr:8.519630913744725e-05 tokens_per_second_per_gpu:4761.998961044532 mfu:22.463161580050176 +Step 9189 | loss:1.0641788244247437 lr:8.519322910327195e-05 tokens_per_second_per_gpu:4230.381630534879 mfu:19.955431928766536 +Step 9190 | loss:1.10552179813385 lr:8.519014880440387e-05 tokens_per_second_per_gpu:5323.561080152495 mfu:25.11214590825982 +Step 9191 | loss:0.9885987639427185 lr:8.518706824086621e-05 tokens_per_second_per_gpu:5333.1509447788285 mfu:25.157382935901857 +Step 9192 | loss:1.2223854064941406 lr:8.518398741268212e-05 tokens_per_second_per_gpu:5313.576468586636 mfu:25.065046791953264 +Step 9193 | loss:1.117386817932129 lr:8.518090631987476e-05 tokens_per_second_per_gpu:4048.253965415476 mfu:19.096304658214375 +Step 9194 | loss:1.0680822134017944 lr:8.517782496246733e-05 tokens_per_second_per_gpu:4454.40449096793 mfu:21.012185983670577 +Step 9195 | loss:1.0253382921218872 lr:8.517474334048298e-05 tokens_per_second_per_gpu:4262.15864391368 mfu:20.10532952259189 +Step 9196 | loss:1.140383243560791 lr:8.51716614539449e-05 tokens_per_second_per_gpu:4658.423850312899 mfu:21.97458011099292 +Step 9197 | loss:1.1784039735794067 lr:8.516857930287627e-05 tokens_per_second_per_gpu:4056.9772889246956 mfu:19.137454063559826 +Step 9198 | loss:1.0289382934570312 lr:8.516549688730024e-05 tokens_per_second_per_gpu:4164.265290104324 mfu:19.643549870344707 +Step 9199 | loss:1.157649278640747 lr:8.516241420724005e-05 tokens_per_second_per_gpu:3696.147075190917 mfu:17.435356381394854 +Step 9200 | loss:0.9717020988464355 lr:8.515933126271885e-05 tokens_per_second_per_gpu:4366.703722570971 mfu:20.598486495848068 +Step 9201 | loss:1.1497704982757568 lr:8.515624805375983e-05 tokens_per_second_per_gpu:5371.958112004761 mfu:25.340442964892066 +Step 9202 | loss:0.895268976688385 lr:8.515316458038617e-05 tokens_per_second_per_gpu:3794.841979911559 mfu:17.900917086049258 +Step 9203 | loss:1.113364338874817 lr:8.51500808426211e-05 tokens_per_second_per_gpu:5360.988143662286 mfu:25.288695752551014 +Step 9204 | loss:1.174474835395813 lr:8.514699684048777e-05 tokens_per_second_per_gpu:4398.636628710039 mfu:20.749119462421653 +Step 9205 | loss:1.1173064708709717 lr:8.51439125740094e-05 tokens_per_second_per_gpu:3606.753596529211 mfu:17.01367209044738 +Step 9206 | loss:1.0097177028656006 lr:8.514082804320917e-05 tokens_per_second_per_gpu:5321.030014961978 mfu:25.100206441911897 +Step 9207 | loss:1.0717030763626099 lr:8.51377432481103e-05 tokens_per_second_per_gpu:4018.1257123837595 mfu:18.954184548253735 +Step 9208 | loss:1.2357869148254395 lr:8.513465818873596e-05 tokens_per_second_per_gpu:5354.179859821987 mfu:25.256579916063313 +Step 9209 | loss:1.2607370615005493 lr:8.513157286510939e-05 tokens_per_second_per_gpu:5361.443472375359 mfu:25.290843615777668 +Step 9210 | loss:1.1421098709106445 lr:8.512848727725376e-05 tokens_per_second_per_gpu:3851.997114960224 mfu:18.17052760974541 +Step 9211 | loss:1.1063873767852783 lr:8.512540142519232e-05 tokens_per_second_per_gpu:4420.749436805424 mfu:20.853429351041687 +Step 9212 | loss:0.866214394569397 lr:8.512231530894822e-05 tokens_per_second_per_gpu:5336.593458872369 mfu:25.17362186223456 +Step 9213 | loss:1.0751720666885376 lr:8.511922892854474e-05 tokens_per_second_per_gpu:4649.422082323285 mfu:21.932117235525922 +Step 9214 | loss:1.081624984741211 lr:8.511614228400504e-05 tokens_per_second_per_gpu:3935.804763721118 mfu:18.565862588009896 +Step 9215 | loss:1.1291537284851074 lr:8.511305537535237e-05 tokens_per_second_per_gpu:4001.427159921416 mfu:18.875414627222025 +Step 9216 | loss:1.2221747636795044 lr:8.510996820260991e-05 tokens_per_second_per_gpu:5340.875699157925 mfu:25.193821920287057 +Step 9217 | loss:1.085336446762085 lr:8.510688076580092e-05 tokens_per_second_per_gpu:4506.489552169245 mfu:21.257880103984984 +Step 9218 | loss:1.1254054307937622 lr:8.510379306494858e-05 tokens_per_second_per_gpu:4353.171772321137 mfu:20.534653977730816 +Step 9219 | loss:1.1306811571121216 lr:8.510070510007613e-05 tokens_per_second_per_gpu:5262.183805124591 mfu:24.822618829910233 +Step 9220 | loss:1.1703871488571167 lr:8.509761687120681e-05 tokens_per_second_per_gpu:3900.4019917049527 mfu:18.39886167207388 +Step 9221 | loss:0.9793281555175781 lr:8.509452837836383e-05 tokens_per_second_per_gpu:5318.108256387151 mfu:25.086424008211026 +Step 9222 | loss:1.173500895500183 lr:8.509143962157042e-05 tokens_per_second_per_gpu:5338.226854171371 mfu:25.181326866545927 +Step 9223 | loss:0.9582245349884033 lr:8.508835060084983e-05 tokens_per_second_per_gpu:5358.616322819618 mfu:25.277507468961783 +Step 9224 | loss:1.1071484088897705 lr:8.508526131622527e-05 tokens_per_second_per_gpu:5271.491210610247 mfu:24.866523449593203 +Step 9225 | loss:0.9892641305923462 lr:8.508217176771997e-05 tokens_per_second_per_gpu:4812.0060787273505 mfu:22.69905368625411 +Step 9226 | loss:1.0292396545410156 lr:8.50790819553572e-05 tokens_per_second_per_gpu:4936.082253243382 mfu:23.28434217102454 +Step 9227 | loss:1.012030005455017 lr:8.507599187916017e-05 tokens_per_second_per_gpu:5365.994321200692 mfu:25.312310746141637 +Step 9228 | loss:0.9158966541290283 lr:8.507290153915213e-05 tokens_per_second_per_gpu:4767.255347116853 mfu:22.48795684998526 +Step 9229 | loss:1.1678386926651 lr:8.506981093535632e-05 tokens_per_second_per_gpu:3653.2416706126264 mfu:17.23296426758218 +Step 9230 | loss:1.0157756805419922 lr:8.506672006779599e-05 tokens_per_second_per_gpu:5341.391500027117 mfu:25.19625504099174 +Step 9231 | loss:1.0259164571762085 lr:8.506362893649438e-05 tokens_per_second_per_gpu:4107.9466089401285 mfu:19.37788504233535 +Step 9232 | loss:1.005682110786438 lr:8.506053754147474e-05 tokens_per_second_per_gpu:4855.713666822349 mfu:22.905229836581718 +Step 9233 | loss:1.0301768779754639 lr:8.505744588276033e-05 tokens_per_second_per_gpu:4124.482860546062 mfu:19.455889362535988 +Step 9234 | loss:1.0959089994430542 lr:8.50543539603744e-05 tokens_per_second_per_gpu:5393.116524012324 mfu:25.44025080432908 +Step 9235 | loss:0.9525368809700012 lr:8.50512617743402e-05 tokens_per_second_per_gpu:3839.295661027292 mfu:18.110612684452267 +Step 9236 | loss:1.1433570384979248 lr:8.504816932468099e-05 tokens_per_second_per_gpu:5337.243675827457 mfu:25.17668904655016 +Step 9237 | loss:0.9691361784934998 lr:8.504507661142e-05 tokens_per_second_per_gpu:4878.212867669498 mfu:23.011362405737227 +Step 9238 | loss:1.110011100769043 lr:8.504198363458054e-05 tokens_per_second_per_gpu:4336.072145590759 mfu:20.453992120946456 +Step 9239 | loss:1.055845022201538 lr:8.503889039418584e-05 tokens_per_second_per_gpu:3814.461280277455 mfu:17.993464673273007 +Step 9240 | loss:1.2359766960144043 lr:8.503579689025917e-05 tokens_per_second_per_gpu:3207.0427461841273 mfu:15.128167811666227 +Step 9241 | loss:1.1219854354858398 lr:8.503270312282383e-05 tokens_per_second_per_gpu:4903.5852790065455 mfu:23.131048398993844 +Step 9242 | loss:1.0545510053634644 lr:8.502960909190303e-05 tokens_per_second_per_gpu:5310.886231884645 mfu:25.05235648642826 +Step 9243 | loss:1.1083626747131348 lr:8.502651479752008e-05 tokens_per_second_per_gpu:4664.308981643037 mfu:22.00234127958413 +Step 9244 | loss:1.2396360635757446 lr:8.502342023969825e-05 tokens_per_second_per_gpu:4594.738353890825 mfu:21.67416475850386 +Step 9245 | loss:0.960582971572876 lr:8.50203254184608e-05 tokens_per_second_per_gpu:3724.1145371202533 mfu:17.567283671056586 +Step 9246 | loss:1.1299654245376587 lr:8.501723033383102e-05 tokens_per_second_per_gpu:3781.66240645577 mfu:17.838746789391195 +Step 9247 | loss:1.1023731231689453 lr:8.501413498583217e-05 tokens_per_second_per_gpu:4399.797361742359 mfu:20.754594838176583 +Step 9248 | loss:1.0637385845184326 lr:8.501103937448756e-05 tokens_per_second_per_gpu:3150.092331799823 mfu:14.859522990272811 +Step 9249 | loss:1.2339192628860474 lr:8.500794349982045e-05 tokens_per_second_per_gpu:5115.991520272538 mfu:24.133004879287263 +Step 9250 | loss:1.1622103452682495 lr:8.500484736185412e-05 tokens_per_second_per_gpu:3915.0327439391635 mfu:18.46787742662639 +Step 9251 | loss:0.9763438701629639 lr:8.500175096061187e-05 tokens_per_second_per_gpu:5347.200781609994 mfu:25.223658413383813 +Step 9252 | loss:1.1638561487197876 lr:8.499865429611699e-05 tokens_per_second_per_gpu:3918.5784230647273 mfu:18.484603000016914 +Step 9253 | loss:0.9796821475028992 lr:8.499555736839275e-05 tokens_per_second_per_gpu:5351.605491413172 mfu:25.24443618104639 +Step 9254 | loss:1.1196444034576416 lr:8.499246017746248e-05 tokens_per_second_per_gpu:4644.9250509137555 mfu:21.91090397109449 +Step 9255 | loss:1.1371949911117554 lr:8.498936272334944e-05 tokens_per_second_per_gpu:4481.717749463499 mfu:21.141027284116614 +Step 9256 | loss:1.1236767768859863 lr:8.498626500607694e-05 tokens_per_second_per_gpu:3938.442914053642 mfu:18.578307193243276 +Step 9257 | loss:1.2098472118377686 lr:8.498316702566828e-05 tokens_per_second_per_gpu:5338.320152562673 mfu:25.181766970976046 +Step 9258 | loss:1.1298943758010864 lr:8.498006878214674e-05 tokens_per_second_per_gpu:3909.072278769752 mfu:18.439760895462054 +Step 9259 | loss:0.952367901802063 lr:8.497697027553566e-05 tokens_per_second_per_gpu:5322.167021150493 mfu:25.105569894096384 +Step 9260 | loss:0.8858911991119385 lr:8.497387150585831e-05 tokens_per_second_per_gpu:5128.6060761073395 mfu:24.192509891425185 +Step 9261 | loss:1.2064930200576782 lr:8.497077247313802e-05 tokens_per_second_per_gpu:4381.0091032132905 mfu:20.66596741708751 +Step 9262 | loss:1.0641337633132935 lr:8.496767317739809e-05 tokens_per_second_per_gpu:4425.244022729565 mfu:20.874631079701295 +Step 9263 | loss:1.0926389694213867 lr:8.496457361866182e-05 tokens_per_second_per_gpu:4990.487118347785 mfu:23.540979202149174 +Step 9264 | loss:1.2007392644882202 lr:8.496147379695253e-05 tokens_per_second_per_gpu:4194.22614453543 mfu:19.78488033254512 +Step 9265 | loss:1.1125664710998535 lr:8.495837371229354e-05 tokens_per_second_per_gpu:5069.151240089877 mfu:23.912051285889767 +Step 9266 | loss:1.284885048866272 lr:8.495527336470815e-05 tokens_per_second_per_gpu:5355.380023985547 mfu:25.262241295191952 +Step 9267 | loss:1.0896594524383545 lr:8.49521727542197e-05 tokens_per_second_per_gpu:5371.849750448217 mfu:25.339931804941457 +Step 9268 | loss:1.0122184753417969 lr:8.49490718808515e-05 tokens_per_second_per_gpu:4837.849547898193 mfu:22.820961739683494 +Step 9269 | loss:1.2101246118545532 lr:8.494597074462687e-05 tokens_per_second_per_gpu:4146.906993719195 mfu:19.5616678246653 +Step 9270 | loss:1.183910846710205 lr:8.494286934556913e-05 tokens_per_second_per_gpu:3950.2626608784517 mfu:18.634062955672665 +Step 9271 | loss:1.1314027309417725 lr:8.49397676837016e-05 tokens_per_second_per_gpu:3932.665320882695 mfu:18.55105330049688 +Step 9272 | loss:1.0586204528808594 lr:8.493666575904764e-05 tokens_per_second_per_gpu:3244.1382341912986 mfu:15.303153557738948 +Step 9273 | loss:1.095072627067566 lr:8.493356357163053e-05 tokens_per_second_per_gpu:3941.945237091292 mfu:18.5948282485698 +Step 9274 | loss:1.1904993057250977 lr:8.493046112147366e-05 tokens_per_second_per_gpu:3398.5334291766735 mfu:16.03146203502158 +Step 9275 | loss:1.2090975046157837 lr:8.492735840860033e-05 tokens_per_second_per_gpu:5336.477057402908 mfu:25.17307277664297 +Step 9276 | loss:1.0320067405700684 lr:8.492425543303387e-05 tokens_per_second_per_gpu:5359.123902984438 mfu:25.27990181120135 +Step 9277 | loss:1.0070136785507202 lr:8.492115219479764e-05 tokens_per_second_per_gpu:5312.806201421853 mfu:25.061413310315853 +Step 9278 | loss:1.032781958580017 lr:8.491804869391495e-05 tokens_per_second_per_gpu:4800.007237823167 mfu:22.64245310649572 +Step 9279 | loss:1.173632025718689 lr:8.491494493040918e-05 tokens_per_second_per_gpu:5314.39089254527 mfu:25.068888568721192 +Step 9280 | loss:1.0227134227752686 lr:8.491184090430364e-05 tokens_per_second_per_gpu:5279.268303179248 mfu:24.903209322149994 +Step 9281 | loss:1.030990719795227 lr:8.490873661562171e-05 tokens_per_second_per_gpu:5356.909033647198 mfu:25.26945389464191 +Step 9282 | loss:1.0331707000732422 lr:8.490563206438668e-05 tokens_per_second_per_gpu:5328.663753311353 mfu:25.136216088155965 +Step 9283 | loss:1.0540540218353271 lr:8.490252725062198e-05 tokens_per_second_per_gpu:5295.733007657466 mfu:24.980876142342005 +Step 9284 | loss:1.0754203796386719 lr:8.48994221743509e-05 tokens_per_second_per_gpu:5322.738409641008 mfu:25.10826523109507 +Step 9285 | loss:1.1044590473175049 lr:8.489631683559682e-05 tokens_per_second_per_gpu:3103.932684587017 mfu:14.641780058722079 +Step 9286 | loss:1.1242114305496216 lr:8.489321123438307e-05 tokens_per_second_per_gpu:4588.551523793872 mfu:21.64498042535447 +Step 9287 | loss:0.9927685856819153 lr:8.489010537073304e-05 tokens_per_second_per_gpu:4915.932268122824 mfu:23.189291253268898 +Step 9288 | loss:1.213314414024353 lr:8.488699924467007e-05 tokens_per_second_per_gpu:4486.459432203374 mfu:21.163394610615104 +Step 9289 | loss:1.0639930963516235 lr:8.488389285621753e-05 tokens_per_second_per_gpu:4727.102251173385 mfu:22.29854784559532 +Step 9290 | loss:1.1889601945877075 lr:8.488078620539879e-05 tokens_per_second_per_gpu:5339.221232363608 mfu:25.18601751813712 +Step 9291 | loss:1.1255247592926025 lr:8.487767929223719e-05 tokens_per_second_per_gpu:4947.681131458826 mfu:23.339056058539477 +Step 9292 | loss:1.0940734148025513 lr:8.487457211675611e-05 tokens_per_second_per_gpu:5367.574217137234 mfu:25.31976338483235 +Step 9293 | loss:1.1913834810256958 lr:8.487146467897892e-05 tokens_per_second_per_gpu:4447.664788716896 mfu:20.980393658240672 +Step 9294 | loss:1.0342997312545776 lr:8.486835697892903e-05 tokens_per_second_per_gpu:5356.872638432584 mfu:25.269282212204907 +Step 9295 | loss:1.1206860542297363 lr:8.486524901662975e-05 tokens_per_second_per_gpu:5367.598373750347 mfu:25.319877335697942 +Step 9296 | loss:1.2196868658065796 lr:8.486214079210448e-05 tokens_per_second_per_gpu:2941.9579669206596 mfu:13.877717679753331 +Step 9297 | loss:0.9423394799232483 lr:8.485903230537662e-05 tokens_per_second_per_gpu:5322.018207111813 mfu:25.104867912885638 +Step 9298 | loss:1.0029510259628296 lr:8.485592355646951e-05 tokens_per_second_per_gpu:4629.360080335806 mfu:21.837481349220482 +Step 9299 | loss:1.0542192459106445 lr:8.485281454540657e-05 tokens_per_second_per_gpu:5333.980631481772 mfu:25.161296709638794 +Step 9300 | loss:1.1033064126968384 lr:8.484970527221114e-05 tokens_per_second_per_gpu:5372.832903705589 mfu:25.344569506599726 +Step 9301 | loss:1.1554735898971558 lr:8.484659573690666e-05 tokens_per_second_per_gpu:4408.679317120723 mfu:20.796492537114176 +Step 9302 | loss:1.2369612455368042 lr:8.484348593951648e-05 tokens_per_second_per_gpu:5243.490109519957 mfu:24.73443747826981 +Step 9303 | loss:1.1225186586380005 lr:8.484037588006397e-05 tokens_per_second_per_gpu:3785.13247775176 mfu:17.855115707749594 +Step 9304 | loss:1.2571905851364136 lr:8.483726555857257e-05 tokens_per_second_per_gpu:5353.436133322674 mfu:25.253071631273233 +Step 9305 | loss:1.1591335535049438 lr:8.483415497506567e-05 tokens_per_second_per_gpu:5320.911932225612 mfu:25.099649425497468 +Step 9306 | loss:1.3058913946151733 lr:8.483104412956659e-05 tokens_per_second_per_gpu:5340.536815307165 mfu:25.19222334734348 +Step 9307 | loss:1.1584911346435547 lr:8.482793302209881e-05 tokens_per_second_per_gpu:5348.521514993936 mfu:25.229888538096223 +Step 9308 | loss:1.0907495021820068 lr:8.482482165268571e-05 tokens_per_second_per_gpu:5309.622830843114 mfu:25.04639680815691 +Step 9309 | loss:1.123205304145813 lr:8.482171002135068e-05 tokens_per_second_per_gpu:5324.350284468732 mfu:25.11586871967165 +Step 9310 | loss:0.9876810908317566 lr:8.481859812811713e-05 tokens_per_second_per_gpu:4433.461470081298 mfu:20.91339418090051 +Step 9311 | loss:1.1344242095947266 lr:8.481548597300845e-05 tokens_per_second_per_gpu:3760.876371393569 mfu:17.740695515539514 +Step 9312 | loss:1.1015269756317139 lr:8.481237355604807e-05 tokens_per_second_per_gpu:5352.030876646992 mfu:25.246442795772527 +Step 9313 | loss:1.0899684429168701 lr:8.480926087725938e-05 tokens_per_second_per_gpu:3844.29063634207 mfu:18.134174835242742 +Step 9314 | loss:1.0568275451660156 lr:8.48061479366658e-05 tokens_per_second_per_gpu:5302.746495982579 mfu:25.013959963396065 +Step 9315 | loss:1.0331127643585205 lr:8.480303473429071e-05 tokens_per_second_per_gpu:5365.123449900189 mfu:25.3082026976312 +Step 9316 | loss:1.0421159267425537 lr:8.47999212701576e-05 tokens_per_second_per_gpu:5316.917959689817 mfu:25.080809175604138 +Step 9317 | loss:1.1677495241165161 lr:8.479680754428983e-05 tokens_per_second_per_gpu:3872.040360203514 mfu:18.26507501728828 +Step 9318 | loss:1.2560428380966187 lr:8.47936935567108e-05 tokens_per_second_per_gpu:5335.807739398936 mfu:25.16991548942081 +Step 9319 | loss:0.9647753834724426 lr:8.479057930744401e-05 tokens_per_second_per_gpu:5338.0619717049485 mfu:25.180549087819962 +Step 9320 | loss:1.0747332572937012 lr:8.478746479651279e-05 tokens_per_second_per_gpu:4624.541969655623 mfu:21.814753499087608 +Step 9321 | loss:1.2034662961959839 lr:8.478435002394062e-05 tokens_per_second_per_gpu:5279.488188669577 mfu:24.904246559524207 +Step 9322 | loss:1.2666374444961548 lr:8.478123498975092e-05 tokens_per_second_per_gpu:5317.072363948831 mfu:25.08153752683799 +Step 9323 | loss:1.156802773475647 lr:8.477811969396713e-05 tokens_per_second_per_gpu:5365.7032045298365 mfu:25.310937499135587 +Step 9324 | loss:1.1623331308364868 lr:8.477500413661264e-05 tokens_per_second_per_gpu:4127.100711078802 mfu:19.468238210150286 +Step 9325 | loss:1.1013641357421875 lr:8.477188831771093e-05 tokens_per_second_per_gpu:5358.6586064764615 mfu:25.27770692818545 +Step 9326 | loss:1.240696907043457 lr:8.476877223728539e-05 tokens_per_second_per_gpu:4889.338133342816 mfu:23.063842182084322 +Step 9327 | loss:1.0836211442947388 lr:8.476565589535947e-05 tokens_per_second_per_gpu:5052.489301803938 mfu:23.833454080175184 +Step 9328 | loss:1.0564020872116089 lr:8.476253929195663e-05 tokens_per_second_per_gpu:5355.258149055533 mfu:25.261666390353067 +Step 9329 | loss:1.168816328048706 lr:8.47594224271003e-05 tokens_per_second_per_gpu:5345.559641179749 mfu:25.215916873966915 +Step 9330 | loss:1.104833960533142 lr:8.475630530081393e-05 tokens_per_second_per_gpu:3999.04295008713 mfu:18.86416790264555 +Step 9331 | loss:1.1481406688690186 lr:8.475318791312093e-05 tokens_per_second_per_gpu:4259.746208430997 mfu:20.093949652816143 +Step 9332 | loss:1.0911173820495605 lr:8.475007026404477e-05 tokens_per_second_per_gpu:3816.3076053407794 mfu:18.002174103612244 +Step 9333 | loss:1.0907069444656372 lr:8.47469523536089e-05 tokens_per_second_per_gpu:4866.112841052422 mfu:22.954284515707055 +Step 9334 | loss:1.0888981819152832 lr:8.474383418183678e-05 tokens_per_second_per_gpu:3533.459272966291 mfu:16.66793026089998 +Step 9335 | loss:1.123233675956726 lr:8.474071574875184e-05 tokens_per_second_per_gpu:5310.155369786017 mfu:25.0489088852111 +Step 9336 | loss:1.0162943601608276 lr:8.473759705437753e-05 tokens_per_second_per_gpu:3768.232865292885 mfu:17.775397352409872 +Step 9337 | loss:1.0812397003173828 lr:8.473447809873736e-05 tokens_per_second_per_gpu:4688.923116141085 mfu:22.118450351615817 +Step 9338 | loss:1.0321705341339111 lr:8.473135888185471e-05 tokens_per_second_per_gpu:5308.44010939563 mfu:25.040817709296043 +Step 9339 | loss:1.1659150123596191 lr:8.47282394037531e-05 tokens_per_second_per_gpu:5286.250247177072 mfu:24.93614434323021 +Step 9340 | loss:1.10482919216156 lr:8.472511966445596e-05 tokens_per_second_per_gpu:3172.3676020080725 mfu:14.96459924039812 +Step 9341 | loss:1.2630631923675537 lr:8.472199966398677e-05 tokens_per_second_per_gpu:4580.074111681167 mfu:21.604991026023182 +Step 9342 | loss:1.2666652202606201 lr:8.471887940236898e-05 tokens_per_second_per_gpu:5368.963836109923 mfu:25.32631845461997 +Step 9343 | loss:0.9821869730949402 lr:8.471575887962605e-05 tokens_per_second_per_gpu:5205.432570830386 mfu:24.55491357498544 +Step 9344 | loss:1.0441328287124634 lr:8.47126380957815e-05 tokens_per_second_per_gpu:5265.812518639361 mfu:24.839736090681768 +Step 9345 | loss:1.192281723022461 lr:8.470951705085877e-05 tokens_per_second_per_gpu:3624.677433192382 mfu:17.098221886109435 +Step 9346 | loss:1.2161056995391846 lr:8.470639574488132e-05 tokens_per_second_per_gpu:5339.216372634135 mfu:25.185994593964043 +Step 9347 | loss:1.0106819868087769 lr:8.470327417787264e-05 tokens_per_second_per_gpu:5283.494830843915 mfu:24.92314657426421 +Step 9348 | loss:1.178020715713501 lr:8.470015234985623e-05 tokens_per_second_per_gpu:5206.309657729766 mfu:24.55905094353627 +Step 9349 | loss:1.0892852544784546 lr:8.469703026085551e-05 tokens_per_second_per_gpu:5346.867612896571 mfu:25.222086799718156 +Step 9350 | loss:0.989338755607605 lr:8.469390791089402e-05 tokens_per_second_per_gpu:3896.383166603451 mfu:18.37990418838783 +Step 9351 | loss:1.2207077741622925 lr:8.469078529999522e-05 tokens_per_second_per_gpu:5381.496263282322 mfu:25.385436051844735 +Step 9352 | loss:1.0404764413833618 lr:8.468766242818258e-05 tokens_per_second_per_gpu:4106.9648070312505 mfu:19.373253715218492 +Step 9353 | loss:1.078449010848999 lr:8.468453929547962e-05 tokens_per_second_per_gpu:4182.763249789082 mfu:19.730807902255076 +Step 9354 | loss:1.078961968421936 lr:8.46814159019098e-05 tokens_per_second_per_gpu:3585.4166900934288 mfu:16.913022262338167 +Step 9355 | loss:1.1400177478790283 lr:8.467829224749665e-05 tokens_per_second_per_gpu:4376.014128036686 mfu:20.64240526694882 +Step 9356 | loss:1.191652536392212 lr:8.467516833226363e-05 tokens_per_second_per_gpu:3574.2303314993223 mfu:16.86025430026018 +Step 9357 | loss:1.0909643173217773 lr:8.467204415623424e-05 tokens_per_second_per_gpu:5297.776502001243 mfu:24.99051565381769 +Step 9358 | loss:1.0887680053710938 lr:8.466891971943198e-05 tokens_per_second_per_gpu:5315.192161729614 mfu:25.072668292174367 +Step 9359 | loss:1.1434657573699951 lr:8.466579502188035e-05 tokens_per_second_per_gpu:3560.6548296459214 mfu:16.79621634179823 +Step 9360 | loss:1.0689302682876587 lr:8.466267006360287e-05 tokens_per_second_per_gpu:3528.1536292345995 mfu:16.642902634181425 +Step 9361 | loss:1.0203273296356201 lr:8.4659544844623e-05 tokens_per_second_per_gpu:5286.418875500138 mfu:24.936939791801073 +Step 9362 | loss:1.172296404838562 lr:8.465641936496427e-05 tokens_per_second_per_gpu:5336.538823015233 mfu:25.17336413557461 +Step 9363 | loss:0.9843528270721436 lr:8.46532936246502e-05 tokens_per_second_per_gpu:5312.660118005339 mfu:25.060724209915758 +Step 9364 | loss:1.1126855611801147 lr:8.46501676237043e-05 tokens_per_second_per_gpu:4675.823278021847 mfu:22.056656180144515 +Step 9365 | loss:0.9367577433586121 lr:8.464704136215005e-05 tokens_per_second_per_gpu:5311.887737882109 mfu:25.05708076109284 +Step 9366 | loss:0.9712511301040649 lr:8.464391484001098e-05 tokens_per_second_per_gpu:4684.842841392042 mfu:22.099202999457923 +Step 9367 | loss:0.9979421496391296 lr:8.464078805731061e-05 tokens_per_second_per_gpu:3540.2877896838436 mfu:16.700141539321766 +Step 9368 | loss:1.0737040042877197 lr:8.463766101407247e-05 tokens_per_second_per_gpu:4268.421871969709 mfu:20.13487423794908 +Step 9369 | loss:1.1688427925109863 lr:8.463453371032004e-05 tokens_per_second_per_gpu:5362.584633651426 mfu:25.296226668219067 +Step 9370 | loss:1.1985018253326416 lr:8.463140614607688e-05 tokens_per_second_per_gpu:4051.462154873627 mfu:19.111438235262035 +Step 9371 | loss:1.025952696800232 lr:8.462827832136646e-05 tokens_per_second_per_gpu:5327.075380983774 mfu:25.128723464882512 +Step 9372 | loss:0.9971767663955688 lr:8.462515023621237e-05 tokens_per_second_per_gpu:5091.776468888249 mfu:24.01877835039378 +Step 9373 | loss:1.1681301593780518 lr:8.462202189063811e-05 tokens_per_second_per_gpu:4289.5126523201625 mfu:20.234363047320358 +Step 9374 | loss:1.1348071098327637 lr:8.461889328466719e-05 tokens_per_second_per_gpu:3347.5380132662667 mfu:15.79090795745705 +Step 9375 | loss:1.0305992364883423 lr:8.461576441832316e-05 tokens_per_second_per_gpu:5366.417681922549 mfu:25.31430781089942 +Step 9376 | loss:1.1435576677322388 lr:8.461263529162955e-05 tokens_per_second_per_gpu:4250.4603712476355 mfu:20.05014677449503 +Step 9377 | loss:1.0612843036651611 lr:8.46095059046099e-05 tokens_per_second_per_gpu:3311.5181484504524 mfu:15.6209961094982 +Step 9378 | loss:1.0235220193862915 lr:8.460637625728773e-05 tokens_per_second_per_gpu:5341.710370212935 mfu:25.19775920606311 +Step 9379 | loss:1.2704946994781494 lr:8.46032463496866e-05 tokens_per_second_per_gpu:4835.956260215944 mfu:22.812030778658137 +Step 9380 | loss:1.1876410245895386 lr:8.460011618183001e-05 tokens_per_second_per_gpu:4425.082768924217 mfu:20.873870417988734 +Step 9381 | loss:0.9881111979484558 lr:8.459698575374157e-05 tokens_per_second_per_gpu:4999.949891743737 mfu:23.5856167187735 +Step 9382 | loss:1.0344293117523193 lr:8.459385506544477e-05 tokens_per_second_per_gpu:5096.411975224795 mfu:24.040644824682204 +Step 9383 | loss:1.1779392957687378 lr:8.459072411696316e-05 tokens_per_second_per_gpu:5159.156371528437 mfu:24.33662084734439 +Step 9384 | loss:1.0387084484100342 lr:8.458759290832032e-05 tokens_per_second_per_gpu:4667.401944479389 mfu:22.01693131299708 +Step 9385 | loss:1.3247113227844238 lr:8.458446143953978e-05 tokens_per_second_per_gpu:3902.433025003505 mfu:18.408442402673163 +Step 9386 | loss:1.1556119918823242 lr:8.458132971064508e-05 tokens_per_second_per_gpu:4337.934925497922 mfu:20.46277917159157 +Step 9387 | loss:1.0468474626541138 lr:8.457819772165978e-05 tokens_per_second_per_gpu:5379.280892625765 mfu:25.37498577047655 +Step 9388 | loss:1.1799207925796509 lr:8.457506547260745e-05 tokens_per_second_per_gpu:4453.408310552337 mfu:21.007486830684318 +Step 9389 | loss:1.1841477155685425 lr:8.457193296351165e-05 tokens_per_second_per_gpu:5359.728716902219 mfu:25.282754821643415 +Step 9390 | loss:1.0577261447906494 lr:8.456880019439593e-05 tokens_per_second_per_gpu:5362.263054037138 mfu:25.294709722311328 +Step 9391 | loss:1.1173175573349 lr:8.456566716528383e-05 tokens_per_second_per_gpu:5305.724136582112 mfu:25.028006002141932 +Step 9392 | loss:1.276936650276184 lr:8.456253387619897e-05 tokens_per_second_per_gpu:4011.2148132242532 mfu:18.921584657797364 +Step 9393 | loss:0.9544692635536194 lr:8.455940032716485e-05 tokens_per_second_per_gpu:5329.028035268348 mfu:25.13793446830002 +Step 9394 | loss:1.1220552921295166 lr:8.45562665182051e-05 tokens_per_second_per_gpu:4633.122251928898 mfu:21.85522815451776 +Step 9395 | loss:1.0217489004135132 lr:8.455313244934324e-05 tokens_per_second_per_gpu:5141.681486998406 mfu:24.254188835493483 +Step 9396 | loss:1.0753854513168335 lr:8.454999812060287e-05 tokens_per_second_per_gpu:3960.574275221981 mfu:18.68270459987392 +Step 9397 | loss:0.9816305041313171 lr:8.454686353200755e-05 tokens_per_second_per_gpu:4613.824726631352 mfu:21.764198435191908 +Step 9398 | loss:1.2302515506744385 lr:8.454372868358088e-05 tokens_per_second_per_gpu:4971.065711596519 mfu:23.449365112869764 +Step 9399 | loss:1.1150832176208496 lr:8.454059357534639e-05 tokens_per_second_per_gpu:5348.367488495389 mfu:25.229161968822943 +Step 9400 | loss:1.063356637954712 lr:8.45374582073277e-05 tokens_per_second_per_gpu:5250.292134318617 mfu:24.766523789790195 +Step 9401 | loss:1.1153616905212402 lr:8.453432257954839e-05 tokens_per_second_per_gpu:4118.073595630463 mfu:19.425655766395757 +Step 9402 | loss:1.0588524341583252 lr:8.453118669203201e-05 tokens_per_second_per_gpu:5268.639286873368 mfu:24.853070438737976 +Step 9403 | loss:1.1571425199508667 lr:8.452805054480219e-05 tokens_per_second_per_gpu:4276.696567403464 mfu:20.17390738811872 +Step 9404 | loss:1.0479907989501953 lr:8.452491413788249e-05 tokens_per_second_per_gpu:5015.675791351895 mfu:23.65979847034033 +Step 9405 | loss:1.1603716611862183 lr:8.45217774712965e-05 tokens_per_second_per_gpu:5311.010657119415 mfu:25.052943421489804 +Step 9406 | loss:1.2351489067077637 lr:8.451864054506783e-05 tokens_per_second_per_gpu:3104.5892730526 mfu:14.644877298539779 +Step 9407 | loss:0.9551414847373962 lr:8.451550335922005e-05 tokens_per_second_per_gpu:5311.0285476144 mfu:25.05302781400708 +Step 9408 | loss:1.0444159507751465 lr:8.451236591377676e-05 tokens_per_second_per_gpu:4246.425880068618 mfu:20.031115391248427 +Step 9409 | loss:0.957041323184967 lr:8.450922820876157e-05 tokens_per_second_per_gpu:5327.613746729079 mfu:25.131263027957782 +Step 9410 | loss:0.9889757633209229 lr:8.450609024419808e-05 tokens_per_second_per_gpu:5321.941524286016 mfu:25.10450618691291 +Step 9411 | loss:1.138017177581787 lr:8.450295202010987e-05 tokens_per_second_per_gpu:4708.507070664827 mfu:22.2108312064715 +Step 9412 | loss:1.1458203792572021 lr:8.449981353652056e-05 tokens_per_second_per_gpu:5347.275617908307 mfu:25.224011428971348 +Step 9413 | loss:1.0749694108963013 lr:8.449667479345374e-05 tokens_per_second_per_gpu:4338.573440601153 mfu:20.465791156275426 +Step 9414 | loss:1.0343925952911377 lr:8.449353579093304e-05 tokens_per_second_per_gpu:5013.841254449946 mfu:23.651144646769566 +Step 9415 | loss:1.1048964262008667 lr:8.449039652898205e-05 tokens_per_second_per_gpu:5316.567563443854 mfu:25.079156296728165 +Step 9416 | loss:0.979032576084137 lr:8.448725700762439e-05 tokens_per_second_per_gpu:5228.311272058395 mfu:24.662836312186908 +Step 9417 | loss:1.0099437236785889 lr:8.448411722688366e-05 tokens_per_second_per_gpu:3775.5850153338315 mfu:17.810078698559163 +Step 9418 | loss:1.028808832168579 lr:8.44809771867835e-05 tokens_per_second_per_gpu:4664.751022869774 mfu:22.004426463470658 +Step 9419 | loss:0.9627152681350708 lr:8.447783688734749e-05 tokens_per_second_per_gpu:5046.7289592374555 mfu:23.806281561473295 +Step 9420 | loss:1.02428138256073 lr:8.447469632859928e-05 tokens_per_second_per_gpu:5365.801159693778 mfu:25.31139957035677 +Step 9421 | loss:1.1168204545974731 lr:8.447155551056249e-05 tokens_per_second_per_gpu:3641.1572298728256 mfu:17.17595989879434 +Step 9422 | loss:0.9908252954483032 lr:8.446841443326072e-05 tokens_per_second_per_gpu:5365.0446756199 mfu:25.307831105910793 +Step 9423 | loss:1.1348038911819458 lr:8.446527309671762e-05 tokens_per_second_per_gpu:3670.744495650993 mfu:17.31552807957804 +Step 9424 | loss:1.1522111892700195 lr:8.446213150095679e-05 tokens_per_second_per_gpu:4388.145672995329 mfu:20.699631834373747 +Step 9425 | loss:0.9240638017654419 lr:8.445898964600188e-05 tokens_per_second_per_gpu:4652.984715000355 mfu:21.948922781711474 +Step 9426 | loss:0.9815917015075684 lr:8.44558475318765e-05 tokens_per_second_per_gpu:5309.035143735289 mfu:25.043624587799542 +Step 9427 | loss:1.1038087606430054 lr:8.44527051586043e-05 tokens_per_second_per_gpu:3665.628313678786 mfu:17.291394176304486 +Step 9428 | loss:1.0564844608306885 lr:8.44495625262089e-05 tokens_per_second_per_gpu:4300.5549270368665 mfu:20.28645134116539 +Step 9429 | loss:1.286470890045166 lr:8.444641963471395e-05 tokens_per_second_per_gpu:4713.247402071443 mfu:22.23319215850044 +Step 9430 | loss:1.142990231513977 lr:8.44432764841431e-05 tokens_per_second_per_gpu:4887.414994199732 mfu:23.054770406625025 +Step 9431 | loss:1.149938702583313 lr:8.444013307451996e-05 tokens_per_second_per_gpu:5362.503789290682 mfu:25.295845311575963 +Step 9432 | loss:0.9802963733673096 lr:8.443698940586818e-05 tokens_per_second_per_gpu:5338.127390883042 mfu:25.180857681244692 +Step 9433 | loss:1.0211098194122314 lr:8.44338454782114e-05 tokens_per_second_per_gpu:4944.868975002195 mfu:23.325790636730183 +Step 9434 | loss:1.1491119861602783 lr:8.443070129157329e-05 tokens_per_second_per_gpu:4238.05849906384 mfu:19.99164502742648 +Step 9435 | loss:1.0839706659317017 lr:8.44275568459775e-05 tokens_per_second_per_gpu:5372.7198205860395 mfu:25.344036074230843 +Step 9436 | loss:1.2469849586486816 lr:8.442441214144763e-05 tokens_per_second_per_gpu:4934.357048811652 mfu:23.276204087369845 +Step 9437 | loss:1.083639144897461 lr:8.442126717800738e-05 tokens_per_second_per_gpu:4188.011246627887 mfu:19.755563598744317 +Step 9438 | loss:1.135360836982727 lr:8.441812195568037e-05 tokens_per_second_per_gpu:4657.8926040900815 mfu:21.97207413191994 +Step 9439 | loss:1.1492234468460083 lr:8.441497647449029e-05 tokens_per_second_per_gpu:5181.371978276743 mfu:24.44141565474951 +Step 9440 | loss:1.0766998529434204 lr:8.441183073446077e-05 tokens_per_second_per_gpu:3963.034234552613 mfu:18.694308647748844 +Step 9441 | loss:1.1809030771255493 lr:8.44086847356155e-05 tokens_per_second_per_gpu:4705.828698432709 mfu:22.198196867675172 +Step 9442 | loss:1.123928189277649 lr:8.44055384779781e-05 tokens_per_second_per_gpu:4130.428469147326 mfu:19.483935812733584 +Step 9443 | loss:1.0574116706848145 lr:8.440239196157226e-05 tokens_per_second_per_gpu:3626.3125489640624 mfu:17.10593500618397 +Step 9444 | loss:1.1322226524353027 lr:8.439924518642164e-05 tokens_per_second_per_gpu:3497.568261956359 mfu:16.498626238328203 +Step 9445 | loss:1.218931794166565 lr:8.439609815254991e-05 tokens_per_second_per_gpu:4845.022660602359 mfu:22.85479853616827 +Step 9446 | loss:1.1208633184432983 lr:8.439295085998073e-05 tokens_per_second_per_gpu:5306.171838218074 mfu:25.03011788714454 +Step 9447 | loss:0.9477361440658569 lr:8.438980330873779e-05 tokens_per_second_per_gpu:5365.3896715640485 mfu:25.309458510641722 +Step 9448 | loss:1.1686025857925415 lr:8.438665549884475e-05 tokens_per_second_per_gpu:4368.207420409848 mfu:20.605579695111356 +Step 9449 | loss:1.1749597787857056 lr:8.438350743032528e-05 tokens_per_second_per_gpu:4769.178921823693 mfu:22.49703068007311 +Step 9450 | loss:1.0903394222259521 lr:8.438035910320307e-05 tokens_per_second_per_gpu:4174.414952005614 mfu:19.691427557243983 +Step 9451 | loss:1.0941307544708252 lr:8.437721051750178e-05 tokens_per_second_per_gpu:4639.297498866785 mfu:21.884357847929547 +Step 9452 | loss:0.9725204110145569 lr:8.437406167324511e-05 tokens_per_second_per_gpu:5138.744493773954 mfu:24.240334537351025 +Step 9453 | loss:1.235386848449707 lr:8.437091257045674e-05 tokens_per_second_per_gpu:4168.813055635134 mfu:19.665002456282355 +Step 9454 | loss:1.1989775896072388 lr:8.436776320916035e-05 tokens_per_second_per_gpu:3405.213670683971 mfu:16.062973873978233 +Step 9455 | loss:1.1581377983093262 lr:8.436461358937962e-05 tokens_per_second_per_gpu:3748.644467963209 mfu:17.682995540080576 +Step 9456 | loss:1.0875842571258545 lr:8.436146371113824e-05 tokens_per_second_per_gpu:5339.848355379547 mfu:25.188975764401526 +Step 9457 | loss:1.1237362623214722 lr:8.435831357445993e-05 tokens_per_second_per_gpu:5337.177781209516 mfu:25.176378210394553 +Step 9458 | loss:1.0404465198516846 lr:8.435516317936835e-05 tokens_per_second_per_gpu:5326.95731197152 mfu:25.12816651320706 +Step 9459 | loss:1.1602230072021484 lr:8.435201252588721e-05 tokens_per_second_per_gpu:4358.891292546519 mfu:20.56163393048472 +Step 9460 | loss:0.9935887455940247 lr:8.434886161404019e-05 tokens_per_second_per_gpu:5301.708513774211 mfu:25.00906362422116 +Step 9461 | loss:1.108091950416565 lr:8.434571044385099e-05 tokens_per_second_per_gpu:4436.236420691078 mfu:20.926484096382236 +Step 9462 | loss:1.1293387413024902 lr:8.434255901534332e-05 tokens_per_second_per_gpu:2954.986689130219 mfu:13.939176385345052 +Step 9463 | loss:1.2740422487258911 lr:8.43394073285409e-05 tokens_per_second_per_gpu:4808.811431289479 mfu:22.68398398922661 +Step 9464 | loss:1.0487159490585327 lr:8.433625538346741e-05 tokens_per_second_per_gpu:4494.964835557097 mfu:21.203516049406065 +Step 9465 | loss:0.9556743502616882 lr:8.433310318014655e-05 tokens_per_second_per_gpu:5381.756754677485 mfu:25.386664834200136 +Step 9466 | loss:1.1381816864013672 lr:8.432995071860205e-05 tokens_per_second_per_gpu:4575.045904404902 mfu:21.581272114400416 +Step 9467 | loss:1.1973077058792114 lr:8.43267979988576e-05 tokens_per_second_per_gpu:3846.647905961738 mfu:18.145294478230387 +Step 9468 | loss:1.0756795406341553 lr:8.432364502093694e-05 tokens_per_second_per_gpu:5333.933509444683 mfu:25.161074426949995 +Step 9469 | loss:1.1607990264892578 lr:8.432049178486376e-05 tokens_per_second_per_gpu:4132.384655733172 mfu:19.493163478618868 +Step 9470 | loss:1.2185428142547607 lr:8.431733829066177e-05 tokens_per_second_per_gpu:4178.292892730856 mfu:19.709720465289866 +Step 9471 | loss:1.1004693508148193 lr:8.431418453835472e-05 tokens_per_second_per_gpu:5366.437242527347 mfu:25.314400081609637 +Step 9472 | loss:1.0388115644454956 lr:8.43110305279663e-05 tokens_per_second_per_gpu:4287.509756122843 mfu:20.224915044228272 +Step 9473 | loss:1.0784398317337036 lr:8.430787625952023e-05 tokens_per_second_per_gpu:4940.62460762251 mfu:23.30576923163646 +Step 9474 | loss:1.112295150756836 lr:8.430472173304027e-05 tokens_per_second_per_gpu:4870.633699349254 mfu:22.975610175631175 +Step 9475 | loss:1.0327966213226318 lr:8.430156694855011e-05 tokens_per_second_per_gpu:5154.459259537984 mfu:24.314463768675473 +Step 9476 | loss:1.0589793920516968 lr:8.429841190607348e-05 tokens_per_second_per_gpu:5340.739067432017 mfu:25.193177405124175 +Step 9477 | loss:1.1662285327911377 lr:8.429525660563411e-05 tokens_per_second_per_gpu:4086.8624978425846 mfu:19.278427692967934 +Step 9478 | loss:1.0978357791900635 lr:8.429210104725575e-05 tokens_per_second_per_gpu:4941.960280122287 mfu:23.312069826707138 +Step 9479 | loss:1.0737872123718262 lr:8.428894523096214e-05 tokens_per_second_per_gpu:5307.98106094781 mfu:25.038652299445932 +Step 9480 | loss:1.1222048997879028 lr:8.428578915677698e-05 tokens_per_second_per_gpu:3816.126975314825 mfu:18.001322040961156 +Step 9481 | loss:1.0880378484725952 lr:8.428263282472402e-05 tokens_per_second_per_gpu:4029.9353700569013 mfu:19.009892718433317 +Step 9482 | loss:1.136552333831787 lr:8.427947623482702e-05 tokens_per_second_per_gpu:4703.120399343211 mfu:22.18542135878654 +Step 9483 | loss:1.1562480926513672 lr:8.427631938710969e-05 tokens_per_second_per_gpu:4588.844805766059 mfu:21.646363886456335 +Step 9484 | loss:1.1833164691925049 lr:8.427316228159578e-05 tokens_per_second_per_gpu:4522.5646773510425 mfu:21.333709212166692 +Step 9485 | loss:1.2236425876617432 lr:8.427000491830905e-05 tokens_per_second_per_gpu:5362.040367021003 mfu:25.293659269661894 +Step 9486 | loss:1.0911896228790283 lr:8.426684729727326e-05 tokens_per_second_per_gpu:5342.826395269863 mfu:25.203023686670143 +Step 9487 | loss:0.9708923101425171 lr:8.426368941851211e-05 tokens_per_second_per_gpu:5356.585371422912 mfu:25.267927124707484 +Step 9488 | loss:1.0926414728164673 lr:8.426053128204941e-05 tokens_per_second_per_gpu:3425.156497155671 mfu:16.157047589042303 +Step 9489 | loss:1.1789292097091675 lr:8.425737288790885e-05 tokens_per_second_per_gpu:4265.393604952867 mfu:20.120589385755 +Step 9490 | loss:1.0843058824539185 lr:8.425421423611424e-05 tokens_per_second_per_gpu:4205.529531521406 mfu:19.838200337514657 +Step 9491 | loss:1.085300326347351 lr:8.42510553266893e-05 tokens_per_second_per_gpu:3963.7646202048604 mfu:18.69775400148714 +Step 9492 | loss:1.1377484798431396 lr:8.424789615965783e-05 tokens_per_second_per_gpu:4650.996506756044 mfu:21.939544064199755 +Step 9493 | loss:1.1710219383239746 lr:8.424473673504354e-05 tokens_per_second_per_gpu:3560.2304023445427 mfu:16.794214245802863 +Step 9494 | loss:1.1915431022644043 lr:8.424157705287022e-05 tokens_per_second_per_gpu:5365.773464858275 mfu:25.311268929092478 +Step 9495 | loss:0.9897809028625488 lr:8.423841711316161e-05 tokens_per_second_per_gpu:4263.011002572684 mfu:20.1093502438138 +Step 9496 | loss:1.1474452018737793 lr:8.423525691594152e-05 tokens_per_second_per_gpu:4628.5856462515 mfu:21.83382821151273 +Step 9497 | loss:1.0297034978866577 lr:8.423209646123368e-05 tokens_per_second_per_gpu:5343.447482817751 mfu:25.205953462601997 +Step 9498 | loss:0.9718718528747559 lr:8.42289357490619e-05 tokens_per_second_per_gpu:5332.797379248574 mfu:25.155715106971147 +Step 9499 | loss:0.9586193561553955 lr:8.422577477944991e-05 tokens_per_second_per_gpu:5379.352974596209 mfu:25.375325793429585 +Step 9500 | loss:1.0671517848968506 lr:8.422261355242151e-05 tokens_per_second_per_gpu:2859.6339467844778 mfu:13.489381230844755 +Step 9501 | loss:0.9827224612236023 lr:8.421945206800044e-05 tokens_per_second_per_gpu:4704.022379115607 mfu:22.189676151266546 +Step 9502 | loss:1.1099884510040283 lr:8.421629032621052e-05 tokens_per_second_per_gpu:5356.954350556491 mfu:25.269667662234905 +Step 9503 | loss:1.1363166570663452 lr:8.421312832707552e-05 tokens_per_second_per_gpu:3549.3931766833657 mfu:16.74309320333773 +Step 9504 | loss:1.0712709426879883 lr:8.420996607061922e-05 tokens_per_second_per_gpu:5077.439569283182 mfu:23.95114874882937 +Step 9505 | loss:1.1247665882110596 lr:8.420680355686539e-05 tokens_per_second_per_gpu:5312.182067092491 mfu:25.05846916219595 +Step 9506 | loss:1.1239328384399414 lr:8.420364078583782e-05 tokens_per_second_per_gpu:4835.94434753831 mfu:22.81197458452522 +Step 9507 | loss:1.1640112400054932 lr:8.420047775756032e-05 tokens_per_second_per_gpu:5360.304756604153 mfu:25.285472099199943 +Step 9508 | loss:1.0342578887939453 lr:8.419731447205665e-05 tokens_per_second_per_gpu:4930.804260948615 mfu:23.259444980851782 +Step 9509 | loss:1.052398920059204 lr:8.419415092935062e-05 tokens_per_second_per_gpu:5346.988802311132 mfu:25.222658470863685 +Step 9510 | loss:1.0610617399215698 lr:8.419098712946601e-05 tokens_per_second_per_gpu:5325.609633116333 mfu:25.121809282109414 +Step 9511 | loss:1.0914217233657837 lr:8.418782307242663e-05 tokens_per_second_per_gpu:5262.428413201647 mfu:24.82377268794415 +Step 9512 | loss:0.9969432950019836 lr:8.418465875825627e-05 tokens_per_second_per_gpu:5346.723306630864 mfu:25.221406082441625 +Step 9513 | loss:1.141784429550171 lr:8.418149418697873e-05 tokens_per_second_per_gpu:5346.109364460994 mfu:25.218510012477015 +Step 9514 | loss:1.192559838294983 lr:8.41783293586178e-05 tokens_per_second_per_gpu:5199.22885920598 mfu:24.525649608790776 +Step 9515 | loss:1.079669713973999 lr:8.417516427319731e-05 tokens_per_second_per_gpu:3995.3638377798466 mfu:18.84681290217081 +Step 9516 | loss:1.023077130317688 lr:8.417199893074105e-05 tokens_per_second_per_gpu:5274.966554746765 mfu:24.882917240840005 +Step 9517 | loss:1.193814992904663 lr:8.416883333127281e-05 tokens_per_second_per_gpu:4470.995756837768 mfu:21.09044981554046 +Step 9518 | loss:1.2162960767745972 lr:8.416566747481642e-05 tokens_per_second_per_gpu:3186.984017747073 mfu:15.033547367256665 +Step 9519 | loss:1.0619944334030151 lr:8.416250136139569e-05 tokens_per_second_per_gpu:4513.618392761687 mfu:21.291508061365157 +Step 9520 | loss:1.120158076286316 lr:8.415933499103442e-05 tokens_per_second_per_gpu:5328.913826492714 mfu:25.13739572601933 +Step 9521 | loss:1.117427945137024 lr:8.415616836375643e-05 tokens_per_second_per_gpu:4059.0103859398523 mfu:19.14704452906244 +Step 9522 | loss:1.1712381839752197 lr:8.415300147958556e-05 tokens_per_second_per_gpu:4945.939259609117 mfu:23.33083935183042 +Step 9523 | loss:1.3282296657562256 lr:8.41498343385456e-05 tokens_per_second_per_gpu:5211.333792160263 mfu:24.58275064285066 +Step 9524 | loss:1.123163104057312 lr:8.414666694066036e-05 tokens_per_second_per_gpu:5292.682608180533 mfu:24.966486887557494 +Step 9525 | loss:0.9976801872253418 lr:8.414349928595371e-05 tokens_per_second_per_gpu:4762.939463758897 mfu:22.467598091820147 +Step 9526 | loss:1.0242818593978882 lr:8.414033137444944e-05 tokens_per_second_per_gpu:5321.253169931621 mfu:25.101259101977867 +Step 9527 | loss:1.051853060722351 lr:8.413716320617135e-05 tokens_per_second_per_gpu:4358.632654465785 mfu:20.56041389052969 +Step 9528 | loss:1.0245599746704102 lr:8.413399478114333e-05 tokens_per_second_per_gpu:4416.4709155228675 mfu:20.83324683616086 +Step 9529 | loss:1.0174883604049683 lr:8.413082609938917e-05 tokens_per_second_per_gpu:5320.236130067222 mfu:25.096461551412602 +Step 9530 | loss:1.0972857475280762 lr:8.412765716093272e-05 tokens_per_second_per_gpu:5254.888478830408 mfu:24.788205531069554 +Step 9531 | loss:1.108756184577942 lr:8.412448796579778e-05 tokens_per_second_per_gpu:5029.6323433176685 mfu:23.725633907196194 +Step 9532 | loss:1.2721422910690308 lr:8.412131851400824e-05 tokens_per_second_per_gpu:3832.8805536115046 mfu:18.080351528241806 +Step 9533 | loss:1.190547227859497 lr:8.41181488055879e-05 tokens_per_second_per_gpu:3431.7135449399907 mfu:16.187978302188654 +Step 9534 | loss:1.1246742010116577 lr:8.411497884056061e-05 tokens_per_second_per_gpu:4536.127915490646 mfu:21.397689320598257 +Step 9535 | loss:1.1197539567947388 lr:8.411180861895021e-05 tokens_per_second_per_gpu:5343.232342618256 mfu:25.204938609574274 +Step 9536 | loss:0.9794138669967651 lr:8.410863814078054e-05 tokens_per_second_per_gpu:4283.6288269535125 mfu:20.20660803918174 +Step 9537 | loss:1.0918290615081787 lr:8.410546740607545e-05 tokens_per_second_per_gpu:5334.5492181696145 mfu:25.163978830056237 +Step 9538 | loss:1.1516423225402832 lr:8.410229641485879e-05 tokens_per_second_per_gpu:5354.51398608688 mfu:25.258156046662638 +Step 9539 | loss:1.1042389869689941 lr:8.409912516715441e-05 tokens_per_second_per_gpu:5329.714924690766 mfu:25.141174642900985 +Step 9540 | loss:1.048632025718689 lr:8.409595366298617e-05 tokens_per_second_per_gpu:4804.680668779223 mfu:22.66449847768498 +Step 9541 | loss:1.115402102470398 lr:8.40927819023779e-05 tokens_per_second_per_gpu:5348.704473946625 mfu:25.23075158669207 +Step 9542 | loss:1.049721360206604 lr:8.408960988535346e-05 tokens_per_second_per_gpu:4821.996050920383 mfu:22.74617808955366 +Step 9543 | loss:1.1942017078399658 lr:8.408643761193671e-05 tokens_per_second_per_gpu:5310.654748223568 mfu:25.051264538503776 +Step 9544 | loss:1.103531002998352 lr:8.408326508215154e-05 tokens_per_second_per_gpu:5275.330517398711 mfu:24.884634114767668 +Step 9545 | loss:1.0002309083938599 lr:8.408009229602176e-05 tokens_per_second_per_gpu:4218.243376675988 mfu:19.898173714314485 +Step 9546 | loss:1.186934471130371 lr:8.407691925357127e-05 tokens_per_second_per_gpu:5316.693203047697 mfu:25.079748960175824 +Step 9547 | loss:1.1182986497879028 lr:8.407374595482392e-05 tokens_per_second_per_gpu:4020.4276470625605 mfu:18.965043166884197 +Step 9548 | loss:1.1288831233978271 lr:8.407057239980358e-05 tokens_per_second_per_gpu:4226.361836704567 mfu:19.936469875421402 +Step 9549 | loss:1.1209158897399902 lr:8.40673985885341e-05 tokens_per_second_per_gpu:4385.072982776933 mfu:20.68513743035856 +Step 9550 | loss:1.018493413925171 lr:8.406422452103939e-05 tokens_per_second_per_gpu:5342.381718441137 mfu:25.20092607019936 +Step 9551 | loss:1.0872174501419067 lr:8.406105019734329e-05 tokens_per_second_per_gpu:4371.0762929843495 mfu:20.619112656525417 +Step 9552 | loss:0.9927384853363037 lr:8.40578756174697e-05 tokens_per_second_per_gpu:4282.0389425627045 mfu:20.19910828324805 +Step 9553 | loss:1.0503343343734741 lr:8.405470078144247e-05 tokens_per_second_per_gpu:5312.302382253045 mfu:25.05903670933617 +Step 9554 | loss:1.1605831384658813 lr:8.405152568928549e-05 tokens_per_second_per_gpu:3829.8628645891336 mfu:18.066116574252444 +Step 9555 | loss:1.093666672706604 lr:8.404835034102264e-05 tokens_per_second_per_gpu:5346.29788078406 mfu:25.21939927613712 +Step 9556 | loss:1.1105619668960571 lr:8.404517473667779e-05 tokens_per_second_per_gpu:4510.902196583159 mfu:21.278695300560187 +Step 9557 | loss:1.2509082555770874 lr:8.404199887627486e-05 tokens_per_second_per_gpu:4137.0807246239965 mfu:19.51531563680886 +Step 9558 | loss:1.0467393398284912 lr:8.40388227598377e-05 tokens_per_second_per_gpu:4858.572114494007 mfu:22.918713621950396 +Step 9559 | loss:1.2021206617355347 lr:8.40356463873902e-05 tokens_per_second_per_gpu:5362.2116210915865 mfu:25.294467104331797 +Step 9560 | loss:1.075467586517334 lr:8.403246975895629e-05 tokens_per_second_per_gpu:5083.1599432198445 mfu:23.97813272868504 +Step 9561 | loss:1.0219210386276245 lr:8.40292928745598e-05 tokens_per_second_per_gpu:5288.735333714134 mfu:24.94786692042502 +Step 9562 | loss:1.154651165008545 lr:8.402611573422469e-05 tokens_per_second_per_gpu:5242.065253146252 mfu:24.727716187649484 +Step 9563 | loss:1.1336814165115356 lr:8.40229383379748e-05 tokens_per_second_per_gpu:5280.7526032365895 mfu:24.910211018768177 +Step 9564 | loss:1.1523332595825195 lr:8.401976068583405e-05 tokens_per_second_per_gpu:3478.1125296261207 mfu:16.406850229436237 +Step 9565 | loss:1.1780052185058594 lr:8.401658277782635e-05 tokens_per_second_per_gpu:5210.721146874663 mfu:24.579860690510415 +Step 9566 | loss:1.135408878326416 lr:8.401340461397558e-05 tokens_per_second_per_gpu:4970.7841194471275 mfu:23.448036794656513 +Step 9567 | loss:1.0420818328857422 lr:8.401022619430565e-05 tokens_per_second_per_gpu:5278.491814753968 mfu:24.899546493765364 +Step 9568 | loss:0.9832069873809814 lr:8.400704751884048e-05 tokens_per_second_per_gpu:4143.085001440988 mfu:19.543638834989874 +Step 9569 | loss:1.068551778793335 lr:8.400386858760396e-05 tokens_per_second_per_gpu:3359.840184805469 mfu:15.84893939957437 +Step 9570 | loss:1.1454548835754395 lr:8.400068940062e-05 tokens_per_second_per_gpu:4223.491557038732 mfu:19.92293027651745 +Step 9571 | loss:0.9963129758834839 lr:8.399750995791253e-05 tokens_per_second_per_gpu:5007.4369767023545 mfu:23.62093457595172 +Step 9572 | loss:1.2545280456542969 lr:8.399433025950544e-05 tokens_per_second_per_gpu:4352.175641165035 mfu:20.52995505710998 +Step 9573 | loss:0.9613363146781921 lr:8.399115030542267e-05 tokens_per_second_per_gpu:5293.497373805864 mfu:24.970330276025532 +Step 9574 | loss:1.008423089981079 lr:8.39879700956881e-05 tokens_per_second_per_gpu:5129.287032931513 mfu:24.195722080948023 +Step 9575 | loss:1.1309350728988647 lr:8.398478963032567e-05 tokens_per_second_per_gpu:3770.3167390638914 mfu:17.785227340533236 +Step 9576 | loss:1.005264401435852 lr:8.398160890935932e-05 tokens_per_second_per_gpu:5327.784732085387 mfu:25.1320695950566 +Step 9577 | loss:1.0131399631500244 lr:8.397842793281294e-05 tokens_per_second_per_gpu:4442.568897391316 mfu:20.956355469409193 +Step 9578 | loss:1.0553398132324219 lr:8.397524670071046e-05 tokens_per_second_per_gpu:5302.128508154247 mfu:25.01104480937023 +Step 9579 | loss:0.9986594915390015 lr:8.397206521307584e-05 tokens_per_second_per_gpu:4334.827057462938 mfu:20.448118827813584 +Step 9580 | loss:1.2690542936325073 lr:8.396888346993297e-05 tokens_per_second_per_gpu:4481.567978182554 mfu:21.140320787430777 +Step 9581 | loss:1.072107195854187 lr:8.396570147130578e-05 tokens_per_second_per_gpu:4894.8600075726745 mfu:23.089889804955458 +Step 9582 | loss:1.094319462776184 lr:8.396251921721823e-05 tokens_per_second_per_gpu:3351.5608748462155 mfu:15.809884481900632 +Step 9583 | loss:1.245195746421814 lr:8.395933670769424e-05 tokens_per_second_per_gpu:4593.082436081655 mfu:21.66635351166929 +Step 9584 | loss:0.9937330484390259 lr:8.395615394275773e-05 tokens_per_second_per_gpu:5052.0589182749545 mfu:23.83142388763796 +Step 9585 | loss:1.0627455711364746 lr:8.395297092243266e-05 tokens_per_second_per_gpu:4839.369501747435 mfu:22.828131621321234 +Step 9586 | loss:1.156489372253418 lr:8.394978764674297e-05 tokens_per_second_per_gpu:3655.787680594809 mfu:17.24497423106223 +Step 9587 | loss:1.0660768747329712 lr:8.39466041157126e-05 tokens_per_second_per_gpu:4788.281863720308 mfu:22.58714251629742 +Step 9588 | loss:1.0266531705856323 lr:8.394342032936548e-05 tokens_per_second_per_gpu:4312.265210405114 mfu:20.34169074579378 +Step 9589 | loss:1.030548095703125 lr:8.394023628772556e-05 tokens_per_second_per_gpu:4965.143365931626 mfu:23.421428397912102 +Step 9590 | loss:1.0851905345916748 lr:8.39370519908168e-05 tokens_per_second_per_gpu:5293.863426182449 mfu:24.972057007540858 +Step 9591 | loss:1.1684216260910034 lr:8.393386743866313e-05 tokens_per_second_per_gpu:3852.7939544482897 mfu:18.174286437565453 +Step 9592 | loss:1.1615455150604248 lr:8.393068263128852e-05 tokens_per_second_per_gpu:4183.646143771679 mfu:19.734972663808236 +Step 9593 | loss:1.0503051280975342 lr:8.392749756871691e-05 tokens_per_second_per_gpu:5313.439223975744 mfu:25.064399385707343 +Step 9594 | loss:1.0328114032745361 lr:8.392431225097228e-05 tokens_per_second_per_gpu:4138.209877738016 mfu:19.52064204470108 +Step 9595 | loss:1.1777077913284302 lr:8.392112667807856e-05 tokens_per_second_per_gpu:4711.394985434158 mfu:22.224453993167504 +Step 9596 | loss:1.128993034362793 lr:8.391794085005973e-05 tokens_per_second_per_gpu:4019.6100498626743 mfu:18.961186421395865 +Step 9597 | loss:1.1392537355422974 lr:8.391475476693971e-05 tokens_per_second_per_gpu:3161.3175520739446 mfu:14.912474269525912 +Step 9598 | loss:1.1671030521392822 lr:8.391156842874251e-05 tokens_per_second_per_gpu:5380.350408226635 mfu:25.380030858044034 +Step 9599 | loss:1.0558675527572632 lr:8.390838183549208e-05 tokens_per_second_per_gpu:5334.464411105276 mfu:25.163578780664174 +Step 9600 | loss:1.0885947942733765 lr:8.390519498721236e-05 tokens_per_second_per_gpu:5315.610280694915 mfu:25.074640630672143 +Step 9601 | loss:1.0432798862457275 lr:8.390200788392738e-05 tokens_per_second_per_gpu:5337.854614780372 mfu:25.179570949827998 +Step 9602 | loss:0.9658728241920471 lr:8.389882052566105e-05 tokens_per_second_per_gpu:4492.624027440872 mfu:21.192474058138785 +Step 9603 | loss:1.0797871351242065 lr:8.389563291243737e-05 tokens_per_second_per_gpu:5344.855495445603 mfu:25.2125952983993 +Step 9604 | loss:1.125569224357605 lr:8.389244504428031e-05 tokens_per_second_per_gpu:3947.137708638586 mfu:18.619322022785383 +Step 9605 | loss:1.0347539186477661 lr:8.388925692121383e-05 tokens_per_second_per_gpu:4785.3751180890085 mfu:22.573430901212635 +Step 9606 | loss:1.0354723930358887 lr:8.388606854326196e-05 tokens_per_second_per_gpu:5320.10592752824 mfu:25.095847363821377 +Step 9607 | loss:0.9905132055282593 lr:8.38828799104486e-05 tokens_per_second_per_gpu:5339.801511027708 mfu:25.188754791601433 +Step 9608 | loss:0.9852927327156067 lr:8.387969102279781e-05 tokens_per_second_per_gpu:4508.952328714858 mfu:21.269497441143645 +Step 9609 | loss:1.085754632949829 lr:8.387650188033354e-05 tokens_per_second_per_gpu:4689.262611915418 mfu:22.120051811107295 +Step 9610 | loss:1.1386021375656128 lr:8.387331248307975e-05 tokens_per_second_per_gpu:4882.873343658882 mfu:23.033346666138346 +Step 9611 | loss:1.1414629220962524 lr:8.387012283106048e-05 tokens_per_second_per_gpu:4604.420668613957 mfu:21.719837889026106 +Step 9612 | loss:1.1338889598846436 lr:8.386693292429967e-05 tokens_per_second_per_gpu:4213.531584300371 mfu:19.87594738576402 +Step 9613 | loss:1.0787692070007324 lr:8.386374276282136e-05 tokens_per_second_per_gpu:2958.565916719076 mfu:13.95606021256089 +Step 9614 | loss:1.177465796470642 lr:8.386055234664951e-05 tokens_per_second_per_gpu:5001.626276511764 mfu:23.593524511745024 +Step 9615 | loss:1.153518557548523 lr:8.385736167580814e-05 tokens_per_second_per_gpu:4414.460331521652 mfu:20.823762568386122 +Step 9616 | loss:1.005547285079956 lr:8.385417075032122e-05 tokens_per_second_per_gpu:5362.51106497083 mfu:25.29587963220058 +Step 9617 | loss:1.0750148296356201 lr:8.385097957021275e-05 tokens_per_second_per_gpu:5335.652180381545 mfu:25.16918169099467 +Step 9618 | loss:1.267430067062378 lr:8.384778813550675e-05 tokens_per_second_per_gpu:4976.808696134667 mfu:23.476455750790414 +Step 9619 | loss:1.1540523767471313 lr:8.384459644622723e-05 tokens_per_second_per_gpu:4232.050968841221 mfu:19.963306482376037 +Step 9620 | loss:1.3172814846038818 lr:8.384140450239817e-05 tokens_per_second_per_gpu:5328.564562137637 mfu:25.135748186465786 +Step 9621 | loss:1.1318180561065674 lr:8.383821230404359e-05 tokens_per_second_per_gpu:3908.8685690372754 mfu:18.438799961897498 +Step 9622 | loss:1.026470422744751 lr:8.383501985118749e-05 tokens_per_second_per_gpu:3699.94833820774 mfu:17.453287587607555 +Step 9623 | loss:0.9547116160392761 lr:8.38318271438539e-05 tokens_per_second_per_gpu:5355.861190611639 mfu:25.264511040262718 +Step 9624 | loss:1.06285560131073 lr:8.382863418206682e-05 tokens_per_second_per_gpu:3862.7785125994783 mfu:18.221385301905894 +Step 9625 | loss:1.029017448425293 lr:8.382544096585027e-05 tokens_per_second_per_gpu:5331.7772440157705 mfu:25.150902955024996 +Step 9626 | loss:1.0842818021774292 lr:8.382224749522825e-05 tokens_per_second_per_gpu:4147.706644523887 mfu:19.56543991394553 +Step 9627 | loss:1.187515139579773 lr:8.38190537702248e-05 tokens_per_second_per_gpu:5364.717621433538 mfu:25.30628833551253 +Step 9628 | loss:1.1254950761795044 lr:8.381585979086392e-05 tokens_per_second_per_gpu:5347.899257164099 mfu:25.22695324174475 +Step 9629 | loss:1.1475144624710083 lr:8.381266555716966e-05 tokens_per_second_per_gpu:5349.675824649106 mfu:25.235333613685196 +Step 9630 | loss:1.0830864906311035 lr:8.380947106916601e-05 tokens_per_second_per_gpu:4456.976568841234 mfu:21.02431891384069 +Step 9631 | loss:1.172694444656372 lr:8.380627632687703e-05 tokens_per_second_per_gpu:4319.1240728690145 mfu:20.37404517027861 +Step 9632 | loss:1.0085381269454956 lr:8.380308133032672e-05 tokens_per_second_per_gpu:5317.529235038847 mfu:25.08369266571984 +Step 9633 | loss:1.0940762758255005 lr:8.379988607953913e-05 tokens_per_second_per_gpu:5323.05388721243 mfu:25.109753392625493 +Step 9634 | loss:1.1954411268234253 lr:8.379669057453829e-05 tokens_per_second_per_gpu:3207.149202768877 mfu:15.128669985539856 +Step 9635 | loss:1.0902267694473267 lr:8.379349481534822e-05 tokens_per_second_per_gpu:4079.183489165033 mfu:19.242204498861945 +Step 9636 | loss:1.077841877937317 lr:8.379029880199296e-05 tokens_per_second_per_gpu:5379.3707807294 mfu:25.37540978799789 +Step 9637 | loss:1.0394526720046997 lr:8.378710253449655e-05 tokens_per_second_per_gpu:5325.263890307441 mfu:25.120178354290072 +Step 9638 | loss:0.9583799242973328 lr:8.378390601288304e-05 tokens_per_second_per_gpu:5307.710822107065 mfu:25.03737753672664 +Step 9639 | loss:1.0950578451156616 lr:8.378070923717645e-05 tokens_per_second_per_gpu:4730.840151537465 mfu:22.316180159364247 +Step 9640 | loss:1.0836352109909058 lr:8.377751220740084e-05 tokens_per_second_per_gpu:5339.943352947695 mfu:25.189423884138733 +Step 9641 | loss:1.0277836322784424 lr:8.377431492358026e-05 tokens_per_second_per_gpu:4618.98021309532 mfu:21.788517744459142 +Step 9642 | loss:1.0933181047439575 lr:8.377111738573872e-05 tokens_per_second_per_gpu:4302.112103621196 mfu:20.293796808795392 +Step 9643 | loss:1.073211669921875 lr:8.376791959390034e-05 tokens_per_second_per_gpu:3851.315946080207 mfu:18.167314419918977 +Step 9644 | loss:1.067610263824463 lr:8.376472154808909e-05 tokens_per_second_per_gpu:5326.277172312236 mfu:25.124958178390422 +Step 9645 | loss:0.9741713404655457 lr:8.376152324832909e-05 tokens_per_second_per_gpu:3820.585398073091 mfu:18.02235318179726 +Step 9646 | loss:1.0569875240325928 lr:8.375832469464435e-05 tokens_per_second_per_gpu:5396.442865741609 mfu:25.45594172579854 +Step 9647 | loss:1.0471656322479248 lr:8.375512588705894e-05 tokens_per_second_per_gpu:4726.943041069531 mfu:22.297796824371392 +Step 9648 | loss:1.0108208656311035 lr:8.375192682559692e-05 tokens_per_second_per_gpu:4059.6390261928714 mfu:19.15000993239302 +Step 9649 | loss:1.0877089500427246 lr:8.374872751028235e-05 tokens_per_second_per_gpu:4587.919677154317 mfu:21.641999896951532 +Step 9650 | loss:1.2087912559509277 lr:8.37455279411393e-05 tokens_per_second_per_gpu:3715.438810246981 mfu:17.526358787163716 +Step 9651 | loss:1.0601941347122192 lr:8.374232811819182e-05 tokens_per_second_per_gpu:4945.533068520746 mfu:23.328923279163227 +Step 9652 | loss:1.002242088317871 lr:8.373912804146401e-05 tokens_per_second_per_gpu:4207.287865605817 mfu:19.846494699393553 +Step 9653 | loss:1.3004916906356812 lr:8.373592771097989e-05 tokens_per_second_per_gpu:5384.762014497994 mfu:25.400841157523676 +Step 9654 | loss:1.0671513080596924 lr:8.373272712676356e-05 tokens_per_second_per_gpu:4749.8105701426775 mfu:22.405666860612207 +Step 9655 | loss:1.0044816732406616 lr:8.372952628883907e-05 tokens_per_second_per_gpu:5321.460793980209 mfu:25.102238500039373 +Step 9656 | loss:1.0310689210891724 lr:8.372632519723053e-05 tokens_per_second_per_gpu:4981.836065547173 mfu:23.500170710066875 +Step 9657 | loss:0.9723007082939148 lr:8.3723123851962e-05 tokens_per_second_per_gpu:5224.2124282724435 mfu:24.64350136671371 +Step 9658 | loss:1.097792387008667 lr:8.371992225305754e-05 tokens_per_second_per_gpu:5299.950624310664 mfu:25.000771359694745 +Step 9659 | loss:1.0581719875335693 lr:8.371672040054123e-05 tokens_per_second_per_gpu:4007.1919211464224 mfu:18.90260798949007 +Step 9660 | loss:0.9772898554801941 lr:8.371351829443718e-05 tokens_per_second_per_gpu:4677.700585852788 mfu:22.06551176148481 +Step 9661 | loss:1.1012942790985107 lr:8.371031593476946e-05 tokens_per_second_per_gpu:4136.806985464376 mfu:19.514024362488215 +Step 9662 | loss:1.1122020483016968 lr:8.370711332156214e-05 tokens_per_second_per_gpu:5353.713652019261 mfu:25.254380734314605 +Step 9663 | loss:1.0269596576690674 lr:8.370391045483934e-05 tokens_per_second_per_gpu:4408.014189320429 mfu:20.793355015797687 +Step 9664 | loss:0.9848129153251648 lr:8.370070733462509e-05 tokens_per_second_per_gpu:4596.885371072576 mfu:21.684292604869086 +Step 9665 | loss:1.2066839933395386 lr:8.369750396094356e-05 tokens_per_second_per_gpu:5375.207831239822 mfu:25.355772444983124 +Step 9666 | loss:0.9679281711578369 lr:8.369430033381877e-05 tokens_per_second_per_gpu:5338.15148209209 mfu:25.180971323588157 +Step 9667 | loss:1.1059571504592896 lr:8.369109645327487e-05 tokens_per_second_per_gpu:4970.074488360012 mfu:23.444689343743036 +Step 9668 | loss:0.999656617641449 lr:8.368789231933593e-05 tokens_per_second_per_gpu:4522.787683629492 mfu:21.334761170830937 +Step 9669 | loss:1.0201804637908936 lr:8.368468793202605e-05 tokens_per_second_per_gpu:4904.569443263118 mfu:23.135690869707833 +Step 9670 | loss:1.0785897970199585 lr:8.368148329136934e-05 tokens_per_second_per_gpu:5345.367139189582 mfu:25.215008809235076 +Step 9671 | loss:1.120988368988037 lr:8.367827839738989e-05 tokens_per_second_per_gpu:5366.242634854128 mfu:25.31348208401157 +Step 9672 | loss:1.2801274061203003 lr:8.367507325011181e-05 tokens_per_second_per_gpu:4915.309867860903 mfu:23.18635528504098 +Step 9673 | loss:1.1734845638275146 lr:8.36718678495592e-05 tokens_per_second_per_gpu:5333.005814619151 mfu:25.1566983321769 +Step 9674 | loss:1.122204303741455 lr:8.36686621957562e-05 tokens_per_second_per_gpu:3241.3108101534317 mfu:15.289816116144044 +Step 9675 | loss:1.1309802532196045 lr:8.366545628872689e-05 tokens_per_second_per_gpu:4610.4954569773845 mfu:21.74849370220217 +Step 9676 | loss:1.0943334102630615 lr:8.366225012849536e-05 tokens_per_second_per_gpu:4229.010636911015 mfu:19.94896471792714 +Step 9677 | loss:1.1696175336837769 lr:8.365904371508577e-05 tokens_per_second_per_gpu:5380.54392586788 mfu:25.380943713775295 +Step 9678 | loss:1.1337604522705078 lr:8.365583704852222e-05 tokens_per_second_per_gpu:3406.5832295401083 mfu:16.06943431677347 +Step 9679 | loss:1.0536693334579468 lr:8.365263012882883e-05 tokens_per_second_per_gpu:4511.793645855539 mfu:21.28290041887493 +Step 9680 | loss:1.0715185403823853 lr:8.36494229560297e-05 tokens_per_second_per_gpu:4112.992918662329 mfu:19.40168934628419 +Step 9681 | loss:1.1079903841018677 lr:8.364621553014898e-05 tokens_per_second_per_gpu:3844.345100226544 mfu:18.134431750678303 +Step 9682 | loss:1.091273307800293 lr:8.364300785121077e-05 tokens_per_second_per_gpu:4305.716311940454 mfu:20.310798474378682 +Step 9683 | loss:1.0474975109100342 lr:8.363979991923921e-05 tokens_per_second_per_gpu:4670.364092188492 mfu:22.03090426913543 +Step 9684 | loss:1.0883091688156128 lr:8.363659173425842e-05 tokens_per_second_per_gpu:3659.8592270690233 mfu:17.26418041045856 +Step 9685 | loss:1.1888442039489746 lr:8.363338329629254e-05 tokens_per_second_per_gpu:3792.084183762113 mfu:17.887908091083776 +Step 9686 | loss:1.083554983139038 lr:8.363017460536567e-05 tokens_per_second_per_gpu:4557.874611509703 mfu:21.500272196089092 +Step 9687 | loss:1.1094002723693848 lr:8.362696566150199e-05 tokens_per_second_per_gpu:4781.819965398177 mfu:22.556660639398388 +Step 9688 | loss:1.1626883745193481 lr:8.36237564647256e-05 tokens_per_second_per_gpu:5229.941922221221 mfu:24.670528367222563 +Step 9689 | loss:1.2557423114776611 lr:8.362054701506065e-05 tokens_per_second_per_gpu:4664.843164559095 mfu:22.004861111540176 +Step 9690 | loss:0.9807554483413696 lr:8.361733731253127e-05 tokens_per_second_per_gpu:5314.351303318625 mfu:25.06870181958449 +Step 9691 | loss:1.0143685340881348 lr:8.36141273571616e-05 tokens_per_second_per_gpu:5329.645366528969 mfu:25.140846525183946 +Step 9692 | loss:1.1922264099121094 lr:8.361091714897579e-05 tokens_per_second_per_gpu:3806.1420419915944 mfu:17.954221406870182 +Step 9693 | loss:1.1176823377609253 lr:8.3607706687998e-05 tokens_per_second_per_gpu:3230.251784619739 mfu:15.237648805837292 +Step 9694 | loss:1.1135833263397217 lr:8.360449597425236e-05 tokens_per_second_per_gpu:4517.258923714175 mfu:21.308681067006674 +Step 9695 | loss:1.0847079753875732 lr:8.360128500776301e-05 tokens_per_second_per_gpu:4421.893229886372 mfu:20.858824818155743 +Step 9696 | loss:0.9903737902641296 lr:8.35980737885541e-05 tokens_per_second_per_gpu:4694.981690522363 mfu:22.14702968067169 +Step 9697 | loss:1.2699722051620483 lr:8.359486231664981e-05 tokens_per_second_per_gpu:3186.8186014078838 mfu:15.032767070161773 +Step 9698 | loss:0.953689455986023 lr:8.359165059207425e-05 tokens_per_second_per_gpu:5353.380909324058 mfu:25.252811130249572 +Step 9699 | loss:1.0863333940505981 lr:8.358843861485162e-05 tokens_per_second_per_gpu:3454.1516931248866 mfu:16.293822875519655 +Step 9700 | loss:1.086969017982483 lr:8.358522638500604e-05 tokens_per_second_per_gpu:4986.101968817802 mfu:23.520293703632582 +Step 9701 | loss:1.0782588720321655 lr:8.358201390256169e-05 tokens_per_second_per_gpu:4496.212984378409 mfu:21.20940378035234 +Step 9702 | loss:1.0128462314605713 lr:8.357880116754274e-05 tokens_per_second_per_gpu:4516.167652866295 mfu:21.303533356227216 +Step 9703 | loss:1.141774296760559 lr:8.357558817997334e-05 tokens_per_second_per_gpu:3736.583996672612 mfu:17.62610423927424 +Step 9704 | loss:1.0670104026794434 lr:8.357237493987765e-05 tokens_per_second_per_gpu:3776.2445465289597 mfu:17.81318981973421 +Step 9705 | loss:0.93277907371521 lr:8.356916144727985e-05 tokens_per_second_per_gpu:4235.1770171448225 mfu:19.97805258558296 +Step 9706 | loss:1.2830077409744263 lr:8.35659477022041e-05 tokens_per_second_per_gpu:3711.971135518603 mfu:17.510001173823643 +Step 9707 | loss:1.117841124534607 lr:8.356273370467457e-05 tokens_per_second_per_gpu:3495.8641117760358 mfu:16.490587471169615 +Step 9708 | loss:1.0038702487945557 lr:8.355951945471544e-05 tokens_per_second_per_gpu:4778.462952786379 mfu:22.540825038143122 +Step 9709 | loss:1.1916841268539429 lr:8.355630495235087e-05 tokens_per_second_per_gpu:4315.6096102851325 mfu:20.357466850641288 +Step 9710 | loss:1.0823098421096802 lr:8.355309019760507e-05 tokens_per_second_per_gpu:4796.61842631163 mfu:22.626467504404935 +Step 9711 | loss:1.0829254388809204 lr:8.354987519050219e-05 tokens_per_second_per_gpu:4120.416788990716 mfu:19.436709009266078 +Step 9712 | loss:1.1069186925888062 lr:8.354665993106642e-05 tokens_per_second_per_gpu:3632.677851806659 mfu:17.13596122572515 +Step 9713 | loss:1.11199152469635 lr:8.354344441932194e-05 tokens_per_second_per_gpu:4922.1400953419825 mfu:23.21857463342621 +Step 9714 | loss:1.1356388330459595 lr:8.354022865529293e-05 tokens_per_second_per_gpu:3918.9745093229994 mfu:18.48647140647647 +Step 9715 | loss:1.0124462842941284 lr:8.353701263900357e-05 tokens_per_second_per_gpu:3900.85975664446 mfu:18.40102102739629 +Step 9716 | loss:1.083660364151001 lr:8.353379637047807e-05 tokens_per_second_per_gpu:3977.7136228961353 mfu:18.763553827127154 +Step 9717 | loss:1.2425580024719238 lr:8.353057984974062e-05 tokens_per_second_per_gpu:3939.1079088548177 mfu:18.581444087180234 +Step 9718 | loss:1.2962323427200317 lr:8.352736307681539e-05 tokens_per_second_per_gpu:5349.070774136696 mfu:25.23247948718589 +Step 9719 | loss:1.1559011936187744 lr:8.352414605172658e-05 tokens_per_second_per_gpu:5091.523818779491 mfu:24.017586556724215 +Step 9720 | loss:1.0205323696136475 lr:8.352092877449839e-05 tokens_per_second_per_gpu:4475.8699351053265 mfu:21.1134421460491 +Step 9721 | loss:1.0475778579711914 lr:8.351771124515502e-05 tokens_per_second_per_gpu:5308.15314903157 mfu:25.039464068298287 +Step 9722 | loss:1.1044847965240479 lr:8.351449346372066e-05 tokens_per_second_per_gpu:4293.328307523135 mfu:20.252362143930686 +Step 9723 | loss:1.0749324560165405 lr:8.351127543021953e-05 tokens_per_second_per_gpu:4486.8626211990395 mfu:21.165296521898792 +Step 9724 | loss:1.0232794284820557 lr:8.350805714467582e-05 tokens_per_second_per_gpu:4431.096085747492 mfu:20.902236259421763 +Step 9725 | loss:0.9365602731704712 lr:8.350483860711373e-05 tokens_per_second_per_gpu:4159.146166215214 mfu:19.61940208954718 +Step 9726 | loss:0.9677881002426147 lr:8.350161981755748e-05 tokens_per_second_per_gpu:5355.168632984019 mfu:25.261244127770716 +Step 9727 | loss:0.9361383318901062 lr:8.349840077603126e-05 tokens_per_second_per_gpu:5360.405596862543 mfu:25.285947779903886 +Step 9728 | loss:1.052431344985962 lr:8.349518148255931e-05 tokens_per_second_per_gpu:5300.558060158553 mfu:25.003636738228483 +Step 9729 | loss:1.1816123723983765 lr:8.349196193716582e-05 tokens_per_second_per_gpu:5379.734546765009 mfu:25.37712573445274 +Step 9730 | loss:1.118389368057251 lr:8.348874213987501e-05 tokens_per_second_per_gpu:5013.547877683746 mfu:23.64976073850816 +Step 9731 | loss:0.9929024577140808 lr:8.34855220907111e-05 tokens_per_second_per_gpu:4673.153691487582 mfu:22.04406328498396 +Step 9732 | loss:1.0755010843276978 lr:8.348230178969832e-05 tokens_per_second_per_gpu:5363.82882312498 mfu:25.30209572224643 +Step 9733 | loss:1.0941336154937744 lr:8.347908123686083e-05 tokens_per_second_per_gpu:3411.169066285559 mfu:16.091066491126192 +Step 9734 | loss:1.1807796955108643 lr:8.347586043222293e-05 tokens_per_second_per_gpu:5119.574366034375 mfu:24.14990577404237 +Step 9735 | loss:1.1347569227218628 lr:8.347263937580882e-05 tokens_per_second_per_gpu:5236.146501896133 mfu:24.699796428159576 +Step 9736 | loss:1.20503830909729 lr:8.34694180676427e-05 tokens_per_second_per_gpu:4596.991893272105 mfu:21.68479508825882 +Step 9737 | loss:0.9527069926261902 lr:8.346619650774882e-05 tokens_per_second_per_gpu:5349.043637269418 mfu:25.232351477952914 +Step 9738 | loss:1.142568826675415 lr:8.346297469615141e-05 tokens_per_second_per_gpu:3769.75644243993 mfu:17.782584325761547 +Step 9739 | loss:0.9561999440193176 lr:8.345975263287471e-05 tokens_per_second_per_gpu:4738.229274273286 mfu:22.351035912023455 +Step 9740 | loss:1.1924139261245728 lr:8.345653031794292e-05 tokens_per_second_per_gpu:5340.793443968748 mfu:25.193433908525527 +Step 9741 | loss:1.1943491697311401 lr:8.345330775138029e-05 tokens_per_second_per_gpu:4646.134187943811 mfu:21.91660767676532 +Step 9742 | loss:0.9622727036476135 lr:8.345008493321108e-05 tokens_per_second_per_gpu:4387.835985674384 mfu:20.698170986442744 +Step 9743 | loss:1.0174672603607178 lr:8.344686186345951e-05 tokens_per_second_per_gpu:5349.196046842407 mfu:25.23307041991246 +Step 9744 | loss:1.0634868144989014 lr:8.344363854214983e-05 tokens_per_second_per_gpu:4161.1895579191205 mfu:19.62904111685308 +Step 9745 | loss:1.1330686807632446 lr:8.344041496930627e-05 tokens_per_second_per_gpu:3116.467551634607 mfu:14.70090916525441 +Step 9746 | loss:1.0230857133865356 lr:8.343719114495309e-05 tokens_per_second_per_gpu:5137.8845660866555 mfu:24.23627811173125 +Step 9747 | loss:1.1700156927108765 lr:8.343396706911451e-05 tokens_per_second_per_gpu:5314.462894755682 mfu:25.069228215432553 +Step 9748 | loss:1.0946791172027588 lr:8.343074274181482e-05 tokens_per_second_per_gpu:4759.372856049201 mfu:22.450773794727485 +Step 9749 | loss:1.1969223022460938 lr:8.342751816307824e-05 tokens_per_second_per_gpu:3932.979946639474 mfu:18.552537443872293 +Step 9750 | loss:1.2239892482757568 lr:8.342429333292903e-05 tokens_per_second_per_gpu:4079.412929443121 mfu:19.243286807800235 +Step 9751 | loss:1.072668194770813 lr:8.342106825139144e-05 tokens_per_second_per_gpu:5367.500824604867 mfu:25.31941717973512 +Step 9752 | loss:1.0273199081420898 lr:8.341784291848973e-05 tokens_per_second_per_gpu:4231.326574548864 mfu:19.95988939090458 +Step 9753 | loss:1.053540825843811 lr:8.341461733424816e-05 tokens_per_second_per_gpu:4929.42785147671 mfu:23.252952222533136 +Step 9754 | loss:1.0816537141799927 lr:8.341139149869099e-05 tokens_per_second_per_gpu:4077.176677454825 mfu:19.232738025627512 +Step 9755 | loss:1.1218197345733643 lr:8.340816541184249e-05 tokens_per_second_per_gpu:4438.295966188843 mfu:20.936199323889888 +Step 9756 | loss:1.2102065086364746 lr:8.34049390737269e-05 tokens_per_second_per_gpu:3333.5665745758265 mfu:15.725002297380994 +Step 9757 | loss:1.0427974462509155 lr:8.340171248436852e-05 tokens_per_second_per_gpu:4388.139796941824 mfu:20.69960411602681 +Step 9758 | loss:1.0119205713272095 lr:8.339848564379158e-05 tokens_per_second_per_gpu:3916.410297796727 mfu:18.474375583207443 +Step 9759 | loss:1.3184915781021118 lr:8.339525855202037e-05 tokens_per_second_per_gpu:5327.185851094891 mfu:25.129244571244797 +Step 9760 | loss:1.1267153024673462 lr:8.339203120907918e-05 tokens_per_second_per_gpu:4317.6027276383575 mfu:20.366868725257333 +Step 9761 | loss:1.086512804031372 lr:8.338880361499224e-05 tokens_per_second_per_gpu:4237.476684862552 mfu:19.988900510571263 +Step 9762 | loss:0.930506706237793 lr:8.338557576978385e-05 tokens_per_second_per_gpu:4467.629322828255 mfu:21.074569771944716 +Step 9763 | loss:1.2643697261810303 lr:8.338234767347829e-05 tokens_per_second_per_gpu:3819.6935827939024 mfu:18.018146336965952 +Step 9764 | loss:1.0422227382659912 lr:8.337911932609983e-05 tokens_per_second_per_gpu:4583.241099583235 mfu:21.619930248301092 +Step 9765 | loss:1.140384554862976 lr:8.337589072767275e-05 tokens_per_second_per_gpu:5208.2737085883555 mfu:24.568315706537874 +Step 9766 | loss:1.130216121673584 lr:8.337266187822135e-05 tokens_per_second_per_gpu:5084.408832506183 mfu:23.98402395253118 +Step 9767 | loss:1.1896610260009766 lr:8.33694327777699e-05 tokens_per_second_per_gpu:5221.881307341434 mfu:24.632505071551893 +Step 9768 | loss:1.1491138935089111 lr:8.33662034263427e-05 tokens_per_second_per_gpu:3594.9801916940464 mfu:16.95813493108991 +Step 9769 | loss:1.0910804271697998 lr:8.336297382396399e-05 tokens_per_second_per_gpu:4856.633505194173 mfu:22.909568871122225 +Step 9770 | loss:1.1767940521240234 lr:8.335974397065811e-05 tokens_per_second_per_gpu:5399.581298578768 mfu:25.470746248963277 +Step 9771 | loss:1.0478650331497192 lr:8.335651386644936e-05 tokens_per_second_per_gpu:4702.517023220262 mfu:22.182575130668003 +Step 9772 | loss:1.18239164352417 lr:8.3353283511362e-05 tokens_per_second_per_gpu:3864.847628831676 mfu:18.2311456762008 +Step 9773 | loss:1.2161166667938232 lr:8.335005290542033e-05 tokens_per_second_per_gpu:5324.6661249727695 mfu:25.11735859321714 +Step 9774 | loss:1.0736161470413208 lr:8.334682204864866e-05 tokens_per_second_per_gpu:5357.584256848142 mfu:25.272639037685718 +Step 9775 | loss:1.0695126056671143 lr:8.334359094107131e-05 tokens_per_second_per_gpu:4637.421865938983 mfu:21.875510167393124 +Step 9776 | loss:1.1691339015960693 lr:8.334035958271254e-05 tokens_per_second_per_gpu:5216.102916912166 mfu:24.60524741800197 +Step 9777 | loss:1.120139241218567 lr:8.333712797359665e-05 tokens_per_second_per_gpu:5354.98310108882 mfu:25.260368942166057 +Step 9778 | loss:0.9576810002326965 lr:8.3333896113748e-05 tokens_per_second_per_gpu:5354.601922183449 mfu:25.258570856233593 +Step 9779 | loss:1.088891625404358 lr:8.333066400319085e-05 tokens_per_second_per_gpu:5301.485075230514 mfu:25.008009626488 +Step 9780 | loss:1.1417317390441895 lr:8.332743164194952e-05 tokens_per_second_per_gpu:4713.4259611625 mfu:22.234034452198284 +Step 9781 | loss:1.0171904563903809 lr:8.332419903004832e-05 tokens_per_second_per_gpu:4656.294592287893 mfu:21.964536037600084 +Step 9782 | loss:1.0710395574569702 lr:8.332096616751158e-05 tokens_per_second_per_gpu:3765.9079689884647 mfu:17.764430419873708 +Step 9783 | loss:1.1398136615753174 lr:8.331773305436359e-05 tokens_per_second_per_gpu:3569.3072028987285 mfu:16.837031062678825 +Step 9784 | loss:1.0917268991470337 lr:8.331449969062869e-05 tokens_per_second_per_gpu:5308.165704419427 mfu:25.03952329420498 +Step 9785 | loss:1.2146334648132324 lr:8.331126607633118e-05 tokens_per_second_per_gpu:4707.37516608601 mfu:22.205491819450383 +Step 9786 | loss:0.9994463920593262 lr:8.330803221149539e-05 tokens_per_second_per_gpu:4720.704003386568 mfu:22.26836621913126 +Step 9787 | loss:1.1948057413101196 lr:8.330479809614564e-05 tokens_per_second_per_gpu:4761.133547213188 mfu:22.459079275352902 +Step 9788 | loss:1.1796064376831055 lr:8.330156373030624e-05 tokens_per_second_per_gpu:4222.956657772163 mfu:19.92040706541382 +Step 9789 | loss:1.0916086435317993 lr:8.329832911400156e-05 tokens_per_second_per_gpu:4254.59771047024 mfu:20.069663309510915 +Step 9790 | loss:1.0797291994094849 lr:8.329509424725587e-05 tokens_per_second_per_gpu:5343.130286066263 mfu:25.20445719140588 +Step 9791 | loss:1.0192233324050903 lr:8.329185913009354e-05 tokens_per_second_per_gpu:4822.751840966642 mfu:22.74974328015294 +Step 9792 | loss:1.2105655670166016 lr:8.328862376253887e-05 tokens_per_second_per_gpu:3408.638085128602 mfu:16.079127421178914 +Step 9793 | loss:1.1299744844436646 lr:8.328538814461623e-05 tokens_per_second_per_gpu:4169.437532720774 mfu:19.667948221242092 +Step 9794 | loss:1.0997346639633179 lr:8.328215227634994e-05 tokens_per_second_per_gpu:3404.3757127491876 mfu:16.059021083429496 +Step 9795 | loss:1.0802462100982666 lr:8.327891615776434e-05 tokens_per_second_per_gpu:4301.920782495469 mfu:20.292894314402588 +Step 9796 | loss:1.1417783498764038 lr:8.327567978888376e-05 tokens_per_second_per_gpu:5357.318158234377 mfu:25.27138380512348 +Step 9797 | loss:1.060365080833435 lr:8.327244316973254e-05 tokens_per_second_per_gpu:4483.7572109629145 mfu:21.150647771977372 +Step 9798 | loss:1.0572525262832642 lr:8.326920630033503e-05 tokens_per_second_per_gpu:3812.3418784588307 mfu:17.983467093287317 +Step 9799 | loss:1.035700798034668 lr:8.326596918071559e-05 tokens_per_second_per_gpu:5312.980098185981 mfu:25.06223361102209 +Step 9800 | loss:1.0762858390808105 lr:8.326273181089855e-05 tokens_per_second_per_gpu:5335.855857510706 mfu:25.170142470763825 +Step 9801 | loss:1.0964010953903198 lr:8.325949419090824e-05 tokens_per_second_per_gpu:3696.7792717433963 mfu:17.43833856039675 +Step 9802 | loss:1.0316957235336304 lr:8.325625632076903e-05 tokens_per_second_per_gpu:5379.309537995193 mfu:25.37512089557154 +Step 9803 | loss:1.2418854236602783 lr:8.325301820050528e-05 tokens_per_second_per_gpu:4483.453835055993 mfu:21.149216696063156 +Step 9804 | loss:1.105053186416626 lr:8.324977983014135e-05 tokens_per_second_per_gpu:5351.103558174423 mfu:25.242068476320572 +Step 9805 | loss:1.0293467044830322 lr:8.324654120970156e-05 tokens_per_second_per_gpu:4794.338487212367 mfu:22.615712642675987 +Step 9806 | loss:1.064000129699707 lr:8.324330233921031e-05 tokens_per_second_per_gpu:4606.427679918932 mfu:21.729305303783114 +Step 9807 | loss:1.1208068132400513 lr:8.324006321869194e-05 tokens_per_second_per_gpu:4196.699918154988 mfu:19.796549544777875 +Step 9808 | loss:1.1391358375549316 lr:8.323682384817082e-05 tokens_per_second_per_gpu:4356.888601878412 mfu:20.552186896908996 +Step 9809 | loss:1.3130183219909668 lr:8.32335842276713e-05 tokens_per_second_per_gpu:4045.074722086203 mfu:19.081307624995343 +Step 9810 | loss:1.1771388053894043 lr:8.323034435721772e-05 tokens_per_second_per_gpu:5369.1305020374575 mfu:25.327104646236286 +Step 9811 | loss:1.000411033630371 lr:8.322710423683453e-05 tokens_per_second_per_gpu:3717.4564004171334 mfu:17.535876104232706 +Step 9812 | loss:1.3915822505950928 lr:8.322386386654602e-05 tokens_per_second_per_gpu:4361.657025388768 mfu:20.57468036418429 +Step 9813 | loss:1.0239266157150269 lr:8.322062324637659e-05 tokens_per_second_per_gpu:5333.4593790647 mfu:25.15883787305263 +Step 9814 | loss:1.2651511430740356 lr:8.321738237635062e-05 tokens_per_second_per_gpu:5014.487412892381 mfu:23.65419268638853 +Step 9815 | loss:0.9861427545547485 lr:8.32141412564925e-05 tokens_per_second_per_gpu:5336.200118618179 mfu:25.171766409144777 +Step 9816 | loss:1.1119341850280762 lr:8.321089988682655e-05 tokens_per_second_per_gpu:5328.028934544899 mfu:25.13322153972335 +Step 9817 | loss:1.1776957511901855 lr:8.32076582673772e-05 tokens_per_second_per_gpu:4743.791121593891 mfu:22.377272094783752 +Step 9818 | loss:0.9318009614944458 lr:8.32044163981688e-05 tokens_per_second_per_gpu:5348.726354730448 mfu:25.230854802082614 +Step 9819 | loss:1.031597375869751 lr:8.320117427922578e-05 tokens_per_second_per_gpu:3938.694938141317 mfu:18.579496033864583 +Step 9820 | loss:1.0445159673690796 lr:8.319793191057247e-05 tokens_per_second_per_gpu:4693.594740950902 mfu:22.140487202905103 +Step 9821 | loss:1.1509134769439697 lr:8.319468929223328e-05 tokens_per_second_per_gpu:4979.981179306277 mfu:23.49142089519234 +Step 9822 | loss:1.066358208656311 lr:8.31914464242326e-05 tokens_per_second_per_gpu:5328.856421284458 mfu:25.137124935857646 +Step 9823 | loss:1.0762773752212524 lr:8.31882033065948e-05 tokens_per_second_per_gpu:4852.888066413934 mfu:22.891900997358082 +Step 9824 | loss:1.0222270488739014 lr:8.31849599393443e-05 tokens_per_second_per_gpu:5130.772901923458 mfu:24.20273117849842 +Step 9825 | loss:1.1648818254470825 lr:8.318171632250549e-05 tokens_per_second_per_gpu:5297.106722997814 mfu:24.987356191982435 +Step 9826 | loss:1.055385708808899 lr:8.317847245610275e-05 tokens_per_second_per_gpu:4762.786230431835 mfu:22.466875264072094 +Step 9827 | loss:1.046610951423645 lr:8.317522834016049e-05 tokens_per_second_per_gpu:4550.480046416395 mfu:21.46539077090106 +Step 9828 | loss:1.0499403476715088 lr:8.317198397470309e-05 tokens_per_second_per_gpu:4222.536631630792 mfu:19.918425730441122 +Step 9829 | loss:1.0286210775375366 lr:8.3168739359755e-05 tokens_per_second_per_gpu:4317.593954584022 mfu:20.3668273412632 +Step 9830 | loss:1.1035816669464111 lr:8.316549449534056e-05 tokens_per_second_per_gpu:4714.27987167391 mfu:22.23806249377276 +Step 9831 | loss:1.034514307975769 lr:8.31622493814842e-05 tokens_per_second_per_gpu:5026.951662471478 mfu:23.712988678273398 +Step 9832 | loss:1.1389704942703247 lr:8.315900401821035e-05 tokens_per_second_per_gpu:4899.471264040373 mfu:23.111641888475074 +Step 9833 | loss:1.2202554941177368 lr:8.315575840554338e-05 tokens_per_second_per_gpu:5302.090412155773 mfu:25.01086510404558 +Step 9834 | loss:1.0197314023971558 lr:8.315251254350774e-05 tokens_per_second_per_gpu:4735.373420607748 mfu:22.33756436302423 +Step 9835 | loss:1.199947714805603 lr:8.31492664321278e-05 tokens_per_second_per_gpu:3996.33502201249 mfu:18.851394143897306 +Step 9836 | loss:0.9932597875595093 lr:8.3146020071428e-05 tokens_per_second_per_gpu:3808.339023670104 mfu:17.96458494429144 +Step 9837 | loss:1.1378264427185059 lr:8.314277346143277e-05 tokens_per_second_per_gpu:5312.013323209836 mfu:25.057673168510796 +Step 9838 | loss:1.1896589994430542 lr:8.31395266021665e-05 tokens_per_second_per_gpu:4265.309447570188 mfu:20.12019240102218 +Step 9839 | loss:1.134747862815857 lr:8.313627949365363e-05 tokens_per_second_per_gpu:3781.4654992752094 mfu:17.83781794462466 +Step 9840 | loss:1.0407333374023438 lr:8.313303213591859e-05 tokens_per_second_per_gpu:4722.228496553689 mfu:22.275557513505927 +Step 9841 | loss:1.1233625411987305 lr:8.312978452898577e-05 tokens_per_second_per_gpu:4507.445290040352 mfu:21.26238848258869 +Step 9842 | loss:1.0116153955459595 lr:8.31265366728796e-05 tokens_per_second_per_gpu:4448.269766009956 mfu:20.983247439351807 +Step 9843 | loss:1.0846142768859863 lr:8.312328856762452e-05 tokens_per_second_per_gpu:4918.215882001792 mfu:23.200063449560695 +Step 9844 | loss:1.0196651220321655 lr:8.312004021324498e-05 tokens_per_second_per_gpu:5358.004295719286 mfu:25.27462043270693 +Step 9845 | loss:1.2089232206344604 lr:8.311679160976539e-05 tokens_per_second_per_gpu:4761.562535185303 mfu:22.461102884810266 +Step 9846 | loss:1.182889461517334 lr:8.311354275721018e-05 tokens_per_second_per_gpu:5315.452421589259 mfu:25.07389598233722 +Step 9847 | loss:0.9938019514083862 lr:8.311029365560379e-05 tokens_per_second_per_gpu:4682.644965443875 mfu:22.08883524361411 +Step 9848 | loss:1.181794285774231 lr:8.310704430497065e-05 tokens_per_second_per_gpu:4648.9226262623315 mfu:21.92976121607041 +Step 9849 | loss:1.141847848892212 lr:8.31037947053352e-05 tokens_per_second_per_gpu:4710.812598687228 mfu:22.221706775516807 +Step 9850 | loss:1.0571589469909668 lr:8.31005448567219e-05 tokens_per_second_per_gpu:4523.20180924154 mfu:21.33671467200008 +Step 9851 | loss:1.1492960453033447 lr:8.309729475915517e-05 tokens_per_second_per_gpu:5371.528303551381 mfu:25.33841548508459 +Step 9852 | loss:1.1519917249679565 lr:8.309404441265946e-05 tokens_per_second_per_gpu:3762.3982232741755 mfu:17.747874350515037 +Step 9853 | loss:0.9765157699584961 lr:8.309079381725923e-05 tokens_per_second_per_gpu:4889.209976390871 mfu:23.063237643876167 +Step 9854 | loss:1.0981000661849976 lr:8.308754297297892e-05 tokens_per_second_per_gpu:5361.828719543752 mfu:25.29266089240092 +Step 9855 | loss:1.1350375413894653 lr:8.308429187984297e-05 tokens_per_second_per_gpu:4607.780490820406 mfu:21.73568674361871 +Step 9856 | loss:1.0158255100250244 lr:8.308104053787584e-05 tokens_per_second_per_gpu:5346.776059799073 mfu:25.221654928136687 +Step 9857 | loss:0.9319481253623962 lr:8.307778894710199e-05 tokens_per_second_per_gpu:5194.234772358537 mfu:24.502091649052304 +Step 9858 | loss:1.045667290687561 lr:8.307453710754587e-05 tokens_per_second_per_gpu:4777.5893118491 mfu:22.536703924783463 +Step 9859 | loss:1.1703097820281982 lr:8.307128501923192e-05 tokens_per_second_per_gpu:5355.449386549791 mfu:25.26256849024196 +Step 9860 | loss:1.1255617141723633 lr:8.306803268218462e-05 tokens_per_second_per_gpu:5323.109977174533 mfu:25.110017978546665 +Step 9861 | loss:0.9468568563461304 lr:8.306478009642843e-05 tokens_per_second_per_gpu:5305.765332000784 mfu:25.028200327960473 +Step 9862 | loss:1.1332191228866577 lr:8.306152726198781e-05 tokens_per_second_per_gpu:4412.269838687738 mfu:20.813429639951938 +Step 9863 | loss:1.13796067237854 lr:8.305827417888723e-05 tokens_per_second_per_gpu:3364.368373598717 mfu:15.870299638700796 +Step 9864 | loss:1.0928250551223755 lr:8.305502084715114e-05 tokens_per_second_per_gpu:4213.959611575649 mfu:19.87796646345049 +Step 9865 | loss:1.199887990951538 lr:8.305176726680402e-05 tokens_per_second_per_gpu:5385.889819242191 mfu:25.406161204925386 +Step 9866 | loss:1.0898278951644897 lr:8.304851343787035e-05 tokens_per_second_per_gpu:5052.451142801257 mfu:23.833274077648937 +Step 9867 | loss:1.0440579652786255 lr:8.30452593603746e-05 tokens_per_second_per_gpu:4031.4774427269435 mfu:19.01716693832264 +Step 9868 | loss:1.3013455867767334 lr:8.304200503434123e-05 tokens_per_second_per_gpu:3982.1915414675373 mfu:18.78467693304015 +Step 9869 | loss:1.0032832622528076 lr:8.303875045979472e-05 tokens_per_second_per_gpu:4039.0304026000726 mfu:19.052795538712026 +Step 9870 | loss:1.0447152853012085 lr:8.303549563675955e-05 tokens_per_second_per_gpu:5320.339569736036 mfu:25.096949493979025 +Step 9871 | loss:1.1975760459899902 lr:8.30322405652602e-05 tokens_per_second_per_gpu:3420.1816255008875 mfu:16.13358027064606 +Step 9872 | loss:1.1769660711288452 lr:8.302898524532116e-05 tokens_per_second_per_gpu:5063.703797498906 mfu:23.88635476975861 +Step 9873 | loss:1.1683779954910278 lr:8.302572967696692e-05 tokens_per_second_per_gpu:5043.553764648086 mfu:23.791303626851178 +Step 9874 | loss:1.3868032693862915 lr:8.302247386022193e-05 tokens_per_second_per_gpu:5371.325775907817 mfu:25.33746012763512 +Step 9875 | loss:1.102521538734436 lr:8.301921779511071e-05 tokens_per_second_per_gpu:4787.388864849765 mfu:22.582930088263748 +Step 9876 | loss:1.052038311958313 lr:8.301596148165775e-05 tokens_per_second_per_gpu:5221.20469361238 mfu:24.629313369149738 +Step 9877 | loss:1.3505929708480835 lr:8.301270491988751e-05 tokens_per_second_per_gpu:4885.222553434668 mfu:23.04442828946725 +Step 9878 | loss:1.168483853340149 lr:8.300944810982452e-05 tokens_per_second_per_gpu:5365.957219756984 mfu:25.31213573230151 +Step 9879 | loss:0.9387643337249756 lr:8.300619105149325e-05 tokens_per_second_per_gpu:4942.685953552462 mfu:23.315492952089926 +Step 9880 | loss:1.111232042312622 lr:8.300293374491821e-05 tokens_per_second_per_gpu:4423.343441368264 mfu:20.865665713147422 +Step 9881 | loss:1.3295267820358276 lr:8.29996761901239e-05 tokens_per_second_per_gpu:5371.607555374369 mfu:25.33878932945538 +Step 9882 | loss:1.0449321269989014 lr:8.299641838713481e-05 tokens_per_second_per_gpu:5298.632554966424 mfu:24.994553801712147 +Step 9883 | loss:1.2806577682495117 lr:8.299316033597544e-05 tokens_per_second_per_gpu:3540.763209232433 mfu:16.702384174447403 +Step 9884 | loss:1.1377402544021606 lr:8.298990203667031e-05 tokens_per_second_per_gpu:5379.918452967378 mfu:25.377993251386947 +Step 9885 | loss:1.1313587427139282 lr:8.298664348924391e-05 tokens_per_second_per_gpu:5216.19255262015 mfu:24.60567024492996 +Step 9886 | loss:1.2046421766281128 lr:8.298338469372075e-05 tokens_per_second_per_gpu:3438.304411007605 mfu:16.219068541947298 +Step 9887 | loss:1.0884965658187866 lr:8.298012565012536e-05 tokens_per_second_per_gpu:5323.698222943208 mfu:25.112792834202583 +Step 9888 | loss:1.0274254083633423 lr:8.297686635848223e-05 tokens_per_second_per_gpu:5372.147299189829 mfu:25.341335393122694 +Step 9889 | loss:1.0405954122543335 lr:8.297360681881589e-05 tokens_per_second_per_gpu:4224.254537078152 mfu:19.92652938354121 +Step 9890 | loss:0.9634199738502502 lr:8.297034703115083e-05 tokens_per_second_per_gpu:3322.9063399363963 mfu:15.674716151763224 +Step 9891 | loss:1.0085607767105103 lr:8.296708699551159e-05 tokens_per_second_per_gpu:5338.305916879533 mfu:25.181699818829813 +Step 9892 | loss:1.2517625093460083 lr:8.296382671192269e-05 tokens_per_second_per_gpu:3162.041696653951 mfu:14.915890183061608 +Step 9893 | loss:0.9667710661888123 lr:8.296056618040863e-05 tokens_per_second_per_gpu:5343.215386909567 mfu:25.20485862660345 +Step 9894 | loss:1.200568675994873 lr:8.295730540099396e-05 tokens_per_second_per_gpu:3995.0039451299162 mfu:18.84511522713707 +Step 9895 | loss:1.1319626569747925 lr:8.295404437370318e-05 tokens_per_second_per_gpu:4095.7157405189814 mfu:19.320189949191462 +Step 9896 | loss:1.1050806045532227 lr:8.295078309856082e-05 tokens_per_second_per_gpu:4510.587255648021 mfu:21.277209670434672 +Step 9897 | loss:1.0386152267456055 lr:8.294752157559142e-05 tokens_per_second_per_gpu:5321.34147686301 mfu:25.10167566083995 +Step 9898 | loss:1.1001466512680054 lr:8.294425980481951e-05 tokens_per_second_per_gpu:4491.710994816946 mfu:21.188167127472298 +Step 9899 | loss:1.2253319025039673 lr:8.294099778626963e-05 tokens_per_second_per_gpu:4624.902241094473 mfu:21.81645296093331 +Step 9900 | loss:1.1186091899871826 lr:8.293773551996629e-05 tokens_per_second_per_gpu:3959.3409654114066 mfu:18.676886867073083 +Step 9901 | loss:1.2532650232315063 lr:8.293447300593402e-05 tokens_per_second_per_gpu:3211.0223509273874 mfu:15.146940286230768 +Step 9902 | loss:1.1506673097610474 lr:8.293121024419741e-05 tokens_per_second_per_gpu:4993.646412886567 mfu:23.555882133519837 +Step 9903 | loss:1.236210584640503 lr:8.292794723478094e-05 tokens_per_second_per_gpu:3101.986998428254 mfu:14.632601925142904 +Step 9904 | loss:1.1480262279510498 lr:8.292468397770918e-05 tokens_per_second_per_gpu:4490.77959448581 mfu:21.183773553197042 +Step 9905 | loss:1.1827099323272705 lr:8.29214204730067e-05 tokens_per_second_per_gpu:4275.100237697992 mfu:20.16637722853608 +Step 9906 | loss:1.178645133972168 lr:8.291815672069798e-05 tokens_per_second_per_gpu:3494.868701645527 mfu:16.485891951749526 +Step 9907 | loss:1.108667254447937 lr:8.291489272080762e-05 tokens_per_second_per_gpu:3403.667580365832 mfu:16.055680702157364 +Step 9908 | loss:1.091365933418274 lr:8.291162847336017e-05 tokens_per_second_per_gpu:3587.4859301045562 mfu:16.922783220519406 +Step 9909 | loss:1.1098847389221191 lr:8.290836397838014e-05 tokens_per_second_per_gpu:4712.785130522404 mfu:22.231011544731967 +Step 9910 | loss:1.0655769109725952 lr:8.29050992358921e-05 tokens_per_second_per_gpu:4127.557768827527 mfu:19.470394229533113 +Step 9911 | loss:1.1419346332550049 lr:8.29018342459206e-05 tokens_per_second_per_gpu:5265.7450791459605 mfu:24.839417967085037 +Step 9912 | loss:1.3978426456451416 lr:8.289856900849022e-05 tokens_per_second_per_gpu:3434.076017177278 mfu:16.199122486811266 +Step 9913 | loss:1.135437250137329 lr:8.289530352362552e-05 tokens_per_second_per_gpu:3374.9141453103252 mfu:15.920045843158643 +Step 9914 | loss:1.0401601791381836 lr:8.289203779135104e-05 tokens_per_second_per_gpu:5351.473192216352 mfu:25.243812103162423 +Step 9915 | loss:1.0617866516113281 lr:8.288877181169134e-05 tokens_per_second_per_gpu:4476.69313419355 mfu:21.117325316600496 +Step 9916 | loss:0.9570844173431396 lr:8.288550558467099e-05 tokens_per_second_per_gpu:4992.524844452864 mfu:23.550591503858755 +Step 9917 | loss:1.1343666315078735 lr:8.288223911031455e-05 tokens_per_second_per_gpu:5348.562110929247 mfu:25.230080036049465 +Step 9918 | loss:1.1414780616760254 lr:8.28789723886466e-05 tokens_per_second_per_gpu:4148.46296086923 mfu:19.569007587187386 +Step 9919 | loss:1.0935308933258057 lr:8.287570541969169e-05 tokens_per_second_per_gpu:5329.790740133432 mfu:25.141532277279495 +Step 9920 | loss:0.8059331774711609 lr:8.287243820347443e-05 tokens_per_second_per_gpu:5278.339524397416 mfu:24.898828114170094 +Step 9921 | loss:0.909258246421814 lr:8.286917074001937e-05 tokens_per_second_per_gpu:5321.829961474681 mfu:25.103979926097267 +Step 9922 | loss:0.9856649041175842 lr:8.286590302935106e-05 tokens_per_second_per_gpu:3709.667049115359 mfu:17.49913240514186 +Step 9923 | loss:0.9539598822593689 lr:8.286263507149411e-05 tokens_per_second_per_gpu:4984.737159726868 mfu:23.513855666290205 +Step 9924 | loss:1.0954465866088867 lr:8.285936686647309e-05 tokens_per_second_per_gpu:3963.2017571406427 mfu:18.695098880378882 +Step 9925 | loss:1.2068181037902832 lr:8.285609841431257e-05 tokens_per_second_per_gpu:4001.0908340308674 mfu:18.873828120613194 +Step 9926 | loss:1.0062392950057983 lr:8.285282971503714e-05 tokens_per_second_per_gpu:4530.651951397098 mfu:21.371858263673175 +Step 9927 | loss:1.2424001693725586 lr:8.284956076867141e-05 tokens_per_second_per_gpu:4072.0468396536407 mfu:19.208539705478742 +Step 9928 | loss:1.0701040029525757 lr:8.28462915752399e-05 tokens_per_second_per_gpu:5359.095015081725 mfu:25.279765542035463 +Step 9929 | loss:1.0511292219161987 lr:8.284302213476728e-05 tokens_per_second_per_gpu:5335.011641066986 mfu:25.16616015776062 +Step 9930 | loss:0.9726951122283936 lr:8.283975244727808e-05 tokens_per_second_per_gpu:4309.583736015969 mfu:20.329041773593477 +Step 9931 | loss:1.2427582740783691 lr:8.283648251279692e-05 tokens_per_second_per_gpu:4368.677607155264 mfu:20.607797646211615 +Step 9932 | loss:1.1377975940704346 lr:8.283321233134837e-05 tokens_per_second_per_gpu:5344.917641797339 mfu:25.2128884533437 +Step 9933 | loss:1.2016584873199463 lr:8.282994190295705e-05 tokens_per_second_per_gpu:5081.249557690865 mfu:23.969121114199464 +Step 9934 | loss:1.1139168739318848 lr:8.282667122764756e-05 tokens_per_second_per_gpu:5335.563027272049 mfu:25.16876114056609 +Step 9935 | loss:1.144782543182373 lr:8.282340030544448e-05 tokens_per_second_per_gpu:3383.9444494122154 mfu:15.962643328336014 +Step 9936 | loss:0.9851895570755005 lr:8.282012913637241e-05 tokens_per_second_per_gpu:4158.049891946957 mfu:19.614230776780133 +Step 9937 | loss:1.0874452590942383 lr:8.281685772045595e-05 tokens_per_second_per_gpu:4030.3425040394777 mfu:19.01181323889335 +Step 9938 | loss:1.154716968536377 lr:8.281358605771974e-05 tokens_per_second_per_gpu:4868.456435109429 mfu:22.965339648732332 +Step 9939 | loss:0.9469245672225952 lr:8.281031414818837e-05 tokens_per_second_per_gpu:4739.780037009945 mfu:22.35835112443968 +Step 9940 | loss:1.216404676437378 lr:8.28070419918864e-05 tokens_per_second_per_gpu:5332.625273180891 mfu:25.154903253285564 +Step 9941 | loss:1.0369460582733154 lr:8.280376958883851e-05 tokens_per_second_per_gpu:4939.796654055483 mfu:23.301863633397346 +Step 9942 | loss:1.268999695777893 lr:8.280049693906929e-05 tokens_per_second_per_gpu:5346.181292904562 mfu:25.218849311217582 +Step 9943 | loss:1.0740469694137573 lr:8.279722404260333e-05 tokens_per_second_per_gpu:4684.761485861147 mfu:22.098819231538002 +Step 9944 | loss:1.1352177858352661 lr:8.279395089946527e-05 tokens_per_second_per_gpu:5345.420986309354 mfu:25.215262815286305 +Step 9945 | loss:1.2189348936080933 lr:8.279067750967972e-05 tokens_per_second_per_gpu:5334.77154783287 mfu:25.165027597011168 +Step 9946 | loss:1.0781548023223877 lr:8.278740387327128e-05 tokens_per_second_per_gpu:5349.941096261855 mfu:25.236584945142535 +Step 9947 | loss:1.1684273481369019 lr:8.278412999026462e-05 tokens_per_second_per_gpu:3679.4713165833778 mfu:17.356693982865288 +Step 9948 | loss:0.9814004898071289 lr:8.278085586068432e-05 tokens_per_second_per_gpu:5280.849492839022 mfu:24.91066806355392 +Step 9949 | loss:0.968607485294342 lr:8.277758148455503e-05 tokens_per_second_per_gpu:5102.221168004665 mfu:24.06804777817534 +Step 9950 | loss:1.194861650466919 lr:8.277430686190136e-05 tokens_per_second_per_gpu:4800.17372419866 mfu:22.643238451134422 +Step 9951 | loss:1.2117412090301514 lr:8.277103199274797e-05 tokens_per_second_per_gpu:5347.451999571046 mfu:25.224843451368304 +Step 9952 | loss:1.0464725494384766 lr:8.276775687711944e-05 tokens_per_second_per_gpu:5081.395028204789 mfu:23.96980732343347 +Step 9953 | loss:0.9730833172798157 lr:8.276448151504043e-05 tokens_per_second_per_gpu:4613.373817721649 mfu:21.762071420931846 +Step 9954 | loss:1.1655904054641724 lr:8.276120590653557e-05 tokens_per_second_per_gpu:5361.633470418081 mfu:25.291739868962193 +Step 9955 | loss:1.0932438373565674 lr:8.275793005162951e-05 tokens_per_second_per_gpu:4863.801799802608 mfu:22.943382939827384 +Step 9956 | loss:1.0939640998840332 lr:8.275465395034688e-05 tokens_per_second_per_gpu:3259.520271073192 mfu:15.375713250156954 +Step 9957 | loss:1.0639292001724243 lr:8.275137760271232e-05 tokens_per_second_per_gpu:4336.424930161134 mfu:20.455656265956176 +Step 9958 | loss:0.9608277082443237 lr:8.274810100875045e-05 tokens_per_second_per_gpu:4339.847570575356 mfu:20.471801444752728 +Step 9959 | loss:1.1514194011688232 lr:8.274482416848596e-05 tokens_per_second_per_gpu:3309.535518180318 mfu:15.61164370424237 +Step 9960 | loss:0.9712243676185608 lr:8.274154708194345e-05 tokens_per_second_per_gpu:5352.962989264673 mfu:25.25083973002494 +Step 9961 | loss:1.2290642261505127 lr:8.27382697491476e-05 tokens_per_second_per_gpu:4461.363692559318 mfu:21.045013724939384 +Step 9962 | loss:1.2074335813522339 lr:8.273499217012303e-05 tokens_per_second_per_gpu:4019.3054442990083 mfu:18.959749544980987 +Step 9963 | loss:1.050160527229309 lr:8.273171434489443e-05 tokens_per_second_per_gpu:4650.976731399761 mfu:21.939450780470143 +Step 9964 | loss:0.9733182787895203 lr:8.272843627348641e-05 tokens_per_second_per_gpu:5006.392396099766 mfu:23.616007111025482 +Step 9965 | loss:1.0099900960922241 lr:8.272515795592366e-05 tokens_per_second_per_gpu:5006.745469457663 mfu:23.617672618295202 +Step 9966 | loss:1.1097347736358643 lr:8.272187939223081e-05 tokens_per_second_per_gpu:5333.886727855302 mfu:25.160853750211118 +Step 9967 | loss:1.1691699028015137 lr:8.271860058243253e-05 tokens_per_second_per_gpu:4316.268807211894 mfu:20.360576395015418 +Step 9968 | loss:0.9890275001525879 lr:8.271532152655348e-05 tokens_per_second_per_gpu:4698.252369747512 mfu:22.16245803261213 +Step 9969 | loss:1.1464762687683105 lr:8.271204222461832e-05 tokens_per_second_per_gpu:5318.891960412506 mfu:25.090120873812314 +Step 9970 | loss:0.9112153053283691 lr:8.270876267665173e-05 tokens_per_second_per_gpu:5349.611085156862 mfu:25.23502822645458 +Step 9971 | loss:1.0539109706878662 lr:8.270548288267836e-05 tokens_per_second_per_gpu:4487.096643036002 mfu:21.16640044283199 +Step 9972 | loss:1.0813651084899902 lr:8.270220284272287e-05 tokens_per_second_per_gpu:5058.305691060735 mfu:23.860890980677826 +Step 9973 | loss:1.202327847480774 lr:8.269892255680993e-05 tokens_per_second_per_gpu:4171.314237233104 mfu:19.676800956625563 +Step 9974 | loss:1.1206871271133423 lr:8.269564202496425e-05 tokens_per_second_per_gpu:2732.14062990701 mfu:12.887973502530823 +Step 9975 | loss:1.0996575355529785 lr:8.269236124721045e-05 tokens_per_second_per_gpu:4544.481738262229 mfu:21.43709572792095 +Step 9976 | loss:1.2955923080444336 lr:8.268908022357323e-05 tokens_per_second_per_gpu:3790.710063953411 mfu:17.881426133497115 +Step 9977 | loss:1.0135037899017334 lr:8.268579895407729e-05 tokens_per_second_per_gpu:4031.3598257803365 mfu:19.01661211911797 +Step 9978 | loss:1.038508415222168 lr:8.268251743874726e-05 tokens_per_second_per_gpu:5057.116069084076 mfu:23.85527933084349 +Step 9979 | loss:0.9899863600730896 lr:8.267923567760785e-05 tokens_per_second_per_gpu:4557.728155124837 mfu:21.49958133633364 +Step 9980 | loss:1.096316933631897 lr:8.267595367068375e-05 tokens_per_second_per_gpu:5007.474530317721 mfu:23.621111722762706 +Step 9981 | loss:1.247899055480957 lr:8.267267141799963e-05 tokens_per_second_per_gpu:5009.856415203488 mfu:23.63234748016034 +Step 9982 | loss:1.0699007511138916 lr:8.266938891958018e-05 tokens_per_second_per_gpu:4034.7572297134907 mfu:19.032638253127544 +Step 9983 | loss:1.182121992111206 lr:8.266610617545008e-05 tokens_per_second_per_gpu:5345.425600713493 mfu:25.215284582217926 +Step 9984 | loss:1.0623652935028076 lr:8.266282318563403e-05 tokens_per_second_per_gpu:4782.8154292323225 mfu:22.561356412147692 +Step 9985 | loss:1.1756820678710938 lr:8.265953995015671e-05 tokens_per_second_per_gpu:3822.928044312494 mfu:18.03340384380519 +Step 9986 | loss:1.1785343885421753 lr:8.265625646904285e-05 tokens_per_second_per_gpu:4728.902965371376 mfu:22.307042121700764 +Step 9987 | loss:1.1730278730392456 lr:8.26529727423171e-05 tokens_per_second_per_gpu:5352.884296413186 mfu:25.250468522418092 +Step 9988 | loss:1.0476059913635254 lr:8.264968877000418e-05 tokens_per_second_per_gpu:5348.051411309348 mfu:25.22767097880806 +Step 9989 | loss:1.1287245750427246 lr:8.264640455212877e-05 tokens_per_second_per_gpu:4496.571175100531 mfu:21.21109342710223 +Step 9990 | loss:1.0186131000518799 lr:8.26431200887156e-05 tokens_per_second_per_gpu:5346.907403671166 mfu:25.222274499590927 +Step 9991 | loss:1.027701735496521 lr:8.263983537978934e-05 tokens_per_second_per_gpu:4900.775075090259 mfu:23.117792187650487 +Step 9992 | loss:0.8570379018783569 lr:8.263655042537475e-05 tokens_per_second_per_gpu:4680.131656361739 mfu:22.076979535858932 +Step 9993 | loss:1.0583916902542114 lr:8.263326522549647e-05 tokens_per_second_per_gpu:4480.1627655481625 mfu:21.133692159680354 +Step 9994 | loss:1.1387975215911865 lr:8.262997978017924e-05 tokens_per_second_per_gpu:4467.00593374692 mfu:21.071629139286916 +Step 9995 | loss:1.0116432905197144 lr:8.262669408944779e-05 tokens_per_second_per_gpu:5240.806543598682 mfu:24.7217786399556 +Step 9996 | loss:1.023169994354248 lr:8.262340815332678e-05 tokens_per_second_per_gpu:3683.158932640526 mfu:17.374089096979034 +Step 9997 | loss:1.1293559074401855 lr:8.262012197184098e-05 tokens_per_second_per_gpu:3702.255384665828 mfu:17.46417031937307 +Step 9998 | loss:1.1231322288513184 lr:8.261683554501508e-05 tokens_per_second_per_gpu:3944.1763708979743 mfu:18.6053528874066 +Step 9999 | loss:1.0917425155639648 lr:8.261354887287378e-05 tokens_per_second_per_gpu:5354.951123292142 mfu:25.260218097443154 +Step 10000 | loss:0.9817971587181091 lr:8.261026195544182e-05 tokens_per_second_per_gpu:4274.271525087008 mfu:20.162468049757116 +Step 10001 | loss:1.0634849071502686 lr:8.260697479274394e-05 tokens_per_second_per_gpu:4000.9776897530764 mfu:18.87329439975045 +Step 10002 | loss:1.1978626251220703 lr:8.26036873848048e-05 tokens_per_second_per_gpu:5363.938850076564 mfu:25.30261473814959 +Step 10003 | loss:1.1422752141952515 lr:8.26003997316492e-05 tokens_per_second_per_gpu:3850.8441304295534 mfu:18.165088784994794 +Step 10004 | loss:1.044836163520813 lr:8.259711183330184e-05 tokens_per_second_per_gpu:5282.911537300019 mfu:24.92039507909751 +Step 10005 | loss:1.0663577318191528 lr:8.259382368978742e-05 tokens_per_second_per_gpu:5342.935670985267 mfu:25.203539158864068 +Step 10006 | loss:1.1748754978179932 lr:8.25905353011307e-05 tokens_per_second_per_gpu:5315.513892484208 mfu:25.07418595103671 +Step 10007 | loss:0.9708527326583862 lr:8.258724666735641e-05 tokens_per_second_per_gpu:4493.341915626844 mfu:21.195860459196368 +Step 10008 | loss:1.0724033117294312 lr:8.258395778848927e-05 tokens_per_second_per_gpu:4735.967310162604 mfu:22.340365841382496 +Step 10009 | loss:1.0763461589813232 lr:8.258066866455404e-05 tokens_per_second_per_gpu:3951.540361308785 mfu:18.64009008660072 +Step 10010 | loss:1.003539800643921 lr:8.257737929557544e-05 tokens_per_second_per_gpu:4555.088022157094 mfu:21.487127378671666 +Step 10011 | loss:1.1739681959152222 lr:8.25740896815782e-05 tokens_per_second_per_gpu:5378.663072364365 mfu:25.372071406892516 +Step 10012 | loss:1.0814276933670044 lr:8.257079982258708e-05 tokens_per_second_per_gpu:5325.071295044896 mfu:25.119269849576504 +Step 10013 | loss:1.2739709615707397 lr:8.256750971862683e-05 tokens_per_second_per_gpu:4319.630581936118 mfu:20.37643445996749 +Step 10014 | loss:1.055070400238037 lr:8.256421936972218e-05 tokens_per_second_per_gpu:5081.326211748656 mfu:23.96948270446859 +Step 10015 | loss:1.1602126359939575 lr:8.256092877589789e-05 tokens_per_second_per_gpu:5321.080545049769 mfu:25.10044480095733 +Step 10016 | loss:1.1984087228775024 lr:8.255763793717868e-05 tokens_per_second_per_gpu:5248.730956798245 mfu:24.75915944906225 +Step 10017 | loss:1.067054271697998 lr:8.255434685358933e-05 tokens_per_second_per_gpu:4332.582074007967 mfu:20.437528857822816 +Step 10018 | loss:1.0785001516342163 lr:8.255105552515459e-05 tokens_per_second_per_gpu:3826.1545750766945 mfu:18.048623939922837 +Step 10019 | loss:1.123041033744812 lr:8.254776395189919e-05 tokens_per_second_per_gpu:5083.799418065295 mfu:23.981149240636693 +Step 10020 | loss:1.0049883127212524 lr:8.254447213384791e-05 tokens_per_second_per_gpu:4791.040307127603 mfu:22.600154564488605 +Step 10021 | loss:1.0461663007736206 lr:8.254118007102551e-05 tokens_per_second_per_gpu:4237.100812161753 mfu:19.987127454911104 +Step 10022 | loss:1.2135467529296875 lr:8.253788776345674e-05 tokens_per_second_per_gpu:4320.6741051400195 mfu:20.381356936963712 +Step 10023 | loss:1.0434280633926392 lr:8.253459521116637e-05 tokens_per_second_per_gpu:3570.932555914368 mfu:16.84469813016742 +Step 10024 | loss:1.1122959852218628 lr:8.253130241417915e-05 tokens_per_second_per_gpu:3303.4113487066543 mfu:15.58275495194404 +Step 10025 | loss:1.1141581535339355 lr:8.252800937251986e-05 tokens_per_second_per_gpu:5339.936509559706 mfu:25.18939160270999 +Step 10026 | loss:1.0237653255462646 lr:8.252471608621325e-05 tokens_per_second_per_gpu:5308.082203379338 mfu:25.039129405552135 +Step 10027 | loss:1.2219345569610596 lr:8.25214225552841e-05 tokens_per_second_per_gpu:3566.7222976078897 mfu:16.82483762339162 +Step 10028 | loss:1.0975626707077026 lr:8.25181287797572e-05 tokens_per_second_per_gpu:4324.627631158897 mfu:20.400006393735854 +Step 10029 | loss:1.1593263149261475 lr:8.251483475965729e-05 tokens_per_second_per_gpu:3978.860623910251 mfu:18.76896442660929 +Step 10030 | loss:1.0804805755615234 lr:8.251154049500915e-05 tokens_per_second_per_gpu:3679.615777302171 mfu:17.357375428723376 +Step 10031 | loss:1.0249212980270386 lr:8.250824598583758e-05 tokens_per_second_per_gpu:4433.216506011927 mfu:20.912238643590108 +Step 10032 | loss:1.0940508842468262 lr:8.250495123216734e-05 tokens_per_second_per_gpu:5331.80035685006 mfu:25.151011982207795 +Step 10033 | loss:1.0559147596359253 lr:8.250165623402322e-05 tokens_per_second_per_gpu:3516.503965301901 mfu:16.58794918177326 +Step 10034 | loss:1.052559494972229 lr:8.249836099142998e-05 tokens_per_second_per_gpu:3927.0243283815644 mfu:18.524443776416476 +Step 10035 | loss:1.0756120681762695 lr:8.249506550441243e-05 tokens_per_second_per_gpu:4944.2901961643365 mfu:23.323060438202152 +Step 10036 | loss:1.1779670715332031 lr:8.249176977299535e-05 tokens_per_second_per_gpu:4256.3264546223145 mfu:20.07781809063042 +Step 10037 | loss:1.1642274856567383 lr:8.248847379720351e-05 tokens_per_second_per_gpu:4612.754588313442 mfu:21.7591504101619 +Step 10038 | loss:1.1808513402938843 lr:8.248517757706173e-05 tokens_per_second_per_gpu:4086.421874132212 mfu:19.27634919574795 +Step 10039 | loss:1.0412770509719849 lr:8.248188111259478e-05 tokens_per_second_per_gpu:3697.858934005024 mfu:17.443431511494225 +Step 10040 | loss:1.1789138317108154 lr:8.247858440382745e-05 tokens_per_second_per_gpu:4069.6551724989213 mfu:19.197257803449435 +Step 10041 | loss:1.1574407815933228 lr:8.247528745078454e-05 tokens_per_second_per_gpu:3920.6666597816907 mfu:18.494453568900756 +Step 10042 | loss:1.0382683277130127 lr:8.247199025349086e-05 tokens_per_second_per_gpu:5320.8166406988685 mfu:25.099199919107406 +Step 10043 | loss:1.0216808319091797 lr:8.24686928119712e-05 tokens_per_second_per_gpu:4686.31809767229 mfu:22.106162035036526 +Step 10044 | loss:1.0854625701904297 lr:8.246539512625034e-05 tokens_per_second_per_gpu:5005.525092576813 mfu:23.611915892330472 +Step 10045 | loss:0.9972825050354004 lr:8.246209719635312e-05 tokens_per_second_per_gpu:4423.295284763234 mfu:20.86543855022513 +Step 10046 | loss:1.1944091320037842 lr:8.245879902230431e-05 tokens_per_second_per_gpu:4623.59370871097 mfu:21.810280390421678 +Step 10047 | loss:0.9749220013618469 lr:8.245550060412874e-05 tokens_per_second_per_gpu:4779.86025917076 mfu:22.5474163707633 +Step 10048 | loss:1.079808235168457 lr:8.245220194185121e-05 tokens_per_second_per_gpu:4163.181720105323 mfu:19.638438485783315 +Step 10049 | loss:0.9687970876693726 lr:8.244890303549653e-05 tokens_per_second_per_gpu:4405.7550766270815 mfu:20.782698395778795 +Step 10050 | loss:1.1622949838638306 lr:8.24456038850895e-05 tokens_per_second_per_gpu:4204.341630843064 mfu:19.83259680733816 +Step 10051 | loss:0.919556736946106 lr:8.244230449065495e-05 tokens_per_second_per_gpu:5157.881006313218 mfu:24.330604732024455 +Step 10052 | loss:1.0331039428710938 lr:8.243900485221768e-05 tokens_per_second_per_gpu:4592.083198485245 mfu:21.66163993743951 +Step 10053 | loss:1.2956013679504395 lr:8.243570496980252e-05 tokens_per_second_per_gpu:3847.1942823515974 mfu:18.147871828882778 +Step 10054 | loss:1.023177981376648 lr:8.243240484343429e-05 tokens_per_second_per_gpu:5258.434453555472 mfu:24.804932498853354 +Step 10055 | loss:1.046431541442871 lr:8.24291044731378e-05 tokens_per_second_per_gpu:5345.855511826455 mfu:25.21731254628817 +Step 10056 | loss:1.1300839185714722 lr:8.242580385893788e-05 tokens_per_second_per_gpu:3771.1811181732846 mfu:17.78930476427053 +Step 10057 | loss:1.1386181116104126 lr:8.242250300085935e-05 tokens_per_second_per_gpu:4685.311068889773 mfu:22.101411708452567 +Step 10058 | loss:0.9856400489807129 lr:8.241920189892705e-05 tokens_per_second_per_gpu:5349.687959004615 mfu:25.235390853510463 +Step 10059 | loss:1.4574775695800781 lr:8.241590055316577e-05 tokens_per_second_per_gpu:5015.204588334102 mfu:23.657575725309872 +Step 10060 | loss:1.10464346408844 lr:8.241259896360038e-05 tokens_per_second_per_gpu:4435.963237888704 mfu:20.925195446493646 +Step 10061 | loss:1.0605312585830688 lr:8.240929713025569e-05 tokens_per_second_per_gpu:3628.4650528597526 mfu:17.116088734314975 +Step 10062 | loss:1.3708268404006958 lr:8.240599505315655e-05 tokens_per_second_per_gpu:4360.68503890645 mfu:20.5700953381092 +Step 10063 | loss:1.1709166765213013 lr:8.240269273232777e-05 tokens_per_second_per_gpu:4627.648688539747 mfu:21.82940842212492 +Step 10064 | loss:1.2853926420211792 lr:8.239939016779423e-05 tokens_per_second_per_gpu:4129.1509769789145 mfu:19.47790966421255 +Step 10065 | loss:1.1029714345932007 lr:8.239608735958073e-05 tokens_per_second_per_gpu:4031.8098679836694 mfu:19.01873504497621 +Step 10066 | loss:1.211093783378601 lr:8.239278430771211e-05 tokens_per_second_per_gpu:4256.226566178857 mfu:20.077346899799903 +Step 10067 | loss:1.1395227909088135 lr:8.238948101221324e-05 tokens_per_second_per_gpu:4108.096328213622 mfu:19.378591293693155 +Step 10068 | loss:1.1799272298812866 lr:8.238617747310896e-05 tokens_per_second_per_gpu:5311.011367413701 mfu:25.05294677206914 +Step 10069 | loss:1.0743472576141357 lr:8.238287369042408e-05 tokens_per_second_per_gpu:5342.57266204129 mfu:25.201826783739616 +Step 10070 | loss:0.9767135977745056 lr:8.237956966418349e-05 tokens_per_second_per_gpu:4522.248713378866 mfu:21.332218756200927 +Step 10071 | loss:0.9517942667007446 lr:8.237626539441204e-05 tokens_per_second_per_gpu:4695.16489422869 mfu:22.147893883812205 +Step 10072 | loss:0.989060640335083 lr:8.237296088113455e-05 tokens_per_second_per_gpu:4715.190926894174 mfu:22.24236009668879 +Step 10073 | loss:0.9814720153808594 lr:8.236965612437588e-05 tokens_per_second_per_gpu:4511.908319583432 mfu:21.28344135441476 +Step 10074 | loss:1.0564093589782715 lr:8.23663511241609e-05 tokens_per_second_per_gpu:4915.5758041444205 mfu:23.187609751863747 +Step 10075 | loss:1.0994675159454346 lr:8.236304588051449e-05 tokens_per_second_per_gpu:5333.941438800836 mfu:25.161111831075857 +Step 10076 | loss:1.11008882522583 lr:8.235974039346145e-05 tokens_per_second_per_gpu:3829.4851420357927 mfu:18.06433479252222 +Step 10077 | loss:1.0168951749801636 lr:8.235643466302669e-05 tokens_per_second_per_gpu:4017.966063466136 mfu:18.95343145707039 +Step 10078 | loss:1.029680609703064 lr:8.235312868923507e-05 tokens_per_second_per_gpu:3758.9236666040697 mfu:17.73148427388165 +Step 10079 | loss:1.0641034841537476 lr:8.23498224721114e-05 tokens_per_second_per_gpu:5100.526936228284 mfu:24.060055797820404 +Step 10080 | loss:1.197681188583374 lr:8.234651601168062e-05 tokens_per_second_per_gpu:5250.900046169426 mfu:24.76939141371092 +Step 10081 | loss:1.0959120988845825 lr:8.234320930796755e-05 tokens_per_second_per_gpu:4705.509627229236 mfu:22.19669175436923 +Step 10082 | loss:1.1345043182373047 lr:8.23399023609971e-05 tokens_per_second_per_gpu:4987.3809275864005 mfu:23.526326770356928 +Step 10083 | loss:1.0471488237380981 lr:8.233659517079408e-05 tokens_per_second_per_gpu:4129.155300321712 mfu:19.477930058158165 +Step 10084 | loss:1.0553759336471558 lr:8.233328773738343e-05 tokens_per_second_per_gpu:3150.470214606406 mfu:14.861305527944918 +Step 10085 | loss:1.185871958732605 lr:8.232998006078997e-05 tokens_per_second_per_gpu:4036.386722281157 mfu:19.04032484758932 +Step 10086 | loss:1.0732040405273438 lr:8.232667214103864e-05 tokens_per_second_per_gpu:5348.835780850471 mfu:25.23137098376078 +Step 10087 | loss:1.0355111360549927 lr:8.232336397815428e-05 tokens_per_second_per_gpu:5325.970409820173 mfu:25.123511127376283 +Step 10088 | loss:1.2052379846572876 lr:8.232005557216175e-05 tokens_per_second_per_gpu:3423.7229344470443 mfu:16.150285229154694 +Step 10089 | loss:1.0573179721832275 lr:8.231674692308598e-05 tokens_per_second_per_gpu:4326.412109114049 mfu:20.408424080713637 +Step 10090 | loss:1.0359135866165161 lr:8.231343803095181e-05 tokens_per_second_per_gpu:3246.812297857479 mfu:15.315767572294627 +Step 10091 | loss:1.2919511795043945 lr:8.231012889578417e-05 tokens_per_second_per_gpu:4809.134887255362 mfu:22.685509786204957 +Step 10092 | loss:1.1222020387649536 lr:8.230681951760793e-05 tokens_per_second_per_gpu:4618.757077882073 mfu:21.787465177587457 +Step 10093 | loss:1.2025482654571533 lr:8.230350989644796e-05 tokens_per_second_per_gpu:4794.582457000043 mfu:22.616863489789996 +Step 10094 | loss:0.9379928708076477 lr:8.230020003232918e-05 tokens_per_second_per_gpu:5306.7656197500955 mfu:25.032918855940665 +Step 10095 | loss:1.0902148485183716 lr:8.229688992527647e-05 tokens_per_second_per_gpu:4178.844852643512 mfu:19.712324154371935 +Step 10096 | loss:0.9509448409080505 lr:8.229357957531472e-05 tokens_per_second_per_gpu:4320.227611088799 mfu:20.379250748344365 +Step 10097 | loss:1.1909152269363403 lr:8.229026898246885e-05 tokens_per_second_per_gpu:3439.557736097638 mfu:16.224980690236297 +Step 10098 | loss:1.1752049922943115 lr:8.228695814676374e-05 tokens_per_second_per_gpu:3284.906445056535 mfu:15.4954641641643 +Step 10099 | loss:0.9243073463439941 lr:8.228364706822432e-05 tokens_per_second_per_gpu:5311.737290795212 mfu:25.056371076515028 +Step 10100 | loss:1.0688281059265137 lr:8.228033574687544e-05 tokens_per_second_per_gpu:5342.305184212569 mfu:25.20056504518511 +Step 10101 | loss:1.2436299324035645 lr:8.227702418274206e-05 tokens_per_second_per_gpu:4751.423480746962 mfu:22.413275235123542 +Step 10102 | loss:1.162928819656372 lr:8.227371237584906e-05 tokens_per_second_per_gpu:5213.317194819744 mfu:24.592106691598815 +Step 10103 | loss:1.1522066593170166 lr:8.227040032622133e-05 tokens_per_second_per_gpu:5341.730067400667 mfu:25.19785212105833 +Step 10104 | loss:0.9554898738861084 lr:8.226708803388383e-05 tokens_per_second_per_gpu:4252.065454224302 mfu:20.05771822475136 +Step 10105 | loss:1.1631602048873901 lr:8.226377549886142e-05 tokens_per_second_per_gpu:3813.635002748443 mfu:17.989566981191174 +Step 10106 | loss:1.0783414840698242 lr:8.226046272117904e-05 tokens_per_second_per_gpu:5067.783802077756 mfu:23.90560084747761 +Step 10107 | loss:1.2318556308746338 lr:8.225714970086162e-05 tokens_per_second_per_gpu:3506.427321598097 mfu:16.54041593416972 +Step 10108 | loss:1.0025538206100464 lr:8.225383643793405e-05 tokens_per_second_per_gpu:4593.459219147778 mfu:21.66813086167801 +Step 10109 | loss:1.1550096273422241 lr:8.225052293242126e-05 tokens_per_second_per_gpu:4082.7643024926915 mfu:19.25909580627826 +Step 10110 | loss:1.1930872201919556 lr:8.224720918434819e-05 tokens_per_second_per_gpu:3969.808184233014 mfu:18.726262524146293 +Step 10111 | loss:1.162226676940918 lr:8.224389519373971e-05 tokens_per_second_per_gpu:5265.9624367228325 mfu:24.84044327986017 +Step 10112 | loss:1.0606745481491089 lr:8.224058096062082e-05 tokens_per_second_per_gpu:4596.646347602582 mfu:21.68316509037951 +Step 10113 | loss:1.0365885496139526 lr:8.223726648501638e-05 tokens_per_second_per_gpu:4744.666703149611 mfu:22.381402362371702 +Step 10114 | loss:0.950628936290741 lr:8.223395176695134e-05 tokens_per_second_per_gpu:5200.285247081711 mfu:24.53063277063895 +Step 10115 | loss:1.2961136102676392 lr:8.223063680645065e-05 tokens_per_second_per_gpu:4178.917614885595 mfu:19.71266738628236 +Step 10116 | loss:1.1479300260543823 lr:8.222732160353923e-05 tokens_per_second_per_gpu:5134.020071243307 mfu:24.218048630204365 +Step 10117 | loss:1.0660388469696045 lr:8.2224006158242e-05 tokens_per_second_per_gpu:3133.3424503904803 mfu:14.780510941840483 +Step 10118 | loss:1.156288981437683 lr:8.222069047058392e-05 tokens_per_second_per_gpu:4525.748276599026 mfu:21.348726792997475 +Step 10119 | loss:1.0400093793869019 lr:8.221737454058991e-05 tokens_per_second_per_gpu:4188.394293863593 mfu:19.757370497909736 +Step 10120 | loss:1.098285436630249 lr:8.22140583682849e-05 tokens_per_second_per_gpu:4716.5147433515785 mfu:22.248604764784748 +Step 10121 | loss:1.031777024269104 lr:8.221074195369386e-05 tokens_per_second_per_gpu:5203.545257948679 mfu:24.54601079811365 +Step 10122 | loss:1.1600526571273804 lr:8.220742529684174e-05 tokens_per_second_per_gpu:4919.525236031507 mfu:23.206239895917694 +Step 10123 | loss:1.0016157627105713 lr:8.220410839775343e-05 tokens_per_second_per_gpu:5217.023845248287 mfu:24.60959159409003 +Step 10124 | loss:1.0719807147979736 lr:8.220079125645393e-05 tokens_per_second_per_gpu:4610.622539718768 mfu:21.749093173175826 +Step 10125 | loss:0.9926401376724243 lr:8.219747387296817e-05 tokens_per_second_per_gpu:5260.5601289699625 mfu:24.814959672461132 +Step 10126 | loss:1.0349806547164917 lr:8.219415624732109e-05 tokens_per_second_per_gpu:3188.533217218556 mfu:15.040855205484336 +Step 10127 | loss:1.034813404083252 lr:8.219083837953767e-05 tokens_per_second_per_gpu:4826.279595760538 mfu:22.76638430141278 +Step 10128 | loss:1.1200908422470093 lr:8.218752026964284e-05 tokens_per_second_per_gpu:4775.396849197704 mfu:22.526361704384126 +Step 10129 | loss:1.1327852010726929 lr:8.218420191766156e-05 tokens_per_second_per_gpu:5373.165922383545 mfu:25.34614041252249 +Step 10130 | loss:1.050429105758667 lr:8.21808833236188e-05 tokens_per_second_per_gpu:5312.765426059275 mfu:25.06122096597358 +Step 10131 | loss:1.0214414596557617 lr:8.217756448753948e-05 tokens_per_second_per_gpu:4752.391702268024 mfu:22.41784250123402 +Step 10132 | loss:1.078414797782898 lr:8.217424540944862e-05 tokens_per_second_per_gpu:4475.377777228579 mfu:21.11112055337369 +Step 10133 | loss:1.0370959043502808 lr:8.217092608937116e-05 tokens_per_second_per_gpu:4115.690424666453 mfu:19.414413942347625 +Step 10134 | loss:1.0724279880523682 lr:8.216760652733204e-05 tokens_per_second_per_gpu:5311.528771870575 mfu:25.055387457170514 +Step 10135 | loss:1.0965640544891357 lr:8.216428672335625e-05 tokens_per_second_per_gpu:5328.304367486334 mfu:25.134520803901808 +Step 10136 | loss:1.0629397630691528 lr:8.216096667746876e-05 tokens_per_second_per_gpu:4646.588660199432 mfu:21.91875149993604 +Step 10137 | loss:1.0099400281906128 lr:8.215764638969454e-05 tokens_per_second_per_gpu:5329.320873497148 mfu:25.139315836188622 +Step 10138 | loss:1.2864775657653809 lr:8.215432586005855e-05 tokens_per_second_per_gpu:3330.957057416971 mfu:15.712692759713136 +Step 10139 | loss:1.0210940837860107 lr:8.215100508858577e-05 tokens_per_second_per_gpu:4716.702893848605 mfu:22.249492302781007 +Step 10140 | loss:1.09636390209198 lr:8.21476840753012e-05 tokens_per_second_per_gpu:5311.430535106439 mfu:25.054924057593166 +Step 10141 | loss:1.1827226877212524 lr:8.214436282022977e-05 tokens_per_second_per_gpu:4691.514236911545 mfu:22.130673110380865 +Step 10142 | loss:1.076317310333252 lr:8.21410413233965e-05 tokens_per_second_per_gpu:3638.240649556724 mfu:17.162201891823372 +Step 10143 | loss:1.2721925973892212 lr:8.213771958482634e-05 tokens_per_second_per_gpu:5349.076779017267 mfu:25.232507813232086 +Step 10144 | loss:0.9884395003318787 lr:8.213439760454431e-05 tokens_per_second_per_gpu:4395.656681543648 mfu:20.73506254320637 +Step 10145 | loss:1.096876621246338 lr:8.213107538257536e-05 tokens_per_second_per_gpu:4521.647738480016 mfu:21.329383855066165 +Step 10146 | loss:1.1101012229919434 lr:8.212775291894451e-05 tokens_per_second_per_gpu:4247.386692383707 mfu:20.035647706870183 +Step 10147 | loss:1.172032356262207 lr:8.212443021367672e-05 tokens_per_second_per_gpu:4433.736051985366 mfu:20.914689430590503 +Step 10148 | loss:1.026421308517456 lr:8.2121107266797e-05 tokens_per_second_per_gpu:5328.190233298345 mfu:25.133982413463766 +Step 10149 | loss:1.0070720911026 lr:8.211778407833033e-05 tokens_per_second_per_gpu:4839.680936521651 mfu:22.829600712287466 +Step 10150 | loss:1.0830763578414917 lr:8.21144606483017e-05 tokens_per_second_per_gpu:3964.0577714467954 mfu:18.699136845912108 +Step 10151 | loss:1.1344908475875854 lr:8.211113697673612e-05 tokens_per_second_per_gpu:4576.783601829107 mfu:21.589469129632146 +Step 10152 | loss:0.9330788254737854 lr:8.210781306365857e-05 tokens_per_second_per_gpu:3674.9113978388405 mfu:17.33518406814494 +Step 10153 | loss:1.2036645412445068 lr:8.210448890909406e-05 tokens_per_second_per_gpu:3492.46050566349 mfu:16.474532080421653 +Step 10154 | loss:1.0754930973052979 lr:8.210116451306762e-05 tokens_per_second_per_gpu:5324.235245823614 mfu:25.11532606275501 +Step 10155 | loss:1.101073980331421 lr:8.20978398756042e-05 tokens_per_second_per_gpu:4133.472218159161 mfu:19.498293696139537 +Step 10156 | loss:1.256866693496704 lr:8.209451499672885e-05 tokens_per_second_per_gpu:5324.892861755465 mfu:25.1184281493064 +Step 10157 | loss:1.0392545461654663 lr:8.209118987646653e-05 tokens_per_second_per_gpu:5329.778674881722 mfu:25.14147536342863 +Step 10158 | loss:0.9289472103118896 lr:8.20878645148423e-05 tokens_per_second_per_gpu:5244.322693600937 mfu:24.73836491943329 +Step 10159 | loss:1.1488635540008545 lr:8.208453891188113e-05 tokens_per_second_per_gpu:3604.3764863664273 mfu:17.002458856232714 +Step 10160 | loss:1.2792850732803345 lr:8.208121306760805e-05 tokens_per_second_per_gpu:5328.699075402642 mfu:25.136382708487147 +Step 10161 | loss:0.993627667427063 lr:8.207788698204808e-05 tokens_per_second_per_gpu:4249.61221574579 mfu:20.046145880282573 +Step 10162 | loss:1.113075613975525 lr:8.207456065522624e-05 tokens_per_second_per_gpu:4085.421716944151 mfu:19.271631283647455 +Step 10163 | loss:1.2431905269622803 lr:8.207123408716753e-05 tokens_per_second_per_gpu:4489.786388511097 mfu:21.17908843115591 +Step 10164 | loss:1.0519765615463257 lr:8.206790727789697e-05 tokens_per_second_per_gpu:5322.373785800107 mfu:25.106545238226296 +Step 10165 | loss:1.110178828239441 lr:8.206458022743959e-05 tokens_per_second_per_gpu:5090.483556125614 mfu:24.012679460298767 +Step 10166 | loss:0.9764069318771362 lr:8.206125293582041e-05 tokens_per_second_per_gpu:3893.0452678193965 mfu:18.364158750319188 +Step 10167 | loss:1.2539143562316895 lr:8.205792540306444e-05 tokens_per_second_per_gpu:5360.596695802876 mfu:25.286849226210283 +Step 10168 | loss:1.0339699983596802 lr:8.205459762919674e-05 tokens_per_second_per_gpu:4621.501479636188 mfu:21.800410988904844 +Step 10169 | loss:1.0179479122161865 lr:8.205126961424233e-05 tokens_per_second_per_gpu:4781.2139081794285 mfu:22.553801764093524 +Step 10170 | loss:1.1529873609542847 lr:8.204794135822621e-05 tokens_per_second_per_gpu:4422.034300454125 mfu:20.859490272092952 +Step 10171 | loss:1.2637118101119995 lr:8.204461286117345e-05 tokens_per_second_per_gpu:4189.167146787512 mfu:19.761016177014167 +Step 10172 | loss:1.187643051147461 lr:8.204128412310906e-05 tokens_per_second_per_gpu:4453.2847922276615 mfu:21.006904173672428 +Step 10173 | loss:1.0823606252670288 lr:8.203795514405808e-05 tokens_per_second_per_gpu:5072.038555360422 mfu:23.925671244647713 +Step 10174 | loss:1.0698498487472534 lr:8.203462592404557e-05 tokens_per_second_per_gpu:5333.5695105878995 mfu:25.15935738223794 +Step 10175 | loss:1.0648831129074097 lr:8.203129646309653e-05 tokens_per_second_per_gpu:4372.931786925671 mfu:20.627865338026538 +Step 10176 | loss:1.1779279708862305 lr:8.202796676123603e-05 tokens_per_second_per_gpu:3695.2041527305673 mfu:17.430908455270224 +Step 10177 | loss:1.1449449062347412 lr:8.202463681848911e-05 tokens_per_second_per_gpu:3890.0767114922673 mfu:18.350155563636182 +Step 10178 | loss:1.1251349449157715 lr:8.20213066348808e-05 tokens_per_second_per_gpu:4826.5238017780475 mfu:22.76753626286322 +Step 10179 | loss:1.102798342704773 lr:8.201797621043615e-05 tokens_per_second_per_gpu:3978.137403716225 mfu:18.765552873559983 +Step 10180 | loss:1.03270423412323 lr:8.201464554518024e-05 tokens_per_second_per_gpu:4597.727949781591 mfu:21.688267192398257 +Step 10181 | loss:1.0765883922576904 lr:8.201131463913807e-05 tokens_per_second_per_gpu:3249.64871294742 mfu:15.329147426216196 +Step 10182 | loss:1.0768240690231323 lr:8.200798349233473e-05 tokens_per_second_per_gpu:5320.85781171712 mfu:25.099394129825 +Step 10183 | loss:1.2962110042572021 lr:8.200465210479526e-05 tokens_per_second_per_gpu:4652.396641190541 mfu:21.946148737214845 +Step 10184 | loss:1.076809287071228 lr:8.200132047654475e-05 tokens_per_second_per_gpu:4141.217141032678 mfu:19.534827818754025 +Step 10185 | loss:1.1132750511169434 lr:8.199798860760818e-05 tokens_per_second_per_gpu:5280.767091934833 mfu:24.91027936442982 +Step 10186 | loss:1.2567812204360962 lr:8.199465649801066e-05 tokens_per_second_per_gpu:3618.151021914589 mfu:17.0674356905915 +Step 10187 | loss:1.0833344459533691 lr:8.199132414777726e-05 tokens_per_second_per_gpu:3758.522271066805 mfu:17.729590822647232 +Step 10188 | loss:1.1393451690673828 lr:8.198799155693304e-05 tokens_per_second_per_gpu:3888.3225070362396 mfu:18.341880681918955 +Step 10189 | loss:1.3469057083129883 lr:8.198465872550302e-05 tokens_per_second_per_gpu:3473.5984574819445 mfu:16.38555658095203 +Step 10190 | loss:1.167388677597046 lr:8.198132565351234e-05 tokens_per_second_per_gpu:3983.846598242262 mfu:18.79248411822833 +Step 10191 | loss:1.070906162261963 lr:8.1977992340986e-05 tokens_per_second_per_gpu:3173.0008375974194 mfu:14.967586320714203 +Step 10192 | loss:1.2146546840667725 lr:8.19746587879491e-05 tokens_per_second_per_gpu:5079.315574452299 mfu:23.95999818529158 +Step 10193 | loss:0.9756474494934082 lr:8.197132499442672e-05 tokens_per_second_per_gpu:4811.085536713693 mfu:22.694711332515027 +Step 10194 | loss:1.044700264930725 lr:8.196799096044392e-05 tokens_per_second_per_gpu:5062.641733331935 mfu:23.881344831876177 +Step 10195 | loss:1.0219610929489136 lr:8.196465668602579e-05 tokens_per_second_per_gpu:4887.526606177849 mfu:23.05529689936877 +Step 10196 | loss:1.184564232826233 lr:8.196132217119739e-05 tokens_per_second_per_gpu:5360.064146887448 mfu:25.284337102113994 +Step 10197 | loss:1.2091904878616333 lr:8.19579874159838e-05 tokens_per_second_per_gpu:4140.113006061267 mfu:19.529619425709026 +Step 10198 | loss:1.1040821075439453 lr:8.195465242041014e-05 tokens_per_second_per_gpu:3874.990676223083 mfu:18.278992161328404 +Step 10199 | loss:1.1809715032577515 lr:8.195131718450143e-05 tokens_per_second_per_gpu:4981.73372771276 mfu:23.49968796504131 +Step 10200 | loss:1.0856797695159912 lr:8.19479817082828e-05 tokens_per_second_per_gpu:4639.231553961459 mfu:21.88404677455976 +Step 10201 | loss:1.0980480909347534 lr:8.194464599177932e-05 tokens_per_second_per_gpu:4944.195722505211 mfu:23.32261478983319 +Step 10202 | loss:1.050115942955017 lr:8.194131003501609e-05 tokens_per_second_per_gpu:5133.001342094996 mfu:24.213243110998665 +Step 10203 | loss:1.2702956199645996 lr:8.193797383801817e-05 tokens_per_second_per_gpu:4359.592520617828 mfu:20.564941742938927 +Step 10204 | loss:0.9478901624679565 lr:8.193463740081069e-05 tokens_per_second_per_gpu:5309.196233433946 mfu:25.044384475392846 +Step 10205 | loss:1.188949465751648 lr:8.193130072341873e-05 tokens_per_second_per_gpu:5271.448647721377 mfu:24.86632267318442 +Step 10206 | loss:1.2245018482208252 lr:8.192796380586738e-05 tokens_per_second_per_gpu:4516.942570608177 mfu:21.307188775430276 +Step 10207 | loss:1.1582037210464478 lr:8.192462664818172e-05 tokens_per_second_per_gpu:3730.374998038192 mfu:17.596815333351277 +Step 10208 | loss:1.1302053928375244 lr:8.192128925038688e-05 tokens_per_second_per_gpu:4146.736588130828 mfu:19.56086399243078 +Step 10209 | loss:0.9754875898361206 lr:8.191795161250797e-05 tokens_per_second_per_gpu:4808.836063883124 mfu:22.68410018537359 +Step 10210 | loss:1.02322518825531 lr:8.191461373457007e-05 tokens_per_second_per_gpu:4562.838542200693 mfu:21.5236879040924 +Step 10211 | loss:1.1102700233459473 lr:8.191127561659827e-05 tokens_per_second_per_gpu:5365.720383616548 mfu:25.311018535818658 +Step 10212 | loss:1.067720890045166 lr:8.19079372586177e-05 tokens_per_second_per_gpu:5334.460797466735 mfu:25.16356173451463 +Step 10213 | loss:1.0721279382705688 lr:8.190459866065347e-05 tokens_per_second_per_gpu:5356.358856740687 mfu:25.26685861630403 +Step 10214 | loss:1.0917272567749023 lr:8.190125982273067e-05 tokens_per_second_per_gpu:5204.963201765226 mfu:24.552699481022458 +Step 10215 | loss:0.9185543656349182 lr:8.189792074487443e-05 tokens_per_second_per_gpu:5316.707699689175 mfu:25.079817343307052 +Step 10216 | loss:0.9713739156723022 lr:8.189458142710987e-05 tokens_per_second_per_gpu:4045.1293032138697 mfu:19.08156509348705 +Step 10217 | loss:1.056178331375122 lr:8.189124186946207e-05 tokens_per_second_per_gpu:5361.01386134747 mfu:25.288817067259917 +Step 10218 | loss:0.9914257526397705 lr:8.188790207195619e-05 tokens_per_second_per_gpu:5390.201915182585 mfu:25.426502097195655 +Step 10219 | loss:1.12070894241333 lr:8.188456203461732e-05 tokens_per_second_per_gpu:4018.210575239017 mfu:18.954584860821083 +Step 10220 | loss:1.2295938730239868 lr:8.188122175747059e-05 tokens_per_second_per_gpu:5363.577653591881 mfu:25.300910912704865 +Step 10221 | loss:1.1293948888778687 lr:8.187788124054113e-05 tokens_per_second_per_gpu:5017.500146950775 mfu:23.66840426696741 +Step 10222 | loss:0.9744194149971008 lr:8.187454048385405e-05 tokens_per_second_per_gpu:5356.732270610484 mfu:25.268620073238843 +Step 10223 | loss:1.1746641397476196 lr:8.18711994874345e-05 tokens_per_second_per_gpu:4342.663397810475 mfu:20.48508418225045 +Step 10224 | loss:1.0251277685165405 lr:8.186785825130757e-05 tokens_per_second_per_gpu:4669.578399253444 mfu:22.027198021507772 +Step 10225 | loss:1.1284457445144653 lr:8.186451677549843e-05 tokens_per_second_per_gpu:5306.043601853256 mfu:25.02951297433232 +Step 10226 | loss:1.1451705694198608 lr:8.186117506003217e-05 tokens_per_second_per_gpu:4845.339232355371 mfu:22.85629185913994 +Step 10227 | loss:1.1443620920181274 lr:8.185783310493398e-05 tokens_per_second_per_gpu:5325.252240478108 mfu:25.12012340005744 +Step 10228 | loss:1.0775245428085327 lr:8.185449091022894e-05 tokens_per_second_per_gpu:4981.017235975558 mfu:23.496308151270835 +Step 10229 | loss:1.0584399700164795 lr:8.185114847594219e-05 tokens_per_second_per_gpu:5311.199286208933 mfu:25.053833217088503 +Step 10230 | loss:1.0010826587677002 lr:8.184780580209892e-05 tokens_per_second_per_gpu:5337.923677976459 mfu:25.179896732707334 +Step 10231 | loss:1.2435590028762817 lr:8.184446288872423e-05 tokens_per_second_per_gpu:4330.273125431264 mfu:20.426637153439255 +Step 10232 | loss:1.149596095085144 lr:8.184111973584327e-05 tokens_per_second_per_gpu:4515.336880136729 mfu:21.29961445951741 +Step 10233 | loss:0.9413418769836426 lr:8.183777634348118e-05 tokens_per_second_per_gpu:4169.571322237636 mfu:19.668579329219977 +Step 10234 | loss:1.2661381959915161 lr:8.183443271166313e-05 tokens_per_second_per_gpu:3845.201979600737 mfu:18.13847379688507 +Step 10235 | loss:1.0554039478302002 lr:8.183108884041424e-05 tokens_per_second_per_gpu:3946.3260415740942 mfu:18.615493250757496 +Step 10236 | loss:1.2067497968673706 lr:8.182774472975966e-05 tokens_per_second_per_gpu:3772.3635058403556 mfu:17.79488228863224 +Step 10237 | loss:1.1052768230438232 lr:8.182440037972455e-05 tokens_per_second_per_gpu:4463.133462059235 mfu:21.053362029624306 +Step 10238 | loss:1.325717568397522 lr:8.182105579033408e-05 tokens_per_second_per_gpu:4173.405996034211 mfu:19.686668139781236 +Step 10239 | loss:1.0405380725860596 lr:8.181771096161337e-05 tokens_per_second_per_gpu:4723.35005946101 mfu:22.280848117098124 +Step 10240 | loss:1.0569846630096436 lr:8.18143658935876e-05 tokens_per_second_per_gpu:5346.305383889238 mfu:25.219434669564407 +Step 10241 | loss:1.0592705011367798 lr:8.181102058628194e-05 tokens_per_second_per_gpu:4582.85736585343 mfu:21.618120110825878 +Step 10242 | loss:1.0591901540756226 lr:8.180767503972153e-05 tokens_per_second_per_gpu:5352.403986934114 mfu:25.24820282065606 +Step 10243 | loss:1.1034717559814453 lr:8.180432925393154e-05 tokens_per_second_per_gpu:5346.219208306202 mfu:25.219028164636153 +Step 10244 | loss:1.09600830078125 lr:8.180098322893711e-05 tokens_per_second_per_gpu:3294.5101783294567 mfu:15.54076661257853 +Step 10245 | loss:1.1491636037826538 lr:8.179763696476345e-05 tokens_per_second_per_gpu:4704.1524040866425 mfu:22.19028950123947 +Step 10246 | loss:1.1480200290679932 lr:8.17942904614357e-05 tokens_per_second_per_gpu:5346.580685393783 mfu:25.220733313732623 +Step 10247 | loss:1.1457487344741821 lr:8.179094371897902e-05 tokens_per_second_per_gpu:4173.115026818198 mfu:19.685295588344836 +Step 10248 | loss:1.2743316888809204 lr:8.17875967374186e-05 tokens_per_second_per_gpu:3946.991035246373 mfu:18.618630139369255 +Step 10249 | loss:1.2099196910858154 lr:8.178424951677964e-05 tokens_per_second_per_gpu:4397.725314513674 mfu:20.74482063787116 +Step 10250 | loss:1.008424997329712 lr:8.178090205708724e-05 tokens_per_second_per_gpu:4992.930392118502 mfu:23.552504541391134 +Step 10251 | loss:0.957836389541626 lr:8.177755435836664e-05 tokens_per_second_per_gpu:4664.612674076756 mfu:22.003773848609253 +Step 10252 | loss:0.994799017906189 lr:8.1774206420643e-05 tokens_per_second_per_gpu:5310.7296049833185 mfu:25.051617650611398 +Step 10253 | loss:1.158850073814392 lr:8.17708582439415e-05 tokens_per_second_per_gpu:5341.662532473303 mfu:25.197533547283303 +Step 10254 | loss:1.2231433391571045 lr:8.176750982828731e-05 tokens_per_second_per_gpu:3316.848523998037 mfu:15.646140400412005 +Step 10255 | loss:1.1358004808425903 lr:8.176416117370565e-05 tokens_per_second_per_gpu:5324.558586394597 mfu:25.116851315395927 +Step 10256 | loss:1.2150766849517822 lr:8.176081228022166e-05 tokens_per_second_per_gpu:3705.004251655012 mfu:17.477137193966122 +Step 10257 | loss:1.1457535028457642 lr:8.175746314786054e-05 tokens_per_second_per_gpu:3637.630218505993 mfu:17.1593223844068 +Step 10258 | loss:1.164377212524414 lr:8.17541137766475e-05 tokens_per_second_per_gpu:4249.262741966919 mfu:20.04449735284126 +Step 10259 | loss:1.0896111726760864 lr:8.175076416660771e-05 tokens_per_second_per_gpu:5333.483245440523 mfu:25.158950454819408 +Step 10260 | loss:1.1266940832138062 lr:8.174741431776639e-05 tokens_per_second_per_gpu:4780.358593264134 mfu:22.549767097706304 +Step 10261 | loss:0.926963746547699 lr:8.174406423014872e-05 tokens_per_second_per_gpu:5185.89250833274 mfu:24.462739766305607 +Step 10262 | loss:0.9025357961654663 lr:8.174071390377987e-05 tokens_per_second_per_gpu:5258.933472607206 mfu:24.807286456861814 +Step 10263 | loss:1.0455378293991089 lr:8.173736333868507e-05 tokens_per_second_per_gpu:4871.983375491458 mfu:22.981976828272444 +Step 10264 | loss:1.1300435066223145 lr:8.173401253488951e-05 tokens_per_second_per_gpu:4414.553882312829 mfu:20.824203863429517 +Step 10265 | loss:1.0261894464492798 lr:8.173066149241838e-05 tokens_per_second_per_gpu:4217.946064803921 mfu:19.896771243487308 +Step 10266 | loss:1.1204431056976318 lr:8.172731021129692e-05 tokens_per_second_per_gpu:4500.149672328356 mfu:21.22797380908047 +Step 10267 | loss:1.0354982614517212 lr:8.172395869155029e-05 tokens_per_second_per_gpu:5335.806535324396 mfu:25.169909809595765 +Step 10268 | loss:1.1224709749221802 lr:8.172060693320372e-05 tokens_per_second_per_gpu:4838.195076878858 mfu:22.82259165883862 +Step 10269 | loss:0.9833418130874634 lr:8.171725493628244e-05 tokens_per_second_per_gpu:4235.615194223277 mfu:19.98011954162275 +Step 10270 | loss:1.0156856775283813 lr:8.171390270081161e-05 tokens_per_second_per_gpu:5320.5419298187235 mfu:25.097904061015253 +Step 10271 | loss:1.2128809690475464 lr:8.171055022681649e-05 tokens_per_second_per_gpu:5171.586444173105 mfu:24.395255620798366 +Step 10272 | loss:1.068259835243225 lr:8.170719751432228e-05 tokens_per_second_per_gpu:5331.5719666286095 mfu:25.149934626565823 +Step 10273 | loss:1.0942546129226685 lr:8.170384456335418e-05 tokens_per_second_per_gpu:5222.44393549688 mfu:24.635159084555397 +Step 10274 | loss:1.0660408735275269 lr:8.170049137393742e-05 tokens_per_second_per_gpu:3440.860983793314 mfu:16.231128331973785 +Step 10275 | loss:1.051117181777954 lr:8.169713794609721e-05 tokens_per_second_per_gpu:5196.4597283257335 mfu:24.5125871459662 +Step 10276 | loss:1.2528022527694702 lr:8.169378427985878e-05 tokens_per_second_per_gpu:5251.381621417705 mfu:24.77166308632226 +Step 10277 | loss:1.1360028982162476 lr:8.169043037524739e-05 tokens_per_second_per_gpu:3651.1477814617037 mfu:17.223087035202948 +Step 10278 | loss:1.0261286497116089 lr:8.16870762322882e-05 tokens_per_second_per_gpu:4478.005753550611 mfu:21.123517166063714 +Step 10279 | loss:1.078199028968811 lr:8.168372185100648e-05 tokens_per_second_per_gpu:4703.139601533302 mfu:22.18551193879343 +Step 10280 | loss:0.999867856502533 lr:8.168036723142743e-05 tokens_per_second_per_gpu:5332.746779955612 mfu:25.155476421473118 +Step 10281 | loss:1.190039873123169 lr:8.16770123735763e-05 tokens_per_second_per_gpu:4585.166794799765 mfu:21.62901408119495 +Step 10282 | loss:1.1805046796798706 lr:8.167365727747832e-05 tokens_per_second_per_gpu:4759.817113507747 mfu:22.452869432957588 +Step 10283 | loss:1.0645735263824463 lr:8.167030194315872e-05 tokens_per_second_per_gpu:5233.346741264469 mfu:24.686589479571595 +Step 10284 | loss:1.129764437675476 lr:8.166694637064274e-05 tokens_per_second_per_gpu:4493.049059285871 mfu:21.194479005869557 +Step 10285 | loss:1.1276450157165527 lr:8.166359055995563e-05 tokens_per_second_per_gpu:5106.216581306217 mfu:24.086894824406567 +Step 10286 | loss:1.1268728971481323 lr:8.166023451112259e-05 tokens_per_second_per_gpu:4866.407205299602 mfu:22.95567308208472 +Step 10287 | loss:1.0710948705673218 lr:8.16568782241689e-05 tokens_per_second_per_gpu:3245.5401189154572 mfu:15.309766487169004 +Step 10288 | loss:1.1402974128723145 lr:8.165352169911979e-05 tokens_per_second_per_gpu:3765.809310208927 mfu:17.763965029577705 +Step 10289 | loss:0.9762585163116455 lr:8.16501649360005e-05 tokens_per_second_per_gpu:4967.163190202183 mfu:23.43095625361387 +Step 10290 | loss:1.1746952533721924 lr:8.164680793483627e-05 tokens_per_second_per_gpu:5018.200133749743 mfu:23.671706224128208 +Step 10291 | loss:1.1683223247528076 lr:8.164345069565238e-05 tokens_per_second_per_gpu:4217.1630763713 mfu:19.893077753459032 +Step 10292 | loss:1.0410898923873901 lr:8.164009321847406e-05 tokens_per_second_per_gpu:3443.6048725355095 mfu:16.244071723326414 +Step 10293 | loss:1.0904384851455688 lr:8.163673550332654e-05 tokens_per_second_per_gpu:4359.768343773183 mfu:20.565771130761263 +Step 10294 | loss:1.147011160850525 lr:8.16333775502351e-05 tokens_per_second_per_gpu:3691.3919352369303 mfu:17.412925574921363 +Step 10295 | loss:1.037382960319519 lr:8.1630019359225e-05 tokens_per_second_per_gpu:5356.421779731961 mfu:25.267155434789643 +Step 10296 | loss:1.099075198173523 lr:8.162666093032149e-05 tokens_per_second_per_gpu:4272.981393340306 mfu:20.156382278188747 +Step 10297 | loss:1.0312371253967285 lr:8.162330226354984e-05 tokens_per_second_per_gpu:3669.2269566743657 mfu:17.308369589306444 +Step 10298 | loss:1.080022931098938 lr:8.161994335893529e-05 tokens_per_second_per_gpu:4735.665949312513 mfu:22.33894427083515 +Step 10299 | loss:1.0729917287826538 lr:8.16165842165031e-05 tokens_per_second_per_gpu:5401.211634623379 mfu:25.47843682224972 +Step 10300 | loss:1.093588948249817 lr:8.161322483627855e-05 tokens_per_second_per_gpu:4318.8718911400265 mfu:20.37285558603633 +Step 10301 | loss:1.287004828453064 lr:8.160986521828688e-05 tokens_per_second_per_gpu:4071.1076072024553 mfu:19.204109185756916 +Step 10302 | loss:1.1646512746810913 lr:8.16065053625534e-05 tokens_per_second_per_gpu:4749.706625543682 mfu:22.405176536203967 +Step 10303 | loss:1.1699141263961792 lr:8.160314526910337e-05 tokens_per_second_per_gpu:5343.313226913029 mfu:25.205320154593007 +Step 10304 | loss:1.1089038848876953 lr:8.159978493796206e-05 tokens_per_second_per_gpu:5343.180207878722 mfu:25.204692681112792 +Step 10305 | loss:1.222945213317871 lr:8.159642436915471e-05 tokens_per_second_per_gpu:5249.1887521027675 mfu:24.761318947621618 +Step 10306 | loss:1.2508171796798706 lr:8.159306356270664e-05 tokens_per_second_per_gpu:4720.269421740497 mfu:22.26631622335947 +Step 10307 | loss:1.0104410648345947 lr:8.158970251864309e-05 tokens_per_second_per_gpu:5323.689049811566 mfu:25.11274956297553 +Step 10308 | loss:1.0487382411956787 lr:8.158634123698938e-05 tokens_per_second_per_gpu:4819.251915075505 mfu:22.733233532575518 +Step 10309 | loss:1.16011381149292 lr:8.158297971777073e-05 tokens_per_second_per_gpu:3730.9689682981316 mfu:17.599617192409188 +Step 10310 | loss:1.2323287725448608 lr:8.15796179610125e-05 tokens_per_second_per_gpu:4219.527260698281 mfu:19.90423001430084 +Step 10311 | loss:1.0405446290969849 lr:8.157625596673992e-05 tokens_per_second_per_gpu:5352.069412391072 mfu:25.246624575452227 +Step 10312 | loss:1.1356974840164185 lr:8.157289373497827e-05 tokens_per_second_per_gpu:4312.412271687298 mfu:20.342384458953084 +Step 10313 | loss:1.101718783378601 lr:8.156953126575287e-05 tokens_per_second_per_gpu:4539.7456257858 mfu:21.414754677305524 +Step 10314 | loss:1.0686619281768799 lr:8.156616855908901e-05 tokens_per_second_per_gpu:4355.694906281735 mfu:20.546556031113997 +Step 10315 | loss:1.022632360458374 lr:8.156280561501195e-05 tokens_per_second_per_gpu:4040.4591283871487 mfu:19.05953508201542 +Step 10316 | loss:1.0707508325576782 lr:8.1559442433547e-05 tokens_per_second_per_gpu:5344.152942722353 mfu:25.20928123733588 +Step 10317 | loss:1.2554619312286377 lr:8.155607901471948e-05 tokens_per_second_per_gpu:4918.997998571518 mfu:23.20375282686294 +Step 10318 | loss:1.1198678016662598 lr:8.155271535855464e-05 tokens_per_second_per_gpu:4550.100574456405 mfu:21.463600736921194 +Step 10319 | loss:1.1093618869781494 lr:8.154935146507781e-05 tokens_per_second_per_gpu:5129.133564176341 mfu:24.194998142645407 +Step 10320 | loss:1.2278693914413452 lr:8.154598733431429e-05 tokens_per_second_per_gpu:5365.741241862928 mfu:25.311116927725617 +Step 10321 | loss:1.0655158758163452 lr:8.154262296628937e-05 tokens_per_second_per_gpu:5103.428110929409 mfu:24.073741133877057 +Step 10322 | loss:1.063767671585083 lr:8.153925836102837e-05 tokens_per_second_per_gpu:3729.426517238377 mfu:17.59234118759104 +Step 10323 | loss:0.9501327872276306 lr:8.153589351855655e-05 tokens_per_second_per_gpu:4916.325393029695 mfu:23.191145690528636 +Step 10324 | loss:1.0689342021942139 lr:8.153252843889927e-05 tokens_per_second_per_gpu:4463.315196684202 mfu:21.054219302857405 +Step 10325 | loss:0.9903498291969299 lr:8.152916312208183e-05 tokens_per_second_per_gpu:4295.835185868173 mfu:20.26418751680178 +Step 10326 | loss:1.10609769821167 lr:8.15257975681295e-05 tokens_per_second_per_gpu:3849.685245916315 mfu:18.15962212901976 +Step 10327 | loss:1.079718828201294 lr:8.152243177706764e-05 tokens_per_second_per_gpu:5363.445780525608 mfu:25.30028884495139 +Step 10328 | loss:1.1319812536239624 lr:8.151906574892157e-05 tokens_per_second_per_gpu:3731.74806119427 mfu:17.60329230652723 +Step 10329 | loss:1.0321006774902344 lr:8.151569948371656e-05 tokens_per_second_per_gpu:4219.985243800988 mfu:19.90639039873629 +Step 10330 | loss:1.1687217950820923 lr:8.151233298147796e-05 tokens_per_second_per_gpu:5349.200740959607 mfu:25.233092562864194 +Step 10331 | loss:1.0558910369873047 lr:8.150896624223108e-05 tokens_per_second_per_gpu:4210.825059878119 mfu:19.863180248283577 +Step 10332 | loss:1.0091365575790405 lr:8.150559926600126e-05 tokens_per_second_per_gpu:4235.548949531424 mfu:19.97980705410877 +Step 10333 | loss:1.0207346677780151 lr:8.150223205281378e-05 tokens_per_second_per_gpu:3754.510849473899 mfu:17.710668262574398 +Step 10334 | loss:1.2366191148757935 lr:8.1498864602694e-05 tokens_per_second_per_gpu:4938.549694334867 mfu:23.295981511641937 +Step 10335 | loss:1.0778539180755615 lr:8.149549691566724e-05 tokens_per_second_per_gpu:4985.516783301871 mfu:23.517533283710993 +Step 10336 | loss:1.209816336631775 lr:8.149212899175886e-05 tokens_per_second_per_gpu:3980.5714170411734 mfu:18.77703452969926 +Step 10337 | loss:1.1042721271514893 lr:8.148876083099413e-05 tokens_per_second_per_gpu:5321.882833125018 mfu:25.10422933069274 +Step 10338 | loss:0.9356644749641418 lr:8.148539243339841e-05 tokens_per_second_per_gpu:5363.3810327249685 mfu:25.299983418528658 +Step 10339 | loss:1.0621997117996216 lr:8.148202379899705e-05 tokens_per_second_per_gpu:5316.968873475834 mfu:25.08104934461957 +Step 10340 | loss:0.9340762495994568 lr:8.147865492781537e-05 tokens_per_second_per_gpu:5327.49879511994 mfu:25.13072078160471 +Step 10341 | loss:1.0923188924789429 lr:8.14752858198787e-05 tokens_per_second_per_gpu:4255.669421066941 mfu:20.074718751248238 +Step 10342 | loss:1.154192566871643 lr:8.147191647521239e-05 tokens_per_second_per_gpu:5321.387145297801 mfu:25.101891086638695 +Step 10343 | loss:1.0254931449890137 lr:8.146854689384178e-05 tokens_per_second_per_gpu:5246.569132113425 mfu:24.748961753178634 +Step 10344 | loss:1.208542823791504 lr:8.146517707579221e-05 tokens_per_second_per_gpu:5342.216652217357 mfu:25.20014742465858 +Step 10345 | loss:1.249038577079773 lr:8.146180702108905e-05 tokens_per_second_per_gpu:3598.5854497802875 mfu:16.975141548603165 +Step 10346 | loss:1.1356604099273682 lr:8.14584367297576e-05 tokens_per_second_per_gpu:4035.2449084672558 mfu:19.034938717015628 +Step 10347 | loss:0.9830679297447205 lr:8.145506620182325e-05 tokens_per_second_per_gpu:4557.6349239119 mfu:21.499141548795333 +Step 10348 | loss:1.15579092502594 lr:8.145169543731132e-05 tokens_per_second_per_gpu:3472.228178066876 mfu:16.37909273915738 +Step 10349 | loss:1.0401787757873535 lr:8.144832443624719e-05 tokens_per_second_per_gpu:3656.810036842019 mfu:17.249796859912284 +Step 10350 | loss:0.9972026348114014 lr:8.14449531986562e-05 tokens_per_second_per_gpu:4484.742556478767 mfu:21.155295814892927 +Step 10351 | loss:1.0625712871551514 lr:8.144158172456368e-05 tokens_per_second_per_gpu:5336.9973424012305 mfu:25.175527049749835 +Step 10352 | loss:0.9650688767433167 lr:8.143821001399503e-05 tokens_per_second_per_gpu:5331.669511732558 mfu:25.15039476346406 +Step 10353 | loss:1.0994857549667358 lr:8.143483806697559e-05 tokens_per_second_per_gpu:5326.431806168008 mfu:25.125687612671356 +Step 10354 | loss:1.1364411115646362 lr:8.143146588353073e-05 tokens_per_second_per_gpu:3860.6439304051314 mfu:18.211316113497883 +Step 10355 | loss:0.9578707814216614 lr:8.142809346368579e-05 tokens_per_second_per_gpu:5331.287602172731 mfu:25.148593230909864 +Step 10356 | loss:1.0297479629516602 lr:8.142472080746615e-05 tokens_per_second_per_gpu:5321.162977976762 mfu:25.100833651138462 +Step 10357 | loss:1.0274755954742432 lr:8.142134791489719e-05 tokens_per_second_per_gpu:3144.2504084305797 mfu:14.831965641005183 +Step 10358 | loss:1.0449775457382202 lr:8.141797478600424e-05 tokens_per_second_per_gpu:5342.468675666776 mfu:25.201336262269127 +Step 10359 | loss:1.0326330661773682 lr:8.14146014208127e-05 tokens_per_second_per_gpu:4978.572359759373 mfu:23.484775253020196 +Step 10360 | loss:1.0401527881622314 lr:8.141122781934795e-05 tokens_per_second_per_gpu:5037.530660753001 mfu:23.762891618130574 +Step 10361 | loss:1.0837457180023193 lr:8.140785398163534e-05 tokens_per_second_per_gpu:4350.063415590798 mfu:20.51999132869346 +Step 10362 | loss:1.1512712240219116 lr:8.140447990770023e-05 tokens_per_second_per_gpu:5194.128031833542 mfu:24.50158813578409 +Step 10363 | loss:1.1105339527130127 lr:8.140110559756805e-05 tokens_per_second_per_gpu:3889.759448889922 mfu:18.34865898181034 +Step 10364 | loss:1.1901507377624512 lr:8.139773105126414e-05 tokens_per_second_per_gpu:3632.8233166204095 mfu:17.13664740807047 +Step 10365 | loss:1.09644615650177 lr:8.139435626881389e-05 tokens_per_second_per_gpu:5010.294062980643 mfu:23.634411939395587 +Step 10366 | loss:0.8948419094085693 lr:8.139098125024267e-05 tokens_per_second_per_gpu:4058.425914253664 mfu:19.144287476397537 +Step 10367 | loss:1.0148751735687256 lr:8.138760599557588e-05 tokens_per_second_per_gpu:5325.866454411214 mfu:25.123020751975602 +Step 10368 | loss:1.109666347503662 lr:8.13842305048389e-05 tokens_per_second_per_gpu:5160.604671374066 mfu:24.343452724821653 +Step 10369 | loss:1.1655126810073853 lr:8.138085477805712e-05 tokens_per_second_per_gpu:3897.2623103236665 mfu:18.384051259313473 +Step 10370 | loss:1.0812485218048096 lr:8.137747881525592e-05 tokens_per_second_per_gpu:5339.954736481064 mfu:25.18947758220787 +Step 10371 | loss:1.2055221796035767 lr:8.13741026164607e-05 tokens_per_second_per_gpu:4842.054612644032 mfu:22.840797747547644 +Step 10372 | loss:1.0774548053741455 lr:8.137072618169685e-05 tokens_per_second_per_gpu:5351.925461825723 mfu:25.245945536074977 +Step 10373 | loss:1.1372041702270508 lr:8.136734951098977e-05 tokens_per_second_per_gpu:4392.436843494739 mfu:20.719874017768646 +Step 10374 | loss:1.007997751235962 lr:8.136397260436486e-05 tokens_per_second_per_gpu:5320.7512653342055 mfu:25.098891532358103 +Step 10375 | loss:1.196932315826416 lr:8.13605954618475e-05 tokens_per_second_per_gpu:5318.175947824771 mfu:25.086743320271626 +Step 10376 | loss:1.1678308248519897 lr:8.13572180834631e-05 tokens_per_second_per_gpu:4585.351933791548 mfu:21.629887413407157 +Step 10377 | loss:1.053181767463684 lr:8.135384046923706e-05 tokens_per_second_per_gpu:4395.094323271917 mfu:20.732409803290402 +Step 10378 | loss:1.0580079555511475 lr:8.13504626191948e-05 tokens_per_second_per_gpu:5093.176304765864 mfu:24.025381615065122 +Step 10379 | loss:1.0298528671264648 lr:8.134708453336168e-05 tokens_per_second_per_gpu:5337.530680882774 mfu:25.178042898364204 +Step 10380 | loss:1.3361873626708984 lr:8.134370621176316e-05 tokens_per_second_per_gpu:4875.758197294411 mfu:22.999783306767988 +Step 10381 | loss:1.1217138767242432 lr:8.13403276544246e-05 tokens_per_second_per_gpu:4116.01594014643 mfu:19.415949454405343 +Step 10382 | loss:1.0912530422210693 lr:8.133694886137147e-05 tokens_per_second_per_gpu:4123.167932251919 mfu:19.44968662142262 +Step 10383 | loss:0.9510622024536133 lr:8.133356983262914e-05 tokens_per_second_per_gpu:4925.7576854545105 mfu:23.235639423211524 +Step 10384 | loss:1.0575569868087769 lr:8.133019056822304e-05 tokens_per_second_per_gpu:5376.701506059475 mfu:25.362818363954627 +Step 10385 | loss:1.039636492729187 lr:8.132681106817856e-05 tokens_per_second_per_gpu:4559.4371797114445 mfu:21.507643096898335 +Step 10386 | loss:0.9953164458274841 lr:8.132343133252114e-05 tokens_per_second_per_gpu:4128.144189735075 mfu:19.47316047700774 +Step 10387 | loss:1.19692063331604 lr:8.132005136127621e-05 tokens_per_second_per_gpu:3540.7497166758235 mfu:16.702320527755806 +Step 10388 | loss:0.9997108578681946 lr:8.131667115446916e-05 tokens_per_second_per_gpu:3443.83800545359 mfu:16.245171451077578 +Step 10389 | loss:1.0421719551086426 lr:8.131329071212543e-05 tokens_per_second_per_gpu:3161.684159626568 mfu:14.914203619901334 +Step 10390 | loss:0.953427255153656 lr:8.130991003427046e-05 tokens_per_second_per_gpu:4833.948860253322 mfu:22.802561530536696 +Step 10391 | loss:1.1478363275527954 lr:8.130652912092966e-05 tokens_per_second_per_gpu:3187.0362790416907 mfu:15.033793892700075 +Step 10392 | loss:1.2130876779556274 lr:8.130314797212845e-05 tokens_per_second_per_gpu:4124.33251472308 mfu:19.455180155637343 +Step 10393 | loss:1.0846072435379028 lr:8.129976658789228e-05 tokens_per_second_per_gpu:3712.5843239423525 mfu:17.512893688239203 +Step 10394 | loss:1.0861234664916992 lr:8.129638496824656e-05 tokens_per_second_per_gpu:4661.846552301134 mfu:21.990725580244604 +Step 10395 | loss:1.188224196434021 lr:8.129300311321675e-05 tokens_per_second_per_gpu:4614.991174353735 mfu:21.76970076811489 +Step 10396 | loss:1.0610195398330688 lr:8.128962102282827e-05 tokens_per_second_per_gpu:5330.23099447824 mfu:25.14360903213909 +Step 10397 | loss:1.0402061939239502 lr:8.128623869710655e-05 tokens_per_second_per_gpu:4453.317121816694 mfu:21.00705667785989 +Step 10398 | loss:1.060446858406067 lr:8.128285613607703e-05 tokens_per_second_per_gpu:5367.7015096450805 mfu:25.3203638453102 +Step 10399 | loss:1.1451939344406128 lr:8.127947333976516e-05 tokens_per_second_per_gpu:4376.245021286925 mfu:20.64349442980465 +Step 10400 | loss:1.0430291891098022 lr:8.127609030819639e-05 tokens_per_second_per_gpu:4972.102708714084 mfu:23.454256805203272 +Step 10401 | loss:0.9637937545776367 lr:8.127270704139614e-05 tokens_per_second_per_gpu:5340.299961171798 mfu:25.191106065975465 +Step 10402 | loss:1.1385186910629272 lr:8.126932353938987e-05 tokens_per_second_per_gpu:4115.660707422741 mfu:19.414273761038753 +Step 10403 | loss:1.1635048389434814 lr:8.126593980220305e-05 tokens_per_second_per_gpu:3998.1196225226595 mfu:18.85981240898802 +Step 10404 | loss:1.1693570613861084 lr:8.126255582986108e-05 tokens_per_second_per_gpu:3609.268800353114 mfu:17.02553673602275 +Step 10405 | loss:1.0498502254486084 lr:8.125917162238945e-05 tokens_per_second_per_gpu:4474.949845681481 mfu:21.10910192725338 +Step 10406 | loss:1.1789010763168335 lr:8.125578717981361e-05 tokens_per_second_per_gpu:3989.6614192995567 mfu:18.819913621266366 +Step 10407 | loss:1.288861632347107 lr:8.1252402502159e-05 tokens_per_second_per_gpu:3435.4494219819776 mfu:16.205601071602057 +Step 10408 | loss:1.2468997240066528 lr:8.124901758945108e-05 tokens_per_second_per_gpu:4133.550854266688 mfu:19.49866463607545 +Step 10409 | loss:1.0020095109939575 lr:8.12456324417153e-05 tokens_per_second_per_gpu:2919.475390374522 mfu:13.77166353026209 +Step 10410 | loss:1.0990204811096191 lr:8.124224705897715e-05 tokens_per_second_per_gpu:3381.612629220853 mfu:15.951643734643808 +Step 10411 | loss:1.108708143234253 lr:8.123886144126207e-05 tokens_per_second_per_gpu:4175.457231279306 mfu:19.69634416640917 +Step 10412 | loss:1.0946192741394043 lr:8.123547558859553e-05 tokens_per_second_per_gpu:5357.332110472884 mfu:25.271449620213016 +Step 10413 | loss:0.997181236743927 lr:8.123208950100299e-05 tokens_per_second_per_gpu:4657.786381568331 mfu:21.97157306216141 +Step 10414 | loss:0.9405990839004517 lr:8.12287031785099e-05 tokens_per_second_per_gpu:5343.494399805089 mfu:25.206174778036136 +Step 10415 | loss:1.1905583143234253 lr:8.122531662114178e-05 tokens_per_second_per_gpu:5372.421465199893 mfu:25.34262868096971 +Step 10416 | loss:1.01968252658844 lr:8.122192982892405e-05 tokens_per_second_per_gpu:5359.726760660723 mfu:25.28274559371851 +Step 10417 | loss:1.0656369924545288 lr:8.12185428018822e-05 tokens_per_second_per_gpu:5353.748432064291 mfu:25.254544797721096 +Step 10418 | loss:1.1103626489639282 lr:8.12151555400417e-05 tokens_per_second_per_gpu:5358.626553754796 mfu:25.277555730028595 +Step 10419 | loss:1.0256729125976562 lr:8.121176804342807e-05 tokens_per_second_per_gpu:5047.33272016496 mfu:23.80912960478057 +Step 10420 | loss:1.126965045928955 lr:8.120838031206672e-05 tokens_per_second_per_gpu:4541.995192345815 mfu:21.425366266584746 +Step 10421 | loss:0.9396136403083801 lr:8.120499234598315e-05 tokens_per_second_per_gpu:5328.088561742219 mfu:25.13350281138656 +Step 10422 | loss:1.1735941171646118 lr:8.120160414520288e-05 tokens_per_second_per_gpu:5339.797867463484 mfu:25.188737604287333 +Step 10423 | loss:1.1003247499465942 lr:8.119821570975135e-05 tokens_per_second_per_gpu:4111.560496299556 mfu:19.394932365602532 +Step 10424 | loss:1.040732979774475 lr:8.119482703965406e-05 tokens_per_second_per_gpu:4608.121622893496 mfu:21.73729592181061 +Step 10425 | loss:1.0196399688720703 lr:8.119143813493648e-05 tokens_per_second_per_gpu:5348.178192178621 mfu:25.22826902579947 +Step 10426 | loss:0.9746562838554382 lr:8.118804899562414e-05 tokens_per_second_per_gpu:4680.914817834717 mfu:22.080673842148094 +Step 10427 | loss:1.1011277437210083 lr:8.118465962174248e-05 tokens_per_second_per_gpu:4100.574884336321 mfu:19.343111359633273 +Step 10428 | loss:1.121240496635437 lr:8.118127001331702e-05 tokens_per_second_per_gpu:4851.158060471813 mfu:22.88374026415965 +Step 10429 | loss:1.0916506052017212 lr:8.117788017037327e-05 tokens_per_second_per_gpu:4734.2285198997015 mfu:22.332163671044174 +Step 10430 | loss:1.0552445650100708 lr:8.117449009293668e-05 tokens_per_second_per_gpu:4298.046479933248 mfu:20.27461856819241 +Step 10431 | loss:1.1224255561828613 lr:8.117109978103279e-05 tokens_per_second_per_gpu:4796.452210746103 mfu:22.625683437223042 +Step 10432 | loss:1.1047569513320923 lr:8.116770923468706e-05 tokens_per_second_per_gpu:5307.405753940591 mfu:25.035938478135385 +Step 10433 | loss:1.0348609685897827 lr:8.116431845392504e-05 tokens_per_second_per_gpu:5328.506765807183 mfu:25.135475551313966 +Step 10434 | loss:1.0886117219924927 lr:8.116092743877219e-05 tokens_per_second_per_gpu:4521.309738037752 mfu:21.32778944931118 +Step 10435 | loss:0.8846495747566223 lr:8.115753618925402e-05 tokens_per_second_per_gpu:4108.6704328499745 mfu:19.381299443215187 +Step 10436 | loss:1.0174651145935059 lr:8.115414470539605e-05 tokens_per_second_per_gpu:5234.996468298743 mfu:24.694371523464724 +Step 10437 | loss:0.9635800719261169 lr:8.115075298722378e-05 tokens_per_second_per_gpu:5319.460834711775 mfu:25.09280435094193 +Step 10438 | loss:1.1310957670211792 lr:8.114736103476274e-05 tokens_per_second_per_gpu:3987.7180237347116 mfu:18.810746292809515 +Step 10439 | loss:1.0942670106887817 lr:8.114396884803841e-05 tokens_per_second_per_gpu:4110.1379352391505 mfu:19.388221902366652 +Step 10440 | loss:1.081760048866272 lr:8.11405764270763e-05 tokens_per_second_per_gpu:4597.461376020931 mfu:21.68700971848705 +Step 10441 | loss:1.1656410694122314 lr:8.113718377190199e-05 tokens_per_second_per_gpu:5069.243455207853 mfu:23.912486280334782 +Step 10442 | loss:1.154549479484558 lr:8.113379088254089e-05 tokens_per_second_per_gpu:5302.31094254889 mfu:25.011905383536273 +Step 10443 | loss:1.119037389755249 lr:8.113039775901861e-05 tokens_per_second_per_gpu:5344.042751948217 mfu:25.208761448653792 +Step 10444 | loss:1.054720401763916 lr:8.112700440136063e-05 tokens_per_second_per_gpu:5328.566466645709 mfu:25.135757170355305 +Step 10445 | loss:1.0916845798492432 lr:8.112361080959248e-05 tokens_per_second_per_gpu:5332.192350526965 mfu:25.152861083262774 +Step 10446 | loss:1.001914620399475 lr:8.112021698373966e-05 tokens_per_second_per_gpu:4744.668127269649 mfu:22.381409080188902 +Step 10447 | loss:1.2165535688400269 lr:8.111682292382774e-05 tokens_per_second_per_gpu:3309.3190699938673 mfu:15.610622681217013 +Step 10448 | loss:1.0981714725494385 lr:8.111342862988223e-05 tokens_per_second_per_gpu:4941.268997092296 mfu:23.30880892670962 +Step 10449 | loss:1.163137435913086 lr:8.111003410192864e-05 tokens_per_second_per_gpu:4317.478968973167 mfu:20.366284934518223 +Step 10450 | loss:1.093609094619751 lr:8.11066393399925e-05 tokens_per_second_per_gpu:3924.406801249025 mfu:18.51209645433611 +Step 10451 | loss:0.9826234579086304 lr:8.11032443440994e-05 tokens_per_second_per_gpu:5315.539272087942 mfu:25.07430567095773 +Step 10452 | loss:1.2657763957977295 lr:8.10998491142748e-05 tokens_per_second_per_gpu:4011.887745129399 mfu:18.924758992407956 +Step 10453 | loss:1.119713544845581 lr:8.109645365054425e-05 tokens_per_second_per_gpu:3294.176157940057 mfu:15.539190981411975 +Step 10454 | loss:1.0801900625228882 lr:8.109305795293331e-05 tokens_per_second_per_gpu:3712.121464396004 mfu:17.510710300787093 +Step 10455 | loss:1.1610462665557861 lr:8.108966202146755e-05 tokens_per_second_per_gpu:4403.677252305004 mfu:20.772896943938168 +Step 10456 | loss:1.064207673072815 lr:8.108626585617245e-05 tokens_per_second_per_gpu:4246.703076179064 mfu:20.03242297259593 +Step 10457 | loss:1.1593300104141235 lr:8.108286945707358e-05 tokens_per_second_per_gpu:4272.61871005274 mfu:20.154671439241152 +Step 10458 | loss:1.0980321168899536 lr:8.107947282419648e-05 tokens_per_second_per_gpu:3862.26502302319 mfu:18.218963083964187 +Step 10459 | loss:1.1886672973632812 lr:8.10760759575667e-05 tokens_per_second_per_gpu:3848.8714817288514 mfu:18.155783464505006 +Step 10460 | loss:0.9697736501693726 lr:8.10726788572098e-05 tokens_per_second_per_gpu:3348.5769939223483 mfu:15.795809006480166 +Step 10461 | loss:1.0278701782226562 lr:8.10692815231513e-05 tokens_per_second_per_gpu:5365.1665808008165 mfu:25.30840615344835 +Step 10462 | loss:1.0351481437683105 lr:8.106588395541677e-05 tokens_per_second_per_gpu:5332.187495503439 mfu:25.15283818128845 +Step 10463 | loss:1.1172645092010498 lr:8.106248615403176e-05 tokens_per_second_per_gpu:5287.398075166574 mfu:24.941558843696345 +Step 10464 | loss:1.085943341255188 lr:8.105908811902184e-05 tokens_per_second_per_gpu:4126.2327029879325 mfu:19.4641436678882 +Step 10465 | loss:1.2542340755462646 lr:8.105568985041255e-05 tokens_per_second_per_gpu:3642.5776193966367 mfu:17.1826601185216 +Step 10466 | loss:1.1012611389160156 lr:8.105229134822946e-05 tokens_per_second_per_gpu:4980.152419898737 mfu:23.492228666283633 +Step 10467 | loss:1.1161022186279297 lr:8.104889261249813e-05 tokens_per_second_per_gpu:5349.89251064939 mfu:25.236355758518975 +Step 10468 | loss:1.0814017057418823 lr:8.10454936432441e-05 tokens_per_second_per_gpu:4775.919582296162 mfu:22.528827525597016 +Step 10469 | loss:1.0580347776412964 lr:8.104209444049296e-05 tokens_per_second_per_gpu:3907.0983305309924 mfu:18.430449444829524 +Step 10470 | loss:1.134515404701233 lr:8.103869500427026e-05 tokens_per_second_per_gpu:5183.25770110807 mfu:24.45031093108237 +Step 10471 | loss:1.1481817960739136 lr:8.103529533460157e-05 tokens_per_second_per_gpu:4960.48543543626 mfu:23.399456145040567 +Step 10472 | loss:1.1518268585205078 lr:8.103189543151247e-05 tokens_per_second_per_gpu:4535.993038795046 mfu:21.397053084212775 +Step 10473 | loss:0.9652701020240784 lr:8.102849529502854e-05 tokens_per_second_per_gpu:5165.979763881457 mfu:24.368807953264366 +Step 10474 | loss:1.0876046419143677 lr:8.102509492517531e-05 tokens_per_second_per_gpu:5212.906126416012 mfu:24.590167611802908 +Step 10475 | loss:1.0960235595703125 lr:8.102169432197842e-05 tokens_per_second_per_gpu:3042.132102281141 mfu:14.350256167718632 +Step 10476 | loss:1.1710214614868164 lr:8.101829348546336e-05 tokens_per_second_per_gpu:5352.975798205404 mfu:25.250900151983803 +Step 10477 | loss:1.234951138496399 lr:8.10148924156558e-05 tokens_per_second_per_gpu:4191.4586139447365 mfu:19.77182541855941 +Step 10478 | loss:1.0187461376190186 lr:8.101149111258125e-05 tokens_per_second_per_gpu:5338.3696799835925 mfu:25.182000600270772 +Step 10479 | loss:0.9621291160583496 lr:8.100808957626533e-05 tokens_per_second_per_gpu:4971.066926235762 mfu:23.449370842530318 +Step 10480 | loss:1.093074917793274 lr:8.10046878067336e-05 tokens_per_second_per_gpu:5333.693210840301 mfu:25.159940897433952 +Step 10481 | loss:0.9122405052185059 lr:8.100128580401168e-05 tokens_per_second_per_gpu:5375.701766514688 mfu:25.35810242194871 +Step 10482 | loss:1.0439199209213257 lr:8.099788356812512e-05 tokens_per_second_per_gpu:5307.696696040296 mfu:25.037310901659534 +Step 10483 | loss:1.0468988418579102 lr:8.09944810990995e-05 tokens_per_second_per_gpu:5327.327547412297 mfu:25.129912976949882 +Step 10484 | loss:1.0643038749694824 lr:8.099107839696047e-05 tokens_per_second_per_gpu:4961.071433933804 mfu:23.402220399935135 +Step 10485 | loss:1.1212552785873413 lr:8.098767546173357e-05 tokens_per_second_per_gpu:5381.278545480361 mfu:25.384409039826398 +Step 10486 | loss:0.9757557511329651 lr:8.098427229344441e-05 tokens_per_second_per_gpu:4203.280096436409 mfu:19.82758936842552 +Step 10487 | loss:0.8859764933586121 lr:8.098086889211857e-05 tokens_per_second_per_gpu:5076.444471846942 mfu:23.946454704441496 +Step 10488 | loss:1.0366123914718628 lr:8.097746525778168e-05 tokens_per_second_per_gpu:5101.678055187676 mfu:24.065485822353004 +Step 10489 | loss:0.9985392689704895 lr:8.097406139045931e-05 tokens_per_second_per_gpu:5350.307725392708 mfu:25.238314397306148 +Step 10490 | loss:0.9874551296234131 lr:8.097065729017707e-05 tokens_per_second_per_gpu:3699.577584065589 mfu:17.451538677061848 +Step 10491 | loss:1.089745283126831 lr:8.09672529569606e-05 tokens_per_second_per_gpu:4949.860656653142 mfu:23.34933725074641 +Step 10492 | loss:1.0456334352493286 lr:8.096384839083541e-05 tokens_per_second_per_gpu:4760.829488106248 mfu:22.457644976667638 +Step 10493 | loss:1.0393122434616089 lr:8.096044359182722e-05 tokens_per_second_per_gpu:3215.4643556486976 mfu:15.167893980385346 +Step 10494 | loss:0.920779287815094 lr:8.095703855996156e-05 tokens_per_second_per_gpu:5375.435926541816 mfu:25.3568484094397 +Step 10495 | loss:1.0888055562973022 lr:8.095363329526406e-05 tokens_per_second_per_gpu:4234.053856380151 mfu:19.972754444625018 +Step 10496 | loss:1.2849222421646118 lr:8.095022779776034e-05 tokens_per_second_per_gpu:3897.9026001438947 mfu:18.3870716156402 +Step 10497 | loss:1.1476086378097534 lr:8.094682206747599e-05 tokens_per_second_per_gpu:4650.945821350392 mfu:21.93930497249347 +Step 10498 | loss:1.1039297580718994 lr:8.094341610443665e-05 tokens_per_second_per_gpu:4294.914164050021 mfu:20.259842899743035 +Step 10499 | loss:1.0183589458465576 lr:8.094000990866794e-05 tokens_per_second_per_gpu:3766.064131045628 mfu:17.765167062941345 +Step 10500 | loss:1.1274868249893188 lr:8.093660348019546e-05 tokens_per_second_per_gpu:3522.336069196294 mfu:16.615460210902643 +Step 10501 | loss:1.099034309387207 lr:8.093319681904485e-05 tokens_per_second_per_gpu:3750.279397656165 mfu:17.69070778239002 +Step 10502 | loss:1.0349819660186768 lr:8.092978992524171e-05 tokens_per_second_per_gpu:4885.739354322807 mfu:23.046866127431926 +Step 10503 | loss:1.2101593017578125 lr:8.092638279881167e-05 tokens_per_second_per_gpu:5311.484569848372 mfu:25.055178948690145 +Step 10504 | loss:1.0029823780059814 lr:8.092297543978036e-05 tokens_per_second_per_gpu:5012.027424206976 mfu:23.642588500042127 +Step 10505 | loss:1.0696274042129517 lr:8.09195678481734e-05 tokens_per_second_per_gpu:5089.115778148271 mfu:24.006227418212653 +Step 10506 | loss:1.0454115867614746 lr:8.091616002401645e-05 tokens_per_second_per_gpu:4039.2775590741435 mfu:19.05396141796946 +Step 10507 | loss:1.089109182357788 lr:8.09127519673351e-05 tokens_per_second_per_gpu:4555.0905314523725 mfu:21.487139215445623 +Step 10508 | loss:1.1234369277954102 lr:8.090934367815498e-05 tokens_per_second_per_gpu:4998.202651587608 mfu:23.57737468884713 +Step 10509 | loss:1.0968824625015259 lr:8.090593515650177e-05 tokens_per_second_per_gpu:4952.537130422204 mfu:23.361962634168595 +Step 10510 | loss:1.2614953517913818 lr:8.090252640240106e-05 tokens_per_second_per_gpu:3400.439310674384 mfu:16.040452403223195 +Step 10511 | loss:1.1004745960235596 lr:8.089911741587852e-05 tokens_per_second_per_gpu:5372.821130145281 mfu:25.344513968706977 +Step 10512 | loss:1.1060429811477661 lr:8.089570819695975e-05 tokens_per_second_per_gpu:3696.1978559381496 mfu:17.43559592284364 +Step 10513 | loss:1.0499595403671265 lr:8.089229874567045e-05 tokens_per_second_per_gpu:4144.691157657869 mfu:19.551215347903405 +Step 10514 | loss:1.1524466276168823 lr:8.08888890620362e-05 tokens_per_second_per_gpu:5042.655395618018 mfu:23.787065866858786 +Step 10515 | loss:1.0428317785263062 lr:8.088547914608268e-05 tokens_per_second_per_gpu:5363.4362522957235 mfu:25.30024389866534 +Step 10516 | loss:0.9694697260856628 lr:8.088206899783555e-05 tokens_per_second_per_gpu:4399.162014636507 mfu:20.751597797476883 +Step 10517 | loss:1.0426609516143799 lr:8.087865861732042e-05 tokens_per_second_per_gpu:4812.224326993487 mfu:22.70008320056247 +Step 10518 | loss:1.0519064664840698 lr:8.087524800456297e-05 tokens_per_second_per_gpu:5298.260178124383 mfu:24.992797236614194 +Step 10519 | loss:1.218106746673584 lr:8.087183715958884e-05 tokens_per_second_per_gpu:4241.873950871973 mfu:20.009643164588645 +Step 10520 | loss:1.1311569213867188 lr:8.086842608242368e-05 tokens_per_second_per_gpu:5178.247224575528 mfu:24.42667565840292 +Step 10521 | loss:1.1116797924041748 lr:8.086501477309314e-05 tokens_per_second_per_gpu:4222.145161873563 mfu:19.916579100804643 +Step 10522 | loss:1.1060748100280762 lr:8.086160323162288e-05 tokens_per_second_per_gpu:5081.7012743656505 mfu:23.971251938825237 +Step 10523 | loss:1.14216947555542 lr:8.08581914580386e-05 tokens_per_second_per_gpu:4381.225570755403 mfu:20.666988531416823 +Step 10524 | loss:1.064391851425171 lr:8.085477945236589e-05 tokens_per_second_per_gpu:5209.450948265164 mfu:24.57386894695118 +Step 10525 | loss:1.1502430438995361 lr:8.085136721463045e-05 tokens_per_second_per_gpu:4793.107145067867 mfu:22.60990419169175 +Step 10526 | loss:1.007464051246643 lr:8.084795474485795e-05 tokens_per_second_per_gpu:5328.42691984247 mfu:25.13509890427529 +Step 10527 | loss:1.2873778343200684 lr:8.084454204307405e-05 tokens_per_second_per_gpu:5019.976742928839 mfu:23.68008679274678 +Step 10528 | loss:1.0742366313934326 lr:8.084112910930442e-05 tokens_per_second_per_gpu:3367.285189180413 mfu:15.88405875545971 +Step 10529 | loss:1.0609205961227417 lr:8.08377159435747e-05 tokens_per_second_per_gpu:4867.587810104924 mfu:22.961242196383377 +Step 10530 | loss:1.2360844612121582 lr:8.083430254591058e-05 tokens_per_second_per_gpu:4841.419308820243 mfu:22.83780091101682 +Step 10531 | loss:1.0971322059631348 lr:8.083088891633777e-05 tokens_per_second_per_gpu:3822.1630757976895 mfu:18.02979535680377 +Step 10532 | loss:1.0504099130630493 lr:8.082747505488187e-05 tokens_per_second_per_gpu:4079.051028691333 mfu:19.241579660207414 +Step 10533 | loss:1.054274082183838 lr:8.082406096156862e-05 tokens_per_second_per_gpu:5389.121154812389 mfu:25.421403966132708 +Step 10534 | loss:1.048804521560669 lr:8.082064663642368e-05 tokens_per_second_per_gpu:4416.634209143249 mfu:20.834017120028967 +Step 10535 | loss:0.9806123971939087 lr:8.08172320794727e-05 tokens_per_second_per_gpu:5119.629926879623 mfu:24.150167864029065 +Step 10536 | loss:1.0228831768035889 lr:8.08138172907414e-05 tokens_per_second_per_gpu:5320.993931826296 mfu:25.10003623160447 +Step 10537 | loss:1.0616772174835205 lr:8.081040227025544e-05 tokens_per_second_per_gpu:5139.138824901413 mfu:24.24219466455908 +Step 10538 | loss:1.087395429611206 lr:8.080698701804055e-05 tokens_per_second_per_gpu:5330.121607997436 mfu:25.143093037445876 +Step 10539 | loss:1.0781968832015991 lr:8.080357153412235e-05 tokens_per_second_per_gpu:5037.086991080192 mfu:23.76079875258608 +Step 10540 | loss:1.0846706628799438 lr:8.080015581852655e-05 tokens_per_second_per_gpu:5144.053677400195 mfu:24.26537886235614 +Step 10541 | loss:0.9785837531089783 lr:8.079673987127884e-05 tokens_per_second_per_gpu:4826.190381517525 mfu:22.76596346260697 +Step 10542 | loss:1.2274013757705688 lr:8.079332369240495e-05 tokens_per_second_per_gpu:4680.001255886317 mfu:22.07636441456775 +Step 10543 | loss:1.046553611755371 lr:8.078990728193052e-05 tokens_per_second_per_gpu:3665.2868757197843 mfu:17.28978355519655 +Step 10544 | loss:1.269538164138794 lr:8.078649063988128e-05 tokens_per_second_per_gpu:4324.75593122104 mfu:20.40061160701923 +Step 10545 | loss:1.1240642070770264 lr:8.07830737662829e-05 tokens_per_second_per_gpu:4514.68416355785 mfu:21.296535484049908 +Step 10546 | loss:1.101577639579773 lr:8.077965666116112e-05 tokens_per_second_per_gpu:4566.6244004526325 mfu:21.541546443400822 +Step 10547 | loss:1.1051093339920044 lr:8.077623932454162e-05 tokens_per_second_per_gpu:5327.29530377891 mfu:25.12976087822987 +Step 10548 | loss:0.9496921300888062 lr:8.077282175645007e-05 tokens_per_second_per_gpu:3867.776297465915 mfu:18.244960705830906 +Step 10549 | loss:1.1171844005584717 lr:8.076940395691222e-05 tokens_per_second_per_gpu:4283.00230339285 mfu:20.203652620650157 +Step 10550 | loss:1.0350830554962158 lr:8.076598592595373e-05 tokens_per_second_per_gpu:4225.310102041052 mfu:19.931508663569165 +Step 10551 | loss:1.2688711881637573 lr:8.076256766360035e-05 tokens_per_second_per_gpu:3696.9495681774947 mfu:17.439141877731963 +Step 10552 | loss:1.0048832893371582 lr:8.075914916987779e-05 tokens_per_second_per_gpu:4328.272441150789 mfu:20.417199584336316 +Step 10553 | loss:1.0735973119735718 lr:8.075573044481173e-05 tokens_per_second_per_gpu:4333.056956449874 mfu:20.439768959324585 +Step 10554 | loss:0.9849001169204712 lr:8.07523114884279e-05 tokens_per_second_per_gpu:5314.986113650732 mfu:25.07169632823085 +Step 10555 | loss:1.00796639919281 lr:8.0748892300752e-05 tokens_per_second_per_gpu:4641.081885259997 mfu:21.892775102994037 +Step 10556 | loss:1.056004524230957 lr:8.074547288180976e-05 tokens_per_second_per_gpu:3210.3869940572067 mfu:15.143943199471066 +Step 10557 | loss:1.3170180320739746 lr:8.07420532316269e-05 tokens_per_second_per_gpu:3504.6532180677286 mfu:16.532047185123698 +Step 10558 | loss:1.00648832321167 lr:8.073863335022913e-05 tokens_per_second_per_gpu:5329.425335141997 mfu:25.13980859959037 +Step 10559 | loss:1.1235257387161255 lr:8.073521323764218e-05 tokens_per_second_per_gpu:4439.698597294678 mfu:20.94281577412966 +Step 10560 | loss:1.0068485736846924 lr:8.073179289389178e-05 tokens_per_second_per_gpu:4231.7843631771675 mfu:19.962048857971077 +Step 10561 | loss:1.1462708711624146 lr:8.072837231900363e-05 tokens_per_second_per_gpu:4855.424680799273 mfu:22.903866640204768 +Step 10562 | loss:1.1251033544540405 lr:8.072495151300347e-05 tokens_per_second_per_gpu:4423.323907425206 mfu:20.865573568205182 +Step 10563 | loss:1.1015092134475708 lr:8.072153047591703e-05 tokens_per_second_per_gpu:4606.525727064378 mfu:21.729767808896753 +Step 10564 | loss:1.10719633102417 lr:8.071810920777006e-05 tokens_per_second_per_gpu:4424.650272937446 mfu:20.871830260627707 +Step 10565 | loss:1.1126534938812256 lr:8.071468770858825e-05 tokens_per_second_per_gpu:5348.972168488716 mfu:25.232014347520515 +Step 10566 | loss:1.0782041549682617 lr:8.071126597839735e-05 tokens_per_second_per_gpu:5337.472234838916 mfu:25.1777671984034 +Step 10567 | loss:1.27632474899292 lr:8.070784401722311e-05 tokens_per_second_per_gpu:3280.809213239497 mfu:15.476136822623497 +Step 10568 | loss:0.9744827747344971 lr:8.070442182509126e-05 tokens_per_second_per_gpu:5165.697102710588 mfu:24.367474592294407 +Step 10569 | loss:1.0686707496643066 lr:8.070099940202754e-05 tokens_per_second_per_gpu:5022.889188173566 mfu:23.693825293880558 +Step 10570 | loss:0.9559919238090515 lr:8.069757674805767e-05 tokens_per_second_per_gpu:4708.03755451368 mfu:22.208616418678822 +Step 10571 | loss:1.1095293760299683 lr:8.069415386320742e-05 tokens_per_second_per_gpu:4791.9661553022925 mfu:22.604521948294604 +Step 10572 | loss:1.0301942825317383 lr:8.069073074750251e-05 tokens_per_second_per_gpu:3853.5110926347215 mfu:18.177669300747244 +Step 10573 | loss:0.9947419762611389 lr:8.06873074009687e-05 tokens_per_second_per_gpu:3504.2133401246224 mfu:16.529972205815145 +Step 10574 | loss:1.1788572072982788 lr:8.068388382363174e-05 tokens_per_second_per_gpu:5365.400206564382 mfu:25.309508206035755 +Step 10575 | loss:1.0394408702850342 lr:8.068046001551738e-05 tokens_per_second_per_gpu:5356.839293986035 mfu:25.26912492076139 +Step 10576 | loss:1.0620211362838745 lr:8.067703597665137e-05 tokens_per_second_per_gpu:4527.974018139775 mfu:21.359225995595853 +Step 10577 | loss:1.2731915712356567 lr:8.067361170705944e-05 tokens_per_second_per_gpu:5364.459982456487 mfu:25.305073008500038 +Step 10578 | loss:1.1717900037765503 lr:8.067018720676738e-05 tokens_per_second_per_gpu:4375.27896923798 mfu:20.6389373974643 +Step 10579 | loss:1.1697622537612915 lr:8.066676247580091e-05 tokens_per_second_per_gpu:4560.726568399637 mfu:21.51372536333293 +Step 10580 | loss:1.0791858434677124 lr:8.066333751418583e-05 tokens_per_second_per_gpu:3528.330928956427 mfu:16.643738987219738 +Step 10581 | loss:0.9837260246276855 lr:8.065991232194785e-05 tokens_per_second_per_gpu:5323.5889144595485 mfu:25.112277207435213 +Step 10582 | loss:1.1139549016952515 lr:8.065648689911278e-05 tokens_per_second_per_gpu:4925.081746598784 mfu:23.232450904302002 +Step 10583 | loss:1.224839687347412 lr:8.065306124570634e-05 tokens_per_second_per_gpu:3962.9630985886606 mfu:18.693973087269836 +Step 10584 | loss:1.0634527206420898 lr:8.064963536175432e-05 tokens_per_second_per_gpu:4555.902265178902 mfu:21.49096830192972 +Step 10585 | loss:1.0944300889968872 lr:8.06462092472825e-05 tokens_per_second_per_gpu:5327.469292066161 mfu:25.130581610666294 +Step 10586 | loss:1.0507597923278809 lr:8.06427829023166e-05 tokens_per_second_per_gpu:4448.192978614224 mfu:20.98288522010486 +Step 10587 | loss:1.0665005445480347 lr:8.063935632688244e-05 tokens_per_second_per_gpu:5363.4591400537265 mfu:25.300351864124895 +Step 10588 | loss:1.1830527782440186 lr:8.063592952100576e-05 tokens_per_second_per_gpu:3749.6174378742553 mfu:17.687585205156765 +Step 10589 | loss:1.0469783544540405 lr:8.063250248471236e-05 tokens_per_second_per_gpu:5318.171570055408 mfu:25.086722669586614 +Step 10590 | loss:0.9986142516136169 lr:8.062907521802799e-05 tokens_per_second_per_gpu:5340.454019347922 mfu:25.19183278467728 +Step 10591 | loss:1.0837044715881348 lr:8.062564772097845e-05 tokens_per_second_per_gpu:3136.9404657420246 mfu:14.797483394138439 +Step 10592 | loss:1.0453405380249023 lr:8.062221999358948e-05 tokens_per_second_per_gpu:3933.9841445056445 mfu:18.557274416540842 +Step 10593 | loss:0.9521408081054688 lr:8.061879203588692e-05 tokens_per_second_per_gpu:4398.43171719637 mfu:20.74815285985017 +Step 10594 | loss:1.1314066648483276 lr:8.061536384789649e-05 tokens_per_second_per_gpu:3490.9745123337116 mfu:16.467522396348482 +Step 10595 | loss:1.1583824157714844 lr:8.0611935429644e-05 tokens_per_second_per_gpu:3538.328765518238 mfu:16.690900488088616 +Step 10596 | loss:1.1305747032165527 lr:8.060850678115526e-05 tokens_per_second_per_gpu:4212.06458710603 mfu:19.869027309702783 +Step 10597 | loss:1.0810792446136475 lr:8.060507790245602e-05 tokens_per_second_per_gpu:5044.590016337869 mfu:23.796191802874656 +Step 10598 | loss:1.199082374572754 lr:8.060164879357209e-05 tokens_per_second_per_gpu:5131.434809487163 mfu:24.205853509409494 +Step 10599 | loss:1.1388980150222778 lr:8.059821945452923e-05 tokens_per_second_per_gpu:4541.014146115504 mfu:21.420738504132625 +Step 10600 | loss:1.066400170326233 lr:8.05947898853533e-05 tokens_per_second_per_gpu:4864.671121134165 mfu:22.947483676869858 +Step 10601 | loss:1.1227290630340576 lr:8.059136008607003e-05 tokens_per_second_per_gpu:5246.661389136197 mfu:24.749396945295715 +Step 10602 | loss:1.05572509765625 lr:8.058793005670521e-05 tokens_per_second_per_gpu:5035.3896701202275 mfu:23.75279220002511 +Step 10603 | loss:1.103208065032959 lr:8.058449979728471e-05 tokens_per_second_per_gpu:5076.430915007941 mfu:23.94639075451889 +Step 10604 | loss:1.0568747520446777 lr:8.058106930783427e-05 tokens_per_second_per_gpu:4582.812758919294 mfu:21.617909692306828 +Step 10605 | loss:1.0746270418167114 lr:8.05776385883797e-05 tokens_per_second_per_gpu:4828.64955593963 mfu:22.777563807934463 +Step 10606 | loss:1.0224018096923828 lr:8.057420763894681e-05 tokens_per_second_per_gpu:4770.762268866787 mfu:22.504499598222246 +Step 10607 | loss:1.1385064125061035 lr:8.05707764595614e-05 tokens_per_second_per_gpu:5328.546408047679 mfu:25.135662550526046 +Step 10608 | loss:1.076441764831543 lr:8.056734505024928e-05 tokens_per_second_per_gpu:5343.916102133507 mfu:25.208164019869116 +Step 10609 | loss:1.0768741369247437 lr:8.056391341103627e-05 tokens_per_second_per_gpu:5341.135666858398 mfu:25.195048232285252 +Step 10610 | loss:1.2516322135925293 lr:8.056048154194815e-05 tokens_per_second_per_gpu:4302.377871673641 mfu:20.29505048204348 +Step 10611 | loss:1.1628468036651611 lr:8.055704944301077e-05 tokens_per_second_per_gpu:3605.4925922164603 mfu:17.00772371795461 +Step 10612 | loss:0.9807450175285339 lr:8.055361711424989e-05 tokens_per_second_per_gpu:4021.1228475611238 mfu:18.968322546249354 +Step 10613 | loss:1.1340866088867188 lr:8.055018455569139e-05 tokens_per_second_per_gpu:3285.6393958612034 mfu:15.498921618164223 +Step 10614 | loss:1.2227500677108765 lr:8.054675176736103e-05 tokens_per_second_per_gpu:4867.018257307508 mfu:22.958555518662237 +Step 10615 | loss:0.9711372256278992 lr:8.054331874928468e-05 tokens_per_second_per_gpu:4803.936389749746 mfu:22.660987586515837 +Step 10616 | loss:0.9938088655471802 lr:8.05398855014881e-05 tokens_per_second_per_gpu:5345.137692422155 mfu:25.213926469685465 +Step 10617 | loss:1.1557241678237915 lr:8.053645202399716e-05 tokens_per_second_per_gpu:4530.665137367382 mfu:21.371920464144765 +Step 10618 | loss:1.0417706966400146 lr:8.053301831683767e-05 tokens_per_second_per_gpu:5363.918005840679 mfu:25.302516412332526 +Step 10619 | loss:1.1666043996810913 lr:8.052958438003544e-05 tokens_per_second_per_gpu:5309.80045696163 mfu:25.04723470086409 +Step 10620 | loss:1.0153753757476807 lr:8.052615021361632e-05 tokens_per_second_per_gpu:5350.639217128249 mfu:25.239878100381016 +Step 10621 | loss:1.1802878379821777 lr:8.052271581760613e-05 tokens_per_second_per_gpu:3831.778286533891 mfu:18.075151946370546 +Step 10622 | loss:1.1080377101898193 lr:8.05192811920307e-05 tokens_per_second_per_gpu:5317.536754051141 mfu:25.08372813418371 +Step 10623 | loss:1.1665256023406982 lr:8.051584633691586e-05 tokens_per_second_per_gpu:4932.925036966983 mfu:23.26944904317212 +Step 10624 | loss:0.9576645493507385 lr:8.051241125228745e-05 tokens_per_second_per_gpu:4960.701605286853 mfu:23.400475855108194 +Step 10625 | loss:1.1133027076721191 lr:8.05089759381713e-05 tokens_per_second_per_gpu:5363.866779918024 mfu:25.302274770915318 +Step 10626 | loss:0.9762457013130188 lr:8.050554039459324e-05 tokens_per_second_per_gpu:4969.097967546851 mfu:23.440082928455723 +Step 10627 | loss:0.9895934462547302 lr:8.050210462157912e-05 tokens_per_second_per_gpu:4611.415770029095 mfu:21.752834975888682 +Step 10628 | loss:0.9798979163169861 lr:8.049866861915478e-05 tokens_per_second_per_gpu:4406.1700188425575 mfu:20.784655749005637 +Step 10629 | loss:1.1437960863113403 lr:8.049523238734607e-05 tokens_per_second_per_gpu:5306.750660313258 mfu:25.032848289724758 +Step 10630 | loss:1.0016111135482788 lr:8.049179592617882e-05 tokens_per_second_per_gpu:4572.385745467308 mfu:21.56872369082187 +Step 10631 | loss:1.070704460144043 lr:8.048835923567888e-05 tokens_per_second_per_gpu:3528.3240235696467 mfu:16.643706413332115 +Step 10632 | loss:1.055217981338501 lr:8.04849223158721e-05 tokens_per_second_per_gpu:4176.880907807053 mfu:19.703059891494814 +Step 10633 | loss:1.1103508472442627 lr:8.048148516678433e-05 tokens_per_second_per_gpu:4541.4850199384655 mfu:21.422959696294892 +Step 10634 | loss:1.198732852935791 lr:8.047804778844144e-05 tokens_per_second_per_gpu:5193.080056203772 mfu:24.496644655935384 +Step 10635 | loss:1.0974419116973877 lr:8.047461018086924e-05 tokens_per_second_per_gpu:4885.041526477221 mfu:23.04357435442259 +Step 10636 | loss:1.2913872003555298 lr:8.047117234409362e-05 tokens_per_second_per_gpu:5317.635522221827 mfu:25.084194040496424 +Step 10637 | loss:1.1605939865112305 lr:8.046773427814042e-05 tokens_per_second_per_gpu:4984.443457364254 mfu:23.51247022213496 +Step 10638 | loss:1.1495181322097778 lr:8.046429598303549e-05 tokens_per_second_per_gpu:4614.636993434011 mfu:21.76803003628706 +Step 10639 | loss:1.0170565843582153 lr:8.046085745880472e-05 tokens_per_second_per_gpu:5334.961155061437 mfu:25.165922006658644 +Step 10640 | loss:1.202561378479004 lr:8.045741870547396e-05 tokens_per_second_per_gpu:5299.233837246033 mfu:24.997390152814393 +Step 10641 | loss:1.2475508451461792 lr:8.045397972306905e-05 tokens_per_second_per_gpu:5347.671990707445 mfu:25.22588118709305 +Step 10642 | loss:1.1502765417099 lr:8.045054051161587e-05 tokens_per_second_per_gpu:3898.838341386305 mfu:18.391485666734564 +Step 10643 | loss:1.061976671218872 lr:8.04471010711403e-05 tokens_per_second_per_gpu:5329.930615137136 mfu:25.1421920915371 +Step 10644 | loss:1.1526836156845093 lr:8.04436614016682e-05 tokens_per_second_per_gpu:5299.717902385963 mfu:24.999673570669472 +Step 10645 | loss:1.087306261062622 lr:8.044022150322544e-05 tokens_per_second_per_gpu:3490.8850265135557 mfu:16.46710027646695 +Step 10646 | loss:1.1381522417068481 lr:8.043678137583788e-05 tokens_per_second_per_gpu:4246.362979640013 mfu:20.030818679194773 +Step 10647 | loss:1.0196375846862793 lr:8.043334101953141e-05 tokens_per_second_per_gpu:4252.741727642389 mfu:20.06090832184862 +Step 10648 | loss:1.2103272676467896 lr:8.04299004343319e-05 tokens_per_second_per_gpu:3694.6371684357673 mfu:17.428233893614156 +Step 10649 | loss:1.046242356300354 lr:8.04264596202652e-05 tokens_per_second_per_gpu:4621.202654237777 mfu:21.79900137851567 +Step 10650 | loss:1.1599212884902954 lr:8.042301857735726e-05 tokens_per_second_per_gpu:4219.841425036816 mfu:19.905711981087673 +Step 10651 | loss:1.0021495819091797 lr:8.041957730563388e-05 tokens_per_second_per_gpu:4843.198225621401 mfu:22.84619236508239 +Step 10652 | loss:1.0622963905334473 lr:8.0416135805121e-05 tokens_per_second_per_gpu:4708.485105720915 mfu:22.210727594083462 +Step 10653 | loss:1.1896640062332153 lr:8.041269407584446e-05 tokens_per_second_per_gpu:5333.262070021904 mfu:25.157907132633365 +Step 10654 | loss:1.1693040132522583 lr:8.040925211783019e-05 tokens_per_second_per_gpu:5078.528382406805 mfu:23.95628486610417 +Step 10655 | loss:0.9292709827423096 lr:8.040580993110404e-05 tokens_per_second_per_gpu:4542.209396427682 mfu:21.4263767037859 +Step 10656 | loss:1.1883182525634766 lr:8.040236751569192e-05 tokens_per_second_per_gpu:4057.1282976730945 mfu:19.138166397590773 +Step 10657 | loss:1.0149097442626953 lr:8.03989248716197e-05 tokens_per_second_per_gpu:4274.743654029123 mfu:20.164695162530577 +Step 10658 | loss:0.8783385157585144 lr:8.039548199891331e-05 tokens_per_second_per_gpu:5351.195718552367 mfu:25.242503212547515 +Step 10659 | loss:1.1007845401763916 lr:8.039203889759861e-05 tokens_per_second_per_gpu:4034.704763454898 mfu:19.032390760834094 +Step 10660 | loss:1.2042709589004517 lr:8.038859556770151e-05 tokens_per_second_per_gpu:4299.643920759186 mfu:20.28215396911987 +Step 10661 | loss:0.9154349565505981 lr:8.03851520092479e-05 tokens_per_second_per_gpu:5127.414137981925 mfu:24.186887315921986 +Step 10662 | loss:1.1719741821289062 lr:8.038170822226369e-05 tokens_per_second_per_gpu:4614.6909731180185 mfu:21.768284667666407 +Step 10663 | loss:0.9676084518432617 lr:8.037826420677477e-05 tokens_per_second_per_gpu:5311.089209122525 mfu:25.053313964690826 +Step 10664 | loss:1.1228487491607666 lr:8.037481996280704e-05 tokens_per_second_per_gpu:5096.5156368878115 mfu:24.041133814433156 +Step 10665 | loss:1.1123493909835815 lr:8.037137549038643e-05 tokens_per_second_per_gpu:5326.170490912054 mfu:25.124454944024045 +Step 10666 | loss:1.09767484664917 lr:8.036793078953883e-05 tokens_per_second_per_gpu:3231.8774275647156 mfu:15.245317240972035 +Step 10667 | loss:1.138563632965088 lr:8.036448586029013e-05 tokens_per_second_per_gpu:4620.310994944105 mfu:21.794795269494657 +Step 10668 | loss:0.9459945559501648 lr:8.036104070266628e-05 tokens_per_second_per_gpu:4414.976997288207 mfu:20.826199768959235 +Step 10669 | loss:0.9996180534362793 lr:8.035759531669316e-05 tokens_per_second_per_gpu:5326.8354998031855 mfu:25.127591904425728 +Step 10670 | loss:1.0474677085876465 lr:8.035414970239669e-05 tokens_per_second_per_gpu:5340.507679980473 mfu:25.192085911036468 +Step 10671 | loss:1.060189127922058 lr:8.035070385980277e-05 tokens_per_second_per_gpu:5352.328378453539 mfu:25.247846162553053 +Step 10672 | loss:1.1455157995224 lr:8.034725778893737e-05 tokens_per_second_per_gpu:5241.855865096771 mfu:24.726728468494795 +Step 10673 | loss:1.0594669580459595 lr:8.034381148982634e-05 tokens_per_second_per_gpu:5326.865203909428 mfu:25.127732023762913 +Step 10674 | loss:1.0693265199661255 lr:8.034036496249564e-05 tokens_per_second_per_gpu:5347.096018536994 mfu:25.22316422809405 +Step 10675 | loss:1.0823583602905273 lr:8.033691820697119e-05 tokens_per_second_per_gpu:3304.133012210935 mfu:15.586159161822222 +Step 10676 | loss:1.094772219657898 lr:8.033347122327891e-05 tokens_per_second_per_gpu:4685.091910779564 mfu:22.100377902254483 +Step 10677 | loss:0.9732436537742615 lr:8.033002401144471e-05 tokens_per_second_per_gpu:5382.225700768135 mfu:25.388876932920187 +Step 10678 | loss:1.0460402965545654 lr:8.032657657149454e-05 tokens_per_second_per_gpu:5346.6382696687115 mfu:25.221004948582355 +Step 10679 | loss:0.9922396540641785 lr:8.032312890345431e-05 tokens_per_second_per_gpu:4976.555027033061 mfu:23.475259150357726 +Step 10680 | loss:1.0583539009094238 lr:8.031968100734997e-05 tokens_per_second_per_gpu:3360.98068047636 mfu:15.854319312242588 +Step 10681 | loss:1.266441822052002 lr:8.031623288320744e-05 tokens_per_second_per_gpu:4081.3260812196936 mfu:19.25231147114768 +Step 10682 | loss:1.1133126020431519 lr:8.031278453105263e-05 tokens_per_second_per_gpu:4387.153058639943 mfu:20.69494950310192 +Step 10683 | loss:1.0054434537887573 lr:8.030933595091152e-05 tokens_per_second_per_gpu:5372.198107570151 mfu:25.34157506492147 +Step 10684 | loss:1.1256028413772583 lr:8.030588714281002e-05 tokens_per_second_per_gpu:4991.493023151421 mfu:23.54572422673298 +Step 10685 | loss:1.123906135559082 lr:8.030243810677408e-05 tokens_per_second_per_gpu:5340.222455583239 mfu:25.190740458890417 +Step 10686 | loss:1.075417399406433 lr:8.029898884282965e-05 tokens_per_second_per_gpu:3309.979722817842 mfu:15.613739093306696 +Step 10687 | loss:1.1372785568237305 lr:8.029553935100264e-05 tokens_per_second_per_gpu:4507.493911709668 mfu:21.262617839298567 +Step 10688 | loss:1.0896925926208496 lr:8.0292089631319e-05 tokens_per_second_per_gpu:3739.3705762011195 mfu:17.639249010349705 +Step 10689 | loss:0.9438596963882446 lr:8.02886396838047e-05 tokens_per_second_per_gpu:5318.427238721736 mfu:25.08792870230726 +Step 10690 | loss:1.2000555992126465 lr:8.028518950848571e-05 tokens_per_second_per_gpu:4213.1941268611745 mfu:19.874355541447155 +Step 10691 | loss:1.0284496545791626 lr:8.028173910538791e-05 tokens_per_second_per_gpu:4012.4177654798823 mfu:18.927259188831982 +Step 10692 | loss:1.088527798652649 lr:8.027828847453729e-05 tokens_per_second_per_gpu:3766.766115297117 mfu:17.768478442426478 +Step 10693 | loss:1.182745337486267 lr:8.02748376159598e-05 tokens_per_second_per_gpu:3908.931907061227 mfu:18.439098738163057 +Step 10694 | loss:1.0315488576889038 lr:8.027138652968141e-05 tokens_per_second_per_gpu:4609.442069724421 mfu:21.743524694803543 +Step 10695 | loss:1.0177773237228394 lr:8.026793521572803e-05 tokens_per_second_per_gpu:2912.212483456328 mfu:13.73740315914927 +Step 10696 | loss:1.066010594367981 lr:8.026448367412567e-05 tokens_per_second_per_gpu:5368.200322211695 mfu:25.32271682929312 +Step 10697 | loss:0.9899595379829407 lr:8.026103190490025e-05 tokens_per_second_per_gpu:4851.924583944355 mfu:22.88735608616202 +Step 10698 | loss:1.101628303527832 lr:8.025757990807776e-05 tokens_per_second_per_gpu:5019.586499175776 mfu:23.678245946381136 +Step 10699 | loss:1.026221752166748 lr:8.025412768368415e-05 tokens_per_second_per_gpu:5042.347300648899 mfu:23.785612530323032 +Step 10700 | loss:0.9421087503433228 lr:8.025067523174537e-05 tokens_per_second_per_gpu:5288.029018602568 mfu:24.944535111533977 +Step 10701 | loss:0.9717808961868286 lr:8.024722255228742e-05 tokens_per_second_per_gpu:5042.701215066778 mfu:23.78728200501619 +Step 10702 | loss:1.1248235702514648 lr:8.024376964533624e-05 tokens_per_second_per_gpu:5093.803897381711 mfu:24.02834207651241 +Step 10703 | loss:1.013370156288147 lr:8.024031651091781e-05 tokens_per_second_per_gpu:5334.790399559111 mfu:25.165116523820284 +Step 10704 | loss:1.0541788339614868 lr:8.023686314905811e-05 tokens_per_second_per_gpu:4079.324515043224 mfu:19.242869741990866 +Step 10705 | loss:0.9700561761856079 lr:8.023340955978309e-05 tokens_per_second_per_gpu:5317.248782068136 mfu:25.082369721206792 +Step 10706 | loss:1.0714186429977417 lr:8.022995574311876e-05 tokens_per_second_per_gpu:3373.6321655170477 mfu:15.913998525745447 +Step 10707 | loss:1.2216684818267822 lr:8.022650169909104e-05 tokens_per_second_per_gpu:5344.14902534912 mfu:25.209262758417967 +Step 10708 | loss:1.070469617843628 lr:8.022304742772597e-05 tokens_per_second_per_gpu:5361.765716847016 mfu:25.292363697931048 +Step 10709 | loss:1.104137897491455 lr:8.02195929290495e-05 tokens_per_second_per_gpu:3430.07999124927 mfu:16.180272550716523 +Step 10710 | loss:0.9340886473655701 lr:8.021613820308762e-05 tokens_per_second_per_gpu:4574.299194735573 mfu:21.577749757488384 +Step 10711 | loss:1.1926151514053345 lr:8.021268324986631e-05 tokens_per_second_per_gpu:5303.730584837764 mfu:25.01860207842884 +Step 10712 | loss:1.0510694980621338 lr:8.020922806941153e-05 tokens_per_second_per_gpu:3963.587003771828 mfu:18.696916154468042 +Step 10713 | loss:0.9757986664772034 lr:8.020577266174932e-05 tokens_per_second_per_gpu:4539.718131354271 mfu:21.414624981380967 +Step 10714 | loss:1.0142884254455566 lr:8.020231702690562e-05 tokens_per_second_per_gpu:5357.430166479773 mfu:25.271912167127592 +Step 10715 | loss:0.99904865026474 lr:8.019886116490644e-05 tokens_per_second_per_gpu:4284.740461596852 mfu:20.211851809458803 +Step 10716 | loss:1.1056101322174072 lr:8.01954050757778e-05 tokens_per_second_per_gpu:5320.961304530861 mfu:25.099882323085108 +Step 10717 | loss:1.1080710887908936 lr:8.019194875954564e-05 tokens_per_second_per_gpu:5290.337428775732 mfu:24.95542427617618 +Step 10718 | loss:1.1102406978607178 lr:8.018849221623599e-05 tokens_per_second_per_gpu:3642.7562536774476 mfu:17.183502766902336 +Step 10719 | loss:1.1519173383712769 lr:8.018503544587484e-05 tokens_per_second_per_gpu:4381.340812764176 mfu:20.66753214763452 +Step 10720 | loss:1.0983648300170898 lr:8.018157844848818e-05 tokens_per_second_per_gpu:3395.06006729662 mfu:16.015077594416727 +Step 10721 | loss:0.9345664381980896 lr:8.0178121224102e-05 tokens_per_second_per_gpu:5369.83483521058 mfu:25.330427105985535 +Step 10722 | loss:1.1473084688186646 lr:8.017466377274235e-05 tokens_per_second_per_gpu:5341.869585195603 mfu:25.19851025030123 +Step 10723 | loss:1.0229836702346802 lr:8.017120609443518e-05 tokens_per_second_per_gpu:5196.394125316189 mfu:24.512277685377153 +Step 10724 | loss:1.079205870628357 lr:8.016774818920653e-05 tokens_per_second_per_gpu:5062.034116828337 mfu:23.878478601158637 +Step 10725 | loss:1.16081702709198 lr:8.01642900570824e-05 tokens_per_second_per_gpu:4059.2755200439583 mfu:19.148295211868053 +Step 10726 | loss:0.9223132133483887 lr:8.01608316980888e-05 tokens_per_second_per_gpu:4828.0888353720175 mfu:22.77491879334824 +Step 10727 | loss:1.0797146558761597 lr:8.015737311225172e-05 tokens_per_second_per_gpu:5333.607711405162 mfu:25.159537582010696 +Step 10728 | loss:1.1539926528930664 lr:8.01539142995972e-05 tokens_per_second_per_gpu:4641.726505842321 mfu:21.895815888264472 +Step 10729 | loss:1.1766095161437988 lr:8.015045526015123e-05 tokens_per_second_per_gpu:3814.4796560490754 mfu:17.993551354922968 +Step 10730 | loss:1.0978658199310303 lr:8.014699599393987e-05 tokens_per_second_per_gpu:5340.169995379218 mfu:25.190492995157417 +Step 10731 | loss:1.0862318277359009 lr:8.014353650098907e-05 tokens_per_second_per_gpu:5209.7254732243555 mfu:24.575163928022473 +Step 10732 | loss:1.1457791328430176 lr:8.01400767813249e-05 tokens_per_second_per_gpu:4206.5792744222545 mfu:19.843152153882563 +Step 10733 | loss:1.0851731300354004 lr:8.013661683497336e-05 tokens_per_second_per_gpu:5147.5200651412715 mfu:24.281730404757532 +Step 10734 | loss:1.0642691850662231 lr:8.01331566619605e-05 tokens_per_second_per_gpu:5364.659129789817 mfu:25.30601242044939 +Step 10735 | loss:1.005807638168335 lr:8.012969626231231e-05 tokens_per_second_per_gpu:5336.127858007833 mfu:25.171425543516847 +Step 10736 | loss:1.201593041419983 lr:8.012623563605483e-05 tokens_per_second_per_gpu:4375.098501118379 mfu:20.638086098553106 +Step 10737 | loss:1.1698291301727295 lr:8.012277478321408e-05 tokens_per_second_per_gpu:3487.2880560967815 mfu:16.45013275330459 +Step 10738 | loss:0.9845372438430786 lr:8.011931370381611e-05 tokens_per_second_per_gpu:5362.3899800191575 mfu:25.29530845382432 +Step 10739 | loss:1.0667698383331299 lr:8.011585239788692e-05 tokens_per_second_per_gpu:4817.860134747382 mfu:22.726668267305552 +Step 10740 | loss:1.1728177070617676 lr:8.011239086545259e-05 tokens_per_second_per_gpu:5342.290236161557 mfu:25.200494532678093 +Step 10741 | loss:0.8983042240142822 lr:8.010892910653911e-05 tokens_per_second_per_gpu:4364.059467547201 mfu:20.58601309374452 +Step 10742 | loss:1.1268808841705322 lr:8.010546712117253e-05 tokens_per_second_per_gpu:4138.113131535026 mfu:19.52018567635497 +Step 10743 | loss:1.0205998420715332 lr:8.010200490937889e-05 tokens_per_second_per_gpu:5311.666153559064 mfu:25.056035510034853 +Step 10744 | loss:0.9195539951324463 lr:8.009854247118424e-05 tokens_per_second_per_gpu:5326.402050451267 mfu:25.12554724987864 +Step 10745 | loss:1.1741771697998047 lr:8.00950798066146e-05 tokens_per_second_per_gpu:4351.265156508448 mfu:20.525660145640156 +Step 10746 | loss:1.1125948429107666 lr:8.009161691569602e-05 tokens_per_second_per_gpu:4826.801591376948 mfu:22.768846643797023 +Step 10747 | loss:0.9943527579307556 lr:8.008815379845456e-05 tokens_per_second_per_gpu:4818.9291767239565 mfu:22.73171112070642 +Step 10748 | loss:1.2840319871902466 lr:8.008469045491626e-05 tokens_per_second_per_gpu:3715.993788432848 mfu:17.528976713955494 +Step 10749 | loss:1.1044931411743164 lr:8.008122688510716e-05 tokens_per_second_per_gpu:4495.829824579041 mfu:21.207596350204923 +Step 10750 | loss:1.0819838047027588 lr:8.007776308905332e-05 tokens_per_second_per_gpu:4351.655577069126 mfu:20.527501826037494 +Step 10751 | loss:1.0387027263641357 lr:8.007429906678077e-05 tokens_per_second_per_gpu:5346.506687162479 mfu:25.22038425145016 +Step 10752 | loss:1.1260359287261963 lr:8.00708348183156e-05 tokens_per_second_per_gpu:3985.8553361498434 mfu:18.80195967766442 +Step 10753 | loss:1.0309479236602783 lr:8.006737034368384e-05 tokens_per_second_per_gpu:4466.405597193772 mfu:21.068797249337738 +Step 10754 | loss:1.2250558137893677 lr:8.006390564291153e-05 tokens_per_second_per_gpu:3392.8351218215075 mfu:16.004582146995777 +Step 10755 | loss:1.1254479885101318 lr:8.006044071602477e-05 tokens_per_second_per_gpu:3612.8797885681447 mfu:17.042570383531174 +Step 10756 | loss:1.1148324012756348 lr:8.005697556304958e-05 tokens_per_second_per_gpu:4128.914675402285 mfu:19.476794989358382 +Step 10757 | loss:1.0574816465377808 lr:8.005351018401206e-05 tokens_per_second_per_gpu:4102.012055466829 mfu:19.349890741062563 +Step 10758 | loss:1.1123117208480835 lr:8.005004457893826e-05 tokens_per_second_per_gpu:4101.424290263128 mfu:19.34711815231362 +Step 10759 | loss:1.0723892450332642 lr:8.004657874785422e-05 tokens_per_second_per_gpu:5308.030481644934 mfu:25.038885425306308 +Step 10760 | loss:1.085057020187378 lr:8.004311269078604e-05 tokens_per_second_per_gpu:3406.3453656407733 mfu:16.06831227217654 +Step 10761 | loss:1.0124590396881104 lr:8.003964640775975e-05 tokens_per_second_per_gpu:5347.866064977826 mfu:25.226796668538924 +Step 10762 | loss:1.053076982498169 lr:8.00361798988015e-05 tokens_per_second_per_gpu:4735.385581987346 mfu:22.33762173032674 +Step 10763 | loss:1.035122275352478 lr:8.003271316393727e-05 tokens_per_second_per_gpu:3490.5933537264596 mfu:16.465724406165712 +Step 10764 | loss:0.9753295183181763 lr:8.002924620319317e-05 tokens_per_second_per_gpu:3943.7458494183866 mfu:18.603322044132845 +Step 10765 | loss:1.02468740940094 lr:8.00257790165953e-05 tokens_per_second_per_gpu:4016.774026472838 mfu:18.947808415190522 +Step 10766 | loss:1.0605418682098389 lr:8.002231160416969e-05 tokens_per_second_per_gpu:4982.181946940083 mfu:23.50180229161068 +Step 10767 | loss:1.0060168504714966 lr:8.001884396594245e-05 tokens_per_second_per_gpu:3942.2738209982735 mfu:18.596378234921225 +Step 10768 | loss:1.1007298231124878 lr:8.001537610193967e-05 tokens_per_second_per_gpu:4526.294589931966 mfu:21.35130384619962 +Step 10769 | loss:1.0803608894348145 lr:8.00119080121874e-05 tokens_per_second_per_gpu:5222.977941427456 mfu:24.637678081640733 +Step 10770 | loss:1.0333917140960693 lr:8.000843969671175e-05 tokens_per_second_per_gpu:4279.237179383011 mfu:20.18589188829935 +Step 10771 | loss:1.1049309968948364 lr:8.000497115553879e-05 tokens_per_second_per_gpu:4831.934263217474 mfu:22.793058332592906 +Step 10772 | loss:1.1726369857788086 lr:8.000150238869462e-05 tokens_per_second_per_gpu:3951.881552311495 mfu:18.641699542773686 +Step 10773 | loss:1.1867506504058838 lr:7.999803339620532e-05 tokens_per_second_per_gpu:5341.797615389859 mfu:25.19817075644822 +Step 10774 | loss:1.0461134910583496 lr:7.999456417809697e-05 tokens_per_second_per_gpu:4979.933316470512 mfu:23.491195118029797 +Step 10775 | loss:1.0508688688278198 lr:7.999109473439569e-05 tokens_per_second_per_gpu:5357.088256952547 mfu:25.270299321551892 +Step 10776 | loss:1.2352219820022583 lr:7.998762506512755e-05 tokens_per_second_per_gpu:4367.471831543637 mfu:20.602109796925035 +Step 10777 | loss:0.9618704319000244 lr:7.998415517031868e-05 tokens_per_second_per_gpu:4007.94182611686 mfu:18.906145419182604 +Step 10778 | loss:1.1997783184051514 lr:7.998068504999511e-05 tokens_per_second_per_gpu:5354.944229973735 mfu:25.260185580484105 +Step 10779 | loss:1.038629174232483 lr:7.997721470418302e-05 tokens_per_second_per_gpu:3390.5816639854716 mfu:15.993952201903642 +Step 10780 | loss:1.1023023128509521 lr:7.997374413290845e-05 tokens_per_second_per_gpu:4374.33840699607 mfu:20.634500604892743 +Step 10781 | loss:0.8992505669593811 lr:7.997027333619754e-05 tokens_per_second_per_gpu:5362.3065639282195 mfu:25.294914965891113 +Step 10782 | loss:1.0981402397155762 lr:7.996680231407637e-05 tokens_per_second_per_gpu:3589.8798197008546 mfu:16.934075606184827 +Step 10783 | loss:0.9534328579902649 lr:7.996333106657106e-05 tokens_per_second_per_gpu:3853.791553843598 mfu:18.17899228412113 +Step 10784 | loss:0.9680566787719727 lr:7.995985959370774e-05 tokens_per_second_per_gpu:4460.102471413335 mfu:21.039064329607264 +Step 10785 | loss:1.02451491355896 lr:7.995638789551245e-05 tokens_per_second_per_gpu:5339.64646944784 mfu:25.188023434016372 +Step 10786 | loss:1.093955636024475 lr:7.995291597201136e-05 tokens_per_second_per_gpu:4687.016023634297 mfu:22.109454270877762 +Step 10787 | loss:1.0642117261886597 lr:7.994944382323057e-05 tokens_per_second_per_gpu:4235.707819348638 mfu:19.980556470142552 +Step 10788 | loss:1.1147273778915405 lr:7.99459714491962e-05 tokens_per_second_per_gpu:4753.090849764368 mfu:22.42114049926113 +Step 10789 | loss:1.0977331399917603 lr:7.994249884993434e-05 tokens_per_second_per_gpu:5311.752978204342 mfu:25.05644507670044 +Step 10790 | loss:1.000711441040039 lr:7.993902602547113e-05 tokens_per_second_per_gpu:4679.274324658992 mfu:22.072935355941198 +Step 10791 | loss:1.0698670148849487 lr:7.99355529758327e-05 tokens_per_second_per_gpu:5295.611747340938 mfu:24.980304136739797 +Step 10792 | loss:1.0959163904190063 lr:7.993207970104514e-05 tokens_per_second_per_gpu:4053.397635721144 mfu:19.12056822864686 +Step 10793 | loss:1.0401853322982788 lr:7.992860620113459e-05 tokens_per_second_per_gpu:5181.818184482617 mfu:24.443520485553243 +Step 10794 | loss:1.0662330389022827 lr:7.992513247612719e-05 tokens_per_second_per_gpu:4035.657191246545 mfu:19.03688352522823 +Step 10795 | loss:1.098085641860962 lr:7.992165852604902e-05 tokens_per_second_per_gpu:4426.6700936946645 mfu:20.881358099756465 +Step 10796 | loss:1.1618417501449585 lr:7.991818435092626e-05 tokens_per_second_per_gpu:4891.125217267167 mfu:23.07227216186344 +Step 10797 | loss:1.2298794984817505 lr:7.991470995078502e-05 tokens_per_second_per_gpu:4556.304827307849 mfu:21.492867256176147 +Step 10798 | loss:1.1119343042373657 lr:7.991123532565141e-05 tokens_per_second_per_gpu:5342.390896206422 mfu:25.2009693632841 +Step 10799 | loss:0.9852136969566345 lr:7.990776047555161e-05 tokens_per_second_per_gpu:3486.1423362347314 mfu:16.444728197235335 +Step 10800 | loss:1.0981014966964722 lr:7.990428540051171e-05 tokens_per_second_per_gpu:4144.354604146485 mfu:19.54962776756826 +Step 10801 | loss:1.1286125183105469 lr:7.990081010055786e-05 tokens_per_second_per_gpu:5085.556195700119 mfu:23.989436260477326 +Step 10802 | loss:0.9773209095001221 lr:7.989733457571622e-05 tokens_per_second_per_gpu:5080.807073931822 mfu:23.96703384281254 +Step 10803 | loss:0.9758606553077698 lr:7.98938588260129e-05 tokens_per_second_per_gpu:4874.710159695953 mfu:22.994839534602406 +Step 10804 | loss:1.0098519325256348 lr:7.989038285147406e-05 tokens_per_second_per_gpu:4980.240707332938 mfu:23.492645133174147 +Step 10805 | loss:1.0352087020874023 lr:7.988690665212581e-05 tokens_per_second_per_gpu:4674.465673362433 mfu:22.05025212733482 +Step 10806 | loss:1.0945690870285034 lr:7.988343022799436e-05 tokens_per_second_per_gpu:4892.6695952031205 mfu:23.079557256085383 +Step 10807 | loss:1.1503393650054932 lr:7.987995357910579e-05 tokens_per_second_per_gpu:3795.7499049398234 mfu:17.905199923315564 +Step 10808 | loss:1.111899733543396 lr:7.987647670548628e-05 tokens_per_second_per_gpu:4426.951416563945 mfu:20.882685147729493 +Step 10809 | loss:1.142806053161621 lr:7.987299960716199e-05 tokens_per_second_per_gpu:5303.689806357869 mfu:25.018409719381655 +Step 10810 | loss:1.1534823179244995 lr:7.986952228415905e-05 tokens_per_second_per_gpu:3708.9201261725925 mfu:17.495609042182636 +Step 10811 | loss:1.0753841400146484 lr:7.986604473650362e-05 tokens_per_second_per_gpu:2989.533533231679 mfu:14.102139743271044 +Step 10812 | loss:1.2129416465759277 lr:7.986256696422186e-05 tokens_per_second_per_gpu:4570.063210736125 mfu:21.557767898229592 +Step 10813 | loss:1.1868979930877686 lr:7.985908896733993e-05 tokens_per_second_per_gpu:4100.928971332654 mfu:19.344781648408453 +Step 10814 | loss:1.1215360164642334 lr:7.985561074588397e-05 tokens_per_second_per_gpu:3419.127461182924 mfu:16.128607597699247 +Step 10815 | loss:1.1914817094802856 lr:7.985213229988015e-05 tokens_per_second_per_gpu:5398.7236559839685 mfu:25.46670060251469 +Step 10816 | loss:1.1726243495941162 lr:7.984865362935463e-05 tokens_per_second_per_gpu:4819.8174272848855 mfu:22.735901150153495 +Step 10817 | loss:1.323624849319458 lr:7.984517473433359e-05 tokens_per_second_per_gpu:3622.5962208953297 mfu:17.088404452612806 +Step 10818 | loss:1.0066521167755127 lr:7.984169561484318e-05 tokens_per_second_per_gpu:5361.028135079627 mfu:25.28888439888985 +Step 10819 | loss:1.109083890914917 lr:7.983821627090958e-05 tokens_per_second_per_gpu:3775.567986905198 mfu:17.80999837255595 +Step 10820 | loss:1.254115104675293 lr:7.983473670255894e-05 tokens_per_second_per_gpu:3940.3535942314447 mfu:18.587320197637627 +Step 10821 | loss:1.0731616020202637 lr:7.983125690981744e-05 tokens_per_second_per_gpu:3961.3233825754464 mfu:18.686238267071108 +Step 10822 | loss:1.170750379562378 lr:7.982777689271123e-05 tokens_per_second_per_gpu:5070.659726351949 mfu:23.919167072962257 +Step 10823 | loss:1.1106160879135132 lr:7.982429665126651e-05 tokens_per_second_per_gpu:4446.0169032517 mfu:20.97262030134309 +Step 10824 | loss:1.11275315284729 lr:7.982081618550947e-05 tokens_per_second_per_gpu:5161.111007973172 mfu:24.345841200949565 +Step 10825 | loss:1.117240309715271 lr:7.981733549546626e-05 tokens_per_second_per_gpu:5340.390728456402 mfu:25.191534230743443 +Step 10826 | loss:1.1894207000732422 lr:7.981385458116307e-05 tokens_per_second_per_gpu:5297.937824992802 mfu:24.991276641893414 +Step 10827 | loss:0.9847766160964966 lr:7.981037344262605e-05 tokens_per_second_per_gpu:5332.604928912016 mfu:25.154807285898187 +Step 10828 | loss:1.1534565687179565 lr:7.980689207988143e-05 tokens_per_second_per_gpu:4654.208288566932 mfu:21.954694586988197 +Step 10829 | loss:1.0341955423355103 lr:7.980341049295536e-05 tokens_per_second_per_gpu:4719.523151200743 mfu:22.262795937896566 +Step 10830 | loss:1.0295672416687012 lr:7.979992868187405e-05 tokens_per_second_per_gpu:5034.616830779481 mfu:23.74914658499482 +Step 10831 | loss:1.0660037994384766 lr:7.979644664666366e-05 tokens_per_second_per_gpu:4819.691495154398 mfu:22.73530710680775 +Step 10832 | loss:1.1621896028518677 lr:7.97929643873504e-05 tokens_per_second_per_gpu:3448.4345070840272 mfu:16.266853933512397 +Step 10833 | loss:1.227886438369751 lr:7.978948190396045e-05 tokens_per_second_per_gpu:4181.224206913795 mfu:19.72354797442456 +Step 10834 | loss:1.0943654775619507 lr:7.978599919652001e-05 tokens_per_second_per_gpu:5326.760111959186 mfu:25.12723628710309 +Step 10835 | loss:1.0949866771697998 lr:7.978251626505526e-05 tokens_per_second_per_gpu:5359.93670389232 mfu:25.283735931761957 +Step 10836 | loss:1.2130376100540161 lr:7.977903310959242e-05 tokens_per_second_per_gpu:5046.685319078841 mfu:23.80607570339933 +Step 10837 | loss:0.9534350633621216 lr:7.977554973015766e-05 tokens_per_second_per_gpu:4637.602951623809 mfu:21.876364379464853 +Step 10838 | loss:1.2023417949676514 lr:7.977206612677718e-05 tokens_per_second_per_gpu:4510.992805381365 mfu:21.279122717720778 +Step 10839 | loss:1.1295039653778076 lr:7.976858229947722e-05 tokens_per_second_per_gpu:4868.4078960169245 mfu:22.96511068155137 +Step 10840 | loss:1.080386757850647 lr:7.976509824828393e-05 tokens_per_second_per_gpu:5262.301498015843 mfu:24.823174007358745 +Step 10841 | loss:1.225693941116333 lr:7.976161397322355e-05 tokens_per_second_per_gpu:4500.029604193491 mfu:21.22740742720257 +Step 10842 | loss:1.217545509338379 lr:7.975812947432229e-05 tokens_per_second_per_gpu:5364.423276821834 mfu:25.304899861758756 +Step 10843 | loss:0.943737804889679 lr:7.975464475160631e-05 tokens_per_second_per_gpu:5344.380421593729 mfu:25.210354293984054 +Step 10844 | loss:1.1666537523269653 lr:7.975115980510187e-05 tokens_per_second_per_gpu:4490.225557685074 mfu:21.181160067079546 +Step 10845 | loss:1.3034112453460693 lr:7.974767463483515e-05 tokens_per_second_per_gpu:4319.833474554748 mfu:20.377391539066622 +Step 10846 | loss:0.9846140742301941 lr:7.974418924083238e-05 tokens_per_second_per_gpu:5010.677892121458 mfu:23.63622252694082 +Step 10847 | loss:1.158554196357727 lr:7.974070362311976e-05 tokens_per_second_per_gpu:3991.477407014744 mfu:18.828479945158357 +Step 10848 | loss:0.928735077381134 lr:7.973721778172352e-05 tokens_per_second_per_gpu:5360.484586148239 mfu:25.28632038584148 +Step 10849 | loss:1.0318788290023804 lr:7.973373171666987e-05 tokens_per_second_per_gpu:5035.603854468096 mfu:23.75380254413762 +Step 10850 | loss:1.2006994485855103 lr:7.973024542798502e-05 tokens_per_second_per_gpu:3352.2988064609185 mfu:15.8133654312313 +Step 10851 | loss:1.1801847219467163 lr:7.972675891569521e-05 tokens_per_second_per_gpu:3529.911556973572 mfu:16.65119507925956 +Step 10852 | loss:1.2787775993347168 lr:7.972327217982665e-05 tokens_per_second_per_gpu:4892.709608619367 mfu:23.079746006196814 +Step 10853 | loss:1.0561463832855225 lr:7.971978522040557e-05 tokens_per_second_per_gpu:3938.6708612582206 mfu:18.57938245909908 +Step 10854 | loss:1.1769940853118896 lr:7.971629803745819e-05 tokens_per_second_per_gpu:4681.241887015573 mfu:22.08221668327795 +Step 10855 | loss:0.9133841395378113 lr:7.971281063101075e-05 tokens_per_second_per_gpu:5334.6653542594695 mfu:25.164526663806946 +Step 10856 | loss:1.204659342765808 lr:7.970932300108945e-05 tokens_per_second_per_gpu:4830.311165503916 mfu:22.785401903749023 +Step 10857 | loss:1.2174800634384155 lr:7.970583514772056e-05 tokens_per_second_per_gpu:5079.325947095375 mfu:23.960047114818725 +Step 10858 | loss:1.010385274887085 lr:7.970234707093028e-05 tokens_per_second_per_gpu:5063.7288097849105 mfu:23.88647275697923 +Step 10859 | loss:1.1316858530044556 lr:7.969885877074486e-05 tokens_per_second_per_gpu:4131.396489330954 mfu:19.488502129101033 +Step 10860 | loss:1.0879204273223877 lr:7.969537024719055e-05 tokens_per_second_per_gpu:3829.080964668268 mfu:18.062428218919838 +Step 10861 | loss:1.0314078330993652 lr:7.969188150029356e-05 tokens_per_second_per_gpu:3943.700921444536 mfu:18.60311011121469 +Step 10862 | loss:0.9856798052787781 lr:7.968839253008012e-05 tokens_per_second_per_gpu:3535.5808240580077 mfu:16.67793797937355 +Step 10863 | loss:1.1016757488250732 lr:7.968490333657653e-05 tokens_per_second_per_gpu:3622.029570084947 mfu:17.08573146405963 +Step 10864 | loss:0.9960296154022217 lr:7.968141391980896e-05 tokens_per_second_per_gpu:3134.339669760251 mfu:14.785214995750861 +Step 10865 | loss:1.1694668531417847 lr:7.967792427980371e-05 tokens_per_second_per_gpu:5292.828349383621 mfu:24.96717437367847 +Step 10866 | loss:1.0588276386260986 lr:7.9674434416587e-05 tokens_per_second_per_gpu:4772.3162198906375 mfu:22.511829850333132 +Step 10867 | loss:1.1681841611862183 lr:7.967094433018508e-05 tokens_per_second_per_gpu:4607.460979300456 mfu:21.734179553264816 +Step 10868 | loss:1.2153586149215698 lr:7.96674540206242e-05 tokens_per_second_per_gpu:4251.156678474963 mfu:20.053431374488593 +Step 10869 | loss:1.1680076122283936 lr:7.966396348793062e-05 tokens_per_second_per_gpu:4862.243814818557 mfu:22.93603365883797 +Step 10870 | loss:1.2622768878936768 lr:7.966047273213057e-05 tokens_per_second_per_gpu:4269.251305490443 mfu:20.138786817382332 +Step 10871 | loss:1.000165343284607 lr:7.965698175325033e-05 tokens_per_second_per_gpu:4093.3318349390142 mfu:19.308944659834523 +Step 10872 | loss:1.0166752338409424 lr:7.965349055131614e-05 tokens_per_second_per_gpu:5001.420519845139 mfu:23.592553922442843 +Step 10873 | loss:0.9782384634017944 lr:7.964999912635428e-05 tokens_per_second_per_gpu:4195.989621685445 mfu:19.79319895514197 +Step 10874 | loss:1.1404176950454712 lr:7.964650747839098e-05 tokens_per_second_per_gpu:4446.747543053518 mfu:20.97606685394827 +Step 10875 | loss:1.2544037103652954 lr:7.964301560745251e-05 tokens_per_second_per_gpu:3889.910210551008 mfu:18.34937015028827 +Step 10876 | loss:1.112449288368225 lr:7.963952351356515e-05 tokens_per_second_per_gpu:4058.5241153493257 mfu:19.14475070772061 +Step 10877 | loss:1.0966116189956665 lr:7.963603119675515e-05 tokens_per_second_per_gpu:4653.614738277483 mfu:21.951894709002747 +Step 10878 | loss:1.2742047309875488 lr:7.963253865704879e-05 tokens_per_second_per_gpu:4179.986488362828 mfu:19.717709444843084 +Step 10879 | loss:1.1325197219848633 lr:7.962904589447228e-05 tokens_per_second_per_gpu:5054.736638490637 mfu:23.844055150761516 +Step 10880 | loss:1.043824553489685 lr:7.962555290905197e-05 tokens_per_second_per_gpu:5296.212759858255 mfu:24.98313921532729 +Step 10881 | loss:0.9046157002449036 lr:7.96220597008141e-05 tokens_per_second_per_gpu:5296.116081463132 mfu:24.982683166842463 +Step 10882 | loss:1.018012523651123 lr:7.961856626978492e-05 tokens_per_second_per_gpu:5266.594232018553 mfu:24.843423566065475 +Step 10883 | loss:1.0073528289794922 lr:7.961507261599075e-05 tokens_per_second_per_gpu:3996.297665136667 mfu:18.851217925140308 +Step 10884 | loss:1.12295663356781 lr:7.961157873945781e-05 tokens_per_second_per_gpu:3924.17861077169 mfu:18.51102004092121 +Step 10885 | loss:1.1452633142471313 lr:7.960808464021243e-05 tokens_per_second_per_gpu:4146.945178501221 mfu:19.56184794879711 +Step 10886 | loss:1.0391292572021484 lr:7.960459031828085e-05 tokens_per_second_per_gpu:4408.061731032025 mfu:20.79357927816274 +Step 10887 | loss:1.1050221920013428 lr:7.960109577368938e-05 tokens_per_second_per_gpu:5354.801102678495 mfu:25.259510424612426 +Step 10888 | loss:1.0089813470840454 lr:7.959760100646429e-05 tokens_per_second_per_gpu:4595.153755280275 mfu:21.676124277732676 +Step 10889 | loss:1.1451431512832642 lr:7.959410601663186e-05 tokens_per_second_per_gpu:4232.868042535152 mfu:19.967160758397455 +Step 10890 | loss:1.0465118885040283 lr:7.959061080421839e-05 tokens_per_second_per_gpu:3620.7510988284444 mfu:17.07970069701302 +Step 10891 | loss:0.9903872609138489 lr:7.958711536925016e-05 tokens_per_second_per_gpu:3660.366998355722 mfu:17.266575654252634 +Step 10892 | loss:1.0471588373184204 lr:7.958361971175346e-05 tokens_per_second_per_gpu:3618.072075620449 mfu:17.067063287452413 +Step 10893 | loss:1.1041022539138794 lr:7.958012383175458e-05 tokens_per_second_per_gpu:5336.312343012171 mfu:25.172295790758692 +Step 10894 | loss:1.2028419971466064 lr:7.957662772927981e-05 tokens_per_second_per_gpu:3806.667777709222 mfu:17.956701391949995 +Step 10895 | loss:0.9159339070320129 lr:7.957313140435545e-05 tokens_per_second_per_gpu:5357.310914103499 mfu:25.271349633322156 +Step 10896 | loss:1.0835336446762085 lr:7.956963485700781e-05 tokens_per_second_per_gpu:4673.564250706086 mfu:22.045999962866265 +Step 10897 | loss:1.0223698616027832 lr:7.956613808726316e-05 tokens_per_second_per_gpu:4293.985680164669 mfu:20.25546308284014 +Step 10898 | loss:0.9916409850120544 lr:7.95626410951478e-05 tokens_per_second_per_gpu:4579.989392334923 mfu:21.604591390412338 +Step 10899 | loss:1.2487305402755737 lr:7.955914388068806e-05 tokens_per_second_per_gpu:5361.696927561321 mfu:25.29203920713376 +Step 10900 | loss:1.094421625137329 lr:7.955564644391022e-05 tokens_per_second_per_gpu:4916.3723032723365 mfu:23.191366974146888 +Step 10901 | loss:1.0404402017593384 lr:7.955214878484061e-05 tokens_per_second_per_gpu:5077.416409096915 mfu:23.95103949827921 +Step 10902 | loss:1.1434441804885864 lr:7.95486509035055e-05 tokens_per_second_per_gpu:5362.405447766638 mfu:25.295381417828256 +Step 10903 | loss:1.074378490447998 lr:7.954515279993122e-05 tokens_per_second_per_gpu:5019.665309020087 mfu:23.678617705863093 +Step 10904 | loss:1.0478050708770752 lr:7.954165447414407e-05 tokens_per_second_per_gpu:5129.478107646857 mfu:24.196623412981012 +Step 10905 | loss:0.9960837960243225 lr:7.953815592617038e-05 tokens_per_second_per_gpu:5352.505670111075 mfu:25.248682477550727 +Step 10906 | loss:0.969744861125946 lr:7.953465715603644e-05 tokens_per_second_per_gpu:5021.949761525059 mfu:23.68939385809684 +Step 10907 | loss:1.138630986213684 lr:7.953115816376858e-05 tokens_per_second_per_gpu:4664.282402110108 mfu:22.002215899392365 +Step 10908 | loss:1.0821694135665894 lr:7.952765894939311e-05 tokens_per_second_per_gpu:5387.623902131032 mfu:25.4143411697771 +Step 10909 | loss:1.0052661895751953 lr:7.952415951293634e-05 tokens_per_second_per_gpu:4929.930449270428 mfu:23.255323062078126 +Step 10910 | loss:0.9913446307182312 lr:7.952065985442461e-05 tokens_per_second_per_gpu:5318.291653450762 mfu:25.087289123450844 +Step 10911 | loss:0.996543824672699 lr:7.951715997388424e-05 tokens_per_second_per_gpu:5316.390159072538 mfu:25.078319450040365 +Step 10912 | loss:1.0479779243469238 lr:7.951365987134152e-05 tokens_per_second_per_gpu:4672.047066259833 mfu:22.03884314497493 +Step 10913 | loss:1.3064082860946655 lr:7.951015954682282e-05 tokens_per_second_per_gpu:4166.950751951095 mfu:19.656217652062967 +Step 10914 | loss:1.1037052869796753 lr:7.950665900035443e-05 tokens_per_second_per_gpu:4807.656373253281 mfu:22.678535383396646 +Step 10915 | loss:1.1790506839752197 lr:7.95031582319627e-05 tokens_per_second_per_gpu:3912.7643444068126 mfu:18.457176999003423 +Step 10916 | loss:1.187687635421753 lr:7.949965724167395e-05 tokens_per_second_per_gpu:4142.033253622599 mfu:19.538677561083848 +Step 10917 | loss:1.2294610738754272 lr:7.949615602951452e-05 tokens_per_second_per_gpu:3878.0421317355435 mfu:18.29338639761268 +Step 10918 | loss:1.1154133081436157 lr:7.949265459551072e-05 tokens_per_second_per_gpu:4782.540749408685 mfu:22.56006070055438 +Step 10919 | loss:1.0385441780090332 lr:7.948915293968892e-05 tokens_per_second_per_gpu:5315.150778043768 mfu:25.072473078267446 +Step 10920 | loss:1.089455485343933 lr:7.948565106207542e-05 tokens_per_second_per_gpu:4105.446972941724 mfu:19.366093832848996 +Step 10921 | loss:0.9859910011291504 lr:7.948214896269658e-05 tokens_per_second_per_gpu:4310.32047583152 mfu:20.332517101004203 +Step 10922 | loss:1.0426082611083984 lr:7.947864664157876e-05 tokens_per_second_per_gpu:3747.599646060511 mfu:17.67806693690055 +Step 10923 | loss:1.090911626815796 lr:7.947514409874824e-05 tokens_per_second_per_gpu:5182.02194876258 mfu:24.4444816764276 +Step 10924 | loss:1.0761120319366455 lr:7.947164133423142e-05 tokens_per_second_per_gpu:5312.170764767046 mfu:25.058415847198454 +Step 10925 | loss:1.1418347358703613 lr:7.946813834805463e-05 tokens_per_second_per_gpu:5332.097847240454 mfu:25.152415295136375 +Step 10926 | loss:1.201629638671875 lr:7.94646351402442e-05 tokens_per_second_per_gpu:4812.820938749318 mfu:22.702897520007188 +Step 10927 | loss:1.0741820335388184 lr:7.94611317108265e-05 tokens_per_second_per_gpu:4649.569510801785 mfu:21.932812680812898 +Step 10928 | loss:1.1658048629760742 lr:7.945762805982785e-05 tokens_per_second_per_gpu:4537.582086990515 mfu:21.404548895670377 +Step 10929 | loss:1.1822746992111206 lr:7.945412418727464e-05 tokens_per_second_per_gpu:3172.7957382042714 mfu:14.966618831883158 +Step 10930 | loss:1.2215343713760376 lr:7.945062009319319e-05 tokens_per_second_per_gpu:5394.346085601473 mfu:25.446050856129933 +Step 10931 | loss:0.9980990290641785 lr:7.944711577760988e-05 tokens_per_second_per_gpu:5018.7516337251 mfu:23.674307743607564 +Step 10932 | loss:1.105124831199646 lr:7.944361124055106e-05 tokens_per_second_per_gpu:5377.117582697044 mfu:25.364781068444035 +Step 10933 | loss:1.0402623414993286 lr:7.944010648204307e-05 tokens_per_second_per_gpu:5351.013105085335 mfu:25.241641793666457 +Step 10934 | loss:1.0732758045196533 lr:7.943660150211228e-05 tokens_per_second_per_gpu:4388.083875014848 mfu:20.699340322755997 +Step 10935 | loss:1.1150555610656738 lr:7.943309630078506e-05 tokens_per_second_per_gpu:4103.748149529611 mfu:19.358080192964895 +Step 10936 | loss:1.1862096786499023 lr:7.942959087808777e-05 tokens_per_second_per_gpu:5101.197987552884 mfu:24.063221261411705 +Step 10937 | loss:1.0188783407211304 lr:7.942608523404677e-05 tokens_per_second_per_gpu:4474.418939528453 mfu:21.106597552347615 +Step 10938 | loss:1.1061246395111084 lr:7.942257936868843e-05 tokens_per_second_per_gpu:3526.322028850352 mfu:16.63426266266568 +Step 10939 | loss:1.1643401384353638 lr:7.941907328203911e-05 tokens_per_second_per_gpu:4208.194806780137 mfu:19.850772895652938 +Step 10940 | loss:1.06048583984375 lr:7.94155669741252e-05 tokens_per_second_per_gpu:4129.531001423327 mfu:19.4797023043557 +Step 10941 | loss:1.282074213027954 lr:7.941206044497303e-05 tokens_per_second_per_gpu:5309.767099775078 mfu:25.047077349323807 +Step 10942 | loss:1.1797009706497192 lr:7.940855369460902e-05 tokens_per_second_per_gpu:4161.410193749398 mfu:19.630081893708855 +Step 10943 | loss:1.137935757637024 lr:7.940504672305953e-05 tokens_per_second_per_gpu:4588.704319911777 mfu:21.64570119071233 +Step 10944 | loss:1.060151219367981 lr:7.940153953035092e-05 tokens_per_second_per_gpu:4027.2564524526306 mfu:18.9972558070235 +Step 10945 | loss:0.9621096253395081 lr:7.939803211650958e-05 tokens_per_second_per_gpu:5361.52927833126 mfu:25.291248377111657 +Step 10946 | loss:1.2141772508621216 lr:7.939452448156188e-05 tokens_per_second_per_gpu:4248.734390953034 mfu:20.042005030962788 +Step 10947 | loss:1.0708789825439453 lr:7.939101662553423e-05 tokens_per_second_per_gpu:3783.8038203281267 mfu:17.848848203989984 +Step 10948 | loss:0.9602905511856079 lr:7.9387508548453e-05 tokens_per_second_per_gpu:4753.58824010844 mfu:22.423486774377665 +Step 10949 | loss:1.0053560733795166 lr:7.938400025034454e-05 tokens_per_second_per_gpu:5352.277920792574 mfu:25.247608145157315 +Step 10950 | loss:1.011552333831787 lr:7.938049173123527e-05 tokens_per_second_per_gpu:5361.068267595233 mfu:25.289073710813273 +Step 10951 | loss:1.013545036315918 lr:7.937698299115159e-05 tokens_per_second_per_gpu:4339.775130278337 mfu:20.471459731512102 +Step 10952 | loss:1.0728094577789307 lr:7.937347403011987e-05 tokens_per_second_per_gpu:5341.038606933153 mfu:25.19459038405774 +Step 10953 | loss:0.9318340420722961 lr:7.93699648481665e-05 tokens_per_second_per_gpu:5335.1883596171365 mfu:25.166993769312974 +Step 10954 | loss:1.098670244216919 lr:7.936645544531787e-05 tokens_per_second_per_gpu:3967.351111268869 mfu:18.714672091754544 +Step 10955 | loss:1.0669981241226196 lr:7.936294582160039e-05 tokens_per_second_per_gpu:5329.186541934715 mfu:25.138682171289304 +Step 10956 | loss:1.020187258720398 lr:7.935943597704044e-05 tokens_per_second_per_gpu:4509.0848115546605 mfu:21.2701223853029 +Step 10957 | loss:1.0346816778182983 lr:7.935592591166443e-05 tokens_per_second_per_gpu:4594.590833195142 mfu:21.673468878212695 +Step 10958 | loss:1.0824735164642334 lr:7.935241562549876e-05 tokens_per_second_per_gpu:5331.887415968008 mfu:25.151422654920975 +Step 10959 | loss:1.2143921852111816 lr:7.934890511856983e-05 tokens_per_second_per_gpu:5318.02981814508 mfu:25.08605400164023 +Step 10960 | loss:1.147650122642517 lr:7.934539439090402e-05 tokens_per_second_per_gpu:4603.431254392929 mfu:21.715170653334045 +Step 10961 | loss:1.2962865829467773 lr:7.934188344252778e-05 tokens_per_second_per_gpu:3170.3437536475003 mfu:14.955052402377165 +Step 10962 | loss:1.1500060558319092 lr:7.933837227346747e-05 tokens_per_second_per_gpu:4745.359721927463 mfu:22.38467145018803 +Step 10963 | loss:1.0915313959121704 lr:7.933486088374952e-05 tokens_per_second_per_gpu:4602.585497504506 mfu:21.71118107379035 +Step 10964 | loss:1.1701421737670898 lr:7.933134927340034e-05 tokens_per_second_per_gpu:4397.941414915802 mfu:20.74584002033853 +Step 10965 | loss:1.0024033784866333 lr:7.932783744244634e-05 tokens_per_second_per_gpu:4283.110877068109 mfu:20.204164781200934 +Step 10966 | loss:0.9945278167724609 lr:7.932432539091395e-05 tokens_per_second_per_gpu:5300.557442093345 mfu:25.00363382270944 +Step 10967 | loss:1.0390387773513794 lr:7.932081311882954e-05 tokens_per_second_per_gpu:4721.137967972945 mfu:22.270413304127196 +Step 10968 | loss:1.1007332801818848 lr:7.931730062621956e-05 tokens_per_second_per_gpu:4304.00222606256 mfu:20.30271283883945 +Step 10969 | loss:0.9196701049804688 lr:7.931378791311044e-05 tokens_per_second_per_gpu:5343.16239547685 mfu:25.20460865697405 +Step 10970 | loss:1.1086935997009277 lr:7.931027497952855e-05 tokens_per_second_per_gpu:5351.590312194928 mfu:25.24436457808409 +Step 10971 | loss:1.1604434251785278 lr:7.930676182550036e-05 tokens_per_second_per_gpu:4648.696775805178 mfu:21.928695840929958 +Step 10972 | loss:1.1632965803146362 lr:7.930324845105228e-05 tokens_per_second_per_gpu:4559.47015309939 mfu:21.50779863799511 +Step 10973 | loss:1.14906907081604 lr:7.929973485621071e-05 tokens_per_second_per_gpu:5081.241942812011 mfu:23.969085193516676 +Step 10974 | loss:1.02116060256958 lr:7.929622104100212e-05 tokens_per_second_per_gpu:4935.807484310489 mfu:23.28304603908767 +Step 10975 | loss:1.0135340690612793 lr:7.92927070054529e-05 tokens_per_second_per_gpu:5217.796781334503 mfu:24.613237665485205 +Step 10976 | loss:1.064835786819458 lr:7.92891927495895e-05 tokens_per_second_per_gpu:3952.438289925607 mfu:18.644325769088688 +Step 10977 | loss:1.1316401958465576 lr:7.928567827343833e-05 tokens_per_second_per_gpu:4388.547362881285 mfu:20.701526674101164 +Step 10978 | loss:1.0500928163528442 lr:7.928216357702584e-05 tokens_per_second_per_gpu:4840.122717152002 mfu:22.83168466689638 +Step 10979 | loss:1.1088709831237793 lr:7.927864866037847e-05 tokens_per_second_per_gpu:4237.475232924158 mfu:19.988893661530135 +Step 10980 | loss:0.9327150583267212 lr:7.927513352352264e-05 tokens_per_second_per_gpu:3847.115818488335 mfu:18.147501701452455 +Step 10981 | loss:1.1857945919036865 lr:7.92716181664848e-05 tokens_per_second_per_gpu:4468.465381309876 mfu:21.07851360244881 +Step 10982 | loss:1.0375478267669678 lr:7.926810258929138e-05 tokens_per_second_per_gpu:4793.054291475049 mfu:22.609654872276668 +Step 10983 | loss:1.067391276359558 lr:7.926458679196885e-05 tokens_per_second_per_gpu:5011.96397744763 mfu:23.642289210853157 +Step 10984 | loss:1.0375471115112305 lr:7.92610707745436e-05 tokens_per_second_per_gpu:5350.595273088677 mfu:25.23967080884873 +Step 10985 | loss:0.9301968812942505 lr:7.925755453704212e-05 tokens_per_second_per_gpu:5364.9145955420245 mfu:25.30721749598951 +Step 10986 | loss:1.0299444198608398 lr:7.925403807949081e-05 tokens_per_second_per_gpu:4253.6390135053625 mfu:20.065140972357018 +Step 10987 | loss:1.0902128219604492 lr:7.925052140191618e-05 tokens_per_second_per_gpu:4895.462880954784 mfu:23.092733661559674 +Step 10988 | loss:1.131212592124939 lr:7.924700450434462e-05 tokens_per_second_per_gpu:4127.941913592908 mfu:19.472206305933273 +Step 10989 | loss:1.0783360004425049 lr:7.924348738680263e-05 tokens_per_second_per_gpu:5194.095623158981 mfu:24.501435258536695 +Step 10990 | loss:1.3201358318328857 lr:7.923997004931662e-05 tokens_per_second_per_gpu:5233.561150786028 mfu:24.68760088586683 +Step 10991 | loss:1.1293880939483643 lr:7.923645249191308e-05 tokens_per_second_per_gpu:4970.377283233259 mfu:23.446117678822276 +Step 10992 | loss:1.0258173942565918 lr:7.923293471461844e-05 tokens_per_second_per_gpu:5323.00565987433 mfu:25.109525896043245 +Step 10993 | loss:1.1556870937347412 lr:7.922941671745919e-05 tokens_per_second_per_gpu:4191.922119324542 mfu:19.77401185251812 +Step 10994 | loss:1.0611345767974854 lr:7.922589850046174e-05 tokens_per_second_per_gpu:5295.784725954344 mfu:24.98112010637243 +Step 10995 | loss:0.9907433390617371 lr:7.92223800636526e-05 tokens_per_second_per_gpu:3720.670707032191 mfu:17.55103853695319 +Step 10996 | loss:0.9531105160713196 lr:7.921886140705819e-05 tokens_per_second_per_gpu:4478.478492859653 mfu:21.125747158042103 +Step 10997 | loss:1.1348339319229126 lr:7.921534253070499e-05 tokens_per_second_per_gpu:4181.19031696109 mfu:19.723388109735424 +Step 10998 | loss:1.059219479560852 lr:7.921182343461949e-05 tokens_per_second_per_gpu:4956.30248877686 mfu:23.379724471963883 +Step 10999 | loss:1.1152865886688232 lr:7.920830411882813e-05 tokens_per_second_per_gpu:4375.9860064628965 mfu:20.64227261268722 +Step 11000 | loss:1.0816835165023804 lr:7.920478458335739e-05 tokens_per_second_per_gpu:4809.839676386301 mfu:22.688834396785026 +Step 11001 | loss:1.062595248222351 lr:7.920126482823374e-05 tokens_per_second_per_gpu:3972.3641941696624 mfu:18.738319659117728 +Step 11002 | loss:1.1711686849594116 lr:7.919774485348365e-05 tokens_per_second_per_gpu:4017.223950601661 mfu:18.94993078407115 +Step 11003 | loss:1.1194511651992798 lr:7.91942246591336e-05 tokens_per_second_per_gpu:3665.0149691986867 mfu:17.288500925744497 +Step 11004 | loss:1.2444216012954712 lr:7.919070424521005e-05 tokens_per_second_per_gpu:3252.1703335132897 mfu:15.341042340658237 +Step 11005 | loss:0.9330810308456421 lr:7.91871836117395e-05 tokens_per_second_per_gpu:5352.975863357853 mfu:25.250900459319016 +Step 11006 | loss:0.9619076251983643 lr:7.918366275874841e-05 tokens_per_second_per_gpu:3504.897624595818 mfu:16.533200092416745 +Step 11007 | loss:1.151672601699829 lr:7.918014168626326e-05 tokens_per_second_per_gpu:5187.276290730436 mfu:24.46926730397316 +Step 11008 | loss:1.14651620388031 lr:7.917662039431058e-05 tokens_per_second_per_gpu:5378.496733317788 mfu:25.37128675722945 +Step 11009 | loss:1.1016987562179565 lr:7.917309888291677e-05 tokens_per_second_per_gpu:5105.599734392658 mfu:24.083985052270375 +Step 11010 | loss:1.071083426475525 lr:7.916957715210837e-05 tokens_per_second_per_gpu:3961.1686123092036 mfu:18.685508189318707 +Step 11011 | loss:1.041856288909912 lr:7.916605520191187e-05 tokens_per_second_per_gpu:5303.94632962836 mfu:25.019619783415944 +Step 11012 | loss:1.1919646263122559 lr:7.916253303235374e-05 tokens_per_second_per_gpu:3263.5222899607106 mfu:15.394591456064212 +Step 11013 | loss:1.0662826299667358 lr:7.91590106434605e-05 tokens_per_second_per_gpu:3269.682327380338 mfu:15.423649403583354 +Step 11014 | loss:1.1670129299163818 lr:7.915548803525858e-05 tokens_per_second_per_gpu:3364.7166526311303 mfu:15.871942530319219 +Step 11015 | loss:1.1043426990509033 lr:7.915196520777455e-05 tokens_per_second_per_gpu:5324.090287406708 mfu:25.11464226917001 +Step 11016 | loss:0.9825649261474609 lr:7.914844216103486e-05 tokens_per_second_per_gpu:3790.7712963579565 mfu:17.881714977196683 +Step 11017 | loss:1.160697340965271 lr:7.914491889506601e-05 tokens_per_second_per_gpu:4786.497274339408 mfu:22.578724303705368 +Step 11018 | loss:1.177585244178772 lr:7.91413954098945e-05 tokens_per_second_per_gpu:4092.232352993392 mfu:19.303758215906058 +Step 11019 | loss:1.1349751949310303 lr:7.913787170554685e-05 tokens_per_second_per_gpu:5335.677878500704 mfu:25.169302913407293 +Step 11020 | loss:1.0682499408721924 lr:7.913434778204956e-05 tokens_per_second_per_gpu:4807.643296801104 mfu:22.67847369954065 +Step 11021 | loss:0.9217164516448975 lr:7.91308236394291e-05 tokens_per_second_per_gpu:4722.87755879093 mfu:22.278619250820434 +Step 11022 | loss:1.0983257293701172 lr:7.912729927771199e-05 tokens_per_second_per_gpu:4199.526544944827 mfu:19.809883225617806 +Step 11023 | loss:1.0054882764816284 lr:7.912377469692478e-05 tokens_per_second_per_gpu:4496.191159160817 mfu:21.209300827077193 +Step 11024 | loss:1.153926968574524 lr:7.912024989709392e-05 tokens_per_second_per_gpu:4693.592905710433 mfu:22.14047854576269 +Step 11025 | loss:1.026817798614502 lr:7.911672487824595e-05 tokens_per_second_per_gpu:5351.0658062155435 mfu:25.241890393889392 +Step 11026 | loss:1.144546627998352 lr:7.911319964040738e-05 tokens_per_second_per_gpu:4845.492602840554 mfu:22.857015333886253 +Step 11027 | loss:1.1347546577453613 lr:7.910967418360471e-05 tokens_per_second_per_gpu:4960.261722262269 mfu:23.398400851829454 +Step 11028 | loss:1.1178627014160156 lr:7.910614850786448e-05 tokens_per_second_per_gpu:5330.4651133975785 mfu:25.144713411026416 +Step 11029 | loss:1.2316296100616455 lr:7.910262261321317e-05 tokens_per_second_per_gpu:4469.043010779757 mfu:21.081238379211992 +Step 11030 | loss:1.0117743015289307 lr:7.909909649967734e-05 tokens_per_second_per_gpu:5357.36206056288 mfu:25.27159089989754 +Step 11031 | loss:1.283747911453247 lr:7.90955701672835e-05 tokens_per_second_per_gpu:4140.742651267493 mfu:19.532589569575343 +Step 11032 | loss:1.1164836883544922 lr:7.909204361605815e-05 tokens_per_second_per_gpu:3561.424632481476 mfu:16.79984763311545 +Step 11033 | loss:1.2257201671600342 lr:7.908851684602783e-05 tokens_per_second_per_gpu:4330.4848747106225 mfu:20.427636010918143 +Step 11034 | loss:0.9606500864028931 lr:7.908498985721908e-05 tokens_per_second_per_gpu:4675.606575692578 mfu:22.055633958284126 +Step 11035 | loss:1.0133851766586304 lr:7.90814626496584e-05 tokens_per_second_per_gpu:3791.8501984491263 mfu:17.88680434244041 +Step 11036 | loss:1.0612080097198486 lr:7.907793522337231e-05 tokens_per_second_per_gpu:4426.041499639963 mfu:20.87839291435123 +Step 11037 | loss:1.1823023557662964 lr:7.907440757838738e-05 tokens_per_second_per_gpu:4221.201329073542 mfu:19.912126880452465 +Step 11038 | loss:0.977893054485321 lr:7.907087971473011e-05 tokens_per_second_per_gpu:3972.2237526275567 mfu:18.737657172401818 +Step 11039 | loss:0.9768507480621338 lr:7.906735163242704e-05 tokens_per_second_per_gpu:5366.755803011497 mfu:25.3159027857654 +Step 11040 | loss:1.090920090675354 lr:7.906382333150472e-05 tokens_per_second_per_gpu:4596.645365972658 mfu:21.68316045986367 +Step 11041 | loss:0.9676871299743652 lr:7.906029481198969e-05 tokens_per_second_per_gpu:4962.543962160909 mfu:23.409166566821614 +Step 11042 | loss:1.1743913888931274 lr:7.905676607390845e-05 tokens_per_second_per_gpu:3569.91436385741 mfu:16.839895144513267 +Step 11043 | loss:0.9917012453079224 lr:7.905323711728757e-05 tokens_per_second_per_gpu:4148.977867030862 mfu:19.571436487403854 +Step 11044 | loss:1.1741071939468384 lr:7.904970794215358e-05 tokens_per_second_per_gpu:4361.726176994347 mfu:20.575006564106285 +Step 11045 | loss:1.2369166612625122 lr:7.904617854853306e-05 tokens_per_second_per_gpu:5307.743450263915 mfu:25.03753144930944 +Step 11046 | loss:1.0971583127975464 lr:7.90426489364525e-05 tokens_per_second_per_gpu:4182.22860438319 mfu:19.728285888655574 +Step 11047 | loss:1.0717045068740845 lr:7.903911910593846e-05 tokens_per_second_per_gpu:3952.9973599231607 mfu:18.646962997654324 +Step 11048 | loss:1.0720993280410767 lr:7.903558905701754e-05 tokens_per_second_per_gpu:3975.4556663511207 mfu:18.75290266085721 +Step 11049 | loss:1.1529477834701538 lr:7.903205878971622e-05 tokens_per_second_per_gpu:5242.147403767319 mfu:24.728103706145394 +Step 11050 | loss:1.1567456722259521 lr:7.902852830406108e-05 tokens_per_second_per_gpu:4757.384933969267 mfu:22.441396427101957 +Step 11051 | loss:1.2088128328323364 lr:7.902499760007867e-05 tokens_per_second_per_gpu:3373.9222911050033 mfu:15.915367097644362 +Step 11052 | loss:1.0935956239700317 lr:7.902146667779557e-05 tokens_per_second_per_gpu:3753.9163445322715 mfu:17.707863881331736 +Step 11053 | loss:1.1273040771484375 lr:7.901793553723831e-05 tokens_per_second_per_gpu:4391.790769008673 mfu:20.7168263741862 +Step 11054 | loss:0.9902832508087158 lr:7.901440417843344e-05 tokens_per_second_per_gpu:5188.802189498129 mfu:24.476465228805544 +Step 11055 | loss:1.2412407398223877 lr:7.901087260140756e-05 tokens_per_second_per_gpu:4588.853802776129 mfu:21.64640632688788 +Step 11056 | loss:0.9959854483604431 lr:7.900734080618718e-05 tokens_per_second_per_gpu:5370.791101401053 mfu:25.334937976761836 +Step 11057 | loss:1.0823405981063843 lr:7.900380879279889e-05 tokens_per_second_per_gpu:5315.1425721057485 mfu:25.07243436945773 +Step 11058 | loss:1.1689175367355347 lr:7.900027656126927e-05 tokens_per_second_per_gpu:4411.23609469035 mfu:20.808553293141344 +Step 11059 | loss:1.1845818758010864 lr:7.899674411162486e-05 tokens_per_second_per_gpu:4865.045875555476 mfu:22.949251457414874 +Step 11060 | loss:1.210955262184143 lr:7.899321144389224e-05 tokens_per_second_per_gpu:4118.856263920117 mfu:19.429347746255033 +Step 11061 | loss:1.1613528728485107 lr:7.898967855809797e-05 tokens_per_second_per_gpu:4397.485506529668 mfu:20.743689422695148 +Step 11062 | loss:1.097319483757019 lr:7.898614545426863e-05 tokens_per_second_per_gpu:4228.155024031971 mfu:19.944928645994075 +Step 11063 | loss:1.0297231674194336 lr:7.898261213243081e-05 tokens_per_second_per_gpu:4841.39174052464 mfu:22.837670866662826 +Step 11064 | loss:1.0148173570632935 lr:7.897907859261105e-05 tokens_per_second_per_gpu:5350.2785695982975 mfu:25.238176864449425 +Step 11065 | loss:1.1620887517929077 lr:7.897554483483593e-05 tokens_per_second_per_gpu:5305.46755809692 mfu:25.02679567765048 +Step 11066 | loss:1.0781950950622559 lr:7.897201085913206e-05 tokens_per_second_per_gpu:5338.566428291596 mfu:25.182928695608346 +Step 11067 | loss:1.0313270092010498 lr:7.896847666552599e-05 tokens_per_second_per_gpu:4494.399452383588 mfu:21.200849040512306 +Step 11068 | loss:1.1923179626464844 lr:7.896494225404432e-05 tokens_per_second_per_gpu:3604.792331769423 mfu:17.00442046994919 +Step 11069 | loss:1.0545505285263062 lr:7.896140762471361e-05 tokens_per_second_per_gpu:4319.694539158485 mfu:20.376736157097625 +Step 11070 | loss:1.0554884672164917 lr:7.895787277756047e-05 tokens_per_second_per_gpu:5348.114145773411 mfu:25.22796690797865 +Step 11071 | loss:1.026828408241272 lr:7.895433771261147e-05 tokens_per_second_per_gpu:5019.4800916126005 mfu:23.677744003750632 +Step 11072 | loss:1.1675388813018799 lr:7.895080242989318e-05 tokens_per_second_per_gpu:3586.664060545844 mfu:16.918906321585553 +Step 11073 | loss:1.117140293121338 lr:7.894726692943225e-05 tokens_per_second_per_gpu:3817.312586719802 mfu:18.0069147722446 +Step 11074 | loss:1.121205449104309 lr:7.894373121125521e-05 tokens_per_second_per_gpu:5279.942592740974 mfu:24.906390061058318 +Step 11075 | loss:1.0529561042785645 lr:7.894019527538867e-05 tokens_per_second_per_gpu:3531.192645741396 mfu:16.657238193553667 +Step 11076 | loss:1.0317366123199463 lr:7.893665912185923e-05 tokens_per_second_per_gpu:4009.257883197343 mfu:18.912353484973735 +Step 11077 | loss:1.1523120403289795 lr:7.89331227506935e-05 tokens_per_second_per_gpu:4108.438125849316 mfu:19.380203611457414 +Step 11078 | loss:1.0893248319625854 lr:7.892958616191805e-05 tokens_per_second_per_gpu:4062.217184419094 mfu:19.162171544625263 +Step 11079 | loss:1.0591620206832886 lr:7.89260493555595e-05 tokens_per_second_per_gpu:5338.908077423383 mfu:25.184540312854338 +Step 11080 | loss:1.0905388593673706 lr:7.892251233164442e-05 tokens_per_second_per_gpu:3719.4420192527728 mfu:17.545242607061923 +Step 11081 | loss:1.0191915035247803 lr:7.891897509019945e-05 tokens_per_second_per_gpu:5362.776004061251 mfu:25.297129395093368 +Step 11082 | loss:0.9920344352722168 lr:7.891543763125118e-05 tokens_per_second_per_gpu:4198.909012310463 mfu:19.806970218819583 +Step 11083 | loss:0.9922596216201782 lr:7.89118999548262e-05 tokens_per_second_per_gpu:4970.207404792416 mfu:23.44531633323257 +Step 11084 | loss:1.04246985912323 lr:7.890836206095116e-05 tokens_per_second_per_gpu:4310.219751281948 mfu:20.332041966118496 +Step 11085 | loss:1.202714443206787 lr:7.890482394965262e-05 tokens_per_second_per_gpu:5286.79050601121 mfu:24.938692836328503 +Step 11086 | loss:1.1007251739501953 lr:7.89012856209572e-05 tokens_per_second_per_gpu:5076.228828656239 mfu:23.945437478718425 +Step 11087 | loss:1.2688100337982178 lr:7.889774707489155e-05 tokens_per_second_per_gpu:5072.746216916199 mfu:23.929009404945823 +Step 11088 | loss:1.117302656173706 lr:7.889420831148225e-05 tokens_per_second_per_gpu:5268.398365225747 mfu:24.85193397022067 +Step 11089 | loss:1.081594467163086 lr:7.889066933075591e-05 tokens_per_second_per_gpu:4312.051770862503 mfu:20.340683915054743 +Step 11090 | loss:1.230914831161499 lr:7.888713013273916e-05 tokens_per_second_per_gpu:5380.669363750862 mfu:25.381535425671235 +Step 11091 | loss:1.0649597644805908 lr:7.888359071745861e-05 tokens_per_second_per_gpu:4374.8260990109175 mfu:20.63680113133562 +Step 11092 | loss:1.0993857383728027 lr:7.888005108494091e-05 tokens_per_second_per_gpu:4883.7700489296085 mfu:23.037576577894058 +Step 11093 | loss:1.303938865661621 lr:7.887651123521265e-05 tokens_per_second_per_gpu:3768.1146950139077 mfu:17.774839923042006 +Step 11094 | loss:1.0761141777038574 lr:7.887297116830047e-05 tokens_per_second_per_gpu:3903.9264932404694 mfu:18.415487347158884 +Step 11095 | loss:0.9453962445259094 lr:7.886943088423098e-05 tokens_per_second_per_gpu:4761.267758083563 mfu:22.4597123709272 +Step 11096 | loss:0.9674528241157532 lr:7.886589038303083e-05 tokens_per_second_per_gpu:5336.766907206666 mfu:25.174440047620774 +Step 11097 | loss:1.0689324140548706 lr:7.886234966472664e-05 tokens_per_second_per_gpu:5339.495556478696 mfu:25.187311551792515 +Step 11098 | loss:1.0950226783752441 lr:7.885880872934502e-05 tokens_per_second_per_gpu:5352.469105086272 mfu:25.248509994090092 +Step 11099 | loss:1.1171777248382568 lr:7.885526757691262e-05 tokens_per_second_per_gpu:3881.667568182496 mfu:18.310488199896195 +Step 11100 | loss:0.9546994566917419 lr:7.885172620745606e-05 tokens_per_second_per_gpu:4285.944745141294 mfu:20.217532620409898 +Step 11101 | loss:1.051967740058899 lr:7.8848184621002e-05 tokens_per_second_per_gpu:4418.058914495876 mfu:20.840737698257104 +Step 11102 | loss:1.0946284532546997 lr:7.884464281757706e-05 tokens_per_second_per_gpu:4885.129723219783 mfu:23.04399039350517 +Step 11103 | loss:0.9950097799301147 lr:7.884110079720787e-05 tokens_per_second_per_gpu:3834.700621165146 mfu:18.08893709742788 +Step 11104 | loss:1.0123950242996216 lr:7.88375585599211e-05 tokens_per_second_per_gpu:4181.685126067136 mfu:19.725722208711783 +Step 11105 | loss:1.0123671293258667 lr:7.883401610574336e-05 tokens_per_second_per_gpu:5309.184102618379 mfu:25.044327252266086 +Step 11106 | loss:1.0475566387176514 lr:7.883047343470132e-05 tokens_per_second_per_gpu:4814.24382786982 mfu:22.709609530758513 +Step 11107 | loss:1.1082227230072021 lr:7.88269305468216e-05 tokens_per_second_per_gpu:3990.0301940891095 mfu:18.821653194868155 +Step 11108 | loss:1.2168928384780884 lr:7.882338744213084e-05 tokens_per_second_per_gpu:5374.807146092147 mfu:25.353882344777606 +Step 11109 | loss:1.0962200164794922 lr:7.881984412065573e-05 tokens_per_second_per_gpu:3398.9616095123693 mfu:16.033481834720156 +Step 11110 | loss:1.0525226593017578 lr:7.881630058242288e-05 tokens_per_second_per_gpu:5335.732018085903 mfu:25.169558299067845 +Step 11111 | loss:1.078892707824707 lr:7.881275682745897e-05 tokens_per_second_per_gpu:4726.386025033929 mfu:22.295169284694726 +Step 11112 | loss:1.000605583190918 lr:7.880921285579062e-05 tokens_per_second_per_gpu:5318.540908396056 mfu:25.08846490155507 +Step 11113 | loss:1.0616952180862427 lr:7.880566866744452e-05 tokens_per_second_per_gpu:5177.7115958519535 mfu:24.424149006326136 +Step 11114 | loss:1.1292698383331299 lr:7.88021242624473e-05 tokens_per_second_per_gpu:4114.0259718292655 mfu:19.406562434329675 +Step 11115 | loss:1.0459423065185547 lr:7.879857964082564e-05 tokens_per_second_per_gpu:5356.312442563675 mfu:25.266639672711996 +Step 11116 | loss:1.0586414337158203 lr:7.879503480260618e-05 tokens_per_second_per_gpu:4878.141976827007 mfu:23.01102800153797 +Step 11117 | loss:1.1774036884307861 lr:7.879148974781561e-05 tokens_per_second_per_gpu:3818.2093812653898 mfu:18.011145105125063 +Step 11118 | loss:1.0463345050811768 lr:7.878794447648055e-05 tokens_per_second_per_gpu:4627.57489097977 mfu:21.82906030644335 +Step 11119 | loss:1.175390601158142 lr:7.878439898862768e-05 tokens_per_second_per_gpu:4743.84620380199 mfu:22.37753192695735 +Step 11120 | loss:1.0785309076309204 lr:7.878085328428369e-05 tokens_per_second_per_gpu:3886.385209045739 mfu:18.33274211676064 +Step 11121 | loss:1.157983660697937 lr:7.877730736347522e-05 tokens_per_second_per_gpu:5240.089563285222 mfu:24.718396521487556 +Step 11122 | loss:1.1373646259307861 lr:7.877376122622894e-05 tokens_per_second_per_gpu:5124.408768703477 mfu:24.172710476266676 +Step 11123 | loss:1.171420693397522 lr:7.877021487257155e-05 tokens_per_second_per_gpu:5333.35977482473 mfu:25.158368022858454 +Step 11124 | loss:1.081463098526001 lr:7.87666683025297e-05 tokens_per_second_per_gpu:3105.2756627372646 mfu:14.648115115792498 +Step 11125 | loss:0.9980998039245605 lr:7.876312151613006e-05 tokens_per_second_per_gpu:4654.82984260666 mfu:21.95762656344165 +Step 11126 | loss:1.0257967710494995 lr:7.87595745133993e-05 tokens_per_second_per_gpu:4569.433682726599 mfu:21.5547983072001 +Step 11127 | loss:1.0382570028305054 lr:7.875602729436413e-05 tokens_per_second_per_gpu:3464.2806930847864 mfu:16.3416030965161 +Step 11128 | loss:1.0446205139160156 lr:7.875247985905121e-05 tokens_per_second_per_gpu:4892.779418964922 mfu:23.080075313507667 +Step 11129 | loss:1.10396146774292 lr:7.87489322074872e-05 tokens_per_second_per_gpu:4813.1223236056185 mfu:22.704319203795915 +Step 11130 | loss:0.970132052898407 lr:7.874538433969883e-05 tokens_per_second_per_gpu:4075.845596496075 mfu:19.226459089639256 +Step 11131 | loss:1.0579638481140137 lr:7.874183625571275e-05 tokens_per_second_per_gpu:5311.98372516226 mfu:25.057533549470435 +Step 11132 | loss:1.0722874402999878 lr:7.873828795555563e-05 tokens_per_second_per_gpu:4849.074022351666 mfu:22.873909500773348 +Step 11133 | loss:0.9549729228019714 lr:7.873473943925421e-05 tokens_per_second_per_gpu:3827.6205942854913 mfu:18.055539402659402 +Step 11134 | loss:1.1671521663665771 lr:7.87311907068351e-05 tokens_per_second_per_gpu:3840.3454794675054 mfu:18.11556485715225 +Step 11135 | loss:1.0759196281433105 lr:7.872764175832509e-05 tokens_per_second_per_gpu:3575.0530935742913 mfu:16.864135409345497 +Step 11136 | loss:1.225584626197815 lr:7.87240925937508e-05 tokens_per_second_per_gpu:4918.23251297344 mfu:23.200141900691495 +Step 11137 | loss:1.119482159614563 lr:7.872054321313895e-05 tokens_per_second_per_gpu:4783.14476809428 mfu:22.562909959750552 +Step 11138 | loss:1.2249864339828491 lr:7.871699361651621e-05 tokens_per_second_per_gpu:4485.309313900187 mfu:21.15796930634855 +Step 11139 | loss:1.1559066772460938 lr:7.871344380390933e-05 tokens_per_second_per_gpu:4747.809011781956 mfu:22.396225168323607 +Step 11140 | loss:1.1149317026138306 lr:7.870989377534495e-05 tokens_per_second_per_gpu:4722.472268369936 mfu:22.27670742675492 +Step 11141 | loss:1.1095770597457886 lr:7.87063435308498e-05 tokens_per_second_per_gpu:4386.26403334135 mfu:20.6907558190851 +Step 11142 | loss:1.10927152633667 lr:7.870279307045059e-05 tokens_per_second_per_gpu:5054.345380835932 mfu:23.842209521648833 +Step 11143 | loss:1.0195863246917725 lr:7.869924239417399e-05 tokens_per_second_per_gpu:4804.877183136622 mfu:22.665425469437633 +Step 11144 | loss:1.033934235572815 lr:7.869569150204674e-05 tokens_per_second_per_gpu:4359.032523540261 mfu:20.56230014117869 +Step 11145 | loss:1.160351037979126 lr:7.869214039409553e-05 tokens_per_second_per_gpu:3419.6575309903196 mfu:16.131108027419966 +Step 11146 | loss:1.292000412940979 lr:7.868858907034707e-05 tokens_per_second_per_gpu:4906.513093945431 mfu:23.144859401597305 +Step 11147 | loss:1.1022719144821167 lr:7.868503753082808e-05 tokens_per_second_per_gpu:5374.505729778723 mfu:25.35246051260026 +Step 11148 | loss:1.1480753421783447 lr:7.868148577556524e-05 tokens_per_second_per_gpu:3938.9224265011935 mfu:18.580569135271634 +Step 11149 | loss:1.0165389776229858 lr:7.86779338045853e-05 tokens_per_second_per_gpu:4788.283482229068 mfu:22.587150151079385 +Step 11150 | loss:0.8831307888031006 lr:7.867438161791497e-05 tokens_per_second_per_gpu:5407.814550603793 mfu:25.509583903502964 +Step 11151 | loss:1.225541591644287 lr:7.867082921558094e-05 tokens_per_second_per_gpu:4270.161049382323 mfu:20.143078234538642 +Step 11152 | loss:1.1342196464538574 lr:7.866727659760995e-05 tokens_per_second_per_gpu:4577.284568893034 mfu:21.591832276746484 +Step 11153 | loss:1.146264910697937 lr:7.866372376402873e-05 tokens_per_second_per_gpu:4055.5940910710165 mfu:19.130929283286132 +Step 11154 | loss:1.0598722696304321 lr:7.866017071486397e-05 tokens_per_second_per_gpu:4757.18829363389 mfu:22.440468841089682 +Step 11155 | loss:0.9701348543167114 lr:7.865661745014241e-05 tokens_per_second_per_gpu:3423.086687173246 mfu:16.147283942209206 +Step 11156 | loss:1.0299756526947021 lr:7.865306396989077e-05 tokens_per_second_per_gpu:4779.381022671082 mfu:22.54515573042814 +Step 11157 | loss:1.0647292137145996 lr:7.864951027413578e-05 tokens_per_second_per_gpu:5366.984213072453 mfu:25.316980234993704 +Step 11158 | loss:0.9646443724632263 lr:7.864595636290418e-05 tokens_per_second_per_gpu:3761.486872546413 mfu:17.743575353639716 +Step 11159 | loss:1.1408311128616333 lr:7.864240223622268e-05 tokens_per_second_per_gpu:4909.467554257862 mfu:23.15879609497395 +Step 11160 | loss:1.1926926374435425 lr:7.863884789411802e-05 tokens_per_second_per_gpu:4387.061380766647 mfu:20.69451704293172 +Step 11161 | loss:1.1614805459976196 lr:7.863529333661693e-05 tokens_per_second_per_gpu:4341.713629482333 mfu:20.480603962999417 +Step 11162 | loss:1.2361620664596558 lr:7.863173856374614e-05 tokens_per_second_per_gpu:4060.609059433482 mfu:19.154585744693716 +Step 11163 | loss:1.0298244953155518 lr:7.862818357553239e-05 tokens_per_second_per_gpu:5362.647299229772 mfu:25.296522272443976 +Step 11164 | loss:0.9789228439331055 lr:7.862462837200243e-05 tokens_per_second_per_gpu:5258.80511921258 mfu:24.80668099200002 +Step 11165 | loss:0.9888537526130676 lr:7.862107295318297e-05 tokens_per_second_per_gpu:5104.8266996829625 mfu:24.080338515651416 +Step 11166 | loss:0.8169659376144409 lr:7.861751731910079e-05 tokens_per_second_per_gpu:4487.36697206927 mfu:21.1676756310048 +Step 11167 | loss:1.0564372539520264 lr:7.861396146978259e-05 tokens_per_second_per_gpu:5347.518234283813 mfu:25.225155891809234 +Step 11168 | loss:1.164355754852295 lr:7.861040540525515e-05 tokens_per_second_per_gpu:3967.1805148144454 mfu:18.71386735917227 +Step 11169 | loss:1.3495254516601562 lr:7.860684912554519e-05 tokens_per_second_per_gpu:4286.284269076813 mfu:20.219134212742496 +Step 11170 | loss:1.0522459745407104 lr:7.860329263067947e-05 tokens_per_second_per_gpu:4814.037624520073 mfu:22.708636834375838 +Step 11171 | loss:1.1134889125823975 lr:7.859973592068475e-05 tokens_per_second_per_gpu:5347.795312063439 mfu:25.22646291497007 +Step 11172 | loss:1.019266128540039 lr:7.859617899558775e-05 tokens_per_second_per_gpu:4666.796522043318 mfu:22.01407543207041 +Step 11173 | loss:1.1602387428283691 lr:7.859262185541523e-05 tokens_per_second_per_gpu:3805.0289784469037 mfu:17.948970896221653 +Step 11174 | loss:1.0962350368499756 lr:7.858906450019398e-05 tokens_per_second_per_gpu:4987.7801929000325 mfu:23.528210172963018 +Step 11175 | loss:0.9952630400657654 lr:7.85855069299507e-05 tokens_per_second_per_gpu:3915.6504669517817 mfu:18.470791331471325 +Step 11176 | loss:0.9370854496955872 lr:7.858194914471219e-05 tokens_per_second_per_gpu:5388.002625419239 mfu:25.416127672144423 +Step 11177 | loss:0.9621763825416565 lr:7.85783911445052e-05 tokens_per_second_per_gpu:3609.136175947254 mfu:17.024911124072254 +Step 11178 | loss:1.1248730421066284 lr:7.857483292935648e-05 tokens_per_second_per_gpu:3235.754563198412 mfu:15.263606351263734 +Step 11179 | loss:1.0051822662353516 lr:7.857127449929279e-05 tokens_per_second_per_gpu:3512.0813086754006 mfu:16.56708675588303 +Step 11180 | loss:0.903856635093689 lr:7.85677158543409e-05 tokens_per_second_per_gpu:3788.145659612473 mfu:17.869329400688414 +Step 11181 | loss:1.0544192790985107 lr:7.856415699452757e-05 tokens_per_second_per_gpu:5329.563968718188 mfu:25.140462557822598 +Step 11182 | loss:1.2317562103271484 lr:7.856059791987958e-05 tokens_per_second_per_gpu:4802.644281511766 mfu:22.654892491500885 +Step 11183 | loss:1.230984091758728 lr:7.855703863042368e-05 tokens_per_second_per_gpu:5275.243452345959 mfu:24.884223414059004 +Step 11184 | loss:1.1399834156036377 lr:7.855347912618665e-05 tokens_per_second_per_gpu:5335.04784661708 mfu:25.166330945517714 +Step 11185 | loss:1.031286597251892 lr:7.854991940719524e-05 tokens_per_second_per_gpu:5321.589634982073 mfu:25.102846265027708 +Step 11186 | loss:1.0162017345428467 lr:7.854635947347628e-05 tokens_per_second_per_gpu:4121.386072539997 mfu:19.441281285144864 +Step 11187 | loss:1.1866434812545776 lr:7.85427993250565e-05 tokens_per_second_per_gpu:4390.576877896241 mfu:20.71110024269768 +Step 11188 | loss:1.1362496614456177 lr:7.853923896196267e-05 tokens_per_second_per_gpu:5387.254173648196 mfu:25.412597097441378 +Step 11189 | loss:1.1765069961547852 lr:7.85356783842216e-05 tokens_per_second_per_gpu:4362.094105981031 mfu:20.57674214791163 +Step 11190 | loss:1.0732907056808472 lr:7.853211759186004e-05 tokens_per_second_per_gpu:5336.5103771978975 mfu:25.17322995180088 +Step 11191 | loss:1.1288625001907349 lr:7.852855658490477e-05 tokens_per_second_per_gpu:3436.0039384991383 mfu:16.20821682062378 +Step 11192 | loss:1.0570300817489624 lr:7.85249953633826e-05 tokens_per_second_per_gpu:5323.771694077735 mfu:25.11313941007962 +Step 11193 | loss:1.0637234449386597 lr:7.85214339273203e-05 tokens_per_second_per_gpu:5328.372760937638 mfu:25.134843427480735 +Step 11194 | loss:1.0578604936599731 lr:7.851787227674466e-05 tokens_per_second_per_gpu:5183.3864351547245 mfu:24.450918191544734 +Step 11195 | loss:1.0772085189819336 lr:7.851431041168244e-05 tokens_per_second_per_gpu:5221.822329737514 mfu:24.632226864131592 +Step 11196 | loss:1.0704572200775146 lr:7.851074833216049e-05 tokens_per_second_per_gpu:5205.143241313047 mfu:24.55354875828745 +Step 11197 | loss:1.219556450843811 lr:7.850718603820555e-05 tokens_per_second_per_gpu:4323.418656908736 mfu:20.394303455925566 +Step 11198 | loss:0.9818225502967834 lr:7.850362352984442e-05 tokens_per_second_per_gpu:4668.544642229247 mfu:22.0223216132475 +Step 11199 | loss:1.2775545120239258 lr:7.85000608071039e-05 tokens_per_second_per_gpu:3392.2366758231483 mfu:16.0017591751161 +Step 11200 | loss:1.0742753744125366 lr:7.849649787001079e-05 tokens_per_second_per_gpu:3912.4308982125985 mfu:18.45560407641353 +Step 11201 | loss:1.0862910747528076 lr:7.849293471859187e-05 tokens_per_second_per_gpu:4167.415219669041 mfu:19.658408625534882 +Step 11202 | loss:1.0856518745422363 lr:7.848937135287397e-05 tokens_per_second_per_gpu:4445.423137609586 mfu:20.969819407502374 +Step 11203 | loss:1.1673976182937622 lr:7.848580777288385e-05 tokens_per_second_per_gpu:4816.620769567052 mfu:22.72082197029243 +Step 11204 | loss:1.0539367198944092 lr:7.848224397864836e-05 tokens_per_second_per_gpu:3889.4811503057044 mfu:18.34734619990595 +Step 11205 | loss:1.051433801651001 lr:7.847867997019426e-05 tokens_per_second_per_gpu:5000.581706701709 mfu:23.588597097729004 +Step 11206 | loss:1.117207646369934 lr:7.84751157475484e-05 tokens_per_second_per_gpu:5342.553597894087 mfu:25.201736854904578 +Step 11207 | loss:0.9366761445999146 lr:7.847155131073754e-05 tokens_per_second_per_gpu:3748.704358468126 mfu:17.683278053810643 +Step 11208 | loss:1.3520491123199463 lr:7.846798665978852e-05 tokens_per_second_per_gpu:5305.005607933832 mfu:25.024616579914365 +Step 11209 | loss:0.9387158155441284 lr:7.846442179472813e-05 tokens_per_second_per_gpu:5324.29215671754 mfu:25.11559452115166 +Step 11210 | loss:1.3239701986312866 lr:7.846085671558319e-05 tokens_per_second_per_gpu:3856.254548679062 mfu:18.190610650990266 +Step 11211 | loss:1.0496495962142944 lr:7.845729142238053e-05 tokens_per_second_per_gpu:5350.606742766533 mfu:25.239724913276106 +Step 11212 | loss:1.0566266775131226 lr:7.845372591514695e-05 tokens_per_second_per_gpu:5374.973104273137 mfu:25.354665197832723 +Step 11213 | loss:1.0903924703598022 lr:7.845016019390925e-05 tokens_per_second_per_gpu:4319.386424857111 mfu:20.375282729368323 +Step 11214 | loss:0.9073246121406555 lr:7.844659425869427e-05 tokens_per_second_per_gpu:3319.8929736804116 mfu:15.660501589000445 +Step 11215 | loss:1.1454609632492065 lr:7.844302810952883e-05 tokens_per_second_per_gpu:5338.36069098164 mfu:25.18195819761488 +Step 11216 | loss:0.9954371452331543 lr:7.843946174643975e-05 tokens_per_second_per_gpu:5282.483049513303 mfu:24.918373829100016 +Step 11217 | loss:1.1399250030517578 lr:7.843589516945384e-05 tokens_per_second_per_gpu:3399.177658171323 mfu:16.034500973106145 +Step 11218 | loss:1.0634429454803467 lr:7.843232837859792e-05 tokens_per_second_per_gpu:4205.658301657332 mfu:19.83880776821627 +Step 11219 | loss:1.1725800037384033 lr:7.842876137389885e-05 tokens_per_second_per_gpu:4333.555051519872 mfu:20.442118558753553 +Step 11220 | loss:1.1677230596542358 lr:7.842519415538346e-05 tokens_per_second_per_gpu:5102.619628070638 mfu:24.069927382290835 +Step 11221 | loss:1.3355234861373901 lr:7.842162672307852e-05 tokens_per_second_per_gpu:3559.676712363189 mfu:16.791602395691566 +Step 11222 | loss:1.0815536975860596 lr:7.841805907701092e-05 tokens_per_second_per_gpu:5275.867056062609 mfu:24.88716505918821 +Step 11223 | loss:1.2020971775054932 lr:7.841449121720745e-05 tokens_per_second_per_gpu:4136.584996744044 mfu:19.512977203819222 +Step 11224 | loss:1.0498661994934082 lr:7.841092314369498e-05 tokens_per_second_per_gpu:4496.386290881128 mfu:21.210221296694808 +Step 11225 | loss:1.1096673011779785 lr:7.840735485650034e-05 tokens_per_second_per_gpu:4283.336905828404 mfu:20.20523099742773 +Step 11226 | loss:1.1562589406967163 lr:7.840378635565034e-05 tokens_per_second_per_gpu:3724.6161081458886 mfu:17.569649667161347 +Step 11227 | loss:1.313698410987854 lr:7.840021764117184e-05 tokens_per_second_per_gpu:4060.3249291451575 mfu:19.15324545364606 +Step 11228 | loss:1.0139683485031128 lr:7.839664871309171e-05 tokens_per_second_per_gpu:4611.3351665789305 mfu:21.752454755661894 +Step 11229 | loss:1.117629051208496 lr:7.839307957143674e-05 tokens_per_second_per_gpu:4741.51918197393 mfu:22.366554967963527 +Step 11230 | loss:0.9603145718574524 lr:7.838951021623379e-05 tokens_per_second_per_gpu:4360.189558377236 mfu:20.5677580719152 +Step 11231 | loss:1.0172977447509766 lr:7.83859406475097e-05 tokens_per_second_per_gpu:3367.8751090993046 mfu:15.88684150836784 +Step 11232 | loss:1.0751770734786987 lr:7.838237086529135e-05 tokens_per_second_per_gpu:3766.2180491919735 mfu:17.765893121098795 +Step 11233 | loss:1.109391212463379 lr:7.837880086960558e-05 tokens_per_second_per_gpu:5083.093656948912 mfu:23.97782004503545 +Step 11234 | loss:0.9865619540214539 lr:7.837523066047921e-05 tokens_per_second_per_gpu:4237.06865739782 mfu:19.986975775403458 +Step 11235 | loss:1.0323188304901123 lr:7.83716602379391e-05 tokens_per_second_per_gpu:4517.889588464612 mfu:21.31165602023699 +Step 11236 | loss:1.0901966094970703 lr:7.836808960201212e-05 tokens_per_second_per_gpu:4446.012275041002 mfu:20.97259846928357 +Step 11237 | loss:1.1011993885040283 lr:7.836451875272513e-05 tokens_per_second_per_gpu:4574.835775829713 mfu:21.58028090205962 +Step 11238 | loss:1.1084997653961182 lr:7.836094769010495e-05 tokens_per_second_per_gpu:3926.9973662704624 mfu:18.5243165915382 +Step 11239 | loss:1.1470280885696411 lr:7.835737641417848e-05 tokens_per_second_per_gpu:5001.725549073855 mfu:23.59399279735805 +Step 11240 | loss:1.0818369388580322 lr:7.835380492497257e-05 tokens_per_second_per_gpu:5368.412920176295 mfu:25.323719690165095 +Step 11241 | loss:1.0593963861465454 lr:7.835023322251408e-05 tokens_per_second_per_gpu:4409.7477146680885 mfu:20.801532350632517 +Step 11242 | loss:1.014732003211975 lr:7.834666130682984e-05 tokens_per_second_per_gpu:4419.419044169181 mfu:20.847153661988138 +Step 11243 | loss:1.0519977807998657 lr:7.834308917794676e-05 tokens_per_second_per_gpu:5355.475936392051 mfu:25.262693730377777 +Step 11244 | loss:0.9602765440940857 lr:7.83395168358917e-05 tokens_per_second_per_gpu:3635.061508507056 mfu:17.147205341074695 +Step 11245 | loss:1.0587626695632935 lr:7.833594428069148e-05 tokens_per_second_per_gpu:4420.958504585731 mfu:20.854415559432038 +Step 11246 | loss:0.983453631401062 lr:7.833237151237305e-05 tokens_per_second_per_gpu:5318.460757867213 mfu:25.08808681783543 +Step 11247 | loss:1.1212180852890015 lr:7.832879853096322e-05 tokens_per_second_per_gpu:5055.2874772295845 mfu:23.84665355107591 +Step 11248 | loss:1.0457102060317993 lr:7.83252253364889e-05 tokens_per_second_per_gpu:4683.985541819438 mfu:22.095158971103558 +Step 11249 | loss:1.2177386283874512 lr:7.832165192897692e-05 tokens_per_second_per_gpu:4677.342329417055 mfu:22.063821804752607 +Step 11250 | loss:1.049501895904541 lr:7.83180783084542e-05 tokens_per_second_per_gpu:4443.888424788375 mfu:20.962579905275714 +Step 11251 | loss:1.10225510597229 lr:7.831450447494758e-05 tokens_per_second_per_gpu:4587.269883887605 mfu:21.63893471124629 +Step 11252 | loss:1.0045204162597656 lr:7.831093042848396e-05 tokens_per_second_per_gpu:3158.6029905410137 mfu:14.89966921962312 +Step 11253 | loss:1.0385513305664062 lr:7.830735616909024e-05 tokens_per_second_per_gpu:5322.87274811799 mfu:25.108898928611502 +Step 11254 | loss:1.1450978517532349 lr:7.830378169679327e-05 tokens_per_second_per_gpu:5366.099560957436 mfu:25.31280718002995 +Step 11255 | loss:1.265307903289795 lr:7.830020701161996e-05 tokens_per_second_per_gpu:3962.666441650254 mfu:18.69257370587648 +Step 11256 | loss:1.3760879039764404 lr:7.829663211359717e-05 tokens_per_second_per_gpu:4026.157794645425 mfu:18.992073250696556 +Step 11257 | loss:1.0369449853897095 lr:7.829305700275179e-05 tokens_per_second_per_gpu:3944.858661946732 mfu:18.608571370694445 +Step 11258 | loss:1.1062198877334595 lr:7.828948167911074e-05 tokens_per_second_per_gpu:5367.165080557919 mfu:25.31783341778182 +Step 11259 | loss:0.9520562291145325 lr:7.828590614270086e-05 tokens_per_second_per_gpu:5291.8657672115805 mfu:24.962633709339503 +Step 11260 | loss:1.0523713827133179 lr:7.828233039354909e-05 tokens_per_second_per_gpu:4996.968615522918 mfu:23.571553530181728 +Step 11261 | loss:1.0484617948532104 lr:7.82787544316823e-05 tokens_per_second_per_gpu:5204.063250777262 mfu:24.548454258664762 +Step 11262 | loss:1.1776628494262695 lr:7.827517825712741e-05 tokens_per_second_per_gpu:4016.297351647663 mfu:18.94555985871106 +Step 11263 | loss:1.0744738578796387 lr:7.827160186991126e-05 tokens_per_second_per_gpu:5312.364370653478 mfu:25.059329119197322 +Step 11264 | loss:1.039128303527832 lr:7.826802527006081e-05 tokens_per_second_per_gpu:3988.6318350310767 mfu:18.815056896606755 +Step 11265 | loss:1.1005454063415527 lr:7.826444845760293e-05 tokens_per_second_per_gpu:3802.0220097300594 mfu:17.934786511742534 +Step 11266 | loss:1.1296720504760742 lr:7.826087143256452e-05 tokens_per_second_per_gpu:4521.675057017352 mfu:21.329512721267793 +Step 11267 | loss:0.9722786545753479 lr:7.82572941949725e-05 tokens_per_second_per_gpu:4613.704721027953 mfu:21.76363234828561 +Step 11268 | loss:1.03621506690979 lr:7.825371674485375e-05 tokens_per_second_per_gpu:4989.9921687296455 mfu:23.538644440353295 +Step 11269 | loss:1.0426863431930542 lr:7.82501390822352e-05 tokens_per_second_per_gpu:4902.239243613149 mfu:23.124698920385867 +Step 11270 | loss:1.0131831169128418 lr:7.824656120714374e-05 tokens_per_second_per_gpu:5187.087329188873 mfu:24.468375940141446 +Step 11271 | loss:1.1818581819534302 lr:7.82429831196063e-05 tokens_per_second_per_gpu:4757.7995311921395 mfu:22.443352152939827 +Step 11272 | loss:1.2194770574569702 lr:7.823940481964977e-05 tokens_per_second_per_gpu:3776.940739066135 mfu:17.81647387871476 +Step 11273 | loss:1.1548573970794678 lr:7.823582630730109e-05 tokens_per_second_per_gpu:5043.116368540569 mfu:23.78924035478455 +Step 11274 | loss:1.1482490301132202 lr:7.823224758258714e-05 tokens_per_second_per_gpu:4298.947364088013 mfu:20.278868192457125 +Step 11275 | loss:1.0410176515579224 lr:7.822866864553487e-05 tokens_per_second_per_gpu:3966.1713889626567 mfu:18.709107140354757 +Step 11276 | loss:0.8905699253082275 lr:7.822508949617118e-05 tokens_per_second_per_gpu:4517.259299836013 mfu:21.30868284123756 +Step 11277 | loss:1.1374820470809937 lr:7.822151013452298e-05 tokens_per_second_per_gpu:3779.5281988158845 mfu:17.828679367820484 +Step 11278 | loss:1.0207401514053345 lr:7.82179305606172e-05 tokens_per_second_per_gpu:5144.452796536344 mfu:24.267261575418104 +Step 11279 | loss:1.0253797769546509 lr:7.821435077448076e-05 tokens_per_second_per_gpu:4132.784668884103 mfu:19.495050408901093 +Step 11280 | loss:1.0039410591125488 lr:7.821077077614061e-05 tokens_per_second_per_gpu:5119.824188792143 mfu:24.15108423061492 +Step 11281 | loss:1.153118371963501 lr:7.820719056562362e-05 tokens_per_second_per_gpu:5336.738156215078 mfu:25.174304424288028 +Step 11282 | loss:1.067642092704773 lr:7.820361014295678e-05 tokens_per_second_per_gpu:4774.244290067915 mfu:22.520924886322046 +Step 11283 | loss:1.1584720611572266 lr:7.820002950816698e-05 tokens_per_second_per_gpu:4751.732011808874 mfu:22.414730628783406 +Step 11284 | loss:1.1492637395858765 lr:7.819644866128116e-05 tokens_per_second_per_gpu:5298.237979068609 mfu:24.992692519880563 +Step 11285 | loss:1.0336110591888428 lr:7.819286760232624e-05 tokens_per_second_per_gpu:5297.269038816229 mfu:24.98812186339183 +Step 11286 | loss:1.0142865180969238 lr:7.818928633132918e-05 tokens_per_second_per_gpu:5357.237756449987 mfu:25.271004536188553 +Step 11287 | loss:1.0967323780059814 lr:7.81857048483169e-05 tokens_per_second_per_gpu:5199.909929113782 mfu:24.528862331748382 +Step 11288 | loss:1.1475666761398315 lr:7.818212315331634e-05 tokens_per_second_per_gpu:3225.35497031748 mfu:15.214549697290865 +Step 11289 | loss:1.0517282485961914 lr:7.817854124635443e-05 tokens_per_second_per_gpu:5354.593466789407 mfu:25.25853097069726 +Step 11290 | loss:1.1326887607574463 lr:7.81749591274581e-05 tokens_per_second_per_gpu:4948.65590137007 mfu:23.343654214523713 +Step 11291 | loss:1.087640404701233 lr:7.817137679665434e-05 tokens_per_second_per_gpu:5043.550076869135 mfu:23.791286230968666 +Step 11292 | loss:1.1049665212631226 lr:7.816779425397003e-05 tokens_per_second_per_gpu:5349.721977465147 mfu:25.235551324392937 +Step 11293 | loss:0.9415966868400574 lr:7.816421149943216e-05 tokens_per_second_per_gpu:4296.672875922362 mfu:20.268139043712054 +Step 11294 | loss:1.0497264862060547 lr:7.816062853306766e-05 tokens_per_second_per_gpu:3345.272100896796 mfu:15.78021926220574 +Step 11295 | loss:1.028878092765808 lr:7.815704535490347e-05 tokens_per_second_per_gpu:4641.63401146407 mfu:21.895379576501096 +Step 11296 | loss:1.1330773830413818 lr:7.815346196496656e-05 tokens_per_second_per_gpu:4596.560959029427 mfu:21.682762297911147 +Step 11297 | loss:1.1835259199142456 lr:7.814987836328387e-05 tokens_per_second_per_gpu:5342.031461032006 mfu:25.19927384624041 +Step 11298 | loss:1.0592749118804932 lr:7.814629454988235e-05 tokens_per_second_per_gpu:4656.285916020276 mfu:21.964495110165416 +Step 11299 | loss:1.0777301788330078 lr:7.814271052478896e-05 tokens_per_second_per_gpu:5401.00443326101 mfu:25.477459418071223 +Step 11300 | loss:1.0863839387893677 lr:7.813912628803066e-05 tokens_per_second_per_gpu:5331.640460149698 mfu:25.150257722190997 +Step 11301 | loss:1.1666717529296875 lr:7.813554183963438e-05 tokens_per_second_per_gpu:4904.970152889604 mfu:23.137581085384078 +Step 11302 | loss:1.055058479309082 lr:7.813195717962712e-05 tokens_per_second_per_gpu:5369.902687719775 mfu:25.330747177848238 +Step 11303 | loss:1.0870863199234009 lr:7.812837230803581e-05 tokens_per_second_per_gpu:3862.581652305589 mfu:18.22045667831179 +Step 11304 | loss:1.1752139329910278 lr:7.812478722488741e-05 tokens_per_second_per_gpu:4765.720563137543 mfu:22.480717012093702 +Step 11305 | loss:1.0286366939544678 lr:7.812120193020892e-05 tokens_per_second_per_gpu:4682.170476453067 mfu:22.0865969980883 +Step 11306 | loss:1.0132980346679688 lr:7.811761642402726e-05 tokens_per_second_per_gpu:4632.68960249496 mfu:21.85318727332027 +Step 11307 | loss:0.9456064701080322 lr:7.811403070636945e-05 tokens_per_second_per_gpu:4275.148102741103 mfu:20.16660301611105 +Step 11308 | loss:1.096930742263794 lr:7.81104447772624e-05 tokens_per_second_per_gpu:5388.949222567443 mfu:25.420592932398634 +Step 11309 | loss:1.3098812103271484 lr:7.810685863673313e-05 tokens_per_second_per_gpu:4706.445366684093 mfu:22.201105797031406 +Step 11310 | loss:1.1248552799224854 lr:7.810327228480856e-05 tokens_per_second_per_gpu:4609.414951777296 mfu:21.743396774820155 +Step 11311 | loss:1.0908045768737793 lr:7.809968572151571e-05 tokens_per_second_per_gpu:4534.4278959940275 mfu:21.38967003857947 +Step 11312 | loss:0.9769838452339172 lr:7.809609894688155e-05 tokens_per_second_per_gpu:5253.423593414948 mfu:24.781295416629607 +Step 11313 | loss:1.1119948625564575 lr:7.809251196093304e-05 tokens_per_second_per_gpu:4918.381392503067 mfu:23.20084419083421 +Step 11314 | loss:1.1235637664794922 lr:7.808892476369714e-05 tokens_per_second_per_gpu:4966.80788783175 mfu:23.429280231711708 +Step 11315 | loss:1.1421300172805786 lr:7.808533735520087e-05 tokens_per_second_per_gpu:3671.1835936066414 mfu:17.31759937955267 +Step 11316 | loss:1.117954969406128 lr:7.80817497354712e-05 tokens_per_second_per_gpu:5082.4423041450045 mfu:23.974747502727165 +Step 11317 | loss:0.969099760055542 lr:7.80781619045351e-05 tokens_per_second_per_gpu:5350.181624554801 mfu:25.237719558138718 +Step 11318 | loss:1.0565073490142822 lr:7.807457386241955e-05 tokens_per_second_per_gpu:5373.130175716217 mfu:25.345971789393694 +Step 11319 | loss:1.1099799871444702 lr:7.807098560915157e-05 tokens_per_second_per_gpu:5338.367731202981 mfu:25.181991407540135 +Step 11320 | loss:1.0811337232589722 lr:7.80673971447581e-05 tokens_per_second_per_gpu:5359.080274943256 mfu:25.279696010287378 +Step 11321 | loss:1.193691611289978 lr:7.806380846926617e-05 tokens_per_second_per_gpu:3939.4283560164586 mfu:18.582955691115615 +Step 11322 | loss:1.1249926090240479 lr:7.806021958270275e-05 tokens_per_second_per_gpu:4216.987398885723 mfu:19.892249052785896 +Step 11323 | loss:1.0631651878356934 lr:7.805663048509486e-05 tokens_per_second_per_gpu:5313.6457893028155 mfu:25.065373789598816 +Step 11324 | loss:1.104822039604187 lr:7.805304117646945e-05 tokens_per_second_per_gpu:5192.480663163968 mfu:24.49381721669963 +Step 11325 | loss:0.9717020392417908 lr:7.804945165685353e-05 tokens_per_second_per_gpu:3965.997693120504 mfu:18.70828778743176 +Step 11326 | loss:1.125205159187317 lr:7.804586192627412e-05 tokens_per_second_per_gpu:4167.042903862605 mfu:19.6566523483523 +Step 11327 | loss:1.0661299228668213 lr:7.804227198475822e-05 tokens_per_second_per_gpu:5316.647245585393 mfu:25.079532170984983 +Step 11328 | loss:1.1843608617782593 lr:7.803868183233278e-05 tokens_per_second_per_gpu:5262.339144738851 mfu:24.82335159337437 +Step 11329 | loss:1.1175310611724854 lr:7.803509146902485e-05 tokens_per_second_per_gpu:3919.128495262697 mfu:18.487197784426733 +Step 11330 | loss:1.1333752870559692 lr:7.803150089486144e-05 tokens_per_second_per_gpu:4981.294271461007 mfu:23.497614974922826 +Step 11331 | loss:1.0444636344909668 lr:7.80279101098695e-05 tokens_per_second_per_gpu:3920.141020782399 mfu:18.491974040057837 +Step 11332 | loss:1.0494098663330078 lr:7.802431911407609e-05 tokens_per_second_per_gpu:5363.0825478710185 mfu:25.29857541454581 +Step 11333 | loss:1.0524487495422363 lr:7.80207279075082e-05 tokens_per_second_per_gpu:5307.632246476624 mfu:25.037006882071463 +Step 11334 | loss:1.2010986804962158 lr:7.801713649019284e-05 tokens_per_second_per_gpu:4844.787422582474 mfu:22.853688878292733 +Step 11335 | loss:1.0659502744674683 lr:7.801354486215702e-05 tokens_per_second_per_gpu:5376.361289379044 mfu:25.361213503826086 +Step 11336 | loss:1.101081371307373 lr:7.800995302342776e-05 tokens_per_second_per_gpu:4715.513510580529 mfu:22.243881778975787 +Step 11337 | loss:1.0682998895645142 lr:7.800636097403205e-05 tokens_per_second_per_gpu:5317.690041740505 mfu:25.084451218368017 +Step 11338 | loss:1.1078906059265137 lr:7.800276871399695e-05 tokens_per_second_per_gpu:3655.7849785896897 mfu:17.24496148524307 +Step 11339 | loss:1.389190912246704 lr:7.799917624334942e-05 tokens_per_second_per_gpu:3410.0966582610577 mfu:16.08600775950294 +Step 11340 | loss:1.0854198932647705 lr:7.799558356211653e-05 tokens_per_second_per_gpu:4758.902461399359 mfu:22.448554862906576 +Step 11341 | loss:1.0414873361587524 lr:7.79919906703253e-05 tokens_per_second_per_gpu:4671.351312951965 mfu:22.03556115791396 +Step 11342 | loss:0.8995508551597595 lr:7.798839756800273e-05 tokens_per_second_per_gpu:4600.645041999432 mfu:21.702027614097062 +Step 11343 | loss:1.1201854944229126 lr:7.798480425517585e-05 tokens_per_second_per_gpu:5000.482967865459 mfu:23.588131329791825 +Step 11344 | loss:1.0151551961898804 lr:7.798121073187167e-05 tokens_per_second_per_gpu:3837.7214771130352 mfu:18.103186990345247 +Step 11345 | loss:1.1010891199111938 lr:7.797761699811726e-05 tokens_per_second_per_gpu:4917.626039092155 mfu:23.19728105991836 +Step 11346 | loss:1.1103825569152832 lr:7.79740230539396e-05 tokens_per_second_per_gpu:5164.2857351443 mfu:24.360816930680087 +Step 11347 | loss:1.06023371219635 lr:7.797042889936574e-05 tokens_per_second_per_gpu:3958.110841414991 mfu:18.67108416230119 +Step 11348 | loss:1.062237024307251 lr:7.796683453442273e-05 tokens_per_second_per_gpu:5352.861798507619 mfu:25.250362395958994 +Step 11349 | loss:1.1773216724395752 lr:7.796323995913759e-05 tokens_per_second_per_gpu:3913.0404211960454 mfu:18.45847930032179 +Step 11350 | loss:1.0568705797195435 lr:7.795964517353735e-05 tokens_per_second_per_gpu:5365.2864968570875 mfu:25.30897181794557 +Step 11351 | loss:1.0381710529327393 lr:7.795605017764905e-05 tokens_per_second_per_gpu:4868.129613599773 mfu:22.963797975909788 +Step 11352 | loss:0.9627624154090881 lr:7.795245497149973e-05 tokens_per_second_per_gpu:4578.333476355378 mfu:21.596780152208424 +Step 11353 | loss:1.0765233039855957 lr:7.794885955511642e-05 tokens_per_second_per_gpu:4122.707463562285 mfu:19.447514512049555 +Step 11354 | loss:1.0265858173370361 lr:7.79452639285262e-05 tokens_per_second_per_gpu:5030.734746634682 mfu:23.730834131731946 +Step 11355 | loss:0.9429463148117065 lr:7.794166809175605e-05 tokens_per_second_per_gpu:5156.7258779974845 mfu:24.325155794673925 +Step 11356 | loss:1.1556804180145264 lr:7.793807204483307e-05 tokens_per_second_per_gpu:4415.112105680179 mfu:20.826837098296036 +Step 11357 | loss:1.0708274841308594 lr:7.793447578778427e-05 tokens_per_second_per_gpu:4081.940935649123 mfu:19.25521184439628 +Step 11358 | loss:0.9493144154548645 lr:7.793087932063672e-05 tokens_per_second_per_gpu:5079.270141076454 mfu:23.959783868305987 +Step 11359 | loss:1.081791639328003 lr:7.792728264341746e-05 tokens_per_second_per_gpu:5180.501729056714 mfu:24.437310540698803 +Step 11360 | loss:1.1823476552963257 lr:7.792368575615356e-05 tokens_per_second_per_gpu:5325.110057668921 mfu:25.11945269948762 +Step 11361 | loss:1.1095483303070068 lr:7.792008865887203e-05 tokens_per_second_per_gpu:5070.394888024907 mfu:23.917917785387523 +Step 11362 | loss:1.0233979225158691 lr:7.791649135159997e-05 tokens_per_second_per_gpu:4053.3258638667653 mfu:19.120229668564214 +Step 11363 | loss:1.18533194065094 lr:7.791289383436442e-05 tokens_per_second_per_gpu:5084.533462388036 mfu:23.984611852947936 +Step 11364 | loss:1.0172501802444458 lr:7.790929610719242e-05 tokens_per_second_per_gpu:4505.979476481632 mfu:21.25547398993863 +Step 11365 | loss:1.0580867528915405 lr:7.790569817011105e-05 tokens_per_second_per_gpu:5266.667741211024 mfu:24.84377032146832 +Step 11366 | loss:1.0854971408843994 lr:7.790210002314734e-05 tokens_per_second_per_gpu:5072.6096409558195 mfu:23.928365152838452 +Step 11367 | loss:1.2695852518081665 lr:7.789850166632841e-05 tokens_per_second_per_gpu:3056.31934095982 mfu:14.417179793158478 +Step 11368 | loss:1.1297447681427002 lr:7.789490309968127e-05 tokens_per_second_per_gpu:5311.591168808342 mfu:25.055681794171985 +Step 11369 | loss:1.229034423828125 lr:7.789130432323301e-05 tokens_per_second_per_gpu:3422.470514393936 mfu:16.144377350079285 +Step 11370 | loss:1.301984429359436 lr:7.78877053370107e-05 tokens_per_second_per_gpu:4108.967659282487 mfu:19.382701511009408 +Step 11371 | loss:1.1010210514068604 lr:7.788410614104138e-05 tokens_per_second_per_gpu:4111.019425746106 mfu:19.392380043485755 +Step 11372 | loss:1.0200151205062866 lr:7.788050673535216e-05 tokens_per_second_per_gpu:5381.108357635975 mfu:25.38360623472739 +Step 11373 | loss:1.2263455390930176 lr:7.787690711997007e-05 tokens_per_second_per_gpu:4724.358852580477 mfu:22.285606766360505 +Step 11374 | loss:1.0598008632659912 lr:7.787330729492222e-05 tokens_per_second_per_gpu:4419.880335474567 mfu:20.849329651780057 +Step 11375 | loss:1.0782139301300049 lr:7.786970726023566e-05 tokens_per_second_per_gpu:5203.457074738385 mfu:24.545594822865084 +Step 11376 | loss:1.1114134788513184 lr:7.786610701593748e-05 tokens_per_second_per_gpu:3160.4640517472085 mfu:14.908448162862525 +Step 11377 | loss:1.0854427814483643 lr:7.786250656205476e-05 tokens_per_second_per_gpu:5399.041584137374 mfu:25.468200323858444 +Step 11378 | loss:1.2255343198776245 lr:7.785890589861456e-05 tokens_per_second_per_gpu:3646.01381525184 mfu:17.198869240646058 +Step 11379 | loss:1.2311393022537231 lr:7.785530502564399e-05 tokens_per_second_per_gpu:4851.55454878702 mfu:22.885610567190362 +Step 11380 | loss:1.2232455015182495 lr:7.78517039431701e-05 tokens_per_second_per_gpu:5349.717689140327 mfu:25.2355310956331 +Step 11381 | loss:1.0417839288711548 lr:7.784810265122001e-05 tokens_per_second_per_gpu:4354.741779260998 mfu:20.542059968336943 +Step 11382 | loss:1.034574031829834 lr:7.784450114982076e-05 tokens_per_second_per_gpu:5331.778630060397 mfu:25.150909493233982 +Step 11383 | loss:1.0490264892578125 lr:7.784089943899948e-05 tokens_per_second_per_gpu:5096.6425297131045 mfu:24.04173238954021 +Step 11384 | loss:1.090700387954712 lr:7.783729751878323e-05 tokens_per_second_per_gpu:5354.933119853069 mfu:25.26013317214934 +Step 11385 | loss:1.0935027599334717 lr:7.783369538919913e-05 tokens_per_second_per_gpu:3514.142228554066 mfu:16.576808466579884 +Step 11386 | loss:1.1057772636413574 lr:7.783009305027425e-05 tokens_per_second_per_gpu:4334.380980354956 mfu:20.446014605986633 +Step 11387 | loss:1.2382826805114746 lr:7.78264905020357e-05 tokens_per_second_per_gpu:4203.533120490564 mfu:19.82878292605933 +Step 11388 | loss:1.0883820056915283 lr:7.782288774451055e-05 tokens_per_second_per_gpu:5018.417742679463 mfu:23.672732722577607 +Step 11389 | loss:1.1208266019821167 lr:7.78192847777259e-05 tokens_per_second_per_gpu:5132.0088125555 mfu:24.20856117981786 +Step 11390 | loss:1.05806565284729 lr:7.781568160170888e-05 tokens_per_second_per_gpu:5329.183416523732 mfu:25.138667428192445 +Step 11391 | loss:1.0175834894180298 lr:7.781207821648657e-05 tokens_per_second_per_gpu:5337.043725214478 mfu:25.17574584539365 +Step 11392 | loss:1.1210311651229858 lr:7.780847462208607e-05 tokens_per_second_per_gpu:3885.0856498625585 mfu:18.32661187436592 +Step 11393 | loss:1.2072182893753052 lr:7.780487081853447e-05 tokens_per_second_per_gpu:3784.0560083964338 mfu:17.85003781813591 +Step 11394 | loss:0.9357511401176453 lr:7.780126680585888e-05 tokens_per_second_per_gpu:5203.347730999688 mfu:24.545079029793687 +Step 11395 | loss:1.0583633184432983 lr:7.779766258408645e-05 tokens_per_second_per_gpu:5368.111745555449 mfu:25.32229899809287 +Step 11396 | loss:0.9734466075897217 lr:7.779405815324423e-05 tokens_per_second_per_gpu:5324.026987913145 mfu:25.114343674658862 +Step 11397 | loss:1.14704430103302 lr:7.779045351335937e-05 tokens_per_second_per_gpu:5329.7116533384005 mfu:25.141159211373733 +Step 11398 | loss:1.2613422870635986 lr:7.778684866445893e-05 tokens_per_second_per_gpu:4481.85405862731 mfu:21.141670277699053 +Step 11399 | loss:1.0049043893814087 lr:7.77832436065701e-05 tokens_per_second_per_gpu:5161.0777476278445 mfu:24.345684306225856 +Step 11400 | loss:1.2137352228164673 lr:7.777963833971991e-05 tokens_per_second_per_gpu:3795.7935955615626 mfu:17.905406019432696 +Step 11401 | loss:1.165273666381836 lr:7.777603286393553e-05 tokens_per_second_per_gpu:4088.3782251514303 mfu:19.285577637293507 +Step 11402 | loss:1.1180572509765625 lr:7.777242717924405e-05 tokens_per_second_per_gpu:4867.073615912864 mfu:22.958816654648853 +Step 11403 | loss:1.127020001411438 lr:7.776882128567263e-05 tokens_per_second_per_gpu:5312.178766134607 mfu:25.058453591014416 +Step 11404 | loss:1.238810420036316 lr:7.776521518324833e-05 tokens_per_second_per_gpu:3334.4818835224733 mfu:15.729319965850138 +Step 11405 | loss:0.9833065867424011 lr:7.776160887199833e-05 tokens_per_second_per_gpu:4487.653518995774 mfu:21.16902732174707 +Step 11406 | loss:0.9868090152740479 lr:7.775800235194973e-05 tokens_per_second_per_gpu:3950.834645195234 mfu:18.63676110328551 +Step 11407 | loss:1.0694308280944824 lr:7.775439562312962e-05 tokens_per_second_per_gpu:3759.336041486864 mfu:17.733429516562204 +Step 11408 | loss:1.0673792362213135 lr:7.775078868556517e-05 tokens_per_second_per_gpu:5173.554225991704 mfu:24.404537983374897 +Step 11409 | loss:1.1359329223632812 lr:7.774718153928351e-05 tokens_per_second_per_gpu:5310.232719959664 mfu:25.049273759177492 +Step 11410 | loss:1.0434107780456543 lr:7.774357418431175e-05 tokens_per_second_per_gpu:3956.6352978491495 mfu:18.66412377154743 +Step 11411 | loss:1.1502532958984375 lr:7.773996662067704e-05 tokens_per_second_per_gpu:4100.291234388591 mfu:19.341773334434166 +Step 11412 | loss:0.9790951609611511 lr:7.773635884840649e-05 tokens_per_second_per_gpu:4445.788128100772 mfu:20.971541129923096 +Step 11413 | loss:1.0353864431381226 lr:7.773275086752725e-05 tokens_per_second_per_gpu:5103.511329485052 mfu:24.074133690002572 +Step 11414 | loss:1.1335524320602417 lr:7.772914267806645e-05 tokens_per_second_per_gpu:4073.5524333908124 mfu:19.21564184801776 +Step 11415 | loss:1.067530870437622 lr:7.772553428005122e-05 tokens_per_second_per_gpu:4663.5372453633845 mfu:21.998700867881222 +Step 11416 | loss:1.108300805091858 lr:7.772192567350872e-05 tokens_per_second_per_gpu:5185.658415390529 mfu:24.46163550995682 +Step 11417 | loss:1.0391145944595337 lr:7.771831685846609e-05 tokens_per_second_per_gpu:4341.975801515123 mfu:20.48184067320925 +Step 11418 | loss:1.0818034410476685 lr:7.771470783495045e-05 tokens_per_second_per_gpu:4550.990948996156 mfu:21.467800785538753 +Step 11419 | loss:1.1539959907531738 lr:7.771109860298895e-05 tokens_per_second_per_gpu:4266.562542082803 mfu:20.126103461638152 +Step 11420 | loss:1.1225091218948364 lr:7.770748916260875e-05 tokens_per_second_per_gpu:3558.0284650013755 mfu:16.783827331665893 +Step 11421 | loss:1.0861907005310059 lr:7.770387951383697e-05 tokens_per_second_per_gpu:3573.359332816381 mfu:16.85614565086505 +Step 11422 | loss:1.065102458000183 lr:7.770026965670081e-05 tokens_per_second_per_gpu:5002.60674497603 mfu:23.598149548776135 +Step 11423 | loss:1.0943883657455444 lr:7.769665959122736e-05 tokens_per_second_per_gpu:5327.362906642033 mfu:25.13007977246945 +Step 11424 | loss:1.133155345916748 lr:7.769304931744381e-05 tokens_per_second_per_gpu:4047.4773705732655 mfu:19.092641327842923 +Step 11425 | loss:1.1560651063919067 lr:7.768943883537732e-05 tokens_per_second_per_gpu:3924.6491309777934 mfu:18.513239565012423 +Step 11426 | loss:1.1162468194961548 lr:7.7685828145055e-05 tokens_per_second_per_gpu:4290.283169147767 mfu:20.237997706659893 +Step 11427 | loss:1.0235252380371094 lr:7.768221724650404e-05 tokens_per_second_per_gpu:5364.491054801039 mfu:25.30521958205078 +Step 11428 | loss:1.0863546133041382 lr:7.76786061397516e-05 tokens_per_second_per_gpu:4855.411163682921 mfu:22.90380287766068 +Step 11429 | loss:1.1644456386566162 lr:7.767499482482482e-05 tokens_per_second_per_gpu:3541.9615696499154 mfu:16.708037044998036 +Step 11430 | loss:0.9542574882507324 lr:7.767138330175088e-05 tokens_per_second_per_gpu:5336.582994602629 mfu:25.173572500488806 +Step 11431 | loss:0.9905668497085571 lr:7.766777157055693e-05 tokens_per_second_per_gpu:4866.89573036024 mfu:22.95797753814672 +Step 11432 | loss:1.1653579473495483 lr:7.766415963127014e-05 tokens_per_second_per_gpu:4105.133432210239 mfu:19.36461480772302 +Step 11433 | loss:1.1819963455200195 lr:7.766054748391768e-05 tokens_per_second_per_gpu:4874.287959687718 mfu:22.992847945128815 +Step 11434 | loss:1.1388767957687378 lr:7.765693512852671e-05 tokens_per_second_per_gpu:3981.004096977967 mfu:18.779075554783926 +Step 11435 | loss:1.0780704021453857 lr:7.765332256512442e-05 tokens_per_second_per_gpu:5057.093386179037 mfu:23.855172331710317 +Step 11436 | loss:0.9342321157455444 lr:7.764970979373793e-05 tokens_per_second_per_gpu:3880.6951282158507 mfu:18.305901034658163 +Step 11437 | loss:1.1435635089874268 lr:7.764609681439448e-05 tokens_per_second_per_gpu:4120.164526031731 mfu:19.43551904184805 +Step 11438 | loss:0.866002082824707 lr:7.764248362712118e-05 tokens_per_second_per_gpu:5376.928056790098 mfu:25.363887042404517 +Step 11439 | loss:1.0487276315689087 lr:7.763887023194525e-05 tokens_per_second_per_gpu:4156.388395635416 mfu:19.60639321519809 +Step 11440 | loss:1.158862590789795 lr:7.763525662889383e-05 tokens_per_second_per_gpu:5341.496487091591 mfu:25.196750282887507 +Step 11441 | loss:1.0737125873565674 lr:7.763164281799414e-05 tokens_per_second_per_gpu:3146.872982128195 mfu:14.844336768594047 +Step 11442 | loss:1.0552074909210205 lr:7.762802879927333e-05 tokens_per_second_per_gpu:5000.55421076451 mfu:23.58846739470194 +Step 11443 | loss:1.0612715482711792 lr:7.762441457275858e-05 tokens_per_second_per_gpu:4703.696776285873 mfu:22.188140227165125 +Step 11444 | loss:1.1653084754943848 lr:7.76208001384771e-05 tokens_per_second_per_gpu:5345.166350469026 mfu:25.214061654582125 +Step 11445 | loss:1.0917545557022095 lr:7.761718549645604e-05 tokens_per_second_per_gpu:3975.5977919660972 mfu:18.753573091632187 +Step 11446 | loss:1.0404670238494873 lr:7.761357064672263e-05 tokens_per_second_per_gpu:5335.707403336412 mfu:25.169442187094788 +Step 11447 | loss:1.166242003440857 lr:7.760995558930402e-05 tokens_per_second_per_gpu:4130.631572501986 mfu:19.48489388591052 +Step 11448 | loss:1.0780469179153442 lr:7.76063403242274e-05 tokens_per_second_per_gpu:4119.490465733746 mfu:19.432339384415908 +Step 11449 | loss:1.0903401374816895 lr:7.760272485151999e-05 tokens_per_second_per_gpu:5357.2533837328365 mfu:25.271078252748044 +Step 11450 | loss:0.9859234094619751 lr:7.759910917120894e-05 tokens_per_second_per_gpu:5343.561231195234 mfu:25.206490033106057 +Step 11451 | loss:1.098844051361084 lr:7.759549328332149e-05 tokens_per_second_per_gpu:4247.6798504075 mfu:20.037030583286338 +Step 11452 | loss:1.0401846170425415 lr:7.759187718788478e-05 tokens_per_second_per_gpu:3637.215574384298 mfu:17.15736643733918 +Step 11453 | loss:1.178807020187378 lr:7.758826088492609e-05 tokens_per_second_per_gpu:4587.517413088515 mfu:21.640102348719818 +Step 11454 | loss:1.148250937461853 lr:7.758464437447254e-05 tokens_per_second_per_gpu:4337.4998781107015 mfu:20.46072697883931 +Step 11455 | loss:1.1521660089492798 lr:7.758102765655137e-05 tokens_per_second_per_gpu:4669.450968027505 mfu:22.026596906673007 +Step 11456 | loss:1.0231317281723022 lr:7.757741073118976e-05 tokens_per_second_per_gpu:3867.7644982042875 mfu:18.244905046700644 +Step 11457 | loss:1.0670510530471802 lr:7.757379359841494e-05 tokens_per_second_per_gpu:4415.532366154847 mfu:20.828819538658877 +Step 11458 | loss:1.1624116897583008 lr:7.757017625825409e-05 tokens_per_second_per_gpu:3822.0119234511467 mfu:18.029082345395878 +Step 11459 | loss:1.0326122045516968 lr:7.756655871073442e-05 tokens_per_second_per_gpu:3543.18090827153 mfu:16.713788873316812 +Step 11460 | loss:1.0243345499038696 lr:7.756294095588317e-05 tokens_per_second_per_gpu:3919.2937469593558 mfu:18.48797730487472 +Step 11461 | loss:0.9773693680763245 lr:7.75593229937275e-05 tokens_per_second_per_gpu:5396.510551402325 mfu:25.4562610106085 +Step 11462 | loss:1.0868576765060425 lr:7.755570482429466e-05 tokens_per_second_per_gpu:5003.636964788338 mfu:23.603009271404417 +Step 11463 | loss:1.1840829849243164 lr:7.755208644761182e-05 tokens_per_second_per_gpu:4288.850034207115 mfu:20.231237364626804 +Step 11464 | loss:1.0894237756729126 lr:7.754846786370626e-05 tokens_per_second_per_gpu:5135.897694394664 mfu:24.226905698964856 +Step 11465 | loss:1.0987656116485596 lr:7.754484907260513e-05 tokens_per_second_per_gpu:5349.983985028193 mfu:25.23678725877093 +Step 11466 | loss:1.0190030336380005 lr:7.754123007433569e-05 tokens_per_second_per_gpu:5335.570848197233 mfu:25.16879803320457 +Step 11467 | loss:1.1417332887649536 lr:7.753761086892513e-05 tokens_per_second_per_gpu:5243.900244804575 mfu:24.736372156383652 +Step 11468 | loss:0.9883108139038086 lr:7.753399145640071e-05 tokens_per_second_per_gpu:5352.225542623755 mfu:25.24736106839836 +Step 11469 | loss:1.129258632659912 lr:7.75303718367896e-05 tokens_per_second_per_gpu:3639.8803600943133 mfu:17.16993668619265 +Step 11470 | loss:1.04945707321167 lr:7.752675201011907e-05 tokens_per_second_per_gpu:4150.376514451613 mfu:19.578034145921393 +Step 11471 | loss:1.2251484394073486 lr:7.752313197641632e-05 tokens_per_second_per_gpu:2781.343544286688 mfu:13.12007204454261 +Step 11472 | loss:1.1662112474441528 lr:7.751951173570858e-05 tokens_per_second_per_gpu:5380.301979569977 mfu:25.3798024118078 +Step 11473 | loss:0.996609628200531 lr:7.751589128802307e-05 tokens_per_second_per_gpu:4047.590348922635 mfu:19.093174265992992 +Step 11474 | loss:1.1941893100738525 lr:7.751227063338704e-05 tokens_per_second_per_gpu:3110.0151895526387 mfu:14.670472272427276 +Step 11475 | loss:1.1241713762283325 lr:7.750864977182772e-05 tokens_per_second_per_gpu:4569.415681671677 mfu:21.55471339315274 +Step 11476 | loss:1.0439599752426147 lr:7.750502870337233e-05 tokens_per_second_per_gpu:5147.095460392407 mfu:24.2797274717121 +Step 11477 | loss:1.075596570968628 lr:7.75014074280481e-05 tokens_per_second_per_gpu:4250.123398469118 mfu:20.04855721640528 +Step 11478 | loss:1.0579053163528442 lr:7.749778594588229e-05 tokens_per_second_per_gpu:4950.529148629648 mfu:23.352490641456086 +Step 11479 | loss:1.054340124130249 lr:7.749416425690212e-05 tokens_per_second_per_gpu:4541.851490625647 mfu:21.42468840105267 +Step 11480 | loss:1.0603654384613037 lr:7.749054236113482e-05 tokens_per_second_per_gpu:3982.8856532199625 mfu:18.787951176604015 +Step 11481 | loss:1.2174794673919678 lr:7.748692025860766e-05 tokens_per_second_per_gpu:3254.7969457630893 mfu:15.353432518786803 +Step 11482 | loss:1.0536892414093018 lr:7.748329794934786e-05 tokens_per_second_per_gpu:3668.7868013807606 mfu:17.306293301688267 +Step 11483 | loss:1.1347144842147827 lr:7.747967543338265e-05 tokens_per_second_per_gpu:4753.4428522882135 mfu:22.422800955223938 +Step 11484 | loss:1.0702450275421143 lr:7.747605271073933e-05 tokens_per_second_per_gpu:5305.188454364231 mfu:25.025479097723377 +Step 11485 | loss:0.9180721640586853 lr:7.747242978144509e-05 tokens_per_second_per_gpu:4291.461851763644 mfu:20.243557753662976 +Step 11486 | loss:1.1799211502075195 lr:7.74688066455272e-05 tokens_per_second_per_gpu:5360.629445088314 mfu:25.2870037101773 +Step 11487 | loss:0.8352433443069458 lr:7.746518330301291e-05 tokens_per_second_per_gpu:5007.395967389022 mfu:23.62074112802382 +Step 11488 | loss:1.021356463432312 lr:7.746155975392948e-05 tokens_per_second_per_gpu:5351.200325026864 mfu:25.242524942073665 +Step 11489 | loss:1.2466754913330078 lr:7.745793599830416e-05 tokens_per_second_per_gpu:4928.096155547571 mfu:23.246670385624626 +Step 11490 | loss:1.2162343263626099 lr:7.745431203616418e-05 tokens_per_second_per_gpu:3673.6946084889423 mfu:17.329444265175802 +Step 11491 | loss:0.8957479596138 lr:7.745068786753681e-05 tokens_per_second_per_gpu:4787.384800750938 mfu:22.582910917216175 +Step 11492 | loss:1.034811019897461 lr:7.744706349244934e-05 tokens_per_second_per_gpu:5364.9428705089795 mfu:25.307350873832842 +Step 11493 | loss:1.2092790603637695 lr:7.7443438910929e-05 tokens_per_second_per_gpu:4556.373911877108 mfu:21.49319313987638 +Step 11494 | loss:1.0543761253356934 lr:7.743981412300305e-05 tokens_per_second_per_gpu:4783.419354920877 mfu:22.56420523266095 +Step 11495 | loss:1.158748984336853 lr:7.743618912869874e-05 tokens_per_second_per_gpu:3590.4001002815708 mfu:16.936529858453085 +Step 11496 | loss:1.0941613912582397 lr:7.743256392804337e-05 tokens_per_second_per_gpu:5231.97163969014 mfu:24.680102890828966 +Step 11497 | loss:1.2488597631454468 lr:7.742893852106419e-05 tokens_per_second_per_gpu:3735.312254448615 mfu:17.62010521422205 +Step 11498 | loss:1.0097784996032715 lr:7.742531290778845e-05 tokens_per_second_per_gpu:3849.7087645284096 mfu:18.1597330703257 +Step 11499 | loss:1.0810964107513428 lr:7.742168708824343e-05 tokens_per_second_per_gpu:4251.088166013389 mfu:20.05310818951792 +Step 11500 | loss:1.0712040662765503 lr:7.741806106245641e-05 tokens_per_second_per_gpu:5333.759228053166 mfu:25.160252311890737 +Step 11501 | loss:0.9644748568534851 lr:7.741443483045466e-05 tokens_per_second_per_gpu:5322.960227792814 mfu:25.10931158516322 +Step 11502 | loss:1.0535439252853394 lr:7.741080839226543e-05 tokens_per_second_per_gpu:4033.9671007336365 mfu:19.028911079919663 +Step 11503 | loss:0.872338056564331 lr:7.740718174791602e-05 tokens_per_second_per_gpu:3557.2101197993243 mfu:16.77996705772373 +Step 11504 | loss:0.9968922734260559 lr:7.740355489743371e-05 tokens_per_second_per_gpu:4190.714071002412 mfu:19.76831328246834 +Step 11505 | loss:1.2019356489181519 lr:7.739992784084575e-05 tokens_per_second_per_gpu:4856.1784041445535 mfu:22.90742208182292 +Step 11506 | loss:0.9484568238258362 lr:7.739630057817945e-05 tokens_per_second_per_gpu:5088.536657270542 mfu:24.00349560622419 +Step 11507 | loss:1.1824727058410645 lr:7.739267310946208e-05 tokens_per_second_per_gpu:3583.096940658732 mfu:16.902079608464636 +Step 11508 | loss:1.0003437995910645 lr:7.738904543472091e-05 tokens_per_second_per_gpu:4323.505770934327 mfu:20.394714387647465 +Step 11509 | loss:1.1008398532867432 lr:7.738541755398323e-05 tokens_per_second_per_gpu:5087.393457380679 mfu:23.998102937293694 +Step 11510 | loss:1.0637058019638062 lr:7.738178946727634e-05 tokens_per_second_per_gpu:4438.091604620292 mfu:20.935235315503355 +Step 11511 | loss:1.0739130973815918 lr:7.737816117462752e-05 tokens_per_second_per_gpu:4395.734245034085 mfu:20.73542842342446 +Step 11512 | loss:1.1374982595443726 lr:7.737453267606405e-05 tokens_per_second_per_gpu:5381.384076140601 mfu:25.38490684595614 +Step 11513 | loss:1.017961025238037 lr:7.737090397161323e-05 tokens_per_second_per_gpu:4389.281919742618 mfu:20.704991704144398 +Step 11514 | loss:1.1304612159729004 lr:7.736727506130234e-05 tokens_per_second_per_gpu:5306.179277910919 mfu:25.030152981445035 +Step 11515 | loss:1.2222357988357544 lr:7.73636459451587e-05 tokens_per_second_per_gpu:3621.091007795259 mfu:17.081304105603934 +Step 11516 | loss:1.2087371349334717 lr:7.736001662320957e-05 tokens_per_second_per_gpu:3302.1357779557816 mfu:15.57673786707736 +Step 11517 | loss:1.1002261638641357 lr:7.735638709548226e-05 tokens_per_second_per_gpu:5382.186831633607 mfu:25.388693580580856 +Step 11518 | loss:0.9793661236763 lr:7.735275736200409e-05 tokens_per_second_per_gpu:4793.433512713498 mfu:22.61144372356025 +Step 11519 | loss:1.0798966884613037 lr:7.734912742280234e-05 tokens_per_second_per_gpu:4965.276160816197 mfu:23.422054814039765 +Step 11520 | loss:1.0629160404205322 lr:7.73454972779043e-05 tokens_per_second_per_gpu:4079.0542194135423 mfu:19.24159471138847 +Step 11521 | loss:1.13270902633667 lr:7.73418669273373e-05 tokens_per_second_per_gpu:5293.695394867002 mfu:24.971264375156792 +Step 11522 | loss:0.966053307056427 lr:7.733823637112863e-05 tokens_per_second_per_gpu:5353.824054601377 mfu:25.254901522131064 +Step 11523 | loss:1.0307108163833618 lr:7.733460560930557e-05 tokens_per_second_per_gpu:4642.965451753458 mfu:21.901660207513277 +Step 11524 | loss:1.165257215499878 lr:7.733097464189548e-05 tokens_per_second_per_gpu:5314.136814160147 mfu:25.067690037628328 +Step 11525 | loss:1.2013338804244995 lr:7.732734346892562e-05 tokens_per_second_per_gpu:4239.87813488476 mfu:20.00022856005564 +Step 11526 | loss:1.0575735569000244 lr:7.732371209042333e-05 tokens_per_second_per_gpu:5313.804537860073 mfu:25.066122633628694 +Step 11527 | loss:1.0620510578155518 lr:7.732008050641593e-05 tokens_per_second_per_gpu:3726.8301096826626 mfu:17.580093490157925 +Step 11528 | loss:1.0807219743728638 lr:7.731644871693071e-05 tokens_per_second_per_gpu:5368.419917019699 mfu:25.323752695469214 +Step 11529 | loss:0.993831217288971 lr:7.7312816721995e-05 tokens_per_second_per_gpu:4201.326754782265 mfu:19.818375122568472 +Step 11530 | loss:1.0297478437423706 lr:7.730918452163609e-05 tokens_per_second_per_gpu:4615.878542051726 mfu:21.77388663294686 +Step 11531 | loss:1.061928629875183 lr:7.730555211588134e-05 tokens_per_second_per_gpu:5358.7244971860655 mfu:25.278017745904766 +Step 11532 | loss:1.19441556930542 lr:7.730191950475803e-05 tokens_per_second_per_gpu:3696.7823880093024 mfu:17.43835326035472 +Step 11533 | loss:1.17974853515625 lr:7.729828668829352e-05 tokens_per_second_per_gpu:5361.414213062876 mfu:25.29070559460853 +Step 11534 | loss:1.1231001615524292 lr:7.729465366651509e-05 tokens_per_second_per_gpu:5331.243487506137 mfu:25.148385134500806 +Step 11535 | loss:1.1261489391326904 lr:7.729102043945012e-05 tokens_per_second_per_gpu:5317.911897464437 mfu:25.085497749670292 +Step 11536 | loss:1.1251866817474365 lr:7.728738700712588e-05 tokens_per_second_per_gpu:5352.4428798044955 mfu:25.24838628496146 +Step 11537 | loss:1.0017402172088623 lr:7.728375336956972e-05 tokens_per_second_per_gpu:5319.446255967169 mfu:25.092735580516234 +Step 11538 | loss:1.0014872550964355 lr:7.728011952680896e-05 tokens_per_second_per_gpu:3832.504922173967 mfu:18.07857961066161 +Step 11539 | loss:1.0576828718185425 lr:7.727648547887097e-05 tokens_per_second_per_gpu:5354.719513275413 mfu:25.2591255534775 +Step 11540 | loss:1.0509179830551147 lr:7.727285122578303e-05 tokens_per_second_per_gpu:5280.886944130148 mfu:24.91084472768401 +Step 11541 | loss:1.2067673206329346 lr:7.72692167675725e-05 tokens_per_second_per_gpu:5352.081013766348 mfu:25.246679301118807 +Step 11542 | loss:1.115017056465149 lr:7.726558210426671e-05 tokens_per_second_per_gpu:4759.012592381367 mfu:22.449074369538994 +Step 11543 | loss:1.2104368209838867 lr:7.7261947235893e-05 tokens_per_second_per_gpu:4053.236631164825 mfu:19.119808742684494 +Step 11544 | loss:1.1804478168487549 lr:7.72583121624787e-05 tokens_per_second_per_gpu:5296.361101847362 mfu:24.983838969799663 +Step 11545 | loss:1.121137261390686 lr:7.725467688405116e-05 tokens_per_second_per_gpu:2744.791030488144 mfu:12.947647600452937 +Step 11546 | loss:1.141326904296875 lr:7.725104140063772e-05 tokens_per_second_per_gpu:4836.710394848232 mfu:22.815588160387275 +Step 11547 | loss:1.0791114568710327 lr:7.724740571226572e-05 tokens_per_second_per_gpu:4831.763393935699 mfu:22.79225231303747 +Step 11548 | loss:0.9741920828819275 lr:7.72437698189625e-05 tokens_per_second_per_gpu:4587.803494412361 mfu:21.64145184313491 +Step 11549 | loss:1.1700736284255981 lr:7.724013372075543e-05 tokens_per_second_per_gpu:3936.911354230967 mfu:18.571082564248666 +Step 11550 | loss:1.206038475036621 lr:7.723649741767181e-05 tokens_per_second_per_gpu:3875.9418926508934 mfu:18.283479211512645 +Step 11551 | loss:1.0901825428009033 lr:7.723286090973904e-05 tokens_per_second_per_gpu:4932.515165825317 mfu:23.267515611065658 +Step 11552 | loss:1.0043925046920776 lr:7.722922419698444e-05 tokens_per_second_per_gpu:3980.2432901899247 mfu:18.775486699357707 +Step 11553 | loss:1.1534677743911743 lr:7.722558727943536e-05 tokens_per_second_per_gpu:3843.9444052376766 mfu:18.1325416040502 +Step 11554 | loss:0.9495141506195068 lr:7.722195015711917e-05 tokens_per_second_per_gpu:4971.692245499019 mfu:23.452320580185564 +Step 11555 | loss:1.036155104637146 lr:7.721831283006322e-05 tokens_per_second_per_gpu:4260.3884196298295 mfu:20.09697907261356 +Step 11556 | loss:1.2209614515304565 lr:7.721467529829489e-05 tokens_per_second_per_gpu:4725.730788158675 mfu:22.29207842055921 +Step 11557 | loss:1.2165892124176025 lr:7.721103756184147e-05 tokens_per_second_per_gpu:3944.4929737524863 mfu:18.60684635708935 +Step 11558 | loss:1.0174882411956787 lr:7.720739962073041e-05 tokens_per_second_per_gpu:4227.007937641832 mfu:19.939517643778668 +Step 11559 | loss:1.161887288093567 lr:7.7203761474989e-05 tokens_per_second_per_gpu:4720.607579000811 mfu:22.267911368851948 +Step 11560 | loss:0.9874914884567261 lr:7.720012312464461e-05 tokens_per_second_per_gpu:5342.274992934557 mfu:25.20042262777557 +Step 11561 | loss:1.0825194120407104 lr:7.719648456972466e-05 tokens_per_second_per_gpu:5353.9668119342605 mfu:25.25557493282719 +Step 11562 | loss:1.1785616874694824 lr:7.719284581025647e-05 tokens_per_second_per_gpu:5363.61381455009 mfu:25.30108149011443 +Step 11563 | loss:0.9680994749069214 lr:7.71892068462674e-05 tokens_per_second_per_gpu:4711.381658149919 mfu:22.224391126093877 +Step 11564 | loss:1.0075052976608276 lr:7.718556767778484e-05 tokens_per_second_per_gpu:5347.075179855514 mfu:25.223065928478064 +Step 11565 | loss:1.032883882522583 lr:7.718192830483618e-05 tokens_per_second_per_gpu:4861.949732227833 mfu:22.93464642108186 +Step 11566 | loss:1.0640385150909424 lr:7.717828872744875e-05 tokens_per_second_per_gpu:4392.8579605887 mfu:20.721860498950846 +Step 11567 | loss:1.0202490091323853 lr:7.717464894564995e-05 tokens_per_second_per_gpu:5353.282123455467 mfu:25.25234514045282 +Step 11568 | loss:1.1593575477600098 lr:7.717100895946714e-05 tokens_per_second_per_gpu:4791.172050704753 mfu:22.600776021419794 +Step 11569 | loss:1.0505750179290771 lr:7.716736876892772e-05 tokens_per_second_per_gpu:3851.7627318449 mfu:18.169421984600543 +Step 11570 | loss:1.152328610420227 lr:7.716372837405903e-05 tokens_per_second_per_gpu:4748.317722442591 mfu:22.398624843304813 +Step 11571 | loss:1.0633984804153442 lr:7.716008777488847e-05 tokens_per_second_per_gpu:5300.878075637496 mfu:25.005146305842924 +Step 11572 | loss:1.0802104473114014 lr:7.715644697144344e-05 tokens_per_second_per_gpu:4958.234697848441 mfu:23.388839031823395 +Step 11573 | loss:0.9716285467147827 lr:7.715280596375131e-05 tokens_per_second_per_gpu:5325.998894952291 mfu:25.12364549660455 +Step 11574 | loss:1.1282399892807007 lr:7.714916475183947e-05 tokens_per_second_per_gpu:4554.48538001099 mfu:21.484284612847144 +Step 11575 | loss:1.0265655517578125 lr:7.714552333573527e-05 tokens_per_second_per_gpu:5331.417289464321 mfu:25.149204987991226 +Step 11576 | loss:1.2121813297271729 lr:7.714188171546614e-05 tokens_per_second_per_gpu:4771.364650087697 mfu:22.507341133218112 +Step 11577 | loss:1.129239797592163 lr:7.713823989105946e-05 tokens_per_second_per_gpu:4481.755864973661 mfu:21.141207081481227 +Step 11578 | loss:1.0600425004959106 lr:7.71345978625426e-05 tokens_per_second_per_gpu:4533.439197722382 mfu:21.385006180142803 +Step 11579 | loss:1.2040791511535645 lr:7.713095562994296e-05 tokens_per_second_per_gpu:4470.6508275900005 mfu:21.088822725427747 +Step 11580 | loss:1.259852647781372 lr:7.712731319328798e-05 tokens_per_second_per_gpu:5333.4943669859 mfu:25.159002917046486 +Step 11581 | loss:1.0581300258636475 lr:7.712367055260499e-05 tokens_per_second_per_gpu:3438.293267443796 mfu:16.219015975855523 +Step 11582 | loss:1.119382381439209 lr:7.712002770792141e-05 tokens_per_second_per_gpu:4633.932643191949 mfu:21.859050908372577 +Step 11583 | loss:1.077272653579712 lr:7.711638465926466e-05 tokens_per_second_per_gpu:3528.8170059147674 mfu:16.64603189516551 +Step 11584 | loss:0.9887512922286987 lr:7.711274140666209e-05 tokens_per_second_per_gpu:4596.406140884233 mfu:21.682031994305564 +Step 11585 | loss:0.9692074656486511 lr:7.710909795014116e-05 tokens_per_second_per_gpu:4404.1122053222325 mfu:20.774948691531073 +Step 11586 | loss:1.0972615480422974 lr:7.710545428972922e-05 tokens_per_second_per_gpu:4022.5933799374575 mfu:18.975259298367217 +Step 11587 | loss:1.018890380859375 lr:7.710181042545372e-05 tokens_per_second_per_gpu:4496.089695879616 mfu:21.208822207468348 +Step 11588 | loss:0.9813827872276306 lr:7.709816635734205e-05 tokens_per_second_per_gpu:5337.372979414478 mfu:25.17729899363161 +Step 11589 | loss:1.033303141593933 lr:7.70945220854216e-05 tokens_per_second_per_gpu:4761.41571542591 mfu:22.460410310955176 +Step 11590 | loss:1.1239933967590332 lr:7.70908776097198e-05 tokens_per_second_per_gpu:5312.622081897952 mfu:25.06054478710758 +Step 11591 | loss:1.1195971965789795 lr:7.708723293026404e-05 tokens_per_second_per_gpu:3816.90719239928 mfu:18.005002458067295 +Step 11592 | loss:0.9988967180252075 lr:7.708358804708175e-05 tokens_per_second_per_gpu:3979.291427770711 mfu:18.7709966019218 +Step 11593 | loss:1.1318107843399048 lr:7.707994296020033e-05 tokens_per_second_per_gpu:3645.5254711107164 mfu:17.196565638012277 +Step 11594 | loss:1.057971715927124 lr:7.707629766964724e-05 tokens_per_second_per_gpu:4062.400176607702 mfu:19.163034750000616 +Step 11595 | loss:1.0946390628814697 lr:7.707265217544982e-05 tokens_per_second_per_gpu:5144.74741135489 mfu:24.26865132378365 +Step 11596 | loss:0.9776950478553772 lr:7.706900647763554e-05 tokens_per_second_per_gpu:5369.765107709385 mfu:25.330098189465662 +Step 11597 | loss:1.1130611896514893 lr:7.706536057623182e-05 tokens_per_second_per_gpu:4919.810621507872 mfu:23.20758610790149 +Step 11598 | loss:1.049412488937378 lr:7.706171447126607e-05 tokens_per_second_per_gpu:4384.70145946243 mfu:20.683384891495635 +Step 11599 | loss:1.1105259656906128 lr:7.705806816276568e-05 tokens_per_second_per_gpu:3518.7445453287924 mfu:16.598518380042478 +Step 11600 | loss:1.07772958278656 lr:7.705442165075813e-05 tokens_per_second_per_gpu:4393.027909958862 mfu:20.722662179126257 +Step 11601 | loss:1.1019333600997925 lr:7.705077493527084e-05 tokens_per_second_per_gpu:4014.776925175795 mfu:18.938387747631754 +Step 11602 | loss:1.1302082538604736 lr:7.70471280163312e-05 tokens_per_second_per_gpu:3603.034330401906 mfu:16.996127677552646 +Step 11603 | loss:1.0212476253509521 lr:7.704348089396666e-05 tokens_per_second_per_gpu:4800.173702233653 mfu:22.643238347521734 +Step 11604 | loss:1.0308095216751099 lr:7.703983356820465e-05 tokens_per_second_per_gpu:5147.312895761867 mfu:24.280753151448142 +Step 11605 | loss:1.043588399887085 lr:7.703618603907261e-05 tokens_per_second_per_gpu:3969.6366442931585 mfu:18.72545334098226 +Step 11606 | loss:1.021571159362793 lr:7.703253830659795e-05 tokens_per_second_per_gpu:5360.48207237146 mfu:25.28630852792752 +Step 11607 | loss:0.9975489377975464 lr:7.702889037080811e-05 tokens_per_second_per_gpu:5236.438718525054 mfu:24.7011748638555 +Step 11608 | loss:1.0211666822433472 lr:7.702524223173057e-05 tokens_per_second_per_gpu:5381.475779520143 mfu:25.385339426443654 +Step 11609 | loss:1.0506103038787842 lr:7.702159388939272e-05 tokens_per_second_per_gpu:4961.753434369356 mfu:23.405437512350908 +Step 11610 | loss:0.948512077331543 lr:7.7017945343822e-05 tokens_per_second_per_gpu:4564.648370354975 mfu:21.532225172284672 +Step 11611 | loss:1.15959632396698 lr:7.701429659504589e-05 tokens_per_second_per_gpu:2976.9671938332476 mfu:14.042862176290054 +Step 11612 | loss:1.015246868133545 lr:7.701064764309181e-05 tokens_per_second_per_gpu:5321.808617516748 mfu:25.10387924300604 +Step 11613 | loss:1.2799749374389648 lr:7.700699848798717e-05 tokens_per_second_per_gpu:3588.939442677506 mfu:16.929639687320826 +Step 11614 | loss:1.1002471446990967 lr:7.700334912975947e-05 tokens_per_second_per_gpu:4284.6390135695165 mfu:20.211373261805043 +Step 11615 | loss:1.024963617324829 lr:7.699969956843614e-05 tokens_per_second_per_gpu:3713.8625716258157 mfu:17.51892340065364 +Step 11616 | loss:1.2263230085372925 lr:7.699604980404463e-05 tokens_per_second_per_gpu:4080.374809410512 mfu:19.247824159720064 +Step 11617 | loss:1.22568678855896 lr:7.699239983661236e-05 tokens_per_second_per_gpu:4177.887619444257 mfu:19.707808722050224 +Step 11618 | loss:1.1474549770355225 lr:7.69887496661668e-05 tokens_per_second_per_gpu:3955.752654730512 mfu:18.65996019336237 +Step 11619 | loss:0.935769259929657 lr:7.698509929273545e-05 tokens_per_second_per_gpu:4166.36740665549 mfu:19.653465912774628 +Step 11620 | loss:1.0345969200134277 lr:7.69814487163457e-05 tokens_per_second_per_gpu:4069.928223695952 mfu:19.19854583253317 +Step 11621 | loss:1.0799236297607422 lr:7.697779793702503e-05 tokens_per_second_per_gpu:4674.880810093437 mfu:22.052210398124593 +Step 11622 | loss:1.1137733459472656 lr:7.69741469548009e-05 tokens_per_second_per_gpu:4195.422303340493 mfu:19.790522817714344 +Step 11623 | loss:1.178850769996643 lr:7.69704957697008e-05 tokens_per_second_per_gpu:5035.497673103313 mfu:23.75330166852353 +Step 11624 | loss:1.0495476722717285 lr:7.696684438175212e-05 tokens_per_second_per_gpu:4622.601330039183 mfu:21.805599170909602 +Step 11625 | loss:1.0880365371704102 lr:7.696319279098236e-05 tokens_per_second_per_gpu:4472.512907615138 mfu:21.09760647460993 +Step 11626 | loss:1.0590463876724243 lr:7.695954099741902e-05 tokens_per_second_per_gpu:5333.248904988063 mfu:25.15784503092255 +Step 11627 | loss:0.9536842703819275 lr:7.69558890010895e-05 tokens_per_second_per_gpu:5348.732627092204 mfu:25.23088438988319 +Step 11628 | loss:1.1221649646759033 lr:7.695223680202133e-05 tokens_per_second_per_gpu:3488.1111346250736 mfu:16.454015355154223 +Step 11629 | loss:1.0243009328842163 lr:7.694858440024191e-05 tokens_per_second_per_gpu:3744.0067859224387 mfu:17.661118802624113 +Step 11630 | loss:1.1194725036621094 lr:7.694493179577879e-05 tokens_per_second_per_gpu:5343.9102647441505 mfu:25.208136483907563 +Step 11631 | loss:1.136613130569458 lr:7.694127898865937e-05 tokens_per_second_per_gpu:4089.1260572556985 mfu:19.289105288922652 +Step 11632 | loss:1.0943799018859863 lr:7.693762597891116e-05 tokens_per_second_per_gpu:5052.21957314057 mfu:23.83218172404962 +Step 11633 | loss:1.107194185256958 lr:7.693397276656163e-05 tokens_per_second_per_gpu:5346.977283132495 mfu:25.222604132932684 +Step 11634 | loss:1.0345083475112915 lr:7.693031935163826e-05 tokens_per_second_per_gpu:4222.376098089216 mfu:19.91766846633566 +Step 11635 | loss:1.1881576776504517 lr:7.692666573416852e-05 tokens_per_second_per_gpu:5378.776703471509 mfu:25.372607424212383 +Step 11636 | loss:1.0685104131698608 lr:7.692301191417987e-05 tokens_per_second_per_gpu:4713.015660430997 mfu:22.23209899364326 +Step 11637 | loss:1.083526372909546 lr:7.691935789169984e-05 tokens_per_second_per_gpu:4784.140937121428 mfu:22.567609059015567 +Step 11638 | loss:1.0881179571151733 lr:7.691570366675586e-05 tokens_per_second_per_gpu:4975.683842309522 mfu:23.47114962337735 +Step 11639 | loss:1.0884970426559448 lr:7.691204923937546e-05 tokens_per_second_per_gpu:5348.811780130105 mfu:25.23125776826787 +Step 11640 | loss:1.1638271808624268 lr:7.69083946095861e-05 tokens_per_second_per_gpu:4199.699880874265 mfu:19.810700880771122 +Step 11641 | loss:1.1112309694290161 lr:7.690473977741526e-05 tokens_per_second_per_gpu:3592.7207213185984 mfu:16.94747662382352 +Step 11642 | loss:1.0244150161743164 lr:7.690108474289045e-05 tokens_per_second_per_gpu:5335.515153935507 mfu:25.16853531386959 +Step 11643 | loss:1.1506223678588867 lr:7.689742950603913e-05 tokens_per_second_per_gpu:4227.182527992917 mfu:19.94034121625292 +Step 11644 | loss:1.120821475982666 lr:7.689377406688882e-05 tokens_per_second_per_gpu:3930.640243249327 mfu:18.541500663786408 +Step 11645 | loss:1.0965651273727417 lr:7.689011842546703e-05 tokens_per_second_per_gpu:5090.562335487991 mfu:24.013051075992244 +Step 11646 | loss:0.9785689115524292 lr:7.68864625818012e-05 tokens_per_second_per_gpu:5369.907395301175 mfu:25.3307693843129 +Step 11647 | loss:1.0852835178375244 lr:7.688280653591885e-05 tokens_per_second_per_gpu:4035.8501952796773 mfu:19.037793958182398 +Step 11648 | loss:1.1766048669815063 lr:7.68791502878475e-05 tokens_per_second_per_gpu:4428.972363015654 mfu:20.892218296951224 +Step 11649 | loss:1.1725502014160156 lr:7.687549383761463e-05 tokens_per_second_per_gpu:5183.308582558065 mfu:24.450550947563265 +Step 11650 | loss:1.0646164417266846 lr:7.687183718524772e-05 tokens_per_second_per_gpu:5319.2715958356675 mfu:25.09191167887585 +Step 11651 | loss:1.1253559589385986 lr:7.686818033077432e-05 tokens_per_second_per_gpu:3384.7323524946905 mfu:15.966360001606178 +Step 11652 | loss:0.8297405242919922 lr:7.686452327422189e-05 tokens_per_second_per_gpu:4401.330040660802 mfu:20.76182474613687 +Step 11653 | loss:1.0976274013519287 lr:7.686086601561796e-05 tokens_per_second_per_gpu:4867.557707373311 mfu:22.9611001966623 +Step 11654 | loss:1.1159998178482056 lr:7.685720855499001e-05 tokens_per_second_per_gpu:5096.183382664108 mfu:24.039566514571526 +Step 11655 | loss:1.0956041812896729 lr:7.685355089236558e-05 tokens_per_second_per_gpu:5312.962940657309 mfu:25.06215267603197 +Step 11656 | loss:1.1350983381271362 lr:7.68498930277722e-05 tokens_per_second_per_gpu:3965.6325758831945 mfu:18.706565467127945 +Step 11657 | loss:1.1019619703292847 lr:7.68462349612373e-05 tokens_per_second_per_gpu:4871.083280897095 mfu:22.977730928498865 +Step 11658 | loss:1.167125940322876 lr:7.684257669278847e-05 tokens_per_second_per_gpu:5337.878737539744 mfu:25.179684740999722 +Step 11659 | loss:1.1040223836898804 lr:7.683891822245319e-05 tokens_per_second_per_gpu:5233.355665040545 mfu:24.686631574545903 +Step 11660 | loss:1.1593745946884155 lr:7.683525955025899e-05 tokens_per_second_per_gpu:4432.968998931883 mfu:20.911071110464476 +Step 11661 | loss:1.0430678129196167 lr:7.683160067623335e-05 tokens_per_second_per_gpu:5207.245009421568 mfu:24.563463157053963 +Step 11662 | loss:1.183314561843872 lr:7.682794160040382e-05 tokens_per_second_per_gpu:5328.668425988473 mfu:25.136238129971183 +Step 11663 | loss:1.1735751628875732 lr:7.682428232279795e-05 tokens_per_second_per_gpu:5098.885337345523 mfu:24.05231209580477 +Step 11664 | loss:1.2234903573989868 lr:7.682062284344323e-05 tokens_per_second_per_gpu:4296.856332412405 mfu:20.26900443927649 +Step 11665 | loss:1.077967643737793 lr:7.681696316236715e-05 tokens_per_second_per_gpu:5062.804306027357 mfu:23.882111715018244 +Step 11666 | loss:1.1335768699645996 lr:7.681330327959728e-05 tokens_per_second_per_gpu:5298.4427290100975 mfu:24.993658360288507 +Step 11667 | loss:1.1860252618789673 lr:7.680964319516114e-05 tokens_per_second_per_gpu:5338.888004080744 mfu:25.184445623472246 +Step 11668 | loss:1.1071768999099731 lr:7.680598290908626e-05 tokens_per_second_per_gpu:4694.212942523725 mfu:22.14340336520018 +Step 11669 | loss:1.1655789613723755 lr:7.680232242140015e-05 tokens_per_second_per_gpu:4599.191771898575 mfu:21.695172291077878 +Step 11670 | loss:1.0523302555084229 lr:7.679866173213035e-05 tokens_per_second_per_gpu:4987.016133903192 mfu:23.524605976313012 +Step 11671 | loss:1.092254400253296 lr:7.679500084130442e-05 tokens_per_second_per_gpu:5362.276425653154 mfu:25.29477279850551 +Step 11672 | loss:1.1759960651397705 lr:7.679133974894983e-05 tokens_per_second_per_gpu:4970.945583141491 mfu:23.44879844645137 +Step 11673 | loss:1.154768943786621 lr:7.678767845509418e-05 tokens_per_second_per_gpu:5379.906758564496 mfu:25.37793808689328 +Step 11674 | loss:1.1611478328704834 lr:7.678401695976499e-05 tokens_per_second_per_gpu:4700.828502920673 mfu:22.174610092323572 +Step 11675 | loss:1.0739942789077759 lr:7.678035526298977e-05 tokens_per_second_per_gpu:3956.5250686495606 mfu:18.663603801605927 +Step 11676 | loss:1.042392373085022 lr:7.677669336479609e-05 tokens_per_second_per_gpu:4543.585347993947 mfu:21.432867302085516 +Step 11677 | loss:1.0956724882125854 lr:7.67730312652115e-05 tokens_per_second_per_gpu:5321.106436539035 mfu:25.100566935529738 +Step 11678 | loss:0.9254493713378906 lr:7.67693689642635e-05 tokens_per_second_per_gpu:4361.048047501308 mfu:20.5718077115868 +Step 11679 | loss:1.0634187459945679 lr:7.676570646197966e-05 tokens_per_second_per_gpu:5322.382207374893 mfu:25.106584964231434 +Step 11680 | loss:1.0306373834609985 lr:7.676204375838753e-05 tokens_per_second_per_gpu:4560.699016834155 mfu:21.51359539789777 +Step 11681 | loss:1.0663338899612427 lr:7.675838085351467e-05 tokens_per_second_per_gpu:5372.491483201051 mfu:25.342958967827474 +Step 11682 | loss:0.9113630056381226 lr:7.67547177473886e-05 tokens_per_second_per_gpu:4142.117125231591 mfu:19.539073197773437 +Step 11683 | loss:1.0075006484985352 lr:7.675105444003687e-05 tokens_per_second_per_gpu:4265.517469337 mfu:20.121173675188402 +Step 11684 | loss:1.0594263076782227 lr:7.674739093148707e-05 tokens_per_second_per_gpu:5378.031596722208 mfu:25.36909262854761 +Step 11685 | loss:1.0105810165405273 lr:7.674372722176673e-05 tokens_per_second_per_gpu:5349.059847837963 mfu:25.232427945970553 +Step 11686 | loss:1.2656292915344238 lr:7.67400633109034e-05 tokens_per_second_per_gpu:4498.623973915625 mfu:21.220776829356627 +Step 11687 | loss:1.216307520866394 lr:7.673639919892464e-05 tokens_per_second_per_gpu:4506.818283660663 mfu:21.259430786517374 +Step 11688 | loss:1.1679390668869019 lr:7.673273488585801e-05 tokens_per_second_per_gpu:3815.5386696993596 mfu:17.99854690299772 +Step 11689 | loss:0.9637775421142578 lr:7.672907037173107e-05 tokens_per_second_per_gpu:5383.377319965018 mfu:25.394309317158378 +Step 11690 | loss:0.9349147081375122 lr:7.672540565657137e-05 tokens_per_second_per_gpu:3325.9583989817716 mfu:15.689113235015228 +Step 11691 | loss:1.1490190029144287 lr:7.67217407404065e-05 tokens_per_second_per_gpu:4656.84733238614 mfu:21.967143406950704 +Step 11692 | loss:1.2021617889404297 lr:7.6718075623264e-05 tokens_per_second_per_gpu:5350.712470190304 mfu:25.24022364757299 +Step 11693 | loss:1.157157063484192 lr:7.671441030517144e-05 tokens_per_second_per_gpu:5344.183213674938 mfu:25.209424030583975 +Step 11694 | loss:0.972660481929779 lr:7.67107447861564e-05 tokens_per_second_per_gpu:4590.267123863257 mfu:21.653073203593316 +Step 11695 | loss:1.0816305875778198 lr:7.670707906624644e-05 tokens_per_second_per_gpu:4516.573501980505 mfu:21.305447815743946 +Step 11696 | loss:1.1195755004882812 lr:7.670341314546914e-05 tokens_per_second_per_gpu:4977.794098348997 mfu:23.481104061162362 +Step 11697 | loss:1.0750083923339844 lr:7.669974702385203e-05 tokens_per_second_per_gpu:4931.038999994697 mfu:23.26055228498276 +Step 11698 | loss:1.029787302017212 lr:7.669608070142274e-05 tokens_per_second_per_gpu:5332.101079503303 mfu:25.152430542271716 +Step 11699 | loss:1.2834137678146362 lr:7.669241417820882e-05 tokens_per_second_per_gpu:3812.654098018713 mfu:17.98493988622143 +Step 11700 | loss:1.0916643142700195 lr:7.668874745423784e-05 tokens_per_second_per_gpu:4300.65174047646 mfu:20.286908026678027 +Step 11701 | loss:1.0933834314346313 lr:7.668508052953739e-05 tokens_per_second_per_gpu:3833.7371802101093 mfu:18.084392382062994 +Step 11702 | loss:1.0111852884292603 lr:7.668141340413504e-05 tokens_per_second_per_gpu:3065.388673499968 mfu:14.45996138213805 +Step 11703 | loss:1.1239720582962036 lr:7.667774607805838e-05 tokens_per_second_per_gpu:5152.869330916791 mfu:24.306963804099514 +Step 11704 | loss:0.9381946921348572 lr:7.6674078551335e-05 tokens_per_second_per_gpu:5328.433301234173 mfu:25.135129006388727 +Step 11705 | loss:1.2105966806411743 lr:7.667041082399244e-05 tokens_per_second_per_gpu:5367.954934454166 mfu:25.321559293373387 +Step 11706 | loss:0.95320063829422 lr:7.666674289605835e-05 tokens_per_second_per_gpu:4731.178230845963 mfu:22.317774937144712 +Step 11707 | loss:0.9783822894096375 lr:7.666307476756027e-05 tokens_per_second_per_gpu:4731.298412120751 mfu:22.318341852722924 +Step 11708 | loss:1.1097612380981445 lr:7.665940643852578e-05 tokens_per_second_per_gpu:4901.485980473423 mfu:23.121145649635842 +Step 11709 | loss:1.0513501167297363 lr:7.665573790898252e-05 tokens_per_second_per_gpu:5289.9152691366835 mfu:24.95343287713093 +Step 11710 | loss:1.0328502655029297 lr:7.665206917895805e-05 tokens_per_second_per_gpu:5369.589832363569 mfu:25.32927138575439 +Step 11711 | loss:0.9784418344497681 lr:7.664840024847996e-05 tokens_per_second_per_gpu:4386.844196507887 mfu:20.693492547727697 +Step 11712 | loss:1.1303175687789917 lr:7.664473111757586e-05 tokens_per_second_per_gpu:3871.4324549854887 mfu:18.262207424655532 +Step 11713 | loss:1.2072384357452393 lr:7.664106178627332e-05 tokens_per_second_per_gpu:4001.8906375658107 mfu:18.877600930348077 +Step 11714 | loss:1.0203778743743896 lr:7.663739225459998e-05 tokens_per_second_per_gpu:5008.007833189487 mfu:23.623627403399684 +Step 11715 | loss:1.017755150794983 lr:7.663372252258341e-05 tokens_per_second_per_gpu:5289.9154261391395 mfu:24.953433617738302 +Step 11716 | loss:1.2647188901901245 lr:7.66300525902512e-05 tokens_per_second_per_gpu:5046.935780624614 mfu:23.80725717324389 +Step 11717 | loss:1.146263599395752 lr:7.662638245763098e-05 tokens_per_second_per_gpu:4583.117116241584 mfu:21.619345397724786 +Step 11718 | loss:1.330319881439209 lr:7.662271212475034e-05 tokens_per_second_per_gpu:3711.9048975370447 mfu:17.509688717963265 +Step 11719 | loss:1.1344881057739258 lr:7.661904159163687e-05 tokens_per_second_per_gpu:5011.549766812073 mfu:23.640335308613615 +Step 11720 | loss:1.3342467546463013 lr:7.661537085831819e-05 tokens_per_second_per_gpu:5126.37062220193 mfu:24.181964873945603 +Step 11721 | loss:1.0309646129608154 lr:7.66116999248219e-05 tokens_per_second_per_gpu:5343.238868255271 mfu:25.204969392117466 +Step 11722 | loss:1.0958373546600342 lr:7.660802879117566e-05 tokens_per_second_per_gpu:4477.694698873765 mfu:21.12204986808163 +Step 11723 | loss:1.0262812376022339 lr:7.660435745740702e-05 tokens_per_second_per_gpu:4012.2609542608902 mfu:18.926519483557247 +Step 11724 | loss:1.0581729412078857 lr:7.66006859235436e-05 tokens_per_second_per_gpu:4572.780105823967 mfu:21.570583955909044 +Step 11725 | loss:1.075151801109314 lr:7.659701418961305e-05 tokens_per_second_per_gpu:2975.7020848099687 mfu:14.036894441177392 +Step 11726 | loss:0.9527822136878967 lr:7.659334225564294e-05 tokens_per_second_per_gpu:5353.611494461624 mfu:25.253898839685352 +Step 11727 | loss:1.105695128440857 lr:7.658967012166092e-05 tokens_per_second_per_gpu:4975.38887481856 mfu:23.46975821139584 +Step 11728 | loss:1.0752614736557007 lr:7.65859977876946e-05 tokens_per_second_per_gpu:4925.694459490433 mfu:23.235341175551508 +Step 11729 | loss:1.0168089866638184 lr:7.658232525377161e-05 tokens_per_second_per_gpu:4796.206233788342 mfu:22.624523121945288 +Step 11730 | loss:1.1804885864257812 lr:7.657865251991954e-05 tokens_per_second_per_gpu:4058.7685897055303 mfu:19.145903934970452 +Step 11731 | loss:1.1217308044433594 lr:7.657497958616602e-05 tokens_per_second_per_gpu:5357.922925249374 mfu:25.274236594317117 +Step 11732 | loss:1.1041287183761597 lr:7.657130645253873e-05 tokens_per_second_per_gpu:4541.964939095743 mfu:21.42522355684242 +Step 11733 | loss:1.193516492843628 lr:7.656763311906522e-05 tokens_per_second_per_gpu:5112.075229222246 mfu:24.114531066214884 +Step 11734 | loss:1.1520196199417114 lr:7.656395958577317e-05 tokens_per_second_per_gpu:5009.295421259699 mfu:23.62970117601205 +Step 11735 | loss:1.0889832973480225 lr:7.656028585269018e-05 tokens_per_second_per_gpu:4809.760418281004 mfu:22.68846052277951 +Step 11736 | loss:1.035573124885559 lr:7.655661191984389e-05 tokens_per_second_per_gpu:5349.43999244888 mfu:25.234221152959602 +Step 11737 | loss:1.056836724281311 lr:7.655293778726193e-05 tokens_per_second_per_gpu:4744.081133905846 mfu:22.37864013234045 +Step 11738 | loss:1.1038765907287598 lr:7.654926345497194e-05 tokens_per_second_per_gpu:5335.6598543036625 mfu:25.169217890194606 +Step 11739 | loss:1.1182693243026733 lr:7.654558892300157e-05 tokens_per_second_per_gpu:4865.015660597726 mfu:22.949108928303968 +Step 11740 | loss:1.0410702228546143 lr:7.654191419137841e-05 tokens_per_second_per_gpu:4481.317053391916 mfu:21.13913713238116 +Step 11741 | loss:1.0537450313568115 lr:7.653823926013015e-05 tokens_per_second_per_gpu:3652.048931865257 mfu:17.227337915408622 +Step 11742 | loss:1.010609745979309 lr:7.653456412928438e-05 tokens_per_second_per_gpu:5242.158772869783 mfu:24.728157336141493 +Step 11743 | loss:1.0533740520477295 lr:7.65308887988688e-05 tokens_per_second_per_gpu:5232.8509714019865 mfu:24.684250848543208 +Step 11744 | loss:1.0164495706558228 lr:7.652721326891099e-05 tokens_per_second_per_gpu:5347.869295251843 mfu:25.226811906292607 +Step 11745 | loss:1.176221251487732 lr:7.652353753943863e-05 tokens_per_second_per_gpu:3822.1736823527626 mfu:18.029845389733712 +Step 11746 | loss:1.175275206565857 lr:7.651986161047936e-05 tokens_per_second_per_gpu:4809.511163668419 mfu:22.68728474624473 +Step 11747 | loss:1.0996947288513184 lr:7.651618548206086e-05 tokens_per_second_per_gpu:4081.7837961460973 mfu:19.254470590549744 +Step 11748 | loss:1.1059352159500122 lr:7.651250915421071e-05 tokens_per_second_per_gpu:4786.961810659651 mfu:22.58091560078602 +Step 11749 | loss:1.073912262916565 lr:7.65088326269566e-05 tokens_per_second_per_gpu:4606.44589193079 mfu:21.729391212950336 +Step 11750 | loss:1.1168670654296875 lr:7.650515590032619e-05 tokens_per_second_per_gpu:5248.263006557098 mfu:24.756952047934217 +Step 11751 | loss:1.1036525964736938 lr:7.65014789743471e-05 tokens_per_second_per_gpu:4101.112052091423 mfu:19.34564527158436 +Step 11752 | loss:1.0716161727905273 lr:7.649780184904702e-05 tokens_per_second_per_gpu:4264.799567093551 mfu:20.1177872078193 +Step 11753 | loss:1.108416199684143 lr:7.649412452445359e-05 tokens_per_second_per_gpu:5106.250899425819 mfu:24.087056708832 +Step 11754 | loss:1.0954279899597168 lr:7.649044700059448e-05 tokens_per_second_per_gpu:4996.329788372874 mfu:23.568540073519813 +Step 11755 | loss:1.1321333646774292 lr:7.648676927749732e-05 tokens_per_second_per_gpu:5295.324562477427 mfu:24.978949436739825 +Step 11756 | loss:1.2982474565505981 lr:7.648309135518979e-05 tokens_per_second_per_gpu:4835.906241414653 mfu:22.81179483143836 +Step 11757 | loss:1.0389299392700195 lr:7.647941323369956e-05 tokens_per_second_per_gpu:5353.647861668597 mfu:25.254070390005534 +Step 11758 | loss:1.222418189048767 lr:7.647573491305427e-05 tokens_per_second_per_gpu:4612.474510260837 mfu:21.75782923420152 +Step 11759 | loss:1.2083536386489868 lr:7.647205639328161e-05 tokens_per_second_per_gpu:4341.180501939681 mfu:20.47808910941979 +Step 11760 | loss:1.0072003602981567 lr:7.646837767440924e-05 tokens_per_second_per_gpu:5345.174503273404 mfu:25.21410011275139 +Step 11761 | loss:1.116141438484192 lr:7.646469875646483e-05 tokens_per_second_per_gpu:3442.2823680316446 mfu:16.23783324393915 +Step 11762 | loss:1.1270641088485718 lr:7.646101963947601e-05 tokens_per_second_per_gpu:5334.391244775405 mfu:25.16323364260271 +Step 11763 | loss:1.054531216621399 lr:7.645734032347051e-05 tokens_per_second_per_gpu:5203.817120553017 mfu:24.547293220402857 +Step 11764 | loss:1.091237187385559 lr:7.645366080847598e-05 tokens_per_second_per_gpu:5363.047149420886 mfu:25.298408434016935 +Step 11765 | loss:0.9716815948486328 lr:7.644998109452009e-05 tokens_per_second_per_gpu:4675.635309708612 mfu:22.055769501540283 +Step 11766 | loss:1.013342261314392 lr:7.644630118163052e-05 tokens_per_second_per_gpu:5364.833751264771 mfu:25.306836139740252 +Step 11767 | loss:1.1421148777008057 lr:7.644262106983494e-05 tokens_per_second_per_gpu:4519.133538282686 mfu:21.317523943768567 +Step 11768 | loss:1.067449927330017 lr:7.643894075916103e-05 tokens_per_second_per_gpu:5366.053565420236 mfu:25.31259021123333 +Step 11769 | loss:1.1329011917114258 lr:7.643526024963645e-05 tokens_per_second_per_gpu:3843.026737191976 mfu:18.128212807308316 +Step 11770 | loss:1.1643930673599243 lr:7.643157954128894e-05 tokens_per_second_per_gpu:3944.7645237845354 mfu:18.60812730492174 +Step 11771 | loss:1.0585379600524902 lr:7.642789863414612e-05 tokens_per_second_per_gpu:5324.500282278763 mfu:25.11657628493383 +Step 11772 | loss:1.0674116611480713 lr:7.642421752823574e-05 tokens_per_second_per_gpu:4654.303423261607 mfu:21.955143353574634 +Step 11773 | loss:0.968007504940033 lr:7.642053622358541e-05 tokens_per_second_per_gpu:5309.719739508351 mfu:25.046853942865162 +Step 11774 | loss:1.046323537826538 lr:7.641685472022287e-05 tokens_per_second_per_gpu:4438.3880669539785 mfu:20.93663377891305 +Step 11775 | loss:1.0611083507537842 lr:7.641317301817581e-05 tokens_per_second_per_gpu:5347.760277692417 mfu:25.226297651864478 +Step 11776 | loss:1.0315868854522705 lr:7.640949111747189e-05 tokens_per_second_per_gpu:3984.0760826595438 mfu:18.793566635378728 +Step 11777 | loss:1.098422646522522 lr:7.640580901813881e-05 tokens_per_second_per_gpu:3418.251111775387 mfu:16.124473708023267 +Step 11778 | loss:1.0253593921661377 lr:7.640212672020428e-05 tokens_per_second_per_gpu:5368.230936572524 mfu:25.3228612424565 +Step 11779 | loss:1.0274925231933594 lr:7.639844422369599e-05 tokens_per_second_per_gpu:4519.130339068604 mfu:21.3175088525299 +Step 11780 | loss:1.1251275539398193 lr:7.639476152864162e-05 tokens_per_second_per_gpu:4032.109048698732 mfu:19.020146331454654 +Step 11781 | loss:1.1872695684432983 lr:7.63910786350689e-05 tokens_per_second_per_gpu:4965.334738929149 mfu:23.42233113699302 +Step 11782 | loss:0.9438248872756958 lr:7.638739554300551e-05 tokens_per_second_per_gpu:3847.990150956217 mfu:18.151626076879953 +Step 11783 | loss:1.3027575016021729 lr:7.638371225247914e-05 tokens_per_second_per_gpu:3289.5321542866045 mfu:15.517284423829913 +Step 11784 | loss:1.023451805114746 lr:7.63800287635175e-05 tokens_per_second_per_gpu:4935.2117463002205 mfu:23.280235841249848 +Step 11785 | loss:1.1645808219909668 lr:7.63763450761483e-05 tokens_per_second_per_gpu:4887.889382650772 mfu:23.057008177887596 +Step 11786 | loss:0.9620839357376099 lr:7.637266119039926e-05 tokens_per_second_per_gpu:3152.9724606119253 mfu:14.873109049280144 +Step 11787 | loss:1.0073684453964233 lr:7.636897710629804e-05 tokens_per_second_per_gpu:4396.468527593174 mfu:20.738892159537205 +Step 11788 | loss:0.925926923751831 lr:7.636529282387239e-05 tokens_per_second_per_gpu:4537.736983959975 mfu:21.405279571103474 +Step 11789 | loss:0.9499145150184631 lr:7.636160834315001e-05 tokens_per_second_per_gpu:5349.0081360283775 mfu:25.23218401254179 +Step 11790 | loss:1.238669991493225 lr:7.635792366415861e-05 tokens_per_second_per_gpu:4020.3040677056742 mfu:18.96446022197295 +Step 11791 | loss:1.1390643119812012 lr:7.63542387869259e-05 tokens_per_second_per_gpu:4908.758268842933 mfu:23.15545027465516 +Step 11792 | loss:1.1699961423873901 lr:7.63505537114796e-05 tokens_per_second_per_gpu:3564.6375730218356 mfu:16.81500362182851 +Step 11793 | loss:0.9948276281356812 lr:7.634686843784744e-05 tokens_per_second_per_gpu:5320.482304215069 mfu:25.097622796869636 +Step 11794 | loss:1.0582971572875977 lr:7.63431829660571e-05 tokens_per_second_per_gpu:3870.8520049120834 mfu:18.259469342624282 +Step 11795 | loss:1.066213846206665 lr:7.633949729613632e-05 tokens_per_second_per_gpu:4701.190662130294 mfu:22.176318459106092 +Step 11796 | loss:1.0421972274780273 lr:7.633581142811282e-05 tokens_per_second_per_gpu:5328.044088514645 mfu:25.133293023584184 +Step 11797 | loss:1.1073778867721558 lr:7.633212536201432e-05 tokens_per_second_per_gpu:4636.1037360704195 mfu:21.86929232390697 +Step 11798 | loss:1.0756733417510986 lr:7.632843909786856e-05 tokens_per_second_per_gpu:4789.193773468331 mfu:22.59144415016669 +Step 11799 | loss:1.1263347864151 lr:7.632475263570324e-05 tokens_per_second_per_gpu:5332.2041244805905 mfu:25.152916623010867 +Step 11800 | loss:0.9321837425231934 lr:7.63210659755461e-05 tokens_per_second_per_gpu:4506.171876350963 mfu:21.256381572949063 +Step 11801 | loss:1.0644580125808716 lr:7.631737911742487e-05 tokens_per_second_per_gpu:5347.623776287728 mfu:25.22565375144902 +Step 11802 | loss:1.1548526287078857 lr:7.631369206136726e-05 tokens_per_second_per_gpu:5337.215539865258 mfu:25.176556324415976 +Step 11803 | loss:1.1181859970092773 lr:7.631000480740101e-05 tokens_per_second_per_gpu:4212.772744089337 mfu:19.872367807017085 +Step 11804 | loss:1.121166467666626 lr:7.630631735555388e-05 tokens_per_second_per_gpu:5333.990644515079 mfu:25.1613439428253 +Step 11805 | loss:1.1113568544387817 lr:7.630262970585356e-05 tokens_per_second_per_gpu:3741.869914593908 mfu:17.65103881597959 +Step 11806 | loss:1.0493437051773071 lr:7.629894185832783e-05 tokens_per_second_per_gpu:4469.292160430517 mfu:21.082413660623935 +Step 11807 | loss:1.0710480213165283 lr:7.629525381300436e-05 tokens_per_second_per_gpu:4083.706523125045 mfu:19.26354042176049 +Step 11808 | loss:1.0810023546218872 lr:7.629156556991098e-05 tokens_per_second_per_gpu:3453.9921795703594 mfu:16.29307042286713 +Step 11809 | loss:1.1016064882278442 lr:7.628787712907534e-05 tokens_per_second_per_gpu:5322.542467131813 mfu:25.107340936847958 +Step 11810 | loss:1.2151482105255127 lr:7.628418849052523e-05 tokens_per_second_per_gpu:5131.111518967968 mfu:24.204328492871397 +Step 11811 | loss:1.0479774475097656 lr:7.628049965428841e-05 tokens_per_second_per_gpu:4292.59524483615 mfu:20.24890416216279 +Step 11812 | loss:1.0563231706619263 lr:7.627681062039257e-05 tokens_per_second_per_gpu:5392.9638994308 mfu:25.43953084813759 +Step 11813 | loss:1.0819950103759766 lr:7.62731213888655e-05 tokens_per_second_per_gpu:4160.650751513399 mfu:19.62649947510744 +Step 11814 | loss:1.0670278072357178 lr:7.626943195973493e-05 tokens_per_second_per_gpu:5141.390981934933 mfu:24.252818473543805 +Step 11815 | loss:1.1494100093841553 lr:7.626574233302861e-05 tokens_per_second_per_gpu:4150.488923853058 mfu:19.57856440024703 +Step 11816 | loss:1.1797810792922974 lr:7.626205250877427e-05 tokens_per_second_per_gpu:4067.1429202166423 mfu:19.185407080799486 +Step 11817 | loss:1.019668698310852 lr:7.625836248699971e-05 tokens_per_second_per_gpu:3896.2378063100305 mfu:18.379218499082743 +Step 11818 | loss:1.0148080587387085 lr:7.625467226773264e-05 tokens_per_second_per_gpu:4467.161502426913 mfu:21.072362983293196 +Step 11819 | loss:1.012125849723816 lr:7.625098185100085e-05 tokens_per_second_per_gpu:4567.478823536616 mfu:21.545576902867584 +Step 11820 | loss:1.0694845914840698 lr:7.624729123683206e-05 tokens_per_second_per_gpu:4397.181613905927 mfu:20.742255909339534 +Step 11821 | loss:1.2601618766784668 lr:7.624360042525404e-05 tokens_per_second_per_gpu:3705.416983965719 mfu:17.47908412269512 +Step 11822 | loss:1.039302945137024 lr:7.623990941629456e-05 tokens_per_second_per_gpu:3869.499333382164 mfu:18.25308856022809 +Step 11823 | loss:1.0635262727737427 lr:7.623621820998137e-05 tokens_per_second_per_gpu:5341.985686620174 mfu:25.19905792052988 +Step 11824 | loss:0.9523932337760925 lr:7.623252680634223e-05 tokens_per_second_per_gpu:4645.961518758207 mfu:21.91579316675642 +Step 11825 | loss:1.0926119089126587 lr:7.622883520540491e-05 tokens_per_second_per_gpu:5383.7227962199595 mfu:25.395938987597265 +Step 11826 | loss:1.079305648803711 lr:7.622514340719719e-05 tokens_per_second_per_gpu:5367.43282807754 mfu:25.31909642851431 +Step 11827 | loss:1.04988694190979 lr:7.62214514117468e-05 tokens_per_second_per_gpu:3941.811116310095 mfu:18.59419557796153 +Step 11828 | loss:1.0393961668014526 lr:7.621775921908153e-05 tokens_per_second_per_gpu:4371.990320120254 mfu:20.623424278474587 +Step 11829 | loss:1.0054059028625488 lr:7.621406682922916e-05 tokens_per_second_per_gpu:5044.307388384578 mfu:23.79485859859768 +Step 11830 | loss:1.0076895952224731 lr:7.621037424221745e-05 tokens_per_second_per_gpu:3523.1640028666716 mfu:16.619365715285866 +Step 11831 | loss:1.108815312385559 lr:7.620668145807414e-05 tokens_per_second_per_gpu:3190.1344090890966 mfu:15.04840830072918 +Step 11832 | loss:1.138178825378418 lr:7.620298847682706e-05 tokens_per_second_per_gpu:4482.785025223941 mfu:21.14606180597464 +Step 11833 | loss:1.079572319984436 lr:7.619929529850397e-05 tokens_per_second_per_gpu:5072.584864505104 mfu:23.92824827809313 +Step 11834 | loss:1.1158262491226196 lr:7.619560192313262e-05 tokens_per_second_per_gpu:5065.543414680018 mfu:23.895032557872973 +Step 11835 | loss:1.0658962726593018 lr:7.61919083507408e-05 tokens_per_second_per_gpu:5354.556044389783 mfu:25.258354442853314 +Step 11836 | loss:1.1883907318115234 lr:7.618821458135631e-05 tokens_per_second_per_gpu:4158.330986249672 mfu:19.61555674656567 +Step 11837 | loss:1.1536641120910645 lr:7.618452061500692e-05 tokens_per_second_per_gpu:5037.533641585827 mfu:23.762905679227597 +Step 11838 | loss:0.9995328783988953 lr:7.61808264517204e-05 tokens_per_second_per_gpu:3649.783913847934 mfu:17.216653438968986 +Step 11839 | loss:1.0299299955368042 lr:7.617713209152453e-05 tokens_per_second_per_gpu:5374.9402468269345 mfu:25.354510203652925 +Step 11840 | loss:1.1423863172531128 lr:7.617343753444714e-05 tokens_per_second_per_gpu:5056.492697651869 mfu:23.852338781439634 +Step 11841 | loss:1.1225754022598267 lr:7.616974278051596e-05 tokens_per_second_per_gpu:3586.8672396791508 mfu:16.919864752223297 +Step 11842 | loss:1.1401517391204834 lr:7.616604782975882e-05 tokens_per_second_per_gpu:4722.401234514812 mfu:22.27637234794068 +Step 11843 | loss:1.0067578554153442 lr:7.616235268220349e-05 tokens_per_second_per_gpu:4528.179553030551 mfu:21.36019553874351 +Step 11844 | loss:0.9600426554679871 lr:7.615865733787777e-05 tokens_per_second_per_gpu:4984.318578740553 mfu:23.511881148360473 +Step 11845 | loss:0.9084405303001404 lr:7.615496179680945e-05 tokens_per_second_per_gpu:5348.644482000121 mfu:25.230468594444726 +Step 11846 | loss:1.0457806587219238 lr:7.615126605902632e-05 tokens_per_second_per_gpu:4872.569105891909 mfu:22.984739818507283 +Step 11847 | loss:1.1907471418380737 lr:7.614757012455618e-05 tokens_per_second_per_gpu:4592.988082083637 mfu:21.665908427762187 +Step 11848 | loss:1.0554475784301758 lr:7.614387399342683e-05 tokens_per_second_per_gpu:5358.651331452729 mfu:25.27767261065726 +Step 11849 | loss:0.8340036869049072 lr:7.614017766566607e-05 tokens_per_second_per_gpu:5359.412297893203 mfu:25.28126221919123 +Step 11850 | loss:1.1321141719818115 lr:7.613648114130169e-05 tokens_per_second_per_gpu:5346.580512831273 mfu:25.220732499725823 +Step 11851 | loss:0.9835475087165833 lr:7.613278442036152e-05 tokens_per_second_per_gpu:5117.552399593497 mfu:24.140367813357702 +Step 11852 | loss:1.1630743741989136 lr:7.612908750287333e-05 tokens_per_second_per_gpu:3667.0844842877214 mfu:17.298263181515157 +Step 11853 | loss:0.9965770244598389 lr:7.612539038886493e-05 tokens_per_second_per_gpu:4597.095902476434 mfu:21.685285717421444 +Step 11854 | loss:1.0124742984771729 lr:7.612169307836415e-05 tokens_per_second_per_gpu:4137.907945201354 mfu:19.519217777410642 +Step 11855 | loss:1.0900123119354248 lr:7.611799557139878e-05 tokens_per_second_per_gpu:3563.1663040599733 mfu:16.808063395111034 +Step 11856 | loss:1.0228773355484009 lr:7.611429786799664e-05 tokens_per_second_per_gpu:4207.3398350665075 mfu:19.846739848206532 +Step 11857 | loss:1.1141698360443115 lr:7.611059996818553e-05 tokens_per_second_per_gpu:3979.7305834128792 mfu:18.773068174013584 +Step 11858 | loss:1.0613305568695068 lr:7.610690187199326e-05 tokens_per_second_per_gpu:4853.63316283481 mfu:22.895415744301914 +Step 11859 | loss:1.1108776330947876 lr:7.610320357944765e-05 tokens_per_second_per_gpu:4320.900697757263 mfu:20.382425812999934 +Step 11860 | loss:1.1209396123886108 lr:7.60995050905765e-05 tokens_per_second_per_gpu:3804.3553176795813 mfu:17.945793123443817 +Step 11861 | loss:1.0443373918533325 lr:7.609580640540768e-05 tokens_per_second_per_gpu:4009.503038385771 mfu:18.913509923825337 +Step 11862 | loss:1.0094953775405884 lr:7.609210752396895e-05 tokens_per_second_per_gpu:4838.514356669088 mfu:22.824097756083873 +Step 11863 | loss:0.9318138360977173 lr:7.608840844628815e-05 tokens_per_second_per_gpu:4562.165865190996 mfu:21.520514771867127 +Step 11864 | loss:1.0815062522888184 lr:7.608470917239309e-05 tokens_per_second_per_gpu:5316.664573136079 mfu:25.07961390799795 +Step 11865 | loss:1.1640104055404663 lr:7.608100970231162e-05 tokens_per_second_per_gpu:3434.3139681396037 mfu:16.200244942099175 +Step 11866 | loss:1.047179937362671 lr:7.607731003607153e-05 tokens_per_second_per_gpu:4820.097264128544 mfu:22.737221188290125 +Step 11867 | loss:1.0768648386001587 lr:7.607361017370068e-05 tokens_per_second_per_gpu:5307.236622073588 mfu:25.03514065426186 +Step 11868 | loss:1.186128854751587 lr:7.606991011522687e-05 tokens_per_second_per_gpu:4573.612233722977 mfu:21.574509245184334 +Step 11869 | loss:1.0774266719818115 lr:7.606620986067796e-05 tokens_per_second_per_gpu:4041.012638531071 mfu:19.06214608380287 +Step 11870 | loss:0.9967185854911804 lr:7.606250941008172e-05 tokens_per_second_per_gpu:3890.498959614603 mfu:18.3521473800722 +Step 11871 | loss:1.222621202468872 lr:7.605880876346607e-05 tokens_per_second_per_gpu:4733.463082521337 mfu:22.32855297233334 +Step 11872 | loss:1.159255862236023 lr:7.605510792085878e-05 tokens_per_second_per_gpu:4345.269619737567 mfu:20.49737817574786 +Step 11873 | loss:1.0189980268478394 lr:7.605140688228768e-05 tokens_per_second_per_gpu:4947.233404111839 mfu:23.336944052253582 +Step 11874 | loss:1.2767701148986816 lr:7.604770564778063e-05 tokens_per_second_per_gpu:3696.5895253134227 mfu:17.437443494112486 +Step 11875 | loss:1.1315209865570068 lr:7.604400421736546e-05 tokens_per_second_per_gpu:5288.712499817428 mfu:24.94775920903839 +Step 11876 | loss:0.9633477330207825 lr:7.604030259107002e-05 tokens_per_second_per_gpu:4142.400407000777 mfu:19.540409486211537 +Step 11877 | loss:1.1171000003814697 lr:7.603660076892213e-05 tokens_per_second_per_gpu:3467.298560770138 mfu:16.355838893289224 +Step 11878 | loss:1.21870756149292 lr:7.603289875094966e-05 tokens_per_second_per_gpu:5320.523996961211 mfu:25.097819468666707 +Step 11879 | loss:1.0416605472564697 lr:7.602919653718044e-05 tokens_per_second_per_gpu:4089.7337491359067 mfu:19.29197187520452 +Step 11880 | loss:1.232085943222046 lr:7.60254941276423e-05 tokens_per_second_per_gpu:4933.488096240079 mfu:23.27210508983114 +Step 11881 | loss:1.0510001182556152 lr:7.60217915223631e-05 tokens_per_second_per_gpu:4559.142657044786 mfu:21.506253783228953 +Step 11882 | loss:1.1914585828781128 lr:7.601808872137067e-05 tokens_per_second_per_gpu:5202.357981143315 mfu:24.540410210852432 +Step 11883 | loss:1.1662648916244507 lr:7.601438572469289e-05 tokens_per_second_per_gpu:4902.860340163942 mfu:23.127628738785962 +Step 11884 | loss:1.178769826889038 lr:7.601068253235759e-05 tokens_per_second_per_gpu:5237.908946135923 mfu:24.70811017834266 +Step 11885 | loss:1.1670581102371216 lr:7.600697914439264e-05 tokens_per_second_per_gpu:5345.787691418517 mfu:25.21699262585257 +Step 11886 | loss:0.9929070472717285 lr:7.600327556082588e-05 tokens_per_second_per_gpu:5348.424568572042 mfu:25.22943122528338 +Step 11887 | loss:1.2366952896118164 lr:7.599957178168515e-05 tokens_per_second_per_gpu:4485.285825424636 mfu:21.157858507201826 +Step 11888 | loss:1.0789071321487427 lr:7.599586780699833e-05 tokens_per_second_per_gpu:4024.6001346795133 mfu:18.98472550287316 +Step 11889 | loss:1.084341049194336 lr:7.599216363679328e-05 tokens_per_second_per_gpu:5350.14806722334 mfu:25.23756126248075 +Step 11890 | loss:1.0757027864456177 lr:7.598845927109783e-05 tokens_per_second_per_gpu:5225.71187973535 mfu:24.65057453509704 +Step 11891 | loss:1.135987401008606 lr:7.598475470993988e-05 tokens_per_second_per_gpu:4021.9483109306516 mfu:18.97221639780199 +Step 11892 | loss:0.9647881388664246 lr:7.598104995334729e-05 tokens_per_second_per_gpu:5341.131003101418 mfu:25.195026232547864 +Step 11893 | loss:1.213606357574463 lr:7.597734500134788e-05 tokens_per_second_per_gpu:5332.883061247937 mfu:25.156119283581 +Step 11894 | loss:1.105269193649292 lr:7.597363985396954e-05 tokens_per_second_per_gpu:4648.775474893048 mfu:21.92906707795489 +Step 11895 | loss:1.0453054904937744 lr:7.596993451124016e-05 tokens_per_second_per_gpu:5317.709784003599 mfu:25.08454434599141 +Step 11896 | loss:1.2053937911987305 lr:7.596622897318758e-05 tokens_per_second_per_gpu:3483.6839323152176 mfu:16.433131486500027 +Step 11897 | loss:1.0879663228988647 lr:7.596252323983968e-05 tokens_per_second_per_gpu:5337.229233764226 mfu:25.176620920873827 +Step 11898 | loss:1.0391943454742432 lr:7.595881731122434e-05 tokens_per_second_per_gpu:4978.044293607737 mfu:23.48228427488555 +Step 11899 | loss:1.0848090648651123 lr:7.59551111873694e-05 tokens_per_second_per_gpu:3720.3959014500165 mfu:17.549742232135408 +Step 11900 | loss:0.9648069143295288 lr:7.595140486830276e-05 tokens_per_second_per_gpu:5327.646152333796 mfu:25.1314158907242 +Step 11901 | loss:1.0148496627807617 lr:7.59476983540523e-05 tokens_per_second_per_gpu:5308.898575796279 mfu:25.042980373530337 +Step 11902 | loss:1.1972219944000244 lr:7.59439916446459e-05 tokens_per_second_per_gpu:4389.776029235509 mfu:20.707322502926132 +Step 11903 | loss:1.0970340967178345 lr:7.594028474011142e-05 tokens_per_second_per_gpu:4953.848219517138 mfu:23.368147265124264 +Step 11904 | loss:1.3282274007797241 lr:7.593657764047675e-05 tokens_per_second_per_gpu:3384.9962604453735 mfu:15.967604900436887 +Step 11905 | loss:1.1855089664459229 lr:7.593287034576977e-05 tokens_per_second_per_gpu:4030.6568367604814 mfu:19.01329599996902 +Step 11906 | loss:1.215938687324524 lr:7.592916285601835e-05 tokens_per_second_per_gpu:3925.245964380497 mfu:18.516054930003257 +Step 11907 | loss:0.9468547105789185 lr:7.59254551712504e-05 tokens_per_second_per_gpu:4445.805987927223 mfu:20.97162537777167 +Step 11908 | loss:1.0258970260620117 lr:7.592174729149379e-05 tokens_per_second_per_gpu:3971.0480299818364 mfu:18.732111088082984 +Step 11909 | loss:0.8470361232757568 lr:7.591803921677643e-05 tokens_per_second_per_gpu:4551.43058652653 mfu:21.469874630779426 +Step 11910 | loss:1.0465034246444702 lr:7.591433094712617e-05 tokens_per_second_per_gpu:4888.524553603799 mfu:23.060004387644366 +Step 11911 | loss:1.0213711261749268 lr:7.591062248257091e-05 tokens_per_second_per_gpu:5043.491202669081 mfu:23.791008511322037 +Step 11912 | loss:1.060411810874939 lr:7.590691382313857e-05 tokens_per_second_per_gpu:3303.884833285792 mfu:15.584988459488633 +Step 11913 | loss:1.1489510536193848 lr:7.590320496885703e-05 tokens_per_second_per_gpu:4966.764086684153 mfu:23.429073614225246 +Step 11914 | loss:0.9973045587539673 lr:7.589949591975415e-05 tokens_per_second_per_gpu:5075.108567707593 mfu:23.940153016687432 +Step 11915 | loss:1.0684866905212402 lr:7.58957866758579e-05 tokens_per_second_per_gpu:5360.346790883699 mfu:25.285670382068346 +Step 11916 | loss:1.0740211009979248 lr:7.589207723719612e-05 tokens_per_second_per_gpu:4201.18777260526 mfu:19.817719519926733 +Step 11917 | loss:1.194787859916687 lr:7.58883676037967e-05 tokens_per_second_per_gpu:5092.335219688142 mfu:24.021414073249595 +Step 11918 | loss:0.9504303336143494 lr:7.588465777568758e-05 tokens_per_second_per_gpu:4916.964682749068 mfu:23.194161329208992 +Step 11919 | loss:1.1376557350158691 lr:7.588094775289664e-05 tokens_per_second_per_gpu:4287.544229717895 mfu:20.225077662057934 +Step 11920 | loss:1.1827811002731323 lr:7.587723753545179e-05 tokens_per_second_per_gpu:5291.908708125008 mfu:24.962836268954618 +Step 11921 | loss:1.035403847694397 lr:7.587352712338095e-05 tokens_per_second_per_gpu:5320.184127767344 mfu:25.096216247691572 +Step 11922 | loss:1.070290446281433 lr:7.586981651671199e-05 tokens_per_second_per_gpu:4385.664119978789 mfu:20.687925925398122 +Step 11923 | loss:1.0514658689498901 lr:7.586610571547286e-05 tokens_per_second_per_gpu:3570.324719504206 mfu:16.8418308621131 +Step 11924 | loss:0.9019290804862976 lr:7.586239471969143e-05 tokens_per_second_per_gpu:4890.6717641902105 mfu:23.070133146332367 +Step 11925 | loss:0.9663509130477905 lr:7.585868352939563e-05 tokens_per_second_per_gpu:4384.440916340726 mfu:20.68215586513732 +Step 11926 | loss:1.0478984117507935 lr:7.58549721446134e-05 tokens_per_second_per_gpu:5346.963675668006 mfu:25.22253994420101 +Step 11927 | loss:1.1309096813201904 lr:7.58512605653726e-05 tokens_per_second_per_gpu:5304.832523923669 mfu:25.023800113107253 +Step 11928 | loss:1.008339524269104 lr:7.584754879170116e-05 tokens_per_second_per_gpu:4911.816499785475 mfu:23.169876471799135 +Step 11929 | loss:1.0067850351333618 lr:7.5843836823627e-05 tokens_per_second_per_gpu:4267.09459233978 mfu:20.12861323347745 +Step 11930 | loss:1.0734107494354248 lr:7.584012466117809e-05 tokens_per_second_per_gpu:5340.143579247402 mfu:25.190368385756585 +Step 11931 | loss:1.2063920497894287 lr:7.583641230438227e-05 tokens_per_second_per_gpu:4552.1938004094845 mfu:21.473474840883043 +Step 11932 | loss:1.135891318321228 lr:7.583269975326753e-05 tokens_per_second_per_gpu:5334.47992360208 mfu:25.163651955758304 +Step 11933 | loss:1.1633707284927368 lr:7.582898700786173e-05 tokens_per_second_per_gpu:5024.298294572337 mfu:23.700472289181906 +Step 11934 | loss:1.0597447156906128 lr:7.582527406819284e-05 tokens_per_second_per_gpu:4025.560490959971 mfu:18.989255667301755 +Step 11935 | loss:1.1488555669784546 lr:7.582156093428876e-05 tokens_per_second_per_gpu:5345.151894731013 mfu:25.213993464399515 +Step 11936 | loss:1.0948482751846313 lr:7.581784760617741e-05 tokens_per_second_per_gpu:4731.851676087817 mfu:22.320951693252145 +Step 11937 | loss:1.0651696920394897 lr:7.581413408388675e-05 tokens_per_second_per_gpu:5338.758750210971 mfu:25.183835910915544 +Step 11938 | loss:0.986090362071991 lr:7.581042036744469e-05 tokens_per_second_per_gpu:5354.534970797054 mfu:25.258255035120886 +Step 11939 | loss:1.1398009061813354 lr:7.580670645687917e-05 tokens_per_second_per_gpu:3772.8018769565465 mfu:17.796950159981137 +Step 11940 | loss:1.1237914562225342 lr:7.58029923522181e-05 tokens_per_second_per_gpu:4509.702852499249 mfu:21.273037789887184 +Step 11941 | loss:1.133845329284668 lr:7.579927805348945e-05 tokens_per_second_per_gpu:4761.326298651323 mfu:22.45998851677337 +Step 11942 | loss:1.0003833770751953 lr:7.579556356072112e-05 tokens_per_second_per_gpu:5162.533879536577 mfu:24.352553128881073 +Step 11943 | loss:1.111321210861206 lr:7.579184887394109e-05 tokens_per_second_per_gpu:5326.558064028566 mfu:25.126283192541425 +Step 11944 | loss:1.1409229040145874 lr:7.578813399317727e-05 tokens_per_second_per_gpu:4200.735631615537 mfu:19.815586693735003 +Step 11945 | loss:0.9892489314079285 lr:7.57844189184576e-05 tokens_per_second_per_gpu:5133.236430254878 mfu:24.214352061959 +Step 11946 | loss:1.1569300889968872 lr:7.578070364981002e-05 tokens_per_second_per_gpu:5207.807720449132 mfu:24.56611756097938 +Step 11947 | loss:1.0099833011627197 lr:7.577698818726246e-05 tokens_per_second_per_gpu:5324.765928553202 mfu:25.117829383734286 +Step 11948 | loss:0.999789834022522 lr:7.577327253084293e-05 tokens_per_second_per_gpu:4804.87496523437 mfu:22.6654150072143 +Step 11949 | loss:1.1493120193481445 lr:7.576955668057928e-05 tokens_per_second_per_gpu:5323.314360223057 mfu:25.110982088257895 +Step 11950 | loss:0.9956602454185486 lr:7.576584063649954e-05 tokens_per_second_per_gpu:4474.690236679224 mfu:21.107877307295865 +Step 11951 | loss:1.0325326919555664 lr:7.576212439863161e-05 tokens_per_second_per_gpu:5341.405782985487 mfu:25.196322416143296 +Step 11952 | loss:0.9092851877212524 lr:7.575840796700346e-05 tokens_per_second_per_gpu:4335.002564530143 mfu:20.448946724594183 +Step 11953 | loss:0.9701122641563416 lr:7.575469134164304e-05 tokens_per_second_per_gpu:4808.705510160588 mfu:22.683484341189626 +Step 11954 | loss:1.147475004196167 lr:7.575097452257827e-05 tokens_per_second_per_gpu:5005.6683947948495 mfu:23.612591873342826 +Step 11955 | loss:1.0749893188476562 lr:7.574725750983716e-05 tokens_per_second_per_gpu:3189.437814380425 mfu:15.045122344637067 +Step 11956 | loss:1.1013702154159546 lr:7.574354030344765e-05 tokens_per_second_per_gpu:5250.410587038686 mfu:24.76708255148088 +Step 11957 | loss:0.9870961904525757 lr:7.573982290343767e-05 tokens_per_second_per_gpu:4587.9023250055625 mfu:21.641918043905275 +Step 11958 | loss:1.049012541770935 lr:7.57361053098352e-05 tokens_per_second_per_gpu:4690.051717711002 mfu:22.1237741577804 +Step 11959 | loss:1.1366536617279053 lr:7.57323875226682e-05 tokens_per_second_per_gpu:5221.944831963561 mfu:24.63280472803289 +Step 11960 | loss:0.9935544729232788 lr:7.572866954196465e-05 tokens_per_second_per_gpu:4621.311444025932 mfu:21.799514558507852 +Step 11961 | loss:1.104640245437622 lr:7.572495136775248e-05 tokens_per_second_per_gpu:5358.210893886984 mfu:25.275594991513238 +Step 11962 | loss:1.1203073263168335 lr:7.572123300005965e-05 tokens_per_second_per_gpu:4653.066690739252 mfu:21.949309475258357 +Step 11963 | loss:1.3023935556411743 lr:7.571751443891418e-05 tokens_per_second_per_gpu:3170.091998286076 mfu:14.953864829383434 +Step 11964 | loss:1.0324591398239136 lr:7.571379568434397e-05 tokens_per_second_per_gpu:4940.053387391236 mfu:23.303074688345323 +Step 11965 | loss:1.0069286823272705 lr:7.571007673637702e-05 tokens_per_second_per_gpu:5337.040384951434 mfu:25.17573008880296 +Step 11966 | loss:1.0613930225372314 lr:7.570635759504132e-05 tokens_per_second_per_gpu:5313.205747730279 mfu:25.063298038422325 +Step 11967 | loss:1.039711594581604 lr:7.570263826036482e-05 tokens_per_second_per_gpu:5253.485331215182 mfu:24.781586644366865 +Step 11968 | loss:1.103188395500183 lr:7.569891873237552e-05 tokens_per_second_per_gpu:3446.8051015062783 mfu:16.259167749396894 +Step 11969 | loss:0.9923791289329529 lr:7.569519901110134e-05 tokens_per_second_per_gpu:4013.155072421054 mfu:18.930737191471025 +Step 11970 | loss:1.3356307744979858 lr:7.569147909657031e-05 tokens_per_second_per_gpu:3691.6348205532954 mfu:17.414071306399165 +Step 11971 | loss:1.1390544176101685 lr:7.568775898881038e-05 tokens_per_second_per_gpu:5075.446107076971 mfu:23.94174524748295 +Step 11972 | loss:1.0972219705581665 lr:7.568403868784952e-05 tokens_per_second_per_gpu:5339.94262678916 mfu:25.189420458725035 +Step 11973 | loss:1.0824004411697388 lr:7.568031819371574e-05 tokens_per_second_per_gpu:5323.304481141368 mfu:25.11093548694403 +Step 11974 | loss:1.1000202894210815 lr:7.567659750643701e-05 tokens_per_second_per_gpu:4618.5167816308785 mfu:21.78633165917181 +Step 11975 | loss:1.1656936407089233 lr:7.567287662604132e-05 tokens_per_second_per_gpu:4783.02428667083 mfu:22.562341628319885 +Step 11976 | loss:1.1230320930480957 lr:7.566915555255663e-05 tokens_per_second_per_gpu:5356.498270455635 mfu:25.267516254583878 +Step 11977 | loss:1.198329210281372 lr:7.566543428601098e-05 tokens_per_second_per_gpu:5352.106593958877 mfu:25.246799967251395 +Step 11978 | loss:1.1029263734817505 lr:7.56617128264323e-05 tokens_per_second_per_gpu:4853.769107521355 mfu:22.896057018582905 +Step 11979 | loss:1.0330921411514282 lr:7.56579911738486e-05 tokens_per_second_per_gpu:3300.8830972886312 mfu:15.570828758641053 +Step 11980 | loss:1.0742868185043335 lr:7.56542693282879e-05 tokens_per_second_per_gpu:4707.557861515235 mfu:22.206353624961096 +Step 11981 | loss:1.1908801794052124 lr:7.565054728977814e-05 tokens_per_second_per_gpu:4068.555061864691 mfu:19.192068393889546 +Step 11982 | loss:1.0722205638885498 lr:7.564682505834735e-05 tokens_per_second_per_gpu:3527.343327972661 mfu:16.63908030487721 +Step 11983 | loss:1.023963451385498 lr:7.564310263402351e-05 tokens_per_second_per_gpu:5319.959942738906 mfu:25.09515872866252 +Step 11984 | loss:1.0638116598129272 lr:7.563938001683464e-05 tokens_per_second_per_gpu:5357.123620829626 mfu:25.27046613899377 +Step 11985 | loss:0.9676260352134705 lr:7.56356572068087e-05 tokens_per_second_per_gpu:4873.407840605775 mfu:22.988696273255492 +Step 11986 | loss:1.0094133615493774 lr:7.563193420397372e-05 tokens_per_second_per_gpu:4445.5690126935515 mfu:20.970507525162294 +Step 11987 | loss:1.0855965614318848 lr:7.56282110083577e-05 tokens_per_second_per_gpu:3867.240939782684 mfu:18.242435332297664 +Step 11988 | loss:0.9041460752487183 lr:7.562448761998861e-05 tokens_per_second_per_gpu:5260.41374639359 mfu:24.814269160872943 +Step 11989 | loss:1.1260539293289185 lr:7.562076403889451e-05 tokens_per_second_per_gpu:4394.577751696086 mfu:20.72997304703101 +Step 11990 | loss:1.0419511795043945 lr:7.561704026510336e-05 tokens_per_second_per_gpu:5331.548669957647 mfu:25.14982473219407 +Step 11991 | loss:0.9743475914001465 lr:7.561331629864318e-05 tokens_per_second_per_gpu:4555.339357071729 mfu:21.488312968346154 +Step 11992 | loss:1.0104012489318848 lr:7.560959213954199e-05 tokens_per_second_per_gpu:4387.956531486872 mfu:20.698739621607487 +Step 11993 | loss:0.9535189867019653 lr:7.56058677878278e-05 tokens_per_second_per_gpu:5347.762493439002 mfu:25.226308103919166 +Step 11994 | loss:1.0682363510131836 lr:7.560214324352858e-05 tokens_per_second_per_gpu:3846.691104718352 mfu:18.145498254135802 +Step 11995 | loss:1.1027677059173584 lr:7.559841850667238e-05 tokens_per_second_per_gpu:5335.642013168406 mfu:25.16913373051559 +Step 11996 | loss:1.022073745727539 lr:7.559469357728722e-05 tokens_per_second_per_gpu:3891.735682145503 mfu:18.357981211257076 +Step 11997 | loss:0.9391825199127197 lr:7.55909684554011e-05 tokens_per_second_per_gpu:4337.923328277203 mfu:20.46272446552275 +Step 11998 | loss:1.0519078969955444 lr:7.558724314104204e-05 tokens_per_second_per_gpu:3694.9749322774896 mfu:17.42982718328401 +Step 11999 | loss:0.9543440937995911 lr:7.558351763423807e-05 tokens_per_second_per_gpu:4213.626316736985 mfu:19.87639425483055 +Step 12000 | loss:1.023808240890503 lr:7.557979193501719e-05 tokens_per_second_per_gpu:4617.6201695314385 mfu:21.78210218692095 +Step 12001 | loss:1.0047338008880615 lr:7.557606604340742e-05 tokens_per_second_per_gpu:4527.118172756734 mfu:21.355188826901767 +Step 12002 | loss:1.1386773586273193 lr:7.55723399594368e-05 tokens_per_second_per_gpu:3743.2878007929958 mfu:17.657727227097016 +Step 12003 | loss:1.1382639408111572 lr:7.556861368313336e-05 tokens_per_second_per_gpu:5179.2458871451645 mfu:24.431386520133298 +Step 12004 | loss:1.0425969362258911 lr:7.556488721452511e-05 tokens_per_second_per_gpu:3878.035725041379 mfu:18.293356176143213 +Step 12005 | loss:1.0175679922103882 lr:7.556116055364008e-05 tokens_per_second_per_gpu:3254.587792471887 mfu:15.35244590702697 +Step 12006 | loss:1.1288114786148071 lr:7.555743370050629e-05 tokens_per_second_per_gpu:4100.953808934461 mfu:19.34489881161391 +Step 12007 | loss:1.099389910697937 lr:7.555370665515179e-05 tokens_per_second_per_gpu:3924.2683801592093 mfu:18.511443498438375 +Step 12008 | loss:1.1424864530563354 lr:7.55499794176046e-05 tokens_per_second_per_gpu:4340.127300762264 mfu:20.473120979771263 +Step 12009 | loss:1.0554286241531372 lr:7.554625198789275e-05 tokens_per_second_per_gpu:4855.224843409229 mfu:22.902923973140158 +Step 12010 | loss:1.0648137331008911 lr:7.554252436604428e-05 tokens_per_second_per_gpu:4485.089068690052 mfu:21.156930372114637 +Step 12011 | loss:1.0336995124816895 lr:7.553879655208724e-05 tokens_per_second_per_gpu:4168.344804158022 mfu:19.662793634172875 +Step 12012 | loss:1.0488091707229614 lr:7.553506854604962e-05 tokens_per_second_per_gpu:5369.053869688553 mfu:25.32674315837168 +Step 12013 | loss:1.1217544078826904 lr:7.553134034795951e-05 tokens_per_second_per_gpu:5061.879153581434 mfu:23.877747613083542 +Step 12014 | loss:1.109598994255066 lr:7.552761195784494e-05 tokens_per_second_per_gpu:3314.1712813027116 mfu:15.633511389833448 +Step 12015 | loss:1.093003511428833 lr:7.552388337573393e-05 tokens_per_second_per_gpu:5365.011052657049 mfu:25.30767250065836 +Step 12016 | loss:0.986571192741394 lr:7.552015460165455e-05 tokens_per_second_per_gpu:5360.554250880676 mfu:25.286649006270416 +Step 12017 | loss:0.9507870674133301 lr:7.551642563563481e-05 tokens_per_second_per_gpu:3954.701079807653 mfu:18.654999735034064 +Step 12018 | loss:1.0406131744384766 lr:7.55126964777028e-05 tokens_per_second_per_gpu:5358.170059959799 mfu:25.27540237091172 +Step 12019 | loss:1.1756454706192017 lr:7.550896712788653e-05 tokens_per_second_per_gpu:5342.875732110419 mfu:25.203256416964784 +Step 12020 | loss:1.0743417739868164 lr:7.550523758621407e-05 tokens_per_second_per_gpu:3753.0677480444565 mfu:17.703860906913224 +Step 12021 | loss:1.0804097652435303 lr:7.550150785271346e-05 tokens_per_second_per_gpu:5304.695236357797 mfu:25.023152504235355 +Step 12022 | loss:1.0215708017349243 lr:7.549777792741275e-05 tokens_per_second_per_gpu:5355.384248030535 mfu:25.262261220732853 +Step 12023 | loss:1.125715732574463 lr:7.549404781034e-05 tokens_per_second_per_gpu:5273.431782854741 mfu:24.87567745996624 +Step 12024 | loss:1.1166212558746338 lr:7.549031750152327e-05 tokens_per_second_per_gpu:3979.426561671649 mfu:18.771634051588745 +Step 12025 | loss:1.038451910018921 lr:7.548658700099061e-05 tokens_per_second_per_gpu:5294.780560601145 mfu:24.976383287073116 +Step 12026 | loss:0.89742112159729 lr:7.548285630877007e-05 tokens_per_second_per_gpu:5180.562570223909 mfu:24.437597538865035 +Step 12027 | loss:1.1975961923599243 lr:7.547912542488973e-05 tokens_per_second_per_gpu:5084.349263139025 mfu:23.983742953662716 +Step 12028 | loss:0.9430655837059021 lr:7.547539434937762e-05 tokens_per_second_per_gpu:5000.430029235104 mfu:23.58788160924018 +Step 12029 | loss:1.0484274625778198 lr:7.547166308226184e-05 tokens_per_second_per_gpu:3496.6123794610226 mfu:16.49411717750733 +Step 12030 | loss:1.137219786643982 lr:7.54679316235704e-05 tokens_per_second_per_gpu:5340.79145288302 mfu:25.193424516234433 +Step 12031 | loss:1.1616339683532715 lr:7.546419997333141e-05 tokens_per_second_per_gpu:3741.7928728375073 mfu:17.650675396869968 +Step 12032 | loss:1.1420936584472656 lr:7.546046813157293e-05 tokens_per_second_per_gpu:5376.006550141209 mfu:25.359540138316465 +Step 12033 | loss:1.0929731130599976 lr:7.545673609832302e-05 tokens_per_second_per_gpu:4942.893482114505 mfu:23.31647189972511 +Step 12034 | loss:1.2595529556274414 lr:7.545300387360975e-05 tokens_per_second_per_gpu:4551.959889526582 mfu:21.472371443339078 +Step 12035 | loss:1.1048612594604492 lr:7.544927145746118e-05 tokens_per_second_per_gpu:4260.182577150595 mfu:20.09600807851866 +Step 12036 | loss:0.9892339110374451 lr:7.54455388499054e-05 tokens_per_second_per_gpu:5370.884395381747 mfu:25.33537806038636 +Step 12037 | loss:1.1773091554641724 lr:7.544180605097047e-05 tokens_per_second_per_gpu:4533.143281736497 mfu:21.383610293948905 +Step 12038 | loss:1.0316020250320435 lr:7.543807306068447e-05 tokens_per_second_per_gpu:5358.581122041472 mfu:25.277341420885396 +Step 12039 | loss:0.9963098764419556 lr:7.543433987907548e-05 tokens_per_second_per_gpu:4116.735392565191 mfu:19.419343234215884 +Step 12040 | loss:1.1276146173477173 lr:7.543060650617158e-05 tokens_per_second_per_gpu:5331.277809800798 mfu:25.14854703862071 +Step 12041 | loss:1.126718282699585 lr:7.542687294200085e-05 tokens_per_second_per_gpu:4663.977485628302 mfu:22.00077755632357 +Step 12042 | loss:1.0730646848678589 lr:7.542313918659134e-05 tokens_per_second_per_gpu:5305.3645727355015 mfu:25.026309878129553 +Step 12043 | loss:1.0702232122421265 lr:7.541940523997118e-05 tokens_per_second_per_gpu:5334.861509281954 mfu:25.165451960515497 +Step 12044 | loss:1.03079092502594 lr:7.541567110216841e-05 tokens_per_second_per_gpu:4253.73767834723 mfu:20.065606391250068 +Step 12045 | loss:1.0310770273208618 lr:7.541193677321114e-05 tokens_per_second_per_gpu:4743.127109793512 mfu:22.37413983782927 +Step 12046 | loss:1.2011560201644897 lr:7.540820225312746e-05 tokens_per_second_per_gpu:4418.92602511955 mfu:20.844828007013305 +Step 12047 | loss:1.0582753419876099 lr:7.540446754194545e-05 tokens_per_second_per_gpu:3598.6437439323417 mfu:16.97541653206441 +Step 12048 | loss:0.9143121838569641 lr:7.540073263969318e-05 tokens_per_second_per_gpu:4906.959904332478 mfu:23.146967082426716 +Step 12049 | loss:1.0392067432403564 lr:7.539699754639878e-05 tokens_per_second_per_gpu:4936.589935648286 mfu:23.286736995548026 +Step 12050 | loss:1.04270601272583 lr:7.539326226209031e-05 tokens_per_second_per_gpu:3804.195229234267 mfu:17.945037958933362 +Step 12051 | loss:1.0379588603973389 lr:7.538952678679587e-05 tokens_per_second_per_gpu:4158.162487759742 mfu:19.614761910439867 +Step 12052 | loss:1.1380056142807007 lr:7.538579112054356e-05 tokens_per_second_per_gpu:5202.433090896218 mfu:24.540764516371926 +Step 12053 | loss:1.0525574684143066 lr:7.538205526336148e-05 tokens_per_second_per_gpu:3464.8526582364952 mfu:16.34430115372403 +Step 12054 | loss:1.1461920738220215 lr:7.537831921527774e-05 tokens_per_second_per_gpu:5332.141075848047 mfu:25.152619211853953 +Step 12055 | loss:1.09938383102417 lr:7.537458297632039e-05 tokens_per_second_per_gpu:3699.3001005032033 mfu:17.450229739754498 +Step 12056 | loss:1.175752878189087 lr:7.537084654651758e-05 tokens_per_second_per_gpu:3640.7349195750903 mfu:17.173967789066356 +Step 12057 | loss:1.226853370666504 lr:7.536710992589739e-05 tokens_per_second_per_gpu:4790.039653537127 mfu:22.595434310773225 +Step 12058 | loss:0.9554533362388611 lr:7.536337311448794e-05 tokens_per_second_per_gpu:5357.699704393629 mfu:25.273183623455164 +Step 12059 | loss:0.8882778286933899 lr:7.535963611231733e-05 tokens_per_second_per_gpu:4835.2225997603455 mfu:22.8085699771137 +Step 12060 | loss:1.2573447227478027 lr:7.535589891941364e-05 tokens_per_second_per_gpu:4218.415632837088 mfu:19.898986276016323 +Step 12061 | loss:1.100352168083191 lr:7.535216153580502e-05 tokens_per_second_per_gpu:4537.1395753979405 mfu:21.402461492987943 +Step 12062 | loss:1.0496786832809448 lr:7.534842396151953e-05 tokens_per_second_per_gpu:4293.340415895802 mfu:20.252419261190457 +Step 12063 | loss:1.1100713014602661 lr:7.534468619658534e-05 tokens_per_second_per_gpu:5110.728312556672 mfu:24.10817743049569 +Step 12064 | loss:0.9467974901199341 lr:7.53409482410305e-05 tokens_per_second_per_gpu:4636.377107499981 mfu:21.870581863582185 +Step 12065 | loss:0.9856460094451904 lr:7.533721009488318e-05 tokens_per_second_per_gpu:5336.343591932959 mfu:25.172443197249663 +Step 12066 | loss:0.8956771492958069 lr:7.533347175817146e-05 tokens_per_second_per_gpu:4087.9574745818954 mfu:19.28359288506956 +Step 12067 | loss:1.1404787302017212 lr:7.532973323092347e-05 tokens_per_second_per_gpu:3547.0917323138765 mfu:16.73223689757995 +Step 12068 | loss:1.0830411911010742 lr:7.532599451316734e-05 tokens_per_second_per_gpu:5142.456829141624 mfu:24.257846256670053 +Step 12069 | loss:1.0968265533447266 lr:7.532225560493114e-05 tokens_per_second_per_gpu:4971.549313190751 mfu:23.451646344100695 +Step 12070 | loss:1.1079233884811401 lr:7.531851650624305e-05 tokens_per_second_per_gpu:3708.469724387243 mfu:17.493484420114672 +Step 12071 | loss:1.0452494621276855 lr:7.531477721713114e-05 tokens_per_second_per_gpu:5343.587680619853 mfu:25.206614799554725 +Step 12072 | loss:1.2001702785491943 lr:7.531103773762358e-05 tokens_per_second_per_gpu:4611.037775924287 mfu:21.751051913203863 +Step 12073 | loss:0.9655498266220093 lr:7.530729806774848e-05 tokens_per_second_per_gpu:4699.548923834905 mfu:22.168574099457302 +Step 12074 | loss:1.0639959573745728 lr:7.530355820753394e-05 tokens_per_second_per_gpu:5308.850705543067 mfu:25.04275456137843 +Step 12075 | loss:1.1901830434799194 lr:7.529981815700812e-05 tokens_per_second_per_gpu:3684.9987917951653 mfu:17.38276802652381 +Step 12076 | loss:1.0564748048782349 lr:7.529607791619916e-05 tokens_per_second_per_gpu:5380.316644157646 mfu:25.379871587169866 +Step 12077 | loss:1.156860113143921 lr:7.529233748513516e-05 tokens_per_second_per_gpu:5280.16580447101 mfu:24.90744298887275 +Step 12078 | loss:1.0128192901611328 lr:7.528859686384424e-05 tokens_per_second_per_gpu:4533.871628933749 mfu:21.38704603194719 +Step 12079 | loss:0.9352198839187622 lr:7.528485605235459e-05 tokens_per_second_per_gpu:4598.363799546793 mfu:21.691266604227874 +Step 12080 | loss:1.0101187229156494 lr:7.528111505069429e-05 tokens_per_second_per_gpu:2965.1246897276897 mfu:13.986999063884586 +Step 12081 | loss:1.0032366514205933 lr:7.52773738588915e-05 tokens_per_second_per_gpu:4246.914506926268 mfu:20.0334203275043 +Step 12082 | loss:1.074954867362976 lr:7.527363247697434e-05 tokens_per_second_per_gpu:4581.626373306616 mfu:21.61231330895357 +Step 12083 | loss:1.066546082496643 lr:7.526989090497101e-05 tokens_per_second_per_gpu:4992.2265598917475 mfu:23.549184444691374 +Step 12084 | loss:1.1027926206588745 lr:7.526614914290956e-05 tokens_per_second_per_gpu:3655.1203072193284 mfu:17.24182611698445 +Step 12085 | loss:1.1832261085510254 lr:7.526240719081821e-05 tokens_per_second_per_gpu:4841.672734011405 mfu:22.83899636088233 +Step 12086 | loss:1.1648019552230835 lr:7.525866504872506e-05 tokens_per_second_per_gpu:3981.756309136213 mfu:18.782623867875 +Step 12087 | loss:1.0593361854553223 lr:7.525492271665826e-05 tokens_per_second_per_gpu:3864.3519213140016 mfu:18.22880733926351 +Step 12088 | loss:1.0924320220947266 lr:7.525118019464598e-05 tokens_per_second_per_gpu:5372.486423258075 mfu:25.34293509921314 +Step 12089 | loss:0.9861123561859131 lr:7.524743748271632e-05 tokens_per_second_per_gpu:5037.591562320025 mfu:23.763178901213106 +Step 12090 | loss:1.1779377460479736 lr:7.52436945808975e-05 tokens_per_second_per_gpu:3602.7972185239732 mfu:16.995009180368978 +Step 12091 | loss:1.1199169158935547 lr:7.52399514892176e-05 tokens_per_second_per_gpu:5319.116177666331 mfu:25.091178544854017 +Step 12092 | loss:1.069930911064148 lr:7.523620820770483e-05 tokens_per_second_per_gpu:4457.514164927639 mfu:21.026854846303614 +Step 12093 | loss:0.976844310760498 lr:7.523246473638731e-05 tokens_per_second_per_gpu:5022.447427015405 mfu:23.691741431125287 +Step 12094 | loss:1.2045176029205322 lr:7.522872107529319e-05 tokens_per_second_per_gpu:5349.521541419357 mfu:25.234605833367034 +Step 12095 | loss:1.12909996509552 lr:7.522497722445065e-05 tokens_per_second_per_gpu:4639.501812492761 mfu:21.885321630162775 +Step 12096 | loss:0.977999210357666 lr:7.522123318388784e-05 tokens_per_second_per_gpu:5335.812327510533 mfu:25.169937132326023 +Step 12097 | loss:1.0743913650512695 lr:7.521748895363292e-05 tokens_per_second_per_gpu:5337.053475632993 mfu:25.175791839781372 +Step 12098 | loss:1.120769739151001 lr:7.521374453371403e-05 tokens_per_second_per_gpu:4645.220941208865 mfu:21.912299736100614 +Step 12099 | loss:1.014642596244812 lr:7.520999992415937e-05 tokens_per_second_per_gpu:4789.218729286992 mfu:22.591561871021216 +Step 12100 | loss:0.9419457912445068 lr:7.520625512499707e-05 tokens_per_second_per_gpu:3217.5536552571266 mfu:15.177749563109575 +Step 12101 | loss:1.0787898302078247 lr:7.520251013625532e-05 tokens_per_second_per_gpu:4868.6722280314425 mfu:22.966357580763543 +Step 12102 | loss:1.1036779880523682 lr:7.519876495796227e-05 tokens_per_second_per_gpu:4386.439227514133 mfu:20.69158223988932 +Step 12103 | loss:1.0977109670639038 lr:7.51950195901461e-05 tokens_per_second_per_gpu:4267.91356194253 mfu:20.13247645282424 +Step 12104 | loss:1.0343961715698242 lr:7.519127403283496e-05 tokens_per_second_per_gpu:4105.460925505506 mfu:19.366159649472905 +Step 12105 | loss:1.090646505355835 lr:7.518752828605704e-05 tokens_per_second_per_gpu:4066.0854864134685 mfu:19.180418985133105 +Step 12106 | loss:1.1130579710006714 lr:7.518378234984051e-05 tokens_per_second_per_gpu:4696.90721291992 mfu:22.156112698348963 +Step 12107 | loss:1.0965375900268555 lr:7.518003622421352e-05 tokens_per_second_per_gpu:4926.680401651614 mfu:23.239992032944883 +Step 12108 | loss:1.1480772495269775 lr:7.51762899092043e-05 tokens_per_second_per_gpu:4258.187479803188 mfu:20.086596863932236 +Step 12109 | loss:1.1273854970932007 lr:7.517254340484097e-05 tokens_per_second_per_gpu:5022.162506669572 mfu:23.690397413241573 +Step 12110 | loss:1.0253607034683228 lr:7.516879671115172e-05 tokens_per_second_per_gpu:5356.433877842135 mfu:25.267212503639488 +Step 12111 | loss:0.9990745782852173 lr:7.516504982816475e-05 tokens_per_second_per_gpu:4590.27806229159 mfu:21.653124802026046 +Step 12112 | loss:1.0300871133804321 lr:7.516130275590823e-05 tokens_per_second_per_gpu:3897.7220148629435 mfu:18.386219764058925 +Step 12113 | loss:1.074817419052124 lr:7.515755549441034e-05 tokens_per_second_per_gpu:4215.346714724211 mfu:19.88450966566646 +Step 12114 | loss:1.1211178302764893 lr:7.515380804369926e-05 tokens_per_second_per_gpu:4803.77132255979 mfu:22.660208936417447 +Step 12115 | loss:1.029367446899414 lr:7.515006040380319e-05 tokens_per_second_per_gpu:5303.454870410378 mfu:25.017301486433134 +Step 12116 | loss:1.0823391675949097 lr:7.514631257475029e-05 tokens_per_second_per_gpu:4207.537298047255 mfu:19.847671314777376 +Step 12117 | loss:0.9948790669441223 lr:7.514256455656878e-05 tokens_per_second_per_gpu:4271.297713838217 mfu:20.14844007471205 +Step 12118 | loss:1.0711232423782349 lr:7.513881634928683e-05 tokens_per_second_per_gpu:5271.400620875484 mfu:24.866096122358115 +Step 12119 | loss:0.9315612316131592 lr:7.513506795293263e-05 tokens_per_second_per_gpu:4823.292700241595 mfu:22.752294605633008 +Step 12120 | loss:1.0671709775924683 lr:7.51313193675344e-05 tokens_per_second_per_gpu:5383.301598251039 mfu:25.393952124914083 +Step 12121 | loss:1.1255567073822021 lr:7.512757059312027e-05 tokens_per_second_per_gpu:3698.6194522702845 mfu:17.447019005909258 +Step 12122 | loss:1.2473677396774292 lr:7.51238216297185e-05 tokens_per_second_per_gpu:5323.3483292089795 mfu:25.111142325760202 +Step 12123 | loss:1.2138746976852417 lr:7.512007247735726e-05 tokens_per_second_per_gpu:3252.9727187976873 mfu:15.344827328945788 +Step 12124 | loss:1.1013797521591187 lr:7.511632313606473e-05 tokens_per_second_per_gpu:5344.410370402823 mfu:25.210495567626346 +Step 12125 | loss:1.1780098676681519 lr:7.511257360586914e-05 tokens_per_second_per_gpu:4072.428427763047 mfu:19.210339721696453 +Step 12126 | loss:1.175957202911377 lr:7.510882388679871e-05 tokens_per_second_per_gpu:5379.647081225609 mfu:25.37671314458023 +Step 12127 | loss:1.156317114830017 lr:7.510507397888158e-05 tokens_per_second_per_gpu:3418.0121712889904 mfu:16.123346584981487 +Step 12128 | loss:1.065098524093628 lr:7.5101323882146e-05 tokens_per_second_per_gpu:4624.632774162319 mfu:21.81518183943854 +Step 12129 | loss:1.0116606950759888 lr:7.509757359662014e-05 tokens_per_second_per_gpu:5345.647564422379 mfu:25.216331622903656 +Step 12130 | loss:1.0699845552444458 lr:7.509382312233224e-05 tokens_per_second_per_gpu:5374.264071723859 mfu:25.351320570324347 +Step 12131 | loss:1.0977565050125122 lr:7.50900724593105e-05 tokens_per_second_per_gpu:3681.3916653015463 mfu:17.365752595414165 +Step 12132 | loss:1.153565764427185 lr:7.50863216075831e-05 tokens_per_second_per_gpu:4114.599504068866 mfu:19.40926788375856 +Step 12133 | loss:1.2029057741165161 lr:7.50825705671783e-05 tokens_per_second_per_gpu:3294.081538679363 mfu:15.538744646215571 +Step 12134 | loss:0.9355360865592957 lr:7.507881933812427e-05 tokens_per_second_per_gpu:4159.227678854939 mfu:19.61978659857625 +Step 12135 | loss:1.0398927927017212 lr:7.507506792044925e-05 tokens_per_second_per_gpu:3964.6584469840313 mfu:18.701970334907067 +Step 12136 | loss:1.130860447883606 lr:7.507131631418144e-05 tokens_per_second_per_gpu:4321.887971201516 mfu:20.387082950282696 +Step 12137 | loss:0.9636344909667969 lr:7.506756451934906e-05 tokens_per_second_per_gpu:3713.7512558669655 mfu:17.518398305226714 +Step 12138 | loss:1.256441593170166 lr:7.506381253598034e-05 tokens_per_second_per_gpu:5202.022370984883 mfu:24.538827080474114 +Step 12139 | loss:1.1204208135604858 lr:7.506006036410346e-05 tokens_per_second_per_gpu:3991.787285320504 mfu:18.82994169399732 +Step 12140 | loss:0.9961352348327637 lr:7.50563080037467e-05 tokens_per_second_per_gpu:4534.966808103191 mfu:21.392212178946284 +Step 12141 | loss:1.0375367403030396 lr:7.505255545493823e-05 tokens_per_second_per_gpu:5212.375821198308 mfu:24.5876660716116 +Step 12142 | loss:1.1598775386810303 lr:7.50488027177063e-05 tokens_per_second_per_gpu:3875.5456311010935 mfu:18.281609978173048 +Step 12143 | loss:1.071736216545105 lr:7.504504979207912e-05 tokens_per_second_per_gpu:3910.5791527928823 mfu:18.44686907732838 +Step 12144 | loss:1.04285728931427 lr:7.504129667808495e-05 tokens_per_second_per_gpu:4848.119505572662 mfu:22.869406882269427 +Step 12145 | loss:1.1990214586257935 lr:7.503754337575198e-05 tokens_per_second_per_gpu:3733.701592390641 mfu:17.612507446487303 +Step 12146 | loss:0.9784530401229858 lr:7.503378988510842e-05 tokens_per_second_per_gpu:5292.956843794651 mfu:24.96778050373871 +Step 12147 | loss:1.2500907182693481 lr:7.503003620618257e-05 tokens_per_second_per_gpu:4163.994438830592 mfu:19.642272218674524 +Step 12148 | loss:1.0458580255508423 lr:7.502628233900262e-05 tokens_per_second_per_gpu:4175.0906298431255 mfu:19.69461484488605 +Step 12149 | loss:1.0237693786621094 lr:7.50225282835968e-05 tokens_per_second_per_gpu:4893.205383159754 mfu:23.082084659291677 +Step 12150 | loss:1.0268619060516357 lr:7.501877403999335e-05 tokens_per_second_per_gpu:5306.734340091517 mfu:25.03277130445429 +Step 12151 | loss:1.0227062702178955 lr:7.501501960822051e-05 tokens_per_second_per_gpu:3358.658303731429 mfu:15.843364264898419 +Step 12152 | loss:1.0791244506835938 lr:7.501126498830652e-05 tokens_per_second_per_gpu:4145.867179225974 mfu:19.556762842290127 +Step 12153 | loss:1.2050620317459106 lr:7.500751018027961e-05 tokens_per_second_per_gpu:5358.4979517164775 mfu:25.276949092272076 +Step 12154 | loss:1.1534820795059204 lr:7.500375518416802e-05 tokens_per_second_per_gpu:4746.368641327195 mfu:22.389430695135964 +Step 12155 | loss:0.9697938561439514 lr:7.500000000000001e-05 tokens_per_second_per_gpu:3208.698600188894 mfu:15.135978757524535 +Step 12156 | loss:0.8742004036903381 lr:7.49962446278038e-05 tokens_per_second_per_gpu:4578.982412974131 mfu:21.59984129695909 +Step 12157 | loss:1.085792064666748 lr:7.499248906760764e-05 tokens_per_second_per_gpu:5347.201322834499 mfu:25.223660966432146 +Step 12158 | loss:1.1588093042373657 lr:7.498873331943978e-05 tokens_per_second_per_gpu:3510.7451654090455 mfu:16.560783940126083 +Step 12159 | loss:1.1267837285995483 lr:7.498497738332849e-05 tokens_per_second_per_gpu:4338.388689845284 mfu:20.46491965543826 +Step 12160 | loss:1.0224729776382446 lr:7.498122125930198e-05 tokens_per_second_per_gpu:4804.662558933966 mfu:22.66441305045504 +Step 12161 | loss:1.0825108289718628 lr:7.49774649473885e-05 tokens_per_second_per_gpu:5347.514705343801 mfu:25.225139245197095 +Step 12162 | loss:1.0294843912124634 lr:7.497370844761637e-05 tokens_per_second_per_gpu:4608.185527929636 mfu:21.737597372769404 +Step 12163 | loss:1.0250450372695923 lr:7.496995176001374e-05 tokens_per_second_per_gpu:3684.8218769050063 mfu:17.381933488802137 +Step 12164 | loss:1.2955890893936157 lr:7.496619488460894e-05 tokens_per_second_per_gpu:5308.149212383671 mfu:25.039445498458484 +Step 12165 | loss:1.0578030347824097 lr:7.496243782143021e-05 tokens_per_second_per_gpu:4600.9735640074205 mfu:21.703577308460385 +Step 12166 | loss:0.9699429869651794 lr:7.495868057050579e-05 tokens_per_second_per_gpu:4005.8745126449253 mfu:18.89639354881544 +Step 12167 | loss:1.1459369659423828 lr:7.495492313186395e-05 tokens_per_second_per_gpu:4276.045734814747 mfu:20.170837299754258 +Step 12168 | loss:0.9206417798995972 lr:7.495116550553294e-05 tokens_per_second_per_gpu:4255.078605833646 mfu:20.07193177498931 +Step 12169 | loss:1.1992459297180176 lr:7.494740769154105e-05 tokens_per_second_per_gpu:4433.187859959689 mfu:20.912103515274183 +Step 12170 | loss:1.1119166612625122 lr:7.49436496899165e-05 tokens_per_second_per_gpu:5328.1385933969605 mfu:25.133738819238257 +Step 12171 | loss:1.1220903396606445 lr:7.493989150068758e-05 tokens_per_second_per_gpu:4289.959256662842 mfu:20.236469756203334 +Step 12172 | loss:1.0542834997177124 lr:7.493613312388257e-05 tokens_per_second_per_gpu:4672.765953005067 mfu:22.042234256406925 +Step 12173 | loss:0.9938806891441345 lr:7.493237455952972e-05 tokens_per_second_per_gpu:5370.137739211159 mfu:25.331855955837018 +Step 12174 | loss:1.0053826570510864 lr:7.492861580765729e-05 tokens_per_second_per_gpu:3609.3532579952225 mfu:17.02593513713054 +Step 12175 | loss:1.0352060794830322 lr:7.492485686829355e-05 tokens_per_second_per_gpu:4348.354586176317 mfu:20.51193048878899 +Step 12176 | loss:0.953784167766571 lr:7.49210977414668e-05 tokens_per_second_per_gpu:4806.628969928898 mfu:22.67368894662245 +Step 12177 | loss:1.2594870328903198 lr:7.49173384272053e-05 tokens_per_second_per_gpu:5299.365882396443 mfu:24.998013032318106 +Step 12178 | loss:1.149532437324524 lr:7.49135789255373e-05 tokens_per_second_per_gpu:4428.693570835648 mfu:20.890903186671284 +Step 12179 | loss:1.194809079170227 lr:7.49098192364911e-05 tokens_per_second_per_gpu:2988.2911354298767 mfu:14.09627914086502 +Step 12180 | loss:0.9870853424072266 lr:7.490605936009498e-05 tokens_per_second_per_gpu:5082.947559855873 mfu:23.97713088012085 +Step 12181 | loss:1.0809415578842163 lr:7.490229929637719e-05 tokens_per_second_per_gpu:4490.975786146619 mfu:21.184699022734826 +Step 12182 | loss:1.0502893924713135 lr:7.489853904536603e-05 tokens_per_second_per_gpu:3797.668589672934 mfu:17.91425068656302 +Step 12183 | loss:1.1262537240982056 lr:7.48947786070898e-05 tokens_per_second_per_gpu:3406.005144455736 mfu:16.066707390799 +Step 12184 | loss:1.0351766347885132 lr:7.489101798157676e-05 tokens_per_second_per_gpu:5318.639996877974 mfu:25.08893231883052 +Step 12185 | loss:0.9722188711166382 lr:7.488725716885518e-05 tokens_per_second_per_gpu:5227.109836387106 mfu:24.657168935139374 +Step 12186 | loss:0.9944711327552795 lr:7.488349616895338e-05 tokens_per_second_per_gpu:5182.614273152018 mfu:24.447275771633542 +Step 12187 | loss:1.0176959037780762 lr:7.487973498189964e-05 tokens_per_second_per_gpu:4166.206671061273 mfu:19.652707695552383 +Step 12188 | loss:1.0020484924316406 lr:7.487597360772221e-05 tokens_per_second_per_gpu:4538.145106923232 mfu:21.407204756754112 +Step 12189 | loss:1.1804627180099487 lr:7.487221204644943e-05 tokens_per_second_per_gpu:4200.311078721038 mfu:19.81358400529546 +Step 12190 | loss:1.1163139343261719 lr:7.486845029810957e-05 tokens_per_second_per_gpu:3711.385675546896 mfu:17.507239459247327 +Step 12191 | loss:0.9968069195747375 lr:7.48646883627309e-05 tokens_per_second_per_gpu:5276.411670782759 mfu:24.889734099744008 +Step 12192 | loss:1.0000689029693604 lr:7.486092624034176e-05 tokens_per_second_per_gpu:4716.434469568486 mfu:22.248226099653785 +Step 12193 | loss:1.058278203010559 lr:7.485716393097039e-05 tokens_per_second_per_gpu:5308.867294604694 mfu:25.04283281481251 +Step 12194 | loss:0.9949429631233215 lr:7.485340143464514e-05 tokens_per_second_per_gpu:3980.96549911585 mfu:18.778893482082843 +Step 12195 | loss:0.9361467361450195 lr:7.48496387513943e-05 tokens_per_second_per_gpu:5341.67482027094 mfu:25.19759151092127 +Step 12196 | loss:1.026997685432434 lr:7.484587588124615e-05 tokens_per_second_per_gpu:5365.005061105671 mfu:25.307644237488248 +Step 12197 | loss:1.1213181018829346 lr:7.484211282422898e-05 tokens_per_second_per_gpu:5359.578893838491 mfu:25.28204808068936 +Step 12198 | loss:1.0547996759414673 lr:7.483834958037112e-05 tokens_per_second_per_gpu:3643.2180310558842 mfu:17.185681049583767 +Step 12199 | loss:1.1154898405075073 lr:7.483458614970085e-05 tokens_per_second_per_gpu:4606.713627837816 mfu:21.730654168904234 +Step 12200 | loss:1.0344014167785645 lr:7.48308225322465e-05 tokens_per_second_per_gpu:5185.463561760674 mfu:24.460716352139336 +Step 12201 | loss:1.0448601245880127 lr:7.482705872803637e-05 tokens_per_second_per_gpu:4220.367274545039 mfu:19.90819250293714 +Step 12202 | loss:1.1625438928604126 lr:7.482329473709877e-05 tokens_per_second_per_gpu:5364.876360512173 mfu:25.307037134830132 +Step 12203 | loss:1.1016207933425903 lr:7.4819530559462e-05 tokens_per_second_per_gpu:5338.672574720602 mfu:25.183429406424416 +Step 12204 | loss:1.1379843950271606 lr:7.481576619515434e-05 tokens_per_second_per_gpu:5331.436350156815 mfu:25.149294900529814 +Step 12205 | loss:1.0152816772460938 lr:7.481200164420417e-05 tokens_per_second_per_gpu:5329.935128554356 mfu:25.142213382096198 +Step 12206 | loss:1.069451928138733 lr:7.480823690663977e-05 tokens_per_second_per_gpu:4313.0171967947135 mfu:20.345237993895626 +Step 12207 | loss:1.0309641361236572 lr:7.480447198248944e-05 tokens_per_second_per_gpu:5376.829027751292 mfu:25.363419905532375 +Step 12208 | loss:1.1789147853851318 lr:7.480070687178153e-05 tokens_per_second_per_gpu:3357.1491799307582 mfu:15.836245470447583 +Step 12209 | loss:1.1387220621109009 lr:7.479694157454432e-05 tokens_per_second_per_gpu:5155.910878090562 mfu:24.321311301059648 +Step 12210 | loss:1.1615082025527954 lr:7.479317609080616e-05 tokens_per_second_per_gpu:3370.5815743917906 mfu:15.899608366923092 +Step 12211 | loss:1.0269203186035156 lr:7.478941042059534e-05 tokens_per_second_per_gpu:3682.319969347887 mfu:17.370131563985755 +Step 12212 | loss:0.9038802981376648 lr:7.478564456394021e-05 tokens_per_second_per_gpu:4290.195680396626 mfu:20.237585007293642 +Step 12213 | loss:0.9820168614387512 lr:7.47818785208691e-05 tokens_per_second_per_gpu:5319.012777521671 mfu:25.09069078872977 +Step 12214 | loss:1.0213838815689087 lr:7.477811229141031e-05 tokens_per_second_per_gpu:4241.095974205701 mfu:20.005973315917334 +Step 12215 | loss:0.9938483834266663 lr:7.477434587559216e-05 tokens_per_second_per_gpu:5338.408103549365 mfu:25.182181850786257 +Step 12216 | loss:1.168073058128357 lr:7.477057927344302e-05 tokens_per_second_per_gpu:4054.7989001040037 mfu:19.127178231821166 +Step 12217 | loss:1.0973339080810547 lr:7.476681248499117e-05 tokens_per_second_per_gpu:5326.954069811717 mfu:25.128151219386098 +Step 12218 | loss:0.9757630228996277 lr:7.476304551026498e-05 tokens_per_second_per_gpu:5187.6685626161525 mfu:24.47111771738657 +Step 12219 | loss:1.1856179237365723 lr:7.475927834929276e-05 tokens_per_second_per_gpu:3882.3859697916623 mfu:18.313877022858588 +Step 12220 | loss:1.2516148090362549 lr:7.475551100210285e-05 tokens_per_second_per_gpu:5363.680092100219 mfu:25.301394132626502 +Step 12221 | loss:1.1570947170257568 lr:7.475174346872357e-05 tokens_per_second_per_gpu:2865.0511421502024 mfu:13.514935065653715 +Step 12222 | loss:1.0159112215042114 lr:7.474797574918328e-05 tokens_per_second_per_gpu:5174.4195077940785 mfu:24.408619665268994 +Step 12223 | loss:1.0194921493530273 lr:7.474420784351031e-05 tokens_per_second_per_gpu:4625.273403778944 mfu:21.818203798642845 +Step 12224 | loss:0.9498599171638489 lr:7.4740439751733e-05 tokens_per_second_per_gpu:5064.643625755659 mfu:23.89078809999695 +Step 12225 | loss:1.162245750427246 lr:7.473667147387966e-05 tokens_per_second_per_gpu:4190.801097518913 mfu:19.768723801394927 +Step 12226 | loss:1.0563561916351318 lr:7.473290300997868e-05 tokens_per_second_per_gpu:4763.089630891098 mfu:22.468306455803894 +Step 12227 | loss:1.1851061582565308 lr:7.472913436005837e-05 tokens_per_second_per_gpu:5336.640258813589 mfu:25.17384262554219 +Step 12228 | loss:1.1936012506484985 lr:7.47253655241471e-05 tokens_per_second_per_gpu:4651.695158462946 mfu:21.942839723503482 +Step 12229 | loss:1.0757262706756592 lr:7.472159650227319e-05 tokens_per_second_per_gpu:5324.026701112141 mfu:25.1143423217696 +Step 12230 | loss:1.092437744140625 lr:7.4717827294465e-05 tokens_per_second_per_gpu:4413.573264619323 mfu:20.819578122458616 +Step 12231 | loss:1.266668677330017 lr:7.471405790075086e-05 tokens_per_second_per_gpu:5264.803380030733 mfu:24.83497581168807 +Step 12232 | loss:1.1172494888305664 lr:7.471028832115916e-05 tokens_per_second_per_gpu:4398.072489860054 mfu:20.74645832321426 +Step 12233 | loss:1.074991226196289 lr:7.470651855571822e-05 tokens_per_second_per_gpu:5142.390557294684 mfu:24.2575336410609 +Step 12234 | loss:1.097906231880188 lr:7.470274860445641e-05 tokens_per_second_per_gpu:4131.986889177336 mfu:19.491287145909016 +Step 12235 | loss:1.1345174312591553 lr:7.469897846740205e-05 tokens_per_second_per_gpu:4934.291800455256 mfu:23.275896299740204 +Step 12236 | loss:1.1680917739868164 lr:7.469520814458355e-05 tokens_per_second_per_gpu:3389.7643539371575 mfu:15.99009681095827 +Step 12237 | loss:1.080600380897522 lr:7.469143763602923e-05 tokens_per_second_per_gpu:5332.502586113944 mfu:25.1543245174582 +Step 12238 | loss:1.144689679145813 lr:7.468766694176745e-05 tokens_per_second_per_gpu:4133.182005337801 mfu:19.49692471274577 +Step 12239 | loss:0.8997467756271362 lr:7.468389606182659e-05 tokens_per_second_per_gpu:3610.932689121507 mfu:17.033385583231958 +Step 12240 | loss:0.9914911389350891 lr:7.468012499623498e-05 tokens_per_second_per_gpu:5342.114571194868 mfu:25.199665891058686 +Step 12241 | loss:1.171776294708252 lr:7.467635374502102e-05 tokens_per_second_per_gpu:3793.8421040621542 mfu:17.896200501071206 +Step 12242 | loss:1.1042819023132324 lr:7.467258230821303e-05 tokens_per_second_per_gpu:5355.561910125452 mfu:25.26309928314686 +Step 12243 | loss:1.004770278930664 lr:7.466881068583942e-05 tokens_per_second_per_gpu:2878.432378385683 mfu:13.578056640050804 +Step 12244 | loss:0.8800702691078186 lr:7.466503887792854e-05 tokens_per_second_per_gpu:4617.0340294364 mfu:21.779337264086838 +Step 12245 | loss:1.0398198366165161 lr:7.466126688450874e-05 tokens_per_second_per_gpu:5319.653194857678 mfu:25.09371174657018 +Step 12246 | loss:1.100631594657898 lr:7.465749470560843e-05 tokens_per_second_per_gpu:4603.9545695019915 mfu:21.717639219990257 +Step 12247 | loss:1.0059502124786377 lr:7.465372234125592e-05 tokens_per_second_per_gpu:4619.751840749098 mfu:21.792157643755264 +Step 12248 | loss:0.9838089942932129 lr:7.464994979147963e-05 tokens_per_second_per_gpu:4321.633960726091 mfu:20.385884739531463 +Step 12249 | loss:1.001754879951477 lr:7.464617705630795e-05 tokens_per_second_per_gpu:4237.356286864497 mfu:19.98833257267242 +Step 12250 | loss:1.1920819282531738 lr:7.46424041357692e-05 tokens_per_second_per_gpu:4510.373881574393 mfu:21.276203148524 +Step 12251 | loss:1.0792295932769775 lr:7.463863102989178e-05 tokens_per_second_per_gpu:4879.488937680578 mfu:23.017381845699482 +Step 12252 | loss:1.0237065553665161 lr:7.463485773870408e-05 tokens_per_second_per_gpu:5352.039553531875 mfu:25.24648372611895 +Step 12253 | loss:1.2021230459213257 lr:7.463108426223448e-05 tokens_per_second_per_gpu:5081.014131219142 mfu:23.968010567364445 +Step 12254 | loss:1.0477455854415894 lr:7.462731060051135e-05 tokens_per_second_per_gpu:4468.374500480519 mfu:21.078084902070863 +Step 12255 | loss:1.0286931991577148 lr:7.462353675356307e-05 tokens_per_second_per_gpu:4632.499670409137 mfu:21.85229133126569 +Step 12256 | loss:1.0994967222213745 lr:7.461976272141803e-05 tokens_per_second_per_gpu:5338.151600337651 mfu:25.180971881372642 +Step 12257 | loss:1.143198013305664 lr:7.461598850410461e-05 tokens_per_second_per_gpu:5322.815213191035 mfu:25.10862752654458 +Step 12258 | loss:1.0083640813827515 lr:7.46122141016512e-05 tokens_per_second_per_gpu:5156.352832975561 mfu:24.323396077657467 +Step 12259 | loss:1.0527478456497192 lr:7.460843951408618e-05 tokens_per_second_per_gpu:4819.278045356915 mfu:22.73335679357121 +Step 12260 | loss:1.2137155532836914 lr:7.460466474143797e-05 tokens_per_second_per_gpu:4558.679671059567 mfu:21.50406979934312 +Step 12261 | loss:1.0833265781402588 lr:7.460088978373492e-05 tokens_per_second_per_gpu:4211.794033951596 mfu:19.86775106431219 +Step 12262 | loss:1.0073096752166748 lr:7.459711464100543e-05 tokens_per_second_per_gpu:5012.608398516563 mfu:23.645329054984884 +Step 12263 | loss:1.210981845855713 lr:7.459333931327792e-05 tokens_per_second_per_gpu:3911.8987962036576 mfu:18.453094060451388 +Step 12264 | loss:1.0034213066101074 lr:7.458956380058077e-05 tokens_per_second_per_gpu:5333.220133684055 mfu:25.15770931177266 +Step 12265 | loss:1.0166155099868774 lr:7.458578810294234e-05 tokens_per_second_per_gpu:4396.865753852411 mfu:20.740765943575813 +Step 12266 | loss:0.9680191278457642 lr:7.458201222039109e-05 tokens_per_second_per_gpu:4341.220918356681 mfu:20.478279760554532 +Step 12267 | loss:0.9866182804107666 lr:7.457823615295539e-05 tokens_per_second_per_gpu:5353.22796471896 mfu:25.25208966445228 +Step 12268 | loss:1.1401329040527344 lr:7.457445990066363e-05 tokens_per_second_per_gpu:4860.052353005528 mfu:22.925696159563966 +Step 12269 | loss:1.2033295631408691 lr:7.457068346354422e-05 tokens_per_second_per_gpu:4683.238337112098 mfu:22.091634279012766 +Step 12270 | loss:1.0958963632583618 lr:7.456690684162557e-05 tokens_per_second_per_gpu:3688.5418978526977 mfu:17.399481462313418 +Step 12271 | loss:0.8730965852737427 lr:7.456313003493608e-05 tokens_per_second_per_gpu:4562.853808263825 mfu:21.523759916716852 +Step 12272 | loss:1.0636556148529053 lr:7.455935304350416e-05 tokens_per_second_per_gpu:5354.3893752387685 mfu:25.257568236031293 +Step 12273 | loss:1.07382333278656 lr:7.45555758673582e-05 tokens_per_second_per_gpu:4953.52737250174 mfu:23.36663377501048 +Step 12274 | loss:0.9491668343544006 lr:7.455179850652663e-05 tokens_per_second_per_gpu:5319.331060958609 mfu:25.09219218600656 +Step 12275 | loss:1.0340580940246582 lr:7.454802096103784e-05 tokens_per_second_per_gpu:4496.383942712922 mfu:21.21021021998474 +Step 12276 | loss:1.019574761390686 lr:7.454424323092025e-05 tokens_per_second_per_gpu:3995.8363722125778 mfu:18.84904192771269 +Step 12277 | loss:0.8871393203735352 lr:7.454046531620229e-05 tokens_per_second_per_gpu:5297.477480091867 mfu:24.98910511645276 +Step 12278 | loss:1.0192519426345825 lr:7.453668721691234e-05 tokens_per_second_per_gpu:3848.339173215773 mfu:18.153272474427617 +Step 12279 | loss:1.14335036277771 lr:7.453290893307885e-05 tokens_per_second_per_gpu:3538.8108398591457 mfu:16.69317451500504 +Step 12280 | loss:1.12256920337677 lr:7.452913046473022e-05 tokens_per_second_per_gpu:3540.186106404298 mfu:16.69966188194319 +Step 12281 | loss:1.1052448749542236 lr:7.452535181189488e-05 tokens_per_second_per_gpu:5154.829364254781 mfu:24.31620961577098 +Step 12282 | loss:1.0533576011657715 lr:7.452157297460122e-05 tokens_per_second_per_gpu:4603.196784525848 mfu:21.714064618966336 +Step 12283 | loss:0.9830137491226196 lr:7.451779395287768e-05 tokens_per_second_per_gpu:4341.0126471964895 mfu:20.477297309957184 +Step 12284 | loss:1.2365649938583374 lr:7.451401474675269e-05 tokens_per_second_per_gpu:4063.320045418738 mfu:19.16737392812807 +Step 12285 | loss:1.0338739156723022 lr:7.451023535625467e-05 tokens_per_second_per_gpu:5361.996992194116 mfu:25.29345466320296 +Step 12286 | loss:1.083311915397644 lr:7.450645578141204e-05 tokens_per_second_per_gpu:4104.441975763877 mfu:19.36135308969089 +Step 12287 | loss:1.0870327949523926 lr:7.450267602225323e-05 tokens_per_second_per_gpu:4679.488423915716 mfu:22.073945298664267 +Step 12288 | loss:1.220627784729004 lr:7.449889607880666e-05 tokens_per_second_per_gpu:4168.287646656645 mfu:19.662524012486802 +Step 12289 | loss:0.910054624080658 lr:7.449511595110077e-05 tokens_per_second_per_gpu:5350.650179246832 mfu:25.239929810564703 +Step 12290 | loss:1.2193410396575928 lr:7.449133563916399e-05 tokens_per_second_per_gpu:4143.13468210969 mfu:19.543873187180537 +Step 12291 | loss:0.9290018081665039 lr:7.448755514302475e-05 tokens_per_second_per_gpu:3641.4274592492907 mfu:17.17723461686901 +Step 12292 | loss:1.167952299118042 lr:7.448377446271149e-05 tokens_per_second_per_gpu:5333.9775732222315 mfu:25.161282283306747 +Step 12293 | loss:0.9181132316589355 lr:7.447999359825263e-05 tokens_per_second_per_gpu:3836.0999763472582 mfu:18.09553809457652 +Step 12294 | loss:0.9598256945610046 lr:7.44762125496766e-05 tokens_per_second_per_gpu:5349.782153018309 mfu:25.235835182744214 +Step 12295 | loss:1.061852216720581 lr:7.447243131701187e-05 tokens_per_second_per_gpu:3978.6837249801665 mfu:18.768129964174015 +Step 12296 | loss:1.2436639070510864 lr:7.446864990028687e-05 tokens_per_second_per_gpu:4533.60819470798 mfu:21.385803367757728 +Step 12297 | loss:1.0352884531021118 lr:7.446486829953003e-05 tokens_per_second_per_gpu:5335.457517416381 mfu:25.168263432575067 +Step 12298 | loss:1.0631861686706543 lr:7.446108651476977e-05 tokens_per_second_per_gpu:5017.856862959503 mfu:23.670086957242603 +Step 12299 | loss:1.0543221235275269 lr:7.445730454603457e-05 tokens_per_second_per_gpu:3707.7826480264575 mfu:17.490243363693533 +Step 12300 | loss:0.9629374742507935 lr:7.445352239335287e-05 tokens_per_second_per_gpu:4638.134955277775 mfu:21.878873931469794 +Step 12301 | loss:1.0982598066329956 lr:7.44497400567531e-05 tokens_per_second_per_gpu:5325.628686161157 mfu:25.12189915857264 +Step 12302 | loss:1.1144903898239136 lr:7.444595753626371e-05 tokens_per_second_per_gpu:4078.0251623103613 mfu:19.236740473457516 +Step 12303 | loss:1.0525957345962524 lr:7.444217483191316e-05 tokens_per_second_per_gpu:4411.685908415222 mfu:20.8106751412277 +Step 12304 | loss:1.0625004768371582 lr:7.443839194372991e-05 tokens_per_second_per_gpu:5326.94151736755 mfu:25.128092007365293 +Step 12305 | loss:1.0932884216308594 lr:7.443460887174236e-05 tokens_per_second_per_gpu:4427.221513903902 mfu:20.883959242965428 +Step 12306 | loss:1.064272165298462 lr:7.443082561597903e-05 tokens_per_second_per_gpu:5374.799970019614 mfu:25.3538484940191 +Step 12307 | loss:1.123969316482544 lr:7.442704217646833e-05 tokens_per_second_per_gpu:4557.469522739329 mfu:21.498361323243962 +Step 12308 | loss:1.098279595375061 lr:7.442325855323873e-05 tokens_per_second_per_gpu:5333.551044252895 mfu:25.15927027338504 +Step 12309 | loss:1.0953433513641357 lr:7.441947474631868e-05 tokens_per_second_per_gpu:5256.607679360649 mfu:24.79631529329656 +Step 12310 | loss:1.27628493309021 lr:7.441569075573665e-05 tokens_per_second_per_gpu:4263.800367814697 mfu:20.113073814339966 +Step 12311 | loss:1.0122030973434448 lr:7.44119065815211e-05 tokens_per_second_per_gpu:4379.1605952245545 mfu:20.657247689516655 +Step 12312 | loss:1.0472424030303955 lr:7.440812222370048e-05 tokens_per_second_per_gpu:3132.9171554978802 mfu:14.778504753269296 +Step 12313 | loss:1.1229443550109863 lr:7.440433768230326e-05 tokens_per_second_per_gpu:3311.977627892265 mfu:15.623163552420506 +Step 12314 | loss:1.2625300884246826 lr:7.44005529573579e-05 tokens_per_second_per_gpu:3801.723746088596 mfu:17.933379551256724 +Step 12315 | loss:0.9196680784225464 lr:7.439676804889287e-05 tokens_per_second_per_gpu:4061.50665063824 mfu:19.15881983555141 +Step 12316 | loss:1.127396821975708 lr:7.439298295693665e-05 tokens_per_second_per_gpu:4875.765318525079 mfu:22.99981689882806 +Step 12317 | loss:0.9442082047462463 lr:7.438919768151766e-05 tokens_per_second_per_gpu:3524.2179846759705 mfu:16.62433752730817 +Step 12318 | loss:1.374847412109375 lr:7.438541222266442e-05 tokens_per_second_per_gpu:3417.7240347107495 mfu:16.121987395580753 +Step 12319 | loss:1.1255971193313599 lr:7.438162658040538e-05 tokens_per_second_per_gpu:5350.100312534154 mfu:25.23733599546407 +Step 12320 | loss:1.2148734331130981 lr:7.437784075476902e-05 tokens_per_second_per_gpu:5335.019843818594 mfu:25.16619885153949 +Step 12321 | loss:1.156445860862732 lr:7.437405474578381e-05 tokens_per_second_per_gpu:5300.446564804994 mfu:25.003110795622668 +Step 12322 | loss:1.038680076599121 lr:7.437026855347822e-05 tokens_per_second_per_gpu:5037.7363183268735 mfu:23.763861739995328 +Step 12323 | loss:1.0709284543991089 lr:7.436648217788072e-05 tokens_per_second_per_gpu:3688.588908814291 mfu:17.39970322104015 +Step 12324 | loss:0.9602987170219421 lr:7.436269561901981e-05 tokens_per_second_per_gpu:5321.953848708802 mfu:25.10456432331796 +Step 12325 | loss:1.056808590888977 lr:7.435890887692397e-05 tokens_per_second_per_gpu:2899.906826487668 mfu:13.679355275666522 +Step 12326 | loss:1.1734825372695923 lr:7.435512195162165e-05 tokens_per_second_per_gpu:4188.2259383628525 mfu:19.756576336288337 +Step 12327 | loss:1.055059552192688 lr:7.435133484314137e-05 tokens_per_second_per_gpu:4702.788166127635 mfu:22.183854158023834 +Step 12328 | loss:1.0237900018692017 lr:7.434754755151156e-05 tokens_per_second_per_gpu:4250.244154166311 mfu:20.049126841631985 +Step 12329 | loss:1.0951865911483765 lr:7.434376007676076e-05 tokens_per_second_per_gpu:5214.5903916422385 mfu:24.5981125782403 +Step 12330 | loss:1.002011775970459 lr:7.433997241891742e-05 tokens_per_second_per_gpu:4846.5756239860475 mfu:22.862124129412003 +Step 12331 | loss:1.0439127683639526 lr:7.433618457801006e-05 tokens_per_second_per_gpu:5249.4821960093905 mfu:24.76270317259585 +Step 12332 | loss:1.2629657983779907 lr:7.433239655406715e-05 tokens_per_second_per_gpu:3398.7600032932296 mfu:16.032530823786907 +Step 12333 | loss:0.9912328720092773 lr:7.432860834711717e-05 tokens_per_second_per_gpu:5080.3195209159885 mfu:23.964733972051732 +Step 12334 | loss:1.0760598182678223 lr:7.432481995718863e-05 tokens_per_second_per_gpu:4825.286834422308 mfu:22.761701276797805 +Step 12335 | loss:1.0928878784179688 lr:7.432103138431001e-05 tokens_per_second_per_gpu:5353.760637039408 mfu:25.254602370671105 +Step 12336 | loss:1.1811197996139526 lr:7.431724262850981e-05 tokens_per_second_per_gpu:5332.229442293429 mfu:25.153036051453693 +Step 12337 | loss:1.1406500339508057 lr:7.431345368981653e-05 tokens_per_second_per_gpu:4020.4683905598795 mfu:18.965235360912608 +Step 12338 | loss:1.0295021533966064 lr:7.430966456825866e-05 tokens_per_second_per_gpu:5323.722211606631 mfu:25.112905992820842 +Step 12339 | loss:1.1051006317138672 lr:7.43058752638647e-05 tokens_per_second_per_gpu:4923.644394931033 mfu:23.225670671247737 +Step 12340 | loss:1.0392621755599976 lr:7.430208577666314e-05 tokens_per_second_per_gpu:4683.3639957304695 mfu:22.092227032155193 +Step 12341 | loss:1.0089459419250488 lr:7.429829610668251e-05 tokens_per_second_per_gpu:4769.310620945989 mfu:22.497651927303167 +Step 12342 | loss:1.0305336713790894 lr:7.429450625395128e-05 tokens_per_second_per_gpu:4968.611449403052 mfu:23.43778793936287 +Step 12343 | loss:1.1451828479766846 lr:7.429071621849798e-05 tokens_per_second_per_gpu:5340.239037841504 mfu:25.190818680231878 +Step 12344 | loss:1.112411379814148 lr:7.428692600035109e-05 tokens_per_second_per_gpu:5344.734199738437 mfu:25.21202312585342 +Step 12345 | loss:0.9754660129547119 lr:7.428313559953915e-05 tokens_per_second_per_gpu:4966.764017297853 mfu:23.42907328691823 +Step 12346 | loss:1.0926975011825562 lr:7.427934501609062e-05 tokens_per_second_per_gpu:5363.719448288006 mfu:25.301579782479156 +Step 12347 | loss:0.9739785194396973 lr:7.427555425003407e-05 tokens_per_second_per_gpu:5309.385661443576 mfu:25.045278039634024 +Step 12348 | loss:1.0506011247634888 lr:7.427176330139795e-05 tokens_per_second_per_gpu:5327.196489403012 mfu:25.129294753959282 +Step 12349 | loss:1.0932132005691528 lr:7.426797217021081e-05 tokens_per_second_per_gpu:4380.72222935662 mfu:20.664614184159866 +Step 12350 | loss:1.1004763841629028 lr:7.426418085650118e-05 tokens_per_second_per_gpu:4456.371913484904 mfu:21.02146665136004 +Step 12351 | loss:1.1135271787643433 lr:7.426038936029753e-05 tokens_per_second_per_gpu:5370.476168586664 mfu:25.333452384943286 +Step 12352 | loss:0.9855892658233643 lr:7.42565976816284e-05 tokens_per_second_per_gpu:5356.220609771537 mfu:25.266206481763188 +Step 12353 | loss:1.1069602966308594 lr:7.425280582052228e-05 tokens_per_second_per_gpu:4787.379945490472 mfu:22.582888014124165 +Step 12354 | loss:1.0551315546035767 lr:7.424901377700775e-05 tokens_per_second_per_gpu:4457.297460934248 mfu:21.025832616593274 +Step 12355 | loss:1.3010972738265991 lr:7.424522155111326e-05 tokens_per_second_per_gpu:5353.9950840521715 mfu:25.2557082972311 +Step 12356 | loss:0.9906714558601379 lr:7.42414291428674e-05 tokens_per_second_per_gpu:5339.986139918718 mfu:25.189625717581265 +Step 12357 | loss:0.9835292100906372 lr:7.423763655229862e-05 tokens_per_second_per_gpu:5336.816146771789 mfu:25.174672319050597 +Step 12358 | loss:1.1178432703018188 lr:7.423384377943552e-05 tokens_per_second_per_gpu:4773.72875594732 mfu:22.518493023916008 +Step 12359 | loss:1.1114170551300049 lr:7.423005082430657e-05 tokens_per_second_per_gpu:4188.319877584354 mfu:19.757019463623838 +Step 12360 | loss:1.158203363418579 lr:7.42262576869403e-05 tokens_per_second_per_gpu:5341.380017926802 mfu:25.196200877965506 +Step 12361 | loss:0.9804285168647766 lr:7.422246436736529e-05 tokens_per_second_per_gpu:5323.132352928023 mfu:25.110123528793572 +Step 12362 | loss:1.0660771131515503 lr:7.421867086561001e-05 tokens_per_second_per_gpu:5329.895285989885 mfu:25.14202543792175 +Step 12363 | loss:1.0204516649246216 lr:7.421487718170301e-05 tokens_per_second_per_gpu:3924.0053344201556 mfu:18.510202666807025 +Step 12364 | loss:1.008790373802185 lr:7.421108331567284e-05 tokens_per_second_per_gpu:5111.523895036815 mfu:24.11193032879489 +Step 12365 | loss:1.0275495052337646 lr:7.420728926754803e-05 tokens_per_second_per_gpu:4977.170099530879 mfu:23.478160552272122 +Step 12366 | loss:1.1314001083374023 lr:7.42034950373571e-05 tokens_per_second_per_gpu:5329.798235730862 mfu:25.14156763529145 +Step 12367 | loss:1.0438004732131958 lr:7.41997006251286e-05 tokens_per_second_per_gpu:5299.069774499787 mfu:24.99661624084836 +Step 12368 | loss:0.9763349890708923 lr:7.419590603089106e-05 tokens_per_second_per_gpu:5131.753551953205 mfu:24.207357072004886 +Step 12369 | loss:1.067531943321228 lr:7.419211125467302e-05 tokens_per_second_per_gpu:4185.724923076001 mfu:19.744778620463045 +Step 12370 | loss:1.2382186651229858 lr:7.418831629650303e-05 tokens_per_second_per_gpu:3536.2826203183627 mfu:16.681248472072433 +Step 12371 | loss:1.1350977420806885 lr:7.418452115640962e-05 tokens_per_second_per_gpu:3937.3175624731034 mfu:18.572998717833094 +Step 12372 | loss:1.2103365659713745 lr:7.418072583442136e-05 tokens_per_second_per_gpu:5371.64341136007 mfu:25.33895846825759 +Step 12373 | loss:1.0923665761947632 lr:7.417693033056675e-05 tokens_per_second_per_gpu:4724.84567448785 mfu:22.287903188357966 +Step 12374 | loss:1.2472805976867676 lr:7.417313464487439e-05 tokens_per_second_per_gpu:5184.590177666503 mfu:24.45659645035244 +Step 12375 | loss:1.1270595788955688 lr:7.416933877737277e-05 tokens_per_second_per_gpu:5326.84370485198 mfu:25.127630609040818 +Step 12376 | loss:0.9802757501602173 lr:7.416554272809049e-05 tokens_per_second_per_gpu:4920.591784299812 mfu:23.21127098607119 +Step 12377 | loss:1.1080451011657715 lr:7.416174649705607e-05 tokens_per_second_per_gpu:5299.555759018213 mfu:24.99890871273929 +Step 12378 | loss:0.9771250486373901 lr:7.415795008429806e-05 tokens_per_second_per_gpu:4342.548036395606 mfu:20.484540002773823 +Step 12379 | loss:1.0925358533859253 lr:7.415415348984506e-05 tokens_per_second_per_gpu:4380.993288264028 mfu:20.66589281527353 +Step 12380 | loss:1.049489140510559 lr:7.415035671372556e-05 tokens_per_second_per_gpu:5357.747718853698 mfu:25.273410115855416 +Step 12381 | loss:1.1426550149917603 lr:7.414655975596816e-05 tokens_per_second_per_gpu:5298.101914432723 mfu:24.992050679778067 +Step 12382 | loss:1.146186113357544 lr:7.414276261660137e-05 tokens_per_second_per_gpu:4835.531474051554 mfu:22.810026989844715 +Step 12383 | loss:1.0556198358535767 lr:7.413896529565382e-05 tokens_per_second_per_gpu:4669.631671386264 mfu:22.027449315247445 +Step 12384 | loss:1.0769767761230469 lr:7.413516779315403e-05 tokens_per_second_per_gpu:4983.803965494483 mfu:23.50945362987669 +Step 12385 | loss:1.1358236074447632 lr:7.413137010913054e-05 tokens_per_second_per_gpu:5359.187878675858 mfu:25.28020359545315 +Step 12386 | loss:1.1747945547103882 lr:7.412757224361196e-05 tokens_per_second_per_gpu:4291.49810637812 mfu:20.243728772865133 +Step 12387 | loss:1.110214114189148 lr:7.412377419662682e-05 tokens_per_second_per_gpu:4615.740935775705 mfu:21.773237520664818 +Step 12388 | loss:1.0857269763946533 lr:7.41199759682037e-05 tokens_per_second_per_gpu:5323.564525997503 mfu:25.112162162898642 +Step 12389 | loss:0.8680679202079773 lr:7.411617755837114e-05 tokens_per_second_per_gpu:4926.334531963429 mfu:23.238360506614256 +Step 12390 | loss:1.2175662517547607 lr:7.411237896715777e-05 tokens_per_second_per_gpu:5339.975385164174 mfu:25.189574985569134 +Step 12391 | loss:1.1924512386322021 lr:7.41085801945921e-05 tokens_per_second_per_gpu:3554.0323295788003 mfu:16.764976879066598 +Step 12392 | loss:1.0563896894454956 lr:7.410478124070273e-05 tokens_per_second_per_gpu:4246.991683354318 mfu:20.03378438188307 +Step 12393 | loss:1.065295934677124 lr:7.410098210551822e-05 tokens_per_second_per_gpu:3787.267950836943 mfu:17.86518909864168 +Step 12394 | loss:1.10939359664917 lr:7.409718278906713e-05 tokens_per_second_per_gpu:3772.8719957781364 mfu:17.79728092242608 +Step 12395 | loss:1.0003896951675415 lr:7.409338329137807e-05 tokens_per_second_per_gpu:4240.539913725404 mfu:20.003350283758074 +Step 12396 | loss:1.141793131828308 lr:7.40895836124796e-05 tokens_per_second_per_gpu:5320.164791878607 mfu:25.096125037005354 +Step 12397 | loss:1.1317691802978516 lr:7.408578375240031e-05 tokens_per_second_per_gpu:5290.833337962914 mfu:24.957763564422947 +Step 12398 | loss:1.0487561225891113 lr:7.408198371116874e-05 tokens_per_second_per_gpu:5309.987902207315 mfu:25.048118912069445 +Step 12399 | loss:1.3275294303894043 lr:7.407818348881353e-05 tokens_per_second_per_gpu:4289.690529040813 mfu:20.235202122160562 +Step 12400 | loss:0.9896508455276489 lr:7.407438308536319e-05 tokens_per_second_per_gpu:3919.589386446745 mfu:18.48937188677803 +Step 12401 | loss:1.0432806015014648 lr:7.407058250084637e-05 tokens_per_second_per_gpu:3808.718589214731 mfu:17.966375419726006 +Step 12402 | loss:1.126120686531067 lr:7.406678173529162e-05 tokens_per_second_per_gpu:4969.906341447315 mfu:23.443896166067024 +Step 12403 | loss:1.0261894464492798 lr:7.406298078872753e-05 tokens_per_second_per_gpu:3752.0338428312925 mfu:17.698983799620308 +Step 12404 | loss:0.9005021452903748 lr:7.40591796611827e-05 tokens_per_second_per_gpu:4792.723337075924 mfu:22.608093703909354 +Step 12405 | loss:0.969353973865509 lr:7.405537835268571e-05 tokens_per_second_per_gpu:4575.3348723568515 mfu:21.58263522553278 +Step 12406 | loss:1.0460690259933472 lr:7.405157686326516e-05 tokens_per_second_per_gpu:4394.125258135728 mfu:20.727838557703432 +Step 12407 | loss:1.1988601684570312 lr:7.40477751929496e-05 tokens_per_second_per_gpu:4693.225493176056 mfu:22.13874539815099 +Step 12408 | loss:1.0129523277282715 lr:7.40439733417677e-05 tokens_per_second_per_gpu:5352.359655646269 mfu:25.247993702407598 +Step 12409 | loss:1.1285730600357056 lr:7.404017130974797e-05 tokens_per_second_per_gpu:4498.337204155763 mfu:21.21942408747139 +Step 12410 | loss:1.0587220191955566 lr:7.403636909691905e-05 tokens_per_second_per_gpu:5353.54912459748 mfu:25.25360463039479 +Step 12411 | loss:1.1588507890701294 lr:7.403256670330954e-05 tokens_per_second_per_gpu:3480.948274942279 mfu:16.4202269239206 +Step 12412 | loss:1.027254343032837 lr:7.402876412894801e-05 tokens_per_second_per_gpu:4858.7997606616855 mfu:22.919787467763772 +Step 12413 | loss:1.163629174232483 lr:7.40249613738631e-05 tokens_per_second_per_gpu:5327.034494283241 mfu:25.128530595340106 +Step 12414 | loss:1.039343237876892 lr:7.402115843808339e-05 tokens_per_second_per_gpu:4461.012870941353 mfu:21.043358839511104 +Step 12415 | loss:1.0965999364852905 lr:7.401735532163748e-05 tokens_per_second_per_gpu:5013.958249013476 mfu:23.65169653008707 +Step 12416 | loss:0.9219169020652771 lr:7.401355202455395e-05 tokens_per_second_per_gpu:5356.818756088397 mfu:25.269028039994105 +Step 12417 | loss:1.1491169929504395 lr:7.400974854686147e-05 tokens_per_second_per_gpu:4907.928263476447 mfu:23.15153499772793 +Step 12418 | loss:1.0516479015350342 lr:7.400594488858859e-05 tokens_per_second_per_gpu:4906.517748768846 mfu:23.144881359193548 +Step 12419 | loss:1.2112966775894165 lr:7.400214104976392e-05 tokens_per_second_per_gpu:5347.617196045948 mfu:25.22562271132584 +Step 12420 | loss:1.0939973592758179 lr:7.399833703041611e-05 tokens_per_second_per_gpu:3311.1534968495816 mfu:15.619275985680915 +Step 12421 | loss:1.0176193714141846 lr:7.399453283057374e-05 tokens_per_second_per_gpu:5323.921917694429 mfu:25.113848040510355 +Step 12422 | loss:1.1427536010742188 lr:7.399072845026543e-05 tokens_per_second_per_gpu:5122.97320039621 mfu:24.165938655628086 +Step 12423 | loss:1.118133306503296 lr:7.398692388951977e-05 tokens_per_second_per_gpu:4259.364763738508 mfu:20.092150313120495 +Step 12424 | loss:1.066122055053711 lr:7.398311914836541e-05 tokens_per_second_per_gpu:4755.96294885805 mfu:22.434688680716672 +Step 12425 | loss:1.0905208587646484 lr:7.397931422683095e-05 tokens_per_second_per_gpu:4411.992635924606 mfu:20.81212202722257 +Step 12426 | loss:1.02498459815979 lr:7.397550912494501e-05 tokens_per_second_per_gpu:5350.702004181905 mfu:25.24017427762569 +Step 12427 | loss:1.2055250406265259 lr:7.39717038427362e-05 tokens_per_second_per_gpu:4588.11930050608 mfu:21.642941554361055 +Step 12428 | loss:1.0219049453735352 lr:7.396789838023315e-05 tokens_per_second_per_gpu:4427.087570473868 mfu:20.883327408952855 +Step 12429 | loss:1.0654867887496948 lr:7.396409273746448e-05 tokens_per_second_per_gpu:4864.340408412589 mfu:22.94592364853677 +Step 12430 | loss:1.0163321495056152 lr:7.39602869144588e-05 tokens_per_second_per_gpu:4272.254403786889 mfu:20.152952944428073 +Step 12431 | loss:1.1018908023834229 lr:7.395648091124475e-05 tokens_per_second_per_gpu:3727.57991136499 mfu:17.583630432622822 +Step 12432 | loss:0.944556713104248 lr:7.395267472785094e-05 tokens_per_second_per_gpu:5381.760607193815 mfu:25.386683007176973 +Step 12433 | loss:0.9395146369934082 lr:7.394886836430603e-05 tokens_per_second_per_gpu:5361.785259895372 mfu:25.292455885824534 +Step 12434 | loss:1.0464857816696167 lr:7.394506182063861e-05 tokens_per_second_per_gpu:4361.05014654896 mfu:20.571817613152703 +Step 12435 | loss:0.9957312941551208 lr:7.394125509687734e-05 tokens_per_second_per_gpu:4435.843129225494 mfu:20.924628873436635 +Step 12436 | loss:1.0774202346801758 lr:7.393744819305081e-05 tokens_per_second_per_gpu:4256.997935756615 mfu:20.08098558170741 +Step 12437 | loss:0.9698317646980286 lr:7.39336411091877e-05 tokens_per_second_per_gpu:4962.353295619888 mfu:23.40826716021655 +Step 12438 | loss:1.035995364189148 lr:7.392983384531661e-05 tokens_per_second_per_gpu:4813.024092729925 mfu:22.70385583199535 +Step 12439 | loss:0.998342752456665 lr:7.39260264014662e-05 tokens_per_second_per_gpu:5362.979314733188 mfu:25.298088446221275 +Step 12440 | loss:1.2424261569976807 lr:7.392221877766508e-05 tokens_per_second_per_gpu:4934.1624040577935 mfu:23.275285914856074 +Step 12441 | loss:1.0751906633377075 lr:7.39184109739419e-05 tokens_per_second_per_gpu:2958.5620383756595 mfu:13.95604191775328 +Step 12442 | loss:0.9150295257568359 lr:7.39146029903253e-05 tokens_per_second_per_gpu:5351.241855275496 mfu:25.24272084734225 +Step 12443 | loss:1.0416250228881836 lr:7.391079482684391e-05 tokens_per_second_per_gpu:4119.684517595632 mfu:19.433254760157087 +Step 12444 | loss:1.0822426080703735 lr:7.390698648352641e-05 tokens_per_second_per_gpu:5362.34817194357 mfu:25.29511123799858 +Step 12445 | loss:1.2411437034606934 lr:7.390317796040138e-05 tokens_per_second_per_gpu:5382.673244249104 mfu:25.39098807187871 +Step 12446 | loss:1.1072578430175781 lr:7.389936925749752e-05 tokens_per_second_per_gpu:3598.5413912275403 mfu:16.974933716892693 +Step 12447 | loss:1.1116427183151245 lr:7.389556037484345e-05 tokens_per_second_per_gpu:4664.690744479012 mfu:22.004142120016883 +Step 12448 | loss:1.0787200927734375 lr:7.389175131246782e-05 tokens_per_second_per_gpu:5360.882044074345 mfu:25.288195262692234 +Step 12449 | loss:0.9793509840965271 lr:7.388794207039929e-05 tokens_per_second_per_gpu:5236.457747234687 mfu:24.701264625525468 +Step 12450 | loss:1.0773646831512451 lr:7.388413264866646e-05 tokens_per_second_per_gpu:3811.765601822561 mfu:17.9807486980711 +Step 12451 | loss:0.8864344358444214 lr:7.388032304729808e-05 tokens_per_second_per_gpu:4747.465304815653 mfu:22.394603843921104 +Step 12452 | loss:1.1047114133834839 lr:7.38765132663227e-05 tokens_per_second_per_gpu:4843.883319485908 mfu:22.849424069730333 +Step 12453 | loss:1.0042848587036133 lr:7.387270330576902e-05 tokens_per_second_per_gpu:3565.4127935135984 mfu:16.81866046915441 +Step 12454 | loss:1.04762864112854 lr:7.386889316566571e-05 tokens_per_second_per_gpu:4897.986458514461 mfu:23.104637807475402 +Step 12455 | loss:0.9396135210990906 lr:7.386508284604141e-05 tokens_per_second_per_gpu:5331.043562946643 mfu:25.147442056243023 +Step 12456 | loss:1.176104187965393 lr:7.386127234692476e-05 tokens_per_second_per_gpu:3484.6008736702593 mfu:16.437456855318086 +Step 12457 | loss:1.1017837524414062 lr:7.385746166834444e-05 tokens_per_second_per_gpu:3778.767727143392 mfu:17.82509209319149 +Step 12458 | loss:1.231522798538208 lr:7.385365081032912e-05 tokens_per_second_per_gpu:2787.4060500938926 mfu:13.148669918805346 +Step 12459 | loss:1.0201821327209473 lr:7.384983977290744e-05 tokens_per_second_per_gpu:4045.890025893413 mfu:19.08515355215913 +Step 12460 | loss:1.0775158405303955 lr:7.384602855610808e-05 tokens_per_second_per_gpu:3796.985811102644 mfu:17.911029903552926 +Step 12461 | loss:1.2991153001785278 lr:7.384221715995967e-05 tokens_per_second_per_gpu:4343.407016551401 mfu:20.48859195872584 +Step 12462 | loss:1.1827447414398193 lr:7.383840558449094e-05 tokens_per_second_per_gpu:3781.0333128348107 mfu:17.83577924744669 +Step 12463 | loss:1.0571558475494385 lr:7.383459382973048e-05 tokens_per_second_per_gpu:3943.7447684057415 mfu:18.603316944811763 +Step 12464 | loss:0.9840012192726135 lr:7.383078189570702e-05 tokens_per_second_per_gpu:5310.592596118327 mfu:25.050971356419105 +Step 12465 | loss:0.9777094721794128 lr:7.38269697824492e-05 tokens_per_second_per_gpu:5101.874116888554 mfu:24.066410678847625 +Step 12466 | loss:1.1168450117111206 lr:7.382315748998572e-05 tokens_per_second_per_gpu:5001.444730943129 mfu:23.5926681303229 +Step 12467 | loss:1.0051791667938232 lr:7.381934501834523e-05 tokens_per_second_per_gpu:4593.995901340854 mfu:21.670662483150174 +Step 12468 | loss:1.01634681224823 lr:7.381553236755637e-05 tokens_per_second_per_gpu:3685.686574136764 mfu:17.386012413176896 +Step 12469 | loss:1.1102045774459839 lr:7.381171953764788e-05 tokens_per_second_per_gpu:5086.243254521372 mfu:23.9926772341615 +Step 12470 | loss:1.238107442855835 lr:7.380790652864842e-05 tokens_per_second_per_gpu:4801.725593534327 mfu:22.650558883566976 +Step 12471 | loss:1.018166422843933 lr:7.380409334058664e-05 tokens_per_second_per_gpu:5246.845010377498 mfu:24.750263118019948 +Step 12472 | loss:0.989051342010498 lr:7.380027997349124e-05 tokens_per_second_per_gpu:4954.597403871383 mfu:23.37168129554727 +Step 12473 | loss:1.0819884538650513 lr:7.37964664273909e-05 tokens_per_second_per_gpu:4172.527397903936 mfu:19.682523642496427 +Step 12474 | loss:1.091110110282898 lr:7.379265270231429e-05 tokens_per_second_per_gpu:4486.440296129249 mfu:21.163304342488505 +Step 12475 | loss:1.0335077047348022 lr:7.378883879829012e-05 tokens_per_second_per_gpu:3679.441691300303 mfu:17.35655423535043 +Step 12476 | loss:1.0145665407180786 lr:7.378502471534705e-05 tokens_per_second_per_gpu:4048.8719651340075 mfu:19.099219868328312 +Step 12477 | loss:0.9144405126571655 lr:7.378121045351378e-05 tokens_per_second_per_gpu:5318.491045568968 mfu:25.088229690092223 +Step 12478 | loss:1.0572963953018188 lr:7.377739601281898e-05 tokens_per_second_per_gpu:3976.7960097181517 mfu:18.759225289205656 +Step 12479 | loss:1.1668612957000732 lr:7.377358139329134e-05 tokens_per_second_per_gpu:5327.600995953705 mfu:25.13120288037484 +Step 12480 | loss:1.0774561166763306 lr:7.37697665949596e-05 tokens_per_second_per_gpu:5335.335977992266 mfu:25.16769011037475 +Step 12481 | loss:1.2198865413665771 lr:7.376595161785237e-05 tokens_per_second_per_gpu:4248.480776011466 mfu:20.04080868601224 +Step 12482 | loss:1.0722874402999878 lr:7.376213646199842e-05 tokens_per_second_per_gpu:5372.540859163055 mfu:25.343191882664655 +Step 12483 | loss:1.0413442850112915 lr:7.375832112742638e-05 tokens_per_second_per_gpu:4126.955588221645 mfu:19.46755364087256 +Step 12484 | loss:1.1802079677581787 lr:7.3754505614165e-05 tokens_per_second_per_gpu:4932.586407610112 mfu:23.267851670719637 +Step 12485 | loss:1.2126981019973755 lr:7.375068992224293e-05 tokens_per_second_per_gpu:5156.754378884453 mfu:24.32529023822051 +Step 12486 | loss:1.020269751548767 lr:7.374687405168889e-05 tokens_per_second_per_gpu:4107.09961381894 mfu:19.373889621836433 +Step 12487 | loss:1.1194454431533813 lr:7.37430580025316e-05 tokens_per_second_per_gpu:5287.977043194462 mfu:24.94428993466602 +Step 12488 | loss:1.1122465133666992 lr:7.373924177479973e-05 tokens_per_second_per_gpu:4221.536433285825 mfu:19.91370762419648 +Step 12489 | loss:1.125590205192566 lr:7.373542536852199e-05 tokens_per_second_per_gpu:5262.3589967801 mfu:24.823445238840048 +Step 12490 | loss:1.1848371028900146 lr:7.373160878372709e-05 tokens_per_second_per_gpu:5202.989222135715 mfu:24.543387882314406 +Step 12491 | loss:1.2375608682632446 lr:7.372779202044376e-05 tokens_per_second_per_gpu:4992.169251585805 mfu:23.548914111634435 +Step 12492 | loss:0.9514119625091553 lr:7.372397507870063e-05 tokens_per_second_per_gpu:5328.10403395207 mfu:25.133575796440255 +Step 12493 | loss:1.0243394374847412 lr:7.372015795852649e-05 tokens_per_second_per_gpu:5347.329602633653 mfu:25.224266084131546 +Step 12494 | loss:0.894507646560669 lr:7.371634065995001e-05 tokens_per_second_per_gpu:4793.495398374755 mfu:22.611735648783178 +Step 12495 | loss:1.1326326131820679 lr:7.37125231829999e-05 tokens_per_second_per_gpu:4139.532825372742 mfu:19.526882614412223 +Step 12496 | loss:1.1431223154067993 lr:7.370870552770488e-05 tokens_per_second_per_gpu:4196.385629716256 mfu:19.795066992589238 +Step 12497 | loss:0.9882674813270569 lr:7.370488769409364e-05 tokens_per_second_per_gpu:4253.50413631589 mfu:20.064504733641858 +Step 12498 | loss:1.1693480014801025 lr:7.370106968219494e-05 tokens_per_second_per_gpu:5316.830125709869 mfu:25.080394847734702 +Step 12499 | loss:0.9175753593444824 lr:7.369725149203746e-05 tokens_per_second_per_gpu:5382.8559596743435 mfu:25.39184997171403 +Step 12500 | loss:0.9759883284568787 lr:7.369343312364993e-05 tokens_per_second_per_gpu:5315.461539184325 mfu:25.073938991588772 +Step 12501 | loss:1.0585743188858032 lr:7.368961457706107e-05 tokens_per_second_per_gpu:5306.105239420687 mfu:25.029803729254343 +Step 12502 | loss:1.2891266345977783 lr:7.36857958522996e-05 tokens_per_second_per_gpu:4692.542594794096 mfu:22.135524049968804 +Step 12503 | loss:1.1256535053253174 lr:7.368197694939424e-05 tokens_per_second_per_gpu:5353.7620398574945 mfu:25.25460898800337 +Step 12504 | loss:1.0997503995895386 lr:7.367815786837368e-05 tokens_per_second_per_gpu:4206.977781893541 mfu:19.84503198161701 +Step 12505 | loss:1.3261817693710327 lr:7.36743386092667e-05 tokens_per_second_per_gpu:3273.980574203165 mfu:15.44392496720316 +Step 12506 | loss:1.1887294054031372 lr:7.367051917210199e-05 tokens_per_second_per_gpu:5207.750681362749 mfu:24.565848497877038 +Step 12507 | loss:1.0418317317962646 lr:7.36666995569083e-05 tokens_per_second_per_gpu:5340.271301787277 mfu:25.190970874768887 +Step 12508 | loss:1.1462814807891846 lr:7.366287976371433e-05 tokens_per_second_per_gpu:4310.42993494749 mfu:20.33303343832989 +Step 12509 | loss:1.0445505380630493 lr:7.365905979254882e-05 tokens_per_second_per_gpu:5270.522596085188 mfu:24.861954329615813 +Step 12510 | loss:1.0555124282836914 lr:7.36552396434405e-05 tokens_per_second_per_gpu:3995.6082449730784 mfu:18.84796581260145 +Step 12511 | loss:1.0845471620559692 lr:7.365141931641811e-05 tokens_per_second_per_gpu:5389.6057983537985 mfu:25.423690112406252 +Step 12512 | loss:1.0884732007980347 lr:7.364759881151038e-05 tokens_per_second_per_gpu:4394.952330081351 mfu:20.731739997183965 +Step 12513 | loss:1.0883853435516357 lr:7.364377812874603e-05 tokens_per_second_per_gpu:5044.290564883821 mfu:23.794779239274224 +Step 12514 | loss:1.0997675657272339 lr:7.363995726815383e-05 tokens_per_second_per_gpu:4033.8273717949655 mfu:19.02825195467574 +Step 12515 | loss:1.0744104385375977 lr:7.363613622976247e-05 tokens_per_second_per_gpu:4754.919781429046 mfu:22.42976788197134 +Step 12516 | loss:1.116005778312683 lr:7.363231501360072e-05 tokens_per_second_per_gpu:3320.2207990215798 mfu:15.662047997067505 +Step 12517 | loss:1.00084388256073 lr:7.362849361969733e-05 tokens_per_second_per_gpu:4335.984473441638 mfu:20.45357855646064 +Step 12518 | loss:1.0407285690307617 lr:7.3624672048081e-05 tokens_per_second_per_gpu:4226.454590398246 mfu:19.9369074104199 +Step 12519 | loss:1.0222315788269043 lr:7.362085029878051e-05 tokens_per_second_per_gpu:5318.895980658487 mfu:25.09013983799853 +Step 12520 | loss:1.046775221824646 lr:7.36170283718246e-05 tokens_per_second_per_gpu:4711.098816557615 mfu:22.223056914045227 +Step 12521 | loss:1.1483160257339478 lr:7.361320626724199e-05 tokens_per_second_per_gpu:4359.784579737561 mfu:20.565847718575373 +Step 12522 | loss:1.076431393623352 lr:7.360938398506145e-05 tokens_per_second_per_gpu:5317.1958144218315 mfu:25.08211986378196 +Step 12523 | loss:0.99329674243927 lr:7.36055615253117e-05 tokens_per_second_per_gpu:4700.801463748482 mfu:22.174482543934992 +Step 12524 | loss:1.0036814212799072 lr:7.360173888802154e-05 tokens_per_second_per_gpu:4032.139436394839 mfu:19.020289675401866 +Step 12525 | loss:0.9421168565750122 lr:7.359791607321969e-05 tokens_per_second_per_gpu:4328.651372504053 mfu:20.418987068182023 +Step 12526 | loss:1.0084134340286255 lr:7.359409308093487e-05 tokens_per_second_per_gpu:5177.506224121773 mfu:24.42318023283484 +Step 12527 | loss:1.0510870218276978 lr:7.359026991119591e-05 tokens_per_second_per_gpu:5199.826348171207 mfu:24.528468066181876 +Step 12528 | loss:1.1079484224319458 lr:7.358644656403149e-05 tokens_per_second_per_gpu:5309.575511635459 mfu:25.046173595380917 +Step 12529 | loss:1.236093282699585 lr:7.358262303947039e-05 tokens_per_second_per_gpu:4339.077900021334 mfu:20.46817077743007 +Step 12530 | loss:1.1263984441757202 lr:7.35787993375414e-05 tokens_per_second_per_gpu:5397.319527108174 mfu:25.460077087038478 +Step 12531 | loss:1.1603187322616577 lr:7.357497545827324e-05 tokens_per_second_per_gpu:4642.890436807346 mfu:21.9013063492136 +Step 12532 | loss:0.9952216148376465 lr:7.357115140169467e-05 tokens_per_second_per_gpu:5042.40639948666 mfu:23.78589130962405 +Step 12533 | loss:1.0206526517868042 lr:7.356732716783446e-05 tokens_per_second_per_gpu:5347.368852524588 mfu:25.22445123256381 +Step 12534 | loss:0.9883928894996643 lr:7.356350275672139e-05 tokens_per_second_per_gpu:5365.524975336742 mfu:25.310096761622724 +Step 12535 | loss:1.0946166515350342 lr:7.35596781683842e-05 tokens_per_second_per_gpu:4372.45382892338 mfu:20.62561072858169 +Step 12536 | loss:1.029503583908081 lr:7.355585340285166e-05 tokens_per_second_per_gpu:4668.195036952443 mfu:22.020672465508 +Step 12537 | loss:1.153618574142456 lr:7.355202846015257e-05 tokens_per_second_per_gpu:4128.414143280136 mfu:19.47443389393862 +Step 12538 | loss:1.1061177253723145 lr:7.354820334031565e-05 tokens_per_second_per_gpu:5381.7042786859265 mfu:25.38641729599463 +Step 12539 | loss:1.1766341924667358 lr:7.354437804336968e-05 tokens_per_second_per_gpu:4315.568839521025 mfu:20.357274527990782 +Step 12540 | loss:1.1451711654663086 lr:7.354055256934344e-05 tokens_per_second_per_gpu:5076.988211226121 mfu:23.949019615864586 +Step 12541 | loss:1.1762362718582153 lr:7.35367269182657e-05 tokens_per_second_per_gpu:5341.697436713634 mfu:25.197698196540223 +Step 12542 | loss:1.2412054538726807 lr:7.353290109016524e-05 tokens_per_second_per_gpu:3767.001567389993 mfu:17.769589110120844 +Step 12543 | loss:1.1176002025604248 lr:7.352907508507083e-05 tokens_per_second_per_gpu:5347.8905059099825 mfu:25.226911960585962 +Step 12544 | loss:0.9966076612472534 lr:7.352524890301124e-05 tokens_per_second_per_gpu:4644.906298276769 mfu:21.910815511706296 +Step 12545 | loss:0.9108753204345703 lr:7.352142254401524e-05 tokens_per_second_per_gpu:4006.3507539247926 mfu:18.898640060187713 +Step 12546 | loss:1.005973219871521 lr:7.351759600811163e-05 tokens_per_second_per_gpu:5350.383688485245 mfu:25.23867272817429 +Step 12547 | loss:1.1550545692443848 lr:7.351376929532918e-05 tokens_per_second_per_gpu:5361.649494295753 mfu:25.29181545632713 +Step 12548 | loss:1.150171160697937 lr:7.350994240569668e-05 tokens_per_second_per_gpu:4510.161955288622 mfu:21.27520345607567 +Step 12549 | loss:1.1386737823486328 lr:7.35061153392429e-05 tokens_per_second_per_gpu:4250.028422955518 mfu:20.048109200703134 +Step 12550 | loss:1.1452584266662598 lr:7.350228809599662e-05 tokens_per_second_per_gpu:3976.359075929621 mfu:18.757164197976568 +Step 12551 | loss:1.1042159795761108 lr:7.349846067598662e-05 tokens_per_second_per_gpu:5358.429827291841 mfu:25.27662773773784 +Step 12552 | loss:0.9436778426170349 lr:7.349463307924172e-05 tokens_per_second_per_gpu:5330.762976550111 mfu:25.146118482337602 +Step 12553 | loss:1.166211485862732 lr:7.349080530579068e-05 tokens_per_second_per_gpu:5346.73095555635 mfu:25.221442163728184 +Step 12554 | loss:0.9739713072776794 lr:7.34869773556623e-05 tokens_per_second_per_gpu:4315.434117803321 mfu:20.356639022662474 +Step 12555 | loss:1.0008593797683716 lr:7.348314922888537e-05 tokens_per_second_per_gpu:5317.628117314518 mfu:25.08415911028524 +Step 12556 | loss:0.8557114005088806 lr:7.347932092548866e-05 tokens_per_second_per_gpu:5032.5861841593 mfu:23.739567678423384 +Step 12557 | loss:0.9550932049751282 lr:7.3475492445501e-05 tokens_per_second_per_gpu:5360.927916690969 mfu:25.288411651651526 +Step 12558 | loss:1.0629500150680542 lr:7.347166378895115e-05 tokens_per_second_per_gpu:3622.052241036299 mfu:17.085838406805223 +Step 12559 | loss:0.9349461793899536 lr:7.346783495586793e-05 tokens_per_second_per_gpu:4141.842731800986 mfu:19.537778837144927 +Step 12560 | loss:1.0617454051971436 lr:7.346400594628013e-05 tokens_per_second_per_gpu:5359.767353780864 mfu:25.282937078392113 +Step 12561 | loss:0.9004453420639038 lr:7.346017676021655e-05 tokens_per_second_per_gpu:4464.590516394679 mfu:21.0602352035237 +Step 12562 | loss:0.986778974533081 lr:7.345634739770598e-05 tokens_per_second_per_gpu:5345.853433303082 mfu:25.217302741538788 +Step 12563 | loss:1.021661400794983 lr:7.345251785877722e-05 tokens_per_second_per_gpu:5102.7019869742035 mfu:24.07031588329083 +Step 12564 | loss:1.1088086366653442 lr:7.344868814345909e-05 tokens_per_second_per_gpu:4169.884805645518 mfu:19.67005808394074 +Step 12565 | loss:1.1340786218643188 lr:7.344485825178039e-05 tokens_per_second_per_gpu:3783.434618409451 mfu:17.847106615547247 +Step 12566 | loss:0.9746537208557129 lr:7.344102818376991e-05 tokens_per_second_per_gpu:4591.193875372007 mfu:21.657444848580198 +Step 12567 | loss:1.0019092559814453 lr:7.343719793945647e-05 tokens_per_second_per_gpu:3156.494574736023 mfu:14.889723462538148 +Step 12568 | loss:1.0531293153762817 lr:7.34333675188689e-05 tokens_per_second_per_gpu:5339.734937345873 mfu:25.188440752185574 +Step 12569 | loss:1.082992672920227 lr:7.342953692203593e-05 tokens_per_second_per_gpu:5124.421581703081 mfu:24.17277091737193 +Step 12570 | loss:1.2135900259017944 lr:7.342570614898645e-05 tokens_per_second_per_gpu:5345.830502246143 mfu:25.217194571830763 +Step 12571 | loss:1.0675126314163208 lr:7.342187519974925e-05 tokens_per_second_per_gpu:5342.886243790513 mfu:25.203306002353262 +Step 12572 | loss:1.0098202228546143 lr:7.341804407435313e-05 tokens_per_second_per_gpu:4394.147693371353 mfu:20.727944388537725 +Step 12573 | loss:1.0539861917495728 lr:7.341421277282692e-05 tokens_per_second_per_gpu:3434.1847359416133 mfu:16.19963533177197 +Step 12574 | loss:0.9782347679138184 lr:7.341038129519942e-05 tokens_per_second_per_gpu:5397.810416277612 mfu:25.462392695005043 +Step 12575 | loss:1.1495319604873657 lr:7.340654964149946e-05 tokens_per_second_per_gpu:3846.7348296100804 mfu:18.14570451191032 +Step 12576 | loss:0.8922840356826782 lr:7.340271781175585e-05 tokens_per_second_per_gpu:5360.194630440559 mfu:25.28495261529681 +Step 12577 | loss:0.9536164402961731 lr:7.339888580599742e-05 tokens_per_second_per_gpu:5323.066793404141 mfu:25.109814273333814 +Step 12578 | loss:1.0145009756088257 lr:7.339505362425299e-05 tokens_per_second_per_gpu:5359.419509673893 mfu:25.281296238391207 +Step 12579 | loss:1.0203049182891846 lr:7.339122126655136e-05 tokens_per_second_per_gpu:3642.923095467033 mfu:17.184289788089952 +Step 12580 | loss:1.1556535959243774 lr:7.338738873292138e-05 tokens_per_second_per_gpu:5396.609699194282 mfu:25.45672870765952 +Step 12581 | loss:0.9733050465583801 lr:7.338355602339184e-05 tokens_per_second_per_gpu:4372.614319158194 mfu:20.62636778840177 +Step 12582 | loss:0.9979856610298157 lr:7.337972313799163e-05 tokens_per_second_per_gpu:5371.4523473349955 mfu:25.338057186652428 +Step 12583 | loss:1.0690853595733643 lr:7.337589007674951e-05 tokens_per_second_per_gpu:5311.595402641542 mfu:25.05570176588555 +Step 12584 | loss:1.1002452373504639 lr:7.337205683969436e-05 tokens_per_second_per_gpu:5119.235372095537 mfu:24.148306681794544 +Step 12585 | loss:1.1326022148132324 lr:7.336822342685497e-05 tokens_per_second_per_gpu:4187.405748267021 mfu:19.752707359667436 +Step 12586 | loss:1.1063786745071411 lr:7.336438983826021e-05 tokens_per_second_per_gpu:5090.241326109677 mfu:24.01153681998489 +Step 12587 | loss:1.1489969491958618 lr:7.336055607393885e-05 tokens_per_second_per_gpu:5327.173545517773 mfu:25.12918652373798 +Step 12588 | loss:0.932788610458374 lr:7.335672213391982e-05 tokens_per_second_per_gpu:5319.108956540807 mfu:25.09114448157287 +Step 12589 | loss:1.1308904886245728 lr:7.335288801823188e-05 tokens_per_second_per_gpu:5024.094093917547 mfu:23.699509039853034 +Step 12590 | loss:1.1260367631912231 lr:7.334905372690389e-05 tokens_per_second_per_gpu:4460.629084465261 mfu:21.041548453222653 +Step 12591 | loss:1.0001296997070312 lr:7.334521925996469e-05 tokens_per_second_per_gpu:5353.24207159023 mfu:25.252156208970945 +Step 12592 | loss:1.1472829580307007 lr:7.334138461744309e-05 tokens_per_second_per_gpu:5344.440085746162 mfu:25.210635739970837 +Step 12593 | loss:1.1121145486831665 lr:7.3337549799368e-05 tokens_per_second_per_gpu:3994.285741314648 mfu:18.841727337202244 +Step 12594 | loss:1.1086190938949585 lr:7.333371480576819e-05 tokens_per_second_per_gpu:4061.4464975949913 mfu:19.15853608338324 +Step 12595 | loss:1.1520130634307861 lr:7.332987963667255e-05 tokens_per_second_per_gpu:5355.488418251049 mfu:25.26275260943629 +Step 12596 | loss:0.9804570078849792 lr:7.33260442921099e-05 tokens_per_second_per_gpu:5103.416437467621 mfu:24.073686068166104 +Step 12597 | loss:1.0985758304595947 lr:7.33222087721091e-05 tokens_per_second_per_gpu:5132.023230808438 mfu:24.208629193176975 +Step 12598 | loss:1.0363068580627441 lr:7.331837307669899e-05 tokens_per_second_per_gpu:5277.571195918069 mfu:24.895203777642656 +Step 12599 | loss:1.0100839138031006 lr:7.331453720590842e-05 tokens_per_second_per_gpu:3622.4042945757296 mfu:17.087499103417247 +Step 12600 | loss:1.0334826707839966 lr:7.331070115976625e-05 tokens_per_second_per_gpu:3579.0042368433474 mfu:16.88277362628019 +Step 12601 | loss:1.0985643863677979 lr:7.33068649383013e-05 tokens_per_second_per_gpu:4223.128454307882 mfu:19.921217458984316 +Step 12602 | loss:1.121985673904419 lr:7.330302854154246e-05 tokens_per_second_per_gpu:4651.209803970165 mfu:21.94055022355102 +Step 12603 | loss:1.1722296476364136 lr:7.329919196951857e-05 tokens_per_second_per_gpu:4244.717744603259 mfu:20.023057824820366 +Step 12604 | loss:1.0332130193710327 lr:7.329535522225849e-05 tokens_per_second_per_gpu:5319.474567941783 mfu:25.09286913293101 +Step 12605 | loss:1.1026256084442139 lr:7.329151829979105e-05 tokens_per_second_per_gpu:3882.4815858214733 mfu:18.314328059984984 +Step 12606 | loss:1.0489057302474976 lr:7.328768120214514e-05 tokens_per_second_per_gpu:5341.775852894024 mfu:25.198068099042302 +Step 12607 | loss:1.3499846458435059 lr:7.328384392934962e-05 tokens_per_second_per_gpu:3777.793496330902 mfu:17.82049648022806 +Step 12608 | loss:1.0163280963897705 lr:7.328000648143331e-05 tokens_per_second_per_gpu:4463.965907593051 mfu:21.057288817237172 +Step 12609 | loss:1.0192102193832397 lr:7.327616885842513e-05 tokens_per_second_per_gpu:3729.814686670153 mfu:17.59417224902926 +Step 12610 | loss:0.9652742743492126 lr:7.32723310603539e-05 tokens_per_second_per_gpu:4515.866115379963 mfu:21.30211095245659 +Step 12611 | loss:1.042991042137146 lr:7.326849308724851e-05 tokens_per_second_per_gpu:3847.9808999332095 mfu:18.15158243822604 +Step 12612 | loss:1.0085902214050293 lr:7.32646549391378e-05 tokens_per_second_per_gpu:5347.458446118947 mfu:25.224873860834645 +Step 12613 | loss:1.0435917377471924 lr:7.326081661605066e-05 tokens_per_second_per_gpu:3938.3764984558616 mfu:18.577993899536878 +Step 12614 | loss:1.0368274450302124 lr:7.325697811801593e-05 tokens_per_second_per_gpu:5348.491562369658 mfu:25.229747246457055 +Step 12615 | loss:0.9576606750488281 lr:7.325313944506254e-05 tokens_per_second_per_gpu:5339.576458953444 mfu:25.187693182569284 +Step 12616 | loss:1.0640000104904175 lr:7.324930059721929e-05 tokens_per_second_per_gpu:3680.7679716781554 mfu:17.362810526179526 +Step 12617 | loss:1.0681021213531494 lr:7.324546157451508e-05 tokens_per_second_per_gpu:4612.570366551888 mfu:21.758281404681178 +Step 12618 | loss:1.086496353149414 lr:7.324162237697881e-05 tokens_per_second_per_gpu:4694.201319727675 mfu:22.143348538488155 +Step 12619 | loss:0.9950044751167297 lr:7.323778300463932e-05 tokens_per_second_per_gpu:4996.4062384859835 mfu:23.568900701747086 +Step 12620 | loss:1.0153419971466064 lr:7.32339434575255e-05 tokens_per_second_per_gpu:4640.982332447682 mfu:21.89230549539289 +Step 12621 | loss:1.1440767049789429 lr:7.323010373566623e-05 tokens_per_second_per_gpu:5322.66836513582 mfu:25.107934819213273 +Step 12622 | loss:1.194443941116333 lr:7.322626383909038e-05 tokens_per_second_per_gpu:4115.836000402099 mfu:19.415100647930448 +Step 12623 | loss:1.0963060855865479 lr:7.322242376782683e-05 tokens_per_second_per_gpu:5344.796453690439 mfu:25.212316788366632 +Step 12624 | loss:1.0952588319778442 lr:7.321858352190448e-05 tokens_per_second_per_gpu:5199.957687571696 mfu:24.529087616542796 +Step 12625 | loss:1.0982664823532104 lr:7.32147431013522e-05 tokens_per_second_per_gpu:5339.802609932841 mfu:25.188759975324437 +Step 12626 | loss:0.9989909529685974 lr:7.321090250619887e-05 tokens_per_second_per_gpu:4540.032470410754 mfu:21.416107772342173 +Step 12627 | loss:1.078667402267456 lr:7.320706173647337e-05 tokens_per_second_per_gpu:4057.10021848985 mfu:19.138033943292605 +Step 12628 | loss:1.0110623836517334 lr:7.320322079220463e-05 tokens_per_second_per_gpu:5343.885904008043 mfu:25.208021570158976 +Step 12629 | loss:0.9700129628181458 lr:7.319937967342149e-05 tokens_per_second_per_gpu:5328.4512189210145 mfu:25.13521352717463 +Step 12630 | loss:1.1230558156967163 lr:7.319553838015284e-05 tokens_per_second_per_gpu:5346.164612377788 mfu:25.218770626326798 +Step 12631 | loss:0.9524018168449402 lr:7.31916969124276e-05 tokens_per_second_per_gpu:3471.6332165284375 mfu:16.376286204069658 +Step 12632 | loss:1.1021610498428345 lr:7.318785527027464e-05 tokens_per_second_per_gpu:3725.6778606805556 mfu:17.574658135021032 +Step 12633 | loss:1.088955044746399 lr:7.318401345372287e-05 tokens_per_second_per_gpu:4775.915739100834 mfu:22.528809396588933 +Step 12634 | loss:1.0253028869628906 lr:7.318017146280118e-05 tokens_per_second_per_gpu:3957.5057669361922 mfu:18.66822992274835 +Step 12635 | loss:0.9836981296539307 lr:7.317632929753843e-05 tokens_per_second_per_gpu:5332.625111831358 mfu:25.154902492172287 +Step 12636 | loss:1.1960562467575073 lr:7.31724869579636e-05 tokens_per_second_per_gpu:5334.871175691401 mfu:25.16549755861812 +Step 12637 | loss:1.084837794303894 lr:7.31686444441055e-05 tokens_per_second_per_gpu:4470.290707492807 mfu:21.08712397748646 +Step 12638 | loss:1.0454388856887817 lr:7.316480175599309e-05 tokens_per_second_per_gpu:5326.014329229041 mfu:25.123718302721343 +Step 12639 | loss:0.8869019150733948 lr:7.316095889365524e-05 tokens_per_second_per_gpu:5237.261854788266 mfu:24.70505773805059 +Step 12640 | loss:1.0675859451293945 lr:7.315711585712087e-05 tokens_per_second_per_gpu:4571.656509590125 mfu:21.56528376076873 +Step 12641 | loss:1.2086557149887085 lr:7.315327264641887e-05 tokens_per_second_per_gpu:5091.928198740769 mfu:24.0194940859959 +Step 12642 | loss:1.1039209365844727 lr:7.314942926157815e-05 tokens_per_second_per_gpu:3701.45522861858 mfu:17.4603958467775 +Step 12643 | loss:1.0930196046829224 lr:7.314558570262764e-05 tokens_per_second_per_gpu:5162.119986455063 mfu:24.350600724598127 +Step 12644 | loss:1.1060384511947632 lr:7.31417419695962e-05 tokens_per_second_per_gpu:5294.868169565433 mfu:24.97679655350529 +Step 12645 | loss:1.0329418182373047 lr:7.313789806251278e-05 tokens_per_second_per_gpu:5323.107251894056 mfu:25.11000512293368 +Step 12646 | loss:0.9943048357963562 lr:7.313405398140627e-05 tokens_per_second_per_gpu:5326.949464661169 mfu:25.128129496105245 +Step 12647 | loss:1.0327740907669067 lr:7.31302097263056e-05 tokens_per_second_per_gpu:4399.076796122949 mfu:20.751195807208656 +Step 12648 | loss:0.9188512563705444 lr:7.312636529723965e-05 tokens_per_second_per_gpu:4299.50672795615 mfu:20.281506807260445 +Step 12649 | loss:1.0170198678970337 lr:7.312252069423737e-05 tokens_per_second_per_gpu:4505.441849983175 mfu:21.252937914016865 +Step 12650 | loss:1.1586410999298096 lr:7.311867591732766e-05 tokens_per_second_per_gpu:4730.780488458188 mfu:22.31589871843973 +Step 12651 | loss:1.1152873039245605 lr:7.311483096653944e-05 tokens_per_second_per_gpu:5333.44086498396 mfu:25.15875053897495 +Step 12652 | loss:1.1323002576828003 lr:7.311098584190163e-05 tokens_per_second_per_gpu:4581.47129893028 mfu:21.611581796661973 +Step 12653 | loss:1.2255722284317017 lr:7.310714054344312e-05 tokens_per_second_per_gpu:3895.1100618684686 mfu:18.37389873613885 +Step 12654 | loss:1.1328413486480713 lr:7.310329507119288e-05 tokens_per_second_per_gpu:3944.8095673164626 mfu:18.60833978294706 +Step 12655 | loss:1.0865236520767212 lr:7.309944942517981e-05 tokens_per_second_per_gpu:3135.6542504295503 mfu:14.791416097058535 +Step 12656 | loss:1.1234790086746216 lr:7.309560360543284e-05 tokens_per_second_per_gpu:5125.774024462231 mfu:24.17915062061737 +Step 12657 | loss:1.0059616565704346 lr:7.309175761198088e-05 tokens_per_second_per_gpu:5089.958763625164 mfu:24.010203924535308 +Step 12658 | loss:1.0631605386734009 lr:7.308791144485286e-05 tokens_per_second_per_gpu:4639.429586380872 mfu:21.88498092726995 +Step 12659 | loss:0.9869714975357056 lr:7.308406510407772e-05 tokens_per_second_per_gpu:4833.702069888542 mfu:22.80139737827925 +Step 12660 | loss:1.057774305343628 lr:7.308021858968437e-05 tokens_per_second_per_gpu:5329.507999377247 mfu:25.14019854089202 +Step 12661 | loss:1.029367446899414 lr:7.307637190170175e-05 tokens_per_second_per_gpu:4161.056320714044 mfu:19.628412614224125 +Step 12662 | loss:1.0178674459457397 lr:7.307252504015882e-05 tokens_per_second_per_gpu:4327.643046774456 mfu:20.4142306236774 +Step 12663 | loss:0.9117783308029175 lr:7.306867800508446e-05 tokens_per_second_per_gpu:4921.6412164794865 mfu:23.216221336714305 +Step 12664 | loss:0.9803830981254578 lr:7.306483079650762e-05 tokens_per_second_per_gpu:3872.8405451602353 mfu:18.268849626254767 +Step 12665 | loss:1.072069525718689 lr:7.306098341445727e-05 tokens_per_second_per_gpu:3491.0701589950622 mfu:16.467973577969076 +Step 12666 | loss:0.9722598791122437 lr:7.30571358589623e-05 tokens_per_second_per_gpu:5116.860860926036 mfu:24.137105707474625 +Step 12667 | loss:0.9590818881988525 lr:7.305328813005168e-05 tokens_per_second_per_gpu:4350.055809964833 mfu:20.51995545165813 +Step 12668 | loss:1.0391830205917358 lr:7.304944022775435e-05 tokens_per_second_per_gpu:4441.478243648706 mfu:20.951210669620586 +Step 12669 | loss:1.1654142141342163 lr:7.304559215209924e-05 tokens_per_second_per_gpu:4407.898799322072 mfu:20.79281070148787 +Step 12670 | loss:1.0691066980361938 lr:7.304174390311526e-05 tokens_per_second_per_gpu:5297.756659332898 mfu:24.9904220525656 +Step 12671 | loss:1.0281189680099487 lr:7.30378954808314e-05 tokens_per_second_per_gpu:4802.2724688791195 mfu:22.653138587874917 +Step 12672 | loss:0.9906653761863708 lr:7.303404688527659e-05 tokens_per_second_per_gpu:3957.394331925785 mfu:18.667704264791485 +Step 12673 | loss:0.8839671015739441 lr:7.303019811647978e-05 tokens_per_second_per_gpu:5327.6511959301315 mfu:25.13143968222864 +Step 12674 | loss:1.145883321762085 lr:7.30263491744699e-05 tokens_per_second_per_gpu:3840.542731073219 mfu:18.116495326630993 +Step 12675 | loss:0.8835662603378296 lr:7.30225000592759e-05 tokens_per_second_per_gpu:4211.2048302955145 mfu:19.864971690137946 +Step 12676 | loss:1.1465343236923218 lr:7.301865077092677e-05 tokens_per_second_per_gpu:5334.457491169691 mfu:25.163546138147357 +Step 12677 | loss:1.0475834608078003 lr:7.30148013094514e-05 tokens_per_second_per_gpu:4912.185451745738 mfu:23.171616881145386 +Step 12678 | loss:1.0859280824661255 lr:7.301095167487879e-05 tokens_per_second_per_gpu:5351.014903602735 mfu:25.24165027757989 +Step 12679 | loss:1.0538074970245361 lr:7.300710186723787e-05 tokens_per_second_per_gpu:5329.584984366966 mfu:25.140561692223518 +Step 12680 | loss:1.186829924583435 lr:7.300325188655761e-05 tokens_per_second_per_gpu:2500.9475114317715 mfu:11.79739611706961 +Step 12681 | loss:1.1501816511154175 lr:7.299940173286695e-05 tokens_per_second_per_gpu:3732.5957144009544 mfu:17.60729083132771 +Step 12682 | loss:1.1800769567489624 lr:7.299555140619485e-05 tokens_per_second_per_gpu:5230.303210449162 mfu:24.672232625435893 +Step 12683 | loss:1.0359411239624023 lr:7.29917009065703e-05 tokens_per_second_per_gpu:5321.3655165958935 mfu:25.101789060361558 +Step 12684 | loss:1.110634446144104 lr:7.298785023402221e-05 tokens_per_second_per_gpu:2735.2986882329537 mfu:12.90287060247461 +Step 12685 | loss:0.8837476968765259 lr:7.298399938857956e-05 tokens_per_second_per_gpu:4598.383185807203 mfu:21.69135805252586 +Step 12686 | loss:0.9934972524642944 lr:7.298014837027133e-05 tokens_per_second_per_gpu:5349.597516350941 mfu:25.234964220081977 +Step 12687 | loss:1.1830573081970215 lr:7.297629717912647e-05 tokens_per_second_per_gpu:3875.317596167241 mfu:18.280534298483293 +Step 12688 | loss:1.1422364711761475 lr:7.297244581517395e-05 tokens_per_second_per_gpu:4469.203760352587 mfu:21.081996662373736 +Step 12689 | loss:1.0117679834365845 lr:7.296859427844271e-05 tokens_per_second_per_gpu:4544.920298863882 mfu:21.439164493105103 +Step 12690 | loss:0.9186370968818665 lr:7.296474256896178e-05 tokens_per_second_per_gpu:5330.101718560626 mfu:25.142999215578957 +Step 12691 | loss:1.2322280406951904 lr:7.296089068676007e-05 tokens_per_second_per_gpu:4632.713988091426 mfu:21.853302304339408 +Step 12692 | loss:1.1125999689102173 lr:7.295703863186657e-05 tokens_per_second_per_gpu:5355.043010362357 mfu:25.260651544430907 +Step 12693 | loss:1.0049374103546143 lr:7.295318640431025e-05 tokens_per_second_per_gpu:5308.97177185161 mfu:25.04332565181187 +Step 12694 | loss:0.9926654696464539 lr:7.294933400412009e-05 tokens_per_second_per_gpu:4910.166326101688 mfu:23.162092320984048 +Step 12695 | loss:1.0617876052856445 lr:7.294548143132506e-05 tokens_per_second_per_gpu:4559.6587507205795 mfu:21.508688285152335 +Step 12696 | loss:0.996286153793335 lr:7.294162868595414e-05 tokens_per_second_per_gpu:4015.646372407586 mfu:18.942489078567107 +Step 12697 | loss:1.1096678972244263 lr:7.293777576803631e-05 tokens_per_second_per_gpu:5339.194705705178 mfu:25.1858923873634 +Step 12698 | loss:1.0864750146865845 lr:7.293392267760053e-05 tokens_per_second_per_gpu:3689.3960505248683 mfu:17.403510646201518 +Step 12699 | loss:1.1965210437774658 lr:7.293006941467581e-05 tokens_per_second_per_gpu:4861.078125346605 mfu:22.930534902711635 +Step 12700 | loss:0.9928515553474426 lr:7.292621597929108e-05 tokens_per_second_per_gpu:5323.961446631008 mfu:25.114034505248526 +Step 12701 | loss:1.1404573917388916 lr:7.292236237147539e-05 tokens_per_second_per_gpu:3638.121297321597 mfu:17.16163888696669 +Step 12702 | loss:1.164101481437683 lr:7.291850859125766e-05 tokens_per_second_per_gpu:5341.723843891118 mfu:25.197822763701947 +Step 12703 | loss:1.0078904628753662 lr:7.291465463866691e-05 tokens_per_second_per_gpu:5135.1996777814575 mfu:24.223613035506446 +Step 12704 | loss:1.145652413368225 lr:7.291080051373213e-05 tokens_per_second_per_gpu:5041.701102232975 mfu:23.782564302142298 +Step 12705 | loss:1.1212157011032104 lr:7.290694621648228e-05 tokens_per_second_per_gpu:4090.409377282416 mfu:19.295158928445655 +Step 12706 | loss:1.0622934103012085 lr:7.290309174694637e-05 tokens_per_second_per_gpu:4546.639420626731 mfu:21.447273883772883 +Step 12707 | loss:1.0783820152282715 lr:7.289923710515339e-05 tokens_per_second_per_gpu:4200.875237982392 mfu:19.816245240786763 +Step 12708 | loss:1.0030078887939453 lr:7.289538229113232e-05 tokens_per_second_per_gpu:2959.40654318666 mfu:13.960025591034148 +Step 12709 | loss:1.0746036767959595 lr:7.289152730491215e-05 tokens_per_second_per_gpu:4924.771360285954 mfu:23.23098675910702 +Step 12710 | loss:1.1156954765319824 lr:7.28876721465219e-05 tokens_per_second_per_gpu:4745.1520628752405 mfu:22.383691887008403 +Step 12711 | loss:1.1105620861053467 lr:7.288381681599052e-05 tokens_per_second_per_gpu:3700.8028622693805 mfu:17.457318523403153 +Step 12712 | loss:1.1353365182876587 lr:7.287996131334707e-05 tokens_per_second_per_gpu:4375.1390036427965 mfu:20.63827715587123 +Step 12713 | loss:0.7905287742614746 lr:7.287610563862049e-05 tokens_per_second_per_gpu:5004.294169527692 mfu:23.60610941829064 +Step 12714 | loss:1.0057590007781982 lr:7.28722497918398e-05 tokens_per_second_per_gpu:4776.4513168182175 mfu:22.531335808060817 +Step 12715 | loss:0.9123148918151855 lr:7.2868393773034e-05 tokens_per_second_per_gpu:4222.584901996632 mfu:19.918653429992563 +Step 12716 | loss:1.0322585105895996 lr:7.28645375822321e-05 tokens_per_second_per_gpu:4770.586243257665 mfu:22.503669255390953 +Step 12717 | loss:0.9151368737220764 lr:7.28606812194631e-05 tokens_per_second_per_gpu:5353.72707715759 mfu:25.25444406298267 +Step 12718 | loss:1.0431963205337524 lr:7.285682468475599e-05 tokens_per_second_per_gpu:4704.6929763194385 mfu:22.192839472692704 +Step 12719 | loss:1.039433479309082 lr:7.28529679781398e-05 tokens_per_second_per_gpu:5358.002046999501 mfu:25.274609825112023 +Step 12720 | loss:0.994476318359375 lr:7.28491110996435e-05 tokens_per_second_per_gpu:4866.996232775736 mfu:22.958451625188097 +Step 12721 | loss:0.9850298166275024 lr:7.284525404929612e-05 tokens_per_second_per_gpu:5164.299961395869 mfu:24.36088403833599 +Step 12722 | loss:1.054427981376648 lr:7.284139682712669e-05 tokens_per_second_per_gpu:4944.301914088153 mfu:23.323115713648082 +Step 12723 | loss:1.1774317026138306 lr:7.283753943316419e-05 tokens_per_second_per_gpu:3715.565433686148 mfu:17.526956091530455 +Step 12724 | loss:0.952156662940979 lr:7.283368186743765e-05 tokens_per_second_per_gpu:5319.2926075945725 mfu:25.09201079492758 +Step 12725 | loss:0.8484160900115967 lr:7.282982412997604e-05 tokens_per_second_per_gpu:5291.912613960526 mfu:24.96285469344716 +Step 12726 | loss:1.1142780780792236 lr:7.282596622080845e-05 tokens_per_second_per_gpu:3467.837156458526 mfu:16.358379541045252 +Step 12727 | loss:0.9646527171134949 lr:7.282210813996382e-05 tokens_per_second_per_gpu:4805.613527585785 mfu:22.668898931837653 +Step 12728 | loss:1.0498549938201904 lr:7.281824988747124e-05 tokens_per_second_per_gpu:4332.636965277417 mfu:20.437787789306228 +Step 12729 | loss:1.0915948152542114 lr:7.281439146335966e-05 tokens_per_second_per_gpu:3524.9584816152446 mfu:16.627830577712533 +Step 12730 | loss:1.1029548645019531 lr:7.281053286765815e-05 tokens_per_second_per_gpu:5379.869319706649 mfu:25.37776148141309 +Step 12731 | loss:1.0246602296829224 lr:7.28066741003957e-05 tokens_per_second_per_gpu:5336.046835026343 mfu:25.17104334428921 +Step 12732 | loss:0.9165564179420471 lr:7.280281516160136e-05 tokens_per_second_per_gpu:5137.424992726301 mfu:24.234110225779155 +Step 12733 | loss:1.0416308641433716 lr:7.279895605130413e-05 tokens_per_second_per_gpu:5333.754122754293 mfu:25.1602282293249 +Step 12734 | loss:1.22020423412323 lr:7.279509676953304e-05 tokens_per_second_per_gpu:3718.904372116553 mfu:17.542706433788315 +Step 12735 | loss:1.0850900411605835 lr:7.279123731631711e-05 tokens_per_second_per_gpu:5342.710250809016 mfu:25.20247581343204 +Step 12736 | loss:1.038136601448059 lr:7.278737769168539e-05 tokens_per_second_per_gpu:5308.978491243935 mfu:25.043357348331917 +Step 12737 | loss:1.1000778675079346 lr:7.27835178956669e-05 tokens_per_second_per_gpu:4361.042577997951 mfu:20.571781911006273 +Step 12738 | loss:0.9521230459213257 lr:7.277965792829064e-05 tokens_per_second_per_gpu:5373.288385649413 mfu:25.346718092641975 +Step 12739 | loss:1.1428629159927368 lr:7.277579778958568e-05 tokens_per_second_per_gpu:3417.0508038997204 mfu:16.11881165098011 +Step 12740 | loss:1.0315965414047241 lr:7.277193747958104e-05 tokens_per_second_per_gpu:4293.944228484124 mfu:20.255267548190623 +Step 12741 | loss:1.1235276460647583 lr:7.276807699830575e-05 tokens_per_second_per_gpu:3881.085926389329 mfu:18.307744496319504 +Step 12742 | loss:1.0728905200958252 lr:7.276421634578886e-05 tokens_per_second_per_gpu:5312.691530131271 mfu:25.06087238627324 +Step 12743 | loss:0.9025060534477234 lr:7.276035552205937e-05 tokens_per_second_per_gpu:4606.355118843926 mfu:21.728963020812113 +Step 12744 | loss:1.0522265434265137 lr:7.275649452714637e-05 tokens_per_second_per_gpu:3715.265452476369 mfu:17.525541028981856 +Step 12745 | loss:1.0862058401107788 lr:7.275263336107884e-05 tokens_per_second_per_gpu:3392.930979101408 mfu:16.005034322140006 +Step 12746 | loss:1.1112855672836304 lr:7.274877202388588e-05 tokens_per_second_per_gpu:5374.539010464231 mfu:25.352617503272068 +Step 12747 | loss:1.0715237855911255 lr:7.274491051559647e-05 tokens_per_second_per_gpu:5071.810756152548 mfu:23.924596676917613 +Step 12748 | loss:0.8865442276000977 lr:7.274104883623971e-05 tokens_per_second_per_gpu:5349.58338638595 mfu:25.234897566626294 +Step 12749 | loss:1.0562670230865479 lr:7.273718698584462e-05 tokens_per_second_per_gpu:3368.3345183711626 mfu:15.8890086202865 +Step 12750 | loss:1.008266806602478 lr:7.273332496444023e-05 tokens_per_second_per_gpu:5092.159180417985 mfu:24.020583665976865 +Step 12751 | loss:1.1716644763946533 lr:7.272946277205562e-05 tokens_per_second_per_gpu:3705.6839507422555 mfu:17.480343450529276 +Step 12752 | loss:1.1050477027893066 lr:7.272560040871982e-05 tokens_per_second_per_gpu:4634.987662911294 mfu:21.864027616393653 +Step 12753 | loss:1.1309391260147095 lr:7.272173787446187e-05 tokens_per_second_per_gpu:4662.813784628909 mfu:21.99528818016157 +Step 12754 | loss:1.076797366142273 lr:7.271787516931083e-05 tokens_per_second_per_gpu:5382.900866756045 mfu:25.392061806080353 +Step 12755 | loss:1.028290867805481 lr:7.271401229329576e-05 tokens_per_second_per_gpu:4337.691892392089 mfu:20.461632742965634 +Step 12756 | loss:1.042211651802063 lr:7.271014924644569e-05 tokens_per_second_per_gpu:5362.421396165278 mfu:25.295456649145716 +Step 12757 | loss:1.1600332260131836 lr:7.27062860287897e-05 tokens_per_second_per_gpu:4065.0727787245187 mfu:19.17564187017858 +Step 12758 | loss:1.1508406400680542 lr:7.270242264035684e-05 tokens_per_second_per_gpu:3938.086293978248 mfu:18.57662495550198 +Step 12759 | loss:1.0649869441986084 lr:7.269855908117616e-05 tokens_per_second_per_gpu:3708.3179264624855 mfu:17.492768363404195 +Step 12760 | loss:1.0303261280059814 lr:7.269469535127673e-05 tokens_per_second_per_gpu:4031.7614825709406 mfu:19.018506802728968 +Step 12761 | loss:1.0682951211929321 lr:7.269083145068757e-05 tokens_per_second_per_gpu:4938.504505960228 mfu:23.29576835036883 +Step 12762 | loss:1.0808366537094116 lr:7.268696737943779e-05 tokens_per_second_per_gpu:5023.692380678042 mfu:23.697614089962943 +Step 12763 | loss:0.9917669892311096 lr:7.268310313755645e-05 tokens_per_second_per_gpu:4541.4804944364605 mfu:21.422938348729804 +Step 12764 | loss:1.1149142980575562 lr:7.267923872507256e-05 tokens_per_second_per_gpu:5237.922706118254 mfu:24.70817508652701 +Step 12765 | loss:1.0746392011642456 lr:7.267537414201524e-05 tokens_per_second_per_gpu:5221.777901890875 mfu:24.632017290398817 +Step 12766 | loss:1.1181703805923462 lr:7.267150938841355e-05 tokens_per_second_per_gpu:5338.035596178174 mfu:25.18042466996004 +Step 12767 | loss:1.0417670011520386 lr:7.266764446429652e-05 tokens_per_second_per_gpu:4772.5477518833495 mfu:22.512922026246187 +Step 12768 | loss:0.8741092085838318 lr:7.266377936969326e-05 tokens_per_second_per_gpu:4444.833499487706 mfu:20.96703798387883 +Step 12769 | loss:1.1663227081298828 lr:7.265991410463283e-05 tokens_per_second_per_gpu:4031.0013620692534 mfu:19.014921184632573 +Step 12770 | loss:1.1108906269073486 lr:7.265604866914428e-05 tokens_per_second_per_gpu:3852.921296670185 mfu:18.174887132552954 +Step 12771 | loss:1.0971174240112305 lr:7.265218306325671e-05 tokens_per_second_per_gpu:4662.334333646631 mfu:21.993026528075394 +Step 12772 | loss:1.189656138420105 lr:7.264831728699916e-05 tokens_per_second_per_gpu:4606.848148411767 mfu:21.731288725403143 +Step 12773 | loss:1.1153464317321777 lr:7.264445134040077e-05 tokens_per_second_per_gpu:4351.038090100509 mfu:20.52458903465239 +Step 12774 | loss:0.9770025014877319 lr:7.264058522349053e-05 tokens_per_second_per_gpu:5351.389511484152 mfu:25.243417366871228 +Step 12775 | loss:1.1851211786270142 lr:7.263671893629757e-05 tokens_per_second_per_gpu:4191.936152323942 mfu:19.774078048570573 +Step 12776 | loss:0.9342637658119202 lr:7.263285247885097e-05 tokens_per_second_per_gpu:4255.109859244699 mfu:20.07207920266163 +Step 12777 | loss:1.1006853580474854 lr:7.262898585117978e-05 tokens_per_second_per_gpu:5211.424849958442 mfu:24.58318017802073 +Step 12778 | loss:0.9861356019973755 lr:7.262511905331311e-05 tokens_per_second_per_gpu:5322.792924332369 mfu:25.108522386195386 +Step 12779 | loss:1.0854295492172241 lr:7.262125208528003e-05 tokens_per_second_per_gpu:4745.513229032805 mfu:22.385395569394937 +Step 12780 | loss:1.1918039321899414 lr:7.261738494710963e-05 tokens_per_second_per_gpu:3204.4032870585543 mfu:15.115717032632734 +Step 12781 | loss:1.0145223140716553 lr:7.261351763883097e-05 tokens_per_second_per_gpu:5344.284472714506 mfu:25.209901686750193 +Step 12782 | loss:1.2517476081848145 lr:7.260965016047318e-05 tokens_per_second_per_gpu:4155.046877080947 mfu:19.600065043289995 +Step 12783 | loss:1.1034646034240723 lr:7.260578251206532e-05 tokens_per_second_per_gpu:3937.6484530896682 mfu:18.574559585327194 +Step 12784 | loss:1.1081390380859375 lr:7.260191469363651e-05 tokens_per_second_per_gpu:4424.02919340071 mfu:20.868900522485827 +Step 12785 | loss:1.1561390161514282 lr:7.259804670521578e-05 tokens_per_second_per_gpu:3862.5932474748174 mfu:18.220511374703378 +Step 12786 | loss:1.001232385635376 lr:7.259417854683228e-05 tokens_per_second_per_gpu:5321.205007407852 mfu:25.101031911135873 +Step 12787 | loss:0.987957239151001 lr:7.259031021851507e-05 tokens_per_second_per_gpu:5081.983154326465 mfu:23.972581614694082 +Step 12788 | loss:0.9581739902496338 lr:7.258644172029325e-05 tokens_per_second_per_gpu:5016.477878007969 mfu:23.663582049947127 +Step 12789 | loss:1.169944405555725 lr:7.258257305219594e-05 tokens_per_second_per_gpu:5343.937791166024 mfu:25.20826633073603 +Step 12790 | loss:0.9883098602294922 lr:7.257870421425219e-05 tokens_per_second_per_gpu:4729.569202601906 mfu:22.310184876388966 +Step 12791 | loss:0.933245062828064 lr:7.257483520649114e-05 tokens_per_second_per_gpu:5347.739562257595 mfu:25.22619993362401 +Step 12792 | loss:1.0191223621368408 lr:7.257096602894186e-05 tokens_per_second_per_gpu:5334.428240971765 mfu:25.163408159973145 +Step 12793 | loss:1.018500804901123 lr:7.256709668163348e-05 tokens_per_second_per_gpu:4393.6455912219835 mfu:20.725575887031688 +Step 12794 | loss:1.0518301725387573 lr:7.25632271645951e-05 tokens_per_second_per_gpu:5326.296072097468 mfu:25.125047331902 +Step 12795 | loss:1.1111689805984497 lr:7.255935747785578e-05 tokens_per_second_per_gpu:4362.59930712431 mfu:20.579125267900682 +Step 12796 | loss:1.1260385513305664 lr:7.255548762144466e-05 tokens_per_second_per_gpu:5139.91973815706 mfu:24.24587836562351 +Step 12797 | loss:1.2538539171218872 lr:7.255161759539085e-05 tokens_per_second_per_gpu:4880.18387127801 mfu:23.020659966046466 +Step 12798 | loss:1.01059091091156 lr:7.254774739972343e-05 tokens_per_second_per_gpu:4760.475696418242 mfu:22.455976080912222 +Step 12799 | loss:1.0866903066635132 lr:7.254387703447154e-05 tokens_per_second_per_gpu:4270.685235529709 mfu:20.14555091003124 +Step 12800 | loss:1.114343523979187 lr:7.254000649966427e-05 tokens_per_second_per_gpu:4036.907637556331 mfu:19.042782093819632 +Step 12801 | loss:0.9963944554328918 lr:7.253613579533073e-05 tokens_per_second_per_gpu:3866.334218236082 mfu:18.23815817205969 +Step 12802 | loss:1.2402405738830566 lr:7.253226492150005e-05 tokens_per_second_per_gpu:3725.1611919632032 mfu:17.57222092052851 +Step 12803 | loss:1.0518723726272583 lr:7.252839387820132e-05 tokens_per_second_per_gpu:5004.538210397993 mfu:23.607260600713513 +Step 12804 | loss:0.9398328065872192 lr:7.252452266546367e-05 tokens_per_second_per_gpu:5341.258299346526 mfu:25.195626710655013 +Step 12805 | loss:1.0431941747665405 lr:7.25206512833162e-05 tokens_per_second_per_gpu:4698.3622170031695 mfu:22.16297620085894 +Step 12806 | loss:1.1021358966827393 lr:7.251677973178806e-05 tokens_per_second_per_gpu:4171.080925227609 mfu:19.67570038408851 +Step 12807 | loss:0.9621074199676514 lr:7.251290801090834e-05 tokens_per_second_per_gpu:5337.510998458457 mfu:25.17795005301053 +Step 12808 | loss:1.0517830848693848 lr:7.250903612070614e-05 tokens_per_second_per_gpu:4304.998319030658 mfu:20.307411579320956 +Step 12809 | loss:1.0429480075836182 lr:7.250516406121063e-05 tokens_per_second_per_gpu:4574.185134289498 mfu:21.57721171490353 +Step 12810 | loss:1.0709518194198608 lr:7.250129183245091e-05 tokens_per_second_per_gpu:3997.628086914784 mfu:18.857493751661046 +Step 12811 | loss:1.088710904121399 lr:7.249741943445609e-05 tokens_per_second_per_gpu:3302.1647378402054 mfu:15.576874475793254 +Step 12812 | loss:1.1006736755371094 lr:7.249354686725531e-05 tokens_per_second_per_gpu:4145.958376462702 mfu:19.55719303521563 +Step 12813 | loss:1.004464864730835 lr:7.248967413087771e-05 tokens_per_second_per_gpu:5095.983989189471 mfu:24.03862594153163 +Step 12814 | loss:0.9519121646881104 lr:7.248580122535239e-05 tokens_per_second_per_gpu:4259.723856739219 mfu:20.09384421607243 +Step 12815 | loss:1.0525016784667969 lr:7.248192815070848e-05 tokens_per_second_per_gpu:3924.5891085375206 mfu:18.512956428920774 +Step 12816 | loss:1.1231796741485596 lr:7.247805490697513e-05 tokens_per_second_per_gpu:4138.561309313134 mfu:19.522299807401318 +Step 12817 | loss:1.1365852355957031 lr:7.247418149418147e-05 tokens_per_second_per_gpu:5163.154661949898 mfu:24.35548146544111 +Step 12818 | loss:1.060982584953308 lr:7.247030791235661e-05 tokens_per_second_per_gpu:3844.407142976531 mfu:18.134724416915574 +Step 12819 | loss:1.0786406993865967 lr:7.246643416152969e-05 tokens_per_second_per_gpu:3727.2271197168534 mfu:17.5819662542261 +Step 12820 | loss:1.1609504222869873 lr:7.246256024172988e-05 tokens_per_second_per_gpu:5361.911426210646 mfu:25.29305103385987 +Step 12821 | loss:1.0632565021514893 lr:7.245868615298627e-05 tokens_per_second_per_gpu:4163.581971787417 mfu:19.640326541257735 +Step 12822 | loss:0.9917661547660828 lr:7.245481189532801e-05 tokens_per_second_per_gpu:5325.079861961677 mfu:25.119310261184626 +Step 12823 | loss:0.9882680773735046 lr:7.245093746878427e-05 tokens_per_second_per_gpu:5338.600456122868 mfu:25.183089210694206 +Step 12824 | loss:1.0259846448898315 lr:7.244706287338414e-05 tokens_per_second_per_gpu:3622.0807717494285 mfu:17.085972991046898 +Step 12825 | loss:0.9774003028869629 lr:7.244318810915679e-05 tokens_per_second_per_gpu:4588.377869475347 mfu:21.644161268305506 +Step 12826 | loss:0.9929634928703308 lr:7.243931317613135e-05 tokens_per_second_per_gpu:3716.8261916250494 mfu:17.532903301835738 +Step 12827 | loss:1.0535492897033691 lr:7.243543807433699e-05 tokens_per_second_per_gpu:4724.427220804707 mfu:22.285929270939018 +Step 12828 | loss:0.9639573693275452 lr:7.243156280380283e-05 tokens_per_second_per_gpu:4122.674234821297 mfu:19.44735776640891 +Step 12829 | loss:0.9134345650672913 lr:7.242768736455803e-05 tokens_per_second_per_gpu:5338.433906303329 mfu:25.18230356677911 +Step 12830 | loss:1.0230674743652344 lr:7.242381175663172e-05 tokens_per_second_per_gpu:4153.423225822166 mfu:19.592406003278857 +Step 12831 | loss:1.0064427852630615 lr:7.241993598005306e-05 tokens_per_second_per_gpu:4675.748653606888 mfu:22.056304164046917 +Step 12832 | loss:0.9795631766319275 lr:7.241606003485121e-05 tokens_per_second_per_gpu:4525.781504895342 mfu:21.34888353654053 +Step 12833 | loss:0.9674827456474304 lr:7.24121839210553e-05 tokens_per_second_per_gpu:5330.764561180837 mfu:25.146125957311103 +Step 12834 | loss:1.0015512704849243 lr:7.240830763869452e-05 tokens_per_second_per_gpu:5339.262933386894 mfu:25.186214228978887 +Step 12835 | loss:1.190986156463623 lr:7.240443118779798e-05 tokens_per_second_per_gpu:5356.307579396042 mfu:25.266616732320525 +Step 12836 | loss:1.0591901540756226 lr:7.240055456839485e-05 tokens_per_second_per_gpu:4215.627260103478 mfu:19.88583304608644 +Step 12837 | loss:1.146425485610962 lr:7.239667778051429e-05 tokens_per_second_per_gpu:4537.448077009713 mfu:21.40391674772648 +Step 12838 | loss:1.1637665033340454 lr:7.239280082418548e-05 tokens_per_second_per_gpu:5297.885487762628 mfu:24.99102975824903 +Step 12839 | loss:1.216273546218872 lr:7.238892369943755e-05 tokens_per_second_per_gpu:5091.119566783016 mfu:24.01567963108434 +Step 12840 | loss:0.9882495403289795 lr:7.238504640629965e-05 tokens_per_second_per_gpu:4629.531307388216 mfu:21.83828905644105 +Step 12841 | loss:1.087904453277588 lr:7.238116894480098e-05 tokens_per_second_per_gpu:5301.857349939069 mfu:25.009765709804867 +Step 12842 | loss:1.1056694984436035 lr:7.237729131497067e-05 tokens_per_second_per_gpu:4040.413106726069 mfu:19.059317989988 +Step 12843 | loss:1.0012998580932617 lr:7.237341351683791e-05 tokens_per_second_per_gpu:5274.378706417287 mfu:24.88014425997259 +Step 12844 | loss:1.0475916862487793 lr:7.236953555043184e-05 tokens_per_second_per_gpu:5367.355978894603 mfu:25.318733917806576 +Step 12845 | loss:1.0439410209655762 lr:7.236565741578163e-05 tokens_per_second_per_gpu:5164.167895795574 mfu:24.360261062366682 +Step 12846 | loss:1.0503414869308472 lr:7.236177911291647e-05 tokens_per_second_per_gpu:4066.570525583749 mfu:19.182706997655536 +Step 12847 | loss:1.1336151361465454 lr:7.235790064186552e-05 tokens_per_second_per_gpu:5291.793682527602 mfu:24.96229367358626 +Step 12848 | loss:1.1254994869232178 lr:7.235402200265793e-05 tokens_per_second_per_gpu:4112.326338616577 mfu:19.398544974478096 +Step 12849 | loss:1.125601887702942 lr:7.235014319532292e-05 tokens_per_second_per_gpu:5362.8864897224885 mfu:25.297650574808223 +Step 12850 | loss:1.0263795852661133 lr:7.234626421988961e-05 tokens_per_second_per_gpu:4839.347997922464 mfu:22.828030184109917 +Step 12851 | loss:1.138146162033081 lr:7.234238507638718e-05 tokens_per_second_per_gpu:5247.136773894438 mfu:24.75163941630934 +Step 12852 | loss:1.060685396194458 lr:7.233850576484484e-05 tokens_per_second_per_gpu:5352.27505809057 mfu:25.247594641303536 +Step 12853 | loss:1.1101146936416626 lr:7.233462628529175e-05 tokens_per_second_per_gpu:4769.254240947294 mfu:22.497385973229903 +Step 12854 | loss:1.0522640943527222 lr:7.233074663775708e-05 tokens_per_second_per_gpu:5034.426567425352 mfu:23.748249080291117 +Step 12855 | loss:1.0352247953414917 lr:7.232686682227001e-05 tokens_per_second_per_gpu:5290.9453245385475 mfu:24.958291824207095 +Step 12856 | loss:0.9923484921455383 lr:7.232298683885972e-05 tokens_per_second_per_gpu:5132.699753153045 mfu:24.211820464503102 +Step 12857 | loss:1.159602165222168 lr:7.231910668755542e-05 tokens_per_second_per_gpu:4430.989225224083 mfu:20.901732180100563 +Step 12858 | loss:0.9844362735748291 lr:7.231522636838626e-05 tokens_per_second_per_gpu:5345.34224056244 mfu:25.214891358162703 +Step 12859 | loss:1.1195343732833862 lr:7.231134588138143e-05 tokens_per_second_per_gpu:4044.414864610078 mfu:19.07819496469717 +Step 12860 | loss:1.0597667694091797 lr:7.230746522657013e-05 tokens_per_second_per_gpu:4135.90879142657 mfu:19.509787427964177 +Step 12861 | loss:0.9577576518058777 lr:7.230358440398152e-05 tokens_per_second_per_gpu:4080.412319703016 mfu:19.24800110216972 +Step 12862 | loss:1.0638703107833862 lr:7.229970341364483e-05 tokens_per_second_per_gpu:4997.129390826882 mfu:23.57231193472162 +Step 12863 | loss:0.9135401844978333 lr:7.22958222555892e-05 tokens_per_second_per_gpu:5115.948559071362 mfu:24.13280222397138 +Step 12864 | loss:0.9742019176483154 lr:7.229194092984386e-05 tokens_per_second_per_gpu:4777.6098146872755 mfu:22.536800640169282 +Step 12865 | loss:1.119905948638916 lr:7.2288059436438e-05 tokens_per_second_per_gpu:4155.242360073171 mfu:19.600987169917154 +Step 12866 | loss:1.0989195108413696 lr:7.228417777540078e-05 tokens_per_second_per_gpu:4360.784964728823 mfu:20.57056670526244 +Step 12867 | loss:1.0974929332733154 lr:7.228029594676142e-05 tokens_per_second_per_gpu:4074.5012659026065 mfu:19.220117652863415 +Step 12868 | loss:1.110037922859192 lr:7.227641395054913e-05 tokens_per_second_per_gpu:5368.450042480494 mfu:25.323894802407718 +Step 12869 | loss:1.1351757049560547 lr:7.227253178679306e-05 tokens_per_second_per_gpu:5316.846836521943 mfu:25.080473675486413 +Step 12870 | loss:1.0376086235046387 lr:7.226864945552245e-05 tokens_per_second_per_gpu:5393.0195909603935 mfu:25.43979355458464 +Step 12871 | loss:0.917789101600647 lr:7.226476695676649e-05 tokens_per_second_per_gpu:4254.8079861509095 mfu:20.070655215773364 +Step 12872 | loss:1.0216503143310547 lr:7.226088429055436e-05 tokens_per_second_per_gpu:5317.476226444831 mfu:25.08344261513767 +Step 12873 | loss:1.1122949123382568 lr:7.22570014569153e-05 tokens_per_second_per_gpu:5345.5408253415535 mfu:25.215828116447838 +Step 12874 | loss:0.9728981256484985 lr:7.225311845587848e-05 tokens_per_second_per_gpu:5364.863311317355 mfu:25.30697557955175 +Step 12875 | loss:1.0857847929000854 lr:7.224923528747313e-05 tokens_per_second_per_gpu:4857.948473905813 mfu:22.915771802891825 +Step 12876 | loss:1.0468063354492188 lr:7.224535195172842e-05 tokens_per_second_per_gpu:5311.060177631695 mfu:25.0531770181953 +Step 12877 | loss:0.9161093831062317 lr:7.224146844867358e-05 tokens_per_second_per_gpu:5171.318943961186 mfu:24.393993776658505 +Step 12878 | loss:0.9305408596992493 lr:7.223758477833785e-05 tokens_per_second_per_gpu:4358.812294443128 mfu:20.561261282952557 +Step 12879 | loss:0.9612746834754944 lr:7.223370094075039e-05 tokens_per_second_per_gpu:4709.79523221896 mfu:22.216907684305372 +Step 12880 | loss:0.9310559630393982 lr:7.22298169359404e-05 tokens_per_second_per_gpu:5368.824731270352 mfu:25.325662273357974 +Step 12881 | loss:1.1814014911651611 lr:7.222593276393716e-05 tokens_per_second_per_gpu:3821.071419050774 mfu:18.02464582567272 +Step 12882 | loss:1.1147431135177612 lr:7.222204842476982e-05 tokens_per_second_per_gpu:3939.216078861235 mfu:18.58195434355621 +Step 12883 | loss:1.1438559293746948 lr:7.221816391846763e-05 tokens_per_second_per_gpu:5375.500039332526 mfu:25.35715084041225 +Step 12884 | loss:1.19297456741333 lr:7.221427924505978e-05 tokens_per_second_per_gpu:5053.98416875901 mfu:23.840505622652852 +Step 12885 | loss:1.1566767692565918 lr:7.221039440457552e-05 tokens_per_second_per_gpu:5340.949828284556 mfu:25.19417160002511 +Step 12886 | loss:1.0150513648986816 lr:7.220650939704406e-05 tokens_per_second_per_gpu:4496.924833440787 mfu:21.212761693833553 +Step 12887 | loss:1.1741876602172852 lr:7.220262422249457e-05 tokens_per_second_per_gpu:5345.013254516181 mfu:25.213339474851693 +Step 12888 | loss:0.9624560475349426 lr:7.219873888095633e-05 tokens_per_second_per_gpu:5352.469997505995 mfu:25.24851420378619 +Step 12889 | loss:0.9919891953468323 lr:7.219485337245855e-05 tokens_per_second_per_gpu:5359.2842132401265 mfu:25.280658022029176 +Step 12890 | loss:1.03249990940094 lr:7.219096769703044e-05 tokens_per_second_per_gpu:3712.997171400791 mfu:17.514841160139696 +Step 12891 | loss:1.15040123462677 lr:7.218708185470122e-05 tokens_per_second_per_gpu:3608.8977940198215 mfu:17.023786635848403 +Step 12892 | loss:1.1323119401931763 lr:7.218319584550013e-05 tokens_per_second_per_gpu:5374.741787065361 mfu:25.35357403509685 +Step 12893 | loss:0.954628050327301 lr:7.21793096694564e-05 tokens_per_second_per_gpu:4671.958175311593 mfu:22.038423831205723 +Step 12894 | loss:1.0024864673614502 lr:7.217542332659926e-05 tokens_per_second_per_gpu:3895.234696165511 mfu:18.374486657382807 +Step 12895 | loss:0.9925719499588013 lr:7.217153681695792e-05 tokens_per_second_per_gpu:3936.483744283181 mfu:18.56906545516718 +Step 12896 | loss:1.0455116033554077 lr:7.216765014056163e-05 tokens_per_second_per_gpu:4976.520426392209 mfu:23.475095933231344 +Step 12897 | loss:1.221108078956604 lr:7.216376329743961e-05 tokens_per_second_per_gpu:5347.393397815056 mfu:25.224567016886805 +Step 12898 | loss:0.9875115752220154 lr:7.215987628762109e-05 tokens_per_second_per_gpu:4602.029775654431 mfu:21.708559639007625 +Step 12899 | loss:0.9687064290046692 lr:7.215598911113533e-05 tokens_per_second_per_gpu:5315.6897558561095 mfu:25.075015528567356 +Step 12900 | loss:1.0626437664031982 lr:7.215210176801152e-05 tokens_per_second_per_gpu:3988.1462293959603 mfu:18.812766211973045 +Step 12901 | loss:0.961906373500824 lr:7.214821425827895e-05 tokens_per_second_per_gpu:5205.4662036713935 mfu:24.555072226834827 +Step 12902 | loss:1.0700798034667969 lr:7.214432658196684e-05 tokens_per_second_per_gpu:3547.547249517805 mfu:16.73438564995013 +Step 12903 | loss:0.9902385473251343 lr:7.21404387391044e-05 tokens_per_second_per_gpu:4082.286791770065 mfu:19.2568433067279 +Step 12904 | loss:1.1593066453933716 lr:7.213655072972092e-05 tokens_per_second_per_gpu:4003.371593141152 mfu:18.88458685047404 +Step 12905 | loss:0.9561723470687866 lr:7.21326625538456e-05 tokens_per_second_per_gpu:4156.964430259214 mfu:19.60911046880019 +Step 12906 | loss:1.0147109031677246 lr:7.212877421150771e-05 tokens_per_second_per_gpu:4532.2867246953 mfu:21.37956976824188 +Step 12907 | loss:1.1683480739593506 lr:7.212488570273648e-05 tokens_per_second_per_gpu:4202.0316091652285 mfu:19.821700040953473 +Step 12908 | loss:0.9853211045265198 lr:7.212099702756116e-05 tokens_per_second_per_gpu:4355.066804192781 mfu:20.543593166395137 +Step 12909 | loss:1.1153905391693115 lr:7.211710818601101e-05 tokens_per_second_per_gpu:3682.8311788183746 mfu:17.372543026279374 +Step 12910 | loss:1.1920802593231201 lr:7.211321917811525e-05 tokens_per_second_per_gpu:4833.712466231914 mfu:22.801446419604737 +Step 12911 | loss:1.0501487255096436 lr:7.210933000390317e-05 tokens_per_second_per_gpu:5368.686465938793 mfu:25.325010052198515 +Step 12912 | loss:1.1011323928833008 lr:7.210544066340398e-05 tokens_per_second_per_gpu:4809.149715178183 mfu:22.685579732046815 +Step 12913 | loss:1.0877182483673096 lr:7.210155115664695e-05 tokens_per_second_per_gpu:3902.181237435561 mfu:18.407254677755784 +Step 12914 | loss:1.0676379203796387 lr:7.209766148366136e-05 tokens_per_second_per_gpu:3943.8363052999484 mfu:18.60374873995954 +Step 12915 | loss:1.0533932447433472 lr:7.20937716444764e-05 tokens_per_second_per_gpu:4832.003526732312 mfu:22.79338506040997 +Step 12916 | loss:1.138112187385559 lr:7.208988163912138e-05 tokens_per_second_per_gpu:5386.401776447872 mfu:25.408576194413488 +Step 12917 | loss:1.2944316864013672 lr:7.208599146762554e-05 tokens_per_second_per_gpu:5261.039149775156 mfu:24.817219295328606 +Step 12918 | loss:0.996986985206604 lr:7.208210113001814e-05 tokens_per_second_per_gpu:3515.516569679793 mfu:16.583291468156713 +Step 12919 | loss:1.0947020053863525 lr:7.207821062632844e-05 tokens_per_second_per_gpu:5354.901851696072 mfu:25.259985674917893 +Step 12920 | loss:1.315875768661499 lr:7.207431995658571e-05 tokens_per_second_per_gpu:3846.9874872903297 mfu:18.146896341295903 +Step 12921 | loss:1.1148499250411987 lr:7.207042912081918e-05 tokens_per_second_per_gpu:3861.3896057193047 mfu:18.214833591180167 +Step 12922 | loss:1.0782395601272583 lr:7.206653811905817e-05 tokens_per_second_per_gpu:5040.25662460132 mfu:23.775750454700486 +Step 12923 | loss:1.0457135438919067 lr:7.206264695133189e-05 tokens_per_second_per_gpu:3933.368127874589 mfu:18.554368560988877 +Step 12924 | loss:0.995585024356842 lr:7.205875561766962e-05 tokens_per_second_per_gpu:4304.692787299182 mfu:20.305970334014493 +Step 12925 | loss:1.107498049736023 lr:7.205486411810065e-05 tokens_per_second_per_gpu:4688.331866692237 mfu:22.11566132708804 +Step 12926 | loss:1.085051417350769 lr:7.205097245265423e-05 tokens_per_second_per_gpu:5138.028419671392 mfu:24.2369566936344 +Step 12927 | loss:0.9064997434616089 lr:7.204708062135963e-05 tokens_per_second_per_gpu:4707.490862578261 mfu:22.20603757954419 +Step 12928 | loss:1.0303707122802734 lr:7.204318862424612e-05 tokens_per_second_per_gpu:4323.602099072353 mfu:20.395168783909806 +Step 12929 | loss:1.0042613744735718 lr:7.203929646134299e-05 tokens_per_second_per_gpu:3013.7489627801274 mfu:14.216368056029392 +Step 12930 | loss:1.1266298294067383 lr:7.203540413267948e-05 tokens_per_second_per_gpu:4618.898722670439 mfu:21.78813334022128 +Step 12931 | loss:1.032065510749817 lr:7.203151163828488e-05 tokens_per_second_per_gpu:4403.3259645741555 mfu:20.77123985977589 +Step 12932 | loss:0.9444746375083923 lr:7.20276189781885e-05 tokens_per_second_per_gpu:5355.874561299579 mfu:25.264574112079003 +Step 12933 | loss:1.0312362909317017 lr:7.202372615241956e-05 tokens_per_second_per_gpu:4433.978901570362 mfu:20.91583499351734 +Step 12934 | loss:1.1172159910202026 lr:7.201983316100736e-05 tokens_per_second_per_gpu:4507.9605399803795 mfu:21.264819004466503 +Step 12935 | loss:0.97935950756073 lr:7.201594000398121e-05 tokens_per_second_per_gpu:5314.065132172097 mfu:25.06735190146049 +Step 12936 | loss:0.9635738730430603 lr:7.201204668137036e-05 tokens_per_second_per_gpu:3931.343277261157 mfu:18.54481699517018 +Step 12937 | loss:1.00527024269104 lr:7.200815319320409e-05 tokens_per_second_per_gpu:5178.8810444764595 mfu:24.42966549501644 +Step 12938 | loss:1.0136750936508179 lr:7.200425953951169e-05 tokens_per_second_per_gpu:5384.989777439055 mfu:25.401915554177023 +Step 12939 | loss:1.1328198909759521 lr:7.200036572032245e-05 tokens_per_second_per_gpu:5289.278061834325 mfu:24.950427061566923 +Step 12940 | loss:1.0829579830169678 lr:7.199647173566566e-05 tokens_per_second_per_gpu:5298.909254910915 mfu:24.995859042560188 +Step 12941 | loss:1.0251494646072388 lr:7.199257758557058e-05 tokens_per_second_per_gpu:4328.791005147347 mfu:20.419645739184226 +Step 12942 | loss:1.1828123331069946 lr:7.198868327006653e-05 tokens_per_second_per_gpu:3728.875722829121 mfu:17.58974299638736 +Step 12943 | loss:1.068471908569336 lr:7.19847887891828e-05 tokens_per_second_per_gpu:5338.01860330847 mfu:25.18034451169436 +Step 12944 | loss:1.1446943283081055 lr:7.198089414294865e-05 tokens_per_second_per_gpu:3782.5924192719967 mfu:17.843133818521885 +Step 12945 | loss:1.2558683156967163 lr:7.19769993313934e-05 tokens_per_second_per_gpu:3555.9513320420883 mfu:16.774029141101508 +Step 12946 | loss:1.1119520664215088 lr:7.197310435454632e-05 tokens_per_second_per_gpu:4938.401515097803 mfu:23.29528252489872 +Step 12947 | loss:1.0431077480316162 lr:7.196920921243674e-05 tokens_per_second_per_gpu:4732.933308141308 mfu:22.32605393619392 +Step 12948 | loss:1.1320234537124634 lr:7.196531390509391e-05 tokens_per_second_per_gpu:5170.267469844884 mfu:24.389033793851922 +Step 12949 | loss:1.1106336116790771 lr:7.196141843254717e-05 tokens_per_second_per_gpu:3941.222785963074 mfu:18.59142032333512 +Step 12950 | loss:1.0304044485092163 lr:7.195752279482578e-05 tokens_per_second_per_gpu:4653.269886772909 mfu:21.950267985617934 +Step 12951 | loss:1.0450457334518433 lr:7.195362699195909e-05 tokens_per_second_per_gpu:3759.9825131177604 mfu:17.736479033544573 +Step 12952 | loss:1.105493187904358 lr:7.194973102397634e-05 tokens_per_second_per_gpu:5343.648903672323 mfu:25.206903599138958 +Step 12953 | loss:1.0977354049682617 lr:7.194583489090687e-05 tokens_per_second_per_gpu:5132.318773897739 mfu:24.21002332035406 +Step 12954 | loss:1.0850224494934082 lr:7.194193859277999e-05 tokens_per_second_per_gpu:5319.444228825388 mfu:25.09272601814259 +Step 12955 | loss:1.1900521516799927 lr:7.193804212962499e-05 tokens_per_second_per_gpu:4675.428288815282 mfu:22.05479294866504 +Step 12956 | loss:0.9990271329879761 lr:7.193414550147116e-05 tokens_per_second_per_gpu:5314.916094791856 mfu:25.071366037327085 +Step 12957 | loss:1.0297836065292358 lr:7.193024870834782e-05 tokens_per_second_per_gpu:4188.1772109897265 mfu:19.75634648095555 +Step 12958 | loss:1.1188586950302124 lr:7.192635175028431e-05 tokens_per_second_per_gpu:4319.770087750234 mfu:20.377092532694824 +Step 12959 | loss:1.0202001333236694 lr:7.192245462730987e-05 tokens_per_second_per_gpu:3702.595881353733 mfu:17.465776500344646 +Step 12960 | loss:0.9805970788002014 lr:7.191855733945387e-05 tokens_per_second_per_gpu:4214.826777867675 mfu:19.882057034803807 +Step 12961 | loss:1.2283421754837036 lr:7.191465988674562e-05 tokens_per_second_per_gpu:5340.867961581748 mfu:25.193785420820063 +Step 12962 | loss:1.0625468492507935 lr:7.191076226921439e-05 tokens_per_second_per_gpu:5331.6580471538655 mfu:25.15034068309031 +Step 12963 | loss:1.0046114921569824 lr:7.190686448688954e-05 tokens_per_second_per_gpu:4800.1139276826125 mfu:22.642956380765888 +Step 12964 | loss:1.1099311113357544 lr:7.190296653980035e-05 tokens_per_second_per_gpu:4640.398178642948 mfu:21.88954994222904 +Step 12965 | loss:1.1337714195251465 lr:7.189906842797617e-05 tokens_per_second_per_gpu:4723.897800897255 mfu:22.283431906907218 +Step 12966 | loss:1.161840558052063 lr:7.189517015144629e-05 tokens_per_second_per_gpu:5349.731025918148 mfu:25.23559400748957 +Step 12967 | loss:0.9611109495162964 lr:7.189127171024006e-05 tokens_per_second_per_gpu:2879.9391628046887 mfu:13.585164399239183 +Step 12968 | loss:1.2330106496810913 lr:7.188737310438675e-05 tokens_per_second_per_gpu:3617.4797072773913 mfu:17.06426898490975 +Step 12969 | loss:1.0371447801589966 lr:7.188347433391576e-05 tokens_per_second_per_gpu:5338.899867571647 mfu:25.184501585582954 +Step 12970 | loss:1.1001805067062378 lr:7.187957539885632e-05 tokens_per_second_per_gpu:4299.205898729033 mfu:20.280087744469377 +Step 12971 | loss:1.1505131721496582 lr:7.187567629923783e-05 tokens_per_second_per_gpu:3368.296563184441 mfu:15.88882957919491 +Step 12972 | loss:1.132693886756897 lr:7.187177703508958e-05 tokens_per_second_per_gpu:5317.005150560034 mfu:25.081220469815246 +Step 12973 | loss:1.0968430042266846 lr:7.186787760644091e-05 tokens_per_second_per_gpu:3840.222809542688 mfu:18.114986202187236 +Step 12974 | loss:1.070634126663208 lr:7.186397801332113e-05 tokens_per_second_per_gpu:4356.123266108012 mfu:20.54857667750027 +Step 12975 | loss:1.1250889301300049 lr:7.186007825575957e-05 tokens_per_second_per_gpu:4381.074701327191 mfu:20.66627685458297 +Step 12976 | loss:0.9649242758750916 lr:7.18561783337856e-05 tokens_per_second_per_gpu:5373.259308734501 mfu:25.346580931873305 +Step 12977 | loss:0.9486515522003174 lr:7.185227824742852e-05 tokens_per_second_per_gpu:5342.82479372415 mfu:25.203016131905766 +Step 12978 | loss:1.0431292057037354 lr:7.184837799671765e-05 tokens_per_second_per_gpu:5320.873879256817 mfu:25.099469923151158 +Step 12979 | loss:1.1954671144485474 lr:7.184447758168236e-05 tokens_per_second_per_gpu:4396.146220836002 mfu:20.73737178357242 +Step 12980 | loss:1.0250012874603271 lr:7.184057700235196e-05 tokens_per_second_per_gpu:5053.988210772872 mfu:23.84052468952188 +Step 12981 | loss:1.1119887828826904 lr:7.183667625875579e-05 tokens_per_second_per_gpu:4462.931674621013 mfu:21.05241016384973 +Step 12982 | loss:1.0861356258392334 lr:7.183277535092318e-05 tokens_per_second_per_gpu:3723.4017356032405 mfu:17.563921264684357 +Step 12983 | loss:1.088799238204956 lr:7.18288742788835e-05 tokens_per_second_per_gpu:3957.8738798002305 mfu:18.66996637393455 +Step 12984 | loss:0.9688656330108643 lr:7.182497304266606e-05 tokens_per_second_per_gpu:4825.882149985086 mfu:22.764509481878733 +Step 12985 | loss:1.0225328207015991 lr:7.182107164230024e-05 tokens_per_second_per_gpu:4809.492485038333 mfu:22.687196635959708 +Step 12986 | loss:1.0499775409698486 lr:7.181717007781531e-05 tokens_per_second_per_gpu:5325.98347527683 mfu:25.123572759364535 +Step 12987 | loss:0.9757276773452759 lr:7.18132683492407e-05 tokens_per_second_per_gpu:5325.405078620254 mfu:25.120844363650818 +Step 12988 | loss:1.188544750213623 lr:7.180936645660569e-05 tokens_per_second_per_gpu:5167.326336017411 mfu:24.37515996377856 +Step 12989 | loss:1.1073325872421265 lr:7.180546439993965e-05 tokens_per_second_per_gpu:5325.220795704099 mfu:25.119975069693407 +Step 12990 | loss:1.077065110206604 lr:7.180156217927195e-05 tokens_per_second_per_gpu:4441.078436027845 mfu:20.94932470885877 +Step 12991 | loss:1.0516260862350464 lr:7.179765979463191e-05 tokens_per_second_per_gpu:3943.138086972998 mfu:18.600455124982982 +Step 12992 | loss:0.9918451309204102 lr:7.179375724604889e-05 tokens_per_second_per_gpu:5327.059942079616 mfu:25.12865063693745 +Step 12993 | loss:0.9596347212791443 lr:7.178985453355224e-05 tokens_per_second_per_gpu:4396.245503970538 mfu:20.73784011905748 +Step 12994 | loss:0.8632500171661377 lr:7.178595165717132e-05 tokens_per_second_per_gpu:5001.287455107986 mfu:23.591926233374505 +Step 12995 | loss:1.1793290376663208 lr:7.178204861693545e-05 tokens_per_second_per_gpu:3665.5858195434894 mfu:17.291193724218044 +Step 12996 | loss:1.0491931438446045 lr:7.177814541287405e-05 tokens_per_second_per_gpu:5034.182930106131 mfu:23.74709980148754 +Step 12997 | loss:1.0633131265640259 lr:7.177424204501643e-05 tokens_per_second_per_gpu:5094.654537060904 mfu:24.032354689013257 +Step 12998 | loss:1.0433323383331299 lr:7.177033851339194e-05 tokens_per_second_per_gpu:3740.1086695133795 mfu:17.64273072243534 +Step 12999 | loss:1.0443369150161743 lr:7.176643481802997e-05 tokens_per_second_per_gpu:5325.674762100933 mfu:25.122116506641927 +Step 13000 | loss:1.034757137298584 lr:7.176253095895986e-05 tokens_per_second_per_gpu:4356.43174331678 mfu:20.550031817125667 +Step 13001 | loss:1.1599630117416382 lr:7.1758626936211e-05 tokens_per_second_per_gpu:4479.221732339706 mfu:21.129253145479247 +Step 13002 | loss:1.079777479171753 lr:7.17547227498127e-05 tokens_per_second_per_gpu:3969.1498007488963 mfu:18.723156816919904 +Step 13003 | loss:1.0430610179901123 lr:7.175081839979437e-05 tokens_per_second_per_gpu:4406.616695458199 mfu:20.786762798812507 +Step 13004 | loss:0.9902039170265198 lr:7.174691388618534e-05 tokens_per_second_per_gpu:5380.989956912096 mfu:25.38304771831161 +Step 13005 | loss:0.9751105308532715 lr:7.174300920901503e-05 tokens_per_second_per_gpu:4872.474917865495 mfu:22.984295517516554 +Step 13006 | loss:1.0315039157867432 lr:7.173910436831274e-05 tokens_per_second_per_gpu:4145.257860114369 mfu:19.553888580080127 +Step 13007 | loss:1.0216631889343262 lr:7.173519936410788e-05 tokens_per_second_per_gpu:3707.7015286205938 mfu:17.489860709615677 +Step 13008 | loss:1.1024433374404907 lr:7.173129419642983e-05 tokens_per_second_per_gpu:4205.588339599307 mfu:19.838477745251804 +Step 13009 | loss:0.9431674480438232 lr:7.172738886530791e-05 tokens_per_second_per_gpu:4329.203038840371 mfu:20.421589372414413 +Step 13010 | loss:1.1509995460510254 lr:7.172348337077155e-05 tokens_per_second_per_gpu:4093.0894813606724 mfu:19.3078014366557 +Step 13011 | loss:1.1165213584899902 lr:7.17195777128501e-05 tokens_per_second_per_gpu:5352.136469641526 mfu:25.246940895943766 +Step 13012 | loss:1.1032949686050415 lr:7.171567189157295e-05 tokens_per_second_per_gpu:5310.570516431766 mfu:25.05086720277042 +Step 13013 | loss:1.0755021572113037 lr:7.171176590696943e-05 tokens_per_second_per_gpu:5258.6332875135095 mfu:24.805870432557988 +Step 13014 | loss:0.8871867060661316 lr:7.170785975906896e-05 tokens_per_second_per_gpu:5348.799639991403 mfu:25.231200501162288 +Step 13015 | loss:1.1930729150772095 lr:7.17039534479009e-05 tokens_per_second_per_gpu:4636.579470546951 mfu:21.871536444601436 +Step 13016 | loss:0.9833652377128601 lr:7.170004697349466e-05 tokens_per_second_per_gpu:4624.177706996934 mfu:21.813035209976956 +Step 13017 | loss:0.8762626647949219 lr:7.169614033587958e-05 tokens_per_second_per_gpu:4392.868609358147 mfu:20.721910731013196 +Step 13018 | loss:1.0401908159255981 lr:7.169223353508505e-05 tokens_per_second_per_gpu:4542.4287711095785 mfu:21.427411531589584 +Step 13019 | loss:1.030077576637268 lr:7.168832657114049e-05 tokens_per_second_per_gpu:4095.0914320255083 mfu:19.317244979509734 +Step 13020 | loss:1.0631697177886963 lr:7.168441944407523e-05 tokens_per_second_per_gpu:3329.90770274313 mfu:15.707742774678259 +Step 13021 | loss:1.0979788303375244 lr:7.168051215391871e-05 tokens_per_second_per_gpu:3659.0453438999807 mfu:17.260341184687398 +Step 13022 | loss:1.0905555486679077 lr:7.167660470070026e-05 tokens_per_second_per_gpu:5295.179960051008 mfu:24.97826732242269 +Step 13023 | loss:1.1855107545852661 lr:7.167269708444934e-05 tokens_per_second_per_gpu:4392.748827334319 mfu:20.721345698769877 +Step 13024 | loss:0.8682725429534912 lr:7.166878930519528e-05 tokens_per_second_per_gpu:5351.823994437267 mfu:25.24546689709151 +Step 13025 | loss:0.8595626950263977 lr:7.166488136296748e-05 tokens_per_second_per_gpu:4368.297885031647 mfu:20.60600643216724 +Step 13026 | loss:1.0965734720230103 lr:7.166097325779536e-05 tokens_per_second_per_gpu:3213.1649692312144 mfu:15.1570473823383 +Step 13027 | loss:1.099584937095642 lr:7.165706498970829e-05 tokens_per_second_per_gpu:4241.786060755337 mfu:20.009228571912875 +Step 13028 | loss:1.0154649019241333 lr:7.165315655873567e-05 tokens_per_second_per_gpu:5023.457742840771 mfu:23.696507263251195 +Step 13029 | loss:0.9552692770957947 lr:7.164924796490689e-05 tokens_per_second_per_gpu:4026.952217693165 mfu:18.995820679755298 +Step 13030 | loss:1.1886873245239258 lr:7.164533920825137e-05 tokens_per_second_per_gpu:4131.78029735069 mfu:19.490312617014528 +Step 13031 | loss:1.1450542211532593 lr:7.164143028879849e-05 tokens_per_second_per_gpu:4127.40281039712 mfu:19.46966326417825 +Step 13032 | loss:1.0017646551132202 lr:7.163752120657763e-05 tokens_per_second_per_gpu:4021.4140368623875 mfu:18.969696135865437 +Step 13033 | loss:0.9760352373123169 lr:7.163361196161823e-05 tokens_per_second_per_gpu:3913.224971307504 mfu:18.459349854685037 +Step 13034 | loss:0.9775369763374329 lr:7.162970255394967e-05 tokens_per_second_per_gpu:3928.6642369634055 mfu:18.53217950499428 +Step 13035 | loss:1.1691147089004517 lr:7.162579298360136e-05 tokens_per_second_per_gpu:5232.738855329119 mfu:24.683721977898514 +Step 13036 | loss:1.1070361137390137 lr:7.162188325060269e-05 tokens_per_second_per_gpu:4991.693807175499 mfu:23.54667135923184 +Step 13037 | loss:1.0126256942749023 lr:7.161797335498311e-05 tokens_per_second_per_gpu:3682.7081131866994 mfu:17.371962504697603 +Step 13038 | loss:0.944644033908844 lr:7.161406329677195e-05 tokens_per_second_per_gpu:5161.896510236851 mfu:24.349546549147846 +Step 13039 | loss:0.9380233883857727 lr:7.161015307599868e-05 tokens_per_second_per_gpu:4859.268328191173 mfu:22.92199778074623 +Step 13040 | loss:1.140993356704712 lr:7.160624269269268e-05 tokens_per_second_per_gpu:5323.1666207959415 mfu:25.11028517617325 +Step 13041 | loss:1.1943721771240234 lr:7.16023321468834e-05 tokens_per_second_per_gpu:3640.9060254924757 mfu:17.17477492487234 +Step 13042 | loss:1.082250952720642 lr:7.15984214386002e-05 tokens_per_second_per_gpu:4368.033029675449 mfu:20.60475706426307 +Step 13043 | loss:1.1055272817611694 lr:7.159451056787253e-05 tokens_per_second_per_gpu:4152.16392157968 mfu:19.58646565030759 +Step 13044 | loss:1.0825310945510864 lr:7.159059953472979e-05 tokens_per_second_per_gpu:5345.614947581531 mfu:25.21617776370037 +Step 13045 | loss:1.0113649368286133 lr:7.15866883392014e-05 tokens_per_second_per_gpu:4782.8194435948835 mfu:22.561375348580814 +Step 13046 | loss:0.9869050979614258 lr:7.158277698131675e-05 tokens_per_second_per_gpu:3463.9825789702672 mfu:16.3401968413744 +Step 13047 | loss:1.157453179359436 lr:7.157886546110529e-05 tokens_per_second_per_gpu:5341.898128727307 mfu:25.19864489501031 +Step 13048 | loss:1.0239355564117432 lr:7.157495377859644e-05 tokens_per_second_per_gpu:4040.7696892006106 mfu:19.061000050359656 +Step 13049 | loss:1.0765883922576904 lr:7.15710419338196e-05 tokens_per_second_per_gpu:5385.549431317924 mfu:25.404555537010452 +Step 13050 | loss:1.0830795764923096 lr:7.15671299268042e-05 tokens_per_second_per_gpu:3345.821231408129 mfu:15.782809604519024 +Step 13051 | loss:1.004453182220459 lr:7.156321775757967e-05 tokens_per_second_per_gpu:3962.947773563162 mfu:18.69390079650984 +Step 13052 | loss:1.14120352268219 lr:7.155930542617542e-05 tokens_per_second_per_gpu:5379.310297928183 mfu:25.375124480305118 +Step 13053 | loss:1.1220808029174805 lr:7.155539293262089e-05 tokens_per_second_per_gpu:4284.11380659983 mfu:20.208895770919487 +Step 13054 | loss:1.1644198894500732 lr:7.15514802769455e-05 tokens_per_second_per_gpu:4655.320332704529 mfu:21.95994028891954 +Step 13055 | loss:0.9941264390945435 lr:7.154756745917867e-05 tokens_per_second_per_gpu:5325.089947341116 mfu:25.119357835640187 +Step 13056 | loss:0.9898106455802917 lr:7.154365447934984e-05 tokens_per_second_per_gpu:4360.216371018425 mfu:20.5678845517184 +Step 13057 | loss:1.0694923400878906 lr:7.153974133748845e-05 tokens_per_second_per_gpu:4177.2629332596725 mfu:19.704861970735084 +Step 13058 | loss:1.1514499187469482 lr:7.153582803362388e-05 tokens_per_second_per_gpu:3973.481468359578 mfu:18.743590032098194 +Step 13059 | loss:0.9912546277046204 lr:7.153191456778563e-05 tokens_per_second_per_gpu:4851.021996868062 mfu:22.88309842892591 +Step 13060 | loss:1.1952029466629028 lr:7.152800094000309e-05 tokens_per_second_per_gpu:3979.9189596424067 mfu:18.773956776828566 +Step 13061 | loss:0.9999629855155945 lr:7.152408715030572e-05 tokens_per_second_per_gpu:5355.00829499759 mfu:25.260487786132256 +Step 13062 | loss:1.0204797983169556 lr:7.152017319872293e-05 tokens_per_second_per_gpu:5342.50701368226 mfu:25.201517109229304 +Step 13063 | loss:1.0931166410446167 lr:7.151625908528419e-05 tokens_per_second_per_gpu:5361.043743640593 mfu:25.288958027135017 +Step 13064 | loss:0.992847204208374 lr:7.15123448100189e-05 tokens_per_second_per_gpu:4802.7764105342985 mfu:22.65551576664371 +Step 13065 | loss:1.0629361867904663 lr:7.150843037295653e-05 tokens_per_second_per_gpu:4340.337203787013 mfu:20.474111128152153 +Step 13066 | loss:1.0209203958511353 lr:7.150451577412652e-05 tokens_per_second_per_gpu:4703.430264455269 mfu:22.186883045388555 +Step 13067 | loss:1.026958703994751 lr:7.150060101355827e-05 tokens_per_second_per_gpu:5380.620065371157 mfu:25.381302876802994 +Step 13068 | loss:0.9872680902481079 lr:7.149668609128129e-05 tokens_per_second_per_gpu:4815.140924875947 mfu:22.713841290396946 +Step 13069 | loss:1.3380639553070068 lr:7.149277100732497e-05 tokens_per_second_per_gpu:3705.039657093001 mfu:17.47730420745791 +Step 13070 | loss:0.8334227204322815 lr:7.14888557617188e-05 tokens_per_second_per_gpu:5320.134510852816 mfu:25.09598219624022 +Step 13071 | loss:1.0669398307800293 lr:7.148494035449219e-05 tokens_per_second_per_gpu:5357.049551024417 mfu:25.270116739084987 +Step 13072 | loss:1.0594426393508911 lr:7.148102478567459e-05 tokens_per_second_per_gpu:5111.857379519749 mfu:24.113503431999213 +Step 13073 | loss:1.2092772722244263 lr:7.147710905529548e-05 tokens_per_second_per_gpu:4093.9801008897066 mfu:19.312002640929514 +Step 13074 | loss:1.1268872022628784 lr:7.147319316338428e-05 tokens_per_second_per_gpu:5378.698415783171 mfu:25.372238127829238 +Step 13075 | loss:1.1195026636123657 lr:7.146927710997047e-05 tokens_per_second_per_gpu:5193.153292684267 mfu:24.496990124909313 +Step 13076 | loss:1.138963222503662 lr:7.146536089508345e-05 tokens_per_second_per_gpu:2594.3747270574304 mfu:12.238108233502542 +Step 13077 | loss:0.9928668737411499 lr:7.146144451875277e-05 tokens_per_second_per_gpu:4430.464025363818 mfu:20.89925472275135 +Step 13078 | loss:0.9551813006401062 lr:7.145752798100778e-05 tokens_per_second_per_gpu:5345.7298327502485 mfu:25.21671969664264 +Step 13079 | loss:0.9749350547790527 lr:7.1453611281878e-05 tokens_per_second_per_gpu:5344.899692769227 mfu:25.212803784715675 +Step 13080 | loss:1.2606066465377808 lr:7.144969442139287e-05 tokens_per_second_per_gpu:5339.125228823308 mfu:25.185564653057632 +Step 13081 | loss:0.9110838174819946 lr:7.144577739958186e-05 tokens_per_second_per_gpu:4223.627642705336 mfu:19.92357221582543 +Step 13082 | loss:0.9712315201759338 lr:7.144186021647441e-05 tokens_per_second_per_gpu:5238.017955614825 mfu:24.708624394653576 +Step 13083 | loss:1.1062674522399902 lr:7.143794287209999e-05 tokens_per_second_per_gpu:4539.6824267774755 mfu:21.414456556800392 +Step 13084 | loss:1.0358905792236328 lr:7.143402536648809e-05 tokens_per_second_per_gpu:5366.867755993136 mfu:25.316430887080934 +Step 13085 | loss:0.934695839881897 lr:7.143010769966813e-05 tokens_per_second_per_gpu:4548.17389207525 mfu:21.45451224740337 +Step 13086 | loss:1.1061861515045166 lr:7.14261898716696e-05 tokens_per_second_per_gpu:5326.41323438847 mfu:25.125600006418605 +Step 13087 | loss:1.2021087408065796 lr:7.142227188252196e-05 tokens_per_second_per_gpu:5334.753419369824 mfu:25.164942081957943 +Step 13088 | loss:1.2013002634048462 lr:7.141835373225468e-05 tokens_per_second_per_gpu:3931.622411190079 mfu:18.546133717538066 +Step 13089 | loss:1.0491863489151 lr:7.141443542089723e-05 tokens_per_second_per_gpu:4526.889423879089 mfu:21.354109779417335 +Step 13090 | loss:1.0313972234725952 lr:7.141051694847908e-05 tokens_per_second_per_gpu:4608.930611172089 mfu:21.741112057548364 +Step 13091 | loss:1.1652551889419556 lr:7.14065983150297e-05 tokens_per_second_per_gpu:5267.331220300529 mfu:24.846900065534722 +Step 13092 | loss:1.0664559602737427 lr:7.140267952057855e-05 tokens_per_second_per_gpu:3915.4989433330934 mfu:18.47007656870938 +Step 13093 | loss:1.221941590309143 lr:7.139876056515514e-05 tokens_per_second_per_gpu:4457.383886853143 mfu:21.026240302398485 +Step 13094 | loss:1.0913041830062866 lr:7.13948414487889e-05 tokens_per_second_per_gpu:5313.703547782972 mfu:25.065646246204327 +Step 13095 | loss:1.031704068183899 lr:7.139092217150934e-05 tokens_per_second_per_gpu:4528.175031725469 mfu:21.360174210976027 +Step 13096 | loss:1.055775761604309 lr:7.138700273334592e-05 tokens_per_second_per_gpu:5314.960787728842 mfu:25.07157686153625 +Step 13097 | loss:0.991730809211731 lr:7.138308313432813e-05 tokens_per_second_per_gpu:4416.816525888861 mfu:20.834877139224464 +Step 13098 | loss:1.0618449449539185 lr:7.137916337448543e-05 tokens_per_second_per_gpu:4374.931087339945 mfu:20.6372963791965 +Step 13099 | loss:1.016038179397583 lr:7.137524345384733e-05 tokens_per_second_per_gpu:5184.524809403631 mfu:24.456288097103496 +Step 13100 | loss:0.9866613149642944 lr:7.137132337244329e-05 tokens_per_second_per_gpu:3414.9833667733437 mfu:16.109059197313844 +Step 13101 | loss:1.1330159902572632 lr:7.136740313030279e-05 tokens_per_second_per_gpu:4680.369234530027 mfu:22.07810023261376 +Step 13102 | loss:1.0207315683364868 lr:7.136348272745534e-05 tokens_per_second_per_gpu:4560.407995135582 mfu:21.512222598892176 +Step 13103 | loss:1.0161324739456177 lr:7.13595621639304e-05 tokens_per_second_per_gpu:5360.95477736745 mfu:25.288538358045397 +Step 13104 | loss:0.9522259831428528 lr:7.135564143975748e-05 tokens_per_second_per_gpu:4744.875220403432 mfu:22.382385973834687 +Step 13105 | loss:1.173691987991333 lr:7.135172055496604e-05 tokens_per_second_per_gpu:5359.68666292343 mfu:25.282556445850318 +Step 13106 | loss:1.1223328113555908 lr:7.13477995095856e-05 tokens_per_second_per_gpu:5317.0642574578815 mfu:25.08149928713707 +Step 13107 | loss:1.1677156686782837 lr:7.134387830364563e-05 tokens_per_second_per_gpu:5211.845982468843 mfu:24.585166731924907 +Step 13108 | loss:0.9713121056556702 lr:7.133995693717561e-05 tokens_per_second_per_gpu:3718.507493882572 mfu:17.540834291444217 +Step 13109 | loss:1.1252920627593994 lr:7.133603541020506e-05 tokens_per_second_per_gpu:4306.332442543362 mfu:20.31370486755532 +Step 13110 | loss:1.035508632659912 lr:7.133211372276348e-05 tokens_per_second_per_gpu:4820.16962481351 mfu:22.737562525987116 +Step 13111 | loss:0.9973744750022888 lr:7.132819187488036e-05 tokens_per_second_per_gpu:4791.067603381957 mfu:22.600283325577614 +Step 13112 | loss:1.0050336122512817 lr:7.132426986658516e-05 tokens_per_second_per_gpu:5307.258760119591 mfu:25.035245083202 +Step 13113 | loss:1.0680629014968872 lr:7.132034769790741e-05 tokens_per_second_per_gpu:4525.935936633144 mfu:21.349612017396513 +Step 13114 | loss:1.1031014919281006 lr:7.131642536887662e-05 tokens_per_second_per_gpu:3966.8460879700187 mfu:18.71228981068801 +Step 13115 | loss:1.0082088708877563 lr:7.131250287952226e-05 tokens_per_second_per_gpu:3547.314728155678 mfu:16.73328880701274 +Step 13116 | loss:1.1834276914596558 lr:7.130858022987386e-05 tokens_per_second_per_gpu:3473.5933498984487 mfu:16.38553248760924 +Step 13117 | loss:1.157772183418274 lr:7.130465741996088e-05 tokens_per_second_per_gpu:5358.709507711152 mfu:25.277947037994128 +Step 13118 | loss:1.1268330812454224 lr:7.13007344498129e-05 tokens_per_second_per_gpu:4229.730211741057 mfu:19.95235907517257 +Step 13119 | loss:1.1316804885864258 lr:7.129681131945933e-05 tokens_per_second_per_gpu:5037.715741513012 mfu:23.763764675653576 +Step 13120 | loss:1.1598119735717773 lr:7.129288802892975e-05 tokens_per_second_per_gpu:5336.436787274677 mfu:25.172882815577314 +Step 13121 | loss:1.0192543268203735 lr:7.128896457825364e-05 tokens_per_second_per_gpu:3064.492960597239 mfu:14.455736151551498 +Step 13122 | loss:1.1975542306900024 lr:7.128504096746051e-05 tokens_per_second_per_gpu:5361.595350772603 mfu:25.291560052090592 +Step 13123 | loss:1.31490159034729 lr:7.128111719657987e-05 tokens_per_second_per_gpu:5322.839630702509 mfu:25.108742708112263 +Step 13124 | loss:0.9444905519485474 lr:7.127719326564123e-05 tokens_per_second_per_gpu:4601.547616852256 mfu:21.70628521367299 +Step 13125 | loss:1.048451542854309 lr:7.127326917467411e-05 tokens_per_second_per_gpu:4754.201190320424 mfu:22.426378165107685 +Step 13126 | loss:1.1612318754196167 lr:7.1269344923708e-05 tokens_per_second_per_gpu:4897.135541240124 mfu:23.100623885510895 +Step 13127 | loss:0.9513354897499084 lr:7.126542051277245e-05 tokens_per_second_per_gpu:5314.340214909017 mfu:25.068649513664493 +Step 13128 | loss:1.089660882949829 lr:7.126149594189696e-05 tokens_per_second_per_gpu:4346.101930535634 mfu:20.501304327789207 +Step 13129 | loss:1.1166993379592896 lr:7.125757121111106e-05 tokens_per_second_per_gpu:5203.041940881954 mfu:24.54363656563516 +Step 13130 | loss:1.0256699323654175 lr:7.125364632044422e-05 tokens_per_second_per_gpu:3632.410761810033 mfu:17.134701316640815 +Step 13131 | loss:1.0206124782562256 lr:7.124972126992602e-05 tokens_per_second_per_gpu:4946.483939299996 mfu:23.33340869886423 +Step 13132 | loss:0.9261381030082703 lr:7.124579605958594e-05 tokens_per_second_per_gpu:4349.099478581801 mfu:20.515444273357407 +Step 13133 | loss:1.0644968748092651 lr:7.124187068945351e-05 tokens_per_second_per_gpu:4559.424602267137 mfu:21.5075837669476 +Step 13134 | loss:1.1025776863098145 lr:7.123794515955828e-05 tokens_per_second_per_gpu:5341.64237029214 mfu:25.197438438834734 +Step 13135 | loss:1.0123709440231323 lr:7.123401946992974e-05 tokens_per_second_per_gpu:3436.9398236315537 mfu:16.21263155047185 +Step 13136 | loss:1.1108732223510742 lr:7.123009362059745e-05 tokens_per_second_per_gpu:4725.812937540074 mfu:22.292465933207403 +Step 13137 | loss:1.2378078699111938 lr:7.122616761159089e-05 tokens_per_second_per_gpu:4113.056934615796 mfu:19.40199132045894 +Step 13138 | loss:1.0123393535614014 lr:7.122224144293963e-05 tokens_per_second_per_gpu:4365.498482628634 mfu:20.592801173404926 +Step 13139 | loss:1.1003729104995728 lr:7.121831511467319e-05 tokens_per_second_per_gpu:4702.092984792078 mfu:22.180574869053846 +Step 13140 | loss:1.1394399404525757 lr:7.121438862682109e-05 tokens_per_second_per_gpu:5333.674520490791 mfu:25.159852731866415 +Step 13141 | loss:1.0251003503799438 lr:7.121046197941286e-05 tokens_per_second_per_gpu:5377.9612533966865 mfu:25.368760807079315 +Step 13142 | loss:0.8909276127815247 lr:7.120653517247805e-05 tokens_per_second_per_gpu:5329.6417271236105 mfu:25.140829357487924 +Step 13143 | loss:0.9801769852638245 lr:7.120260820604617e-05 tokens_per_second_per_gpu:4745.679636716987 mfu:22.3861805428333 +Step 13144 | loss:1.214115858078003 lr:7.119868108014677e-05 tokens_per_second_per_gpu:4589.906457681013 mfu:21.651371879675672 +Step 13145 | loss:1.0582804679870605 lr:7.119475379480939e-05 tokens_per_second_per_gpu:4863.361145812411 mfu:22.941304299772302 +Step 13146 | loss:0.9342369437217712 lr:7.119082635006357e-05 tokens_per_second_per_gpu:4316.173299960752 mfu:20.360125871016557 +Step 13147 | loss:1.071956753730774 lr:7.118689874593883e-05 tokens_per_second_per_gpu:3433.4716593673543 mfu:16.19627162790778 +Step 13148 | loss:1.1721515655517578 lr:7.118297098246473e-05 tokens_per_second_per_gpu:3694.5967443419804 mfu:17.4280432062667 +Step 13149 | loss:0.9975677728652954 lr:7.11790430596708e-05 tokens_per_second_per_gpu:3844.565325245269 mfu:18.13547058966589 +Step 13150 | loss:0.9996687769889832 lr:7.117511497758657e-05 tokens_per_second_per_gpu:3987.072249854022 mfu:18.807700067233917 +Step 13151 | loss:1.1036477088928223 lr:7.11711867362416e-05 tokens_per_second_per_gpu:4106.926149587764 mfu:19.373071361461925 +Step 13152 | loss:1.114848256111145 lr:7.116725833566545e-05 tokens_per_second_per_gpu:4720.918897139093 mfu:22.269379909626576 +Step 13153 | loss:1.0576227903366089 lr:7.116332977588764e-05 tokens_per_second_per_gpu:3707.169915226632 mfu:17.487352998533762 +Step 13154 | loss:1.2888445854187012 lr:7.115940105693773e-05 tokens_per_second_per_gpu:5355.667104152882 mfu:25.263595501322197 +Step 13155 | loss:0.9553711414337158 lr:7.115547217884524e-05 tokens_per_second_per_gpu:3745.826756894461 mfu:17.66970391621821 +Step 13156 | loss:1.0217581987380981 lr:7.115154314163976e-05 tokens_per_second_per_gpu:4725.918015636114 mfu:22.29296160451459 +Step 13157 | loss:1.0864378213882446 lr:7.114761394535084e-05 tokens_per_second_per_gpu:5316.261510271489 mfu:25.077712591696066 +Step 13158 | loss:1.1497149467468262 lr:7.114368459000802e-05 tokens_per_second_per_gpu:5056.245454119913 mfu:23.851172491515303 +Step 13159 | loss:0.9775829911231995 lr:7.113975507564083e-05 tokens_per_second_per_gpu:4263.667006342758 mfu:20.112444725523023 +Step 13160 | loss:1.1864197254180908 lr:7.113582540227885e-05 tokens_per_second_per_gpu:3783.508401979258 mfu:17.8474546652348 +Step 13161 | loss:1.0509052276611328 lr:7.113189556995165e-05 tokens_per_second_per_gpu:5360.401903892728 mfu:25.285930359535186 +Step 13162 | loss:1.1615005731582642 lr:7.112796557868873e-05 tokens_per_second_per_gpu:4282.002702079825 mfu:20.19893733070705 +Step 13163 | loss:1.106714129447937 lr:7.11240354285197e-05 tokens_per_second_per_gpu:5368.269946763559 mfu:25.323045260183438 +Step 13164 | loss:1.180946946144104 lr:7.112010511947412e-05 tokens_per_second_per_gpu:5031.592555641749 mfu:23.73488056317559 +Step 13165 | loss:1.12922203540802 lr:7.111617465158153e-05 tokens_per_second_per_gpu:4759.341950350624 mfu:22.45062800727424 +Step 13166 | loss:1.1193349361419678 lr:7.111224402487148e-05 tokens_per_second_per_gpu:5330.547398018374 mfu:25.145101561621914 +Step 13167 | loss:0.9667667150497437 lr:7.110831323937356e-05 tokens_per_second_per_gpu:4972.713656787336 mfu:23.457138751503294 +Step 13168 | loss:1.2233529090881348 lr:7.110438229511733e-05 tokens_per_second_per_gpu:4728.689554631887 mfu:22.30603542683097 +Step 13169 | loss:1.0912431478500366 lr:7.110045119213233e-05 tokens_per_second_per_gpu:4832.052766210441 mfu:22.793617331429427 +Step 13170 | loss:1.0484819412231445 lr:7.109651993044814e-05 tokens_per_second_per_gpu:3642.5300283897514 mfu:17.182435623622258 +Step 13171 | loss:1.1216908693313599 lr:7.109258851009437e-05 tokens_per_second_per_gpu:4237.143560656316 mfu:19.9873291068536 +Step 13172 | loss:1.025746464729309 lr:7.108865693110052e-05 tokens_per_second_per_gpu:5385.520261121817 mfu:25.40441793621847 +Step 13173 | loss:1.1409183740615845 lr:7.108472519349618e-05 tokens_per_second_per_gpu:4019.3639715320423 mfu:18.96002562792498 +Step 13174 | loss:1.1897284984588623 lr:7.108079329731094e-05 tokens_per_second_per_gpu:5363.096821176263 mfu:25.298642744161924 +Step 13175 | loss:0.8608458638191223 lr:7.107686124257438e-05 tokens_per_second_per_gpu:5168.795808610217 mfu:24.3820917167167 +Step 13176 | loss:1.1369935274124146 lr:7.107292902931606e-05 tokens_per_second_per_gpu:5344.14738729463 mfu:25.209255031435454 +Step 13177 | loss:1.1848688125610352 lr:7.106899665756552e-05 tokens_per_second_per_gpu:4623.098775973534 mfu:21.807945708254966 +Step 13178 | loss:0.9769415259361267 lr:7.106506412735239e-05 tokens_per_second_per_gpu:5322.861020145548 mfu:25.108843605764495 +Step 13179 | loss:1.0297720432281494 lr:7.106113143870624e-05 tokens_per_second_per_gpu:3853.088282156805 mfu:18.175674831584004 +Step 13180 | loss:1.1615262031555176 lr:7.105719859165661e-05 tokens_per_second_per_gpu:4885.761950033169 mfu:23.04697271525293 +Step 13181 | loss:1.1578147411346436 lr:7.10532655862331e-05 tokens_per_second_per_gpu:5313.119207747404 mfu:25.062889814557863 +Step 13182 | loss:0.9694413542747498 lr:7.10493324224653e-05 tokens_per_second_per_gpu:5340.588825517051 mfu:25.192468688377367 +Step 13183 | loss:1.0388879776000977 lr:7.104539910038278e-05 tokens_per_second_per_gpu:3648.164533965176 mfu:17.209014547767318 +Step 13184 | loss:1.002813458442688 lr:7.104146562001513e-05 tokens_per_second_per_gpu:4857.278403046282 mfu:22.912610964321537 +Step 13185 | loss:0.978587806224823 lr:7.103753198139194e-05 tokens_per_second_per_gpu:5332.569504911926 mfu:25.154640184845807 +Step 13186 | loss:1.1231824159622192 lr:7.103359818454279e-05 tokens_per_second_per_gpu:5251.0732747888605 mfu:24.770208562664642 +Step 13187 | loss:1.1883337497711182 lr:7.102966422949725e-05 tokens_per_second_per_gpu:4271.156125955256 mfu:20.147772180510742 +Step 13188 | loss:0.9754204154014587 lr:7.102573011628493e-05 tokens_per_second_per_gpu:5305.834857121401 mfu:25.02852828981663 +Step 13189 | loss:0.9547399282455444 lr:7.102179584493541e-05 tokens_per_second_per_gpu:5158.982360485005 mfu:24.335800007563176 +Step 13190 | loss:1.0580989122390747 lr:7.101786141547828e-05 tokens_per_second_per_gpu:5362.32747573266 mfu:25.295013610440577 +Step 13191 | loss:1.0538644790649414 lr:7.101392682794312e-05 tokens_per_second_per_gpu:4616.850774272634 mfu:21.778472818212805 +Step 13192 | loss:1.0989468097686768 lr:7.100999208235955e-05 tokens_per_second_per_gpu:4857.344789682853 mfu:22.912924121412974 +Step 13193 | loss:1.0063135623931885 lr:7.100605717875716e-05 tokens_per_second_per_gpu:3958.034536326284 mfu:18.670724218178776 +Step 13194 | loss:0.9407925605773926 lr:7.100212211716551e-05 tokens_per_second_per_gpu:5378.358171618862 mfu:25.370633138054547 +Step 13195 | loss:0.9292521476745605 lr:7.099818689761423e-05 tokens_per_second_per_gpu:5369.033680445422 mfu:25.32664792226719 +Step 13196 | loss:1.0838398933410645 lr:7.09942515201329e-05 tokens_per_second_per_gpu:5264.505820078441 mfu:24.833572170624144 +Step 13197 | loss:1.161013126373291 lr:7.099031598475115e-05 tokens_per_second_per_gpu:5304.588218232823 mfu:25.022647681480684 +Step 13198 | loss:1.1016664505004883 lr:7.098638029149853e-05 tokens_per_second_per_gpu:3770.8311261352987 mfu:17.787653792112547 +Step 13199 | loss:1.0320360660552979 lr:7.098244444040468e-05 tokens_per_second_per_gpu:3945.818895364522 mfu:18.61310095555888 +Step 13200 | loss:1.0323468446731567 lr:7.097850843149918e-05 tokens_per_second_per_gpu:3807.8852527182717 mfu:17.96244442929034 +Step 13201 | loss:0.972695529460907 lr:7.097457226481164e-05 tokens_per_second_per_gpu:5298.844782974047 mfu:24.99555491743393 +Step 13202 | loss:0.9784993529319763 lr:7.097063594037167e-05 tokens_per_second_per_gpu:3592.12025839762 mfu:16.944644137775725 +Step 13203 | loss:1.095022201538086 lr:7.096669945820887e-05 tokens_per_second_per_gpu:3509.7278456675003 mfu:16.555985069332568 +Step 13204 | loss:1.2475285530090332 lr:7.096276281835285e-05 tokens_per_second_per_gpu:4437.118851943497 mfu:20.930646675158197 +Step 13205 | loss:1.0365601778030396 lr:7.095882602083322e-05 tokens_per_second_per_gpu:5156.92139049967 mfu:24.32607806050461 +Step 13206 | loss:1.274631381034851 lr:7.095488906567957e-05 tokens_per_second_per_gpu:4806.6199144894545 mfu:22.673646230569574 +Step 13207 | loss:0.8862629532814026 lr:7.095095195292154e-05 tokens_per_second_per_gpu:5339.311254820765 mfu:25.186442169426943 +Step 13208 | loss:0.9298970699310303 lr:7.094701468258872e-05 tokens_per_second_per_gpu:4363.240059731077 mfu:20.58214780727047 +Step 13209 | loss:0.9788686633110046 lr:7.094307725471073e-05 tokens_per_second_per_gpu:5302.282024646034 mfu:25.0117689728546 +Step 13210 | loss:1.2945235967636108 lr:7.093913966931717e-05 tokens_per_second_per_gpu:4539.913376598872 mfu:21.415545986512026 +Step 13211 | loss:1.0196934938430786 lr:7.093520192643769e-05 tokens_per_second_per_gpu:3785.065354522314 mfu:17.85479907602389 +Step 13212 | loss:1.1045879125595093 lr:7.093126402610189e-05 tokens_per_second_per_gpu:5342.499716048626 mfu:25.20148268504635 +Step 13213 | loss:1.1751796007156372 lr:7.092732596833936e-05 tokens_per_second_per_gpu:3752.9926781375098 mfu:17.70350678935391 +Step 13214 | loss:1.0433720350265503 lr:7.092338775317974e-05 tokens_per_second_per_gpu:4944.045026929105 mfu:23.32190393308938 +Step 13215 | loss:1.1094985008239746 lr:7.091944938065267e-05 tokens_per_second_per_gpu:4805.028531365215 mfu:22.666139404854576 +Step 13216 | loss:1.0803639888763428 lr:7.091551085078775e-05 tokens_per_second_per_gpu:5317.670097092054 mfu:25.08435713605836 +Step 13217 | loss:0.9458326101303101 lr:7.09115721636146e-05 tokens_per_second_per_gpu:4072.164101855916 mfu:19.209092851293867 +Step 13218 | loss:0.9619130492210388 lr:7.090763331916285e-05 tokens_per_second_per_gpu:4431.037862801499 mfu:20.901961611851657 +Step 13219 | loss:1.0192255973815918 lr:7.090369431746211e-05 tokens_per_second_per_gpu:5363.80876147028 mfu:25.30200108799834 +Step 13220 | loss:1.049048900604248 lr:7.089975515854204e-05 tokens_per_second_per_gpu:4406.7467391393375 mfu:20.787376237044178 +Step 13221 | loss:1.1047132015228271 lr:7.089581584243222e-05 tokens_per_second_per_gpu:4828.411049447665 mfu:22.776438732118464 +Step 13222 | loss:1.0232059955596924 lr:7.089187636916231e-05 tokens_per_second_per_gpu:3392.313314618341 mfu:16.00212069338895 +Step 13223 | loss:1.0625406503677368 lr:7.088793673876193e-05 tokens_per_second_per_gpu:5335.215735285723 mfu:25.167122905012466 +Step 13224 | loss:1.130082607269287 lr:7.088399695126073e-05 tokens_per_second_per_gpu:4891.719819940345 mfu:23.075077004122498 +Step 13225 | loss:0.9393652081489563 lr:7.088005700668831e-05 tokens_per_second_per_gpu:5324.646840295347 mfu:25.117267624103434 +Step 13226 | loss:1.0210925340652466 lr:7.087611690507432e-05 tokens_per_second_per_gpu:4332.148767599873 mfu:20.43548487756614 +Step 13227 | loss:1.0483523607254028 lr:7.087217664644837e-05 tokens_per_second_per_gpu:5366.044386376058 mfu:25.312546912115838 +Step 13228 | loss:1.107938289642334 lr:7.086823623084015e-05 tokens_per_second_per_gpu:4680.358291625037 mfu:22.07804861306387 +Step 13229 | loss:1.2445682287216187 lr:7.086429565827925e-05 tokens_per_second_per_gpu:3053.35073988262 mfu:14.403176395381758 +Step 13230 | loss:0.9046065211296082 lr:7.086035492879531e-05 tokens_per_second_per_gpu:5353.266400416456 mfu:25.252270972195184 +Step 13231 | loss:1.1312482357025146 lr:7.085641404241799e-05 tokens_per_second_per_gpu:3626.266791518197 mfu:17.105719160504798 +Step 13232 | loss:1.1169912815093994 lr:7.085247299917691e-05 tokens_per_second_per_gpu:4717.570551139631 mfu:22.253585190260377 +Step 13233 | loss:1.020561933517456 lr:7.084853179910174e-05 tokens_per_second_per_gpu:5330.3396182166725 mfu:25.144121428846393 +Step 13234 | loss:0.9223624467849731 lr:7.084459044222207e-05 tokens_per_second_per_gpu:3898.0784336023735 mfu:18.387901052064127 +Step 13235 | loss:1.067081332206726 lr:7.08406489285676e-05 tokens_per_second_per_gpu:3921.385094193159 mfu:18.4978425465972 +Step 13236 | loss:0.9914550185203552 lr:7.083670725816795e-05 tokens_per_second_per_gpu:3626.830785530727 mfu:17.108379616489202 +Step 13237 | loss:0.9241053462028503 lr:7.083276543105274e-05 tokens_per_second_per_gpu:4961.160412052549 mfu:23.402640124902486 +Step 13238 | loss:0.9402649998664856 lr:7.082882344725168e-05 tokens_per_second_per_gpu:3658.8959843808852 mfu:17.259636630352464 +Step 13239 | loss:0.8881091475486755 lr:7.082488130679435e-05 tokens_per_second_per_gpu:4388.761430010181 mfu:20.702536465271784 +Step 13240 | loss:0.9141035676002502 lr:7.082093900971046e-05 tokens_per_second_per_gpu:4116.04111941263 mfu:19.416068229300038 +Step 13241 | loss:1.0929394960403442 lr:7.08169965560296e-05 tokens_per_second_per_gpu:5269.4780978703975 mfu:24.857027253326876 +Step 13242 | loss:1.0573182106018066 lr:7.081305394578148e-05 tokens_per_second_per_gpu:4588.958020069992 mfu:21.64689793764434 +Step 13243 | loss:1.0549681186676025 lr:7.080911117899569e-05 tokens_per_second_per_gpu:3806.196386424427 mfu:17.95447775883197 +Step 13244 | loss:1.1070563793182373 lr:7.080516825570196e-05 tokens_per_second_per_gpu:4049.9030128482814 mfu:19.10408349631142 +Step 13245 | loss:1.016003966331482 lr:7.080122517592988e-05 tokens_per_second_per_gpu:4674.7339522970415 mfu:22.05151764484247 +Step 13246 | loss:1.1140180826187134 lr:7.079728193970913e-05 tokens_per_second_per_gpu:3514.361579288098 mfu:16.577843181417407 +Step 13247 | loss:1.0954370498657227 lr:7.079333854706938e-05 tokens_per_second_per_gpu:4304.660000760889 mfu:20.3058156743194 +Step 13248 | loss:0.9925057291984558 lr:7.078939499804029e-05 tokens_per_second_per_gpu:4811.674620073452 mfu:22.697490139230997 +Step 13249 | loss:0.9703966975212097 lr:7.078545129265148e-05 tokens_per_second_per_gpu:5325.993306256924 mfu:25.123619133774977 +Step 13250 | loss:1.0276532173156738 lr:7.078150743093265e-05 tokens_per_second_per_gpu:4553.953328457777 mfu:21.48177483489328 +Step 13251 | loss:1.2052744626998901 lr:7.077756341291347e-05 tokens_per_second_per_gpu:5367.5279935648905 mfu:25.31954534035505 +Step 13252 | loss:1.0538250207901 lr:7.077361923862355e-05 tokens_per_second_per_gpu:4936.619140718443 mfu:23.286874760846857 +Step 13253 | loss:0.9449910521507263 lr:7.07696749080926e-05 tokens_per_second_per_gpu:5387.183448635239 mfu:25.412263475488327 +Step 13254 | loss:1.0968674421310425 lr:7.076573042135028e-05 tokens_per_second_per_gpu:5326.547851962977 mfu:25.12623502048568 +Step 13255 | loss:0.9386679530143738 lr:7.076178577842625e-05 tokens_per_second_per_gpu:3953.460378463081 mfu:18.649147135106237 +Step 13256 | loss:1.235336184501648 lr:7.075784097935019e-05 tokens_per_second_per_gpu:3913.164139677789 mfu:18.459062901508727 +Step 13257 | loss:1.0087528228759766 lr:7.075389602415174e-05 tokens_per_second_per_gpu:4321.490775349378 mfu:20.385209309679563 +Step 13258 | loss:0.9662161469459534 lr:7.07499509128606e-05 tokens_per_second_per_gpu:5135.754830578717 mfu:24.226231785969954 +Step 13259 | loss:1.046642780303955 lr:7.074600564550643e-05 tokens_per_second_per_gpu:3714.4820161918956 mfu:17.52184542635998 +Step 13260 | loss:1.03097403049469 lr:7.074206022211891e-05 tokens_per_second_per_gpu:3585.471966147022 mfu:16.913283008914032 +Step 13261 | loss:1.031316876411438 lr:7.07381146427277e-05 tokens_per_second_per_gpu:5140.925805354949 mfu:24.250624156248715 +Step 13262 | loss:1.199641466140747 lr:7.073416890736249e-05 tokens_per_second_per_gpu:4137.428103685822 mfu:19.516954283112472 +Step 13263 | loss:1.0516871213912964 lr:7.073022301605294e-05 tokens_per_second_per_gpu:4120.005899666488 mfu:19.43477077421875 +Step 13264 | loss:1.0616614818572998 lr:7.072627696882876e-05 tokens_per_second_per_gpu:4266.921617627721 mfu:20.12779728224814 +Step 13265 | loss:1.0273170471191406 lr:7.072233076571959e-05 tokens_per_second_per_gpu:5335.006026969478 mfu:25.16613367510481 +Step 13266 | loss:1.035368800163269 lr:7.071838440675512e-05 tokens_per_second_per_gpu:5352.538836859255 mfu:25.24883893076072 +Step 13267 | loss:1.265450358390808 lr:7.071443789196505e-05 tokens_per_second_per_gpu:4235.843097354559 mfu:19.98119459957729 +Step 13268 | loss:1.1207300424575806 lr:7.071049122137904e-05 tokens_per_second_per_gpu:5332.079383014064 mfu:25.152328196230172 +Step 13269 | loss:1.2818231582641602 lr:7.07065443950268e-05 tokens_per_second_per_gpu:3758.8480571172854 mfu:17.731127611032182 +Step 13270 | loss:1.02128267288208 lr:7.070259741293797e-05 tokens_per_second_per_gpu:5334.029750101078 mfu:25.16152841054656 +Step 13271 | loss:0.9741270542144775 lr:7.069865027514229e-05 tokens_per_second_per_gpu:4165.590768723962 mfu:19.649802379143516 +Step 13272 | loss:1.0119929313659668 lr:7.06947029816694e-05 tokens_per_second_per_gpu:4089.7952278733956 mfu:19.292261880898614 +Step 13273 | loss:1.0644257068634033 lr:7.069075553254902e-05 tokens_per_second_per_gpu:4794.3128454899415 mfu:22.615591686296376 +Step 13274 | loss:0.935824453830719 lr:7.068680792781081e-05 tokens_per_second_per_gpu:5331.962588634072 mfu:25.151777257212576 +Step 13275 | loss:0.9179502129554749 lr:7.068286016748449e-05 tokens_per_second_per_gpu:4562.922294379903 mfu:21.52408297741133 +Step 13276 | loss:1.083357810974121 lr:7.067891225159976e-05 tokens_per_second_per_gpu:5388.268547984135 mfu:25.417382074254107 +Step 13277 | loss:1.0959612131118774 lr:7.067496418018626e-05 tokens_per_second_per_gpu:3367.7024747686864 mfu:15.88602716277586 +Step 13278 | loss:0.8913787007331848 lr:7.067101595327374e-05 tokens_per_second_per_gpu:4203.928476618502 mfu:19.83064788836952 +Step 13279 | loss:1.0799225568771362 lr:7.066706757089184e-05 tokens_per_second_per_gpu:5298.475170093877 mfu:24.993811390415708 +Step 13280 | loss:1.0265370607376099 lr:7.066311903307032e-05 tokens_per_second_per_gpu:5305.684527559868 mfu:25.027819159626013 +Step 13281 | loss:1.2133980989456177 lr:7.065917033983883e-05 tokens_per_second_per_gpu:3453.8984671125427 mfu:16.292628365214615 +Step 13282 | loss:0.9882715344429016 lr:7.06552214912271e-05 tokens_per_second_per_gpu:4437.941262500403 mfu:20.934526126072832 +Step 13283 | loss:1.0100446939468384 lr:7.065127248726479e-05 tokens_per_second_per_gpu:5316.890061722094 mfu:25.08067757613051 +Step 13284 | loss:1.1521332263946533 lr:7.064732332798165e-05 tokens_per_second_per_gpu:4942.02222348361 mfu:23.312362024111128 +Step 13285 | loss:0.9697527289390564 lr:7.064337401340734e-05 tokens_per_second_per_gpu:5337.221720489449 mfu:25.17658547947481 +Step 13286 | loss:1.1104516983032227 lr:7.063942454357156e-05 tokens_per_second_per_gpu:5330.700312509849 mfu:25.1458228853681 +Step 13287 | loss:1.1054149866104126 lr:7.063547491850408e-05 tokens_per_second_per_gpu:4747.946257382227 mfu:22.396872579236604 +Step 13288 | loss:0.9917324781417847 lr:7.063152513823453e-05 tokens_per_second_per_gpu:4058.050915338671 mfu:19.142518542534102 +Step 13289 | loss:1.1357074975967407 lr:7.062757520279265e-05 tokens_per_second_per_gpu:4996.091071584028 mfu:23.567414005697547 +Step 13290 | loss:1.0192807912826538 lr:7.062362511220815e-05 tokens_per_second_per_gpu:5308.273196866836 mfu:25.040030354419507 +Step 13291 | loss:1.0607703924179077 lr:7.061967486651073e-05 tokens_per_second_per_gpu:4285.00447033559 mfu:20.213097183723967 +Step 13292 | loss:1.1019366979599 lr:7.061572446573011e-05 tokens_per_second_per_gpu:5252.964298554859 mfu:24.77912884441078 +Step 13293 | loss:1.134094476699829 lr:7.061177390989599e-05 tokens_per_second_per_gpu:2933.9229940471387 mfu:13.839815341801197 +Step 13294 | loss:1.067196249961853 lr:7.06078231990381e-05 tokens_per_second_per_gpu:4950.38945246347 mfu:23.3518316708056 +Step 13295 | loss:1.0872876644134521 lr:7.060387233318612e-05 tokens_per_second_per_gpu:5303.259788961281 mfu:25.016381253953412 +Step 13296 | loss:1.020171880722046 lr:7.05999213123698e-05 tokens_per_second_per_gpu:4913.54485623249 mfu:23.178029424047722 +Step 13297 | loss:1.0030772686004639 lr:7.059597013661884e-05 tokens_per_second_per_gpu:4419.167635455189 mfu:20.84596772418949 +Step 13298 | loss:1.2136731147766113 lr:7.059201880596296e-05 tokens_per_second_per_gpu:4389.69035297192 mfu:20.70691835337291 +Step 13299 | loss:1.0982111692428589 lr:7.058806732043189e-05 tokens_per_second_per_gpu:3703.0075487630584 mfu:17.467718405752134 +Step 13300 | loss:0.9872325658798218 lr:7.058411568005532e-05 tokens_per_second_per_gpu:4358.527897835345 mfu:20.55991973563058 +Step 13301 | loss:1.0397181510925293 lr:7.058016388486302e-05 tokens_per_second_per_gpu:5078.4999098523285 mfu:23.95615055620679 +Step 13302 | loss:0.9876658916473389 lr:7.057621193488465e-05 tokens_per_second_per_gpu:4511.67687986212 mfu:21.282349613761642 +Step 13303 | loss:1.046635627746582 lr:7.057225983014998e-05 tokens_per_second_per_gpu:4467.153923608824 mfu:21.072327232715192 +Step 13304 | loss:0.9485180974006653 lr:7.05683075706887e-05 tokens_per_second_per_gpu:4058.6943011648486 mfu:19.145553503249126 +Step 13305 | loss:0.9969161748886108 lr:7.056435515653059e-05 tokens_per_second_per_gpu:5364.723816838967 mfu:25.306317560296982 +Step 13306 | loss:1.0047903060913086 lr:7.056040258770531e-05 tokens_per_second_per_gpu:3434.787742189608 mfu:16.202479815127237 +Step 13307 | loss:1.1381261348724365 lr:7.055644986424264e-05 tokens_per_second_per_gpu:5324.508615503935 mfu:25.11661559417876 +Step 13308 | loss:0.9210423827171326 lr:7.055249698617227e-05 tokens_per_second_per_gpu:4160.151166695142 mfu:19.624142848281764 +Step 13309 | loss:1.0247372388839722 lr:7.054854395352395e-05 tokens_per_second_per_gpu:5203.801816715237 mfu:24.547221029588943 +Step 13310 | loss:1.0915517807006836 lr:7.054459076632743e-05 tokens_per_second_per_gpu:4608.490525449042 mfu:21.739036098106386 +Step 13311 | loss:1.0412780046463013 lr:7.054063742461239e-05 tokens_per_second_per_gpu:4000.5615426916897 mfu:18.871331363059877 +Step 13312 | loss:1.0389763116836548 lr:7.053668392840862e-05 tokens_per_second_per_gpu:5332.118414195822 mfu:25.152512312973933 +Step 13313 | loss:1.118791937828064 lr:7.053273027774581e-05 tokens_per_second_per_gpu:4002.8268988159307 mfu:18.882017434407533 +Step 13314 | loss:1.147538661956787 lr:7.052877647265372e-05 tokens_per_second_per_gpu:5361.427652698573 mfu:25.29076899166315 +Step 13315 | loss:1.0984387397766113 lr:7.052482251316208e-05 tokens_per_second_per_gpu:5314.439803553569 mfu:25.06911929029242 +Step 13316 | loss:1.1897257566452026 lr:7.052086839930065e-05 tokens_per_second_per_gpu:5079.986252271178 mfu:23.963161886992673 +Step 13317 | loss:1.0146169662475586 lr:7.051691413109913e-05 tokens_per_second_per_gpu:4736.413685405343 mfu:22.342471469561964 +Step 13318 | loss:1.041215181350708 lr:7.051295970858729e-05 tokens_per_second_per_gpu:4385.382954154348 mfu:20.6865996182324 +Step 13319 | loss:1.2049517631530762 lr:7.050900513179486e-05 tokens_per_second_per_gpu:5327.217029856884 mfu:25.129391646784764 +Step 13320 | loss:1.2314013242721558 lr:7.05050504007516e-05 tokens_per_second_per_gpu:4190.888030888031 mfu:19.76913388092942 +Step 13321 | loss:1.0870468616485596 lr:7.050109551548722e-05 tokens_per_second_per_gpu:4567.989186575957 mfu:21.54798437240107 +Step 13322 | loss:0.9879761338233948 lr:7.049714047603148e-05 tokens_per_second_per_gpu:5362.670126309116 mfu:25.296629951671942 +Step 13323 | loss:1.0307384729385376 lr:7.049318528241416e-05 tokens_per_second_per_gpu:4028.946607886807 mfu:19.005228558576764 +Step 13324 | loss:0.9893767833709717 lr:7.048922993466495e-05 tokens_per_second_per_gpu:4294.903453016867 mfu:20.259792373972157 +Step 13325 | loss:0.8766083717346191 lr:7.048527443281365e-05 tokens_per_second_per_gpu:4682.383599704038 mfu:22.087602336825707 +Step 13326 | loss:1.0625026226043701 lr:7.048131877688997e-05 tokens_per_second_per_gpu:5242.746316121495 mfu:24.730928877905026 +Step 13327 | loss:1.0921578407287598 lr:7.047736296692367e-05 tokens_per_second_per_gpu:5315.899935349556 mfu:25.076006981098264 +Step 13328 | loss:0.9596568942070007 lr:7.047340700294453e-05 tokens_per_second_per_gpu:5300.523359294241 mfu:25.003473048330946 +Step 13329 | loss:1.0902433395385742 lr:7.046945088498227e-05 tokens_per_second_per_gpu:3668.9762608907554 mfu:17.30718701452204 +Step 13330 | loss:1.0049622058868408 lr:7.046549461306669e-05 tokens_per_second_per_gpu:4569.4092292913665 mfu:21.554682956173924 +Step 13331 | loss:1.1223976612091064 lr:7.046153818722746e-05 tokens_per_second_per_gpu:4183.893267609384 mfu:19.736138389114362 +Step 13332 | loss:0.9205254912376404 lr:7.045758160749443e-05 tokens_per_second_per_gpu:3953.1367900090404 mfu:18.647620713158645 +Step 13333 | loss:0.9804721474647522 lr:7.04536248738973e-05 tokens_per_second_per_gpu:4219.26360584416 mfu:19.902986309369794 +Step 13334 | loss:0.9932006001472473 lr:7.044966798646587e-05 tokens_per_second_per_gpu:4493.170698632446 mfu:21.195052799421166 +Step 13335 | loss:0.9796943664550781 lr:7.044571094522985e-05 tokens_per_second_per_gpu:5371.3863810581 mfu:25.337746012469307 +Step 13336 | loss:1.183104395866394 lr:7.044175375021904e-05 tokens_per_second_per_gpu:4668.194409042721 mfu:22.020669503550714 +Step 13337 | loss:1.1182385683059692 lr:7.04377964014632e-05 tokens_per_second_per_gpu:5011.778340943538 mfu:23.641413531790498 +Step 13338 | loss:1.0131890773773193 lr:7.043383889899208e-05 tokens_per_second_per_gpu:5351.292558732878 mfu:25.242960024201626 +Step 13339 | loss:1.017606496810913 lr:7.042988124283544e-05 tokens_per_second_per_gpu:5342.481715114213 mfu:25.20139777156745 +Step 13340 | loss:0.9784247279167175 lr:7.042592343302307e-05 tokens_per_second_per_gpu:5372.360142703634 mfu:25.34233941229216 +Step 13341 | loss:1.0196146965026855 lr:7.042196546958473e-05 tokens_per_second_per_gpu:5311.806219767149 mfu:25.05669622623615 +Step 13342 | loss:1.1613565683364868 lr:7.041800735255016e-05 tokens_per_second_per_gpu:4310.042553910307 mfu:20.331206095883473 +Step 13343 | loss:1.0040785074234009 lr:7.041404908194915e-05 tokens_per_second_per_gpu:4046.378674482064 mfu:19.08745859092384 +Step 13344 | loss:1.0874214172363281 lr:7.041009065781148e-05 tokens_per_second_per_gpu:5210.612801772075 mfu:24.579349608175928 +Step 13345 | loss:0.9415175914764404 lr:7.040613208016693e-05 tokens_per_second_per_gpu:5163.518662950515 mfu:24.357198520265992 +Step 13346 | loss:1.0905009508132935 lr:7.040217334904524e-05 tokens_per_second_per_gpu:3523.0827605618124 mfu:16.61898248147248 +Step 13347 | loss:0.995329737663269 lr:7.03982144644762e-05 tokens_per_second_per_gpu:4403.97523011587 mfu:20.77430255611228 +Step 13348 | loss:1.0110276937484741 lr:7.03942554264896e-05 tokens_per_second_per_gpu:5320.923106013162 mfu:25.099702134159784 +Step 13349 | loss:1.0307602882385254 lr:7.03902962351152e-05 tokens_per_second_per_gpu:5279.270187223775 mfu:24.90321820950948 +Step 13350 | loss:1.0607887506484985 lr:7.038633689038276e-05 tokens_per_second_per_gpu:4428.897446850776 mfu:20.891864904619478 +Step 13351 | loss:1.1833221912384033 lr:7.03823773923221e-05 tokens_per_second_per_gpu:5328.945471475104 mfu:25.137545000800454 +Step 13352 | loss:1.192849040031433 lr:7.037841774096298e-05 tokens_per_second_per_gpu:4951.020037897287 mfu:23.35480624988588 +Step 13353 | loss:1.2113609313964844 lr:7.037445793633516e-05 tokens_per_second_per_gpu:4260.182576752247 mfu:20.09600807663959 +Step 13354 | loss:1.1212071180343628 lr:7.037049797846846e-05 tokens_per_second_per_gpu:5325.013513218831 mfu:25.11899728284436 +Step 13355 | loss:0.987136721611023 lr:7.036653786739263e-05 tokens_per_second_per_gpu:5320.1346508259485 mfu:25.09598285651737 +Step 13356 | loss:1.145564079284668 lr:7.03625776031375e-05 tokens_per_second_per_gpu:4130.716338605307 mfu:19.485293742082476 +Step 13357 | loss:1.08046293258667 lr:7.03586171857328e-05 tokens_per_second_per_gpu:4317.085599845208 mfu:20.36442934522595 +Step 13358 | loss:0.9483441114425659 lr:7.035465661520834e-05 tokens_per_second_per_gpu:4861.881639972854 mfu:22.93432521829739 +Step 13359 | loss:1.0293653011322021 lr:7.035069589159392e-05 tokens_per_second_per_gpu:4716.463198000158 mfu:22.248361616567554 +Step 13360 | loss:1.0331133604049683 lr:7.034673501491931e-05 tokens_per_second_per_gpu:4877.892263752271 mfu:23.009850062358996 +Step 13361 | loss:0.9908257126808167 lr:7.034277398521431e-05 tokens_per_second_per_gpu:4140.380600714298 mfu:19.53088171534373 +Step 13362 | loss:1.1586767435073853 lr:7.033881280250872e-05 tokens_per_second_per_gpu:3526.2492570382656 mfu:16.633919385611918 +Step 13363 | loss:1.0555636882781982 lr:7.033485146683232e-05 tokens_per_second_per_gpu:4931.229081035959 mfu:23.261448929685564 +Step 13364 | loss:0.9550963044166565 lr:7.033088997821492e-05 tokens_per_second_per_gpu:4213.374718829227 mfu:19.875207424572626 +Step 13365 | loss:1.1810259819030762 lr:7.032692833668628e-05 tokens_per_second_per_gpu:5347.384191476138 mfu:25.224523589015366 +Step 13366 | loss:1.0187926292419434 lr:7.032296654227625e-05 tokens_per_second_per_gpu:4459.501737966004 mfu:21.03623056744051 +Step 13367 | loss:1.1064069271087646 lr:7.031900459501456e-05 tokens_per_second_per_gpu:3669.135879820066 mfu:17.307939964245396 +Step 13368 | loss:1.127878189086914 lr:7.031504249493106e-05 tokens_per_second_per_gpu:3921.9940399286793 mfu:18.500715047528505 +Step 13369 | loss:1.1281911134719849 lr:7.031108024205552e-05 tokens_per_second_per_gpu:4872.535364393941 mfu:22.984580654104473 +Step 13370 | loss:1.1212191581726074 lr:7.030711783641778e-05 tokens_per_second_per_gpu:4815.566069487024 mfu:22.715846770064545 +Step 13371 | loss:1.200342059135437 lr:7.03031552780476e-05 tokens_per_second_per_gpu:5125.587880097129 mfu:24.178272545887676 +Step 13372 | loss:1.1043646335601807 lr:7.029919256697479e-05 tokens_per_second_per_gpu:5367.396738426271 mfu:25.318926187471746 +Step 13373 | loss:1.0066219568252563 lr:7.029522970322917e-05 tokens_per_second_per_gpu:4264.080634587397 mfu:20.114395880525237 +Step 13374 | loss:1.156791090965271 lr:7.029126668684055e-05 tokens_per_second_per_gpu:5202.50631455738 mfu:24.54110992487487 +Step 13375 | loss:0.9229744076728821 lr:7.028730351783871e-05 tokens_per_second_per_gpu:5336.829440347335 mfu:25.174735027114565 +Step 13376 | loss:1.1086899042129517 lr:7.028334019625347e-05 tokens_per_second_per_gpu:4857.121263982331 mfu:22.911869712546192 +Step 13377 | loss:0.935614824295044 lr:7.027937672211465e-05 tokens_per_second_per_gpu:3685.1043919654785 mfu:17.383266160544423 +Step 13378 | loss:1.061630368232727 lr:7.027541309545205e-05 tokens_per_second_per_gpu:4917.344566444747 mfu:23.195953305416353 +Step 13379 | loss:1.0508642196655273 lr:7.027144931629547e-05 tokens_per_second_per_gpu:5360.414550824634 mfu:25.285990017270773 +Step 13380 | loss:1.0220742225646973 lr:7.026748538467474e-05 tokens_per_second_per_gpu:5265.01174496035 mfu:24.835958704611798 +Step 13381 | loss:0.9841227531433105 lr:7.026352130061967e-05 tokens_per_second_per_gpu:3858.13227676663 mfu:18.199468214752773 +Step 13382 | loss:1.0726741552352905 lr:7.025955706416007e-05 tokens_per_second_per_gpu:4576.885777938407 mfu:21.589951111772752 +Step 13383 | loss:0.9728444218635559 lr:7.025559267532574e-05 tokens_per_second_per_gpu:4195.568339626827 mfu:19.791211695793336 +Step 13384 | loss:0.90568608045578 lr:7.025162813414653e-05 tokens_per_second_per_gpu:5336.7637971824415 mfu:25.174425377105884 +Step 13385 | loss:1.188096284866333 lr:7.024766344065224e-05 tokens_per_second_per_gpu:3961.8405434070364 mfu:18.68867780295045 +Step 13386 | loss:1.0452866554260254 lr:7.024369859487268e-05 tokens_per_second_per_gpu:4141.323123860721 mfu:19.53532775783644 +Step 13387 | loss:1.0050934553146362 lr:7.023973359683767e-05 tokens_per_second_per_gpu:5317.088820038059 mfu:25.081615153018575 +Step 13388 | loss:1.157981038093567 lr:7.023576844657708e-05 tokens_per_second_per_gpu:3291.166200601136 mfu:15.524992499092608 +Step 13389 | loss:1.1151975393295288 lr:7.023180314412067e-05 tokens_per_second_per_gpu:5321.831188228026 mfu:25.103985712902098 +Step 13390 | loss:1.2863086462020874 lr:7.022783768949828e-05 tokens_per_second_per_gpu:3865.2236815393367 mfu:18.232919580984046 +Step 13391 | loss:1.0879660844802856 lr:7.022387208273974e-05 tokens_per_second_per_gpu:3851.7200490902137 mfu:18.169220642764262 +Step 13392 | loss:1.0143574476242065 lr:7.021990632387489e-05 tokens_per_second_per_gpu:4013.2210170885924 mfu:18.931048263719127 +Step 13393 | loss:1.0851287841796875 lr:7.021594041293355e-05 tokens_per_second_per_gpu:3956.337867495944 mfu:18.662720741824504 +Step 13394 | loss:1.1975550651550293 lr:7.021197434994552e-05 tokens_per_second_per_gpu:5347.349479442386 mfu:25.224359846429675 +Step 13395 | loss:1.1362682580947876 lr:7.020800813494068e-05 tokens_per_second_per_gpu:3762.3153938009327 mfu:17.747483629757586 +Step 13396 | loss:1.134233832359314 lr:7.020404176794881e-05 tokens_per_second_per_gpu:5076.510918353097 mfu:23.946768143947967 +Step 13397 | loss:1.1332874298095703 lr:7.020007524899977e-05 tokens_per_second_per_gpu:4703.168489747904 mfu:22.185648209430543 +Step 13398 | loss:1.0310574769973755 lr:7.019610857812338e-05 tokens_per_second_per_gpu:5094.91953336091 mfu:24.033604721773415 +Step 13399 | loss:1.1576708555221558 lr:7.019214175534949e-05 tokens_per_second_per_gpu:4160.353125021355 mfu:19.625095520164148 +Step 13400 | loss:1.0854486227035522 lr:7.018817478070793e-05 tokens_per_second_per_gpu:5329.822747547383 mfu:25.141683261712128 +Step 13401 | loss:1.1312882900238037 lr:7.018420765422852e-05 tokens_per_second_per_gpu:4065.034269906067 mfu:19.17546021751166 +Step 13402 | loss:1.1038531064987183 lr:7.018024037594111e-05 tokens_per_second_per_gpu:4680.536442641538 mfu:22.078888981804404 +Step 13403 | loss:1.1105955839157104 lr:7.017627294587554e-05 tokens_per_second_per_gpu:4655.613449468337 mfu:21.961322970705297 +Step 13404 | loss:1.1322718858718872 lr:7.017230536406165e-05 tokens_per_second_per_gpu:3677.9862832627555 mfu:17.3496888273191 +Step 13405 | loss:1.1561086177825928 lr:7.016833763052928e-05 tokens_per_second_per_gpu:4390.813581790631 mfu:20.71221681535337 +Step 13406 | loss:1.0978732109069824 lr:7.016436974530827e-05 tokens_per_second_per_gpu:4917.311745926308 mfu:23.195798485431116 +Step 13407 | loss:1.1340994834899902 lr:7.016040170842845e-05 tokens_per_second_per_gpu:5314.876780344537 mfu:25.071180584371344 +Step 13408 | loss:0.9923089146614075 lr:7.015643351991967e-05 tokens_per_second_per_gpu:5300.424745073974 mfu:25.003007868228618 +Step 13409 | loss:1.025123119354248 lr:7.01524651798118e-05 tokens_per_second_per_gpu:4855.038490791338 mfu:22.902044916046528 +Step 13410 | loss:1.009148359298706 lr:7.014849668813465e-05 tokens_per_second_per_gpu:4362.882155208941 mfu:20.580459510574745 +Step 13411 | loss:1.0571752786636353 lr:7.01445280449181e-05 tokens_per_second_per_gpu:4943.530471901242 mfu:23.31947668923081 +Step 13412 | loss:1.1045904159545898 lr:7.014055925019199e-05 tokens_per_second_per_gpu:5352.309796648908 mfu:25.24775850901022 +Step 13413 | loss:0.9530995488166809 lr:7.013659030398615e-05 tokens_per_second_per_gpu:5327.660590984852 mfu:25.13148400030472 +Step 13414 | loss:1.0580296516418457 lr:7.013262120633045e-05 tokens_per_second_per_gpu:5339.327432382109 mfu:25.186518481743985 +Step 13415 | loss:0.9443415403366089 lr:7.012865195725473e-05 tokens_per_second_per_gpu:5277.201444416173 mfu:24.893459596722103 +Step 13416 | loss:1.1280035972595215 lr:7.012468255678886e-05 tokens_per_second_per_gpu:3438.7182236284148 mfu:16.22102056668337 +Step 13417 | loss:1.0052915811538696 lr:7.012071300496269e-05 tokens_per_second_per_gpu:5081.558476185036 mfu:23.9705783354434 +Step 13418 | loss:1.0939677953720093 lr:7.011674330180607e-05 tokens_per_second_per_gpu:4775.699084394479 mfu:22.527787399373953 +Step 13419 | loss:1.1961597204208374 lr:7.011277344734883e-05 tokens_per_second_per_gpu:4469.209364434363 mfu:21.08202309778363 +Step 13420 | loss:1.1185498237609863 lr:7.010880344162088e-05 tokens_per_second_per_gpu:4275.907146933013 mfu:20.170183557070583 +Step 13421 | loss:1.1541990041732788 lr:7.010483328465205e-05 tokens_per_second_per_gpu:3159.9425934425462 mfu:14.905988355069471 +Step 13422 | loss:1.0608175992965698 lr:7.01008629764722e-05 tokens_per_second_per_gpu:4561.009782235439 mfu:21.515061331317675 +Step 13423 | loss:1.3063220977783203 lr:7.009689251711119e-05 tokens_per_second_per_gpu:4350.171288435416 mfu:20.520500183306527 +Step 13424 | loss:1.1145799160003662 lr:7.009292190659889e-05 tokens_per_second_per_gpu:5349.803808253666 mfu:25.235937334184158 +Step 13425 | loss:0.9972395300865173 lr:7.008895114496517e-05 tokens_per_second_per_gpu:4727.628209269857 mfu:22.301028879674078 +Step 13426 | loss:1.1355862617492676 lr:7.008498023223986e-05 tokens_per_second_per_gpu:3839.437617624737 mfu:18.111282317942706 +Step 13427 | loss:0.970815122127533 lr:7.008100916845288e-05 tokens_per_second_per_gpu:5013.478799379193 mfu:23.64943488435961 +Step 13428 | loss:1.171930193901062 lr:7.007703795363405e-05 tokens_per_second_per_gpu:4147.2774753609 mfu:19.563415449779907 +Step 13429 | loss:1.0681228637695312 lr:7.007306658781326e-05 tokens_per_second_per_gpu:5075.743389042137 mfu:23.943147577234168 +Step 13430 | loss:1.1947416067123413 lr:7.006909507102037e-05 tokens_per_second_per_gpu:5360.530313123048 mfu:25.286536087783478 +Step 13431 | loss:1.0531190633773804 lr:7.006512340328526e-05 tokens_per_second_per_gpu:5360.32751311296 mfu:25.285579445534648 +Step 13432 | loss:0.9713581800460815 lr:7.00611515846378e-05 tokens_per_second_per_gpu:4207.346870607276 mfu:19.846773036052728 +Step 13433 | loss:1.1232342720031738 lr:7.005717961510786e-05 tokens_per_second_per_gpu:4187.890301024909 mfu:19.75499307769962 +Step 13434 | loss:1.1871976852416992 lr:7.005320749472532e-05 tokens_per_second_per_gpu:3139.5295911140606 mfu:14.809696740268672 +Step 13435 | loss:1.03724205493927 lr:7.004923522352003e-05 tokens_per_second_per_gpu:4476.276517947416 mfu:21.115360066685422 +Step 13436 | loss:1.164431095123291 lr:7.004526280152189e-05 tokens_per_second_per_gpu:4000.545448742897 mfu:18.871255445157512 +Step 13437 | loss:1.0862935781478882 lr:7.004129022876077e-05 tokens_per_second_per_gpu:5377.843402117422 mfu:25.36820488248752 +Step 13438 | loss:1.0614478588104248 lr:7.003731750526655e-05 tokens_per_second_per_gpu:4263.083740017599 mfu:20.109693358751745 +Step 13439 | loss:0.9335734248161316 lr:7.003334463106911e-05 tokens_per_second_per_gpu:4104.77952778766 mfu:19.36294538017939 +Step 13440 | loss:1.0266262292861938 lr:7.002937160619833e-05 tokens_per_second_per_gpu:4024.7635081617645 mfu:18.985496163463328 +Step 13441 | loss:0.935695230960846 lr:7.002539843068408e-05 tokens_per_second_per_gpu:4556.4923477630955 mfu:21.49375182215783 +Step 13442 | loss:0.960918128490448 lr:7.002142510455626e-05 tokens_per_second_per_gpu:3535.8715925904485 mfu:16.6793095841513 +Step 13443 | loss:1.1835893392562866 lr:7.001745162784477e-05 tokens_per_second_per_gpu:3793.772609357875 mfu:17.895872682694097 +Step 13444 | loss:1.0000957250595093 lr:7.001347800057942e-05 tokens_per_second_per_gpu:5017.506312908852 mfu:23.66843335284368 +Step 13445 | loss:0.9874255657196045 lr:7.000950422279019e-05 tokens_per_second_per_gpu:3639.489725503815 mfu:17.168093996180133 +Step 13446 | loss:0.9322124123573303 lr:7.000553029450692e-05 tokens_per_second_per_gpu:5325.718000939882 mfu:25.122320471622555 +Step 13447 | loss:1.0763777494430542 lr:7.00015562157595e-05 tokens_per_second_per_gpu:5063.559558133462 mfu:23.885674368062038 +Step 13448 | loss:0.9998568296432495 lr:6.999758198657781e-05 tokens_per_second_per_gpu:4592.054879816633 mfu:21.661506353448004 +Step 13449 | loss:1.0164599418640137 lr:6.999360760699176e-05 tokens_per_second_per_gpu:5386.2782579096465 mfu:25.407993536394244 +Step 13450 | loss:1.1069048643112183 lr:6.998963307703125e-05 tokens_per_second_per_gpu:5327.455509458811 mfu:25.13051659575587 +Step 13451 | loss:1.0122838020324707 lr:6.998565839672615e-05 tokens_per_second_per_gpu:4771.847401171346 mfu:22.509618352445514 +Step 13452 | loss:1.0815086364746094 lr:6.998168356610635e-05 tokens_per_second_per_gpu:5324.61246795243 mfu:25.117105483897365 +Step 13453 | loss:1.0697895288467407 lr:6.997770858520178e-05 tokens_per_second_per_gpu:5051.766469687221 mfu:23.830044357752747 +Step 13454 | loss:1.1049888134002686 lr:6.997373345404231e-05 tokens_per_second_per_gpu:3885.76917199872 mfu:18.3298361649031 +Step 13455 | loss:1.0029164552688599 lr:6.996975817265783e-05 tokens_per_second_per_gpu:4441.820742968462 mfu:20.952826297348288 +Step 13456 | loss:0.8688454031944275 lr:6.996578274107826e-05 tokens_per_second_per_gpu:5312.746283510857 mfu:25.06113066730664 +Step 13457 | loss:1.0856229066848755 lr:6.996180715933349e-05 tokens_per_second_per_gpu:4590.189765212911 mfu:21.65270828964088 +Step 13458 | loss:0.9450793862342834 lr:6.995783142745342e-05 tokens_per_second_per_gpu:5348.07654003631 mfu:25.227789515300543 +Step 13459 | loss:1.015912413597107 lr:6.995385554546797e-05 tokens_per_second_per_gpu:3650.443855585236 mfu:17.21976649674151 +Step 13460 | loss:1.1479079723358154 lr:6.994987951340699e-05 tokens_per_second_per_gpu:5332.058582024322 mfu:25.152230074412547 +Step 13461 | loss:1.0889711380004883 lr:6.994590333130045e-05 tokens_per_second_per_gpu:3831.951896541158 mfu:18.07597089439575 +Step 13462 | loss:1.145523190498352 lr:6.994192699917821e-05 tokens_per_second_per_gpu:5109.3310540622615 mfu:24.101586323780676 +Step 13463 | loss:1.0954755544662476 lr:6.993795051707021e-05 tokens_per_second_per_gpu:5322.538531455783 mfu:25.10732237159263 +Step 13464 | loss:1.1286975145339966 lr:6.993397388500633e-05 tokens_per_second_per_gpu:3767.117594583995 mfu:17.770136430191233 +Step 13465 | loss:1.0562019348144531 lr:6.992999710301649e-05 tokens_per_second_per_gpu:5160.816464920454 mfu:24.344451791115667 +Step 13466 | loss:0.8791375160217285 lr:6.992602017113057e-05 tokens_per_second_per_gpu:4935.3553010146725 mfu:23.28091301333088 +Step 13467 | loss:1.056524634361267 lr:6.992204308937855e-05 tokens_per_second_per_gpu:2870.6702646532785 mfu:13.541441425221885 +Step 13468 | loss:1.0089963674545288 lr:6.991806585779028e-05 tokens_per_second_per_gpu:3721.901022082283 mfu:17.55684214295787 +Step 13469 | loss:1.0376629829406738 lr:6.99140884763957e-05 tokens_per_second_per_gpu:4914.779558173887 mfu:23.183853723767168 +Step 13470 | loss:1.0287724733352661 lr:6.991011094522474e-05 tokens_per_second_per_gpu:4140.862087791931 mfu:19.533152972039094 +Step 13471 | loss:0.9970017075538635 lr:6.990613326430727e-05 tokens_per_second_per_gpu:4687.263164301688 mfu:22.110620075572392 +Step 13472 | loss:1.0217951536178589 lr:6.990215543367324e-05 tokens_per_second_per_gpu:4298.964691966772 mfu:20.278949931017674 +Step 13473 | loss:1.0154772996902466 lr:6.989817745335255e-05 tokens_per_second_per_gpu:5371.0885099151155 mfu:25.336340903465793 +Step 13474 | loss:0.9858899116516113 lr:6.989419932337514e-05 tokens_per_second_per_gpu:4190.6168647385075 mfu:19.767854743936404 +Step 13475 | loss:0.9471259117126465 lr:6.989022104377092e-05 tokens_per_second_per_gpu:4868.159140516998 mfu:22.963937259416152 +Step 13476 | loss:1.0571051836013794 lr:6.98862426145698e-05 tokens_per_second_per_gpu:4275.542509626525 mfu:20.168463500682343 +Step 13477 | loss:1.0487339496612549 lr:6.988226403580172e-05 tokens_per_second_per_gpu:4786.804836093778 mfu:22.58017512497618 +Step 13478 | loss:1.0461647510528564 lr:6.987828530749658e-05 tokens_per_second_per_gpu:5319.4534292522985 mfu:25.09276941812608 +Step 13479 | loss:1.0568214654922485 lr:6.987430642968435e-05 tokens_per_second_per_gpu:5313.405346028306 mfu:25.0642395776491 +Step 13480 | loss:1.128953218460083 lr:6.98703274023949e-05 tokens_per_second_per_gpu:5365.541163219497 mfu:25.310173122627617 +Step 13481 | loss:1.039570927619934 lr:6.986634822565819e-05 tokens_per_second_per_gpu:4852.695133857222 mfu:22.890990901570383 +Step 13482 | loss:1.0997709035873413 lr:6.986236889950414e-05 tokens_per_second_per_gpu:5320.15378699221 mfu:25.096073125078597 +Step 13483 | loss:1.016404628753662 lr:6.985838942396268e-05 tokens_per_second_per_gpu:3967.618045646622 mfu:18.715931266758105 +Step 13484 | loss:1.1731698513031006 lr:6.985440979906373e-05 tokens_per_second_per_gpu:5400.228364996018 mfu:25.473798571654356 +Step 13485 | loss:1.0496127605438232 lr:6.985043002483725e-05 tokens_per_second_per_gpu:4552.553816361042 mfu:21.47517309755159 +Step 13486 | loss:1.0177327394485474 lr:6.984645010131314e-05 tokens_per_second_per_gpu:5337.204566337484 mfu:25.17650456041319 +Step 13487 | loss:1.0828813314437866 lr:6.984247002852134e-05 tokens_per_second_per_gpu:4489.125804647755 mfu:21.1759723443653 +Step 13488 | loss:1.038949728012085 lr:6.98384898064918e-05 tokens_per_second_per_gpu:4489.517477605387 mfu:21.17781993253275 +Step 13489 | loss:1.013852596282959 lr:6.983450943525445e-05 tokens_per_second_per_gpu:4927.0064068099455 mfu:23.241529854898943 +Step 13490 | loss:0.9851270914077759 lr:6.983052891483921e-05 tokens_per_second_per_gpu:5316.419695387181 mfu:25.078458777875948 +Step 13491 | loss:1.0967899560928345 lr:6.982654824527603e-05 tokens_per_second_per_gpu:4167.868920021818 mfu:19.66054880750819 +Step 13492 | loss:1.0092964172363281 lr:6.982256742659487e-05 tokens_per_second_per_gpu:5325.022097886384 mfu:25.119037778185902 +Step 13493 | loss:1.107858657836914 lr:6.981858645882564e-05 tokens_per_second_per_gpu:3409.422687470124 mfu:16.082828524290832 +Step 13494 | loss:0.9991598725318909 lr:6.98146053419983e-05 tokens_per_second_per_gpu:5074.493723321308 mfu:23.93725269081406 +Step 13495 | loss:1.0038186311721802 lr:6.981062407614278e-05 tokens_per_second_per_gpu:5325.056435221525 mfu:25.119199753254318 +Step 13496 | loss:1.096165418624878 lr:6.980664266128903e-05 tokens_per_second_per_gpu:3971.5186690235737 mfu:18.73433117274216 +Step 13497 | loss:1.0322952270507812 lr:6.980266109746699e-05 tokens_per_second_per_gpu:5330.190177306677 mfu:25.143416490576694 +Step 13498 | loss:1.1170533895492554 lr:6.97986793847066e-05 tokens_per_second_per_gpu:4381.180390426491 mfu:20.66677540809679 +Step 13499 | loss:0.999323308467865 lr:6.979469752303783e-05 tokens_per_second_per_gpu:4355.1257775198 mfu:20.543871353640576 +Step 13500 | loss:0.9100920557975769 lr:6.979071551249061e-05 tokens_per_second_per_gpu:5363.085577467717 mfu:25.298589705670338 +Step 13501 | loss:0.9426431059837341 lr:6.97867333530949e-05 tokens_per_second_per_gpu:5329.104194870167 mfu:25.13829372613597 +Step 13502 | loss:0.972232460975647 lr:6.978275104488063e-05 tokens_per_second_per_gpu:3878.5106054353196 mfu:18.295596267984376 +Step 13503 | loss:1.2436822652816772 lr:6.977876858787776e-05 tokens_per_second_per_gpu:5364.295986311742 mfu:25.304299410704647 +Step 13504 | loss:0.966327965259552 lr:6.977478598211626e-05 tokens_per_second_per_gpu:5336.025702026955 mfu:25.17094365632543 +Step 13505 | loss:1.095752477645874 lr:6.977080322762609e-05 tokens_per_second_per_gpu:5329.019857388606 mfu:25.13789589184599 +Step 13506 | loss:0.9945278763771057 lr:6.976682032443716e-05 tokens_per_second_per_gpu:4182.29811614161 mfu:19.728613787479972 +Step 13507 | loss:1.205604910850525 lr:6.976283727257945e-05 tokens_per_second_per_gpu:5321.271386923001 mfu:25.101345034634345 +Step 13508 | loss:0.8945350050926208 lr:6.975885407208293e-05 tokens_per_second_per_gpu:5357.714923184079 mfu:25.27325541308631 +Step 13509 | loss:1.0361735820770264 lr:6.975487072297755e-05 tokens_per_second_per_gpu:5355.193603676339 mfu:25.261361918786832 +Step 13510 | loss:1.0644584894180298 lr:6.975088722529326e-05 tokens_per_second_per_gpu:4727.602516830692 mfu:22.300907684054955 +Step 13511 | loss:0.9304081201553345 lr:6.974690357906002e-05 tokens_per_second_per_gpu:4981.373751731295 mfu:23.497989896918504 +Step 13512 | loss:1.0929888486862183 lr:6.974291978430782e-05 tokens_per_second_per_gpu:4995.787442702968 mfu:23.565981736461158 +Step 13513 | loss:0.9555384516716003 lr:6.973893584106659e-05 tokens_per_second_per_gpu:4218.152155428291 mfu:19.897743408125145 +Step 13514 | loss:1.0813273191452026 lr:6.97349517493663e-05 tokens_per_second_per_gpu:4959.781416585677 mfu:23.396135168004363 +Step 13515 | loss:1.0988361835479736 lr:6.973096750923694e-05 tokens_per_second_per_gpu:4310.563488135642 mfu:20.333663431504924 +Step 13516 | loss:0.9958725571632385 lr:6.972698312070842e-05 tokens_per_second_per_gpu:5376.689242275401 mfu:25.362760513592736 +Step 13517 | loss:1.1029781103134155 lr:6.972299858381075e-05 tokens_per_second_per_gpu:5366.109329443206 mfu:25.312853259644005 +Step 13518 | loss:1.0280311107635498 lr:6.971901389857392e-05 tokens_per_second_per_gpu:3516.2139647539675 mfu:16.586581199709464 +Step 13519 | loss:0.9921611547470093 lr:6.971502906502784e-05 tokens_per_second_per_gpu:4245.006046737596 mfu:20.024417795177452 +Step 13520 | loss:0.9256665706634521 lr:6.971104408320253e-05 tokens_per_second_per_gpu:5356.262612637591 mfu:25.266404616448792 +Step 13521 | loss:1.1861484050750732 lr:6.970705895312793e-05 tokens_per_second_per_gpu:5012.853814849984 mfu:23.646486725701973 +Step 13522 | loss:0.943540632724762 lr:6.970307367483404e-05 tokens_per_second_per_gpu:5341.736793738084 mfu:25.197883850339554 +Step 13523 | loss:0.9814172983169556 lr:6.96990882483508e-05 tokens_per_second_per_gpu:4698.915997973162 mfu:22.165588480179174 +Step 13524 | loss:0.9391011595726013 lr:6.969510267370823e-05 tokens_per_second_per_gpu:4274.767268316944 mfu:20.16480655515524 +Step 13525 | loss:0.9929628372192383 lr:6.969111695093625e-05 tokens_per_second_per_gpu:4258.374198240105 mfu:20.087477646656627 +Step 13526 | loss:1.2668700218200684 lr:6.968713108006489e-05 tokens_per_second_per_gpu:5334.203130484754 mfu:25.162346275398104 +Step 13527 | loss:0.9652963280677795 lr:6.968314506112409e-05 tokens_per_second_per_gpu:4062.2742434340507 mfu:19.162440701734088 +Step 13528 | loss:0.86989426612854 lr:6.967915889414385e-05 tokens_per_second_per_gpu:5054.236931669247 mfu:23.84169794842624 +Step 13529 | loss:1.0493136644363403 lr:6.967517257915415e-05 tokens_per_second_per_gpu:4967.026606972403 mfu:23.4303119672155 +Step 13530 | loss:1.0099211931228638 lr:6.967118611618495e-05 tokens_per_second_per_gpu:3873.15525183022 mfu:18.27033415131158 +Step 13531 | loss:0.9904293417930603 lr:6.966719950526627e-05 tokens_per_second_per_gpu:5063.029513148936 mfu:23.883174055435028 +Step 13532 | loss:1.2633448839187622 lr:6.966321274642806e-05 tokens_per_second_per_gpu:5322.255654242195 mfu:25.105987991512308 +Step 13533 | loss:1.2012537717819214 lr:6.965922583970033e-05 tokens_per_second_per_gpu:5062.836001426079 mfu:23.882261227621807 +Step 13534 | loss:1.0027817487716675 lr:6.965523878511303e-05 tokens_per_second_per_gpu:4676.700915658469 mfu:22.06079614661688 +Step 13535 | loss:1.2678136825561523 lr:6.965125158269619e-05 tokens_per_second_per_gpu:4698.716586320185 mfu:22.164647821388947 +Step 13536 | loss:1.249341607093811 lr:6.964726423247977e-05 tokens_per_second_per_gpu:5370.377808224182 mfu:25.33298840233145 +Step 13537 | loss:1.1424521207809448 lr:6.964327673449378e-05 tokens_per_second_per_gpu:5018.677292716875 mfu:23.673957064387878 +Step 13538 | loss:0.9202658534049988 lr:6.963928908876818e-05 tokens_per_second_per_gpu:4309.5461072050275 mfu:20.328864272072128 +Step 13539 | loss:1.007543683052063 lr:6.9635301295333e-05 tokens_per_second_per_gpu:5368.418129434959 mfu:25.323744263127 +Step 13540 | loss:0.9181068539619446 lr:6.963131335421821e-05 tokens_per_second_per_gpu:3948.2481091122095 mfu:18.62455997127332 +Step 13541 | loss:1.138869285583496 lr:6.96273252654538e-05 tokens_per_second_per_gpu:5358.784891761272 mfu:25.27830263742039 +Step 13542 | loss:0.8769727945327759 lr:6.962333702906978e-05 tokens_per_second_per_gpu:5338.97545198189 mfu:25.1848581301421 +Step 13543 | loss:0.9789108037948608 lr:6.961934864509612e-05 tokens_per_second_per_gpu:5354.910987769451 mfu:25.260028771334795 +Step 13544 | loss:1.1048307418823242 lr:6.961536011356286e-05 tokens_per_second_per_gpu:3684.3985090417186 mfu:17.379936390357003 +Step 13545 | loss:1.0291478633880615 lr:6.961137143449996e-05 tokens_per_second_per_gpu:5288.008182487448 mfu:24.94443682402395 +Step 13546 | loss:1.0446925163269043 lr:6.960738260793744e-05 tokens_per_second_per_gpu:4405.063735681953 mfu:20.7794372225857 +Step 13547 | loss:1.0146845579147339 lr:6.960339363390529e-05 tokens_per_second_per_gpu:4540.757316513926 mfu:21.41952699508233 +Step 13548 | loss:1.159379482269287 lr:6.959940451243351e-05 tokens_per_second_per_gpu:3446.655438930988 mfu:16.258461765494182 +Step 13549 | loss:1.0730688571929932 lr:6.959541524355214e-05 tokens_per_second_per_gpu:4648.687662304058 mfu:21.928652850990254 +Step 13550 | loss:0.9866566061973572 lr:6.95914258272911e-05 tokens_per_second_per_gpu:4633.669567581861 mfu:21.857809935834467 +Step 13551 | loss:0.9878562092781067 lr:6.958743626368048e-05 tokens_per_second_per_gpu:5383.937044052137 mfu:25.396949631176096 +Step 13552 | loss:1.0757193565368652 lr:6.958344655275024e-05 tokens_per_second_per_gpu:3742.1940142314265 mfu:17.652567649266867 +Step 13553 | loss:0.9501335024833679 lr:6.957945669453041e-05 tokens_per_second_per_gpu:4365.923046587196 mfu:20.59480391403554 +Step 13554 | loss:1.1215037107467651 lr:6.957546668905097e-05 tokens_per_second_per_gpu:3526.352669914507 mfu:16.634407201793202 +Step 13555 | loss:1.0868885517120361 lr:6.957147653634198e-05 tokens_per_second_per_gpu:3568.496773271469 mfu:16.83320812785349 +Step 13556 | loss:0.9866301417350769 lr:6.956748623643338e-05 tokens_per_second_per_gpu:5373.694028052285 mfu:25.348631577066055 +Step 13557 | loss:1.1171419620513916 lr:6.956349578935523e-05 tokens_per_second_per_gpu:3534.2042436558777 mfu:16.671444414747867 +Step 13558 | loss:0.9235004782676697 lr:6.955950519513754e-05 tokens_per_second_per_gpu:4709.798076212454 mfu:22.216921099907918 +Step 13559 | loss:1.0854389667510986 lr:6.955551445381032e-05 tokens_per_second_per_gpu:5324.951258282937 mfu:25.118703615689956 +Step 13560 | loss:1.0479637384414673 lr:6.955152356540358e-05 tokens_per_second_per_gpu:3289.601870980358 mfu:15.517613289369244 +Step 13561 | loss:0.9980976581573486 lr:6.954753252994731e-05 tokens_per_second_per_gpu:5305.551666348556 mfu:25.027192430623792 +Step 13562 | loss:0.977190375328064 lr:6.954354134747159e-05 tokens_per_second_per_gpu:5309.839483190265 mfu:25.047418794243125 +Step 13563 | loss:1.0969642400741577 lr:6.953955001800637e-05 tokens_per_second_per_gpu:5314.7242368813695 mfu:25.070461010828986 +Step 13564 | loss:1.2117936611175537 lr:6.953555854158172e-05 tokens_per_second_per_gpu:3915.4645138838673 mfu:18.46991415912311 +Step 13565 | loss:0.9878228902816772 lr:6.953156691822764e-05 tokens_per_second_per_gpu:4186.830977162694 mfu:19.74999606630226 +Step 13566 | loss:1.0731219053268433 lr:6.952757514797416e-05 tokens_per_second_per_gpu:4331.9709869322 mfu:20.434646255824216 +Step 13567 | loss:0.8923670053482056 lr:6.952358323085129e-05 tokens_per_second_per_gpu:4754.821551994609 mfu:22.42930451696943 +Step 13568 | loss:1.0858289003372192 lr:6.951959116688904e-05 tokens_per_second_per_gpu:4005.652180127559 mfu:18.89534476839718 +Step 13569 | loss:0.9590014219284058 lr:6.951559895611748e-05 tokens_per_second_per_gpu:4115.289682894643 mfu:19.4125235750364 +Step 13570 | loss:1.043076992034912 lr:6.951160659856659e-05 tokens_per_second_per_gpu:3668.404270482777 mfu:17.30448883817598 +Step 13571 | loss:1.1168142557144165 lr:6.950761409426643e-05 tokens_per_second_per_gpu:3792.717147930916 mfu:17.890893891062845 +Step 13572 | loss:0.9971835613250732 lr:6.9503621443247e-05 tokens_per_second_per_gpu:5347.947385509318 mfu:25.227180271360687 +Step 13573 | loss:1.0569298267364502 lr:6.949962864553837e-05 tokens_per_second_per_gpu:3703.408675884156 mfu:17.469610590821695 +Step 13574 | loss:0.9844325184822083 lr:6.949563570117052e-05 tokens_per_second_per_gpu:5320.936964278646 mfu:25.099767505962536 +Step 13575 | loss:1.0441426038742065 lr:6.949164261017352e-05 tokens_per_second_per_gpu:4561.552526707612 mfu:21.517621549594665 +Step 13576 | loss:1.0728899240493774 lr:6.948764937257739e-05 tokens_per_second_per_gpu:5325.252682161589 mfu:25.120125483553778 +Step 13577 | loss:1.0654460191726685 lr:6.948365598841215e-05 tokens_per_second_per_gpu:5095.342635854331 mfu:24.035600568423614 +Step 13578 | loss:1.0508041381835938 lr:6.947966245770786e-05 tokens_per_second_per_gpu:4517.5190475360605 mfu:21.30990811545621 +Step 13579 | loss:1.0264537334442139 lr:6.947566878049455e-05 tokens_per_second_per_gpu:4035.8589809219984 mfu:19.037835401556215 +Step 13580 | loss:1.1602712869644165 lr:6.947167495680224e-05 tokens_per_second_per_gpu:4149.894416233939 mfu:19.57576000637419 +Step 13581 | loss:1.0254788398742676 lr:6.946768098666096e-05 tokens_per_second_per_gpu:4308.409393732836 mfu:20.323502200680487 +Step 13582 | loss:1.1574864387512207 lr:6.946368687010079e-05 tokens_per_second_per_gpu:3792.6243824160874 mfu:17.89045630030197 +Step 13583 | loss:1.0141916275024414 lr:6.945969260715174e-05 tokens_per_second_per_gpu:3700.4629330943985 mfu:17.455715019486867 +Step 13584 | loss:1.0056843757629395 lr:6.945569819784386e-05 tokens_per_second_per_gpu:4739.100480494825 mfu:22.355145540413567 +Step 13585 | loss:1.0072036981582642 lr:6.94517036422072e-05 tokens_per_second_per_gpu:4434.361382326071 mfu:20.917639220499968 +Step 13586 | loss:1.0814794301986694 lr:6.944770894027178e-05 tokens_per_second_per_gpu:4052.592469579413 mfu:19.116770122580803 +Step 13587 | loss:0.9057912826538086 lr:6.944371409206767e-05 tokens_per_second_per_gpu:5286.400016176424 mfu:24.936850829153382 +Step 13588 | loss:0.9916778206825256 lr:6.94397190976249e-05 tokens_per_second_per_gpu:5333.605219538691 mfu:25.159525827451397 +Step 13589 | loss:0.9326134324073792 lr:6.943572395697351e-05 tokens_per_second_per_gpu:5355.445597696007 mfu:25.26255061757222 +Step 13590 | loss:1.0001126527786255 lr:6.943172867014359e-05 tokens_per_second_per_gpu:5323.528210333936 mfu:25.111990855717504 +Step 13591 | loss:0.9528810977935791 lr:6.942773323716514e-05 tokens_per_second_per_gpu:5323.856598114011 mfu:25.113539916904852 +Step 13592 | loss:0.9501099586486816 lr:6.942373765806822e-05 tokens_per_second_per_gpu:5329.061245853937 mfu:25.138091128298555 +Step 13593 | loss:1.0756953954696655 lr:6.941974193288291e-05 tokens_per_second_per_gpu:4488.564089579708 mfu:21.173322638550847 +Step 13594 | loss:1.0789437294006348 lr:6.941574606163924e-05 tokens_per_second_per_gpu:3091.276997801305 mfu:14.58208102487105 +Step 13595 | loss:1.0959901809692383 lr:6.941175004436726e-05 tokens_per_second_per_gpu:5365.98247775132 mfu:25.312254878570247 +Step 13596 | loss:1.0852630138397217 lr:6.940775388109704e-05 tokens_per_second_per_gpu:5314.3717544679375 mfu:25.068798291145125 +Step 13597 | loss:1.0823853015899658 lr:6.94037575718586e-05 tokens_per_second_per_gpu:4550.283518327188 mfu:21.464463714373117 +Step 13598 | loss:1.119207501411438 lr:6.939976111668205e-05 tokens_per_second_per_gpu:4261.478114678487 mfu:20.102119350079228 +Step 13599 | loss:1.0681813955307007 lr:6.939576451559742e-05 tokens_per_second_per_gpu:5356.625869672788 mfu:25.26811816186186 +Step 13600 | loss:1.032585620880127 lr:6.939176776863476e-05 tokens_per_second_per_gpu:4976.502309039977 mfu:23.47501047058973 +Step 13601 | loss:0.9613441228866577 lr:6.938777087582414e-05 tokens_per_second_per_gpu:4054.069886511931 mfu:19.12373935032481 +Step 13602 | loss:1.018181562423706 lr:6.938377383719563e-05 tokens_per_second_per_gpu:3603.3461337466656 mfu:16.997598507129013 +Step 13603 | loss:0.9409218430519104 lr:6.937977665277928e-05 tokens_per_second_per_gpu:5348.844131876661 mfu:25.23141037697615 +Step 13604 | loss:0.9609875679016113 lr:6.937577932260515e-05 tokens_per_second_per_gpu:5347.845176400239 mfu:25.226698133554493 +Step 13605 | loss:1.0712734460830688 lr:6.937178184670333e-05 tokens_per_second_per_gpu:3560.968200308431 mfu:16.797694564679837 +Step 13606 | loss:1.0307083129882812 lr:6.936778422510384e-05 tokens_per_second_per_gpu:5248.436393975079 mfu:24.757769945967777 +Step 13607 | loss:0.9946451187133789 lr:6.936378645783678e-05 tokens_per_second_per_gpu:3992.1200098653185 mfu:18.83151121244387 +Step 13608 | loss:1.0282537937164307 lr:6.935978854493223e-05 tokens_per_second_per_gpu:5323.615707433281 mfu:25.112403594463665 +Step 13609 | loss:0.9467650055885315 lr:6.935579048642023e-05 tokens_per_second_per_gpu:4213.35395160227 mfu:19.875109462019815 +Step 13610 | loss:1.0456031560897827 lr:6.935179228233085e-05 tokens_per_second_per_gpu:3556.4516252120075 mfu:16.776389109342833 +Step 13611 | loss:1.0032833814620972 lr:6.934779393269419e-05 tokens_per_second_per_gpu:3867.811354417189 mfu:18.24512607545139 +Step 13612 | loss:1.1972676515579224 lr:6.93437954375403e-05 tokens_per_second_per_gpu:4792.746223084522 mfu:22.608201661116667 +Step 13613 | loss:1.0203508138656616 lr:6.933979679689923e-05 tokens_per_second_per_gpu:5327.066886389298 mfu:25.128683394431043 +Step 13614 | loss:1.1257416009902954 lr:6.933579801080112e-05 tokens_per_second_per_gpu:5316.493381761188 mfu:25.078806369074332 +Step 13615 | loss:1.1466730833053589 lr:6.933179907927598e-05 tokens_per_second_per_gpu:5362.444430547513 mfu:25.29556530625679 +Step 13616 | loss:1.076411485671997 lr:6.932780000235392e-05 tokens_per_second_per_gpu:3732.2134928616447 mfu:17.605487827113066 +Step 13617 | loss:1.0860978364944458 lr:6.932380078006499e-05 tokens_per_second_per_gpu:4896.490066364983 mfu:23.097579070395486 +Step 13618 | loss:1.1407058238983154 lr:6.931980141243932e-05 tokens_per_second_per_gpu:5218.552335681747 mfu:24.616801744252015 +Step 13619 | loss:1.1074655055999756 lr:6.931580189950694e-05 tokens_per_second_per_gpu:4168.254306116355 mfu:19.66236673946977 +Step 13620 | loss:1.121376872062683 lr:6.931180224129796e-05 tokens_per_second_per_gpu:5328.675264736567 mfu:25.136270389512756 +Step 13621 | loss:1.0503673553466797 lr:6.930780243784246e-05 tokens_per_second_per_gpu:4003.8449573662947 mfu:18.8868197902883 +Step 13622 | loss:1.1884829998016357 lr:6.930380248917048e-05 tokens_per_second_per_gpu:5333.717459330277 mfu:25.16005528169839 +Step 13623 | loss:0.9741772413253784 lr:6.929980239531217e-05 tokens_per_second_per_gpu:5315.111938667786 mfu:25.072289866304256 +Step 13624 | loss:1.1133795976638794 lr:6.929580215629757e-05 tokens_per_second_per_gpu:4559.423184058844 mfu:21.507577077017107 +Step 13625 | loss:1.0994211435317993 lr:6.929180177215678e-05 tokens_per_second_per_gpu:4285.341827310382 mfu:20.214688554133097 +Step 13626 | loss:1.0909607410430908 lr:6.928780124291987e-05 tokens_per_second_per_gpu:4104.258836169261 mfu:19.360489188976246 +Step 13627 | loss:1.1116943359375 lr:6.928380056861698e-05 tokens_per_second_per_gpu:3678.963330886534 mfu:17.35429772766198 +Step 13628 | loss:1.0426369905471802 lr:6.927979974927815e-05 tokens_per_second_per_gpu:4068.9105712049486 mfu:19.193745392103686 +Step 13629 | loss:1.065543532371521 lr:6.927579878493348e-05 tokens_per_second_per_gpu:5039.350959050891 mfu:23.771478275776634 +Step 13630 | loss:1.0852056741714478 lr:6.927179767561307e-05 tokens_per_second_per_gpu:4629.876464221305 mfu:21.839917220112554 +Step 13631 | loss:0.9850468635559082 lr:6.926779642134702e-05 tokens_per_second_per_gpu:4535.25824405781 mfu:21.393586932068324 +Step 13632 | loss:1.0002607107162476 lr:6.926379502216541e-05 tokens_per_second_per_gpu:4529.293822227469 mfu:21.365451736659303 +Step 13633 | loss:1.1796371936798096 lr:6.925979347809833e-05 tokens_per_second_per_gpu:4548.64054422701 mfu:21.456713525222288 +Step 13634 | loss:0.999398946762085 lr:6.92557917891759e-05 tokens_per_second_per_gpu:4429.713567711452 mfu:20.895714685963863 +Step 13635 | loss:0.995202362537384 lr:6.92517899554282e-05 tokens_per_second_per_gpu:4016.2164472824766 mfu:18.945178219016768 +Step 13636 | loss:1.1372846364974976 lr:6.924778797688532e-05 tokens_per_second_per_gpu:5338.818727985215 mfu:25.184118836309917 +Step 13637 | loss:1.0377953052520752 lr:6.924378585357738e-05 tokens_per_second_per_gpu:4264.825860053791 mfu:20.117911236198793 +Step 13638 | loss:0.9416550993919373 lr:6.923978358553446e-05 tokens_per_second_per_gpu:5256.381689324029 mfu:24.795249259735947 +Step 13639 | loss:1.1285897493362427 lr:6.923578117278669e-05 tokens_per_second_per_gpu:4496.454353431667 mfu:21.21054235935837 +Step 13640 | loss:1.0324946641921997 lr:6.923177861536413e-05 tokens_per_second_per_gpu:5216.269964243651 mfu:24.606035408765795 +Step 13641 | loss:0.985750138759613 lr:6.922777591329694e-05 tokens_per_second_per_gpu:3534.010425118236 mfu:16.67053013963696 +Step 13642 | loss:0.9873989820480347 lr:6.922377306661518e-05 tokens_per_second_per_gpu:5336.013033960903 mfu:25.170883898896474 +Step 13643 | loss:0.9460760951042175 lr:6.921977007534896e-05 tokens_per_second_per_gpu:4331.451368135768 mfu:20.432195125305334 +Step 13644 | loss:1.116188406944275 lr:6.92157669395284e-05 tokens_per_second_per_gpu:4251.518359658193 mfu:20.055137486339067 +Step 13645 | loss:1.0338993072509766 lr:6.921176365918363e-05 tokens_per_second_per_gpu:5311.084766048533 mfu:25.053293005952746 +Step 13646 | loss:1.0877275466918945 lr:6.920776023434472e-05 tokens_per_second_per_gpu:3983.236623091851 mfu:18.78960676137135 +Step 13647 | loss:1.0245184898376465 lr:6.920375666504179e-05 tokens_per_second_per_gpu:5354.687305557748 mfu:25.258973624178072 +Step 13648 | loss:1.1071274280548096 lr:6.919975295130496e-05 tokens_per_second_per_gpu:4172.936376221825 mfu:19.68445286300081 +Step 13649 | loss:1.0816538333892822 lr:6.919574909316433e-05 tokens_per_second_per_gpu:4950.627132780792 mfu:23.352952849414848 +Step 13650 | loss:1.0084426403045654 lr:6.919174509065004e-05 tokens_per_second_per_gpu:5370.836014004632 mfu:25.335149837175795 +Step 13651 | loss:1.03342866897583 lr:6.918774094379215e-05 tokens_per_second_per_gpu:3795.554380235873 mfu:17.904277599927077 +Step 13652 | loss:0.9950829148292542 lr:6.918373665262084e-05 tokens_per_second_per_gpu:4123.209122462891 mfu:19.449880922675547 +Step 13653 | loss:0.9404895305633545 lr:6.917973221716618e-05 tokens_per_second_per_gpu:5256.898977551099 mfu:24.79768939656151 +Step 13654 | loss:1.0092144012451172 lr:6.917572763745832e-05 tokens_per_second_per_gpu:4368.355242520153 mfu:20.606276997226722 +Step 13655 | loss:1.0882148742675781 lr:6.917172291352737e-05 tokens_per_second_per_gpu:5352.082712049082 mfu:25.24668731220822 +Step 13656 | loss:1.1081821918487549 lr:6.916771804540345e-05 tokens_per_second_per_gpu:3990.8834839026304 mfu:18.825678308504784 +Step 13657 | loss:0.9440582394599915 lr:6.916371303311666e-05 tokens_per_second_per_gpu:3230.9015240980552 mfu:15.240713737813497 +Step 13658 | loss:0.9944558143615723 lr:6.915970787669714e-05 tokens_per_second_per_gpu:5064.001456296471 mfu:23.88775887709255 +Step 13659 | loss:0.9569252729415894 lr:6.915570257617502e-05 tokens_per_second_per_gpu:5019.027244275309 mfu:23.67560784559691 +Step 13660 | loss:1.1355907917022705 lr:6.915169713158041e-05 tokens_per_second_per_gpu:5270.994015011571 mfu:24.86417809312393 +Step 13661 | loss:1.0796042680740356 lr:6.914769154294345e-05 tokens_per_second_per_gpu:5355.431873742971 mfu:25.2624858793442 +Step 13662 | loss:1.0967243909835815 lr:6.914368581029427e-05 tokens_per_second_per_gpu:5369.407203565146 mfu:25.328409894552543 +Step 13663 | loss:0.9872713088989258 lr:6.913967993366296e-05 tokens_per_second_per_gpu:5126.447107829821 mfu:24.182325669702156 +Step 13664 | loss:1.014231562614441 lr:6.913567391307969e-05 tokens_per_second_per_gpu:5054.040889716011 mfu:23.840773185084625 +Step 13665 | loss:0.9723781943321228 lr:6.913166774857457e-05 tokens_per_second_per_gpu:5322.861607991263 mfu:25.108846378733027 +Step 13666 | loss:1.0716588497161865 lr:6.912766144017774e-05 tokens_per_second_per_gpu:4774.952524800188 mfu:22.524265750391482 +Step 13667 | loss:0.9453214406967163 lr:6.912365498791933e-05 tokens_per_second_per_gpu:5368.638597636781 mfu:25.32478424925075 +Step 13668 | loss:1.068402886390686 lr:6.911964839182946e-05 tokens_per_second_per_gpu:3306.1800019555694 mfu:15.595815161700127 +Step 13669 | loss:1.077663779258728 lr:6.911564165193827e-05 tokens_per_second_per_gpu:5346.451737645937 mfu:25.220125045205283 +Step 13670 | loss:0.9416334629058838 lr:6.911163476827593e-05 tokens_per_second_per_gpu:5325.117108117031 mfu:25.119485957654284 +Step 13671 | loss:1.0896363258361816 lr:6.910762774087252e-05 tokens_per_second_per_gpu:4800.7381380860925 mfu:22.645900887740382 +Step 13672 | loss:1.111121654510498 lr:6.910362056975821e-05 tokens_per_second_per_gpu:5300.753880694525 mfu:25.004560457107086 +Step 13673 | loss:0.9848970174789429 lr:6.909961325496312e-05 tokens_per_second_per_gpu:4245.615106081739 mfu:20.02729083202001 +Step 13674 | loss:1.1143560409545898 lr:6.909560579651743e-05 tokens_per_second_per_gpu:4885.908245483779 mfu:23.047662815854036 +Step 13675 | loss:1.0289876461029053 lr:6.909159819445123e-05 tokens_per_second_per_gpu:5142.663159726244 mfu:24.258819553241334 +Step 13676 | loss:1.102473258972168 lr:6.908759044879469e-05 tokens_per_second_per_gpu:5149.341329465937 mfu:24.290321619316718 +Step 13677 | loss:0.9282329678535461 lr:6.908358255957795e-05 tokens_per_second_per_gpu:3857.9536708164587 mfu:18.198625700012496 +Step 13678 | loss:0.9935975074768066 lr:6.907957452683113e-05 tokens_per_second_per_gpu:4764.090621555707 mfu:22.473028299555473 +Step 13679 | loss:1.0575770139694214 lr:6.907556635058443e-05 tokens_per_second_per_gpu:4473.813307759589 mfu:21.10374068396245 +Step 13680 | loss:1.1899176836013794 lr:6.907155803086795e-05 tokens_per_second_per_gpu:4067.526069347585 mfu:19.187214460622123 +Step 13681 | loss:1.0032488107681274 lr:6.906754956771185e-05 tokens_per_second_per_gpu:4480.544799742679 mfu:21.135494280157648 +Step 13682 | loss:1.0715709924697876 lr:6.906354096114625e-05 tokens_per_second_per_gpu:4341.163886263908 mfu:20.478010730442243 +Step 13683 | loss:0.9927610158920288 lr:6.905953221120134e-05 tokens_per_second_per_gpu:5095.890683710541 mfu:24.038185803667965 +Step 13684 | loss:1.0108147859573364 lr:6.905552331790727e-05 tokens_per_second_per_gpu:5172.694451515434 mfu:24.40048228047768 +Step 13685 | loss:0.9986065626144409 lr:6.905151428129417e-05 tokens_per_second_per_gpu:5207.577970175132 mfu:24.56503378973748 +Step 13686 | loss:1.072051763534546 lr:6.90475051013922e-05 tokens_per_second_per_gpu:5349.152485957877 mfu:25.23286493578788 +Step 13687 | loss:0.9178235530853271 lr:6.904349577823151e-05 tokens_per_second_per_gpu:5336.711503663658 mfu:25.174178699655588 +Step 13688 | loss:1.2225220203399658 lr:6.903948631184226e-05 tokens_per_second_per_gpu:4953.894623332253 mfu:23.368366159837468 +Step 13689 | loss:1.0274794101715088 lr:6.90354767022546e-05 tokens_per_second_per_gpu:4732.168320007478 mfu:22.322445356646206 +Step 13690 | loss:1.1573328971862793 lr:6.90314669494987e-05 tokens_per_second_per_gpu:4165.219883081369 mfu:19.648052848287787 +Step 13691 | loss:1.109461784362793 lr:6.902745705360471e-05 tokens_per_second_per_gpu:4473.919498946136 mfu:21.104241605907475 +Step 13692 | loss:1.1693036556243896 lr:6.902344701460278e-05 tokens_per_second_per_gpu:3288.4005312196055 mfu:15.511946364748075 +Step 13693 | loss:1.0034449100494385 lr:6.901943683252307e-05 tokens_per_second_per_gpu:5176.605939685647 mfu:24.418933437545274 +Step 13694 | loss:1.0306520462036133 lr:6.901542650739577e-05 tokens_per_second_per_gpu:4717.691904493361 mfu:22.25415763473494 +Step 13695 | loss:1.0520063638687134 lr:6.9011416039251e-05 tokens_per_second_per_gpu:4649.367269929426 mfu:21.931858676112128 +Step 13696 | loss:0.9787958860397339 lr:6.900740542811896e-05 tokens_per_second_per_gpu:5053.26242389698 mfu:23.83710102899629 +Step 13697 | loss:1.136163353919983 lr:6.900339467402979e-05 tokens_per_second_per_gpu:4431.479854143841 mfu:20.904046560424987 +Step 13698 | loss:1.0337048768997192 lr:6.899938377701368e-05 tokens_per_second_per_gpu:3876.8964451141005 mfu:18.28798199834504 +Step 13699 | loss:1.0055811405181885 lr:6.899537273710075e-05 tokens_per_second_per_gpu:5290.327209750796 mfu:24.95537607129201 +Step 13700 | loss:0.9982743263244629 lr:6.899136155432122e-05 tokens_per_second_per_gpu:5287.308407130955 mfu:24.941135864273384 +Step 13701 | loss:1.0566238164901733 lr:6.898735022870522e-05 tokens_per_second_per_gpu:4761.099542637683 mfu:22.458918869968464 +Step 13702 | loss:1.1257284879684448 lr:6.898333876028294e-05 tokens_per_second_per_gpu:4992.780513784387 mfu:23.551797539717235 +Step 13703 | loss:1.1595512628555298 lr:6.897932714908456e-05 tokens_per_second_per_gpu:4978.500442733593 mfu:23.48443600813928 +Step 13704 | loss:1.0103483200073242 lr:6.897531539514024e-05 tokens_per_second_per_gpu:5302.209807176878 mfu:25.011428310731016 +Step 13705 | loss:1.0723873376846313 lr:6.897130349848013e-05 tokens_per_second_per_gpu:5350.7060581075275 mfu:25.24019340068452 +Step 13706 | loss:1.0941052436828613 lr:6.896729145913445e-05 tokens_per_second_per_gpu:3935.0794744977293 mfu:18.562441274996072 +Step 13707 | loss:1.09418785572052 lr:6.896327927713333e-05 tokens_per_second_per_gpu:5381.9761659166825 mfu:25.387699834450853 +Step 13708 | loss:1.1605454683303833 lr:6.895926695250699e-05 tokens_per_second_per_gpu:5332.828368778828 mfu:25.155861289872718 +Step 13709 | loss:1.0655362606048584 lr:6.895525448528555e-05 tokens_per_second_per_gpu:5371.501751761414 mfu:25.33829023576111 +Step 13710 | loss:1.149471402168274 lr:6.895124187549927e-05 tokens_per_second_per_gpu:5305.822240387059 mfu:25.0284687745281 +Step 13711 | loss:0.9214985966682434 lr:6.894722912317824e-05 tokens_per_second_per_gpu:4885.819117253728 mfu:23.047242382786038 +Step 13712 | loss:1.0889140367507935 lr:6.89432162283527e-05 tokens_per_second_per_gpu:5041.580582433357 mfu:23.781995789684704 +Step 13713 | loss:1.135846495628357 lr:6.893920319105281e-05 tokens_per_second_per_gpu:4726.044248838146 mfu:22.29355706806618 +Step 13714 | loss:0.9602193832397461 lr:6.893519001130875e-05 tokens_per_second_per_gpu:5232.245142868387 mfu:24.68139305198521 +Step 13715 | loss:1.0244982242584229 lr:6.893117668915072e-05 tokens_per_second_per_gpu:5335.595160692087 mfu:25.168912719390935 +Step 13716 | loss:1.1832091808319092 lr:6.892716322460889e-05 tokens_per_second_per_gpu:4131.668010578178 mfu:19.48978294115049 +Step 13717 | loss:1.061702013015747 lr:6.892314961771345e-05 tokens_per_second_per_gpu:5325.937181313395 mfu:25.123354382840446 +Step 13718 | loss:1.0723130702972412 lr:6.891913586849459e-05 tokens_per_second_per_gpu:4064.0119543680025 mfu:19.170637780693593 +Step 13719 | loss:0.8948291540145874 lr:6.891512197698248e-05 tokens_per_second_per_gpu:3792.941971148177 mfu:17.891954420537463 +Step 13720 | loss:1.0752676725387573 lr:6.891110794320734e-05 tokens_per_second_per_gpu:4053.0086981093286 mfu:19.11873354357233 +Step 13721 | loss:1.0172832012176514 lr:6.890709376719935e-05 tokens_per_second_per_gpu:4772.327650992854 mfu:22.511883772792636 +Step 13722 | loss:1.1126432418823242 lr:6.890307944898869e-05 tokens_per_second_per_gpu:5233.585326890631 mfu:24.687714928677106 +Step 13723 | loss:0.9176037907600403 lr:6.889906498860554e-05 tokens_per_second_per_gpu:5219.527016066751 mfu:24.621399477925703 +Step 13724 | loss:1.0660299062728882 lr:6.889505038608012e-05 tokens_per_second_per_gpu:4208.504540595465 mfu:19.85223396290588 +Step 13725 | loss:1.0326796770095825 lr:6.889103564144263e-05 tokens_per_second_per_gpu:5145.556298135543 mfu:24.272466980738315 +Step 13726 | loss:1.1997857093811035 lr:6.888702075472324e-05 tokens_per_second_per_gpu:3478.55626978735 mfu:16.408943427486637 +Step 13727 | loss:1.0474822521209717 lr:6.888300572595215e-05 tokens_per_second_per_gpu:4255.055005123178 mfu:20.071820446411337 +Step 13728 | loss:1.1393786668777466 lr:6.887899055515958e-05 tokens_per_second_per_gpu:4698.091457217855 mfu:22.161698980755233 +Step 13729 | loss:0.9386578798294067 lr:6.887497524237571e-05 tokens_per_second_per_gpu:5293.219101447723 mfu:24.96901761783402 +Step 13730 | loss:0.9357287287712097 lr:6.887095978763072e-05 tokens_per_second_per_gpu:4206.344761119696 mfu:19.8420459146223 +Step 13731 | loss:0.9730162620544434 lr:6.886694419095485e-05 tokens_per_second_per_gpu:4197.830033105951 mfu:19.801880489819062 +Step 13732 | loss:1.0910874605178833 lr:6.88629284523783e-05 tokens_per_second_per_gpu:5351.991331966106 mfu:25.246256256765772 +Step 13733 | loss:1.0059599876403809 lr:6.885891257193124e-05 tokens_per_second_per_gpu:5348.210503742973 mfu:25.228421444961437 +Step 13734 | loss:1.0171858072280884 lr:6.88548965496439e-05 tokens_per_second_per_gpu:4418.397609142176 mfu:20.842335378690972 +Step 13735 | loss:1.0027679204940796 lr:6.885088038554647e-05 tokens_per_second_per_gpu:5364.158564227281 mfu:25.303651167285622 +Step 13736 | loss:1.1376055479049683 lr:6.884686407966917e-05 tokens_per_second_per_gpu:4480.071470210353 mfu:21.13326150399532 +Step 13737 | loss:1.0510601997375488 lr:6.884284763204218e-05 tokens_per_second_per_gpu:4978.669177378824 mfu:23.485231958249933 +Step 13738 | loss:0.9878844618797302 lr:6.883883104269575e-05 tokens_per_second_per_gpu:4233.50574108788 mfu:19.970168891273264 +Step 13739 | loss:1.1112607717514038 lr:6.883481431166007e-05 tokens_per_second_per_gpu:4937.200672101967 mfu:23.289617943602423 +Step 13740 | loss:1.155195951461792 lr:6.883079743896534e-05 tokens_per_second_per_gpu:3391.9637984343103 mfu:16.000471965915267 +Step 13741 | loss:1.0772864818572998 lr:6.882678042464178e-05 tokens_per_second_per_gpu:4715.960755708129 mfu:22.24599151055056 +Step 13742 | loss:0.9789707064628601 lr:6.88227632687196e-05 tokens_per_second_per_gpu:3954.4107611007694 mfu:18.653630252160216 +Step 13743 | loss:1.0197279453277588 lr:6.881874597122902e-05 tokens_per_second_per_gpu:4056.399888616639 mfu:19.134730367792073 +Step 13744 | loss:1.0293893814086914 lr:6.881472853220026e-05 tokens_per_second_per_gpu:4871.205264695749 mfu:22.978306346889696 +Step 13745 | loss:1.0299279689788818 lr:6.881071095166352e-05 tokens_per_second_per_gpu:5301.619890179082 mfu:25.008645571602194 +Step 13746 | loss:1.077304482460022 lr:6.880669322964901e-05 tokens_per_second_per_gpu:3521.145528111326 mfu:16.609844225477595 +Step 13747 | loss:1.1802974939346313 lr:6.880267536618696e-05 tokens_per_second_per_gpu:4172.566265279844 mfu:19.682706986540044 +Step 13748 | loss:0.9161837100982666 lr:6.87986573613076e-05 tokens_per_second_per_gpu:4292.09667281155 mfu:20.246552312857588 +Step 13749 | loss:1.1372147798538208 lr:6.879463921504113e-05 tokens_per_second_per_gpu:5342.114225664549 mfu:25.19966426113322 +Step 13750 | loss:1.1598769426345825 lr:6.87906209274178e-05 tokens_per_second_per_gpu:3652.2148407450045 mfu:17.228120535901365 +Step 13751 | loss:1.1309560537338257 lr:6.87866024984678e-05 tokens_per_second_per_gpu:4407.506698189219 mfu:20.790961093544855 +Step 13752 | loss:1.0960676670074463 lr:6.878258392822136e-05 tokens_per_second_per_gpu:4067.6975414401404 mfu:19.188023323738868 +Step 13753 | loss:0.9363090395927429 lr:6.877856521670871e-05 tokens_per_second_per_gpu:3891.243336401757 mfu:18.355658732380977 +Step 13754 | loss:1.0597028732299805 lr:6.877454636396009e-05 tokens_per_second_per_gpu:3962.421783922312 mfu:18.691419613630863 +Step 13755 | loss:1.1004985570907593 lr:6.877052737000569e-05 tokens_per_second_per_gpu:5340.1874036492 mfu:25.19057511293708 +Step 13756 | loss:1.0468823909759521 lr:6.876650823487577e-05 tokens_per_second_per_gpu:5372.82596427265 mfu:25.344536772110565 +Step 13757 | loss:0.9359637498855591 lr:6.876248895860054e-05 tokens_per_second_per_gpu:5326.858697236621 mfu:25.127701330677134 +Step 13758 | loss:1.0485451221466064 lr:6.875846954121026e-05 tokens_per_second_per_gpu:4265.446616249047 mfu:20.120839449083746 +Step 13759 | loss:1.0534389019012451 lr:6.87544499827351e-05 tokens_per_second_per_gpu:5361.989812698334 mfu:25.293420796296413 +Step 13760 | loss:0.9732911586761475 lr:6.875043028320536e-05 tokens_per_second_per_gpu:5376.977437092764 mfu:25.364119977717348 +Step 13761 | loss:1.0659552812576294 lr:6.874641044265122e-05 tokens_per_second_per_gpu:3441.8799993085645 mfu:16.235935202021196 +Step 13762 | loss:1.121832251548767 lr:6.874239046110296e-05 tokens_per_second_per_gpu:4219.480016535862 mfu:19.904007155526028 +Step 13763 | loss:1.137300729751587 lr:6.873837033859076e-05 tokens_per_second_per_gpu:4007.154233767485 mfu:18.902430211693492 +Step 13764 | loss:0.9215011596679688 lr:6.873435007514491e-05 tokens_per_second_per_gpu:4834.469397531734 mfu:22.805016993691787 +Step 13765 | loss:1.0990339517593384 lr:6.873032967079561e-05 tokens_per_second_per_gpu:5301.8694728338305 mfu:25.009822895567837 +Step 13766 | loss:1.037013292312622 lr:6.872630912557311e-05 tokens_per_second_per_gpu:5337.245108865239 mfu:25.17669580643388 +Step 13767 | loss:1.1345312595367432 lr:6.872228843950765e-05 tokens_per_second_per_gpu:3857.9101826589413 mfu:18.198420558953643 +Step 13768 | loss:1.1020854711532593 lr:6.871826761262947e-05 tokens_per_second_per_gpu:4147.751391356848 mfu:19.565650992391202 +Step 13769 | loss:1.0253539085388184 lr:6.871424664496882e-05 tokens_per_second_per_gpu:3452.5753202870874 mfu:16.286386855886814 +Step 13770 | loss:1.0679280757904053 lr:6.871022553655593e-05 tokens_per_second_per_gpu:3650.397296432767 mfu:17.219546869275504 +Step 13771 | loss:1.0379928350448608 lr:6.870620428742103e-05 tokens_per_second_per_gpu:4364.4405161234035 mfu:20.587810564892056 +Step 13772 | loss:1.108498454093933 lr:6.87021828975944e-05 tokens_per_second_per_gpu:4103.450448228598 mfu:19.35667588513494 +Step 13773 | loss:1.0444657802581787 lr:6.869816136710625e-05 tokens_per_second_per_gpu:5288.813465028004 mfu:24.948235479163124 +Step 13774 | loss:0.957436203956604 lr:6.869413969598686e-05 tokens_per_second_per_gpu:4372.583881595019 mfu:20.626224209223086 +Step 13775 | loss:1.0076707601547241 lr:6.869011788426644e-05 tokens_per_second_per_gpu:4323.859169847053 mfu:20.396381430614863 +Step 13776 | loss:1.0147974491119385 lr:6.868609593197528e-05 tokens_per_second_per_gpu:5299.239820072842 mfu:24.99741837482923 +Step 13777 | loss:0.9295720458030701 lr:6.868207383914358e-05 tokens_per_second_per_gpu:5311.432682363084 mfu:25.05493418656912 +Step 13778 | loss:1.1321923732757568 lr:6.867805160580166e-05 tokens_per_second_per_gpu:3807.8735168949047 mfu:17.96238906940918 +Step 13779 | loss:1.0549919605255127 lr:6.86740292319797e-05 tokens_per_second_per_gpu:4167.892614191283 mfu:19.660660576948203 +Step 13780 | loss:1.1322026252746582 lr:6.867000671770799e-05 tokens_per_second_per_gpu:3459.9557110989426 mfu:16.321201418570855 +Step 13781 | loss:1.081931233406067 lr:6.866598406301678e-05 tokens_per_second_per_gpu:4711.412410018331 mfu:22.224536187903983 +Step 13782 | loss:1.1698302030563354 lr:6.866196126793632e-05 tokens_per_second_per_gpu:4237.69440929771 mfu:19.98992755387952 +Step 13783 | loss:1.008786678314209 lr:6.865793833249687e-05 tokens_per_second_per_gpu:4667.885510825363 mfu:22.019212377955963 +Step 13784 | loss:1.1018314361572266 lr:6.865391525672868e-05 tokens_per_second_per_gpu:4690.824994600573 mfu:22.127421836803144 +Step 13785 | loss:0.9868907928466797 lr:6.864989204066201e-05 tokens_per_second_per_gpu:4598.2271768980145 mfu:21.690622131883355 +Step 13786 | loss:1.120368242263794 lr:6.864586868432713e-05 tokens_per_second_per_gpu:3503.303211681977 mfu:16.525678974667265 +Step 13787 | loss:1.1696228981018066 lr:6.864184518775429e-05 tokens_per_second_per_gpu:5319.872622759718 mfu:25.094746825422366 +Step 13788 | loss:1.193525791168213 lr:6.863782155097376e-05 tokens_per_second_per_gpu:4734.476911441996 mfu:22.333335376329146 +Step 13789 | loss:1.190073013305664 lr:6.863379777401579e-05 tokens_per_second_per_gpu:4177.746278140278 mfu:19.707141991004043 +Step 13790 | loss:1.093990683555603 lr:6.862977385691065e-05 tokens_per_second_per_gpu:4778.501046242257 mfu:22.541004731473908 +Step 13791 | loss:1.0117769241333008 lr:6.862574979968859e-05 tokens_per_second_per_gpu:5365.692306659986 mfu:25.31088609202413 +Step 13792 | loss:1.12165105342865 lr:6.862172560237992e-05 tokens_per_second_per_gpu:5369.844145241757 mfu:25.330471022991052 +Step 13793 | loss:1.0102901458740234 lr:6.861770126501486e-05 tokens_per_second_per_gpu:4724.216542223853 mfu:22.28493546412766 +Step 13794 | loss:0.8371196985244751 lr:6.861367678762368e-05 tokens_per_second_per_gpu:5330.017093217982 mfu:25.14260002339913 +Step 13795 | loss:1.166654109954834 lr:6.860965217023667e-05 tokens_per_second_per_gpu:5125.507494984136 mfu:24.17789335559458 +Step 13796 | loss:1.0464411973953247 lr:6.860562741288408e-05 tokens_per_second_per_gpu:4855.542815803142 mfu:22.904423903173928 +Step 13797 | loss:1.1752257347106934 lr:6.860160251559622e-05 tokens_per_second_per_gpu:5369.389286339138 mfu:25.32832537594047 +Step 13798 | loss:1.1172807216644287 lr:6.859757747840332e-05 tokens_per_second_per_gpu:4938.7404568793645 mfu:23.296881371111777 +Step 13799 | loss:1.0610437393188477 lr:6.859355230133567e-05 tokens_per_second_per_gpu:4382.792917215871 mfu:20.674381972087733 +Step 13800 | loss:0.9946444630622864 lr:6.858952698442352e-05 tokens_per_second_per_gpu:5401.241416161563 mfu:25.47857730684659 +Step 13801 | loss:1.127371907234192 lr:6.85855015276972e-05 tokens_per_second_per_gpu:3929.9896390207477 mfu:18.53843165263567 +Step 13802 | loss:0.9967316389083862 lr:6.858147593118693e-05 tokens_per_second_per_gpu:5334.179192970104 mfu:25.162233358057335 +Step 13803 | loss:1.1530297994613647 lr:6.857745019492301e-05 tokens_per_second_per_gpu:5337.165373222069 mfu:25.176319679800745 +Step 13804 | loss:1.0622299909591675 lr:6.857342431893572e-05 tokens_per_second_per_gpu:5323.501116075397 mfu:25.111863047477218 +Step 13805 | loss:1.0118581056594849 lr:6.856939830325534e-05 tokens_per_second_per_gpu:4528.528124281225 mfu:21.361839808805332 +Step 13806 | loss:1.1016671657562256 lr:6.856537214791214e-05 tokens_per_second_per_gpu:5330.817195223145 mfu:25.146374241068965 +Step 13807 | loss:1.0974864959716797 lr:6.856134585293641e-05 tokens_per_second_per_gpu:4981.787313339248 mfu:23.499940737584105 +Step 13808 | loss:1.1443064212799072 lr:6.855731941835844e-05 tokens_per_second_per_gpu:5351.253524684933 mfu:25.242775893937573 +Step 13809 | loss:1.0061674118041992 lr:6.855329284420847e-05 tokens_per_second_per_gpu:4279.812026817291 mfu:20.18860354172593 +Step 13810 | loss:1.0494444370269775 lr:6.854926613051685e-05 tokens_per_second_per_gpu:5353.349323709543 mfu:25.252662135516843 +Step 13811 | loss:1.0580601692199707 lr:6.854523927731383e-05 tokens_per_second_per_gpu:5302.64099777122 mfu:25.013462310333235 +Step 13812 | loss:1.1678423881530762 lr:6.854121228462967e-05 tokens_per_second_per_gpu:4122.985822491206 mfu:19.448827578610203 +Step 13813 | loss:0.9569199681282043 lr:6.853718515249469e-05 tokens_per_second_per_gpu:4334.027288544188 mfu:20.4443461813692 +Step 13814 | loss:1.0885456800460815 lr:6.85331578809392e-05 tokens_per_second_per_gpu:5356.046446912277 mfu:25.265384925840813 +Step 13815 | loss:0.9453999400138855 lr:6.852913046999346e-05 tokens_per_second_per_gpu:3579.613098942793 mfu:16.88564573268586 +Step 13816 | loss:0.9631577134132385 lr:6.852510291968777e-05 tokens_per_second_per_gpu:4002.4746460407455 mfu:18.880355797966462 +Step 13817 | loss:1.135453224182129 lr:6.85210752300524e-05 tokens_per_second_per_gpu:5330.180912215554 mfu:25.143372785561084 +Step 13818 | loss:0.9998748898506165 lr:6.851704740111767e-05 tokens_per_second_per_gpu:3892.9507021485174 mfu:18.36371266791508 +Step 13819 | loss:1.1452687978744507 lr:6.851301943291386e-05 tokens_per_second_per_gpu:4393.784409668457 mfu:20.72623071732857 +Step 13820 | loss:0.9567068219184875 lr:6.850899132547126e-05 tokens_per_second_per_gpu:4708.860598363526 mfu:22.212498856944308 +Step 13821 | loss:1.1137291193008423 lr:6.850496307882019e-05 tokens_per_second_per_gpu:5334.829282777498 mfu:25.16529994259557 +Step 13822 | loss:1.0623321533203125 lr:6.850093469299093e-05 tokens_per_second_per_gpu:4783.271004878113 mfu:22.563505440198107 +Step 13823 | loss:0.9473366141319275 lr:6.849690616801377e-05 tokens_per_second_per_gpu:5053.801438578882 mfu:23.83964365321621 +Step 13824 | loss:1.0801446437835693 lr:6.849287750391904e-05 tokens_per_second_per_gpu:5345.851327978233 mfu:25.217292810362267 +Step 13825 | loss:1.1089421510696411 lr:6.848884870073698e-05 tokens_per_second_per_gpu:4321.504512430309 mfu:20.385274109834103 +Step 13826 | loss:1.160274863243103 lr:6.848481975849797e-05 tokens_per_second_per_gpu:4630.303634577655 mfu:21.841932255566242 +Step 13827 | loss:0.9498582482337952 lr:6.848079067723225e-05 tokens_per_second_per_gpu:4001.589251258569 mfu:18.876179239715015 +Step 13828 | loss:1.1157728433609009 lr:6.847676145697015e-05 tokens_per_second_per_gpu:4775.016731440965 mfu:22.524568624070806 +Step 13829 | loss:1.1733250617980957 lr:6.847273209774197e-05 tokens_per_second_per_gpu:5333.536382850135 mfu:25.159201113046727 +Step 13830 | loss:1.0085229873657227 lr:6.846870259957801e-05 tokens_per_second_per_gpu:3688.4355612224367 mfu:17.398979854285574 +Step 13831 | loss:0.9429999589920044 lr:6.846467296250858e-05 tokens_per_second_per_gpu:3960.4540266696895 mfu:18.68213736693624 +Step 13832 | loss:1.0883119106292725 lr:6.8460643186564e-05 tokens_per_second_per_gpu:4110.163992975049 mfu:19.388344821152796 +Step 13833 | loss:1.1164063215255737 lr:6.845661327177455e-05 tokens_per_second_per_gpu:5235.849547805374 mfu:24.698395645047874 +Step 13834 | loss:0.9358333945274353 lr:6.845258321817058e-05 tokens_per_second_per_gpu:5332.582124304398 mfu:25.15469971267319 +Step 13835 | loss:1.1026533842086792 lr:6.844855302578236e-05 tokens_per_second_per_gpu:4151.189037475586 mfu:19.581866955656196 +Step 13836 | loss:0.9172858595848083 lr:6.844452269464021e-05 tokens_per_second_per_gpu:4281.734166543839 mfu:20.197670602766912 +Step 13837 | loss:1.0037184953689575 lr:6.844049222477448e-05 tokens_per_second_per_gpu:5312.65767900626 mfu:25.06071270474096 +Step 13838 | loss:0.9779292941093445 lr:6.843646161621543e-05 tokens_per_second_per_gpu:4651.944134113362 mfu:21.94401418412612 +Step 13839 | loss:0.9425421357154846 lr:6.843243086899341e-05 tokens_per_second_per_gpu:5330.226046125969 mfu:25.143585689917153 +Step 13840 | loss:1.1467050313949585 lr:6.842839998313872e-05 tokens_per_second_per_gpu:5299.77258366126 mfu:24.999931511574637 +Step 13841 | loss:1.0893144607543945 lr:6.842436895868169e-05 tokens_per_second_per_gpu:4001.420115278111 mfu:18.875381396437618 +Step 13842 | loss:1.025310754776001 lr:6.842033779565263e-05 tokens_per_second_per_gpu:4166.427484968293 mfu:19.653749312426537 +Step 13843 | loss:1.1173211336135864 lr:6.841630649408184e-05 tokens_per_second_per_gpu:4483.449762860953 mfu:21.149197486824363 +Step 13844 | loss:1.1938318014144897 lr:6.84122750539997e-05 tokens_per_second_per_gpu:4040.9930897361733 mfu:19.062053868801964 +Step 13845 | loss:1.0386654138565063 lr:6.840824347543644e-05 tokens_per_second_per_gpu:5326.275854815143 mfu:25.12495196353185 +Step 13846 | loss:1.069717288017273 lr:6.840421175842246e-05 tokens_per_second_per_gpu:4433.903262979177 mfu:20.91547819337745 +Step 13847 | loss:0.9969120621681213 lr:6.840017990298805e-05 tokens_per_second_per_gpu:4925.516212783916 mfu:23.234500355424842 +Step 13848 | loss:1.0472745895385742 lr:6.839614790916354e-05 tokens_per_second_per_gpu:4267.296565800652 mfu:20.129565976752605 +Step 13849 | loss:1.0511330366134644 lr:6.839211577697925e-05 tokens_per_second_per_gpu:5349.004701078987 mfu:25.232167809299455 +Step 13850 | loss:1.108928918838501 lr:6.838808350646552e-05 tokens_per_second_per_gpu:4994.708232382969 mfu:23.560890917249452 +Step 13851 | loss:1.0397748947143555 lr:6.838405109765265e-05 tokens_per_second_per_gpu:4996.316879894798 mfu:23.568479182026316 +Step 13852 | loss:1.0809900760650635 lr:6.838001855057101e-05 tokens_per_second_per_gpu:4330.641018815648 mfu:20.42837256930248 +Step 13853 | loss:1.0590221881866455 lr:6.83759858652509e-05 tokens_per_second_per_gpu:4671.840987832155 mfu:22.037871037870968 +Step 13854 | loss:0.9887548089027405 lr:6.837195304172263e-05 tokens_per_second_per_gpu:4760.248656444255 mfu:22.454905094617914 +Step 13855 | loss:0.9396153688430786 lr:6.836792008001659e-05 tokens_per_second_per_gpu:4288.264078705529 mfu:20.228473312551504 +Step 13856 | loss:0.9781429171562195 lr:6.836388698016306e-05 tokens_per_second_per_gpu:4605.493059790776 mfu:21.724896541175962 +Step 13857 | loss:0.98092120885849 lr:6.835985374219241e-05 tokens_per_second_per_gpu:4653.0700238815125 mfu:21.94932519825909 +Step 13858 | loss:1.0986599922180176 lr:6.835582036613493e-05 tokens_per_second_per_gpu:4081.718086408913 mfu:19.25416062650821 +Step 13859 | loss:1.0549012422561646 lr:6.835178685202101e-05 tokens_per_second_per_gpu:3760.4525223510514 mfu:17.738696147289854 +Step 13860 | loss:0.996724545955658 lr:6.834775319988094e-05 tokens_per_second_per_gpu:3482.0797740040393 mfu:16.42556439804857 +Step 13861 | loss:0.940372109413147 lr:6.834371940974508e-05 tokens_per_second_per_gpu:5321.822694341139 mfu:25.10394564578845 +Step 13862 | loss:1.0939549207687378 lr:6.833968548164378e-05 tokens_per_second_per_gpu:5330.684222720616 mfu:25.14574698708709 +Step 13863 | loss:1.1106072664260864 lr:6.833565141560734e-05 tokens_per_second_per_gpu:3791.8360674719697 mfu:17.88673768421017 +Step 13864 | loss:1.0255935192108154 lr:6.833161721166614e-05 tokens_per_second_per_gpu:5358.134231846961 mfu:25.275233363590555 +Step 13865 | loss:1.1470741033554077 lr:6.83275828698505e-05 tokens_per_second_per_gpu:3937.619447982569 mfu:18.574422763288226 +Step 13866 | loss:0.9565720558166504 lr:6.832354839019079e-05 tokens_per_second_per_gpu:4875.267387950398 mfu:22.99746807535157 +Step 13867 | loss:0.9796299934387207 lr:6.831951377271733e-05 tokens_per_second_per_gpu:5355.039176441674 mfu:25.26063345917291 +Step 13868 | loss:1.1876800060272217 lr:6.831547901746046e-05 tokens_per_second_per_gpu:5211.393167897688 mfu:24.583030728334634 +Step 13869 | loss:1.1313964128494263 lr:6.831144412445056e-05 tokens_per_second_per_gpu:5316.50531945773 mfu:25.078862681225704 +Step 13870 | loss:1.1395074129104614 lr:6.830740909371791e-05 tokens_per_second_per_gpu:5092.460466388698 mfu:24.022004883305403 +Step 13871 | loss:1.0870956182479858 lr:6.830337392529292e-05 tokens_per_second_per_gpu:5091.848210467903 mfu:24.019116767665384 +Step 13872 | loss:1.1958961486816406 lr:6.829933861920592e-05 tokens_per_second_per_gpu:4286.602436891266 mfu:20.2206350646083 +Step 13873 | loss:1.116510033607483 lr:6.829530317548727e-05 tokens_per_second_per_gpu:3933.044333363546 mfu:18.552841167035247 +Step 13874 | loss:1.0487442016601562 lr:6.829126759416728e-05 tokens_per_second_per_gpu:4765.627493559397 mfu:22.480277987014286 +Step 13875 | loss:1.0293822288513184 lr:6.828723187527636e-05 tokens_per_second_per_gpu:4084.7531794244323 mfu:19.26847767810268 +Step 13876 | loss:1.044419527053833 lr:6.828319601884483e-05 tokens_per_second_per_gpu:4710.776622993563 mfu:22.221537072031705 +Step 13877 | loss:0.9426595568656921 lr:6.827916002490304e-05 tokens_per_second_per_gpu:5215.5599952299945 mfu:24.602686363795176 +Step 13878 | loss:1.0904874801635742 lr:6.827512389348138e-05 tokens_per_second_per_gpu:4546.027410108669 mfu:21.444386925739423 +Step 13879 | loss:1.1878846883773804 lr:6.827108762461015e-05 tokens_per_second_per_gpu:3345.970139878068 mfu:15.783512031178132 +Step 13880 | loss:1.1233668327331543 lr:6.826705121831976e-05 tokens_per_second_per_gpu:5366.733687696874 mfu:25.31579846405306 +Step 13881 | loss:1.101297378540039 lr:6.826301467464054e-05 tokens_per_second_per_gpu:3443.595817693397 mfu:16.244029010091253 +Step 13882 | loss:1.0932095050811768 lr:6.825897799360286e-05 tokens_per_second_per_gpu:3782.886146855478 mfu:17.844519381648258 +Step 13883 | loss:0.9465950727462769 lr:6.825494117523706e-05 tokens_per_second_per_gpu:5350.126934651297 mfu:25.237461576532834 +Step 13884 | loss:1.0976120233535767 lr:6.825090421957354e-05 tokens_per_second_per_gpu:5322.601366088498 mfu:25.107618773275842 +Step 13885 | loss:1.1487334966659546 lr:6.824686712664262e-05 tokens_per_second_per_gpu:4867.941746875039 mfu:22.96291177651585 +Step 13886 | loss:0.9768018126487732 lr:6.824282989647469e-05 tokens_per_second_per_gpu:5261.864003820108 mfu:24.82111027259326 +Step 13887 | loss:0.8693257570266724 lr:6.823879252910013e-05 tokens_per_second_per_gpu:4632.479255389207 mfu:21.85219503013351 +Step 13888 | loss:1.0793561935424805 lr:6.823475502454925e-05 tokens_per_second_per_gpu:5359.629825621063 mfu:25.282288334597606 +Step 13889 | loss:1.0991188287734985 lr:6.823071738285247e-05 tokens_per_second_per_gpu:5338.346482620941 mfu:25.181891174353265 +Step 13890 | loss:1.2049190998077393 lr:6.822667960404014e-05 tokens_per_second_per_gpu:4379.026600313922 mfu:20.656615612661316 +Step 13891 | loss:1.0791877508163452 lr:6.822264168814263e-05 tokens_per_second_per_gpu:4906.597654344044 mfu:23.14525828742505 +Step 13892 | loss:1.0126718282699585 lr:6.82186036351903e-05 tokens_per_second_per_gpu:4477.556836931741 mfu:21.121399549779426 +Step 13893 | loss:1.0192975997924805 lr:6.821456544521351e-05 tokens_per_second_per_gpu:5126.7967843297665 mfu:24.183975153413364 +Step 13894 | loss:0.911063015460968 lr:6.821052711824268e-05 tokens_per_second_per_gpu:3817.943814407992 mfu:18.00989238094834 +Step 13895 | loss:1.0412287712097168 lr:6.820648865430814e-05 tokens_per_second_per_gpu:5352.573863840905 mfu:25.249004159009385 +Step 13896 | loss:1.0773050785064697 lr:6.820245005344028e-05 tokens_per_second_per_gpu:4472.036980872744 mfu:21.09536144696458 +Step 13897 | loss:1.0480036735534668 lr:6.819841131566946e-05 tokens_per_second_per_gpu:5366.248321433116 mfu:25.31350890857489 +Step 13898 | loss:0.9886056780815125 lr:6.819437244102606e-05 tokens_per_second_per_gpu:5358.521464911447 mfu:25.27706000802451 +Step 13899 | loss:0.9819440841674805 lr:6.819033342954047e-05 tokens_per_second_per_gpu:5319.091046947508 mfu:25.09105999896559 +Step 13900 | loss:1.0818171501159668 lr:6.818629428124308e-05 tokens_per_second_per_gpu:4491.802285867386 mfu:21.188597762933078 +Step 13901 | loss:1.1664170026779175 lr:6.818225499616422e-05 tokens_per_second_per_gpu:3856.227286394224 mfu:18.19048205014126 +Step 13902 | loss:0.9769012928009033 lr:6.817821557433433e-05 tokens_per_second_per_gpu:5310.993040919891 mfu:25.052860322871016 +Step 13903 | loss:1.0256071090698242 lr:6.817417601578375e-05 tokens_per_second_per_gpu:5079.919014261667 mfu:23.96284471382982 +Step 13904 | loss:1.01320481300354 lr:6.817013632054286e-05 tokens_per_second_per_gpu:4484.598164354289 mfu:21.15461469260593 +Step 13905 | loss:1.022992730140686 lr:6.816609648864208e-05 tokens_per_second_per_gpu:4644.452441446495 mfu:21.908674591601937 +Step 13906 | loss:1.0445961952209473 lr:6.816205652011174e-05 tokens_per_second_per_gpu:4149.511384641621 mfu:19.573953181001293 +Step 13907 | loss:1.013379693031311 lr:6.815801641498227e-05 tokens_per_second_per_gpu:5030.765974045077 mfu:23.730981436754725 +Step 13908 | loss:1.0269192457199097 lr:6.815397617328403e-05 tokens_per_second_per_gpu:4026.8757011340895 mfu:18.995459738091064 +Step 13909 | loss:1.2206813097000122 lr:6.814993579504746e-05 tokens_per_second_per_gpu:3735.868873272053 mfu:17.62273088018117 +Step 13910 | loss:1.1432920694351196 lr:6.814589528030286e-05 tokens_per_second_per_gpu:4013.471969867235 mfu:18.932232050792944 +Step 13911 | loss:1.0374116897583008 lr:6.814185462908069e-05 tokens_per_second_per_gpu:3193.7975224460533 mfu:15.065687831425532 +Step 13912 | loss:1.1373827457427979 lr:6.813781384141131e-05 tokens_per_second_per_gpu:4305.177721711797 mfu:20.30825785237714 +Step 13913 | loss:1.0065333843231201 lr:6.813377291732512e-05 tokens_per_second_per_gpu:5354.88101406747 mfu:25.259887380268506 +Step 13914 | loss:0.9803777933120728 lr:6.812973185685252e-05 tokens_per_second_per_gpu:5262.1789341641015 mfu:24.822595852758532 +Step 13915 | loss:1.022214412689209 lr:6.812569066002387e-05 tokens_per_second_per_gpu:5351.809211236182 mfu:25.245397162209716 +Step 13916 | loss:1.1045125722885132 lr:6.812164932686962e-05 tokens_per_second_per_gpu:3752.994641109374 mfu:17.703516049027108 +Step 13917 | loss:1.1677500009536743 lr:6.811760785742009e-05 tokens_per_second_per_gpu:5334.6630053786685 mfu:25.164515583735447 +Step 13918 | loss:1.0417472124099731 lr:6.811356625170575e-05 tokens_per_second_per_gpu:3883.6583085225157 mfu:18.319878861735393 +Step 13919 | loss:1.0938057899475098 lr:6.810952450975695e-05 tokens_per_second_per_gpu:5344.05745168938 mfu:25.208830789840892 +Step 13920 | loss:1.1444443464279175 lr:6.810548263160413e-05 tokens_per_second_per_gpu:4072.7939535446426 mfu:19.21206396917337 +Step 13921 | loss:1.1152375936508179 lr:6.810144061727765e-05 tokens_per_second_per_gpu:3799.9772275075375 mfu:17.925140925123078 +Step 13922 | loss:1.014188289642334 lr:6.809739846680791e-05 tokens_per_second_per_gpu:4982.37774658388 mfu:23.50272591193734 +Step 13923 | loss:1.1804485321044922 lr:6.809335618022537e-05 tokens_per_second_per_gpu:3780.954052211421 mfu:17.835405361562906 +Step 13924 | loss:1.2483140230178833 lr:6.808931375756034e-05 tokens_per_second_per_gpu:3679.2236116375757 mfu:17.355525516373117 +Step 13925 | loss:1.061235785484314 lr:6.80852711988433e-05 tokens_per_second_per_gpu:4469.722502042512 mfu:21.08444365543157 +Step 13926 | loss:1.0517287254333496 lr:6.808122850410461e-05 tokens_per_second_per_gpu:5396.107277205654 mfu:25.45435869741713 +Step 13927 | loss:1.1012738943099976 lr:6.807718567337473e-05 tokens_per_second_per_gpu:5334.88386260362 mfu:25.165557404947673 +Step 13928 | loss:1.0321388244628906 lr:6.807314270668399e-05 tokens_per_second_per_gpu:5319.287592678675 mfu:25.09198713871366 +Step 13929 | loss:1.045025110244751 lr:6.806909960406285e-05 tokens_per_second_per_gpu:5371.938061670519 mfu:25.34034838404451 +Step 13930 | loss:0.9442523717880249 lr:6.806505636554172e-05 tokens_per_second_per_gpu:5354.316386503122 mfu:25.257223935712084 +Step 13931 | loss:0.9595493078231812 lr:6.806101299115097e-05 tokens_per_second_per_gpu:4314.30886273082 mfu:20.351331002497176 +Step 13932 | loss:1.1234830617904663 lr:6.805696948092107e-05 tokens_per_second_per_gpu:4888.639223752478 mfu:23.060545306300426 +Step 13933 | loss:1.0085934400558472 lr:6.805292583488237e-05 tokens_per_second_per_gpu:5091.867126593973 mfu:24.01920599825948 +Step 13934 | loss:1.0269019603729248 lr:6.804888205306534e-05 tokens_per_second_per_gpu:3752.618678296414 mfu:17.701742568292506 +Step 13935 | loss:1.0192946195602417 lr:6.804483813550033e-05 tokens_per_second_per_gpu:5370.974659003258 mfu:25.335803849289594 +Step 13936 | loss:0.966593325138092 lr:6.80407940822178e-05 tokens_per_second_per_gpu:4530.330901739297 mfu:21.370343817660608 +Step 13937 | loss:1.1242822408676147 lr:6.803674989324818e-05 tokens_per_second_per_gpu:4268.684413777812 mfu:20.13611269245215 +Step 13938 | loss:1.003103256225586 lr:6.803270556862184e-05 tokens_per_second_per_gpu:5331.398666764578 mfu:25.14911714153918 +Step 13939 | loss:1.0489170551300049 lr:6.802866110836921e-05 tokens_per_second_per_gpu:4540.412070073203 mfu:21.417898408716322 +Step 13940 | loss:0.9351944923400879 lr:6.802461651252073e-05 tokens_per_second_per_gpu:4991.7794559978565 mfu:23.547075379340097 +Step 13941 | loss:0.9741819500923157 lr:6.802057178110683e-05 tokens_per_second_per_gpu:5291.0308459229 mfu:24.95869524316854 +Step 13942 | loss:1.0348860025405884 lr:6.801652691415789e-05 tokens_per_second_per_gpu:3528.2079073755576 mfu:16.64315867343313 +Step 13943 | loss:1.0393186807632446 lr:6.801248191170435e-05 tokens_per_second_per_gpu:5171.283715705775 mfu:24.393827598967167 +Step 13944 | loss:0.9688951969146729 lr:6.800843677377664e-05 tokens_per_second_per_gpu:4763.259525263189 mfu:22.469107876544015 +Step 13945 | loss:1.0346101522445679 lr:6.800439150040518e-05 tokens_per_second_per_gpu:5084.65608648269 mfu:23.985190291724248 +Step 13946 | loss:1.1118009090423584 lr:6.80003460916204e-05 tokens_per_second_per_gpu:4973.172009023678 mfu:23.45930087720507 +Step 13947 | loss:1.2431654930114746 lr:6.799630054745272e-05 tokens_per_second_per_gpu:3669.7194238651277 mfu:17.310692641068833 +Step 13948 | loss:1.029205083847046 lr:6.799225486793255e-05 tokens_per_second_per_gpu:5322.567230031135 mfu:25.10745774766905 +Step 13949 | loss:1.1367501020431519 lr:6.798820905309036e-05 tokens_per_second_per_gpu:4293.9225276780535 mfu:20.25516518178585 +Step 13950 | loss:1.0461483001708984 lr:6.798416310295653e-05 tokens_per_second_per_gpu:4118.554660349162 mfu:19.427925030751965 +Step 13951 | loss:1.1249672174453735 lr:6.798011701756153e-05 tokens_per_second_per_gpu:5325.548300838059 mfu:25.12151996728843 +Step 13952 | loss:1.0319472551345825 lr:6.797607079693578e-05 tokens_per_second_per_gpu:4157.301286545361 mfu:19.61069947737555 +Step 13953 | loss:1.1533652544021606 lr:6.797202444110968e-05 tokens_per_second_per_gpu:3698.2830718472837 mfu:17.445432242061546 +Step 13954 | loss:0.963460385799408 lr:6.796797795011372e-05 tokens_per_second_per_gpu:4884.883166839898 mfu:23.042827344994393 +Step 13955 | loss:0.968966007232666 lr:6.796393132397829e-05 tokens_per_second_per_gpu:5328.348453704007 mfu:25.134728766112453 +Step 13956 | loss:1.08173668384552 lr:6.795988456273386e-05 tokens_per_second_per_gpu:4210.585360469671 mfu:19.862049545276992 +Step 13957 | loss:1.019313931465149 lr:6.795583766641083e-05 tokens_per_second_per_gpu:4001.609944446217 mfu:18.87627685301177 +Step 13958 | loss:1.0031583309173584 lr:6.795179063503964e-05 tokens_per_second_per_gpu:4910.95832170963 mfu:23.165828299395095 +Step 13959 | loss:1.0414433479309082 lr:6.794774346865076e-05 tokens_per_second_per_gpu:4909.614173652424 mfu:23.159487723673948 +Step 13960 | loss:1.0456428527832031 lr:6.794369616727462e-05 tokens_per_second_per_gpu:4760.727038524106 mfu:22.45716170450897 +Step 13961 | loss:1.09849214553833 lr:6.793964873094163e-05 tokens_per_second_per_gpu:3578.3810813062537 mfu:16.87983409529042 +Step 13962 | loss:1.088294506072998 lr:6.793560115968226e-05 tokens_per_second_per_gpu:5123.086467840128 mfu:24.166472957486462 +Step 13963 | loss:1.017696738243103 lr:6.793155345352697e-05 tokens_per_second_per_gpu:5285.3448092762665 mfu:24.93187323816848 +Step 13964 | loss:1.1132769584655762 lr:6.792750561250614e-05 tokens_per_second_per_gpu:3581.293088083559 mfu:16.893570528098557 +Step 13965 | loss:1.1234251260757446 lr:6.792345763665026e-05 tokens_per_second_per_gpu:4173.4551818242235 mfu:19.686900157544684 +Step 13966 | loss:1.0267313718795776 lr:6.791940952598979e-05 tokens_per_second_per_gpu:5370.03408839916 mfu:25.33136701727217 +Step 13967 | loss:1.0603381395339966 lr:6.791536128055514e-05 tokens_per_second_per_gpu:5330.145460660029 mfu:25.143205554525014 +Step 13968 | loss:0.8773064613342285 lr:6.791131290037677e-05 tokens_per_second_per_gpu:3384.5339843858483 mfu:15.965424265391679 +Step 13969 | loss:1.0596815347671509 lr:6.790726438548514e-05 tokens_per_second_per_gpu:5333.341154721135 mfu:25.15828018865288 +Step 13970 | loss:0.99067622423172 lr:6.79032157359107e-05 tokens_per_second_per_gpu:5347.326410395209 mfu:25.224251025798157 +Step 13971 | loss:1.3055081367492676 lr:6.789916695168386e-05 tokens_per_second_per_gpu:4400.546757255123 mfu:20.75812986467035 +Step 13972 | loss:0.9292742609977722 lr:6.789511803283512e-05 tokens_per_second_per_gpu:3556.8030362184113 mfu:16.778046775016943 +Step 13973 | loss:1.0619840621948242 lr:6.78910689793949e-05 tokens_per_second_per_gpu:5352.218454405274 mfu:25.24732763206241 +Step 13974 | loss:1.0791704654693604 lr:6.788701979139369e-05 tokens_per_second_per_gpu:4993.119180737264 mfu:23.5533950895165 +Step 13975 | loss:1.0889095067977905 lr:6.78829704688619e-05 tokens_per_second_per_gpu:3687.6921176758124 mfu:17.3954729042316 +Step 13976 | loss:1.056281566619873 lr:6.787892101183001e-05 tokens_per_second_per_gpu:5378.366741696005 mfu:25.37067356457064 +Step 13977 | loss:0.9951817989349365 lr:6.787487142032848e-05 tokens_per_second_per_gpu:4209.109816726996 mfu:19.855089153689068 +Step 13978 | loss:1.0641586780548096 lr:6.787082169438777e-05 tokens_per_second_per_gpu:4352.810981378325 mfu:20.532952065296286 +Step 13979 | loss:1.0771467685699463 lr:6.786677183403831e-05 tokens_per_second_per_gpu:5350.298152663256 mfu:25.238269241108046 +Step 13980 | loss:1.1260758638381958 lr:6.786272183931058e-05 tokens_per_second_per_gpu:5199.337880349094 mfu:24.526163880123764 +Step 13981 | loss:1.1240357160568237 lr:6.785867171023507e-05 tokens_per_second_per_gpu:4233.141688564501 mfu:19.968451593406723 +Step 13982 | loss:1.0493477582931519 lr:6.785462144684218e-05 tokens_per_second_per_gpu:3670.377013726684 mfu:17.313794604642194 +Step 13983 | loss:1.1049985885620117 lr:6.785057104916242e-05 tokens_per_second_per_gpu:5325.029106193521 mfu:25.11907083756641 +Step 13984 | loss:1.0164486169815063 lr:6.784652051722622e-05 tokens_per_second_per_gpu:4221.245644507126 mfu:19.91233592391367 +Step 13985 | loss:0.8760900497436523 lr:6.784246985106408e-05 tokens_per_second_per_gpu:5360.377920129449 mfu:25.28581722403175 +Step 13986 | loss:1.1401407718658447 lr:6.783841905070644e-05 tokens_per_second_per_gpu:3250.9787961346715 mfu:15.335421655546055 +Step 13987 | loss:0.9085959792137146 lr:6.783436811618376e-05 tokens_per_second_per_gpu:4521.860345926905 mfu:21.330386760667707 +Step 13988 | loss:1.2187390327453613 lr:6.783031704752654e-05 tokens_per_second_per_gpu:4292.463356431094 mfu:20.248282022053658 +Step 13989 | loss:1.1325438022613525 lr:6.782626584476524e-05 tokens_per_second_per_gpu:3814.1872883346487 mfu:17.99217220653094 +Step 13990 | loss:0.9400736689567566 lr:6.78222145079303e-05 tokens_per_second_per_gpu:5318.246096010454 mfu:25.087074221232005 +Step 13991 | loss:1.0515971183776855 lr:6.781816303705222e-05 tokens_per_second_per_gpu:3704.966061136873 mfu:17.476957042776085 +Step 13992 | loss:0.9946157336235046 lr:6.781411143216147e-05 tokens_per_second_per_gpu:5339.232680877643 mfu:25.186071522731147 +Step 13993 | loss:1.1079795360565186 lr:6.78100596932885e-05 tokens_per_second_per_gpu:3238.628829055557 mfu:15.27716475371221 +Step 13994 | loss:1.0791869163513184 lr:6.78060078204638e-05 tokens_per_second_per_gpu:4658.38271958848 mfu:21.97438609034821 +Step 13995 | loss:1.0390663146972656 lr:6.780195581371784e-05 tokens_per_second_per_gpu:4568.208604991595 mfu:21.549019406504275 +Step 13996 | loss:1.017572283744812 lr:6.779790367308111e-05 tokens_per_second_per_gpu:3172.176445330511 mfu:14.963697521735371 +Step 13997 | loss:1.1402385234832764 lr:6.779385139858407e-05 tokens_per_second_per_gpu:4520.385739199604 mfu:21.323430789141177 +Step 13998 | loss:1.1957566738128662 lr:6.778979899025719e-05 tokens_per_second_per_gpu:4577.403902241927 mfu:21.59239519251358 +Step 13999 | loss:0.9259415864944458 lr:6.7785746448131e-05 tokens_per_second_per_gpu:5354.691805585496 mfu:25.25899485157674 +Step 14000 | loss:1.0854127407073975 lr:6.77816937722359e-05 tokens_per_second_per_gpu:4703.330903307991 mfu:22.18641434190409 +Step 14001 | loss:1.008568286895752 lr:6.777764096260242e-05 tokens_per_second_per_gpu:5315.916684567254 mfu:25.07608599001586 +Step 14002 | loss:1.0610368251800537 lr:6.777358801926105e-05 tokens_per_second_per_gpu:3776.2611716604138 mfu:17.813268243315814 +Step 14003 | loss:1.065175175666809 lr:6.776953494224226e-05 tokens_per_second_per_gpu:5406.765974200737 mfu:25.50463758970417 +Step 14004 | loss:1.0877230167388916 lr:6.776548173157652e-05 tokens_per_second_per_gpu:4091.7386598345593 mfu:19.301429381043107 +Step 14005 | loss:1.0962293148040771 lr:6.776142838729431e-05 tokens_per_second_per_gpu:5347.069025145767 mfu:25.223036895662087 +Step 14006 | loss:1.0959303379058838 lr:6.775737490942616e-05 tokens_per_second_per_gpu:4080.679348468177 mfu:19.249260722414792 +Step 14007 | loss:1.1329143047332764 lr:6.77533212980025e-05 tokens_per_second_per_gpu:4734.521229345601 mfu:22.33354443144186 +Step 14008 | loss:0.9688661694526672 lr:6.774926755305387e-05 tokens_per_second_per_gpu:3568.649427291394 mfu:16.833928222910938 +Step 14009 | loss:1.0743664503097534 lr:6.774521367461072e-05 tokens_per_second_per_gpu:5351.031230289754 mfu:25.241727293348177 +Step 14010 | loss:0.9800020456314087 lr:6.774115966270355e-05 tokens_per_second_per_gpu:5271.031964295995 mfu:24.864357106373372 +Step 14011 | loss:0.8916282653808594 lr:6.773710551736286e-05 tokens_per_second_per_gpu:4558.276630154896 mfu:21.50216858662992 +Step 14012 | loss:0.9908738732337952 lr:6.773305123861913e-05 tokens_per_second_per_gpu:5356.288369807138 mfu:25.266526117412173 +Step 14013 | loss:0.9353235363960266 lr:6.772899682650286e-05 tokens_per_second_per_gpu:3541.3956606363354 mfu:16.705367555626946 +Step 14014 | loss:1.1161710023880005 lr:6.772494228104455e-05 tokens_per_second_per_gpu:5256.915549983164 mfu:24.797767571551105 +Step 14015 | loss:0.9378745555877686 lr:6.77208876022747e-05 tokens_per_second_per_gpu:3868.568454461044 mfu:18.248697445532844 +Step 14016 | loss:1.180871844291687 lr:6.771683279022376e-05 tokens_per_second_per_gpu:5372.70165645066 mfu:25.343950390905135 +Step 14017 | loss:0.9287143349647522 lr:6.771277784492228e-05 tokens_per_second_per_gpu:5328.6836885101175 mfu:25.136310125889832 +Step 14018 | loss:1.0305739641189575 lr:6.770872276640075e-05 tokens_per_second_per_gpu:3861.260112299021 mfu:18.21422274862283 +Step 14019 | loss:1.047223448753357 lr:6.770466755468964e-05 tokens_per_second_per_gpu:4211.276564459975 mfu:19.865310072430816 +Step 14020 | loss:0.8285358548164368 lr:6.770061220981947e-05 tokens_per_second_per_gpu:5302.675979285361 mfu:25.013627324103894 +Step 14021 | loss:1.0784831047058105 lr:6.769655673182074e-05 tokens_per_second_per_gpu:3775.1232558622687 mfu:17.80790050034751 +Step 14022 | loss:1.150380253791809 lr:6.769250112072394e-05 tokens_per_second_per_gpu:4660.746594380556 mfu:21.985536891061063 +Step 14023 | loss:1.1057552099227905 lr:6.768844537655957e-05 tokens_per_second_per_gpu:5287.8440453469175 mfu:24.943662561127383 +Step 14024 | loss:0.9812779426574707 lr:6.768438949935817e-05 tokens_per_second_per_gpu:3995.5586396891413 mfu:18.847731816013585 +Step 14025 | loss:0.8966883420944214 lr:6.768033348915021e-05 tokens_per_second_per_gpu:4463.355947791367 mfu:21.05441153278275 +Step 14026 | loss:1.0115290880203247 lr:6.767627734596621e-05 tokens_per_second_per_gpu:5353.683044910722 mfu:25.254236355361513 +Step 14027 | loss:1.2411471605300903 lr:6.767222106983666e-05 tokens_per_second_per_gpu:4697.4143337969745 mfu:22.15850487404977 +Step 14028 | loss:1.033743977546692 lr:6.76681646607921e-05 tokens_per_second_per_gpu:5330.1449632563545 mfu:25.143203208187014 +Step 14029 | loss:1.069331169128418 lr:6.766410811886302e-05 tokens_per_second_per_gpu:5346.211677823046 mfu:25.218992642062272 +Step 14030 | loss:1.0712931156158447 lr:6.76600514440799e-05 tokens_per_second_per_gpu:4765.893935948971 mfu:22.481534841225663 +Step 14031 | loss:1.0617685317993164 lr:6.76559946364733e-05 tokens_per_second_per_gpu:4229.9595156799 mfu:19.953440740975378 +Step 14032 | loss:0.9859328269958496 lr:6.765193769607371e-05 tokens_per_second_per_gpu:5228.428490196122 mfu:24.663389250142043 +Step 14033 | loss:1.070452332496643 lr:6.764788062291164e-05 tokens_per_second_per_gpu:4711.505254481335 mfu:22.22497415107688 +Step 14034 | loss:1.0306874513626099 lr:6.764382341701761e-05 tokens_per_second_per_gpu:4975.019209632052 mfu:23.46801443763969 +Step 14035 | loss:1.0768316984176636 lr:6.763976607842213e-05 tokens_per_second_per_gpu:5361.906221117843 mfu:25.293026480549045 +Step 14036 | loss:0.9041426181793213 lr:6.763570860715573e-05 tokens_per_second_per_gpu:3989.6509054896455 mfu:18.81986402583117 +Step 14037 | loss:1.0421156883239746 lr:6.76316510032489e-05 tokens_per_second_per_gpu:5381.290598110601 mfu:25.384465894139637 +Step 14038 | loss:1.1226270198822021 lr:6.762759326673218e-05 tokens_per_second_per_gpu:4710.054107413579 mfu:22.21812884276705 +Step 14039 | loss:1.0094670057296753 lr:6.762353539763608e-05 tokens_per_second_per_gpu:3260.9230058180947 mfu:15.38233018927989 +Step 14040 | loss:1.0427193641662598 lr:6.761947739599112e-05 tokens_per_second_per_gpu:4685.239888849864 mfu:22.10107594005975 +Step 14041 | loss:0.9681268930435181 lr:6.761541926182783e-05 tokens_per_second_per_gpu:5293.797603704166 mfu:24.971746511680323 +Step 14042 | loss:1.0146198272705078 lr:6.761136099517672e-05 tokens_per_second_per_gpu:4310.132673897798 mfu:20.331631207240523 +Step 14043 | loss:1.0181446075439453 lr:6.760730259606831e-05 tokens_per_second_per_gpu:5311.620881158326 mfu:25.055821952396304 +Step 14044 | loss:0.8680771589279175 lr:6.760324406453315e-05 tokens_per_second_per_gpu:3628.6241718011756 mfu:17.116839325510156 +Step 14045 | loss:0.975825846195221 lr:6.759918540060173e-05 tokens_per_second_per_gpu:4957.927924560909 mfu:23.387391929885187 +Step 14046 | loss:0.9426442980766296 lr:6.759512660430459e-05 tokens_per_second_per_gpu:4851.326541006906 mfu:22.884535015589424 +Step 14047 | loss:1.02970290184021 lr:6.759106767567227e-05 tokens_per_second_per_gpu:5349.969340522148 mfu:25.236718178137302 +Step 14048 | loss:1.1859084367752075 lr:6.758700861473526e-05 tokens_per_second_per_gpu:5283.19354489259 mfu:24.92172535702712 +Step 14049 | loss:1.059023141860962 lr:6.758294942152414e-05 tokens_per_second_per_gpu:4762.949091032534 mfu:22.467643505312473 +Step 14050 | loss:1.0856536626815796 lr:6.75788900960694e-05 tokens_per_second_per_gpu:3735.5497131888114 mfu:17.621225347614192 +Step 14051 | loss:1.1424978971481323 lr:6.757483063840158e-05 tokens_per_second_per_gpu:4116.132637796109 mfu:19.41649993712955 +Step 14052 | loss:1.0997108221054077 lr:6.757077104855122e-05 tokens_per_second_per_gpu:4703.818276409743 mfu:22.188713363979474 +Step 14053 | loss:1.0783592462539673 lr:6.756671132654886e-05 tokens_per_second_per_gpu:5317.491283478324 mfu:25.08351364173365 +Step 14054 | loss:0.9856555461883545 lr:6.756265147242503e-05 tokens_per_second_per_gpu:5299.743195307216 mfu:24.999792881694443 +Step 14055 | loss:1.051506757736206 lr:6.755859148621024e-05 tokens_per_second_per_gpu:4984.200668492571 mfu:23.511324945603178 +Step 14056 | loss:1.019917368888855 lr:6.755453136793504e-05 tokens_per_second_per_gpu:5058.36075326488 mfu:23.861150718489363 +Step 14057 | loss:1.0725712776184082 lr:6.755047111762998e-05 tokens_per_second_per_gpu:4751.7155461485945 mfu:22.414652957454535 +Step 14058 | loss:0.9813076257705688 lr:6.754641073532559e-05 tokens_per_second_per_gpu:4029.8551589741232 mfu:19.009514349070436 +Step 14059 | loss:0.8656866550445557 lr:6.754235022105239e-05 tokens_per_second_per_gpu:5303.610180485839 mfu:25.01803411055771 +Step 14060 | loss:1.1490558385849 lr:6.753828957484094e-05 tokens_per_second_per_gpu:4625.029139379849 mfu:21.81705156179651 +Step 14061 | loss:0.9384263157844543 lr:6.753422879672178e-05 tokens_per_second_per_gpu:5326.761447437729 mfu:25.127242586783233 +Step 14062 | loss:1.1625162363052368 lr:6.753016788672545e-05 tokens_per_second_per_gpu:5325.96092559326 mfu:25.12346638865976 +Step 14063 | loss:0.952243983745575 lr:6.75261068448825e-05 tokens_per_second_per_gpu:5199.430471102124 mfu:24.526600646503418 +Step 14064 | loss:0.9483004808425903 lr:6.752204567122343e-05 tokens_per_second_per_gpu:4324.352888936696 mfu:20.398710387797877 +Step 14065 | loss:1.1175737380981445 lr:6.751798436577887e-05 tokens_per_second_per_gpu:3878.0116546073336 mfu:18.293242631798975 +Step 14066 | loss:1.0292764902114868 lr:6.751392292857927e-05 tokens_per_second_per_gpu:2967.6739934545426 mfu:13.999024564520218 +Step 14067 | loss:1.0673561096191406 lr:6.750986135965524e-05 tokens_per_second_per_gpu:3806.144313631748 mfu:17.95423212258437 +Step 14068 | loss:1.0552375316619873 lr:6.750579965903731e-05 tokens_per_second_per_gpu:4804.760966554398 mfu:22.66487725599069 +Step 14069 | loss:1.0836166143417358 lr:6.750173782675602e-05 tokens_per_second_per_gpu:4386.68910505873 mfu:20.692760954900695 +Step 14070 | loss:1.2293896675109863 lr:6.749767586284192e-05 tokens_per_second_per_gpu:3231.453114773566 mfu:15.243315685140955 +Step 14071 | loss:1.0497976541519165 lr:6.749361376732556e-05 tokens_per_second_per_gpu:5359.618696049119 mfu:25.282235834507848 +Step 14072 | loss:0.9980278611183167 lr:6.748955154023753e-05 tokens_per_second_per_gpu:3890.834335322632 mfu:18.353729404507966 +Step 14073 | loss:1.0702563524246216 lr:6.748548918160834e-05 tokens_per_second_per_gpu:3542.2938367479787 mfu:16.709604405590213 +Step 14074 | loss:1.0874714851379395 lr:6.748142669146855e-05 tokens_per_second_per_gpu:5339.245641335373 mfu:25.18613265942154 +Step 14075 | loss:0.9516269564628601 lr:6.747736406984872e-05 tokens_per_second_per_gpu:5268.91615421991 mfu:24.854376469250056 +Step 14076 | loss:0.9895967245101929 lr:6.747330131677941e-05 tokens_per_second_per_gpu:5337.028440671301 mfu:25.175673745595667 +Step 14077 | loss:1.0657814741134644 lr:6.746923843229116e-05 tokens_per_second_per_gpu:4093.5795747463967 mfu:19.310113290774712 +Step 14078 | loss:0.837672233581543 lr:6.746517541641455e-05 tokens_per_second_per_gpu:4395.165616801677 mfu:20.73274610703421 +Step 14079 | loss:1.0823270082473755 lr:6.746111226918011e-05 tokens_per_second_per_gpu:5363.365987755635 mfu:25.299912448841376 +Step 14080 | loss:0.9724040627479553 lr:6.745704899061843e-05 tokens_per_second_per_gpu:4582.997065643842 mfu:21.618779098572492 +Step 14081 | loss:1.00539231300354 lr:6.745298558076006e-05 tokens_per_second_per_gpu:5344.504773206877 mfu:25.210940881759846 +Step 14082 | loss:1.1314387321472168 lr:6.744892203963556e-05 tokens_per_second_per_gpu:4591.376596078485 mfu:21.65830677332802 +Step 14083 | loss:1.0465373992919922 lr:6.744485836727549e-05 tokens_per_second_per_gpu:5297.095296522729 mfu:24.9873022913499 +Step 14084 | loss:1.0701603889465332 lr:6.744079456371039e-05 tokens_per_second_per_gpu:3628.2216358768096 mfu:17.114940494875214 +Step 14085 | loss:0.8917608857154846 lr:6.743673062897087e-05 tokens_per_second_per_gpu:5361.3315789205335 mfu:25.290315795260263 +Step 14086 | loss:0.9651064872741699 lr:6.743266656308748e-05 tokens_per_second_per_gpu:4365.543932431781 mfu:20.59301556788091 +Step 14087 | loss:0.9975318312644958 lr:6.742860236609077e-05 tokens_per_second_per_gpu:4825.472277133757 mfu:22.762576041707497 +Step 14088 | loss:1.0822792053222656 lr:6.74245380380113e-05 tokens_per_second_per_gpu:4500.658318974963 mfu:21.230373182096578 +Step 14089 | loss:1.0850415229797363 lr:6.742047357887969e-05 tokens_per_second_per_gpu:4997.889304389144 mfu:23.57589657664943 +Step 14090 | loss:0.9616395235061646 lr:6.741640898872645e-05 tokens_per_second_per_gpu:5284.301263931779 mfu:24.926950656731336 +Step 14091 | loss:1.0441391468048096 lr:6.74123442675822e-05 tokens_per_second_per_gpu:5292.487989746111 mfu:24.965568839196997 +Step 14092 | loss:0.9005343914031982 lr:6.740827941547747e-05 tokens_per_second_per_gpu:4806.234935222595 mfu:22.67183021768356 +Step 14093 | loss:1.0889368057250977 lr:6.740421443244285e-05 tokens_per_second_per_gpu:4512.761668178837 mfu:21.287466745335568 +Step 14094 | loss:1.1055444478988647 lr:6.740014931850892e-05 tokens_per_second_per_gpu:4929.815437199945 mfu:23.25478053051858 +Step 14095 | loss:0.9852477312088013 lr:6.739608407370624e-05 tokens_per_second_per_gpu:5326.586411908584 mfu:25.126416914328114 +Step 14096 | loss:0.9595771431922913 lr:6.73920186980654e-05 tokens_per_second_per_gpu:3704.219040604484 mfu:17.47343321946996 +Step 14097 | loss:1.1016267538070679 lr:6.738795319161694e-05 tokens_per_second_per_gpu:5315.139485564341 mfu:25.072419809715292 +Step 14098 | loss:0.9176070094108582 lr:6.73838875543915e-05 tokens_per_second_per_gpu:4516.194823035129 mfu:21.30366152254932 +Step 14099 | loss:0.9727416634559631 lr:6.737982178641961e-05 tokens_per_second_per_gpu:4096.213640720598 mfu:19.32253862939302 +Step 14100 | loss:0.9813141822814941 lr:6.737575588773187e-05 tokens_per_second_per_gpu:5328.215533987877 mfu:25.134101761133028 +Step 14101 | loss:1.2028577327728271 lr:6.737168985835885e-05 tokens_per_second_per_gpu:4352.668254506705 mfu:20.53227879829114 +Step 14102 | loss:1.0473171472549438 lr:6.736762369833112e-05 tokens_per_second_per_gpu:4613.976920423768 mfu:21.764916359277674 +Step 14103 | loss:1.068945050239563 lr:6.736355740767928e-05 tokens_per_second_per_gpu:4588.280503054934 mfu:21.643701974287993 +Step 14104 | loss:1.2050973176956177 lr:6.735949098643393e-05 tokens_per_second_per_gpu:5323.81554823147 mfu:25.11334627760507 +Step 14105 | loss:0.989656925201416 lr:6.735542443462561e-05 tokens_per_second_per_gpu:5287.517091361039 mfu:24.942120263391892 +Step 14106 | loss:1.1407676935195923 lr:6.735135775228492e-05 tokens_per_second_per_gpu:4583.9124800205345 mfu:21.623097264373147 +Step 14107 | loss:1.050248384475708 lr:6.734729093944249e-05 tokens_per_second_per_gpu:5367.922208618195 mfu:25.321404920021514 +Step 14108 | loss:1.014289379119873 lr:6.734322399612885e-05 tokens_per_second_per_gpu:4722.993468081053 mfu:22.279166014718104 +Step 14109 | loss:1.0766937732696533 lr:6.733915692237462e-05 tokens_per_second_per_gpu:5227.087937212739 mfu:24.6570656329975 +Step 14110 | loss:0.9523569941520691 lr:6.733508971821036e-05 tokens_per_second_per_gpu:3917.8373866637044 mfu:18.48110740487944 +Step 14111 | loss:1.0024126768112183 lr:6.733102238366668e-05 tokens_per_second_per_gpu:5358.422247681444 mfu:25.27659198342238 +Step 14112 | loss:0.8691442012786865 lr:6.73269549187742e-05 tokens_per_second_per_gpu:5353.0853168255935 mfu:25.251416770001036 +Step 14113 | loss:1.0156974792480469 lr:6.732288732356344e-05 tokens_per_second_per_gpu:4183.661845729346 mfu:19.73504673262158 +Step 14114 | loss:1.1593748331069946 lr:6.731881959806506e-05 tokens_per_second_per_gpu:5338.816057885187 mfu:25.184106240992513 +Step 14115 | loss:0.9193452596664429 lr:6.731475174230961e-05 tokens_per_second_per_gpu:5338.816733510973 mfu:25.18410942803462 +Step 14116 | loss:1.0963109731674194 lr:6.731068375632773e-05 tokens_per_second_per_gpu:3649.778776924668 mfu:17.21662920722549 +Step 14117 | loss:1.0582014322280884 lr:6.730661564014997e-05 tokens_per_second_per_gpu:5242.410040123897 mfu:24.72934260664976 +Step 14118 | loss:0.8473547697067261 lr:6.730254739380693e-05 tokens_per_second_per_gpu:3869.870603930824 mfu:18.254839906751407 +Step 14119 | loss:1.0638155937194824 lr:6.729847901732925e-05 tokens_per_second_per_gpu:5155.905651248149 mfu:24.3212866451522 +Step 14120 | loss:1.1925350427627563 lr:6.729441051074749e-05 tokens_per_second_per_gpu:5365.364342241257 mfu:25.30933902790449 +Step 14121 | loss:0.9405436515808105 lr:6.729034187409227e-05 tokens_per_second_per_gpu:5359.216532483696 mfu:25.280338760353565 +Step 14122 | loss:0.9703977108001709 lr:6.728627310739417e-05 tokens_per_second_per_gpu:4449.435772698153 mfu:20.98874769184129 +Step 14123 | loss:0.8926826119422913 lr:6.728220421068381e-05 tokens_per_second_per_gpu:5343.227098082344 mfu:25.204913870203573 +Step 14124 | loss:1.0972462892532349 lr:6.727813518399178e-05 tokens_per_second_per_gpu:3765.5196827922196 mfu:17.762598807637563 +Step 14125 | loss:1.0492432117462158 lr:6.72740660273487e-05 tokens_per_second_per_gpu:3641.888267144054 mfu:17.179408326330545 +Step 14126 | loss:0.9727564454078674 lr:6.726999674078517e-05 tokens_per_second_per_gpu:5364.010891096695 mfu:25.30295456793301 +Step 14127 | loss:1.05580735206604 lr:6.726592732433178e-05 tokens_per_second_per_gpu:5316.171607933718 mfu:25.07728850702988 +Step 14128 | loss:1.1911003589630127 lr:6.726185777801915e-05 tokens_per_second_per_gpu:3838.6355935155725 mfu:18.107499033380357 +Step 14129 | loss:1.211918592453003 lr:6.725778810187789e-05 tokens_per_second_per_gpu:5382.0361208646245 mfu:25.38798265216966 +Step 14130 | loss:1.0254884958267212 lr:6.72537182959386e-05 tokens_per_second_per_gpu:5369.298412298524 mfu:25.327896707586184 +Step 14131 | loss:1.013771653175354 lr:6.72496483602319e-05 tokens_per_second_per_gpu:4143.805886584403 mfu:19.54703937320705 +Step 14132 | loss:1.1922539472579956 lr:6.724557829478838e-05 tokens_per_second_per_gpu:4976.221982903045 mfu:23.47368812437325 +Step 14133 | loss:1.0696232318878174 lr:6.724150809963867e-05 tokens_per_second_per_gpu:5324.21921115159 mfu:25.115250424471277 +Step 14134 | loss:1.0117099285125732 lr:6.72374377748134e-05 tokens_per_second_per_gpu:3932.466778115526 mfu:18.55011674038925 +Step 14135 | loss:1.1117807626724243 lr:6.723336732034313e-05 tokens_per_second_per_gpu:4188.1317237291905 mfu:19.756131909786678 +Step 14136 | loss:1.0144996643066406 lr:6.72292967362585e-05 tokens_per_second_per_gpu:5088.738193271185 mfu:24.004446285924814 +Step 14137 | loss:0.9192013740539551 lr:6.722522602259016e-05 tokens_per_second_per_gpu:3634.5332430559506 mfu:17.144713422810447 +Step 14138 | loss:1.1249885559082031 lr:6.722115517936869e-05 tokens_per_second_per_gpu:5333.1229861410875 mfu:25.1572510502374 +Step 14139 | loss:1.0192123651504517 lr:6.721708420662471e-05 tokens_per_second_per_gpu:4827.541669291858 mfu:22.772337717591967 +Step 14140 | loss:1.1257978677749634 lr:6.721301310438884e-05 tokens_per_second_per_gpu:4427.494043705993 mfu:20.88524481254027 +Step 14141 | loss:1.0292612314224243 lr:6.72089418726917e-05 tokens_per_second_per_gpu:4066.281816463225 mfu:19.181345107475167 +Step 14142 | loss:1.0175950527191162 lr:6.720487051156392e-05 tokens_per_second_per_gpu:5329.693927383743 mfu:25.141075595021253 +Step 14143 | loss:1.0715100765228271 lr:6.72007990210361e-05 tokens_per_second_per_gpu:4420.049561126312 mfu:20.850127918052216 +Step 14144 | loss:1.0750937461853027 lr:6.719672740113889e-05 tokens_per_second_per_gpu:5319.562273772071 mfu:25.09328285629664 +Step 14145 | loss:1.042911410331726 lr:6.719265565190289e-05 tokens_per_second_per_gpu:4200.504033770476 mfu:19.814494207185184 +Step 14146 | loss:1.1130695343017578 lr:6.718858377335874e-05 tokens_per_second_per_gpu:5377.103430244615 mfu:25.36471430891127 +Step 14147 | loss:0.9385056495666504 lr:6.718451176553704e-05 tokens_per_second_per_gpu:4348.942283246807 mfu:20.514702756141954 +Step 14148 | loss:1.0026514530181885 lr:6.718043962846847e-05 tokens_per_second_per_gpu:5315.345299818618 mfu:25.07339067066825 +Step 14149 | loss:1.0972687005996704 lr:6.71763673621836e-05 tokens_per_second_per_gpu:5371.38217601299 mfu:25.33772617655407 +Step 14150 | loss:0.9345585107803345 lr:6.717229496671307e-05 tokens_per_second_per_gpu:5364.446693188802 mfu:25.30501032075698 +Step 14151 | loss:1.0351531505584717 lr:6.716822244208754e-05 tokens_per_second_per_gpu:5065.476005157846 mfu:23.894714575655637 +Step 14152 | loss:1.1666456460952759 lr:6.71641497883376e-05 tokens_per_second_per_gpu:4975.7265349818 mfu:23.471351011996607 +Step 14153 | loss:1.0130455493927002 lr:6.71600770054939e-05 tokens_per_second_per_gpu:5323.202673903537 mfu:25.110455244833073 +Step 14154 | loss:1.0917004346847534 lr:6.715600409358707e-05 tokens_per_second_per_gpu:3340.5419688574684 mfu:15.757906422332487 +Step 14155 | loss:1.0130337476730347 lr:6.715193105264776e-05 tokens_per_second_per_gpu:5332.530914866749 mfu:25.15445814901857 +Step 14156 | loss:0.90227872133255 lr:6.714785788270658e-05 tokens_per_second_per_gpu:4057.416509934974 mfu:19.13952594400414 +Step 14157 | loss:0.9729655981063843 lr:6.714378458379418e-05 tokens_per_second_per_gpu:4770.677024417218 mfu:22.504097485609424 +Step 14158 | loss:1.1159892082214355 lr:6.713971115594116e-05 tokens_per_second_per_gpu:3885.91611741506 mfu:18.330529331503445 +Step 14159 | loss:1.0173324346542358 lr:6.713563759917822e-05 tokens_per_second_per_gpu:2891.2003109807006 mfu:13.638285156535433 +Step 14160 | loss:0.9927372932434082 lr:6.713156391353596e-05 tokens_per_second_per_gpu:4131.105604329679 mfu:19.487129974919924 +Step 14161 | loss:1.027953028678894 lr:6.7127490099045e-05 tokens_per_second_per_gpu:5167.536095027215 mfu:24.37614943281638 +Step 14162 | loss:1.0802874565124512 lr:6.712341615573601e-05 tokens_per_second_per_gpu:4262.589330660839 mfu:20.107361145460974 +Step 14163 | loss:1.16571044921875 lr:6.711934208363962e-05 tokens_per_second_per_gpu:4362.553083679838 mfu:20.578907224026576 +Step 14164 | loss:0.997158944606781 lr:6.711526788278648e-05 tokens_per_second_per_gpu:4331.121681245755 mfu:20.430639935994762 +Step 14165 | loss:1.0149734020233154 lr:6.711119355320722e-05 tokens_per_second_per_gpu:4588.193705578578 mfu:21.643292535782845 +Step 14166 | loss:1.1178184747695923 lr:6.71071190949325e-05 tokens_per_second_per_gpu:4561.652835666856 mfu:21.518094724069893 +Step 14167 | loss:1.1729979515075684 lr:6.710304450799294e-05 tokens_per_second_per_gpu:4474.112523989725 mfu:21.10515213797155 +Step 14168 | loss:0.9543499946594238 lr:6.709896979241922e-05 tokens_per_second_per_gpu:4983.87142186461 mfu:23.50977183308382 +Step 14169 | loss:0.9994009137153625 lr:6.709489494824194e-05 tokens_per_second_per_gpu:5354.237715778859 mfu:25.256852832483123 +Step 14170 | loss:1.0216788053512573 lr:6.70908199754918e-05 tokens_per_second_per_gpu:5228.595447038087 mfu:24.664176814051366 +Step 14171 | loss:0.9872812032699585 lr:6.708674487419942e-05 tokens_per_second_per_gpu:5330.279318915722 mfu:25.143836986755687 +Step 14172 | loss:1.0279699563980103 lr:6.708266964439543e-05 tokens_per_second_per_gpu:5326.301256660973 mfu:25.12507178837263 +Step 14173 | loss:1.081045389175415 lr:6.707859428611053e-05 tokens_per_second_per_gpu:5037.374530525964 mfu:23.762155125211073 +Step 14174 | loss:1.2055975198745728 lr:6.707451879937531e-05 tokens_per_second_per_gpu:5379.988898506642 mfu:25.378325555014904 +Step 14175 | loss:1.2050386667251587 lr:6.707044318422048e-05 tokens_per_second_per_gpu:4545.475188197705 mfu:21.441782000766633 +Step 14176 | loss:1.0023078918457031 lr:6.706636744067665e-05 tokens_per_second_per_gpu:4788.3520126015 mfu:22.587473420538632 +Step 14177 | loss:0.9894129633903503 lr:6.706229156877451e-05 tokens_per_second_per_gpu:4733.27992847518 mfu:22.32768900344832 +Step 14178 | loss:1.0759608745574951 lr:6.70582155685447e-05 tokens_per_second_per_gpu:5379.597618975766 mfu:25.37647982270859 +Step 14179 | loss:0.9828417897224426 lr:6.705413944001785e-05 tokens_per_second_per_gpu:3243.9789321524145 mfu:15.302402102841802 +Step 14180 | loss:0.9843165874481201 lr:6.705006318322467e-05 tokens_per_second_per_gpu:5269.183254754366 mfu:24.855636428043134 +Step 14181 | loss:1.1454455852508545 lr:6.704598679819578e-05 tokens_per_second_per_gpu:5321.182351702801 mfu:25.100925040309686 +Step 14182 | loss:1.0464338064193726 lr:6.704191028496185e-05 tokens_per_second_per_gpu:4412.533170492804 mfu:20.814671821005472 +Step 14183 | loss:1.0418617725372314 lr:6.703783364355351e-05 tokens_per_second_per_gpu:3899.4314336117286 mfu:18.394283383953855 +Step 14184 | loss:1.0867773294448853 lr:6.70337568740015e-05 tokens_per_second_per_gpu:4910.641084062244 mfu:23.16433183528605 +Step 14185 | loss:1.1422998905181885 lr:6.702967997633638e-05 tokens_per_second_per_gpu:5178.005809259242 mfu:24.425536861166297 +Step 14186 | loss:1.129869818687439 lr:6.70256029505889e-05 tokens_per_second_per_gpu:4249.433020461447 mfu:20.045300585552457 +Step 14187 | loss:0.993760883808136 lr:6.702152579678967e-05 tokens_per_second_per_gpu:3625.71665301621 mfu:17.103124063327616 +Step 14188 | loss:1.0712579488754272 lr:6.701744851496939e-05 tokens_per_second_per_gpu:5042.60226836388 mfu:23.786815256536574 +Step 14189 | loss:1.0058403015136719 lr:6.701337110515868e-05 tokens_per_second_per_gpu:5341.068545866907 mfu:25.194731611116367 +Step 14190 | loss:0.9924570918083191 lr:6.700929356738825e-05 tokens_per_second_per_gpu:5249.651875728491 mfu:24.763503580781194 +Step 14191 | loss:0.9756236672401428 lr:6.700521590168877e-05 tokens_per_second_per_gpu:4961.796916933379 mfu:23.405642627024296 +Step 14192 | loss:1.0177706480026245 lr:6.700113810809086e-05 tokens_per_second_per_gpu:4143.077380043746 mfu:19.543602883558737 +Step 14193 | loss:0.9876630306243896 lr:6.699706018662524e-05 tokens_per_second_per_gpu:5360.233045384509 mfu:25.285133825141685 +Step 14194 | loss:1.1071912050247192 lr:6.699298213732256e-05 tokens_per_second_per_gpu:4359.489822246872 mfu:20.564457297200978 +Step 14195 | loss:0.9605517387390137 lr:6.698890396021349e-05 tokens_per_second_per_gpu:5341.480330661535 mfu:25.196674070250356 +Step 14196 | loss:1.1000818014144897 lr:6.698482565532871e-05 tokens_per_second_per_gpu:4334.614502690413 mfu:20.447116170686 +Step 14197 | loss:1.0812487602233887 lr:6.698074722269888e-05 tokens_per_second_per_gpu:5296.86861622896 mfu:24.986233001728543 +Step 14198 | loss:1.0912127494812012 lr:6.69766686623547e-05 tokens_per_second_per_gpu:5319.517516510741 mfu:25.09307172865857 +Step 14199 | loss:1.052187442779541 lr:6.69725899743268e-05 tokens_per_second_per_gpu:3552.2947064719037 mfu:16.756780214402294 +Step 14200 | loss:1.041494369506836 lr:6.696851115864591e-05 tokens_per_second_per_gpu:5339.8526381490465 mfu:25.18899596695597 +Step 14201 | loss:0.8956068158149719 lr:6.696443221534265e-05 tokens_per_second_per_gpu:4285.70761560621 mfu:20.21641403993444 +Step 14202 | loss:1.1798015832901 lr:6.696035314444778e-05 tokens_per_second_per_gpu:4429.932283871265 mfu:20.89674640740653 +Step 14203 | loss:0.9378952383995056 lr:6.695627394599188e-05 tokens_per_second_per_gpu:3771.959978770146 mfu:17.792978782593245 +Step 14204 | loss:1.0161020755767822 lr:6.695219462000572e-05 tokens_per_second_per_gpu:4958.624359403227 mfu:23.390677131860436 +Step 14205 | loss:0.9900215268135071 lr:6.694811516651993e-05 tokens_per_second_per_gpu:5319.1041783859 mfu:25.091121942200935 +Step 14206 | loss:0.977218508720398 lr:6.694403558556519e-05 tokens_per_second_per_gpu:5370.956901664649 mfu:25.33572008489366 +Step 14207 | loss:1.0315210819244385 lr:6.69399558771722e-05 tokens_per_second_per_gpu:5345.433117612759 mfu:25.215320040714282 +Step 14208 | loss:1.0948092937469482 lr:6.693587604137164e-05 tokens_per_second_per_gpu:4041.1870453765055 mfu:19.06296879064966 +Step 14209 | loss:0.9299216866493225 lr:6.693179607819422e-05 tokens_per_second_per_gpu:5074.496665804903 mfu:23.937266571011218 +Step 14210 | loss:1.0839941501617432 lr:6.692771598767058e-05 tokens_per_second_per_gpu:4540.428405606906 mfu:21.417975466215932 +Step 14211 | loss:1.050439476966858 lr:6.692363576983143e-05 tokens_per_second_per_gpu:4277.373985030381 mfu:20.177102882644256 +Step 14212 | loss:1.272772192955017 lr:6.691955542470746e-05 tokens_per_second_per_gpu:3783.5356363105952 mfu:17.847583134222376 +Step 14213 | loss:1.1510812044143677 lr:6.691547495232934e-05 tokens_per_second_per_gpu:5383.439749232603 mfu:25.394603806665113 +Step 14214 | loss:1.0235447883605957 lr:6.69113943527278e-05 tokens_per_second_per_gpu:3859.6065743292384 mfu:18.206422727896697 +Step 14215 | loss:1.1359320878982544 lr:6.690731362593348e-05 tokens_per_second_per_gpu:4190.357291171977 mfu:19.76663029113526 +Step 14216 | loss:1.1253976821899414 lr:6.690323277197711e-05 tokens_per_second_per_gpu:5357.304493203122 mfu:25.271319344839554 +Step 14217 | loss:1.0002446174621582 lr:6.689915179088936e-05 tokens_per_second_per_gpu:4859.000157325225 mfu:22.920732773016887 +Step 14218 | loss:1.101731538772583 lr:6.689507068270095e-05 tokens_per_second_per_gpu:3978.6805483517574 mfu:18.76811497947582 +Step 14219 | loss:0.9554999470710754 lr:6.689098944744254e-05 tokens_per_second_per_gpu:4320.523199959648 mfu:20.380645091480822 +Step 14220 | loss:0.9517959952354431 lr:6.688690808514485e-05 tokens_per_second_per_gpu:3390.317566880359 mfu:15.992706410799213 +Step 14221 | loss:1.005877137184143 lr:6.688282659583857e-05 tokens_per_second_per_gpu:5324.943191666477 mfu:25.118665564083813 +Step 14222 | loss:0.9662727117538452 lr:6.68787449795544e-05 tokens_per_second_per_gpu:4576.474323688788 mfu:21.588010211875968 +Step 14223 | loss:1.03061842918396 lr:6.687466323632304e-05 tokens_per_second_per_gpu:5351.563561555724 mfu:25.244238390754838 +Step 14224 | loss:0.9676759243011475 lr:6.687058136617517e-05 tokens_per_second_per_gpu:4593.214149572577 mfu:21.666974826678608 +Step 14225 | loss:1.183957576751709 lr:6.686649936914152e-05 tokens_per_second_per_gpu:5155.271565448647 mfu:24.318295554249755 +Step 14226 | loss:1.009106159210205 lr:6.686241724525275e-05 tokens_per_second_per_gpu:5038.423029747628 mfu:23.76710107493208 +Step 14227 | loss:1.0376684665679932 lr:6.685833499453961e-05 tokens_per_second_per_gpu:4724.999388150867 mfu:22.28862828193267 +Step 14228 | loss:0.9808339476585388 lr:6.685425261703278e-05 tokens_per_second_per_gpu:4227.133758141814 mfu:19.940111160544276 +Step 14229 | loss:0.9541595578193665 lr:6.685017011276296e-05 tokens_per_second_per_gpu:4779.520596539478 mfu:22.545814124179124 +Step 14230 | loss:1.0563305616378784 lr:6.684608748176087e-05 tokens_per_second_per_gpu:4211.2418683894 mfu:19.865146405146152 +Step 14231 | loss:0.9817374348640442 lr:6.68420047240572e-05 tokens_per_second_per_gpu:5277.635554383201 mfu:24.895507367503374 +Step 14232 | loss:1.0622715950012207 lr:6.683792183968266e-05 tokens_per_second_per_gpu:4925.651370640467 mfu:23.235137918094505 +Step 14233 | loss:0.9928628206253052 lr:6.683383882866795e-05 tokens_per_second_per_gpu:4269.846498924496 mfu:20.14159444636123 +Step 14234 | loss:0.9312655329704285 lr:6.68297556910438e-05 tokens_per_second_per_gpu:3157.4178485159778 mfu:14.894078702484947 +Step 14235 | loss:1.180069923400879 lr:6.682567242684092e-05 tokens_per_second_per_gpu:5364.813400696181 mfu:25.306740142636045 +Step 14236 | loss:0.9376345276832581 lr:6.682158903609002e-05 tokens_per_second_per_gpu:4349.053243169375 mfu:20.515226173028417 +Step 14237 | loss:1.1033393144607544 lr:6.681750551882177e-05 tokens_per_second_per_gpu:4471.879986067018 mfu:21.094620875679894 +Step 14238 | loss:1.0294286012649536 lr:6.681342187506694e-05 tokens_per_second_per_gpu:3634.851021263548 mfu:17.14621243683424 +Step 14239 | loss:0.9033608436584473 lr:6.680933810485621e-05 tokens_per_second_per_gpu:4417.730622445416 mfu:20.839189088642687 +Step 14240 | loss:1.1698668003082275 lr:6.68052542082203e-05 tokens_per_second_per_gpu:4938.93336269525 mfu:23.29779134075844 +Step 14241 | loss:1.0183943510055542 lr:6.680117018518994e-05 tokens_per_second_per_gpu:4542.185102801286 mfu:21.42626210660527 +Step 14242 | loss:1.0750349760055542 lr:6.679708603579583e-05 tokens_per_second_per_gpu:4623.397873731652 mfu:21.809356603411345 +Step 14243 | loss:0.9654276967048645 lr:6.679300176006869e-05 tokens_per_second_per_gpu:5345.785896486905 mfu:25.216984158853908 +Step 14244 | loss:1.1327348947525024 lr:6.678891735803922e-05 tokens_per_second_per_gpu:3961.311905996729 mfu:18.68618413009119 +Step 14245 | loss:1.1270160675048828 lr:6.678483282973821e-05 tokens_per_second_per_gpu:4577.763336689328 mfu:21.59409070612781 +Step 14246 | loss:1.0105277299880981 lr:6.67807481751963e-05 tokens_per_second_per_gpu:3203.7996336439337 mfu:15.112869496482077 +Step 14247 | loss:0.9848178029060364 lr:6.677666339444424e-05 tokens_per_second_per_gpu:4533.583046990442 mfu:21.38568474168346 +Step 14248 | loss:1.0166258811950684 lr:6.677257848751277e-05 tokens_per_second_per_gpu:4599.450538814585 mfu:21.696392938770888 +Step 14249 | loss:1.2061893939971924 lr:6.676849345443257e-05 tokens_per_second_per_gpu:4561.584910323828 mfu:21.517774308637573 +Step 14250 | loss:1.0906919240951538 lr:6.676440829523441e-05 tokens_per_second_per_gpu:5344.117215776807 mfu:25.20911270723809 +Step 14251 | loss:0.9997897744178772 lr:6.676032300994899e-05 tokens_per_second_per_gpu:4928.552567995709 mfu:23.24882336101449 +Step 14252 | loss:0.9624983668327332 lr:6.675623759860704e-05 tokens_per_second_per_gpu:4207.230982971796 mfu:19.846226374303697 +Step 14253 | loss:0.8573885560035706 lr:6.67521520612393e-05 tokens_per_second_per_gpu:5351.41801847518 mfu:25.243551839211705 +Step 14254 | loss:1.0253413915634155 lr:6.674806639787647e-05 tokens_per_second_per_gpu:4876.4666549466265 mfu:23.003125222388828 +Step 14255 | loss:1.1276748180389404 lr:6.674398060854931e-05 tokens_per_second_per_gpu:5126.411117425006 mfu:24.182155896822046 +Step 14256 | loss:0.9837454557418823 lr:6.673989469328854e-05 tokens_per_second_per_gpu:3860.8804912992127 mfu:18.212432011596714 +Step 14257 | loss:0.9499375820159912 lr:6.673580865212487e-05 tokens_per_second_per_gpu:5351.160331977415 mfu:25.242336288035894 +Step 14258 | loss:0.9786840677261353 lr:6.673172248508906e-05 tokens_per_second_per_gpu:4637.00918003467 mfu:21.873563457571006 +Step 14259 | loss:0.912518322467804 lr:6.672763619221181e-05 tokens_per_second_per_gpu:4490.728840158516 mfu:21.183534136375616 +Step 14260 | loss:1.023300051689148 lr:6.67235497735239e-05 tokens_per_second_per_gpu:4512.3313726269325 mfu:21.285436967801243 +Step 14261 | loss:1.1617566347122192 lr:6.671946322905602e-05 tokens_per_second_per_gpu:5326.73992341164 mfu:25.127141054279793 +Step 14262 | loss:1.048614501953125 lr:6.671537655883892e-05 tokens_per_second_per_gpu:5266.664146547056 mfu:24.843753364825073 +Step 14263 | loss:0.9825772643089294 lr:6.671128976290337e-05 tokens_per_second_per_gpu:5348.143180430619 mfu:25.228103869410525 +Step 14264 | loss:1.37163507938385 lr:6.670720284128005e-05 tokens_per_second_per_gpu:5375.607550334628 mfu:25.357657988152436 +Step 14265 | loss:0.9725687503814697 lr:6.670311579399974e-05 tokens_per_second_per_gpu:3953.244920110629 mfu:18.648130781296736 +Step 14266 | loss:0.9958723187446594 lr:6.669902862109316e-05 tokens_per_second_per_gpu:3822.5182872170203 mfu:18.03147094967407 +Step 14267 | loss:1.096303939819336 lr:6.669494132259105e-05 tokens_per_second_per_gpu:4356.6777800672435 mfu:20.5511924144558 +Step 14268 | loss:0.9586830735206604 lr:6.669085389852417e-05 tokens_per_second_per_gpu:4176.160628160884 mfu:19.699662209511455 +Step 14269 | loss:1.1586579084396362 lr:6.668676634892321e-05 tokens_per_second_per_gpu:3514.5813362611057 mfu:16.57887981255349 +Step 14270 | loss:1.063767433166504 lr:6.668267867381899e-05 tokens_per_second_per_gpu:4317.769824049655 mfu:20.367656947539015 +Step 14271 | loss:0.9434678554534912 lr:6.667859087324221e-05 tokens_per_second_per_gpu:4077.4217577119657 mfu:19.233894111015445 +Step 14272 | loss:1.0032747983932495 lr:6.66745029472236e-05 tokens_per_second_per_gpu:4744.5005593501255 mfu:22.380618633722474 +Step 14273 | loss:1.0470112562179565 lr:6.667041489579395e-05 tokens_per_second_per_gpu:4495.922587544639 mfu:21.208033928940644 +Step 14274 | loss:1.0304419994354248 lr:6.666632671898397e-05 tokens_per_second_per_gpu:4802.94772285396 mfu:22.65632387608445 +Step 14275 | loss:1.018094778060913 lr:6.666223841682442e-05 tokens_per_second_per_gpu:5335.314292385143 mfu:25.167587815666007 +Step 14276 | loss:0.9703599214553833 lr:6.665814998934604e-05 tokens_per_second_per_gpu:4266.119107040563 mfu:20.124011702886087 +Step 14277 | loss:0.9890981316566467 lr:6.665406143657961e-05 tokens_per_second_per_gpu:4423.171275758802 mfu:20.864853578593102 +Step 14278 | loss:1.0980440378189087 lr:6.664997275855584e-05 tokens_per_second_per_gpu:5132.235014036868 mfu:24.209628210799426 +Step 14279 | loss:1.0080682039260864 lr:6.664588395530551e-05 tokens_per_second_per_gpu:3412.0639835411253 mfu:16.095287968509325 +Step 14280 | loss:1.105756163597107 lr:6.664179502685936e-05 tokens_per_second_per_gpu:5382.527323255216 mfu:25.39029973765385 +Step 14281 | loss:1.0559290647506714 lr:6.663770597324814e-05 tokens_per_second_per_gpu:5318.194967009281 mfu:25.086833037009967 +Step 14282 | loss:1.0519698858261108 lr:6.663361679450262e-05 tokens_per_second_per_gpu:3744.631777099589 mfu:17.66406699264122 +Step 14283 | loss:1.0397286415100098 lr:6.662952749065355e-05 tokens_per_second_per_gpu:3973.9606974924404 mfu:18.745850637682828 +Step 14284 | loss:1.196051836013794 lr:6.662543806173167e-05 tokens_per_second_per_gpu:4272.213635573364 mfu:20.152760633809102 +Step 14285 | loss:0.975691020488739 lr:6.662134850776775e-05 tokens_per_second_per_gpu:4166.243859274116 mfu:19.652883118697332 +Step 14286 | loss:0.94451504945755 lr:6.661725882879255e-05 tokens_per_second_per_gpu:3734.192581477126 mfu:17.61482352577963 +Step 14287 | loss:1.0242624282836914 lr:6.661316902483683e-05 tokens_per_second_per_gpu:4358.703113658239 mfu:20.56074625856207 +Step 14288 | loss:1.1476283073425293 lr:6.660907909593134e-05 tokens_per_second_per_gpu:3736.1765388015933 mfu:17.624182190977166 +Step 14289 | loss:0.9833006262779236 lr:6.660498904210684e-05 tokens_per_second_per_gpu:5033.6220907886955 mfu:23.74445422673814 +Step 14290 | loss:1.04452383518219 lr:6.660089886339411e-05 tokens_per_second_per_gpu:4874.0127658025995 mfu:22.99154980861979 +Step 14291 | loss:1.0194989442825317 lr:6.65968085598239e-05 tokens_per_second_per_gpu:4104.129422689912 mfu:19.359878723513944 +Step 14292 | loss:1.182374119758606 lr:6.659271813142698e-05 tokens_per_second_per_gpu:4308.183412174715 mfu:20.3224362071144 +Step 14293 | loss:1.089396357536316 lr:6.65886275782341e-05 tokens_per_second_per_gpu:4140.073539913202 mfu:19.529433257154984 +Step 14294 | loss:0.9171887636184692 lr:6.658453690027604e-05 tokens_per_second_per_gpu:4077.0796895114786 mfu:19.232280516950897 +Step 14295 | loss:1.1532355546951294 lr:6.658044609758356e-05 tokens_per_second_per_gpu:4581.130086256736 mfu:21.609972238263985 +Step 14296 | loss:1.1013402938842773 lr:6.657635517018743e-05 tokens_per_second_per_gpu:3498.797694442368 mfu:16.504425681127522 +Step 14297 | loss:1.1116883754730225 lr:6.657226411811842e-05 tokens_per_second_per_gpu:5066.871566447355 mfu:23.901297676365562 +Step 14298 | loss:1.0201069116592407 lr:6.656817294140728e-05 tokens_per_second_per_gpu:4809.106059889928 mfu:22.685373802603753 +Step 14299 | loss:0.9067193269729614 lr:6.656408164008482e-05 tokens_per_second_per_gpu:4225.475071301192 mfu:19.932286851715904 +Step 14300 | loss:1.0641579627990723 lr:6.655999021418178e-05 tokens_per_second_per_gpu:4691.017307530013 mfu:22.128329009703283 +Step 14301 | loss:1.0976028442382812 lr:6.655589866372894e-05 tokens_per_second_per_gpu:3892.6844065563632 mfu:18.36245650617218 +Step 14302 | loss:0.9721553325653076 lr:6.655180698875708e-05 tokens_per_second_per_gpu:3569.3113804975255 mfu:16.837050769125124 +Step 14303 | loss:1.0017898082733154 lr:6.654771518929696e-05 tokens_per_second_per_gpu:4806.667780666148 mfu:22.67387202349185 +Step 14304 | loss:0.9709700345993042 lr:6.654362326537936e-05 tokens_per_second_per_gpu:3628.157249253159 mfu:17.114636772185964 +Step 14305 | loss:1.0585623979568481 lr:6.653953121703506e-05 tokens_per_second_per_gpu:3955.670237409325 mfu:18.659571416796485 +Step 14306 | loss:1.2999335527420044 lr:6.653543904429483e-05 tokens_per_second_per_gpu:4171.17081072724 mfu:19.676124389326315 +Step 14307 | loss:1.0342007875442505 lr:6.653134674718948e-05 tokens_per_second_per_gpu:5349.828227951093 mfu:25.236052526063354 +Step 14308 | loss:1.0531864166259766 lr:6.652725432574973e-05 tokens_per_second_per_gpu:4492.019318267831 mfu:21.18962154179553 +Step 14309 | loss:1.0236072540283203 lr:6.65231617800064e-05 tokens_per_second_per_gpu:4979.052937670605 mfu:23.487042221022524 +Step 14310 | loss:1.073346734046936 lr:6.651906910999028e-05 tokens_per_second_per_gpu:3674.772623192702 mfu:17.33452944446171 +Step 14311 | loss:1.008849859237671 lr:6.65149763157321e-05 tokens_per_second_per_gpu:3320.0011161775565 mfu:15.66101171561049 +Step 14312 | loss:1.0724457502365112 lr:6.65108833972627e-05 tokens_per_second_per_gpu:3500.4106720872405 mfu:16.51203437188075 +Step 14313 | loss:1.0441174507141113 lr:6.650679035461284e-05 tokens_per_second_per_gpu:3678.280720536432 mfu:17.35107773817521 +Step 14314 | loss:1.0383223295211792 lr:6.65026971878133e-05 tokens_per_second_per_gpu:4274.687694532236 mfu:20.164431192036083 +Step 14315 | loss:0.9504779577255249 lr:6.649860389689487e-05 tokens_per_second_per_gpu:5394.414644984687 mfu:25.44637426243799 +Step 14316 | loss:1.057633399963379 lr:6.649451048188832e-05 tokens_per_second_per_gpu:4361.822035404016 mfu:20.575458744579826 +Step 14317 | loss:1.018700122833252 lr:6.649041694282446e-05 tokens_per_second_per_gpu:4323.402359338991 mfu:20.39422657750843 +Step 14318 | loss:0.9647848606109619 lr:6.648632327973408e-05 tokens_per_second_per_gpu:5310.750185934591 mfu:25.05171473447002 +Step 14319 | loss:1.0172622203826904 lr:6.648222949264795e-05 tokens_per_second_per_gpu:3094.6127290583436 mfu:14.597816238344942 +Step 14320 | loss:1.0015236139297485 lr:6.647813558159688e-05 tokens_per_second_per_gpu:4328.003748694089 mfu:20.415932116174403 +Step 14321 | loss:0.9291045069694519 lr:6.647404154661164e-05 tokens_per_second_per_gpu:5342.635480852936 mfu:25.20212311079214 +Step 14322 | loss:1.0369793176651 lr:6.646994738772303e-05 tokens_per_second_per_gpu:4340.991282473737 mfu:20.477196528914764 +Step 14323 | loss:1.0670883655548096 lr:6.646585310496183e-05 tokens_per_second_per_gpu:4897.82834719189 mfu:23.10389196939026 +Step 14324 | loss:1.033107876777649 lr:6.646175869835887e-05 tokens_per_second_per_gpu:3938.605190223527 mfu:18.579072677623795 +Step 14325 | loss:0.9754427075386047 lr:6.645766416794492e-05 tokens_per_second_per_gpu:5158.562707951731 mfu:24.333820434963613 +Step 14326 | loss:1.0466450452804565 lr:6.645356951375077e-05 tokens_per_second_per_gpu:5337.027218235453 mfu:25.175667979157204 +Step 14327 | loss:1.0748704671859741 lr:6.644947473580722e-05 tokens_per_second_per_gpu:4880.116366760203 mfu:23.020341535718515 +Step 14328 | loss:1.0905020236968994 lr:6.644537983414509e-05 tokens_per_second_per_gpu:5331.43029602534 mfu:25.1492663421587 +Step 14329 | loss:0.9437128901481628 lr:6.644128480879514e-05 tokens_per_second_per_gpu:4959.224893879573 mfu:23.393509955447104 +Step 14330 | loss:1.0583152770996094 lr:6.64371896597882e-05 tokens_per_second_per_gpu:4157.8640785607595 mfu:19.613354263334408 +Step 14331 | loss:1.079685091972351 lr:6.643309438715505e-05 tokens_per_second_per_gpu:4543.575586103802 mfu:21.432821253584155 +Step 14332 | loss:1.1578127145767212 lr:6.642899899092652e-05 tokens_per_second_per_gpu:4751.227577112844 mfu:22.4123511242568 +Step 14333 | loss:1.0651218891143799 lr:6.642490347113338e-05 tokens_per_second_per_gpu:3982.4803448916346 mfu:18.78603926806656 +Step 14334 | loss:1.004707932472229 lr:6.642080782780644e-05 tokens_per_second_per_gpu:5342.8707440260005 mfu:25.203232887319526 +Step 14335 | loss:0.9143694043159485 lr:6.641671206097652e-05 tokens_per_second_per_gpu:5167.358628477997 mfu:24.375312292824802 +Step 14336 | loss:1.0617212057113647 lr:6.641261617067441e-05 tokens_per_second_per_gpu:5348.029789672536 mfu:25.227568985858188 +Step 14337 | loss:1.0845671892166138 lr:6.64085201569309e-05 tokens_per_second_per_gpu:2951.9886008998496 mfu:13.925033891635799 +Step 14338 | loss:1.0817927122116089 lr:6.640442401977685e-05 tokens_per_second_per_gpu:4493.249541555997 mfu:21.19542471494354 +Step 14339 | loss:0.9396615028381348 lr:6.640032775924302e-05 tokens_per_second_per_gpu:5026.069220735817 mfu:23.7088260500462 +Step 14340 | loss:1.0100606679916382 lr:6.639623137536023e-05 tokens_per_second_per_gpu:4392.722512676353 mfu:20.721221568038512 +Step 14341 | loss:1.092631459236145 lr:6.639213486815928e-05 tokens_per_second_per_gpu:3839.08837703407 mfu:18.109634890489943 +Step 14342 | loss:1.06326162815094 lr:6.638803823767101e-05 tokens_per_second_per_gpu:4195.291058472857 mfu:19.789903713281014 +Step 14343 | loss:1.137742280960083 lr:6.63839414839262e-05 tokens_per_second_per_gpu:3503.591395255807 mfu:16.52703838575421 +Step 14344 | loss:1.0538471937179565 lr:6.637984460695569e-05 tokens_per_second_per_gpu:4433.301358326169 mfu:20.912638906433685 +Step 14345 | loss:1.1511796712875366 lr:6.637574760679026e-05 tokens_per_second_per_gpu:5386.337367066331 mfu:25.40827236437139 +Step 14346 | loss:1.0590814352035522 lr:6.637165048346076e-05 tokens_per_second_per_gpu:5005.993690388472 mfu:23.614126348159093 +Step 14347 | loss:1.197721242904663 lr:6.636755323699799e-05 tokens_per_second_per_gpu:4106.924261302002 mfu:19.373062454095813 +Step 14348 | loss:1.1246285438537598 lr:6.636345586743274e-05 tokens_per_second_per_gpu:4351.337805178333 mfu:20.526002841811145 +Step 14349 | loss:0.9533364772796631 lr:6.635935837479588e-05 tokens_per_second_per_gpu:5226.281423288946 mfu:24.653261169213504 +Step 14350 | loss:1.0236066579818726 lr:6.635526075911818e-05 tokens_per_second_per_gpu:3927.4376268137707 mfu:18.526393375637177 +Step 14351 | loss:0.948103129863739 lr:6.635116302043048e-05 tokens_per_second_per_gpu:3792.266810281979 mfu:17.888769571537285 +Step 14352 | loss:0.9915441870689392 lr:6.634706515876359e-05 tokens_per_second_per_gpu:4614.629383960511 mfu:21.767994141102253 +Step 14353 | loss:1.0735132694244385 lr:6.634296717414834e-05 tokens_per_second_per_gpu:3635.360508792491 mfu:17.1486157764358 +Step 14354 | loss:1.080568790435791 lr:6.633886906661555e-05 tokens_per_second_per_gpu:4232.964761023058 mfu:19.967616996006782 +Step 14355 | loss:0.9721540212631226 lr:6.633477083619605e-05 tokens_per_second_per_gpu:5332.695210110628 mfu:25.155233157715603 +Step 14356 | loss:1.0579643249511719 lr:6.633067248292065e-05 tokens_per_second_per_gpu:5290.93686113224 mfu:24.958251900875535 +Step 14357 | loss:1.065450668334961 lr:6.632657400682017e-05 tokens_per_second_per_gpu:4111.0667559000085 mfu:19.39260330789707 +Step 14358 | loss:1.1533534526824951 lr:6.632247540792545e-05 tokens_per_second_per_gpu:4749.418980364251 mfu:22.403819664815263 +Step 14359 | loss:1.13387930393219 lr:6.63183766862673e-05 tokens_per_second_per_gpu:5060.571201732283 mfu:23.8715777810515 +Step 14360 | loss:1.0614666938781738 lr:6.631427784187658e-05 tokens_per_second_per_gpu:5251.728154008498 mfu:24.773297739677588 +Step 14361 | loss:1.0919718742370605 lr:6.631017887478408e-05 tokens_per_second_per_gpu:4456.603526689006 mfu:21.022559210361077 +Step 14362 | loss:1.1020376682281494 lr:6.630607978502064e-05 tokens_per_second_per_gpu:3475.2117210459137 mfu:16.39316662043442 +Step 14363 | loss:1.2359941005706787 lr:6.63019805726171e-05 tokens_per_second_per_gpu:3096.410427858087 mfu:14.606296290302376 +Step 14364 | loss:1.1335684061050415 lr:6.629788123760427e-05 tokens_per_second_per_gpu:3432.926138679367 mfu:16.193698313747117 +Step 14365 | loss:1.1564347743988037 lr:6.629378178001301e-05 tokens_per_second_per_gpu:5386.920813461282 mfu:25.41102458056259 +Step 14366 | loss:1.0545499324798584 lr:6.628968219987413e-05 tokens_per_second_per_gpu:5329.626728757083 mfu:25.140758607634027 +Step 14367 | loss:1.0913591384887695 lr:6.628558249721849e-05 tokens_per_second_per_gpu:5110.207918122415 mfu:24.105722641160853 +Step 14368 | loss:1.120334506034851 lr:6.628148267207689e-05 tokens_per_second_per_gpu:4537.533095629401 mfu:21.404317795061218 +Step 14369 | loss:0.959852397441864 lr:6.627738272448019e-05 tokens_per_second_per_gpu:5314.309309843637 mfu:25.06850372919815 +Step 14370 | loss:1.0770349502563477 lr:6.62732826544592e-05 tokens_per_second_per_gpu:5178.268964269804 mfu:24.426778208249996 +Step 14371 | loss:1.0901293754577637 lr:6.62691824620448e-05 tokens_per_second_per_gpu:4853.052916859807 mfu:22.89267862503765 +Step 14372 | loss:1.0609456300735474 lr:6.626508214726778e-05 tokens_per_second_per_gpu:5337.948190082941 mfu:25.18001236049581 +Step 14373 | loss:1.201982855796814 lr:6.626098171015901e-05 tokens_per_second_per_gpu:5323.120406391254 mfu:25.110067174941346 +Step 14374 | loss:0.9868783354759216 lr:6.625688115074933e-05 tokens_per_second_per_gpu:4545.866051443785 mfu:21.443625769386536 +Step 14375 | loss:1.0660134553909302 lr:6.625278046906956e-05 tokens_per_second_per_gpu:5094.72462051207 mfu:24.032685284609876 +Step 14376 | loss:1.0796316862106323 lr:6.624867966515056e-05 tokens_per_second_per_gpu:4074.753578517004 mfu:19.221307854514496 +Step 14377 | loss:0.9857377409934998 lr:6.624457873902316e-05 tokens_per_second_per_gpu:4633.186222287926 mfu:21.855529913615765 +Step 14378 | loss:1.0727912187576294 lr:6.624047769071822e-05 tokens_per_second_per_gpu:5348.166463013133 mfu:25.22821369732466 +Step 14379 | loss:1.086904525756836 lr:6.623637652026656e-05 tokens_per_second_per_gpu:3989.68717108522 mfu:18.82003509683305 +Step 14380 | loss:0.9276766180992126 lr:6.623227522769906e-05 tokens_per_second_per_gpu:5292.567870285831 mfu:24.965945649331882 +Step 14381 | loss:0.9813128709793091 lr:6.622817381304653e-05 tokens_per_second_per_gpu:4460.834979873359 mfu:21.042519696992027 +Step 14382 | loss:1.0307698249816895 lr:6.622407227633984e-05 tokens_per_second_per_gpu:5339.727079897041 mfu:25.188403687258816 +Step 14383 | loss:1.099064588546753 lr:6.621997061760984e-05 tokens_per_second_per_gpu:4449.256456538045 mfu:20.98790182691957 +Step 14384 | loss:1.0731685161590576 lr:6.621586883688737e-05 tokens_per_second_per_gpu:5367.332643563532 mfu:25.318623841068494 +Step 14385 | loss:1.0671154260635376 lr:6.621176693420329e-05 tokens_per_second_per_gpu:4538.762413049042 mfu:21.410116695072844 +Step 14386 | loss:0.9288595914840698 lr:6.620766490958842e-05 tokens_per_second_per_gpu:5339.566478904228 mfu:25.18764610497436 +Step 14387 | loss:0.9751115441322327 lr:6.620356276307364e-05 tokens_per_second_per_gpu:4667.576034680089 mfu:22.017752526176633 +Step 14388 | loss:1.1045140027999878 lr:6.619946049468979e-05 tokens_per_second_per_gpu:4724.238541763377 mfu:22.2850392397091 +Step 14389 | loss:0.9458011388778687 lr:6.619535810446774e-05 tokens_per_second_per_gpu:4732.28484048438 mfu:22.322995003616192 +Step 14390 | loss:1.1802737712860107 lr:6.619125559243834e-05 tokens_per_second_per_gpu:4053.6800429459336 mfu:19.121900391709833 +Step 14391 | loss:1.0756648778915405 lr:6.618715295863244e-05 tokens_per_second_per_gpu:4783.473517095008 mfu:22.564460724877346 +Step 14392 | loss:1.2523013353347778 lr:6.61830502030809e-05 tokens_per_second_per_gpu:4841.895007370195 mfu:22.840044862239488 +Step 14393 | loss:1.0310944318771362 lr:6.617894732581456e-05 tokens_per_second_per_gpu:5334.631811725979 mfu:25.16436843795352 +Step 14394 | loss:1.0126616954803467 lr:6.61748443268643e-05 tokens_per_second_per_gpu:4434.287089289778 mfu:20.917288767572074 +Step 14395 | loss:0.9330825209617615 lr:6.617074120626097e-05 tokens_per_second_per_gpu:5361.879805223376 mfu:25.292901872267823 +Step 14396 | loss:1.0881267786026 lr:6.616663796403544e-05 tokens_per_second_per_gpu:5288.956411361319 mfu:24.948909781406524 +Step 14397 | loss:1.0238640308380127 lr:6.616253460021857e-05 tokens_per_second_per_gpu:3465.9595351623557 mfu:16.34952248103599 +Step 14398 | loss:0.8985192775726318 lr:6.615843111484119e-05 tokens_per_second_per_gpu:5335.291121568611 mfu:25.167478514971073 +Step 14399 | loss:1.0910037755966187 lr:6.61543275079342e-05 tokens_per_second_per_gpu:5337.424791992293 mfu:25.177543402401295 +Step 14400 | loss:0.8872900009155273 lr:6.615022377952845e-05 tokens_per_second_per_gpu:3916.7255798620904 mfu:18.47586282250243 +Step 14401 | loss:0.9538431167602539 lr:6.614611992965482e-05 tokens_per_second_per_gpu:4178.509316395535 mfu:19.71074137264184 +Step 14402 | loss:1.0333795547485352 lr:6.614201595834413e-05 tokens_per_second_per_gpu:4465.840756305393 mfu:21.066132798495467 +Step 14403 | loss:1.0105260610580444 lr:6.613791186562731e-05 tokens_per_second_per_gpu:5375.132743182692 mfu:25.355418241806298 +Step 14404 | loss:1.032248854637146 lr:6.613380765153518e-05 tokens_per_second_per_gpu:4896.641677671436 mfu:23.09829424679558 +Step 14405 | loss:1.0289677381515503 lr:6.612970331609863e-05 tokens_per_second_per_gpu:5197.6195421700795 mfu:24.51805818575463 +Step 14406 | loss:1.0386382341384888 lr:6.612559885934851e-05 tokens_per_second_per_gpu:4002.0942205859196 mfu:18.878561266188775 +Step 14407 | loss:1.1078189611434937 lr:6.612149428131572e-05 tokens_per_second_per_gpu:4891.242755497762 mfu:23.072826609751253 +Step 14408 | loss:1.0332621335983276 lr:6.611738958203111e-05 tokens_per_second_per_gpu:4426.750748746151 mfu:20.881738563395494 +Step 14409 | loss:0.9517738223075867 lr:6.611328476152557e-05 tokens_per_second_per_gpu:3787.6764777337125 mfu:17.867116189715617 +Step 14410 | loss:0.9572101831436157 lr:6.610917981982994e-05 tokens_per_second_per_gpu:5349.892261728289 mfu:25.236354584315674 +Step 14411 | loss:1.0869550704956055 lr:6.610507475697511e-05 tokens_per_second_per_gpu:5023.089569546462 mfu:23.6947705270051 +Step 14412 | loss:1.035449743270874 lr:6.610096957299197e-05 tokens_per_second_per_gpu:3417.9987050379364 mfu:16.123283062377684 +Step 14413 | loss:1.0400570631027222 lr:6.609686426791137e-05 tokens_per_second_per_gpu:4290.160268804404 mfu:20.237417964771293 +Step 14414 | loss:0.9582603573799133 lr:6.609275884176421e-05 tokens_per_second_per_gpu:4323.152138452813 mfu:20.393046242896244 +Step 14415 | loss:0.92778080701828 lr:6.608865329458136e-05 tokens_per_second_per_gpu:5206.156580193275 mfu:24.558328850678926 +Step 14416 | loss:1.1100459098815918 lr:6.60845476263937e-05 tokens_per_second_per_gpu:4159.41883028739 mfu:19.620688292496922 +Step 14417 | loss:1.0072641372680664 lr:6.608044183723208e-05 tokens_per_second_per_gpu:5328.9128273025435 mfu:25.137391012668818 +Step 14418 | loss:1.024092674255371 lr:6.607633592712742e-05 tokens_per_second_per_gpu:5349.604021755957 mfu:25.234994907187374 +Step 14419 | loss:0.9999834299087524 lr:6.60722298961106e-05 tokens_per_second_per_gpu:3246.848661501343 mfu:15.315939105807013 +Step 14420 | loss:1.0754516124725342 lr:6.606812374421247e-05 tokens_per_second_per_gpu:4354.376129193275 mfu:20.54033513458116 +Step 14421 | loss:0.9888946413993835 lr:6.606401747146393e-05 tokens_per_second_per_gpu:5326.696658145463 mfu:25.12693696463742 +Step 14422 | loss:1.1133836507797241 lr:6.605991107789586e-05 tokens_per_second_per_gpu:3997.0137599076065 mfu:18.854595866353197 +Step 14423 | loss:1.0362874269485474 lr:6.605580456353918e-05 tokens_per_second_per_gpu:5193.77145236055 mfu:24.499906089571255 +Step 14424 | loss:1.1928153038024902 lr:6.605169792842471e-05 tokens_per_second_per_gpu:4283.881160040173 mfu:20.20779833740399 +Step 14425 | loss:0.9403306841850281 lr:6.60475911725834e-05 tokens_per_second_per_gpu:4750.625578814758 mfu:22.40951139557323 +Step 14426 | loss:1.036515235900879 lr:6.60434842960461e-05 tokens_per_second_per_gpu:3845.202723587092 mfu:18.138477306395643 +Step 14427 | loss:1.0434582233428955 lr:6.603937729884371e-05 tokens_per_second_per_gpu:4670.991229248362 mfu:22.033862581647455 +Step 14428 | loss:1.1170668601989746 lr:6.60352701810071e-05 tokens_per_second_per_gpu:4271.409939532855 mfu:20.148969462461334 +Step 14429 | loss:1.1732121706008911 lr:6.603116294256719e-05 tokens_per_second_per_gpu:4720.628172832902 mfu:22.268008513471592 +Step 14430 | loss:0.985427975654602 lr:6.602705558355486e-05 tokens_per_second_per_gpu:4342.791885488001 mfu:20.4856902805476 +Step 14431 | loss:1.1018437147140503 lr:6.602294810400099e-05 tokens_per_second_per_gpu:4398.546313075875 mfu:20.748693428165865 +Step 14432 | loss:0.8593429923057556 lr:6.601884050393649e-05 tokens_per_second_per_gpu:5321.1835218243605 mfu:25.100930559972728 +Step 14433 | loss:1.0848947763442993 lr:6.601473278339224e-05 tokens_per_second_per_gpu:4388.976346176835 mfu:20.70355026149837 +Step 14434 | loss:0.8898288607597351 lr:6.601062494239915e-05 tokens_per_second_per_gpu:4508.504772915816 mfu:21.267386244078672 +Step 14435 | loss:0.8450299501419067 lr:6.600651698098811e-05 tokens_per_second_per_gpu:4494.118911903211 mfu:21.199525683201227 +Step 14436 | loss:1.1693038940429688 lr:6.600240889918999e-05 tokens_per_second_per_gpu:5370.031983655891 mfu:25.331357088839063 +Step 14437 | loss:1.133763313293457 lr:6.599830069703574e-05 tokens_per_second_per_gpu:5360.243661405074 mfu:25.28518390272197 +Step 14438 | loss:1.088412880897522 lr:6.59941923745562e-05 tokens_per_second_per_gpu:4160.458270502987 mfu:19.62559150934068 +Step 14439 | loss:1.0678954124450684 lr:6.599008393178234e-05 tokens_per_second_per_gpu:4003.4282432687946 mfu:18.884854078791637 +Step 14440 | loss:0.9175920486450195 lr:6.598597536874497e-05 tokens_per_second_per_gpu:4442.303520822024 mfu:20.95510364285394 +Step 14441 | loss:0.9355779886245728 lr:6.598186668547509e-05 tokens_per_second_per_gpu:5090.545862207348 mfu:24.01297336871682 +Step 14442 | loss:0.9967203736305237 lr:6.59777578820035e-05 tokens_per_second_per_gpu:4188.5237553091965 mfu:19.757981189637047 +Step 14443 | loss:1.0996915102005005 lr:6.597364895836118e-05 tokens_per_second_per_gpu:4298.110543622511 mfu:20.274920767545055 +Step 14444 | loss:0.9952660799026489 lr:6.5969539914579e-05 tokens_per_second_per_gpu:5345.697273383316 mfu:25.21656610855362 +Step 14445 | loss:1.1606082916259766 lr:6.596543075068788e-05 tokens_per_second_per_gpu:4491.622980386791 mfu:21.187751948388446 +Step 14446 | loss:1.0423974990844727 lr:6.596132146671872e-05 tokens_per_second_per_gpu:3438.547439512769 mfu:16.220214948871096 +Step 14447 | loss:0.9662747383117676 lr:6.595721206270242e-05 tokens_per_second_per_gpu:4810.756525199332 mfu:22.69315932906663 +Step 14448 | loss:1.052344799041748 lr:6.595310253866991e-05 tokens_per_second_per_gpu:4191.71154597328 mfu:19.773018542092704 +Step 14449 | loss:1.2192221879959106 lr:6.594899289465205e-05 tokens_per_second_per_gpu:4055.0673552080852 mfu:19.12844458035028 +Step 14450 | loss:1.0251383781433105 lr:6.59448831306798e-05 tokens_per_second_per_gpu:5311.683581770701 mfu:25.056117721882703 +Step 14451 | loss:1.11762273311615 lr:6.594077324678404e-05 tokens_per_second_per_gpu:5336.178453502144 mfu:25.17166421109599 +Step 14452 | loss:0.9965612292289734 lr:6.59366632429957e-05 tokens_per_second_per_gpu:4952.697417779075 mfu:23.362718736978795 +Step 14453 | loss:1.0174487829208374 lr:6.593255311934566e-05 tokens_per_second_per_gpu:5020.323695200323 mfu:23.68172342580753 +Step 14454 | loss:1.209805965423584 lr:6.592844287586486e-05 tokens_per_second_per_gpu:5370.6497802376825 mfu:25.33427134072234 +Step 14455 | loss:1.0952328443527222 lr:6.592433251258423e-05 tokens_per_second_per_gpu:4392.629334965351 mfu:20.720782032877953 +Step 14456 | loss:1.0743247270584106 lr:6.592022202953466e-05 tokens_per_second_per_gpu:2842.7412526601292 mfu:13.409695510469383 +Step 14457 | loss:1.1298949718475342 lr:6.591611142674706e-05 tokens_per_second_per_gpu:5060.107690699466 mfu:23.869391320426566 +Step 14458 | loss:1.0868743658065796 lr:6.591200070425235e-05 tokens_per_second_per_gpu:5231.36890297622 mfu:24.677259678911916 +Step 14459 | loss:1.1078561544418335 lr:6.590788986208145e-05 tokens_per_second_per_gpu:4387.7169254596265 mfu:20.697609359095967 +Step 14460 | loss:0.937440812587738 lr:6.590377890026529e-05 tokens_per_second_per_gpu:4969.715519281568 mfu:23.442996025353576 +Step 14461 | loss:1.1028931140899658 lr:6.589966781883478e-05 tokens_per_second_per_gpu:4787.695348813553 mfu:22.584375825412494 +Step 14462 | loss:1.1147676706314087 lr:6.589555661782084e-05 tokens_per_second_per_gpu:3686.861100416556 mfu:17.391552854033524 +Step 14463 | loss:1.153910517692566 lr:6.589144529725439e-05 tokens_per_second_per_gpu:3951.05448310857 mfu:18.637798116230933 +Step 14464 | loss:1.0522125959396362 lr:6.588733385716635e-05 tokens_per_second_per_gpu:5215.675498855869 mfu:24.603231214105374 +Step 14465 | loss:0.9687246680259705 lr:6.588322229758764e-05 tokens_per_second_per_gpu:5311.412799536807 mfu:25.054840395885215 +Step 14466 | loss:1.1010773181915283 lr:6.587911061854919e-05 tokens_per_second_per_gpu:5008.650957025872 mfu:23.626661128264153 +Step 14467 | loss:1.0422992706298828 lr:6.587499882008193e-05 tokens_per_second_per_gpu:5187.275475939696 mfu:24.46926346046622 +Step 14468 | loss:1.1114610433578491 lr:6.587088690221677e-05 tokens_per_second_per_gpu:4858.055659187603 mfu:22.916277414153722 +Step 14469 | loss:0.9710218906402588 lr:6.586677486498464e-05 tokens_per_second_per_gpu:5315.592497447107 mfu:25.0745567440581 +Step 14470 | loss:0.9493911862373352 lr:6.586266270841648e-05 tokens_per_second_per_gpu:5049.741073295775 mfu:23.820490217406043 +Step 14471 | loss:0.9918904900550842 lr:6.58585504325432e-05 tokens_per_second_per_gpu:4611.5499234352355 mfu:21.753467800394382 +Step 14472 | loss:1.0290862321853638 lr:6.585443803739575e-05 tokens_per_second_per_gpu:3686.844388171364 mfu:17.391474019521546 +Step 14473 | loss:1.2502877712249756 lr:6.585032552300506e-05 tokens_per_second_per_gpu:4632.337194365278 mfu:21.851524904045533 +Step 14474 | loss:1.0534511804580688 lr:6.584621288940202e-05 tokens_per_second_per_gpu:4333.363898557879 mfu:20.44121685761777 +Step 14475 | loss:1.040099859237671 lr:6.584210013661762e-05 tokens_per_second_per_gpu:4440.7485758933835 mfu:20.947768702324574 +Step 14476 | loss:0.999445915222168 lr:6.583798726468274e-05 tokens_per_second_per_gpu:5330.731972252224 mfu:25.145972229774596 +Step 14477 | loss:1.0279662609100342 lr:6.583387427362836e-05 tokens_per_second_per_gpu:5344.250497395848 mfu:25.20974141937529 +Step 14478 | loss:1.0586657524108887 lr:6.582976116348537e-05 tokens_per_second_per_gpu:3430.5127714729547 mfu:16.182314048871564 +Step 14479 | loss:0.9425097703933716 lr:6.582564793428474e-05 tokens_per_second_per_gpu:3839.1089327755417 mfu:18.109731855429636 +Step 14480 | loss:1.1470969915390015 lr:6.582153458605738e-05 tokens_per_second_per_gpu:3683.606084456043 mfu:17.376198388384793 +Step 14481 | loss:1.0274672508239746 lr:6.581742111883426e-05 tokens_per_second_per_gpu:3859.968246477546 mfu:18.2081287971279 +Step 14482 | loss:0.9811047911643982 lr:6.581330753264627e-05 tokens_per_second_per_gpu:4321.492362685185 mfu:20.385216797413392 +Step 14483 | loss:1.0113168954849243 lr:6.580919382752439e-05 tokens_per_second_per_gpu:5337.877121142849 mfu:25.179677116179782 +Step 14484 | loss:1.0477039813995361 lr:6.580508000349955e-05 tokens_per_second_per_gpu:4969.010373101114 mfu:23.439669730510033 +Step 14485 | loss:1.161374568939209 lr:6.580096606060269e-05 tokens_per_second_per_gpu:5328.10888961092 mfu:25.133598701411515 +Step 14486 | loss:0.909171462059021 lr:6.579685199886473e-05 tokens_per_second_per_gpu:4889.8112765746855 mfu:23.066074079435612 +Step 14487 | loss:1.0321654081344604 lr:6.579273781831665e-05 tokens_per_second_per_gpu:4714.715804800334 mfu:22.240118864707963 +Step 14488 | loss:1.0285178422927856 lr:6.578862351898936e-05 tokens_per_second_per_gpu:4081.5159598214245 mfu:19.253207160908676 +Step 14489 | loss:1.2756197452545166 lr:6.578450910091383e-05 tokens_per_second_per_gpu:3427.7957181708143 mfu:16.169497244868275 +Step 14490 | loss:1.0198497772216797 lr:6.578039456412097e-05 tokens_per_second_per_gpu:4004.6114733866693 mfu:18.890435577137524 +Step 14491 | loss:1.0228745937347412 lr:6.577627990864179e-05 tokens_per_second_per_gpu:4931.630708534776 mfu:23.263343475122205 +Step 14492 | loss:1.1547868251800537 lr:6.577216513450715e-05 tokens_per_second_per_gpu:4009.4870129570636 mfu:18.913434329143904 +Step 14493 | loss:0.9854830503463745 lr:6.576805024174807e-05 tokens_per_second_per_gpu:3459.4998288712554 mfu:16.319050944321326 +Step 14494 | loss:1.0700387954711914 lr:6.576393523039546e-05 tokens_per_second_per_gpu:5351.228022639312 mfu:25.24265559643728 +Step 14495 | loss:1.0822155475616455 lr:6.57598201004803e-05 tokens_per_second_per_gpu:4004.6510050042 mfu:18.890622054522208 +Step 14496 | loss:1.1406575441360474 lr:6.57557048520335e-05 tokens_per_second_per_gpu:4544.024888222855 mfu:21.43494068833857 +Step 14497 | loss:1.063289999961853 lr:6.575158948508604e-05 tokens_per_second_per_gpu:4169.53251522481 mfu:19.668396269919317 +Step 14498 | loss:1.1958414316177368 lr:6.574747399966888e-05 tokens_per_second_per_gpu:4800.98499183019 mfu:22.64706533896888 +Step 14499 | loss:1.0368186235427856 lr:6.574335839581294e-05 tokens_per_second_per_gpu:4593.408374183325 mfu:21.667891017305642 +Step 14500 | loss:1.077613115310669 lr:6.573924267354921e-05 tokens_per_second_per_gpu:5363.056877367409 mfu:25.298454322400442 +Step 14501 | loss:0.9813715219497681 lr:6.57351268329086e-05 tokens_per_second_per_gpu:5342.241594737775 mfu:25.20026508278303 +Step 14502 | loss:1.047541856765747 lr:6.573101087392212e-05 tokens_per_second_per_gpu:4104.259342832348 mfu:19.360491578992477 +Step 14503 | loss:1.0159385204315186 lr:6.572689479662068e-05 tokens_per_second_per_gpu:5337.308953503335 mfu:25.176996972484798 +Step 14504 | loss:1.0882500410079956 lr:6.572277860103527e-05 tokens_per_second_per_gpu:4017.804040927014 mfu:18.952667169109457 +Step 14505 | loss:1.0890419483184814 lr:6.571866228719684e-05 tokens_per_second_per_gpu:4703.915462201211 mfu:22.189171805939438 +Step 14506 | loss:0.938362717628479 lr:6.571454585513635e-05 tokens_per_second_per_gpu:4572.370515050804 mfu:21.568651846348644 +Step 14507 | loss:0.9302420616149902 lr:6.571042930488474e-05 tokens_per_second_per_gpu:5090.14615495733 mfu:24.011087881421393 +Step 14508 | loss:1.0524893999099731 lr:6.570631263647298e-05 tokens_per_second_per_gpu:3158.456046260384 mfu:14.89897605838011 +Step 14509 | loss:1.005456566810608 lr:6.570219584993206e-05 tokens_per_second_per_gpu:5377.9567717148875 mfu:25.368739666221618 +Step 14510 | loss:1.0040844678878784 lr:6.56980789452929e-05 tokens_per_second_per_gpu:5254.649892141767 mfu:24.78708007695215 +Step 14511 | loss:0.9351404905319214 lr:6.56939619225865e-05 tokens_per_second_per_gpu:4412.286022806117 mfu:20.81350598319963 +Step 14512 | loss:0.9076740145683289 lr:6.56898447818438e-05 tokens_per_second_per_gpu:4318.29444254377 mfu:20.370131662484898 +Step 14513 | loss:1.0275121927261353 lr:6.568572752309578e-05 tokens_per_second_per_gpu:4149.728485803999 mfu:19.57497728422551 +Step 14514 | loss:1.0151673555374146 lr:6.56816101463734e-05 tokens_per_second_per_gpu:5334.026515499529 mfu:25.16151315237917 +Step 14515 | loss:0.9545006155967712 lr:6.567749265170764e-05 tokens_per_second_per_gpu:4542.698719233823 mfu:21.428684922949547 +Step 14516 | loss:1.0850632190704346 lr:6.567337503912945e-05 tokens_per_second_per_gpu:5164.23185984351 mfu:24.36056279169419 +Step 14517 | loss:1.1600703001022339 lr:6.566925730866981e-05 tokens_per_second_per_gpu:5358.635988187207 mfu:25.277600233855956 +Step 14518 | loss:1.0851622819900513 lr:6.566513946035969e-05 tokens_per_second_per_gpu:5267.729321347913 mfu:24.848777976098276 +Step 14519 | loss:1.0877290964126587 lr:6.566102149423004e-05 tokens_per_second_per_gpu:4099.747419383189 mfu:19.33920806626927 +Step 14520 | loss:0.945405125617981 lr:6.565690341031186e-05 tokens_per_second_per_gpu:5033.6998634068805 mfu:23.744821093447495 +Step 14521 | loss:0.966353714466095 lr:6.565278520863611e-05 tokens_per_second_per_gpu:4996.284920947867 mfu:23.568328426220887 +Step 14522 | loss:1.0025678873062134 lr:6.564866688923377e-05 tokens_per_second_per_gpu:4481.933048746973 mfu:21.142042887570607 +Step 14523 | loss:0.7537873387336731 lr:6.56445484521358e-05 tokens_per_second_per_gpu:130.55008903615052 mfu:0.6158270441255482 +Step 14524 | loss:0.8944789171218872 lr:6.56404298973732e-05 tokens_per_second_per_gpu:4365.191704426804 mfu:20.591354048285137 +Step 14525 | loss:0.8958328366279602 lr:6.563631122497691e-05 tokens_per_second_per_gpu:3826.460315988304 mfu:18.05006617196768 +Step 14526 | loss:0.7700703144073486 lr:6.563219243497793e-05 tokens_per_second_per_gpu:5340.430580143566 mfu:25.19172221795119 +Step 14527 | loss:0.6793127059936523 lr:6.562807352740727e-05 tokens_per_second_per_gpu:5349.436093347309 mfu:25.23420276023224 +Step 14528 | loss:0.8041146993637085 lr:6.562395450229584e-05 tokens_per_second_per_gpu:4680.923187903707 mfu:22.0807133251916 +Step 14529 | loss:0.752622127532959 lr:6.561983535967466e-05 tokens_per_second_per_gpu:4123.638059969531 mfu:19.451904294078652 +Step 14530 | loss:0.786491334438324 lr:6.56157160995747e-05 tokens_per_second_per_gpu:3941.447349395732 mfu:18.592479627361445 +Step 14531 | loss:0.8411610722541809 lr:6.561159672202697e-05 tokens_per_second_per_gpu:4669.347911564743 mfu:22.026110771754915 +Step 14532 | loss:0.8344486951828003 lr:6.56074772270624e-05 tokens_per_second_per_gpu:4049.421125895539 mfu:19.101810353336944 +Step 14533 | loss:0.8675512671470642 lr:6.560335761471202e-05 tokens_per_second_per_gpu:4641.048410669342 mfu:21.892617197638565 +Step 14534 | loss:0.6453553438186646 lr:6.55992378850068e-05 tokens_per_second_per_gpu:4928.6292753968755 mfu:23.249185202913427 +Step 14535 | loss:0.7614315152168274 lr:6.55951180379777e-05 tokens_per_second_per_gpu:5060.328726713698 mfu:23.870433985018334 +Step 14536 | loss:0.6899263262748718 lr:6.559099807365575e-05 tokens_per_second_per_gpu:3228.1982429670793 mfu:15.227961899491364 +Step 14537 | loss:0.8533673882484436 lr:6.55868779920719e-05 tokens_per_second_per_gpu:4856.303721763793 mfu:22.908013226414266 +Step 14538 | loss:0.7376543283462524 lr:6.558275779325718e-05 tokens_per_second_per_gpu:3452.93729340555 mfu:16.288094344845586 +Step 14539 | loss:0.8567678332328796 lr:6.557863747724252e-05 tokens_per_second_per_gpu:3610.5319332247213 mfu:17.031495149290816 +Step 14540 | loss:0.9049575924873352 lr:6.557451704405894e-05 tokens_per_second_per_gpu:4149.028918318297 mfu:19.571677305036932 +Step 14541 | loss:0.7466415166854858 lr:6.557039649373744e-05 tokens_per_second_per_gpu:5382.692666613137 mfu:25.391079690483647 +Step 14542 | loss:0.7543728947639465 lr:6.5566275826309e-05 tokens_per_second_per_gpu:5345.001570083988 mfu:25.213284357391466 +Step 14543 | loss:0.9019045233726501 lr:6.556215504180462e-05 tokens_per_second_per_gpu:3561.7626902500524 mfu:16.80144230928843 +Step 14544 | loss:0.6536518931388855 lr:6.555803414025526e-05 tokens_per_second_per_gpu:5053.9096654749255 mfu:23.84015417795026 +Step 14545 | loss:0.7507869005203247 lr:6.555391312169196e-05 tokens_per_second_per_gpu:4611.737432210307 mfu:21.75435231127869 +Step 14546 | loss:0.7926517724990845 lr:6.55497919861457e-05 tokens_per_second_per_gpu:5311.245740205936 mfu:25.054052348518237 +Step 14547 | loss:0.7179577946662903 lr:6.554567073364747e-05 tokens_per_second_per_gpu:4875.535917043566 mfu:22.99873477289983 +Step 14548 | loss:0.8219634294509888 lr:6.554154936422825e-05 tokens_per_second_per_gpu:4827.206013973744 mfu:22.770754374187725 +Step 14549 | loss:0.746239185333252 lr:6.553742787791908e-05 tokens_per_second_per_gpu:4336.096531846383 mfu:20.45410715507496 +Step 14550 | loss:0.8384863138198853 lr:6.55333062747509e-05 tokens_per_second_per_gpu:5382.626218103397 mfu:25.39076624152592 +Step 14551 | loss:0.8252428770065308 lr:6.552918455475477e-05 tokens_per_second_per_gpu:4355.9406127889 mfu:20.547715070630147 +Step 14552 | loss:0.8400297164916992 lr:6.552506271796165e-05 tokens_per_second_per_gpu:4842.924070192272 mfu:22.844899127147446 +Step 14553 | loss:0.6589285731315613 lr:6.552094076440255e-05 tokens_per_second_per_gpu:5376.34577387753 mfu:25.361140314558224 +Step 14554 | loss:0.802765429019928 lr:6.551681869410848e-05 tokens_per_second_per_gpu:5354.502048213557 mfu:25.25809973367736 +Step 14555 | loss:0.7962263226509094 lr:6.551269650711043e-05 tokens_per_second_per_gpu:3345.606988861769 mfu:15.781798985874298 +Step 14556 | loss:0.6964986324310303 lr:6.55085742034394e-05 tokens_per_second_per_gpu:3601.9660748620045 mfu:16.991088533907742 +Step 14557 | loss:0.8762703537940979 lr:6.550445178312643e-05 tokens_per_second_per_gpu:5014.209824207179 mfu:23.652883253198915 +Step 14558 | loss:0.6405079364776611 lr:6.550032924620248e-05 tokens_per_second_per_gpu:4428.114911609642 mfu:20.88817355237277 +Step 14559 | loss:0.698872983455658 lr:6.54962065926986e-05 tokens_per_second_per_gpu:4309.413313922697 mfu:20.328237863502505 +Step 14560 | loss:0.8079282641410828 lr:6.549208382264574e-05 tokens_per_second_per_gpu:5052.665240986298 mfu:23.83428401531669 +Step 14561 | loss:0.7569608688354492 lr:6.548796093607497e-05 tokens_per_second_per_gpu:5339.284552254906 mfu:25.18631620886786 +Step 14562 | loss:0.6478236317634583 lr:6.548383793301724e-05 tokens_per_second_per_gpu:4981.204543932079 mfu:23.497191714859778 +Step 14563 | loss:0.8173530101776123 lr:6.547971481350361e-05 tokens_per_second_per_gpu:4510.515676233567 mfu:21.276872018123992 +Step 14564 | loss:0.6241537928581238 lr:6.547559157756507e-05 tokens_per_second_per_gpu:4827.865784598469 mfu:22.773866624793026 +Step 14565 | loss:0.8095261454582214 lr:6.547146822523262e-05 tokens_per_second_per_gpu:5312.516333536062 mfu:25.060045954042014 +Step 14566 | loss:0.791994571685791 lr:6.546734475653728e-05 tokens_per_second_per_gpu:3401.1205945829915 mfu:16.04366613565912 +Step 14567 | loss:0.7955880761146545 lr:6.54632211715101e-05 tokens_per_second_per_gpu:5314.615155938439 mfu:25.069946457410055 +Step 14568 | loss:0.7390955686569214 lr:6.545909747018202e-05 tokens_per_second_per_gpu:5204.793159293579 mfu:24.551897361671465 +Step 14569 | loss:0.7140532732009888 lr:6.545497365258411e-05 tokens_per_second_per_gpu:4866.706570813073 mfu:22.957085240288787 +Step 14570 | loss:0.9185658693313599 lr:6.545084971874738e-05 tokens_per_second_per_gpu:3870.5695294508387 mfu:18.258136857678306 +Step 14571 | loss:0.7624638676643372 lr:6.544672566870282e-05 tokens_per_second_per_gpu:4756.202278911207 mfu:22.43581764141133 +Step 14572 | loss:0.7520260214805603 lr:6.544260150248148e-05 tokens_per_second_per_gpu:5367.5091668711075 mfu:25.319456531628312 +Step 14573 | loss:0.7649762630462646 lr:6.543847722011436e-05 tokens_per_second_per_gpu:2992.7661787061816 mfu:14.117388683520463 +Step 14574 | loss:0.8829910755157471 lr:6.543435282163249e-05 tokens_per_second_per_gpu:5156.156869917812 mfu:24.322471686479325 +Step 14575 | loss:0.7348055243492126 lr:6.543022830706687e-05 tokens_per_second_per_gpu:4646.342008219771 mfu:21.917588000464814 +Step 14576 | loss:0.7986544370651245 lr:6.542610367644854e-05 tokens_per_second_per_gpu:4542.0542022144755 mfu:21.42564462620334 +Step 14577 | loss:0.7509179711341858 lr:6.542197892980851e-05 tokens_per_second_per_gpu:4043.0065653205347 mfu:19.071551776667757 +Step 14578 | loss:0.8304858207702637 lr:6.541785406717782e-05 tokens_per_second_per_gpu:4928.974541389312 mfu:23.2508138815082 +Step 14579 | loss:0.7999201416969299 lr:6.541372908858747e-05 tokens_per_second_per_gpu:5287.3137933987155 mfu:24.941161272217403 +Step 14580 | loss:0.7958127856254578 lr:6.54096039940685e-05 tokens_per_second_per_gpu:4042.251105231839 mfu:19.0679881425346 +Step 14581 | loss:0.6791903376579285 lr:6.540547878365194e-05 tokens_per_second_per_gpu:5256.713676517935 mfu:24.796815299972458 +Step 14582 | loss:0.7885284423828125 lr:6.54013534573688e-05 tokens_per_second_per_gpu:4302.694869246286 mfu:20.29654581367901 +Step 14583 | loss:0.7516962885856628 lr:6.539722801525012e-05 tokens_per_second_per_gpu:5332.455425534253 mfu:25.154102052957374 +Step 14584 | loss:0.7332075238227844 lr:6.539310245732691e-05 tokens_per_second_per_gpu:4970.653828683601 mfu:23.44742219089506 +Step 14585 | loss:0.7787508368492126 lr:6.538897678363024e-05 tokens_per_second_per_gpu:4774.685247744241 mfu:22.523004958916413 +Step 14586 | loss:0.6715822219848633 lr:6.53848509941911e-05 tokens_per_second_per_gpu:3389.8971610498033 mfu:15.990723284767851 +Step 14587 | loss:0.645034909248352 lr:6.538072508904051e-05 tokens_per_second_per_gpu:5319.555122210863 mfu:25.093249121162238 +Step 14588 | loss:0.6327651143074036 lr:6.537659906820955e-05 tokens_per_second_per_gpu:4226.947794770431 mfu:19.93923393959284 +Step 14589 | loss:0.8196625113487244 lr:6.537247293172922e-05 tokens_per_second_per_gpu:4085.2874000310153 mfu:19.270997687851366 +Step 14590 | loss:0.7794970273971558 lr:6.536834667963056e-05 tokens_per_second_per_gpu:3619.3735980956317 mfu:17.073202791030482 +Step 14591 | loss:0.8109897375106812 lr:6.53642203119446e-05 tokens_per_second_per_gpu:4708.835660644825 mfu:22.212381221470384 +Step 14592 | loss:0.6805787086486816 lr:6.536009382870239e-05 tokens_per_second_per_gpu:5008.612053821173 mfu:23.626477615210014 +Step 14593 | loss:0.7946361899375916 lr:6.535596722993494e-05 tokens_per_second_per_gpu:4823.703504951009 mfu:22.75423244153782 +Step 14594 | loss:0.6375089883804321 lr:6.53518405156733e-05 tokens_per_second_per_gpu:4406.5997008030545 mfu:20.786682632124602 +Step 14595 | loss:0.6743813753128052 lr:6.534771368594852e-05 tokens_per_second_per_gpu:5099.582967098287 mfu:24.055602934376466 +Step 14596 | loss:0.7398751974105835 lr:6.534358674079161e-05 tokens_per_second_per_gpu:5381.37296632883 mfu:25.384854439078488 +Step 14597 | loss:0.8456605672836304 lr:6.533945968023365e-05 tokens_per_second_per_gpu:3842.7472044884576 mfu:18.126894203852622 +Step 14598 | loss:0.7255265116691589 lr:6.533533250430563e-05 tokens_per_second_per_gpu:3742.650604278432 mfu:17.654721462420884 +Step 14599 | loss:0.7632582783699036 lr:6.533120521303863e-05 tokens_per_second_per_gpu:4588.632342266186 mfu:21.645361659877427 +Step 14600 | loss:0.8852510452270508 lr:6.53270778064637e-05 tokens_per_second_per_gpu:5200.590825247965 mfu:24.532074234986183 +Step 14601 | loss:0.6639740467071533 lr:6.532295028461183e-05 tokens_per_second_per_gpu:4812.502283415143 mfu:22.70139436842768 +Step 14602 | loss:0.8828980922698975 lr:6.531882264751412e-05 tokens_per_second_per_gpu:5315.112812250421 mfu:25.072293987142594 +Step 14603 | loss:0.8097681999206543 lr:6.531469489520158e-05 tokens_per_second_per_gpu:3915.2576506283453 mfu:18.468938349852298 +Step 14604 | loss:0.8853675127029419 lr:6.531056702770525e-05 tokens_per_second_per_gpu:4057.0833054223212 mfu:19.137954161467405 +Step 14605 | loss:0.7521494030952454 lr:6.530643904505621e-05 tokens_per_second_per_gpu:3955.785737313405 mfu:18.66011624955035 +Step 14606 | loss:0.8867227435112 lr:6.530231094728549e-05 tokens_per_second_per_gpu:4567.109683471516 mfu:21.54383560619872 +Step 14607 | loss:0.7708849310874939 lr:6.529818273442413e-05 tokens_per_second_per_gpu:5020.326079817799 mfu:23.681734674455026 +Step 14608 | loss:0.7158076763153076 lr:6.529405440650319e-05 tokens_per_second_per_gpu:4392.552679810258 mfu:20.72042043743267 +Step 14609 | loss:0.6977199912071228 lr:6.52899259635537e-05 tokens_per_second_per_gpu:5089.424342258039 mfu:24.007682967764893 +Step 14610 | loss:0.6896757483482361 lr:6.528579740560674e-05 tokens_per_second_per_gpu:4310.867799458092 mfu:20.335098919933785 +Step 14611 | loss:0.7923511862754822 lr:6.528166873269334e-05 tokens_per_second_per_gpu:5338.028699051901 mfu:25.180392135038637 +Step 14612 | loss:0.9572293758392334 lr:6.527753994484458e-05 tokens_per_second_per_gpu:5332.344609456019 mfu:25.15357931460917 +Step 14613 | loss:0.799092710018158 lr:6.527341104209149e-05 tokens_per_second_per_gpu:4582.097808234292 mfu:21.614537147942507 +Step 14614 | loss:0.7620932459831238 lr:6.52692820244651e-05 tokens_per_second_per_gpu:3935.6715645421696 mfu:18.565234264756675 +Step 14615 | loss:0.7596803307533264 lr:6.52651528919965e-05 tokens_per_second_per_gpu:4330.781713084056 mfu:20.429036248171478 +Step 14616 | loss:0.6513038873672485 lr:6.526102364471674e-05 tokens_per_second_per_gpu:4599.573111958301 mfu:21.696971137202926 +Step 14617 | loss:0.8432983160018921 lr:6.525689428265689e-05 tokens_per_second_per_gpu:4651.131335444951 mfu:21.940180074129476 +Step 14618 | loss:0.7522951364517212 lr:6.525276480584797e-05 tokens_per_second_per_gpu:3816.981571370449 mfu:18.00535331636466 +Step 14619 | loss:0.8387033939361572 lr:6.524863521432108e-05 tokens_per_second_per_gpu:4704.513547664823 mfu:22.191993077115196 +Step 14620 | loss:0.7042510509490967 lr:6.524450550810723e-05 tokens_per_second_per_gpu:4234.420160673145 mfu:19.974482364473026 +Step 14621 | loss:0.8023359775543213 lr:6.524037568723754e-05 tokens_per_second_per_gpu:4955.845785657283 mfu:23.37757012542725 +Step 14622 | loss:0.8278599381446838 lr:6.523624575174302e-05 tokens_per_second_per_gpu:4090.7721598569615 mfu:19.296870235746876 +Step 14623 | loss:0.915031373500824 lr:6.523211570165475e-05 tokens_per_second_per_gpu:3822.5935978552384 mfu:18.031826202803863 +Step 14624 | loss:0.6831700205802917 lr:6.522798553700382e-05 tokens_per_second_per_gpu:4889.204276930663 mfu:23.06321075854994 +Step 14625 | loss:0.7831135988235474 lr:6.522385525782124e-05 tokens_per_second_per_gpu:4890.896003650282 mfu:23.071190922124725 +Step 14626 | loss:0.6726461052894592 lr:6.521972486413811e-05 tokens_per_second_per_gpu:3367.785460873838 mfu:15.88641862239275 +Step 14627 | loss:0.8516916036605835 lr:6.521559435598548e-05 tokens_per_second_per_gpu:4850.759272720584 mfu:22.881859114296894 +Step 14628 | loss:0.6911677122116089 lr:6.521146373339444e-05 tokens_per_second_per_gpu:5308.349497519416 mfu:25.040390277615963 +Step 14629 | loss:0.7481224536895752 lr:6.520733299639603e-05 tokens_per_second_per_gpu:5368.16268730214 mfu:25.322539299003573 +Step 14630 | loss:0.807133674621582 lr:6.520320214502134e-05 tokens_per_second_per_gpu:4725.59728975347 mfu:22.291448685804685 +Step 14631 | loss:0.750912070274353 lr:6.519907117930141e-05 tokens_per_second_per_gpu:4714.91008425002 mfu:22.241035314019623 +Step 14632 | loss:0.7529976963996887 lr:6.519494009926733e-05 tokens_per_second_per_gpu:4585.072376940141 mfu:21.628568696041754 +Step 14633 | loss:0.6986599564552307 lr:6.519080890495017e-05 tokens_per_second_per_gpu:4887.7496571896245 mfu:23.05634906904774 +Step 14634 | loss:0.7600091695785522 lr:6.518667759638099e-05 tokens_per_second_per_gpu:4375.668777031157 mfu:20.640776187332786 +Step 14635 | loss:0.8056432604789734 lr:6.518254617359088e-05 tokens_per_second_per_gpu:5332.260121372281 mfu:25.15318076990303 +Step 14636 | loss:0.8118228912353516 lr:6.51784146366109e-05 tokens_per_second_per_gpu:5350.115315712804 mfu:25.237406768017564 +Step 14637 | loss:0.6316445469856262 lr:6.517428298547213e-05 tokens_per_second_per_gpu:4475.736811443459 mfu:21.112814179022834 +Step 14638 | loss:0.7587898969650269 lr:6.517015122020563e-05 tokens_per_second_per_gpu:3146.433884231258 mfu:14.84226546889637 +Step 14639 | loss:0.8612481355667114 lr:6.51660193408425e-05 tokens_per_second_per_gpu:3741.705951312969 mfu:17.65026537320838 +Step 14640 | loss:0.7751033902168274 lr:6.516188734741379e-05 tokens_per_second_per_gpu:5312.02239211891 mfu:25.057715948102214 +Step 14641 | loss:1.0115963220596313 lr:6.51577552399506e-05 tokens_per_second_per_gpu:5304.710398657255 mfu:25.023224027388856 +Step 14642 | loss:0.8024751543998718 lr:6.515362301848402e-05 tokens_per_second_per_gpu:5281.749502794825 mfu:24.91491356407298 +Step 14643 | loss:0.7832466959953308 lr:6.514949068304508e-05 tokens_per_second_per_gpu:4102.563119782039 mfu:19.35249020545868 +Step 14644 | loss:0.9046232104301453 lr:6.514535823366489e-05 tokens_per_second_per_gpu:4352.447802500543 mfu:20.531238888565085 +Step 14645 | loss:0.8479042053222656 lr:6.514122567037452e-05 tokens_per_second_per_gpu:4366.074854477557 mfu:20.595520017755476 +Step 14646 | loss:0.7072458863258362 lr:6.513709299320508e-05 tokens_per_second_per_gpu:4817.052832014233 mfu:22.722860082573238 +Step 14647 | loss:0.8212007284164429 lr:6.513296020218762e-05 tokens_per_second_per_gpu:4222.293444151587 mfu:19.91727857360965 +Step 14648 | loss:0.8064152002334595 lr:6.512882729735324e-05 tokens_per_second_per_gpu:4858.520889755607 mfu:22.918471986119513 +Step 14649 | loss:0.7457682490348816 lr:6.512469427873302e-05 tokens_per_second_per_gpu:3543.9141412759072 mfu:16.717247658500913 +Step 14650 | loss:0.7694646120071411 lr:6.512056114635804e-05 tokens_per_second_per_gpu:3641.1245951868677 mfu:17.175805955412624 +Step 14651 | loss:0.7188667058944702 lr:6.51164279002594e-05 tokens_per_second_per_gpu:4102.982692822873 mfu:19.35446940307883 +Step 14652 | loss:0.7437499761581421 lr:6.511229454046816e-05 tokens_per_second_per_gpu:5354.086768848464 mfu:25.25614079005825 +Step 14653 | loss:0.7690401077270508 lr:6.510816106701543e-05 tokens_per_second_per_gpu:4663.542276808502 mfu:21.998724602066293 +Step 14654 | loss:0.7933407425880432 lr:6.510402747993229e-05 tokens_per_second_per_gpu:5343.446088645829 mfu:25.20594688605517 +Step 14655 | loss:0.828859269618988 lr:6.509989377924985e-05 tokens_per_second_per_gpu:3917.593420857769 mfu:18.479956576547988 +Step 14656 | loss:0.7702929377555847 lr:6.509575996499916e-05 tokens_per_second_per_gpu:4250.349185281161 mfu:20.049622291320887 +Step 14657 | loss:0.7071632742881775 lr:6.509162603721135e-05 tokens_per_second_per_gpu:4159.782009139256 mfu:19.622401469105874 +Step 14658 | loss:0.743722677230835 lr:6.50874919959175e-05 tokens_per_second_per_gpu:3514.291708770983 mfu:16.577513590267035 +Step 14659 | loss:0.8748360872268677 lr:6.508335784114867e-05 tokens_per_second_per_gpu:4314.7001225566255 mfu:20.353176641851313 +Step 14660 | loss:0.6902998685836792 lr:6.507922357293601e-05 tokens_per_second_per_gpu:3869.491160916707 mfu:18.253050009314165 +Step 14661 | loss:0.7377132177352905 lr:6.507508919131057e-05 tokens_per_second_per_gpu:4813.614945464013 mfu:22.706642985151916 +Step 14662 | loss:0.8500828146934509 lr:6.507095469630347e-05 tokens_per_second_per_gpu:4652.551495067639 mfu:21.946879209373908 +Step 14663 | loss:0.8066176176071167 lr:6.506682008794578e-05 tokens_per_second_per_gpu:3863.617071872072 mfu:18.225340929067684 +Step 14664 | loss:0.8394631147384644 lr:6.506268536626862e-05 tokens_per_second_per_gpu:4818.568775502626 mfu:22.730011046654614 +Step 14665 | loss:0.8660210371017456 lr:6.505855053130308e-05 tokens_per_second_per_gpu:5138.593453976843 mfu:24.23962205685801 +Step 14666 | loss:0.7963169813156128 lr:6.505441558308027e-05 tokens_per_second_per_gpu:3730.73647391941 mfu:17.598520476757127 +Step 14667 | loss:0.7705996632575989 lr:6.505028052163128e-05 tokens_per_second_per_gpu:5320.527656847645 mfu:25.097836732975455 +Step 14668 | loss:0.7733199000358582 lr:6.50461453469872e-05 tokens_per_second_per_gpu:4554.786054156302 mfu:21.485702944090875 +Step 14669 | loss:0.8392058610916138 lr:6.504201005917915e-05 tokens_per_second_per_gpu:4344.878806054195 mfu:20.4955346409237 +Step 14670 | loss:0.6980055570602417 lr:6.50378746582382e-05 tokens_per_second_per_gpu:4796.976233193766 mfu:22.628155340516116 +Step 14671 | loss:0.736697256565094 lr:6.50337391441955e-05 tokens_per_second_per_gpu:5294.921086943643 mfu:24.977046173806944 +Step 14672 | loss:0.8593029975891113 lr:6.502960351708212e-05 tokens_per_second_per_gpu:5039.5297311858785 mfu:23.77232157443904 +Step 14673 | loss:0.788333535194397 lr:6.502546777692919e-05 tokens_per_second_per_gpu:3812.5153437613644 mfu:17.9842853587156 +Step 14674 | loss:0.7353000044822693 lr:6.502133192376778e-05 tokens_per_second_per_gpu:5340.072719297032 mfu:25.190034127280732 +Step 14675 | loss:0.8461940884590149 lr:6.501719595762903e-05 tokens_per_second_per_gpu:3735.7548659220834 mfu:17.622193088059706 +Step 14676 | loss:0.744777500629425 lr:6.501305987854403e-05 tokens_per_second_per_gpu:4783.143601769688 mfu:22.562904457998457 +Step 14677 | loss:0.7400062680244446 lr:6.500892368654388e-05 tokens_per_second_per_gpu:5339.767910357701 mfu:25.188596291508144 +Step 14678 | loss:0.6963614225387573 lr:6.50047873816597e-05 tokens_per_second_per_gpu:4615.435388409107 mfu:21.77179620160482 +Step 14679 | loss:0.8846877217292786 lr:6.500065096392261e-05 tokens_per_second_per_gpu:4411.574633071227 mfu:20.810150236444546 +Step 14680 | loss:0.6715012788772583 lr:6.49965144333637e-05 tokens_per_second_per_gpu:5194.838419761789 mfu:24.50493915684631 +Step 14681 | loss:0.7466753125190735 lr:6.49923777900141e-05 tokens_per_second_per_gpu:3407.4810359786347 mfu:16.073669422925573 +Step 14682 | loss:0.7848394513130188 lr:6.498824103390492e-05 tokens_per_second_per_gpu:5337.462129864553 mfu:25.177719531515248 +Step 14683 | loss:0.874406099319458 lr:6.498410416506725e-05 tokens_per_second_per_gpu:5355.148870797419 mfu:25.261150906164747 +Step 14684 | loss:0.8179459571838379 lr:6.497996718353223e-05 tokens_per_second_per_gpu:4451.600962584162 mfu:20.998961261953077 +Step 14685 | loss:0.7470417618751526 lr:6.497583008933097e-05 tokens_per_second_per_gpu:3418.281591244564 mfu:16.12461748487971 +Step 14686 | loss:0.8223628401756287 lr:6.497169288249456e-05 tokens_per_second_per_gpu:5344.172078353302 mfu:25.20937150337194 +Step 14687 | loss:0.7385568022727966 lr:6.496755556305417e-05 tokens_per_second_per_gpu:5115.154536758037 mfu:24.12905668524525 +Step 14688 | loss:0.7672679424285889 lr:6.496341813104085e-05 tokens_per_second_per_gpu:5326.716686690912 mfu:25.127031442703522 +Step 14689 | loss:0.7732924222946167 lr:6.495928058648578e-05 tokens_per_second_per_gpu:5234.141395867678 mfu:24.690338000916988 +Step 14690 | loss:0.7899341583251953 lr:6.495514292942004e-05 tokens_per_second_per_gpu:5259.798391908883 mfu:24.811366428778275 +Step 14691 | loss:0.7723177671432495 lr:6.495100515987476e-05 tokens_per_second_per_gpu:5315.754700107392 mfu:25.07532188168129 +Step 14692 | loss:0.8681325912475586 lr:6.494686727788107e-05 tokens_per_second_per_gpu:4034.841351098065 mfu:19.033035068051166 +Step 14693 | loss:0.7414475679397583 lr:6.494272928347007e-05 tokens_per_second_per_gpu:5294.850629570849 mfu:24.97671381435821 +Step 14694 | loss:0.7813838124275208 lr:6.493859117667293e-05 tokens_per_second_per_gpu:4811.3071324429575 mfu:22.695756637378096 +Step 14695 | loss:0.7754947543144226 lr:6.493445295752069e-05 tokens_per_second_per_gpu:5328.473579885503 mfu:25.135319007659294 +Step 14696 | loss:0.7815119028091431 lr:6.493031462604457e-05 tokens_per_second_per_gpu:4090.2257305475605 mfu:19.29429263546398 +Step 14697 | loss:0.7613714933395386 lr:6.492617618227562e-05 tokens_per_second_per_gpu:4515.118444399289 mfu:21.29858406087493 +Step 14698 | loss:0.7964048385620117 lr:6.4922037626245e-05 tokens_per_second_per_gpu:4774.3072402990065 mfu:22.52122183330252 +Step 14699 | loss:0.6643336415290833 lr:6.491789895798383e-05 tokens_per_second_per_gpu:5314.083083382154 mfu:25.067436580381116 +Step 14700 | loss:0.8245453834533691 lr:6.491376017752324e-05 tokens_per_second_per_gpu:5324.132140196837 mfu:25.11483969592187 +Step 14701 | loss:0.841705322265625 lr:6.490962128489437e-05 tokens_per_second_per_gpu:3544.238232064467 mfu:16.71877645004614 +Step 14702 | loss:0.7291477918624878 lr:6.490548228012831e-05 tokens_per_second_per_gpu:4439.962259134928 mfu:20.944059512015468 +Step 14703 | loss:0.799131453037262 lr:6.490134316325624e-05 tokens_per_second_per_gpu:4407.632719341539 mfu:20.79155555682177 +Step 14704 | loss:0.8773954510688782 lr:6.489720393430924e-05 tokens_per_second_per_gpu:4891.962962578193 mfu:23.076223949429643 +Step 14705 | loss:0.7575390934944153 lr:6.48930645933185e-05 tokens_per_second_per_gpu:5186.945794879982 mfu:24.46770829865816 +Step 14706 | loss:0.7650465369224548 lr:6.488892514031511e-05 tokens_per_second_per_gpu:4948.257201995217 mfu:23.341773481547406 +Step 14707 | loss:0.8984703421592712 lr:6.48847855753302e-05 tokens_per_second_per_gpu:4482.086761389807 mfu:21.142767976332934 +Step 14708 | loss:0.7542654275894165 lr:6.488064589839494e-05 tokens_per_second_per_gpu:4639.257480996728 mfu:21.884169076808735 +Step 14709 | loss:0.7828595042228699 lr:6.487650610954042e-05 tokens_per_second_per_gpu:4117.6443540308 mfu:19.423630960533977 +Step 14710 | loss:0.6419900059700012 lr:6.487236620879782e-05 tokens_per_second_per_gpu:5249.223372943217 mfu:24.761482260032942 +Step 14711 | loss:0.8935638070106506 lr:6.486822619619824e-05 tokens_per_second_per_gpu:4055.331667383087 mfu:19.12969138597607 +Step 14712 | loss:0.7908053398132324 lr:6.486408607177286e-05 tokens_per_second_per_gpu:3994.1598821432135 mfu:18.841133638016796 +Step 14713 | loss:0.7574391961097717 lr:6.485994583555275e-05 tokens_per_second_per_gpu:5381.574016736661 mfu:25.385802828154965 +Step 14714 | loss:0.8550200462341309 lr:6.485580548756912e-05 tokens_per_second_per_gpu:4991.0862723226 mfu:23.543805513674585 +Step 14715 | loss:0.7848623991012573 lr:6.485166502785306e-05 tokens_per_second_per_gpu:5351.971114532103 mfu:25.246160887680126 +Step 14716 | loss:0.8042016625404358 lr:6.484752445643577e-05 tokens_per_second_per_gpu:5229.347633711455 mfu:24.667725006925924 +Step 14717 | loss:0.8149064779281616 lr:6.484338377334832e-05 tokens_per_second_per_gpu:3803.4716370574006 mfu:17.94162465117753 +Step 14718 | loss:0.7517837882041931 lr:6.483924297862189e-05 tokens_per_second_per_gpu:4626.408817716225 mfu:21.823559739906592 +Step 14719 | loss:0.827961802482605 lr:6.483510207228765e-05 tokens_per_second_per_gpu:3744.9084959787674 mfu:17.665372322807414 +Step 14720 | loss:0.7756011486053467 lr:6.483096105437668e-05 tokens_per_second_per_gpu:4659.076028475134 mfu:21.977656546657254 +Step 14721 | loss:0.7770676016807556 lr:6.482681992492017e-05 tokens_per_second_per_gpu:5332.787466554675 mfu:25.155668347102747 +Step 14722 | loss:0.7134742140769958 lr:6.482267868394926e-05 tokens_per_second_per_gpu:4227.541871794309 mfu:19.942036302274467 +Step 14723 | loss:0.789157509803772 lr:6.48185373314951e-05 tokens_per_second_per_gpu:5373.781048971507 mfu:25.349042069589323 +Step 14724 | loss:0.8669243454933167 lr:6.481439586758881e-05 tokens_per_second_per_gpu:3835.6788145677783 mfu:18.0935514026051 +Step 14725 | loss:0.7764497995376587 lr:6.481025429226158e-05 tokens_per_second_per_gpu:4158.1499282288905 mfu:19.614702664989963 +Step 14726 | loss:0.7669510245323181 lr:6.480611260554451e-05 tokens_per_second_per_gpu:3619.745334204791 mfu:17.07495633368211 +Step 14727 | loss:0.8072210550308228 lr:6.48019708074688e-05 tokens_per_second_per_gpu:4864.870654653226 mfu:22.94842491052319 +Step 14728 | loss:0.8041776418685913 lr:6.479782889806557e-05 tokens_per_second_per_gpu:3456.79603546391 mfu:16.306296689504155 +Step 14729 | loss:0.8095958828926086 lr:6.479368687736598e-05 tokens_per_second_per_gpu:5050.030171461825 mfu:23.821853942780535 +Step 14730 | loss:0.7562230825424194 lr:6.47895447454012e-05 tokens_per_second_per_gpu:5356.811481166002 mfu:25.268993722943936 +Step 14731 | loss:0.699954628944397 lr:6.478540250220234e-05 tokens_per_second_per_gpu:4392.80043310486 mfu:20.72158913199437 +Step 14732 | loss:0.7695295810699463 lr:6.478126014780059e-05 tokens_per_second_per_gpu:5382.63924844993 mfu:25.390827707893724 +Step 14733 | loss:0.7620230317115784 lr:6.47771176822271e-05 tokens_per_second_per_gpu:5318.347062315286 mfu:25.087550496518528 +Step 14734 | loss:0.8291916847229004 lr:6.477297510551304e-05 tokens_per_second_per_gpu:5345.133745996386 mfu:25.213907853721782 +Step 14735 | loss:0.7297791838645935 lr:6.476883241768952e-05 tokens_per_second_per_gpu:5320.316757282117 mfu:25.096841883741682 +Step 14736 | loss:0.6766647696495056 lr:6.476468961878774e-05 tokens_per_second_per_gpu:5338.921174028162 mfu:25.184602091773588 +Step 14737 | loss:0.8797612190246582 lr:6.476054670883884e-05 tokens_per_second_per_gpu:3619.531678289815 mfu:17.073948482277707 +Step 14738 | loss:0.8595319390296936 lr:6.4756403687874e-05 tokens_per_second_per_gpu:5321.893228554657 mfu:25.10427836770799 +Step 14739 | loss:0.7136420011520386 lr:6.475226055592435e-05 tokens_per_second_per_gpu:5275.880114689035 mfu:24.887226658957093 +Step 14740 | loss:0.769898533821106 lr:6.474811731302106e-05 tokens_per_second_per_gpu:4935.281492858662 mfu:23.28056484766601 +Step 14741 | loss:0.9295171499252319 lr:6.474397395919532e-05 tokens_per_second_per_gpu:4137.611540810736 mfu:19.51781958732829 +Step 14742 | loss:0.8150200843811035 lr:6.473983049447825e-05 tokens_per_second_per_gpu:5240.590390899816 mfu:24.7207590107956 +Step 14743 | loss:0.7982582449913025 lr:6.473568691890104e-05 tokens_per_second_per_gpu:3546.8546037748265 mfu:16.73111832180295 +Step 14744 | loss:1.1163792610168457 lr:6.473154323249485e-05 tokens_per_second_per_gpu:4937.603198169121 mfu:23.291516727739214 +Step 14745 | loss:0.841486930847168 lr:6.472739943529085e-05 tokens_per_second_per_gpu:5303.015730758948 mfu:25.015229989772383 +Step 14746 | loss:0.7456015348434448 lr:6.472325552732019e-05 tokens_per_second_per_gpu:5352.287025117199 mfu:25.247651091809928 +Step 14747 | loss:0.6700978875160217 lr:6.471911150861404e-05 tokens_per_second_per_gpu:5326.635300896877 mfu:25.126647532027217 +Step 14748 | loss:0.745551347732544 lr:6.47149673792036e-05 tokens_per_second_per_gpu:5348.938974367408 mfu:25.231857765186803 +Step 14749 | loss:0.7603101134300232 lr:6.471082313912e-05 tokens_per_second_per_gpu:5156.034644888101 mfu:24.32189513016043 +Step 14750 | loss:0.7877525091171265 lr:6.470667878839442e-05 tokens_per_second_per_gpu:5347.8640539838425 mfu:25.226787182337194 +Step 14751 | loss:0.7622143626213074 lr:6.470253432705801e-05 tokens_per_second_per_gpu:5367.117877979112 mfu:25.317610755163997 +Step 14752 | loss:0.6899517774581909 lr:6.4698389755142e-05 tokens_per_second_per_gpu:5267.969158716218 mfu:24.849909329885044 +Step 14753 | loss:0.736335039138794 lr:6.46942450726775e-05 tokens_per_second_per_gpu:4909.993841377538 mfu:23.161278681111305 +Step 14754 | loss:0.7288832664489746 lr:6.469010027969573e-05 tokens_per_second_per_gpu:3645.1574750607792 mfu:17.19482973785812 +Step 14755 | loss:0.7891706228256226 lr:6.468595537622783e-05 tokens_per_second_per_gpu:4684.466918577795 mfu:22.09742970740407 +Step 14756 | loss:0.6680988073348999 lr:6.468181036230498e-05 tokens_per_second_per_gpu:2867.713838454941 mfu:13.527495458426005 +Step 14757 | loss:0.971356213092804 lr:6.467766523795837e-05 tokens_per_second_per_gpu:5087.258660247122 mfu:23.997467076215944 +Step 14758 | loss:0.7684952616691589 lr:6.467352000321915e-05 tokens_per_second_per_gpu:5079.781431654931 mfu:23.96219571319983 +Step 14759 | loss:0.7395423650741577 lr:6.466937465811854e-05 tokens_per_second_per_gpu:4461.040246278105 mfu:21.043487973645274 +Step 14760 | loss:0.802239716053009 lr:6.466522920268768e-05 tokens_per_second_per_gpu:4316.808695141661 mfu:20.363123138494505 +Step 14761 | loss:0.9233314990997314 lr:6.466108363695775e-05 tokens_per_second_per_gpu:3156.9836816683874 mfu:14.89203066338836 +Step 14762 | loss:0.7637088298797607 lr:6.465693796095994e-05 tokens_per_second_per_gpu:4922.745464977159 mfu:23.221430265282045 +Step 14763 | loss:0.7284086346626282 lr:6.465279217472544e-05 tokens_per_second_per_gpu:5341.705269145096 mfu:25.19773514345578 +Step 14764 | loss:0.7513992190361023 lr:6.464864627828541e-05 tokens_per_second_per_gpu:4122.807501056247 mfu:19.447986405976728 +Step 14765 | loss:0.9479658007621765 lr:6.464450027167105e-05 tokens_per_second_per_gpu:4104.325879215303 mfu:19.360805442463153 +Step 14766 | loss:0.8515324592590332 lr:6.464035415491354e-05 tokens_per_second_per_gpu:4710.446534218931 mfu:22.219979986962816 +Step 14767 | loss:0.8027881383895874 lr:6.463620792804404e-05 tokens_per_second_per_gpu:4226.266826678665 mfu:19.936021696918843 +Step 14768 | loss:0.825424313545227 lr:6.463206159109377e-05 tokens_per_second_per_gpu:4580.414843197596 mfu:21.60659831472097 +Step 14769 | loss:0.7395482659339905 lr:6.462791514409389e-05 tokens_per_second_per_gpu:4160.819188530639 mfu:19.627294021256077 +Step 14770 | loss:0.8520165085792542 lr:6.46237685870756e-05 tokens_per_second_per_gpu:3604.326911196519 mfu:17.00222500169786 +Step 14771 | loss:0.7157797813415527 lr:6.461962192007008e-05 tokens_per_second_per_gpu:4371.691641986389 mfu:20.622015362757402 +Step 14772 | loss:0.7941854596138 lr:6.461547514310851e-05 tokens_per_second_per_gpu:3501.5360948132434 mfu:16.51734318289622 +Step 14773 | loss:0.6853699684143066 lr:6.46113282562221e-05 tokens_per_second_per_gpu:4386.60309840193 mfu:20.692355246826317 +Step 14774 | loss:0.8913918137550354 lr:6.460718125944201e-05 tokens_per_second_per_gpu:4085.119894266629 mfu:19.270207534581328 +Step 14775 | loss:0.6188100576400757 lr:6.460303415279945e-05 tokens_per_second_per_gpu:5331.795032299751 mfu:25.150986865395527 +Step 14776 | loss:0.8683683276176453 lr:6.459888693632559e-05 tokens_per_second_per_gpu:4106.642209019691 mfu:19.371731965357068 +Step 14777 | loss:0.8509989380836487 lr:6.459473961005168e-05 tokens_per_second_per_gpu:5392.750274905275 mfu:25.438523144802588 +Step 14778 | loss:0.8216965794563293 lr:6.459059217400883e-05 tokens_per_second_per_gpu:3862.578751054919 mfu:18.220442992617375 +Step 14779 | loss:0.7479257583618164 lr:6.45864446282283e-05 tokens_per_second_per_gpu:5098.48239474164 mfu:24.050411346793158 +Step 14780 | loss:0.8217970132827759 lr:6.458229697274125e-05 tokens_per_second_per_gpu:5362.8857017258315 mfu:25.297646857693547 +Step 14781 | loss:0.989063560962677 lr:6.45781492075789e-05 tokens_per_second_per_gpu:2867.633603963183 mfu:13.527116978638965 +Step 14782 | loss:0.744396448135376 lr:6.45740013327724e-05 tokens_per_second_per_gpu:4379.555916862965 mfu:20.659112489133868 +Step 14783 | loss:0.8098858594894409 lr:6.4569853348353e-05 tokens_per_second_per_gpu:5330.486023093047 mfu:25.144812045627503 +Step 14784 | loss:0.7223292589187622 lr:6.456570525435187e-05 tokens_per_second_per_gpu:3541.410922463523 mfu:16.705439548269727 +Step 14785 | loss:0.7649707794189453 lr:6.45615570508002e-05 tokens_per_second_per_gpu:4166.310238314342 mfu:19.653196239955506 +Step 14786 | loss:0.8792482018470764 lr:6.455740873772921e-05 tokens_per_second_per_gpu:5366.367225093083 mfu:25.31406979742601 +Step 14787 | loss:0.732572615146637 lr:6.455326031517009e-05 tokens_per_second_per_gpu:4810.764114197156 mfu:22.693195127664186 +Step 14788 | loss:0.8820160031318665 lr:6.454911178315403e-05 tokens_per_second_per_gpu:3942.674245911842 mfu:18.598267107558062 +Step 14789 | loss:0.7593836188316345 lr:6.454496314171226e-05 tokens_per_second_per_gpu:3280.4452902611 mfu:15.474420135842736 +Step 14790 | loss:0.7246758341789246 lr:6.454081439087596e-05 tokens_per_second_per_gpu:4778.883591556118 mfu:22.542809262988353 +Step 14791 | loss:0.7016324996948242 lr:6.453666553067634e-05 tokens_per_second_per_gpu:5357.675602849298 mfu:25.273069932358407 +Step 14792 | loss:0.7757673859596252 lr:6.453251656114462e-05 tokens_per_second_per_gpu:4139.877911043713 mfu:19.528510442399877 +Step 14793 | loss:0.9327627420425415 lr:6.452836748231197e-05 tokens_per_second_per_gpu:3483.1925714256977 mfu:16.43081365334935 +Step 14794 | loss:0.8036597967147827 lr:6.452421829420961e-05 tokens_per_second_per_gpu:3960.9675191883252 mfu:18.6845595987575 +Step 14795 | loss:0.8830363750457764 lr:6.452006899686878e-05 tokens_per_second_per_gpu:4314.018289932612 mfu:20.349960321031197 +Step 14796 | loss:0.811098039150238 lr:6.451591959032061e-05 tokens_per_second_per_gpu:4110.583995437807 mfu:19.390326044429436 +Step 14797 | loss:0.784382700920105 lr:6.45117700745964e-05 tokens_per_second_per_gpu:4752.258199101455 mfu:22.41721274401932 +Step 14798 | loss:0.8550505638122559 lr:6.450762044972729e-05 tokens_per_second_per_gpu:4353.793525579151 mfu:20.53758689393091 +Step 14799 | loss:0.8304274678230286 lr:6.450347071574454e-05 tokens_per_second_per_gpu:4829.351291354302 mfu:22.780874013613875 +Step 14800 | loss:0.7410193681716919 lr:6.449932087267932e-05 tokens_per_second_per_gpu:5090.898677482988 mfu:24.014637658568507 +Step 14801 | loss:0.8609374761581421 lr:6.449517092056286e-05 tokens_per_second_per_gpu:3440.454458106992 mfu:16.22921068095097 +Step 14802 | loss:0.7219933867454529 lr:6.449102085942637e-05 tokens_per_second_per_gpu:5197.011776379427 mfu:24.515191250824582 +Step 14803 | loss:0.7821269035339355 lr:6.448687068930105e-05 tokens_per_second_per_gpu:4043.3982319250463 mfu:19.073399334866455 +Step 14804 | loss:0.8306114673614502 lr:6.448272041021814e-05 tokens_per_second_per_gpu:4646.59455338807 mfu:21.91877929911233 +Step 14805 | loss:0.7358832359313965 lr:6.447857002220883e-05 tokens_per_second_per_gpu:5356.904608189752 mfu:25.269433019004072 +Step 14806 | loss:0.6679794788360596 lr:6.447441952530438e-05 tokens_per_second_per_gpu:4662.21294194434 mfu:21.992453902704124 +Step 14807 | loss:1.0357301235198975 lr:6.447026891953594e-05 tokens_per_second_per_gpu:4674.379444411386 mfu:22.049845370660552 +Step 14808 | loss:0.6480774283409119 lr:6.446611820493476e-05 tokens_per_second_per_gpu:4976.448619565665 mfu:23.474757208179028 +Step 14809 | loss:0.9840295910835266 lr:6.446196738153209e-05 tokens_per_second_per_gpu:5216.197816830306 mfu:24.60569507710743 +Step 14810 | loss:0.7041829824447632 lr:6.44578164493591e-05 tokens_per_second_per_gpu:5346.908160977337 mfu:25.222278071933342 +Step 14811 | loss:0.7944039106369019 lr:6.445366540844701e-05 tokens_per_second_per_gpu:4814.4532536452325 mfu:22.71059742787286 +Step 14812 | loss:0.8262348771095276 lr:6.444951425882708e-05 tokens_per_second_per_gpu:5333.764236333365 mfu:25.160275936802936 +Step 14813 | loss:0.750735342502594 lr:6.444536300053052e-05 tokens_per_second_per_gpu:5312.067312065834 mfu:25.057927843155987 +Step 14814 | loss:0.7163955569267273 lr:6.444121163358852e-05 tokens_per_second_per_gpu:4313.246086592702 mfu:20.346317706125365 +Step 14815 | loss:0.8778377175331116 lr:6.443706015803233e-05 tokens_per_second_per_gpu:4851.146416785924 mfu:22.88368533890669 +Step 14816 | loss:0.7172548770904541 lr:6.443290857389317e-05 tokens_per_second_per_gpu:4178.930333826714 mfu:19.71272738369769 +Step 14817 | loss:0.7572228312492371 lr:6.442875688120227e-05 tokens_per_second_per_gpu:4861.607031544324 mfu:22.93302984348698 +Step 14818 | loss:0.7639235258102417 lr:6.442460507999084e-05 tokens_per_second_per_gpu:3919.011078911134 mfu:18.486643911463116 +Step 14819 | loss:0.808375358581543 lr:6.442045317029011e-05 tokens_per_second_per_gpu:5358.02731338412 mfu:25.274729010959167 +Step 14820 | loss:0.9074976444244385 lr:6.441630115213132e-05 tokens_per_second_per_gpu:3173.7065822501795 mfu:14.970915438653726 +Step 14821 | loss:0.7769338488578796 lr:6.44121490255457e-05 tokens_per_second_per_gpu:5347.372563317238 mfu:25.224468737005868 +Step 14822 | loss:0.8149071335792542 lr:6.440799679056445e-05 tokens_per_second_per_gpu:5330.030662999759 mfu:25.14266403437502 +Step 14823 | loss:0.7017587423324585 lr:6.44038444472188e-05 tokens_per_second_per_gpu:3598.4054741719647 mfu:16.9742925729517 +Step 14824 | loss:0.7275309562683105 lr:6.439969199554003e-05 tokens_per_second_per_gpu:4349.448021121709 mfu:20.517088407985643 +Step 14825 | loss:0.7269409894943237 lr:6.439553943555933e-05 tokens_per_second_per_gpu:5318.548751855139 mfu:25.0885019004898 +Step 14826 | loss:0.7666370868682861 lr:6.439138676730793e-05 tokens_per_second_per_gpu:4616.454092749266 mfu:21.77660160378623 +Step 14827 | loss:0.7537158727645874 lr:6.43872339908171e-05 tokens_per_second_per_gpu:5303.829344494164 mfu:25.019067944578218 +Step 14828 | loss:0.8129153251647949 lr:6.438308110611802e-05 tokens_per_second_per_gpu:5352.759930787988 mfu:25.249881868545067 +Step 14829 | loss:0.9418352246284485 lr:6.437892811324198e-05 tokens_per_second_per_gpu:4258.212602015206 mfu:20.08671536969262 +Step 14830 | loss:0.696310818195343 lr:6.437477501222016e-05 tokens_per_second_per_gpu:4778.857317830203 mfu:22.542685325340454 +Step 14831 | loss:0.7083160877227783 lr:6.437062180308384e-05 tokens_per_second_per_gpu:5345.709441677436 mfu:25.21662350847311 +Step 14832 | loss:0.9078436493873596 lr:6.436646848586422e-05 tokens_per_second_per_gpu:4168.95768590353 mfu:19.665684701934843 +Step 14833 | loss:0.8194501399993896 lr:6.436231506059258e-05 tokens_per_second_per_gpu:4913.565107341806 mfu:23.178124951985563 +Step 14834 | loss:0.7595046162605286 lr:6.435816152730012e-05 tokens_per_second_per_gpu:4190.986403530375 mfu:19.76959792146746 +Step 14835 | loss:0.8346990346908569 lr:6.43540078860181e-05 tokens_per_second_per_gpu:4090.4078728279414 mfu:19.295151831677213 +Step 14836 | loss:0.7787623405456543 lr:6.434985413677776e-05 tokens_per_second_per_gpu:5335.208334726038 mfu:25.167087995309767 +Step 14837 | loss:0.7909834980964661 lr:6.434570027961034e-05 tokens_per_second_per_gpu:4952.8685412759005 mfu:23.36352595571 +Step 14838 | loss:0.7375765442848206 lr:6.434154631454708e-05 tokens_per_second_per_gpu:4826.225792423552 mfu:22.766130501892423 +Step 14839 | loss:0.7062851786613464 lr:6.43373922416192e-05 tokens_per_second_per_gpu:4740.847858441295 mfu:22.36338822032048 +Step 14840 | loss:0.8757847547531128 lr:6.433323806085797e-05 tokens_per_second_per_gpu:3712.7565189803486 mfu:17.51370596161293 +Step 14841 | loss:0.7051289677619934 lr:6.432908377229462e-05 tokens_per_second_per_gpu:4945.066994785199 mfu:23.326724729832275 +Step 14842 | loss:0.746679961681366 lr:6.432492937596042e-05 tokens_per_second_per_gpu:4768.314958875616 mfu:22.492955219439754 +Step 14843 | loss:0.7165010571479797 lr:6.432077487188659e-05 tokens_per_second_per_gpu:4282.168030885373 mfu:20.199717214890818 +Step 14844 | loss:0.8211251497268677 lr:6.431662026010439e-05 tokens_per_second_per_gpu:5356.569906243958 mfu:25.26785417281972 +Step 14845 | loss:0.8114813566207886 lr:6.431246554064506e-05 tokens_per_second_per_gpu:4751.549030283851 mfu:22.413867473710006 +Step 14846 | loss:0.7810661196708679 lr:6.430831071353984e-05 tokens_per_second_per_gpu:5297.568724503288 mfu:24.989535531909336 +Step 14847 | loss:0.8220496773719788 lr:6.430415577882e-05 tokens_per_second_per_gpu:5316.960490954863 mfu:25.081009802837947 +Step 14848 | loss:0.8616825342178345 lr:6.430000073651678e-05 tokens_per_second_per_gpu:5170.49900992316 mfu:24.390126007905963 +Step 14849 | loss:0.8245160579681396 lr:6.429584558666143e-05 tokens_per_second_per_gpu:5346.617613298798 mfu:25.22090750896113 +Step 14850 | loss:0.7367263436317444 lr:6.429169032928519e-05 tokens_per_second_per_gpu:5037.452000663624 mfu:23.762520565068186 +Step 14851 | loss:0.8506675362586975 lr:6.428753496441935e-05 tokens_per_second_per_gpu:4133.6715345877765 mfu:19.499233905740212 +Step 14852 | loss:0.8728896379470825 lr:6.42833794920951e-05 tokens_per_second_per_gpu:4804.581694424996 mfu:22.664031598769323 +Step 14853 | loss:0.7393991947174072 lr:6.427922391234377e-05 tokens_per_second_per_gpu:4291.407505933516 mfu:20.24330139510991 +Step 14854 | loss:0.781253457069397 lr:6.427506822519656e-05 tokens_per_second_per_gpu:4065.265674528042 mfu:19.176551792595216 +Step 14855 | loss:0.7653231620788574 lr:6.427091243068471e-05 tokens_per_second_per_gpu:3951.710458981848 mfu:18.640892466346212 +Step 14856 | loss:0.9530432820320129 lr:6.426675652883953e-05 tokens_per_second_per_gpu:3894.5605681192924 mfu:18.371306680370687 +Step 14857 | loss:0.8343222737312317 lr:6.426260051969226e-05 tokens_per_second_per_gpu:4896.093617701452 mfu:23.095708954408494 +Step 14858 | loss:0.6957570314407349 lr:6.425844440327416e-05 tokens_per_second_per_gpu:3195.476197350601 mfu:15.073606427361755 +Step 14859 | loss:0.800676703453064 lr:6.425428817961645e-05 tokens_per_second_per_gpu:4985.386521593986 mfu:23.51691881700996 +Step 14860 | loss:0.7335106134414673 lr:6.425013184875045e-05 tokens_per_second_per_gpu:5281.6961595516805 mfu:24.914661934893772 +Step 14861 | loss:0.8143681883811951 lr:6.424597541070736e-05 tokens_per_second_per_gpu:5322.421749996812 mfu:25.10677149352569 +Step 14862 | loss:0.6867474317550659 lr:6.42418188655185e-05 tokens_per_second_per_gpu:4148.2885294646 mfu:19.56818476449068 +Step 14863 | loss:0.851873517036438 lr:6.423766221321509e-05 tokens_per_second_per_gpu:3497.1698554760374 mfu:16.49674688698537 +Step 14864 | loss:0.8268730044364929 lr:6.42335054538284e-05 tokens_per_second_per_gpu:4049.46214732311 mfu:19.10200385840977 +Step 14865 | loss:0.9408943057060242 lr:6.422934858738972e-05 tokens_per_second_per_gpu:5398.03925791218 mfu:25.463472180039712 +Step 14866 | loss:0.6673234701156616 lr:6.422519161393026e-05 tokens_per_second_per_gpu:4985.685808605459 mfu:23.518330604906705 +Step 14867 | loss:0.8673239350318909 lr:6.422103453348135e-05 tokens_per_second_per_gpu:3377.991763073669 mfu:15.934563491244074 +Step 14868 | loss:0.8130471706390381 lr:6.42168773460742e-05 tokens_per_second_per_gpu:5243.206021463606 mfu:24.733097386437585 +Step 14869 | loss:0.7808618545532227 lr:6.421272005174013e-05 tokens_per_second_per_gpu:5093.31455907065 mfu:24.02603378420941 +Step 14870 | loss:0.8639792203903198 lr:6.420856265051036e-05 tokens_per_second_per_gpu:4071.3637071461853 mfu:19.205317252886648 +Step 14871 | loss:0.7808681726455688 lr:6.420440514241618e-05 tokens_per_second_per_gpu:4367.772127406396 mfu:20.603526343745433 +Step 14872 | loss:0.7971009016036987 lr:6.420024752748885e-05 tokens_per_second_per_gpu:4989.069518797558 mfu:23.534292143203324 +Step 14873 | loss:0.7329208850860596 lr:6.419608980575965e-05 tokens_per_second_per_gpu:4663.445387790604 mfu:21.998267560037895 +Step 14874 | loss:0.6915885210037231 lr:6.419193197725986e-05 tokens_per_second_per_gpu:5379.375388136964 mfu:25.37543152192547 +Step 14875 | loss:0.7048168182373047 lr:6.418777404202072e-05 tokens_per_second_per_gpu:4257.894054394188 mfu:20.085212726214447 +Step 14876 | loss:0.8272718787193298 lr:6.418361600007353e-05 tokens_per_second_per_gpu:4404.531915635589 mfu:20.77692853668914 +Step 14877 | loss:0.7966418862342834 lr:6.417945785144955e-05 tokens_per_second_per_gpu:5042.855741001911 mfu:23.78801093021706 +Step 14878 | loss:0.8584891557693481 lr:6.417529959618007e-05 tokens_per_second_per_gpu:5341.8726411572525 mfu:25.19852466579373 +Step 14879 | loss:0.7905568480491638 lr:6.417114123429635e-05 tokens_per_second_per_gpu:3334.884173439146 mfu:15.731217636024821 +Step 14880 | loss:0.8384513854980469 lr:6.416698276582965e-05 tokens_per_second_per_gpu:3844.936210229689 mfu:18.137220117416906 +Step 14881 | loss:0.7875167727470398 lr:6.416282419081129e-05 tokens_per_second_per_gpu:5335.298113800624 mfu:25.16751149852248 +Step 14882 | loss:0.7138777375221252 lr:6.41586655092725e-05 tokens_per_second_per_gpu:5333.897715236827 mfu:25.160905579564417 +Step 14883 | loss:0.729132890701294 lr:6.41545067212446e-05 tokens_per_second_per_gpu:5304.608685480763 mfu:25.022744228981285 +Step 14884 | loss:0.6949734687805176 lr:6.415034782675884e-05 tokens_per_second_per_gpu:4901.160487635048 mfu:23.11961024438242 +Step 14885 | loss:0.782605767250061 lr:6.41461888258465e-05 tokens_per_second_per_gpu:4985.974601357133 mfu:23.519692889589475 +Step 14886 | loss:0.8002961874008179 lr:6.414202971853888e-05 tokens_per_second_per_gpu:5052.161814026611 mfu:23.83190926445435 +Step 14887 | loss:0.8269326090812683 lr:6.413787050486725e-05 tokens_per_second_per_gpu:5133.841955782434 mfu:24.217208429185792 +Step 14888 | loss:0.7427211403846741 lr:6.413371118486289e-05 tokens_per_second_per_gpu:5351.761505035707 mfu:25.24517212392256 +Step 14889 | loss:0.7957946062088013 lr:6.412955175855708e-05 tokens_per_second_per_gpu:5329.5851684322315 mfu:25.140562560490782 +Step 14890 | loss:0.8611185550689697 lr:6.412539222598112e-05 tokens_per_second_per_gpu:3938.83463223363 mfu:18.58015499473229 +Step 14891 | loss:0.7573223114013672 lr:6.412123258716627e-05 tokens_per_second_per_gpu:5326.398967009739 mfu:25.125532704758864 +Step 14892 | loss:0.7913491129875183 lr:6.411707284214384e-05 tokens_per_second_per_gpu:3417.305110654518 mfu:16.120011259331694 +Step 14893 | loss:0.9970658421516418 lr:6.411291299094508e-05 tokens_per_second_per_gpu:5121.74860999193 mfu:24.16016205375443 +Step 14894 | loss:0.7549517154693604 lr:6.410875303360133e-05 tokens_per_second_per_gpu:5354.2171049397175 mfu:25.256755607638322 +Step 14895 | loss:0.7341748476028442 lr:6.410459297014382e-05 tokens_per_second_per_gpu:4654.063613689946 mfu:21.954012130909373 +Step 14896 | loss:0.7255641222000122 lr:6.41004328006039e-05 tokens_per_second_per_gpu:4621.205087714214 mfu:21.799012857639216 +Step 14897 | loss:0.8558547496795654 lr:6.409627252501279e-05 tokens_per_second_per_gpu:4315.819025949815 mfu:20.35845470006159 +Step 14898 | loss:0.8389642238616943 lr:6.409211214340183e-05 tokens_per_second_per_gpu:5342.214525422001 mfu:25.200137392202024 +Step 14899 | loss:0.6703202128410339 lr:6.408795165580231e-05 tokens_per_second_per_gpu:4275.402524202852 mfu:20.1678031655549 +Step 14900 | loss:0.7377381920814514 lr:6.40837910622455e-05 tokens_per_second_per_gpu:5155.683236550228 mfu:24.32023747707424 +Step 14901 | loss:0.8573086261749268 lr:6.407963036276269e-05 tokens_per_second_per_gpu:4523.8831456177095 mfu:21.339928651934446 +Step 14902 | loss:0.6737481951713562 lr:6.407546955738519e-05 tokens_per_second_per_gpu:4931.728305858223 mfu:23.263803858348723 +Step 14903 | loss:0.8179638981819153 lr:6.407130864614428e-05 tokens_per_second_per_gpu:4960.128988296813 mfu:23.397774723067645 +Step 14904 | loss:0.8449999094009399 lr:6.406714762907128e-05 tokens_per_second_per_gpu:4015.1791540702766 mfu:18.94028512995435 +Step 14905 | loss:0.8795105218887329 lr:6.406298650619745e-05 tokens_per_second_per_gpu:4308.468312858249 mfu:20.3237801322477 +Step 14906 | loss:0.7880123257637024 lr:6.405882527755412e-05 tokens_per_second_per_gpu:4069.602206764073 mfu:19.19700795504127 +Step 14907 | loss:0.7806336879730225 lr:6.405466394317257e-05 tokens_per_second_per_gpu:4879.267209933684 mfu:23.01633591808673 +Step 14908 | loss:0.8806031942367554 lr:6.405050250308409e-05 tokens_per_second_per_gpu:4508.4537498439095 mfu:21.267145559543053 +Step 14909 | loss:0.7264794707298279 lr:6.404634095731998e-05 tokens_per_second_per_gpu:5371.43072467639 mfu:25.337955188882585 +Step 14910 | loss:0.7250763177871704 lr:6.404217930591156e-05 tokens_per_second_per_gpu:5329.661191702966 mfu:25.140921175229742 +Step 14911 | loss:0.9004528522491455 lr:6.403801754889012e-05 tokens_per_second_per_gpu:5348.61960812156 mfu:25.23035126011564 +Step 14912 | loss:0.5983407497406006 lr:6.403385568628695e-05 tokens_per_second_per_gpu:5350.817145546202 mfu:25.24071741908622 +Step 14913 | loss:0.7486258149147034 lr:6.402969371813336e-05 tokens_per_second_per_gpu:5043.427027236969 mfu:23.790705784859306 +Step 14914 | loss:0.6380443572998047 lr:6.402553164446067e-05 tokens_per_second_per_gpu:4390.724230487957 mfu:20.71179533001375 +Step 14915 | loss:0.7612570524215698 lr:6.402136946530014e-05 tokens_per_second_per_gpu:4687.159860559801 mfu:22.110132774196472 +Step 14916 | loss:0.6735589504241943 lr:6.401720718068311e-05 tokens_per_second_per_gpu:4809.896444397226 mfu:22.689102181178185 +Step 14917 | loss:0.807692289352417 lr:6.40130447906409e-05 tokens_per_second_per_gpu:5331.376639927006 mfu:25.14901323718819 +Step 14918 | loss:0.7564980387687683 lr:6.400888229520477e-05 tokens_per_second_per_gpu:4506.772456258473 mfu:21.259214610842285 +Step 14919 | loss:0.8385355472564697 lr:6.400471969440605e-05 tokens_per_second_per_gpu:5349.423657577394 mfu:25.23414409858379 +Step 14920 | loss:0.762219250202179 lr:6.400055698827604e-05 tokens_per_second_per_gpu:4610.149438808875 mfu:21.74686147546452 +Step 14921 | loss:0.7390294075012207 lr:6.399639417684606e-05 tokens_per_second_per_gpu:4341.9412108163115 mfu:20.481677502981174 +Step 14922 | loss:0.7579241394996643 lr:6.399223126014742e-05 tokens_per_second_per_gpu:5392.608719812915 mfu:25.43785540528014 +Step 14923 | loss:0.8357146978378296 lr:6.398806823821143e-05 tokens_per_second_per_gpu:5369.257451812418 mfu:25.327703489984643 +Step 14924 | loss:0.692579984664917 lr:6.398390511106937e-05 tokens_per_second_per_gpu:4084.493023575095 mfu:19.26725047857419 +Step 14925 | loss:0.9379348158836365 lr:6.39797418787526e-05 tokens_per_second_per_gpu:3818.989156138616 mfu:18.014823436245386 +Step 14926 | loss:0.7455561757087708 lr:6.39755785412924e-05 tokens_per_second_per_gpu:5296.183994533261 mfu:24.983003524381427 +Step 14927 | loss:0.965201735496521 lr:6.397141509872007e-05 tokens_per_second_per_gpu:4566.209222095565 mfu:21.5395879762538 +Step 14928 | loss:0.8893652558326721 lr:6.396725155106697e-05 tokens_per_second_per_gpu:3879.0398184064434 mfu:18.298092655862213 +Step 14929 | loss:0.8413869738578796 lr:6.396308789836438e-05 tokens_per_second_per_gpu:5337.306024207695 mfu:25.176983154497478 +Step 14930 | loss:0.7682716250419617 lr:6.395892414064362e-05 tokens_per_second_per_gpu:4635.467185024022 mfu:21.8662896040143 +Step 14931 | loss:0.7269125580787659 lr:6.395476027793602e-05 tokens_per_second_per_gpu:5121.662838001502 mfu:24.159757452641372 +Step 14932 | loss:0.827472984790802 lr:6.395059631027288e-05 tokens_per_second_per_gpu:5348.498322006965 mfu:25.229779132819548 +Step 14933 | loss:0.8466381430625916 lr:6.394643223768555e-05 tokens_per_second_per_gpu:4318.750778374013 mfu:20.372284276455076 +Step 14934 | loss:0.8028726577758789 lr:6.394226806020528e-05 tokens_per_second_per_gpu:4237.827285769257 mfu:19.990554354866877 +Step 14935 | loss:0.8220577239990234 lr:6.393810377786346e-05 tokens_per_second_per_gpu:5379.3226460469195 mfu:25.375182728488003 +Step 14936 | loss:0.917935311794281 lr:6.393393939069137e-05 tokens_per_second_per_gpu:5319.096622775093 mfu:25.09108630109564 +Step 14937 | loss:0.830734133720398 lr:6.392977489872036e-05 tokens_per_second_per_gpu:4624.761138149995 mfu:21.81578735426956 +Step 14938 | loss:0.7420396208763123 lr:6.392561030198172e-05 tokens_per_second_per_gpu:4850.434659404702 mfu:22.880327857901136 +Step 14939 | loss:0.8078424334526062 lr:6.39214456005068e-05 tokens_per_second_per_gpu:4222.7432432268215 mfu:19.91940035259117 +Step 14940 | loss:0.9091670513153076 lr:6.39172807943269e-05 tokens_per_second_per_gpu:4540.284629152074 mfu:21.417297248147694 +Step 14941 | loss:0.8560345768928528 lr:6.391311588347336e-05 tokens_per_second_per_gpu:4814.368505421204 mfu:22.710197656040574 +Step 14942 | loss:0.7680415511131287 lr:6.390895086797751e-05 tokens_per_second_per_gpu:4164.715092824336 mfu:19.645671666519263 +Step 14943 | loss:0.7797034978866577 lr:6.390478574787065e-05 tokens_per_second_per_gpu:3608.879767689495 mfu:17.023701602572647 +Step 14944 | loss:0.702002763748169 lr:6.390062052318413e-05 tokens_per_second_per_gpu:5316.408900391659 mfu:25.078407856040254 +Step 14945 | loss:0.877334713935852 lr:6.389645519394925e-05 tokens_per_second_per_gpu:4629.503224828695 mfu:21.83815658621641 +Step 14946 | loss:0.754798948764801 lr:6.38922897601974e-05 tokens_per_second_per_gpu:4097.896911248242 mfu:19.330478903667483 +Step 14947 | loss:0.8526699542999268 lr:6.388812422195982e-05 tokens_per_second_per_gpu:4806.82528396996 mfu:22.674614993448763 +Step 14948 | loss:0.9028587937355042 lr:6.388395857926792e-05 tokens_per_second_per_gpu:4301.655020401286 mfu:20.291640669260634 +Step 14949 | loss:0.6778775453567505 lr:6.387979283215297e-05 tokens_per_second_per_gpu:4743.39751559941 mfu:22.37541538815168 +Step 14950 | loss:0.8411937952041626 lr:6.387562698064634e-05 tokens_per_second_per_gpu:3901.5548675322225 mfu:18.404299984052013 +Step 14951 | loss:0.8082613348960876 lr:6.387146102477937e-05 tokens_per_second_per_gpu:5308.546699921531 mfu:25.04132051499294 +Step 14952 | loss:0.7008744478225708 lr:6.386729496458333e-05 tokens_per_second_per_gpu:5313.883193397194 mfu:25.066493665217315 +Step 14953 | loss:0.8536867499351501 lr:6.386312880008962e-05 tokens_per_second_per_gpu:3788.3831725778396 mfu:17.87044978986994 +Step 14954 | loss:0.7573935389518738 lr:6.385896253132955e-05 tokens_per_second_per_gpu:4523.962394626079 mfu:21.340302483028204 +Step 14955 | loss:0.8853975534439087 lr:6.385479615833445e-05 tokens_per_second_per_gpu:4483.507484295288 mfu:21.149469768678415 +Step 14956 | loss:0.8594241738319397 lr:6.385062968113564e-05 tokens_per_second_per_gpu:5356.437782258728 mfu:25.267230921438713 +Step 14957 | loss:0.7761573791503906 lr:6.384646309976451e-05 tokens_per_second_per_gpu:4271.12704919753 mfu:20.147635020483538 +Step 14958 | loss:0.7842645049095154 lr:6.384229641425234e-05 tokens_per_second_per_gpu:3672.6571706453524 mfu:17.32455049386244 +Step 14959 | loss:0.8283979892730713 lr:6.383812962463052e-05 tokens_per_second_per_gpu:4344.402185786208 mfu:20.493286341799802 +Step 14960 | loss:0.8582634329795837 lr:6.383396273093035e-05 tokens_per_second_per_gpu:5348.5595595382765 mfu:25.230068000702943 +Step 14961 | loss:0.7351390719413757 lr:6.382979573318316e-05 tokens_per_second_per_gpu:4740.10067257943 mfu:22.359863617127008 +Step 14962 | loss:0.6714087128639221 lr:6.382562863142034e-05 tokens_per_second_per_gpu:4363.218472153478 mfu:20.582045974983714 +Step 14963 | loss:0.7980250120162964 lr:6.382146142567318e-05 tokens_per_second_per_gpu:3852.737837442666 mfu:18.17402172407538 +Step 14964 | loss:0.8730633854866028 lr:6.381729411597305e-05 tokens_per_second_per_gpu:3569.1648959653457 mfu:16.836359776593994 +Step 14965 | loss:0.7060199975967407 lr:6.38131267023513e-05 tokens_per_second_per_gpu:5352.4301043551395 mfu:25.24832602098713 +Step 14966 | loss:0.7700615525245667 lr:6.380895918483924e-05 tokens_per_second_per_gpu:4303.185945589694 mfu:20.298862304577128 +Step 14967 | loss:0.7590855360031128 lr:6.380479156346826e-05 tokens_per_second_per_gpu:5367.471179800248 mfu:25.319277340133805 +Step 14968 | loss:0.7699321508407593 lr:6.380062383826966e-05 tokens_per_second_per_gpu:5175.029050576115 mfu:24.411494982570584 +Step 14969 | loss:0.7172346711158752 lr:6.379645600927484e-05 tokens_per_second_per_gpu:5349.978080684313 mfu:25.23675940697346 +Step 14970 | loss:0.7496905326843262 lr:6.379228807651508e-05 tokens_per_second_per_gpu:4481.968610440585 mfu:21.142210638146825 +Step 14971 | loss:0.850622832775116 lr:6.378812004002177e-05 tokens_per_second_per_gpu:4426.8180723588175 mfu:20.882056140363037 +Step 14972 | loss:0.7833065986633301 lr:6.378395189982625e-05 tokens_per_second_per_gpu:5376.3623902387435 mfu:25.361218696769118 +Step 14973 | loss:0.6192318201065063 lr:6.377978365595988e-05 tokens_per_second_per_gpu:4929.879584849017 mfu:23.255083125923967 +Step 14974 | loss:0.9268235564231873 lr:6.377561530845399e-05 tokens_per_second_per_gpu:5309.702043778499 mfu:25.046770469088244 +Step 14975 | loss:0.6786813735961914 lr:6.377144685733993e-05 tokens_per_second_per_gpu:5014.509013822671 mfu:23.654294581662203 +Step 14976 | loss:0.7394200563430786 lr:6.376727830264908e-05 tokens_per_second_per_gpu:5309.696099756289 mfu:25.046742430121327 +Step 14977 | loss:0.670231282711029 lr:6.376310964441276e-05 tokens_per_second_per_gpu:5147.104576698035 mfu:24.279770474881147 +Step 14978 | loss:0.6144916415214539 lr:6.375894088266234e-05 tokens_per_second_per_gpu:5316.02614258873 mfu:25.076602322178626 +Step 14979 | loss:0.8051074147224426 lr:6.375477201742916e-05 tokens_per_second_per_gpu:4751.64097677113 mfu:22.41430120097819 +Step 14980 | loss:0.8723610043525696 lr:6.375060304874461e-05 tokens_per_second_per_gpu:5053.345009294518 mfu:23.837490598406998 +Step 14981 | loss:0.7537726759910583 lr:6.374643397664e-05 tokens_per_second_per_gpu:4463.527728783235 mfu:21.055221853030254 +Step 14982 | loss:0.8631306886672974 lr:6.374226480114671e-05 tokens_per_second_per_gpu:5338.3557048443945 mfu:25.18193467715477 +Step 14983 | loss:0.7271748185157776 lr:6.37380955222961e-05 tokens_per_second_per_gpu:5338.5241209164615 mfu:25.182729124501456 +Step 14984 | loss:0.8318316340446472 lr:6.373392614011952e-05 tokens_per_second_per_gpu:5303.157118217063 mfu:25.0158969385355 +Step 14985 | loss:0.7478427886962891 lr:6.372975665464833e-05 tokens_per_second_per_gpu:3826.4349245177473 mfu:18.04994639606883 +Step 14986 | loss:0.7248412370681763 lr:6.372558706591387e-05 tokens_per_second_per_gpu:4376.701356493545 mfu:20.645647040833094 +Step 14987 | loss:0.7498524188995361 lr:6.372141737394755e-05 tokens_per_second_per_gpu:3891.8173931098236 mfu:18.358366655817104 +Step 14988 | loss:0.8570927977561951 lr:6.371724757878066e-05 tokens_per_second_per_gpu:2967.384929363072 mfu:13.997660999881333 +Step 14989 | loss:0.7889692783355713 lr:6.371307768044463e-05 tokens_per_second_per_gpu:3325.0872110202317 mfu:15.68500369276065 +Step 14990 | loss:0.8729695677757263 lr:6.370890767897077e-05 tokens_per_second_per_gpu:5322.6396788888105 mfu:25.107799501291755 +Step 14991 | loss:0.8581910729408264 lr:6.370473757439048e-05 tokens_per_second_per_gpu:4166.764886349023 mfu:19.65534089230605 +Step 14992 | loss:0.8330857157707214 lr:6.370056736673511e-05 tokens_per_second_per_gpu:5352.588970467509 mfu:25.24907541954453 +Step 14993 | loss:0.6401289105415344 lr:6.369639705603603e-05 tokens_per_second_per_gpu:4439.696283919703 mfu:20.942804861545202 +Step 14994 | loss:0.9153090715408325 lr:6.369222664232459e-05 tokens_per_second_per_gpu:5346.510993089284 mfu:25.22040456324157 +Step 14995 | loss:0.7579131722450256 lr:6.368805612563217e-05 tokens_per_second_per_gpu:5346.252623761058 mfu:25.219185791037972 +Step 14996 | loss:0.7952378988265991 lr:6.368388550599014e-05 tokens_per_second_per_gpu:5330.183501011522 mfu:25.14338499735336 +Step 14997 | loss:0.8962841629981995 lr:6.367971478342984e-05 tokens_per_second_per_gpu:4769.5862060137715 mfu:22.49895190908756 +Step 14998 | loss:0.8161711096763611 lr:6.367554395798269e-05 tokens_per_second_per_gpu:5347.764239377077 mfu:25.226316339806953 +Step 14999 | loss:0.8581723570823669 lr:6.367137302968e-05 tokens_per_second_per_gpu:4476.289001547472 mfu:21.11541895395681 +Step 15000 | loss:0.857249915599823 lr:6.366720199855318e-05 tokens_per_second_per_gpu:4821.9198492367905 mfu:22.745818633210806 +Step 15001 | loss:0.6880503296852112 lr:6.366303086463359e-05 tokens_per_second_per_gpu:4656.694563406394 mfu:21.96642276960817 +Step 15002 | loss:0.8664497137069702 lr:6.365885962795262e-05 tokens_per_second_per_gpu:5031.681262395858 mfu:23.735299008069653 +Step 15003 | loss:0.8518000841140747 lr:6.36546882885416e-05 tokens_per_second_per_gpu:4047.7409680785063 mfu:19.093884762249516 +Step 15004 | loss:0.7850006818771362 lr:6.365051684643193e-05 tokens_per_second_per_gpu:4633.526383089157 mfu:21.85713451015259 +Step 15005 | loss:0.9104964733123779 lr:6.364634530165499e-05 tokens_per_second_per_gpu:4697.347791564386 mfu:22.158190982985374 +Step 15006 | loss:0.7905614376068115 lr:6.364217365424213e-05 tokens_per_second_per_gpu:4198.897946689872 mfu:19.806918020399262 +Step 15007 | loss:0.7770308256149292 lr:6.363800190422476e-05 tokens_per_second_per_gpu:4557.620395399878 mfu:21.499073015325305 +Step 15008 | loss:0.7536109089851379 lr:6.363383005163423e-05 tokens_per_second_per_gpu:4303.72508647146 mfu:20.301405524103338 +Step 15009 | loss:0.6652107238769531 lr:6.362965809650192e-05 tokens_per_second_per_gpu:5202.205164205413 mfu:24.539689347283094 +Step 15010 | loss:0.766619086265564 lr:6.36254860388592e-05 tokens_per_second_per_gpu:5340.908264676074 mfu:25.193975537392422 +Step 15011 | loss:0.9194554686546326 lr:6.362131387873749e-05 tokens_per_second_per_gpu:4949.89434456644 mfu:23.34949616238119 +Step 15012 | loss:0.7199205160140991 lr:6.361714161616813e-05 tokens_per_second_per_gpu:5326.480506854677 mfu:25.125917342119568 +Step 15013 | loss:0.7231128215789795 lr:6.36129692511825e-05 tokens_per_second_per_gpu:4862.076365335735 mfu:22.93524377105767 +Step 15014 | loss:0.6710402965545654 lr:6.3608796783812e-05 tokens_per_second_per_gpu:5361.45105236549 mfu:25.290879371884394 +Step 15015 | loss:0.7645946145057678 lr:6.360462421408801e-05 tokens_per_second_per_gpu:4909.517136759168 mfu:23.15902998409226 +Step 15016 | loss:0.7105209827423096 lr:6.36004515420419e-05 tokens_per_second_per_gpu:4790.944394963885 mfu:22.599702130448044 +Step 15017 | loss:0.8606785535812378 lr:6.359627876770504e-05 tokens_per_second_per_gpu:4912.560087028922 mfu:23.17338409969554 +Step 15018 | loss:0.7130171060562134 lr:6.359210589110885e-05 tokens_per_second_per_gpu:5141.991146251116 mfu:24.255649551021836 +Step 15019 | loss:0.6529573202133179 lr:6.358793291228471e-05 tokens_per_second_per_gpu:4703.501542758124 mfu:22.187219277304454 +Step 15020 | loss:0.7045759558677673 lr:6.358375983126398e-05 tokens_per_second_per_gpu:4967.97472450667 mfu:23.43478439939017 +Step 15021 | loss:0.7468363642692566 lr:6.357958664807805e-05 tokens_per_second_per_gpu:3938.084453222069 mfu:18.576616272341024 +Step 15022 | loss:0.9000024199485779 lr:6.357541336275832e-05 tokens_per_second_per_gpu:5364.911280836218 mfu:25.30720185995667 +Step 15023 | loss:0.7939878702163696 lr:6.357123997533618e-05 tokens_per_second_per_gpu:5365.804630978926 mfu:25.311415945001077 +Step 15024 | loss:0.7870146632194519 lr:6.3567066485843e-05 tokens_per_second_per_gpu:5336.559187894311 mfu:25.17346020018383 +Step 15025 | loss:0.9096018671989441 lr:6.35628928943102e-05 tokens_per_second_per_gpu:5324.497739288537 mfu:25.116564289215052 +Step 15026 | loss:0.6654923558235168 lr:6.355871920076913e-05 tokens_per_second_per_gpu:5142.178686750773 mfu:24.256534211556424 +Step 15027 | loss:0.7344345450401306 lr:6.355454540525123e-05 tokens_per_second_per_gpu:4716.150809067689 mfu:22.24688802467405 +Step 15028 | loss:0.7784050703048706 lr:6.355037150778784e-05 tokens_per_second_per_gpu:4339.573387100761 mfu:20.470508074522535 +Step 15029 | loss:0.8232693672180176 lr:6.354619750841039e-05 tokens_per_second_per_gpu:4605.289109981806 mfu:21.72393447512963 +Step 15030 | loss:0.8293592929840088 lr:6.354202340715026e-05 tokens_per_second_per_gpu:5165.060222387895 mfu:24.364470319149273 +Step 15031 | loss:0.9244430661201477 lr:6.353784920403885e-05 tokens_per_second_per_gpu:3940.42085823918 mfu:18.587637493438535 +Step 15032 | loss:0.7911885976791382 lr:6.353367489910754e-05 tokens_per_second_per_gpu:5344.223229134141 mfu:25.209612790332375 +Step 15033 | loss:0.7444673180580139 lr:6.352950049238772e-05 tokens_per_second_per_gpu:4223.697243684573 mfu:19.92390053551964 +Step 15034 | loss:0.6667052507400513 lr:6.352532598391083e-05 tokens_per_second_per_gpu:3926.697249879331 mfu:18.522900891316098 +Step 15035 | loss:0.7794862389564514 lr:6.352115137370821e-05 tokens_per_second_per_gpu:4968.803656682809 mfu:23.43869461389543 +Step 15036 | loss:0.7467923760414124 lr:6.35169766618113e-05 tokens_per_second_per_gpu:5328.371056688163 mfu:25.134835388245186 +Step 15037 | loss:0.802791953086853 lr:6.351280184825148e-05 tokens_per_second_per_gpu:5157.403452986544 mfu:24.328352031503535 +Step 15038 | loss:0.9314024448394775 lr:6.350862693306016e-05 tokens_per_second_per_gpu:5329.915626448277 mfu:25.14212138733442 +Step 15039 | loss:0.8945967555046082 lr:6.350445191626872e-05 tokens_per_second_per_gpu:4956.1572284370095 mfu:23.37903925415684 +Step 15040 | loss:0.7874159812927246 lr:6.350027679790858e-05 tokens_per_second_per_gpu:4353.1638359737535 mfu:20.53461654062613 +Step 15041 | loss:0.750680685043335 lr:6.349610157801114e-05 tokens_per_second_per_gpu:4454.639472714059 mfu:21.013294432659087 +Step 15042 | loss:0.7899667620658875 lr:6.34919262566078e-05 tokens_per_second_per_gpu:4607.0044341517705 mfu:21.732025951903836 +Step 15043 | loss:0.7337198257446289 lr:6.348775083372996e-05 tokens_per_second_per_gpu:5183.367233671166 mfu:24.45082761487068 +Step 15044 | loss:0.7905373573303223 lr:6.3483575309409e-05 tokens_per_second_per_gpu:4018.490154202483 mfu:18.955903682492842 +Step 15045 | loss:0.8069664835929871 lr:6.347939968367638e-05 tokens_per_second_per_gpu:3813.39225633359 mfu:17.9884219049355 +Step 15046 | loss:0.8036318421363831 lr:6.347522395656346e-05 tokens_per_second_per_gpu:4233.567391357462 mfu:19.970459706113505 +Step 15047 | loss:0.8576605916023254 lr:6.347104812810166e-05 tokens_per_second_per_gpu:4007.037485240628 mfu:18.901879488972956 +Step 15048 | loss:0.5643019080162048 lr:6.34668721983224e-05 tokens_per_second_per_gpu:4884.1364245198365 mfu:23.039304834063287 +Step 15049 | loss:0.7982847094535828 lr:6.346269616725707e-05 tokens_per_second_per_gpu:4111.020223614656 mfu:19.392383807167835 +Step 15050 | loss:0.7864528298377991 lr:6.345852003493708e-05 tokens_per_second_per_gpu:5199.982042521319 mfu:24.529202502995552 +Step 15051 | loss:0.7498315572738647 lr:6.345434380139384e-05 tokens_per_second_per_gpu:5310.7585499236775 mfu:25.051754188833584 +Step 15052 | loss:0.796498715877533 lr:6.34501674666588e-05 tokens_per_second_per_gpu:3776.554575788494 mfu:17.814652280647877 +Step 15053 | loss:0.7028430104255676 lr:6.344599103076329e-05 tokens_per_second_per_gpu:4324.239875317051 mfu:20.3981772832719 +Step 15054 | loss:0.7398856282234192 lr:6.344181449373879e-05 tokens_per_second_per_gpu:3836.365702015516 mfu:18.096791567891163 +Step 15055 | loss:0.7535179257392883 lr:6.343763785561667e-05 tokens_per_second_per_gpu:3782.1906088320834 mfu:17.841238410120884 +Step 15056 | loss:0.7243568897247314 lr:6.343346111642837e-05 tokens_per_second_per_gpu:5338.669084853305 mfu:25.183412944124946 +Step 15057 | loss:0.7126162052154541 lr:6.34292842762053e-05 tokens_per_second_per_gpu:4208.7819979284495 mfu:19.853542776484673 +Step 15058 | loss:0.6641588807106018 lr:6.342510733497886e-05 tokens_per_second_per_gpu:5134.751362306583 mfu:24.22149825492096 +Step 15059 | loss:0.6783604025840759 lr:6.342093029278049e-05 tokens_per_second_per_gpu:5235.353620855659 mfu:24.696056273012207 +Step 15060 | loss:0.8102135062217712 lr:6.341675314964157e-05 tokens_per_second_per_gpu:3102.8416062733963 mfu:14.636633256159573 +Step 15061 | loss:0.858488142490387 lr:6.341257590559355e-05 tokens_per_second_per_gpu:3247.9157115073112 mfu:15.32097256274267 +Step 15062 | loss:0.7209509611129761 lr:6.340839856066783e-05 tokens_per_second_per_gpu:3908.2591461220704 mfu:18.435925210028206 +Step 15063 | loss:0.6980695724487305 lr:6.340422111489583e-05 tokens_per_second_per_gpu:4867.781218268341 mfu:22.962154535689052 +Step 15064 | loss:0.7385332584381104 lr:6.340004356830898e-05 tokens_per_second_per_gpu:5312.226535742824 mfu:25.05867892840671 +Step 15065 | loss:0.694671094417572 lr:6.339586592093868e-05 tokens_per_second_per_gpu:5146.60873701732 mfu:24.277431514508244 +Step 15066 | loss:0.7122064232826233 lr:6.339168817281638e-05 tokens_per_second_per_gpu:3800.5665729267616 mfu:17.927920968018253 +Step 15067 | loss:0.6438330411911011 lr:6.338751032397349e-05 tokens_per_second_per_gpu:5295.278525675585 mfu:24.978732273290852 +Step 15068 | loss:0.8135294318199158 lr:6.338333237444142e-05 tokens_per_second_per_gpu:5210.707722639274 mfu:24.579797366101673 +Step 15069 | loss:0.8410249352455139 lr:6.337915432425158e-05 tokens_per_second_per_gpu:5368.89872070528 mfu:25.32601129414643 +Step 15070 | loss:0.8683427572250366 lr:6.337497617343544e-05 tokens_per_second_per_gpu:5286.553664646653 mfu:24.937575615202583 +Step 15071 | loss:0.7792066335678101 lr:6.337079792202438e-05 tokens_per_second_per_gpu:4770.4452762332685 mfu:22.50300428988542 +Step 15072 | loss:0.8241268992424011 lr:6.336661957004985e-05 tokens_per_second_per_gpu:4287.064460880492 mfu:20.2228145105949 +Step 15073 | loss:0.7637523412704468 lr:6.336244111754327e-05 tokens_per_second_per_gpu:5335.492384303285 mfu:25.168427905629496 +Step 15074 | loss:0.8061885833740234 lr:6.335826256453606e-05 tokens_per_second_per_gpu:5317.18315685664 mfu:25.082060155887348 +Step 15075 | loss:0.7098340392112732 lr:6.335408391105966e-05 tokens_per_second_per_gpu:5016.56040217359 mfu:23.663971330516432 +Step 15076 | loss:0.8421914577484131 lr:6.334990515714548e-05 tokens_per_second_per_gpu:4971.037450135114 mfu:23.44923179873442 +Step 15077 | loss:0.8218561410903931 lr:6.334572630282497e-05 tokens_per_second_per_gpu:4455.825508694012 mfu:21.018889166735132 +Step 15078 | loss:0.814343273639679 lr:6.334154734812954e-05 tokens_per_second_per_gpu:3659.5483128400397 mfu:17.262713774993006 +Step 15079 | loss:0.7642263174057007 lr:6.333736829309064e-05 tokens_per_second_per_gpu:5339.563401270724 mfu:25.187631587252024 +Step 15080 | loss:0.851494312286377 lr:6.333318913773966e-05 tokens_per_second_per_gpu:5337.3940503731465 mfu:25.177398388938723 +Step 15081 | loss:0.8204691410064697 lr:6.332900988210811e-05 tokens_per_second_per_gpu:5326.300497117757 mfu:25.12506820547769 +Step 15082 | loss:0.7598004937171936 lr:6.332483052622735e-05 tokens_per_second_per_gpu:3823.9054658904474 mfu:18.03801450815371 +Step 15083 | loss:0.7789567708969116 lr:6.332065107012885e-05 tokens_per_second_per_gpu:4893.118676372551 mfu:23.08167564858274 +Step 15084 | loss:0.8287897109985352 lr:6.331647151384403e-05 tokens_per_second_per_gpu:5348.338280757203 mfu:25.229024190938567 +Step 15085 | loss:0.8735779523849487 lr:6.331229185740433e-05 tokens_per_second_per_gpu:5345.820157668241 mfu:25.217145774691833 +Step 15086 | loss:0.7311908006668091 lr:6.330811210084118e-05 tokens_per_second_per_gpu:3299.4612384390293 mfu:15.564121607853345 +Step 15087 | loss:0.8167628049850464 lr:6.330393224418601e-05 tokens_per_second_per_gpu:4471.429022691664 mfu:21.092493604496063 +Step 15088 | loss:0.7223342061042786 lr:6.329975228747029e-05 tokens_per_second_per_gpu:5352.972012850822 mfu:25.250882295820386 +Step 15089 | loss:0.7173174619674683 lr:6.329557223072542e-05 tokens_per_second_per_gpu:5352.225358619684 mfu:25.247360200419763 +Step 15090 | loss:0.8016771674156189 lr:6.329139207398287e-05 tokens_per_second_per_gpu:4919.156557455371 mfu:23.20450077617101 +Step 15091 | loss:0.8284558653831482 lr:6.328721181727405e-05 tokens_per_second_per_gpu:5311.610359032102 mfu:25.055772317731645 +Step 15092 | loss:0.6670644283294678 lr:6.328303146063043e-05 tokens_per_second_per_gpu:4764.4260790811295 mfu:22.47461070993785 +Step 15093 | loss:0.8320976495742798 lr:6.327885100408342e-05 tokens_per_second_per_gpu:5213.5594533194235 mfu:24.593249466275363 +Step 15094 | loss:0.7006896138191223 lr:6.327467044766447e-05 tokens_per_second_per_gpu:4647.23200752518 mfu:21.92178627903797 +Step 15095 | loss:0.8414946794509888 lr:6.327048979140505e-05 tokens_per_second_per_gpu:4076.6137275550136 mfu:19.23008249490094 +Step 15096 | loss:0.7665444612503052 lr:6.326630903533657e-05 tokens_per_second_per_gpu:3282.4038339251515 mfu:15.483658920468073 +Step 15097 | loss:0.8345289826393127 lr:6.32621281794905e-05 tokens_per_second_per_gpu:4891.468054858664 mfu:23.073889385276658 +Step 15098 | loss:0.8659443855285645 lr:6.325794722389826e-05 tokens_per_second_per_gpu:3993.038003729069 mfu:18.835841546125085 +Step 15099 | loss:0.9149152040481567 lr:6.325376616859131e-05 tokens_per_second_per_gpu:4962.676142574341 mfu:23.409790084384074 +Step 15100 | loss:0.6994699239730835 lr:6.324958501360108e-05 tokens_per_second_per_gpu:5111.0095982888415 mfu:24.109504303286627 +Step 15101 | loss:0.8047593235969543 lr:6.324540375895906e-05 tokens_per_second_per_gpu:5349.454341256672 mfu:25.23428883873412 +Step 15102 | loss:0.7378642559051514 lr:6.324122240469666e-05 tokens_per_second_per_gpu:4884.175577898998 mfu:23.039489527233048 +Step 15103 | loss:0.7774614691734314 lr:6.323704095084531e-05 tokens_per_second_per_gpu:4209.40595026914 mfu:19.856486066132955 +Step 15104 | loss:0.883405327796936 lr:6.323285939743652e-05 tokens_per_second_per_gpu:5361.510495329085 mfu:25.291159774485685 +Step 15105 | loss:0.8856914043426514 lr:6.322867774450167e-05 tokens_per_second_per_gpu:4770.455166158022 mfu:22.503050942347876 +Step 15106 | loss:0.7866067886352539 lr:6.322449599207229e-05 tokens_per_second_per_gpu:4623.372124212698 mfu:21.809235138537115 +Step 15107 | loss:0.8488138914108276 lr:6.322031414017977e-05 tokens_per_second_per_gpu:4075.6986919368455 mfu:19.225766115768867 +Step 15108 | loss:0.8418217301368713 lr:6.321613218885556e-05 tokens_per_second_per_gpu:4202.661342472292 mfu:19.82467060040647 +Step 15109 | loss:0.8043267130851746 lr:6.321195013813117e-05 tokens_per_second_per_gpu:5348.735869796791 mfu:25.23089968627399 +Step 15110 | loss:0.7469125390052795 lr:6.320776798803798e-05 tokens_per_second_per_gpu:4719.219174585733 mfu:22.261362028339803 +Step 15111 | loss:0.7763589024543762 lr:6.32035857386075e-05 tokens_per_second_per_gpu:5312.715724411731 mfu:25.060986514822137 +Step 15112 | loss:0.8409082293510437 lr:6.319940338987116e-05 tokens_per_second_per_gpu:2980.155885556277 mfu:14.057903779194493 +Step 15113 | loss:0.754907488822937 lr:6.319522094186045e-05 tokens_per_second_per_gpu:4452.686997278828 mfu:21.004084272904446 +Step 15114 | loss:0.7469162940979004 lr:6.319103839460676e-05 tokens_per_second_per_gpu:4461.75708516853 mfu:21.046869424996736 +Step 15115 | loss:0.8622550964355469 lr:6.31868557481416e-05 tokens_per_second_per_gpu:5354.793810416435 mfu:25.2594760257681 +Step 15116 | loss:0.7818483710289001 lr:6.318267300249642e-05 tokens_per_second_per_gpu:5350.082011405974 mfu:25.237249665919947 +Step 15117 | loss:0.7784382104873657 lr:6.317849015770269e-05 tokens_per_second_per_gpu:4129.130902345191 mfu:19.47781496874019 +Step 15118 | loss:0.7787992358207703 lr:6.317430721379183e-05 tokens_per_second_per_gpu:4065.1740123622426 mfu:19.176119406519962 +Step 15119 | loss:0.8497560620307922 lr:6.317012417079533e-05 tokens_per_second_per_gpu:4462.697738107697 mfu:21.051306645402725 +Step 15120 | loss:0.7423574328422546 lr:6.316594102874464e-05 tokens_per_second_per_gpu:4004.3271739480433 mfu:18.88909448817904 +Step 15121 | loss:0.7465425729751587 lr:6.316175778767125e-05 tokens_per_second_per_gpu:5338.757254793802 mfu:25.18382885677761 +Step 15122 | loss:0.7861164212226868 lr:6.31575744476066e-05 tokens_per_second_per_gpu:5351.753177458731 mfu:25.245132841321134 +Step 15123 | loss:0.614193320274353 lr:6.315339100858212e-05 tokens_per_second_per_gpu:5056.555277227863 mfu:23.852633979976776 +Step 15124 | loss:0.9334744811058044 lr:6.314920747062933e-05 tokens_per_second_per_gpu:4181.894252508609 mfu:19.72670869381686 +Step 15125 | loss:0.7487327456474304 lr:6.314502383377967e-05 tokens_per_second_per_gpu:4744.9992524223435 mfu:22.382971054029937 +Step 15126 | loss:0.8136926889419556 lr:6.314084009806461e-05 tokens_per_second_per_gpu:3081.809793991293 mfu:14.537422609227695 +Step 15127 | loss:0.872808575630188 lr:6.313665626351562e-05 tokens_per_second_per_gpu:3251.5987546294155 mfu:15.33834610554159 +Step 15128 | loss:0.8789401054382324 lr:6.313247233016415e-05 tokens_per_second_per_gpu:3597.7273165560337 mfu:16.971093587771737 +Step 15129 | loss:0.8519163131713867 lr:6.31282882980417e-05 tokens_per_second_per_gpu:3364.043308574846 mfu:15.868766251521384 +Step 15130 | loss:0.8801358938217163 lr:6.312410416717968e-05 tokens_per_second_per_gpu:5049.830383073173 mfu:23.820911506863848 +Step 15131 | loss:0.7605176568031311 lr:6.311991993760964e-05 tokens_per_second_per_gpu:4735.265764460917 mfu:22.3370565306116 +Step 15132 | loss:0.748350977897644 lr:6.311573560936298e-05 tokens_per_second_per_gpu:5325.043666058929 mfu:25.119139518935707 +Step 15133 | loss:0.7224109172821045 lr:6.31115511824712e-05 tokens_per_second_per_gpu:3701.860906566908 mfu:17.462309498875562 +Step 15134 | loss:0.7064719796180725 lr:6.310736665696578e-05 tokens_per_second_per_gpu:5148.81065341553 mfu:24.287818329844278 +Step 15135 | loss:0.8124579191207886 lr:6.31031820328782e-05 tokens_per_second_per_gpu:4357.815600381474 mfu:20.55655970701019 +Step 15136 | loss:0.7621766328811646 lr:6.30989973102399e-05 tokens_per_second_per_gpu:4115.874681289668 mfu:19.415283112276754 +Step 15137 | loss:0.8062677979469299 lr:6.309481248908235e-05 tokens_per_second_per_gpu:5144.650839492678 mfu:24.26819577783274 +Step 15138 | loss:0.7014310956001282 lr:6.309062756943707e-05 tokens_per_second_per_gpu:4114.291885517419 mfu:19.407816794566248 +Step 15139 | loss:0.8259396553039551 lr:6.308644255133552e-05 tokens_per_second_per_gpu:3458.1029889976076 mfu:16.312461812311874 +Step 15140 | loss:0.763377845287323 lr:6.308225743480914e-05 tokens_per_second_per_gpu:5121.904866930851 mfu:24.160899144395508 +Step 15141 | loss:0.8647820949554443 lr:6.307807221988943e-05 tokens_per_second_per_gpu:3409.4190196672976 mfu:16.08281122263911 +Step 15142 | loss:0.8217929601669312 lr:6.30738869066079e-05 tokens_per_second_per_gpu:5180.411958947129 mfu:24.43688707977553 +Step 15143 | loss:0.8084078431129456 lr:6.306970149499596e-05 tokens_per_second_per_gpu:4326.01618764183 mfu:20.406556451578012 +Step 15144 | loss:0.8323701024055481 lr:6.306551598508515e-05 tokens_per_second_per_gpu:4657.255803160149 mfu:21.969070233284008 +Step 15145 | loss:0.8983275890350342 lr:6.306133037690693e-05 tokens_per_second_per_gpu:3877.070474004922 mfu:18.288802922315433 +Step 15146 | loss:0.7739138007164001 lr:6.305714467049277e-05 tokens_per_second_per_gpu:5369.584487793413 mfu:25.329246174505087 +Step 15147 | loss:0.708453357219696 lr:6.305295886587416e-05 tokens_per_second_per_gpu:5389.443827501256 mfu:25.422926068259734 +Step 15148 | loss:0.7805679440498352 lr:6.304877296308258e-05 tokens_per_second_per_gpu:4375.263927592563 mfu:20.638866443456497 +Step 15149 | loss:0.8370126485824585 lr:6.30445869621495e-05 tokens_per_second_per_gpu:4212.345356783993 mfu:19.87035174817798 +Step 15150 | loss:0.8372477293014526 lr:6.304040086310644e-05 tokens_per_second_per_gpu:3568.0209637938633 mfu:16.830963653366155 +Step 15151 | loss:0.8035210371017456 lr:6.303621466598486e-05 tokens_per_second_per_gpu:5339.450035328684 mfu:25.187096820761198 +Step 15152 | loss:0.7123772501945496 lr:6.303202837081622e-05 tokens_per_second_per_gpu:3798.821578089879 mfu:17.919689529645858 +Step 15153 | loss:0.719207227230072 lr:6.302784197763206e-05 tokens_per_second_per_gpu:4111.0364050783155 mfu:19.392460137892755 +Step 15154 | loss:0.807454526424408 lr:6.302365548646382e-05 tokens_per_second_per_gpu:4713.740375634385 mfu:22.235517598906803 +Step 15155 | loss:0.7795053720474243 lr:6.301946889734302e-05 tokens_per_second_per_gpu:5359.492954365101 mfu:25.28164268953059 +Step 15156 | loss:0.7021345496177673 lr:6.301528221030113e-05 tokens_per_second_per_gpu:4596.338941737807 mfu:21.681715004466515 +Step 15157 | loss:0.8247815370559692 lr:6.301109542536964e-05 tokens_per_second_per_gpu:3614.813904545594 mfu:17.051693938591015 +Step 15158 | loss:0.7868241667747498 lr:6.300690854258004e-05 tokens_per_second_per_gpu:5323.314126655563 mfu:25.11098098648018 +Step 15159 | loss:0.8103939890861511 lr:6.30027215619638e-05 tokens_per_second_per_gpu:5305.418037252016 mfu:25.02656207937594 +Step 15160 | loss:0.8618976473808289 lr:6.299853448355248e-05 tokens_per_second_per_gpu:4782.526573081495 mfu:22.559993828400298 +Step 15161 | loss:0.7841900587081909 lr:6.299434730737748e-05 tokens_per_second_per_gpu:5314.040753842888 mfu:25.067236904722236 +Step 15162 | loss:0.8540558218955994 lr:6.299016003347035e-05 tokens_per_second_per_gpu:5329.071782346991 mfu:25.138140830734006 +Step 15163 | loss:0.6965039968490601 lr:6.298597266186259e-05 tokens_per_second_per_gpu:3796.316280986558 mfu:17.90787161576173 +Step 15164 | loss:0.7511711716651917 lr:6.298178519258565e-05 tokens_per_second_per_gpu:4841.561137704803 mfu:22.838469942063792 +Step 15165 | loss:0.8207249641418457 lr:6.297759762567105e-05 tokens_per_second_per_gpu:5373.0468600271415 mfu:25.3455787750732 +Step 15166 | loss:0.8021116256713867 lr:6.297340996115027e-05 tokens_per_second_per_gpu:4970.976356038683 mfu:23.448943607457824 +Step 15167 | loss:0.8360006213188171 lr:6.296922219905484e-05 tokens_per_second_per_gpu:5188.438493060108 mfu:24.474749610654392 +Step 15168 | loss:0.7087966203689575 lr:6.296503433941622e-05 tokens_per_second_per_gpu:4864.390197000242 mfu:22.946158509799538 +Step 15169 | loss:0.6954740881919861 lr:6.296084638226593e-05 tokens_per_second_per_gpu:5336.957051356345 mfu:25.17533699001675 +Step 15170 | loss:0.7629055976867676 lr:6.295665832763545e-05 tokens_per_second_per_gpu:5322.722464506169 mfu:25.108190015173513 +Step 15171 | loss:0.8626165986061096 lr:6.295247017555631e-05 tokens_per_second_per_gpu:4439.44685991884 mfu:20.941628285977007 +Step 15172 | loss:0.7496641874313354 lr:6.294828192605997e-05 tokens_per_second_per_gpu:5145.0267172998865 mfu:24.26996885758071 +Step 15173 | loss:0.7507996559143066 lr:6.294409357917795e-05 tokens_per_second_per_gpu:4979.155628397199 mfu:23.487526630700707 +Step 15174 | loss:0.731791079044342 lr:6.293990513494177e-05 tokens_per_second_per_gpu:5190.442501884506 mfu:24.484202862198238 +Step 15175 | loss:0.7609807252883911 lr:6.293571659338289e-05 tokens_per_second_per_gpu:4465.211821188191 mfu:21.063166004240223 +Step 15176 | loss:0.7989628314971924 lr:6.293152795453286e-05 tokens_per_second_per_gpu:4447.022574144861 mfu:20.97736422253128 +Step 15177 | loss:0.9207977056503296 lr:6.292733921842312e-05 tokens_per_second_per_gpu:4493.599266004505 mfu:21.1970744248363 +Step 15178 | loss:0.9042991399765015 lr:6.292315038508525e-05 tokens_per_second_per_gpu:4538.9146815669155 mfu:21.41083497165134 +Step 15179 | loss:0.9225261807441711 lr:6.29189614545507e-05 tokens_per_second_per_gpu:3787.803710506298 mfu:17.867716368412054 +Step 15180 | loss:0.7940137386322021 lr:6.291477242685098e-05 tokens_per_second_per_gpu:4065.01653084403 mfu:19.175376539329434 +Step 15181 | loss:0.834132730960846 lr:6.291058330201763e-05 tokens_per_second_per_gpu:4026.3443089549955 mfu:18.992953070517427 +Step 15182 | loss:0.7233591675758362 lr:6.290639408008214e-05 tokens_per_second_per_gpu:3611.1704083431887 mfu:17.034506945359716 +Step 15183 | loss:0.877115786075592 lr:6.2902204761076e-05 tokens_per_second_per_gpu:5362.009254418599 mfu:25.293512506208 +Step 15184 | loss:0.6736429333686829 lr:6.289801534503073e-05 tokens_per_second_per_gpu:5216.095043082159 mfu:24.60521027580241 +Step 15185 | loss:0.7201879620552063 lr:6.289382583197784e-05 tokens_per_second_per_gpu:5301.702164380982 mfu:25.009033673049984 +Step 15186 | loss:0.7414942979812622 lr:6.288963622194885e-05 tokens_per_second_per_gpu:4152.876981699081 mfu:19.589829276551402 +Step 15187 | loss:0.7263291478157043 lr:6.288544651497525e-05 tokens_per_second_per_gpu:5323.785211133327 mfu:25.11320317233709 +Step 15188 | loss:0.7261125445365906 lr:6.288125671108857e-05 tokens_per_second_per_gpu:5313.765201334475 mfu:25.065937076525948 +Step 15189 | loss:0.7984328269958496 lr:6.287706681032031e-05 tokens_per_second_per_gpu:3296.8936070649365 mfu:15.552009652578663 +Step 15190 | loss:0.8848279118537903 lr:6.2872876812702e-05 tokens_per_second_per_gpu:4339.945078536746 mfu:20.472261406443177 +Step 15191 | loss:0.7516006827354431 lr:6.286868671826512e-05 tokens_per_second_per_gpu:5168.5684475396265 mfu:24.38101921575493 +Step 15192 | loss:0.7938023209571838 lr:6.286449652704123e-05 tokens_per_second_per_gpu:5168.789949085105 mfu:24.382064076337013 +Step 15193 | loss:0.8203286528587341 lr:6.286030623906181e-05 tokens_per_second_per_gpu:3670.107669601689 mfu:17.312524062449747 +Step 15194 | loss:0.8408672213554382 lr:6.285611585435838e-05 tokens_per_second_per_gpu:4276.619493579176 mfu:20.17354381773941 +Step 15195 | loss:0.8543353080749512 lr:6.285192537296245e-05 tokens_per_second_per_gpu:4088.1840967888397 mfu:19.284661900685453 +Step 15196 | loss:0.7478938102722168 lr:6.284773479490556e-05 tokens_per_second_per_gpu:5338.867446438229 mfu:25.184348649565006 +Step 15197 | loss:0.8123207688331604 lr:6.284354412021923e-05 tokens_per_second_per_gpu:4627.326382608814 mfu:21.82788805005765 +Step 15198 | loss:0.6972659826278687 lr:6.283935334893496e-05 tokens_per_second_per_gpu:4740.524393065741 mfu:22.36186237895511 +Step 15199 | loss:0.87205970287323 lr:6.283516248108427e-05 tokens_per_second_per_gpu:4053.0207886980265 mfu:19.118790576942082 +Step 15200 | loss:0.8058000802993774 lr:6.283097151669869e-05 tokens_per_second_per_gpu:5126.548011035514 mfu:24.18280164733927 +Step 15201 | loss:0.8096634745597839 lr:6.282678045580975e-05 tokens_per_second_per_gpu:5255.92715104433 mfu:24.793105125118377 +Step 15202 | loss:0.7296634912490845 lr:6.282258929844892e-05 tokens_per_second_per_gpu:5157.697295440698 mfu:24.32973813649447 +Step 15203 | loss:0.7626543045043945 lr:6.281839804464777e-05 tokens_per_second_per_gpu:5333.787929167734 mfu:25.16038769994507 +Step 15204 | loss:0.8558178544044495 lr:6.281420669443784e-05 tokens_per_second_per_gpu:5345.115563280345 mfu:25.213822082747928 +Step 15205 | loss:0.7823333740234375 lr:6.28100152478506e-05 tokens_per_second_per_gpu:3834.215388048449 mfu:18.086648170027114 +Step 15206 | loss:0.8603727221488953 lr:6.280582370491761e-05 tokens_per_second_per_gpu:3684.7833715720503 mfu:17.381751852576894 +Step 15207 | loss:0.843100368976593 lr:6.28016320656704e-05 tokens_per_second_per_gpu:5355.470243275057 mfu:25.26266687497357 +Step 15208 | loss:0.8410294055938721 lr:6.279744033014046e-05 tokens_per_second_per_gpu:5341.354943980755 mfu:25.196082599883947 +Step 15209 | loss:0.7854827046394348 lr:6.279324849835935e-05 tokens_per_second_per_gpu:5296.674189382727 mfu:24.98531585712222 +Step 15210 | loss:0.8262636065483093 lr:6.278905657035857e-05 tokens_per_second_per_gpu:4118.127736786009 mfu:19.425911159463897 +Step 15211 | loss:0.8181912302970886 lr:6.278486454616968e-05 tokens_per_second_per_gpu:4852.323580401859 mfu:22.88923822052788 +Step 15212 | loss:0.6813924908638 lr:6.278067242582419e-05 tokens_per_second_per_gpu:5346.678221152952 mfu:25.221193406549936 +Step 15213 | loss:0.6704564690589905 lr:6.277648020935363e-05 tokens_per_second_per_gpu:5346.564106048176 mfu:25.220655106130664 +Step 15214 | loss:0.7214062213897705 lr:6.277228789678953e-05 tokens_per_second_per_gpu:5048.817021490473 mfu:23.816131307380935 +Step 15215 | loss:0.8146956562995911 lr:6.276809548816343e-05 tokens_per_second_per_gpu:4758.77439983623 mfu:22.447950774663745 +Step 15216 | loss:0.9576625227928162 lr:6.276390298350685e-05 tokens_per_second_per_gpu:3796.4192221660064 mfu:17.908357206868764 +Step 15217 | loss:0.8203611969947815 lr:6.275971038285133e-05 tokens_per_second_per_gpu:4612.468976199182 mfu:21.75780312908849 +Step 15218 | loss:0.7736063599586487 lr:6.275551768622839e-05 tokens_per_second_per_gpu:5374.343974014131 mfu:25.3516974830603 +Step 15219 | loss:0.8335452675819397 lr:6.275132489366958e-05 tokens_per_second_per_gpu:4919.724027677368 mfu:23.207177630028575 +Step 15220 | loss:0.7374197840690613 lr:6.274713200520642e-05 tokens_per_second_per_gpu:4293.430263635072 mfu:20.2528430883062 +Step 15221 | loss:0.7944996953010559 lr:6.274293902087048e-05 tokens_per_second_per_gpu:5324.676798810075 mfu:25.117408943528854 +Step 15222 | loss:0.8138505816459656 lr:6.273874594069324e-05 tokens_per_second_per_gpu:4655.712975117135 mfu:21.9617924501715 +Step 15223 | loss:0.8418129086494446 lr:6.273455276470628e-05 tokens_per_second_per_gpu:4585.23384860491 mfu:21.62933038543438 +Step 15224 | loss:0.7879402041435242 lr:6.273035949294113e-05 tokens_per_second_per_gpu:5015.074160065729 mfu:23.656960472914605 +Step 15225 | loss:0.8121548295021057 lr:6.272616612542931e-05 tokens_per_second_per_gpu:5336.909809352224 mfu:25.17511414142299 +Step 15226 | loss:1.021582841873169 lr:6.272197266220238e-05 tokens_per_second_per_gpu:2781.3556392230676 mfu:13.120129098421122 +Step 15227 | loss:0.8198775053024292 lr:6.271777910329185e-05 tokens_per_second_per_gpu:3795.268913898493 mfu:17.902931006508076 +Step 15228 | loss:0.7796946167945862 lr:6.27135854487293e-05 tokens_per_second_per_gpu:4277.613116323679 mfu:20.178230905754745 +Step 15229 | loss:0.8145808577537537 lr:6.270939169854625e-05 tokens_per_second_per_gpu:3866.9597567056617 mfu:18.241108943748568 +Step 15230 | loss:0.8524317741394043 lr:6.270519785277425e-05 tokens_per_second_per_gpu:3772.783688013706 mfu:17.796864359634405 +Step 15231 | loss:0.7622677087783813 lr:6.270100391144481e-05 tokens_per_second_per_gpu:3841.501964628516 mfu:18.12102019497345 +Step 15232 | loss:0.7657238841056824 lr:6.269680987458951e-05 tokens_per_second_per_gpu:4398.026422420051 mfu:20.74624101523985 +Step 15233 | loss:0.7335060238838196 lr:6.26926157422399e-05 tokens_per_second_per_gpu:3835.7667071367005 mfu:18.093966006848724 +Step 15234 | loss:0.6461764574050903 lr:6.268842151442748e-05 tokens_per_second_per_gpu:5354.739826454018 mfu:25.259221374206767 +Step 15235 | loss:0.7202883362770081 lr:6.268422719118385e-05 tokens_per_second_per_gpu:5046.059498846603 mfu:23.803123602587924 +Step 15236 | loss:0.796148955821991 lr:6.268003277254052e-05 tokens_per_second_per_gpu:5334.038522360227 mfu:25.161569790789667 +Step 15237 | loss:0.7546111941337585 lr:6.267583825852906e-05 tokens_per_second_per_gpu:4562.11066056496 mfu:21.520254362227234 +Step 15238 | loss:0.8833584189414978 lr:6.267164364918096e-05 tokens_per_second_per_gpu:4794.166428822785 mfu:22.614901013896137 +Step 15239 | loss:0.7930995225906372 lr:6.266744894452784e-05 tokens_per_second_per_gpu:5335.315557724634 mfu:25.167593784488275 +Step 15240 | loss:0.8283724784851074 lr:6.266325414460122e-05 tokens_per_second_per_gpu:5124.242078340231 mfu:24.171924169382937 +Step 15241 | loss:0.8321806192398071 lr:6.265905924943266e-05 tokens_per_second_per_gpu:4789.434972060949 mfu:22.592581925080793 +Step 15242 | loss:0.7972933650016785 lr:6.265486425905367e-05 tokens_per_second_per_gpu:5345.992688953997 mfu:25.217959634203613 +Step 15243 | loss:0.6079341173171997 lr:6.265066917349585e-05 tokens_per_second_per_gpu:5345.723294529843 mfu:25.216688854741452 +Step 15244 | loss:0.7742003798484802 lr:6.264647399279073e-05 tokens_per_second_per_gpu:3835.05685991664 mfu:18.090617536399275 +Step 15245 | loss:0.7456281185150146 lr:6.264227871696986e-05 tokens_per_second_per_gpu:5208.489004403853 mfu:24.56933129363283 +Step 15246 | loss:0.760149359703064 lr:6.263808334606481e-05 tokens_per_second_per_gpu:5313.067466432313 mfu:25.0626457419466 +Step 15247 | loss:0.7300240993499756 lr:6.26338878801071e-05 tokens_per_second_per_gpu:5325.961647954135 mfu:25.123469796159252 +Step 15248 | loss:0.7903676629066467 lr:6.262969231912834e-05 tokens_per_second_per_gpu:4810.250252095897 mfu:22.69077115245863 +Step 15249 | loss:0.7838608026504517 lr:6.262549666316004e-05 tokens_per_second_per_gpu:4522.347278965841 mfu:21.332683706891707 +Step 15250 | loss:0.7624766230583191 lr:6.262130091223377e-05 tokens_per_second_per_gpu:4074.0030842429837 mfu:19.217767644976416 +Step 15251 | loss:0.6569268703460693 lr:6.261710506638108e-05 tokens_per_second_per_gpu:3461.642494801142 mfu:16.32915826508911 +Step 15252 | loss:0.8853024840354919 lr:6.261290912563353e-05 tokens_per_second_per_gpu:4077.502001158532 mfu:19.234272633043844 +Step 15253 | loss:0.8956475853919983 lr:6.260871309002268e-05 tokens_per_second_per_gpu:4372.993221428015 mfu:20.628155135055906 +Step 15254 | loss:0.7000052332878113 lr:6.26045169595801e-05 tokens_per_second_per_gpu:3921.412936132018 mfu:18.497973881773124 +Step 15255 | loss:0.8282644748687744 lr:6.260032073433734e-05 tokens_per_second_per_gpu:5043.8426028344975 mfu:23.79266612585736 +Step 15256 | loss:0.8185927867889404 lr:6.259612441432594e-05 tokens_per_second_per_gpu:4567.009130704987 mfu:21.543361281642927 +Step 15257 | loss:0.7756291031837463 lr:6.259192799957749e-05 tokens_per_second_per_gpu:3931.0971934317827 mfu:18.543656175760788 +Step 15258 | loss:0.8738005757331848 lr:6.258773149012356e-05 tokens_per_second_per_gpu:5105.858395661641 mfu:24.08520520160833 +Step 15259 | loss:0.8004498481750488 lr:6.258353488599569e-05 tokens_per_second_per_gpu:4373.851711503103 mfu:20.632204779213605 +Step 15260 | loss:0.7985506653785706 lr:6.257933818722543e-05 tokens_per_second_per_gpu:3745.398793219307 mfu:17.667685138544375 +Step 15261 | loss:0.6938139200210571 lr:6.257514139384438e-05 tokens_per_second_per_gpu:4335.93358859654 mfu:20.453338523964465 +Step 15262 | loss:0.6879940032958984 lr:6.257094450588407e-05 tokens_per_second_per_gpu:5300.963703602607 mfu:25.005550227563987 +Step 15263 | loss:0.8729948997497559 lr:6.25667475233761e-05 tokens_per_second_per_gpu:4230.43614892567 mfu:19.95568910131769 +Step 15264 | loss:0.7584296464920044 lr:6.256255044635201e-05 tokens_per_second_per_gpu:5316.674901592605 mfu:25.079662629089604 +Step 15265 | loss:0.9129108786582947 lr:6.255835327484339e-05 tokens_per_second_per_gpu:5332.0730672275095 mfu:25.152298403587412 +Step 15266 | loss:0.8855401873588562 lr:6.255415600888177e-05 tokens_per_second_per_gpu:4069.558282799198 mfu:19.196800758204752 +Step 15267 | loss:0.8014329075813293 lr:6.254995864849875e-05 tokens_per_second_per_gpu:3782.304491437102 mfu:17.841775613799133 +Step 15268 | loss:0.7354826331138611 lr:6.254576119372588e-05 tokens_per_second_per_gpu:4893.778767831395 mfu:23.084789412617322 +Step 15269 | loss:0.7127725481987 lr:6.254156364459475e-05 tokens_per_second_per_gpu:5210.359784627401 mfu:24.57815608313532 +Step 15270 | loss:0.7402149438858032 lr:6.253736600113692e-05 tokens_per_second_per_gpu:5313.547517547727 mfu:25.06491022496317 +Step 15271 | loss:0.8752679824829102 lr:6.253316826338396e-05 tokens_per_second_per_gpu:4003.2881478481286 mfu:18.88419322478961 +Step 15272 | loss:0.8136761784553528 lr:6.252897043136743e-05 tokens_per_second_per_gpu:4390.173537253576 mfu:20.70919761606922 +Step 15273 | loss:0.7675890922546387 lr:6.252477250511894e-05 tokens_per_second_per_gpu:5327.140072932749 mfu:25.12902862784341 +Step 15274 | loss:0.7594455480575562 lr:6.252057448467002e-05 tokens_per_second_per_gpu:5319.811354792081 mfu:25.09445781396563 +Step 15275 | loss:0.8021941781044006 lr:6.251637637005227e-05 tokens_per_second_per_gpu:4667.440952738336 mfu:22.017115321610028 +Step 15276 | loss:0.777079701423645 lr:6.251217816129724e-05 tokens_per_second_per_gpu:4390.019105127636 mfu:20.708469133382323 +Step 15277 | loss:0.780083954334259 lr:6.250797985843654e-05 tokens_per_second_per_gpu:4008.246683906163 mfu:18.907583485388848 +Step 15278 | loss:0.7787833213806152 lr:6.250378146150172e-05 tokens_per_second_per_gpu:4214.841860172871 mfu:19.882128180610717 +Step 15279 | loss:0.8505905866622925 lr:6.249958297052437e-05 tokens_per_second_per_gpu:3720.535467887502 mfu:17.550400590833515 +Step 15280 | loss:0.8401354551315308 lr:6.249538438553606e-05 tokens_per_second_per_gpu:4640.955061649179 mfu:21.89217685438347 +Step 15281 | loss:0.9172329306602478 lr:6.249118570656835e-05 tokens_per_second_per_gpu:4295.960169522403 mfu:20.264777086023194 +Step 15282 | loss:0.5970998406410217 lr:6.248698693365286e-05 tokens_per_second_per_gpu:4586.011692430886 mfu:21.632999607475345 +Step 15283 | loss:0.8974513411521912 lr:6.248278806682114e-05 tokens_per_second_per_gpu:4386.8310756394385 mfu:20.693430654352575 +Step 15284 | loss:0.7412649989128113 lr:6.247858910610478e-05 tokens_per_second_per_gpu:4537.944742939669 mfu:21.406259605658025 +Step 15285 | loss:0.8091921806335449 lr:6.247439005153535e-05 tokens_per_second_per_gpu:4434.94555656083 mfu:20.920394870035725 +Step 15286 | loss:0.753807783126831 lr:6.247019090314445e-05 tokens_per_second_per_gpu:4417.014603578872 mfu:20.835811505483786 +Step 15287 | loss:0.8391414284706116 lr:6.246599166096366e-05 tokens_per_second_per_gpu:4757.600648945498 mfu:22.442413991449605 +Step 15288 | loss:0.8971672058105469 lr:6.246179232502454e-05 tokens_per_second_per_gpu:4499.792068551935 mfu:21.226286931053593 +Step 15289 | loss:0.8239011168479919 lr:6.24575928953587e-05 tokens_per_second_per_gpu:5330.614180530351 mfu:25.14541658612518 +Step 15290 | loss:0.8747151494026184 lr:6.245339337199772e-05 tokens_per_second_per_gpu:4646.381545661579 mfu:21.917774505323607 +Step 15291 | loss:0.7853101491928101 lr:6.244919375497318e-05 tokens_per_second_per_gpu:5065.5984002867035 mfu:23.89529193436122 +Step 15292 | loss:0.7904800772666931 lr:6.244499404431665e-05 tokens_per_second_per_gpu:4722.174275466593 mfu:22.275301743387022 +Step 15293 | loss:0.7287480235099792 lr:6.244079424005974e-05 tokens_per_second_per_gpu:5311.704639475385 mfu:25.056217054668515 +Step 15294 | loss:0.9122349619865417 lr:6.243659434223403e-05 tokens_per_second_per_gpu:5325.148335669172 mfu:25.1196332633457 +Step 15295 | loss:0.7156270742416382 lr:6.24323943508711e-05 tokens_per_second_per_gpu:5331.038619547506 mfu:25.14741873738586 +Step 15296 | loss:0.8753398656845093 lr:6.242819426600256e-05 tokens_per_second_per_gpu:5146.094189988029 mfu:24.275004308380304 +Step 15297 | loss:0.8174824118614197 lr:6.242399408765996e-05 tokens_per_second_per_gpu:4003.0385864465366 mfu:18.883016001078733 +Step 15298 | loss:0.836923360824585 lr:6.241979381587494e-05 tokens_per_second_per_gpu:5029.601438801801 mfu:23.725488125321995 +Step 15299 | loss:0.7254959940910339 lr:6.241559345067905e-05 tokens_per_second_per_gpu:4463.33863978375 mfu:21.054329887957774 +Step 15300 | loss:0.9444648623466492 lr:6.241139299210392e-05 tokens_per_second_per_gpu:4512.555252693572 mfu:21.286493048273456 +Step 15301 | loss:0.7823470234870911 lr:6.24071924401811e-05 tokens_per_second_per_gpu:5348.998270319119 mfu:25.232137474307955 +Step 15302 | loss:0.8091468811035156 lr:6.240299179494221e-05 tokens_per_second_per_gpu:4526.51846644291 mfu:21.352359909899015 +Step 15303 | loss:0.7239363193511963 lr:6.239879105641882e-05 tokens_per_second_per_gpu:3989.6386699358436 mfu:18.819806308636288 +Step 15304 | loss:0.7466792464256287 lr:6.239459022464256e-05 tokens_per_second_per_gpu:5290.416865660068 mfu:24.95579899351292 +Step 15305 | loss:0.8298140168190002 lr:6.2390389299645e-05 tokens_per_second_per_gpu:5320.889969424364 mfu:25.09954582321674 +Step 15306 | loss:0.7825111746788025 lr:6.238618828145775e-05 tokens_per_second_per_gpu:3323.054040036354 mfu:15.675412878334958 +Step 15307 | loss:0.920120120048523 lr:6.238198717011238e-05 tokens_per_second_per_gpu:4109.368730270209 mfu:19.38459343128826 +Step 15308 | loss:0.7735350131988525 lr:6.23777859656405e-05 tokens_per_second_per_gpu:2886.895435823333 mfu:13.617978325930691 +Step 15309 | loss:0.8600671887397766 lr:6.237358466807374e-05 tokens_per_second_per_gpu:4804.039076085165 mfu:22.661471975480104 +Step 15310 | loss:0.8754711747169495 lr:6.236938327744365e-05 tokens_per_second_per_gpu:4039.230493835573 mfu:19.05373940320896 +Step 15311 | loss:0.8277496695518494 lr:6.236518179378185e-05 tokens_per_second_per_gpu:5357.849246179537 mfu:25.273889037573745 +Step 15312 | loss:0.7105952501296997 lr:6.236098021711995e-05 tokens_per_second_per_gpu:3731.1169999336594 mfu:17.600315482890775 +Step 15313 | loss:0.8205868601799011 lr:6.235677854748954e-05 tokens_per_second_per_gpu:5077.325769251986 mfu:23.95061193466593 +Step 15314 | loss:0.8231175541877747 lr:6.235257678492221e-05 tokens_per_second_per_gpu:5294.858418693329 mfu:24.97675055697791 +Step 15315 | loss:0.8126239776611328 lr:6.234837492944957e-05 tokens_per_second_per_gpu:3724.3780789972448 mfu:17.568526843055235 +Step 15316 | loss:0.8324025273323059 lr:6.234417298110325e-05 tokens_per_second_per_gpu:4705.50157406754 mfu:22.1966537662315 +Step 15317 | loss:0.8321308493614197 lr:6.233997093991481e-05 tokens_per_second_per_gpu:4509.060692703697 mfu:21.270008612567803 +Step 15318 | loss:0.7997438907623291 lr:6.233576880591588e-05 tokens_per_second_per_gpu:5366.107217954495 mfu:25.3128432993915 +Step 15319 | loss:0.9660413861274719 lr:6.233156657913805e-05 tokens_per_second_per_gpu:4056.7616091562472 mfu:19.136436665293296 +Step 15320 | loss:0.8775877356529236 lr:6.232736425961297e-05 tokens_per_second_per_gpu:4381.885869832788 mfu:20.670103274823486 +Step 15321 | loss:0.8182392120361328 lr:6.232316184737217e-05 tokens_per_second_per_gpu:4126.3505643731605 mfu:19.464699640151547 +Step 15322 | loss:0.8726804852485657 lr:6.231895934244731e-05 tokens_per_second_per_gpu:5107.766065953164 mfu:24.094204007855556 +Step 15323 | loss:0.7720276117324829 lr:6.231475674487e-05 tokens_per_second_per_gpu:5321.819856075276 mfu:25.103932257204118 +Step 15324 | loss:0.7805508375167847 lr:6.231055405467182e-05 tokens_per_second_per_gpu:5322.073178799019 mfu:25.10512722371332 +Step 15325 | loss:0.8390493392944336 lr:6.230635127188439e-05 tokens_per_second_per_gpu:3435.3701183458966 mfu:16.205226982820065 +Step 15326 | loss:0.8112038373947144 lr:6.230214839653932e-05 tokens_per_second_per_gpu:4389.963059738537 mfu:20.70820475771961 +Step 15327 | loss:0.8780132532119751 lr:6.229794542866823e-05 tokens_per_second_per_gpu:5329.66463576155 mfu:25.1409374214417 +Step 15328 | loss:0.7342484593391418 lr:6.22937423683027e-05 tokens_per_second_per_gpu:3784.2708699822474 mfu:17.851051356895358 +Step 15329 | loss:0.7109182476997375 lr:6.228953921547441e-05 tokens_per_second_per_gpu:4810.129003192239 mfu:22.690199200692877 +Step 15330 | loss:0.7273448705673218 lr:6.228533597021488e-05 tokens_per_second_per_gpu:5356.105062887917 mfu:25.26566142739882 +Step 15331 | loss:0.803453266620636 lr:6.22811326325558e-05 tokens_per_second_per_gpu:4079.7057067257892 mfu:19.244667888195536 +Step 15332 | loss:0.7878707051277161 lr:6.227692920252876e-05 tokens_per_second_per_gpu:5352.522318575196 mfu:25.248761011196503 +Step 15333 | loss:0.65632563829422 lr:6.227272568016533e-05 tokens_per_second_per_gpu:5316.906757723128 mfu:25.080756334016016 +Step 15334 | loss:0.8595824837684631 lr:6.226852206549722e-05 tokens_per_second_per_gpu:5225.79017295491 mfu:24.650943857571907 +Step 15335 | loss:0.6759716868400574 lr:6.226431835855593e-05 tokens_per_second_per_gpu:5363.600011089657 mfu:25.301016376836454 +Step 15336 | loss:0.6906886100769043 lr:6.226011455937318e-05 tokens_per_second_per_gpu:5360.262460030803 mfu:25.285272579046907 +Step 15337 | loss:0.8679729104042053 lr:6.225591066798052e-05 tokens_per_second_per_gpu:4289.130771074595 mfu:20.232561648329355 +Step 15338 | loss:0.8968318700790405 lr:6.225170668440961e-05 tokens_per_second_per_gpu:4331.977869525935 mfu:20.43467872219315 +Step 15339 | loss:0.7954443097114563 lr:6.224750260869204e-05 tokens_per_second_per_gpu:5340.059456566995 mfu:25.189971564720295 +Step 15340 | loss:0.7843716740608215 lr:6.224329844085945e-05 tokens_per_second_per_gpu:4482.014763149628 mfu:21.142428348349846 +Step 15341 | loss:0.7567436099052429 lr:6.223909418094344e-05 tokens_per_second_per_gpu:4304.588885262873 mfu:20.305480210381706 +Step 15342 | loss:0.7780480980873108 lr:6.223488982897565e-05 tokens_per_second_per_gpu:4665.73692523815 mfu:22.009077133154577 +Step 15343 | loss:0.6905227899551392 lr:6.223068538498769e-05 tokens_per_second_per_gpu:3788.3334889692105 mfu:17.87021542381114 +Step 15344 | loss:0.8063777089118958 lr:6.222648084901117e-05 tokens_per_second_per_gpu:5271.634269442509 mfu:24.867198282513336 +Step 15345 | loss:0.9229591488838196 lr:6.222227622107775e-05 tokens_per_second_per_gpu:4671.102510146327 mfu:22.034387512629635 +Step 15346 | loss:0.830480694770813 lr:6.221807150121901e-05 tokens_per_second_per_gpu:5270.897996254916 mfu:24.863725156266295 +Step 15347 | loss:0.8431175947189331 lr:6.221386668946661e-05 tokens_per_second_per_gpu:4939.038309357926 mfu:23.298286392072026 +Step 15348 | loss:0.7303512692451477 lr:6.220966178585215e-05 tokens_per_second_per_gpu:5334.925542819181 mfu:25.165754017635848 +Step 15349 | loss:0.8181652426719666 lr:6.220545679040728e-05 tokens_per_second_per_gpu:3827.662660522228 mfu:18.055737836275327 +Step 15350 | loss:0.7538772225379944 lr:6.22012517031636e-05 tokens_per_second_per_gpu:5317.562236556639 mfu:25.083848339509906 +Step 15351 | loss:0.7601209282875061 lr:6.219704652415275e-05 tokens_per_second_per_gpu:4664.9118994527225 mfu:22.0051853457608 +Step 15352 | loss:0.8288061022758484 lr:6.219284125340636e-05 tokens_per_second_per_gpu:4254.287949377124 mfu:20.068202113583997 +Step 15353 | loss:0.8197274804115295 lr:6.218863589095604e-05 tokens_per_second_per_gpu:5021.170049998174 mfu:23.685715825792016 +Step 15354 | loss:0.7129197716712952 lr:6.218443043683346e-05 tokens_per_second_per_gpu:5322.053834006727 mfu:25.10503597102752 +Step 15355 | loss:0.7539252638816833 lr:6.218022489107021e-05 tokens_per_second_per_gpu:5339.496251723895 mfu:25.18731483138274 +Step 15356 | loss:0.7155922055244446 lr:6.217601925369794e-05 tokens_per_second_per_gpu:4006.014336797257 mfu:18.897053123198447 +Step 15357 | loss:0.7382376790046692 lr:6.217181352474826e-05 tokens_per_second_per_gpu:5354.328776586998 mfu:25.25728238185169 +Step 15358 | loss:0.7892054915428162 lr:6.216760770425282e-05 tokens_per_second_per_gpu:3582.879041149981 mfu:16.901051739304368 +Step 15359 | loss:0.832068920135498 lr:6.216340179224327e-05 tokens_per_second_per_gpu:5325.061090291235 mfu:25.119221712012376 +Step 15360 | loss:0.7806599140167236 lr:6.215919578875122e-05 tokens_per_second_per_gpu:5317.624547085066 mfu:25.08414226890377 +Step 15361 | loss:0.7084876298904419 lr:6.21549896938083e-05 tokens_per_second_per_gpu:4370.4976020195845 mfu:20.61638287250976 +Step 15362 | loss:0.7918941974639893 lr:6.215078350744614e-05 tokens_per_second_per_gpu:4017.349394775375 mfu:18.950522525641546 +Step 15363 | loss:0.7667734026908875 lr:6.214657722969641e-05 tokens_per_second_per_gpu:4950.336264856191 mfu:23.35158077578732 +Step 15364 | loss:0.7638955116271973 lr:6.21423708605907e-05 tokens_per_second_per_gpu:4503.479159529846 mfu:21.24367956827899 +Step 15365 | loss:0.763049840927124 lr:6.213816440016069e-05 tokens_per_second_per_gpu:5334.972787547779 mfu:25.165976879081427 +Step 15366 | loss:0.7204607725143433 lr:6.213395784843797e-05 tokens_per_second_per_gpu:3741.3947473675817 mfu:17.64879737110112 +Step 15367 | loss:0.7432758808135986 lr:6.212975120545423e-05 tokens_per_second_per_gpu:3019.387830122205 mfu:14.242967555371946 +Step 15368 | loss:0.7981089353561401 lr:6.212554447124108e-05 tokens_per_second_per_gpu:5366.576939086528 mfu:25.315059054113984 +Step 15369 | loss:0.7256105542182922 lr:6.212133764583014e-05 tokens_per_second_per_gpu:4575.783182477279 mfu:21.58474998086043 +Step 15370 | loss:0.6954776048660278 lr:6.211713072925311e-05 tokens_per_second_per_gpu:4824.714804784082 mfu:22.759002915395968 +Step 15371 | loss:0.7902591824531555 lr:6.211292372154156e-05 tokens_per_second_per_gpu:3360.413129500081 mfu:15.8516420774533 +Step 15372 | loss:0.7702839970588684 lr:6.210871662272718e-05 tokens_per_second_per_gpu:4277.899993368701 mfu:20.179584153722388 +Step 15373 | loss:0.8481869101524353 lr:6.210450943284159e-05 tokens_per_second_per_gpu:5373.573707979858 mfu:25.348064006755695 +Step 15374 | loss:0.7923128604888916 lr:6.210030215191646e-05 tokens_per_second_per_gpu:4585.889503342701 mfu:21.632423220698765 +Step 15375 | loss:0.8099201917648315 lr:6.209609477998338e-05 tokens_per_second_per_gpu:3980.8759778676535 mfu:18.778471195081202 +Step 15376 | loss:0.833320677280426 lr:6.209188731707406e-05 tokens_per_second_per_gpu:5386.514473069926 mfu:25.409107803607746 +Step 15377 | loss:0.7946483492851257 lr:6.20876797632201e-05 tokens_per_second_per_gpu:4449.957613784067 mfu:20.99120930527881 +Step 15378 | loss:0.7852597236633301 lr:6.208347211845316e-05 tokens_per_second_per_gpu:4299.47369744694 mfu:20.281350996712913 +Step 15379 | loss:0.9349352717399597 lr:6.207926438280487e-05 tokens_per_second_per_gpu:3161.5491684560443 mfu:14.913566843518105 +Step 15380 | loss:0.9491159319877625 lr:6.20750565563069e-05 tokens_per_second_per_gpu:4802.046118062398 mfu:22.65207085245301 +Step 15381 | loss:0.7565213441848755 lr:6.207084863899092e-05 tokens_per_second_per_gpu:4162.503554009076 mfu:19.63523946060025 +Step 15382 | loss:0.911742627620697 lr:6.206664063088849e-05 tokens_per_second_per_gpu:4807.899678928165 mfu:22.6796830977773 +Step 15383 | loss:0.7575375437736511 lr:6.206243253203135e-05 tokens_per_second_per_gpu:5318.7127365797405 mfu:25.089275444414422 +Step 15384 | loss:0.920958936214447 lr:6.205822434245112e-05 tokens_per_second_per_gpu:4820.304795067905 mfu:22.738200147139516 +Step 15385 | loss:0.796160101890564 lr:6.205401606217944e-05 tokens_per_second_per_gpu:3902.3221380261853 mfu:18.40791932988188 +Step 15386 | loss:0.8208232522010803 lr:6.204980769124796e-05 tokens_per_second_per_gpu:4319.784628658958 mfu:20.377161124642214 +Step 15387 | loss:0.689017653465271 lr:6.204559922968834e-05 tokens_per_second_per_gpu:4604.076368323754 mfu:21.718213765813534 +Step 15388 | loss:0.8882536888122559 lr:6.204139067753225e-05 tokens_per_second_per_gpu:3225.3707278884895 mfu:15.214624028441836 +Step 15389 | loss:0.8439415097236633 lr:6.203718203481129e-05 tokens_per_second_per_gpu:5352.431190550948 mfu:25.24833114475808 +Step 15390 | loss:0.7510002255439758 lr:6.203297330155718e-05 tokens_per_second_per_gpu:5320.226691995021 mfu:25.09641703041573 +Step 15391 | loss:0.9320406317710876 lr:6.202876447780152e-05 tokens_per_second_per_gpu:4517.790215310832 mfu:21.311187260115798 +Step 15392 | loss:0.821343719959259 lr:6.2024555563576e-05 tokens_per_second_per_gpu:3668.9386182437897 mfu:17.307009447733805 +Step 15393 | loss:0.8068600296974182 lr:6.202034655891224e-05 tokens_per_second_per_gpu:4112.019296746923 mfu:19.39709660559219 +Step 15394 | loss:0.8415922522544861 lr:6.201613746384196e-05 tokens_per_second_per_gpu:4755.804902609009 mfu:22.43394314959447 +Step 15395 | loss:0.8372422456741333 lr:6.201192827839675e-05 tokens_per_second_per_gpu:4306.119055907733 mfu:20.312698286387544 +Step 15396 | loss:0.7009668946266174 lr:6.20077190026083e-05 tokens_per_second_per_gpu:4504.733235895879 mfu:21.24959526046506 +Step 15397 | loss:0.7894718647003174 lr:6.200350963650827e-05 tokens_per_second_per_gpu:5130.2744859536315 mfu:24.200380065330215 +Step 15398 | loss:0.796061635017395 lr:6.19993001801283e-05 tokens_per_second_per_gpu:4797.314192822908 mfu:22.62974955374864 +Step 15399 | loss:0.8369247317314148 lr:6.199509063350007e-05 tokens_per_second_per_gpu:3474.889398463585 mfu:16.391646169819676 +Step 15400 | loss:0.7787806391716003 lr:6.199088099665523e-05 tokens_per_second_per_gpu:5033.351689281021 mfu:23.74317869669119 +Step 15401 | loss:0.8857274651527405 lr:6.198667126962546e-05 tokens_per_second_per_gpu:4246.118673211184 mfu:20.029666244087014 +Step 15402 | loss:0.8643983006477356 lr:6.198246145244237e-05 tokens_per_second_per_gpu:4727.888256008742 mfu:22.30225556450995 +Step 15403 | loss:0.9235061407089233 lr:6.197825154513769e-05 tokens_per_second_per_gpu:4899.047226338232 mfu:23.10964163028578 +Step 15404 | loss:0.9241748452186584 lr:6.197404154774304e-05 tokens_per_second_per_gpu:4825.185278017972 mfu:22.761222217911243 +Step 15405 | loss:0.7618994116783142 lr:6.196983146029009e-05 tokens_per_second_per_gpu:5345.038395300206 mfu:25.213458068219392 +Step 15406 | loss:0.9858636856079102 lr:6.196562128281052e-05 tokens_per_second_per_gpu:5190.441208147235 mfu:24.48419675941879 +Step 15407 | loss:0.9755144715309143 lr:6.196141101533598e-05 tokens_per_second_per_gpu:4710.890309464661 mfu:22.222073350512783 +Step 15408 | loss:0.8165287971496582 lr:6.195720065789815e-05 tokens_per_second_per_gpu:4170.724802972351 mfu:19.674020494649735 +Step 15409 | loss:0.9097132086753845 lr:6.195299021052868e-05 tokens_per_second_per_gpu:3426.376075725333 mfu:16.16280054923696 +Step 15410 | loss:0.804546058177948 lr:6.194877967325925e-05 tokens_per_second_per_gpu:5356.090999628986 mfu:25.265595088607004 +Step 15411 | loss:0.9545283913612366 lr:6.194456904612152e-05 tokens_per_second_per_gpu:4762.625226605328 mfu:22.466115781552375 +Step 15412 | loss:0.8034026622772217 lr:6.194035832914716e-05 tokens_per_second_per_gpu:4118.502563218211 mfu:19.427679279696584 +Step 15413 | loss:0.8068675398826599 lr:6.193614752236785e-05 tokens_per_second_per_gpu:5376.878900895468 mfu:25.36365516566269 +Step 15414 | loss:0.7140009999275208 lr:6.193193662581525e-05 tokens_per_second_per_gpu:5344.231195606928 mfu:25.209650369543738 +Step 15415 | loss:0.6840621829032898 lr:6.192772563952102e-05 tokens_per_second_per_gpu:5338.905627589315 mfu:25.184528756569055 +Step 15416 | loss:0.7824476957321167 lr:6.192351456351685e-05 tokens_per_second_per_gpu:2728.4637467136067 mfu:12.870629017166578 +Step 15417 | loss:0.8344252705574036 lr:6.191930339783442e-05 tokens_per_second_per_gpu:5384.203905167213 mfu:25.39820846058711 +Step 15418 | loss:0.9065377712249756 lr:6.191509214250536e-05 tokens_per_second_per_gpu:5320.244190134576 mfu:25.096499572125502 +Step 15419 | loss:0.7391398549079895 lr:6.191088079756139e-05 tokens_per_second_per_gpu:4448.311549335052 mfu:20.983444538425257 +Step 15420 | loss:0.7560450434684753 lr:6.190666936303417e-05 tokens_per_second_per_gpu:3169.8478845754016 mfu:14.952713303360088 +Step 15421 | loss:0.7745854258537292 lr:6.190245783895536e-05 tokens_per_second_per_gpu:5338.921238838935 mfu:25.184602397497066 +Step 15422 | loss:0.8656876087188721 lr:6.189824622535666e-05 tokens_per_second_per_gpu:4707.1125382031 mfu:22.20425295891776 +Step 15423 | loss:0.8756633400917053 lr:6.18940345222697e-05 tokens_per_second_per_gpu:5205.107141479319 mfu:24.553378469212486 +Step 15424 | loss:0.692383348941803 lr:6.188982272972623e-05 tokens_per_second_per_gpu:4407.607650938104 mfu:20.791437304885658 +Step 15425 | loss:0.7947204113006592 lr:6.188561084775784e-05 tokens_per_second_per_gpu:4777.566226109496 mfu:22.536595025410914 +Step 15426 | loss:0.8433825969696045 lr:6.18813988763963e-05 tokens_per_second_per_gpu:4797.388188217719 mfu:22.63009860265088 +Step 15427 | loss:0.7742084860801697 lr:6.18771868156732e-05 tokens_per_second_per_gpu:4093.5493224707056 mfu:19.30997058562871 +Step 15428 | loss:0.7787386178970337 lr:6.18729746656203e-05 tokens_per_second_per_gpu:4579.1014946895775 mfu:21.600403025728287 +Step 15429 | loss:0.7099461555480957 lr:6.186876242626922e-05 tokens_per_second_per_gpu:4909.4461981762315 mfu:23.158695354693176 +Step 15430 | loss:0.8678514957427979 lr:6.186455009765168e-05 tokens_per_second_per_gpu:4192.149449605164 mfu:19.77508420823836 +Step 15431 | loss:0.7804889678955078 lr:6.186033767979933e-05 tokens_per_second_per_gpu:4316.037359578396 mfu:20.35948461703916 +Step 15432 | loss:0.7224493026733398 lr:6.185612517274387e-05 tokens_per_second_per_gpu:4306.447803858227 mfu:20.31424904656021 +Step 15433 | loss:0.6985799074172974 lr:6.185191257651699e-05 tokens_per_second_per_gpu:4633.608028728643 mfu:21.857519646564228 +Step 15434 | loss:0.8029115200042725 lr:6.184769989115034e-05 tokens_per_second_per_gpu:5299.5728923249735 mfu:24.99898953347052 +Step 15435 | loss:0.7713510990142822 lr:6.184348711667567e-05 tokens_per_second_per_gpu:4651.2507418440955 mfu:21.94074333448707 +Step 15436 | loss:0.7923987507820129 lr:6.183927425312458e-05 tokens_per_second_per_gpu:4640.810416899249 mfu:21.89149454041911 +Step 15437 | loss:0.8108156323432922 lr:6.183506130052881e-05 tokens_per_second_per_gpu:5344.267144283658 mfu:25.209819945585348 +Step 15438 | loss:0.842487633228302 lr:6.183084825892004e-05 tokens_per_second_per_gpu:5271.360757681648 mfu:24.86590808087143 +Step 15439 | loss:0.8486642837524414 lr:6.182663512832996e-05 tokens_per_second_per_gpu:4823.617942068938 mfu:22.75382882682456 +Step 15440 | loss:0.7822614908218384 lr:6.182242190879023e-05 tokens_per_second_per_gpu:4843.981133368645 mfu:22.84988547450397 +Step 15441 | loss:0.8319407105445862 lr:6.181820860033255e-05 tokens_per_second_per_gpu:3587.572460683081 mfu:16.923191400021945 +Step 15442 | loss:0.701824426651001 lr:6.181399520298863e-05 tokens_per_second_per_gpu:5378.707671060037 mfu:25.372281786549326 +Step 15443 | loss:0.7357353568077087 lr:6.180978171679014e-05 tokens_per_second_per_gpu:3745.5464963003337 mfu:17.668381879178323 +Step 15444 | loss:0.722047746181488 lr:6.180556814176877e-05 tokens_per_second_per_gpu:4041.6895473338545 mfu:19.065339178118148 +Step 15445 | loss:0.8816401958465576 lr:6.180135447795622e-05 tokens_per_second_per_gpu:5239.125572081632 mfu:24.71384921050636 +Step 15446 | loss:0.8196773529052734 lr:6.179714072538419e-05 tokens_per_second_per_gpu:4440.3201780884365 mfu:20.945747876786438 +Step 15447 | loss:0.8258664608001709 lr:6.179292688408434e-05 tokens_per_second_per_gpu:5329.028714876144 mfu:25.137937674125947 +Step 15448 | loss:0.8787205815315247 lr:6.178871295408838e-05 tokens_per_second_per_gpu:3397.095884113063 mfu:16.024680889686053 +Step 15449 | loss:0.7327398657798767 lr:6.178449893542804e-05 tokens_per_second_per_gpu:4613.879726745947 mfu:21.764457880116442 +Step 15450 | loss:0.7728046774864197 lr:6.178028482813495e-05 tokens_per_second_per_gpu:5325.009177375856 mfu:25.118976829933274 +Step 15451 | loss:0.7306195497512817 lr:6.177607063224085e-05 tokens_per_second_per_gpu:3965.6147185205045 mfu:18.706481230901353 +Step 15452 | loss:0.7452468276023865 lr:6.177185634777739e-05 tokens_per_second_per_gpu:4154.4719552341785 mfu:19.597353038847004 +Step 15453 | loss:0.8319695591926575 lr:6.176764197477634e-05 tokens_per_second_per_gpu:4665.792165333755 mfu:22.00933771011048 +Step 15454 | loss:0.7848209738731384 lr:6.176342751326932e-05 tokens_per_second_per_gpu:3286.6687855613504 mfu:15.503777425011808 +Step 15455 | loss:0.7263956069946289 lr:6.175921296328807e-05 tokens_per_second_per_gpu:3693.5026829532476 mfu:17.42288233202993 +Step 15456 | loss:0.8991307616233826 lr:6.175499832486428e-05 tokens_per_second_per_gpu:4740.359064037547 mfu:22.361082493721074 +Step 15457 | loss:0.9141559600830078 lr:6.175078359802965e-05 tokens_per_second_per_gpu:4424.650017795863 mfu:20.87182905708133 +Step 15458 | loss:0.7470498085021973 lr:6.174656878281588e-05 tokens_per_second_per_gpu:5351.869197674648 mfu:25.245680128474703 +Step 15459 | loss:0.8108314871788025 lr:6.174235387925464e-05 tokens_per_second_per_gpu:5139.4154120344165 mfu:24.24349937325609 +Step 15460 | loss:0.9613127112388611 lr:6.173813888737768e-05 tokens_per_second_per_gpu:4521.949629515332 mfu:21.330807926587667 +Step 15461 | loss:0.8102557063102722 lr:6.173392380721669e-05 tokens_per_second_per_gpu:3880.804408745146 mfu:18.30641652956003 +Step 15462 | loss:0.8560751080513 lr:6.172970863880335e-05 tokens_per_second_per_gpu:3797.923975175987 mfu:17.91545538355418 +Step 15463 | loss:0.7676035165786743 lr:6.172549338216936e-05 tokens_per_second_per_gpu:5337.4716810076 mfu:25.17776458588659 +Step 15464 | loss:0.887426495552063 lr:6.172127803734647e-05 tokens_per_second_per_gpu:4889.668887680049 mfu:23.06540240672556 +Step 15465 | loss:0.6965815424919128 lr:6.171706260436634e-05 tokens_per_second_per_gpu:5322.085637772085 mfu:25.105185994814992 +Step 15466 | loss:0.8351608514785767 lr:6.171284708326067e-05 tokens_per_second_per_gpu:4684.458814900262 mfu:22.09739148097452 +Step 15467 | loss:0.8439849615097046 lr:6.170863147406119e-05 tokens_per_second_per_gpu:5010.6608129616625 mfu:23.63614196163006 +Step 15468 | loss:0.8214610815048218 lr:6.170441577679962e-05 tokens_per_second_per_gpu:3775.4467323881017 mfu:17.809426394310652 +Step 15469 | loss:0.9285189509391785 lr:6.170019999150763e-05 tokens_per_second_per_gpu:5077.896001132073 mfu:23.95330181573589 +Step 15470 | loss:0.8794867396354675 lr:6.169598411821693e-05 tokens_per_second_per_gpu:4472.9349848220045 mfu:21.099597484808516 +Step 15471 | loss:0.7960372567176819 lr:6.169176815695927e-05 tokens_per_second_per_gpu:4384.222735655937 mfu:20.681126669621708 +Step 15472 | loss:0.7584754824638367 lr:6.16875521077663e-05 tokens_per_second_per_gpu:5188.507953619149 mfu:24.475077267962575 +Step 15473 | loss:0.8090240359306335 lr:6.168333597066978e-05 tokens_per_second_per_gpu:5302.073784151939 mfu:25.010786666914463 +Step 15474 | loss:0.8150227069854736 lr:6.167911974570138e-05 tokens_per_second_per_gpu:4116.416159451135 mfu:19.417837357150095 +Step 15475 | loss:0.694841206073761 lr:6.167490343289285e-05 tokens_per_second_per_gpu:4387.154696714413 mfu:20.694957230178677 +Step 15476 | loss:0.8211259245872498 lr:6.167068703227588e-05 tokens_per_second_per_gpu:5327.288487930997 mfu:25.129728726712354 +Step 15477 | loss:0.8150836825370789 lr:6.166647054388215e-05 tokens_per_second_per_gpu:5060.462097055104 mfu:23.871063115674072 +Step 15478 | loss:0.8161072731018066 lr:6.166225396774345e-05 tokens_per_second_per_gpu:5256.044129168974 mfu:24.79365693089085 +Step 15479 | loss:0.9011901021003723 lr:6.165803730389143e-05 tokens_per_second_per_gpu:4479.679658747053 mfu:21.131413262473004 +Step 15480 | loss:0.867551863193512 lr:6.165382055235783e-05 tokens_per_second_per_gpu:5102.460539255659 mfu:24.069176933207224 +Step 15481 | loss:0.7599889636039734 lr:6.164960371317436e-05 tokens_per_second_per_gpu:5360.9415512033665 mfu:25.288475967972797 +Step 15482 | loss:0.8562391400337219 lr:6.164538678637273e-05 tokens_per_second_per_gpu:5344.862400163567 mfu:25.212627869132003 +Step 15483 | loss:0.9137733578681946 lr:6.164116977198467e-05 tokens_per_second_per_gpu:4201.872856905934 mfu:19.820951179460447 +Step 15484 | loss:0.9554556608200073 lr:6.163695267004186e-05 tokens_per_second_per_gpu:5321.8017962106 mfu:25.103847065741103 +Step 15485 | loss:0.7702134847640991 lr:6.163273548057607e-05 tokens_per_second_per_gpu:5341.036138927915 mfu:25.194578742055953 +Step 15486 | loss:1.039078950881958 lr:6.162851820361897e-05 tokens_per_second_per_gpu:4155.362178951236 mfu:19.601552376008193 +Step 15487 | loss:0.7796617746353149 lr:6.162430083920232e-05 tokens_per_second_per_gpu:5314.162358775926 mfu:25.067810535939316 +Step 15488 | loss:0.6904338598251343 lr:6.16200833873578e-05 tokens_per_second_per_gpu:4405.763902153695 mfu:20.78274002729362 +Step 15489 | loss:0.8393682837486267 lr:6.161586584811717e-05 tokens_per_second_per_gpu:4969.084776644261 mfu:23.44002070471761 +Step 15490 | loss:0.8465255498886108 lr:6.161164822151213e-05 tokens_per_second_per_gpu:3436.2156370944826 mfu:16.209215439017523 +Step 15491 | loss:0.8856301307678223 lr:6.160743050757441e-05 tokens_per_second_per_gpu:3075.41477506021 mfu:14.507256213827036 +Step 15492 | loss:0.9293743968009949 lr:6.16032127063357e-05 tokens_per_second_per_gpu:4527.719693530392 mfu:21.35802630302114 +Step 15493 | loss:0.7597982883453369 lr:6.159899481782775e-05 tokens_per_second_per_gpu:4651.064690075938 mfu:21.93986569655297 +Step 15494 | loss:0.7312889695167542 lr:6.15947768420823e-05 tokens_per_second_per_gpu:5340.465829814847 mfu:25.191888496664838 +Step 15495 | loss:0.8164964318275452 lr:6.159055877913104e-05 tokens_per_second_per_gpu:5354.918061266103 mfu:25.26006213822537 +Step 15496 | loss:0.8105493187904358 lr:6.15863406290057e-05 tokens_per_second_per_gpu:5354.896676503306 mfu:25.259961262650634 +Step 15497 | loss:0.8385931849479675 lr:6.158212239173803e-05 tokens_per_second_per_gpu:4384.8808961276245 mfu:20.684231324860413 +Step 15498 | loss:0.798519492149353 lr:6.157790406735972e-05 tokens_per_second_per_gpu:4263.176610384007 mfu:20.110131444115414 +Step 15499 | loss:0.8332879543304443 lr:6.157368565590255e-05 tokens_per_second_per_gpu:4556.822260730332 mfu:21.49530807791324 +Step 15500 | loss:0.7462329864501953 lr:6.156946715739818e-05 tokens_per_second_per_gpu:4845.276776477151 mfu:22.8559972441063 +Step 15501 | loss:0.7507407069206238 lr:6.15652485718784e-05 tokens_per_second_per_gpu:3756.338809975645 mfu:17.719291064143352 +Step 15502 | loss:0.8656689524650574 lr:6.156102989937488e-05 tokens_per_second_per_gpu:4526.895212349858 mfu:21.35413708462157 +Step 15503 | loss:0.6789586544036865 lr:6.155681113991938e-05 tokens_per_second_per_gpu:5088.844230117448 mfu:24.00494647982039 +Step 15504 | loss:0.7660694122314453 lr:6.155259229354365e-05 tokens_per_second_per_gpu:5313.398689246972 mfu:25.064208176475784 +Step 15505 | loss:0.8090820908546448 lr:6.15483733602794e-05 tokens_per_second_per_gpu:5330.651319799782 mfu:25.145591778395698 +Step 15506 | loss:0.7675656676292419 lr:6.154415434015834e-05 tokens_per_second_per_gpu:4029.0841755234587 mfu:19.005877488590308 +Step 15507 | loss:0.8081279397010803 lr:6.153993523321223e-05 tokens_per_second_per_gpu:5330.203906238072 mfu:25.143481252288485 +Step 15508 | loss:0.7170380353927612 lr:6.15357160394728e-05 tokens_per_second_per_gpu:5355.97047837586 mfu:25.265026569292964 +Step 15509 | loss:0.816079318523407 lr:6.153149675897177e-05 tokens_per_second_per_gpu:4501.122786325822 mfu:21.232564153836883 +Step 15510 | loss:0.8723058104515076 lr:6.15272773917409e-05 tokens_per_second_per_gpu:4050.512806548244 mfu:19.106959997235677 +Step 15511 | loss:0.7422910332679749 lr:6.152305793781188e-05 tokens_per_second_per_gpu:4448.727929600603 mfu:20.985408675179656 +Step 15512 | loss:0.8037607669830322 lr:6.15188383972165e-05 tokens_per_second_per_gpu:4907.973290200055 mfu:23.151747396465538 +Step 15513 | loss:0.8051517605781555 lr:6.151461876998643e-05 tokens_per_second_per_gpu:4350.398229385512 mfu:20.52157070248825 +Step 15514 | loss:0.8196626305580139 lr:6.151039905615346e-05 tokens_per_second_per_gpu:3576.602105156268 mfu:16.871442361266308 +Step 15515 | loss:0.8814484477043152 lr:6.150617925574933e-05 tokens_per_second_per_gpu:4567.208110812339 mfu:21.544299904758816 +Step 15516 | loss:0.7625731825828552 lr:6.150195936880574e-05 tokens_per_second_per_gpu:4733.065248560411 mfu:22.326676321662184 +Step 15517 | loss:0.7776468992233276 lr:6.149773939535444e-05 tokens_per_second_per_gpu:4756.392653679364 mfu:22.43671567167401 +Step 15518 | loss:0.8563228845596313 lr:6.149351933542719e-05 tokens_per_second_per_gpu:4506.069959175055 mfu:21.255900812241443 +Step 15519 | loss:0.718626081943512 lr:6.14892991890557e-05 tokens_per_second_per_gpu:4655.948652220475 mfu:21.962904179278617 +Step 15520 | loss:0.7442306876182556 lr:6.148507895627172e-05 tokens_per_second_per_gpu:5319.031029138563 mfu:25.0907768847207 +Step 15521 | loss:0.7650249600410461 lr:6.148085863710701e-05 tokens_per_second_per_gpu:5312.87788607195 mfu:25.061751459041023 +Step 15522 | loss:0.7592976689338684 lr:6.14766382315933e-05 tokens_per_second_per_gpu:3141.965650508659 mfu:14.821188048067585 +Step 15523 | loss:0.8253648281097412 lr:6.147241773976232e-05 tokens_per_second_per_gpu:5367.623838812471 mfu:25.319997458740772 +Step 15524 | loss:0.8797046542167664 lr:6.146819716164583e-05 tokens_per_second_per_gpu:5390.429243343478 mfu:25.42757444291677 +Step 15525 | loss:0.7118585109710693 lr:6.146397649727555e-05 tokens_per_second_per_gpu:3592.389719373544 mfu:16.94591523117394 +Step 15526 | loss:0.7455561757087708 lr:6.145975574668326e-05 tokens_per_second_per_gpu:4362.7167868019 mfu:20.579679439583973 +Step 15527 | loss:0.7416017055511475 lr:6.145553490990068e-05 tokens_per_second_per_gpu:5366.003781594438 mfu:25.312355372433046 +Step 15528 | loss:0.7126383781433105 lr:6.145131398695955e-05 tokens_per_second_per_gpu:4760.002909753377 mfu:22.45374586554833 +Step 15529 | loss:0.6762894988059998 lr:6.144709297789164e-05 tokens_per_second_per_gpu:5203.994890766694 mfu:24.548131792831494 +Step 15530 | loss:0.7791178226470947 lr:6.144287188272867e-05 tokens_per_second_per_gpu:5048.313353178874 mfu:23.81375541802045 +Step 15531 | loss:0.742798924446106 lr:6.14386507015024e-05 tokens_per_second_per_gpu:4311.660198868018 mfu:20.338836803147686 +Step 15532 | loss:0.8068894743919373 lr:6.143442943424459e-05 tokens_per_second_per_gpu:5328.476770635325 mfu:25.13533405897061 +Step 15533 | loss:0.8194206357002258 lr:6.143020808098697e-05 tokens_per_second_per_gpu:3814.1342124830308 mfu:17.991921838683187 +Step 15534 | loss:0.8097000122070312 lr:6.14259866417613e-05 tokens_per_second_per_gpu:3507.416168894009 mfu:16.54508049557928 +Step 15535 | loss:0.7796715497970581 lr:6.142176511659933e-05 tokens_per_second_per_gpu:3943.2742390758553 mfu:18.601097377682102 +Step 15536 | loss:0.8617669939994812 lr:6.141754350553279e-05 tokens_per_second_per_gpu:5045.0065859119195 mfu:23.798156832629548 +Step 15537 | loss:0.8754819631576538 lr:6.141332180859348e-05 tokens_per_second_per_gpu:5353.061041768884 mfu:25.251302260416786 +Step 15538 | loss:0.7671781182289124 lr:6.140910002581308e-05 tokens_per_second_per_gpu:5319.071157726389 mfu:25.090966178116116 +Step 15539 | loss:0.6257416009902954 lr:6.14048781572234e-05 tokens_per_second_per_gpu:5323.940826547569 mfu:25.113937236796797 +Step 15540 | loss:0.6410924196243286 lr:6.140065620285619e-05 tokens_per_second_per_gpu:4687.791877275109 mfu:22.113114104875752 +Step 15541 | loss:0.960540771484375 lr:6.139643416274317e-05 tokens_per_second_per_gpu:4249.880151502604 mfu:20.047409778962017 +Step 15542 | loss:0.833539605140686 lr:6.139221203691611e-05 tokens_per_second_per_gpu:3681.119200075169 mfu:17.36446733045398 +Step 15543 | loss:0.6759905219078064 lr:6.13879898254068e-05 tokens_per_second_per_gpu:4245.846751829359 mfu:20.02838354453438 +Step 15544 | loss:0.744249165058136 lr:6.138376752824695e-05 tokens_per_second_per_gpu:4638.7536618773265 mfu:21.88179247606211 +Step 15545 | loss:0.7861508727073669 lr:6.137954514546833e-05 tokens_per_second_per_gpu:3715.616523474347 mfu:17.527197090778206 +Step 15546 | loss:0.801029622554779 lr:6.13753226771027e-05 tokens_per_second_per_gpu:4269.538248495125 mfu:20.140140376492944 +Step 15547 | loss:0.9411979913711548 lr:6.137110012318182e-05 tokens_per_second_per_gpu:4457.421162603216 mfu:21.026416138471475 +Step 15548 | loss:0.759257972240448 lr:6.136687748373744e-05 tokens_per_second_per_gpu:4967.777806598531 mfu:23.433855504019817 +Step 15549 | loss:0.6717490553855896 lr:6.136265475880132e-05 tokens_per_second_per_gpu:5237.78584030191 mfu:24.7075294671196 +Step 15550 | loss:0.8362376689910889 lr:6.135843194840523e-05 tokens_per_second_per_gpu:3952.4714325653586 mfu:18.64448210857511 +Step 15551 | loss:0.8936439752578735 lr:6.135420905258092e-05 tokens_per_second_per_gpu:4404.8961652244225 mfu:20.77864676414706 +Step 15552 | loss:0.7802719473838806 lr:6.134998607136017e-05 tokens_per_second_per_gpu:5375.1839250244475 mfu:25.355659675286383 +Step 15553 | loss:0.7924786806106567 lr:6.134576300477471e-05 tokens_per_second_per_gpu:3253.2704505945308 mfu:15.346231780629779 +Step 15554 | loss:1.0583463907241821 lr:6.134153985285631e-05 tokens_per_second_per_gpu:4678.785877813777 mfu:22.070631268834692 +Step 15555 | loss:0.7268081307411194 lr:6.133731661563677e-05 tokens_per_second_per_gpu:4228.773172911431 mfu:19.94784455972556 +Step 15556 | loss:0.8260197639465332 lr:6.13330932931478e-05 tokens_per_second_per_gpu:4327.579003659898 mfu:20.41392852137915 +Step 15557 | loss:0.867683470249176 lr:6.13288698854212e-05 tokens_per_second_per_gpu:5021.08466571191 mfu:23.685313053545645 +Step 15558 | loss:0.7842973470687866 lr:6.132464639248873e-05 tokens_per_second_per_gpu:4985.165258338708 mfu:23.515875080483404 +Step 15559 | loss:0.8380113244056702 lr:6.132042281438214e-05 tokens_per_second_per_gpu:4896.11846049748 mfu:23.09582614211597 +Step 15560 | loss:0.8224282264709473 lr:6.13161991511332e-05 tokens_per_second_per_gpu:3879.554835340086 mfu:18.300522078609152 +Step 15561 | loss:0.8002675771713257 lr:6.13119754027737e-05 tokens_per_second_per_gpu:4754.807360789433 mfu:22.429237574633348 +Step 15562 | loss:0.8664277195930481 lr:6.130775156933537e-05 tokens_per_second_per_gpu:5165.983674622574 mfu:24.36882640089745 +Step 15563 | loss:0.8079264163970947 lr:6.130352765085001e-05 tokens_per_second_per_gpu:3925.0387760838403 mfu:18.515077587457956 +Step 15564 | loss:0.8291031718254089 lr:6.129930364734937e-05 tokens_per_second_per_gpu:4345.959932785047 mfu:20.50063450017237 +Step 15565 | loss:0.7812328934669495 lr:6.129507955886523e-05 tokens_per_second_per_gpu:5338.572427358328 mfu:25.18295699422967 +Step 15566 | loss:0.7398653626441956 lr:6.129085538542936e-05 tokens_per_second_per_gpu:4639.984679224258 mfu:21.887599394920805 +Step 15567 | loss:0.7631993293762207 lr:6.128663112707352e-05 tokens_per_second_per_gpu:4517.773643237173 mfu:21.311109086816863 +Step 15568 | loss:0.6079069375991821 lr:6.128240678382948e-05 tokens_per_second_per_gpu:4593.370079915772 mfu:21.66771037671192 +Step 15569 | loss:0.8071202039718628 lr:6.127818235572903e-05 tokens_per_second_per_gpu:3944.6375684333357 mfu:18.607528434869316 +Step 15570 | loss:0.9128727912902832 lr:6.127395784280393e-05 tokens_per_second_per_gpu:4146.239729923747 mfu:19.558520227495226 +Step 15571 | loss:0.8289549350738525 lr:6.126973324508594e-05 tokens_per_second_per_gpu:5014.407599731953 mfu:23.653816194093263 +Step 15572 | loss:0.815080463886261 lr:6.126550856260687e-05 tokens_per_second_per_gpu:5360.913498098026 mfu:25.28834363668855 +Step 15573 | loss:0.8183929324150085 lr:6.126128379539847e-05 tokens_per_second_per_gpu:5300.45724237332 mfu:25.003161163534244 +Step 15574 | loss:0.790111243724823 lr:6.12570589434925e-05 tokens_per_second_per_gpu:4576.2623080244275 mfu:21.58701009781358 +Step 15575 | loss:0.7973673343658447 lr:6.125283400692076e-05 tokens_per_second_per_gpu:3512.3093169859394 mfu:16.568162309986068 +Step 15576 | loss:0.669477641582489 lr:6.124860898571503e-05 tokens_per_second_per_gpu:5322.310901411484 mfu:25.106248601835983 +Step 15577 | loss:0.7823635339736938 lr:6.124438387990707e-05 tokens_per_second_per_gpu:3097.9568802186313 mfu:14.613591169939044 +Step 15578 | loss:0.8569941520690918 lr:6.124015868952866e-05 tokens_per_second_per_gpu:5402.834728688276 mfu:25.48609323388224 +Step 15579 | loss:0.8119790554046631 lr:6.123593341461157e-05 tokens_per_second_per_gpu:4676.569083341355 mfu:22.06017427108414 +Step 15580 | loss:0.8918256163597107 lr:6.123170805518763e-05 tokens_per_second_per_gpu:4174.971838173793 mfu:19.69405448431386 +Step 15581 | loss:0.8629001975059509 lr:6.122748261128855e-05 tokens_per_second_per_gpu:4269.1856124122805 mfu:20.138476931924245 +Step 15582 | loss:0.8815423250198364 lr:6.122325708294615e-05 tokens_per_second_per_gpu:4983.308466472226 mfu:23.507116276447853 +Step 15583 | loss:0.7843664884567261 lr:6.121903147019219e-05 tokens_per_second_per_gpu:3936.4388377661917 mfu:18.5688536234647 +Step 15584 | loss:0.8660099506378174 lr:6.121480577305849e-05 tokens_per_second_per_gpu:4603.86378841027 mfu:21.717210990091758 +Step 15585 | loss:0.9190479516983032 lr:6.121057999157678e-05 tokens_per_second_per_gpu:5327.460534947671 mfu:25.130540301844256 +Step 15586 | loss:0.7684306502342224 lr:6.120635412577888e-05 tokens_per_second_per_gpu:4579.824983324844 mfu:21.603815845060723 +Step 15587 | loss:0.8149915933609009 lr:6.120212817569655e-05 tokens_per_second_per_gpu:4824.6231035829505 mfu:22.758570345184395 +Step 15588 | loss:0.8143717646598816 lr:6.119790214136159e-05 tokens_per_second_per_gpu:4891.433523661297 mfu:23.07372649572705 +Step 15589 | loss:0.8060919046401978 lr:6.119367602280578e-05 tokens_per_second_per_gpu:3602.192654421692 mfu:16.992157348349245 +Step 15590 | loss:0.7375719547271729 lr:6.118944982006089e-05 tokens_per_second_per_gpu:4471.9673620242165 mfu:21.095033042977878 +Step 15591 | loss:0.7134512066841125 lr:6.118522353315876e-05 tokens_per_second_per_gpu:4308.335899693492 mfu:20.32315551675752 +Step 15592 | loss:0.862856924533844 lr:6.11809971621311e-05 tokens_per_second_per_gpu:4690.3770223191605 mfu:22.12530867511977 +Step 15593 | loss:0.7581678032875061 lr:6.117677070700974e-05 tokens_per_second_per_gpu:4954.653632870445 mfu:23.3719465373295 +Step 15594 | loss:0.8191208243370056 lr:6.117254416782646e-05 tokens_per_second_per_gpu:3848.535967384922 mfu:18.15420078610004 +Step 15595 | loss:0.8292069435119629 lr:6.116831754461306e-05 tokens_per_second_per_gpu:5330.439599139692 mfu:25.144593055918783 +Step 15596 | loss:0.8145799040794373 lr:6.116409083740131e-05 tokens_per_second_per_gpu:4213.468665792402 mfu:19.87565058842693 +Step 15597 | loss:0.7951831817626953 lr:6.115986404622301e-05 tokens_per_second_per_gpu:3725.893075564345 mfu:17.575673340347315 +Step 15598 | loss:0.9125832915306091 lr:6.115563717110995e-05 tokens_per_second_per_gpu:5361.765324647487 mfu:25.292361847858956 +Step 15599 | loss:0.8553460240364075 lr:6.115141021209392e-05 tokens_per_second_per_gpu:5350.238827888691 mfu:25.237989396024574 +Step 15600 | loss:0.8649057149887085 lr:6.11471831692067e-05 tokens_per_second_per_gpu:5353.582693652598 mfu:25.253762981355255 +Step 15601 | loss:0.7343688607215881 lr:6.114295604248012e-05 tokens_per_second_per_gpu:4418.794630838637 mfu:20.844208197772065 +Step 15602 | loss:0.9539359211921692 lr:6.113872883194594e-05 tokens_per_second_per_gpu:4023.625308184394 mfu:18.98012707997313 +Step 15603 | loss:0.6546573638916016 lr:6.113450153763594e-05 tokens_per_second_per_gpu:5365.0698628492855 mfu:25.307949918369193 +Step 15604 | loss:0.7408540844917297 lr:6.113027415958194e-05 tokens_per_second_per_gpu:4958.640147181911 mfu:23.39075160550616 +Step 15605 | loss:0.8645684719085693 lr:6.112604669781572e-05 tokens_per_second_per_gpu:5372.786807434742 mfu:25.3443520626253 +Step 15606 | loss:0.843451976776123 lr:6.112181915236909e-05 tokens_per_second_per_gpu:5254.672655535167 mfu:24.787187455762652 +Step 15607 | loss:0.7845059037208557 lr:6.111759152327385e-05 tokens_per_second_per_gpu:4102.520787425109 mfu:19.352290516508393 +Step 15608 | loss:0.8522829413414001 lr:6.111336381056178e-05 tokens_per_second_per_gpu:4331.345769346072 mfu:20.43169699779733 +Step 15609 | loss:0.7591803073883057 lr:6.110913601426468e-05 tokens_per_second_per_gpu:5358.026441956696 mfu:25.274724900287325 +Step 15610 | loss:0.8372118473052979 lr:6.110490813441434e-05 tokens_per_second_per_gpu:5122.045492113168 mfu:24.161562497373623 +Step 15611 | loss:0.8272819519042969 lr:6.110068017104258e-05 tokens_per_second_per_gpu:5296.243034037753 mfu:24.983282023797294 +Step 15612 | loss:0.9050953984260559 lr:6.109645212418119e-05 tokens_per_second_per_gpu:4216.1177120165585 mfu:19.888146591440165 +Step 15613 | loss:0.9997900128364563 lr:6.109222399386197e-05 tokens_per_second_per_gpu:4882.141437129228 mfu:23.029894138161836 +Step 15614 | loss:0.8014762997627258 lr:6.108799578011671e-05 tokens_per_second_per_gpu:4473.469743881406 mfu:21.102120034531016 +Step 15615 | loss:0.7611982822418213 lr:6.108376748297722e-05 tokens_per_second_per_gpu:5324.0365256333625 mfu:25.114388665712433 +Step 15616 | loss:0.739686906337738 lr:6.10795391024753e-05 tokens_per_second_per_gpu:5331.7499105966945 mfu:25.150774018623654 +Step 15617 | loss:0.8221595287322998 lr:6.107531063864276e-05 tokens_per_second_per_gpu:4241.243889542389 mfu:20.00667105779745 +Step 15618 | loss:0.8267609477043152 lr:6.107108209151139e-05 tokens_per_second_per_gpu:5321.326198071006 mfu:25.10160358817123 +Step 15619 | loss:0.7217467427253723 lr:6.106685346111298e-05 tokens_per_second_per_gpu:5336.90701423891 mfu:25.175100956396594 +Step 15620 | loss:0.742729127407074 lr:6.10626247474794e-05 tokens_per_second_per_gpu:3800.9700850179984 mfu:17.92982440339888 +Step 15621 | loss:0.7597109079360962 lr:6.105839595064236e-05 tokens_per_second_per_gpu:3593.471807629328 mfu:16.95101962609977 +Step 15622 | loss:0.7772513031959534 lr:6.105416707063374e-05 tokens_per_second_per_gpu:4964.7402532535625 mfu:23.41952684663156 +Step 15623 | loss:0.8676155209541321 lr:6.104993810748532e-05 tokens_per_second_per_gpu:4418.530593343332 mfu:20.842962687857266 +Step 15624 | loss:0.8728330731391907 lr:6.104570906122889e-05 tokens_per_second_per_gpu:3555.4442423236087 mfu:16.771637112381015 +Step 15625 | loss:0.8327261209487915 lr:6.104147993189628e-05 tokens_per_second_per_gpu:4660.379265910656 mfu:21.983804139995843 +Step 15626 | loss:0.7822499871253967 lr:6.103725071951929e-05 tokens_per_second_per_gpu:4017.93734645539 mfu:18.953295994031077 +Step 15627 | loss:0.7119497656822205 lr:6.103302142412973e-05 tokens_per_second_per_gpu:5034.705815048107 mfu:23.749566338972212 +Step 15628 | loss:0.8099821209907532 lr:6.102879204575941e-05 tokens_per_second_per_gpu:5009.186572670626 mfu:23.62918771864591 +Step 15629 | loss:0.6306302547454834 lr:6.102456258444014e-05 tokens_per_second_per_gpu:4401.737288236776 mfu:20.76374580243634 +Step 15630 | loss:0.8369097113609314 lr:6.10203330402037e-05 tokens_per_second_per_gpu:5000.015917320353 mfu:23.585928172683523 +Step 15631 | loss:0.8726359009742737 lr:6.101610341308197e-05 tokens_per_second_per_gpu:5347.686551112936 mfu:25.22594987101002 +Step 15632 | loss:0.810122013092041 lr:6.101187370310669e-05 tokens_per_second_per_gpu:5330.71885118173 mfu:25.14591033544639 +Step 15633 | loss:0.8159351944923401 lr:6.100764391030972e-05 tokens_per_second_per_gpu:4132.004715567228 mfu:19.491371236031636 +Step 15634 | loss:0.8255265355110168 lr:6.100341403472285e-05 tokens_per_second_per_gpu:5133.867813306607 mfu:24.217330403539066 +Step 15635 | loss:0.726035475730896 lr:6.09991840763779e-05 tokens_per_second_per_gpu:5304.358079733207 mfu:25.021562078912314 +Step 15636 | loss:0.8019963502883911 lr:6.0994954035306694e-05 tokens_per_second_per_gpu:4118.055408052029 mfu:19.42556997248516 +Step 15637 | loss:0.9361163377761841 lr:6.099072391154101e-05 tokens_per_second_per_gpu:4569.62267999868 mfu:21.555689839578772 +Step 15638 | loss:0.910313069820404 lr:6.098649370511271e-05 tokens_per_second_per_gpu:5078.540288000654 mfu:23.95634102682166 +Step 15639 | loss:0.735990047454834 lr:6.0982263416053584e-05 tokens_per_second_per_gpu:5135.420622409423 mfu:24.224655269013603 +Step 15640 | loss:0.8219272494316101 lr:6.097803304439546e-05 tokens_per_second_per_gpu:4615.128054804156 mfu:21.770346456553742 +Step 15641 | loss:0.6975088119506836 lr:6.097380259017014e-05 tokens_per_second_per_gpu:3425.1196413690473 mfu:16.156873734008553 +Step 15642 | loss:0.833907961845398 lr:6.096957205340945e-05 tokens_per_second_per_gpu:4157.523675343891 mfu:19.611748523281662 +Step 15643 | loss:0.7571641802787781 lr:6.096534143414521e-05 tokens_per_second_per_gpu:3606.659658689017 mfu:17.013228969627736 +Step 15644 | loss:0.7870368361473083 lr:6.096111073240923e-05 tokens_per_second_per_gpu:4202.555713062383 mfu:19.824172328457696 +Step 15645 | loss:0.7453861236572266 lr:6.0956879948233355e-05 tokens_per_second_per_gpu:5323.242562540644 mfu:25.110643406340007 +Step 15646 | loss:0.7013974189758301 lr:6.095264908164937e-05 tokens_per_second_per_gpu:3153.8691497274785 mfu:14.877338884829147 +Step 15647 | loss:0.6666191816329956 lr:6.0948418132689124e-05 tokens_per_second_per_gpu:4610.126147004449 mfu:21.746751604049052 +Step 15648 | loss:0.7834433317184448 lr:6.094418710138442e-05 tokens_per_second_per_gpu:5248.539024200247 mfu:24.758254070250416 +Step 15649 | loss:0.7819276452064514 lr:6.0939955987767105e-05 tokens_per_second_per_gpu:3926.331436353112 mfu:18.521175286498696 +Step 15650 | loss:0.7245180010795593 lr:6.093572479186896e-05 tokens_per_second_per_gpu:5347.3026881520345 mfu:25.22413912392967 +Step 15651 | loss:0.8539300560951233 lr:6.093149351372186e-05 tokens_per_second_per_gpu:4259.654939885149 mfu:20.093519123513357 +Step 15652 | loss:0.8215987086296082 lr:6.092726215335759e-05 tokens_per_second_per_gpu:4068.9726992123165 mfu:19.194038460514573 +Step 15653 | loss:0.8845220804214478 lr:6.0923030710807994e-05 tokens_per_second_per_gpu:4753.2564767897475 mfu:22.421921790198734 +Step 15654 | loss:0.743927538394928 lr:6.091879918610488e-05 tokens_per_second_per_gpu:4347.654365162361 mfu:20.508627426796135 +Step 15655 | loss:0.7777915596961975 lr:6.091456757928008e-05 tokens_per_second_per_gpu:4448.302147066471 mfu:20.983400186320164 +Step 15656 | loss:0.8259071707725525 lr:6.0910335890365446e-05 tokens_per_second_per_gpu:3623.0418810638434 mfu:17.09050670766515 +Step 15657 | loss:0.7523787021636963 lr:6.090610411939276e-05 tokens_per_second_per_gpu:5332.875573137291 mfu:25.156083960885482 +Step 15658 | loss:0.9184946417808533 lr:6.090187226639389e-05 tokens_per_second_per_gpu:4702.719216869742 mfu:22.1835289126104 +Step 15659 | loss:0.8731467127799988 lr:6.089764033140064e-05 tokens_per_second_per_gpu:3702.316608925736 mfu:17.464459124652993 +Step 15660 | loss:0.7073074579238892 lr:6.0893408314444845e-05 tokens_per_second_per_gpu:3776.2768641247935 mfu:17.813342267347707 +Step 15661 | loss:0.7594594955444336 lr:6.088917621555834e-05 tokens_per_second_per_gpu:5306.078126972637 mfu:25.02967583521103 +Step 15662 | loss:1.0498005151748657 lr:6.0884944034772936e-05 tokens_per_second_per_gpu:4546.162987497925 mfu:21.44502646741643 +Step 15663 | loss:0.7994101643562317 lr:6.088071177212051e-05 tokens_per_second_per_gpu:5334.456477740915 mfu:25.163541357630912 +Step 15664 | loss:0.7756951451301575 lr:6.087647942763284e-05 tokens_per_second_per_gpu:5367.552838910688 mfu:25.319662540090228 +Step 15665 | loss:0.783714234828949 lr:6.0872247001341787e-05 tokens_per_second_per_gpu:4091.7116763525423 mfu:19.301302095354576 +Step 15666 | loss:0.6970751285552979 lr:6.086801449327917e-05 tokens_per_second_per_gpu:4143.217595666453 mfu:19.5442643045743 +Step 15667 | loss:0.6321946382522583 lr:6.0863781903476844e-05 tokens_per_second_per_gpu:5344.807294450714 mfu:25.21236792608247 +Step 15668 | loss:0.8300683498382568 lr:6.08595492319666e-05 tokens_per_second_per_gpu:4839.1416459428865 mfu:22.82705678661497 +Step 15669 | loss:0.8548742532730103 lr:6.0855316478780336e-05 tokens_per_second_per_gpu:5186.593585062393 mfu:24.4660468648554 +Step 15670 | loss:0.8296743035316467 lr:6.0851083643949837e-05 tokens_per_second_per_gpu:5355.173091436944 mfu:25.261265159053856 +Step 15671 | loss:0.822392463684082 lr:6.0846850727506954e-05 tokens_per_second_per_gpu:3506.5606239626472 mfu:16.54104474416703 +Step 15672 | loss:0.7099769711494446 lr:6.084261772948352e-05 tokens_per_second_per_gpu:4671.654620434041 mfu:22.03699191105651 +Step 15673 | loss:0.7422515153884888 lr:6.0838384649911365e-05 tokens_per_second_per_gpu:3954.8318237923545 mfu:18.655616476717125 +Step 15674 | loss:0.8035284876823425 lr:6.0834151488822355e-05 tokens_per_second_per_gpu:5266.810174392044 mfu:24.8444422030848 +Step 15675 | loss:0.7883080244064331 lr:6.082991824624829e-05 tokens_per_second_per_gpu:3501.190959493374 mfu:16.51571512070624 +Step 15676 | loss:0.6808274984359741 lr:6.082568492222104e-05 tokens_per_second_per_gpu:5091.566960039975 mfu:24.017790061410576 +Step 15677 | loss:0.779967188835144 lr:6.082145151677243e-05 tokens_per_second_per_gpu:4760.687769885337 mfu:22.456976467639972 +Step 15678 | loss:0.8037226796150208 lr:6.081721802993431e-05 tokens_per_second_per_gpu:5058.23605566252 mfu:23.860562498623427 +Step 15679 | loss:0.9477711319923401 lr:6.0812984461738495e-05 tokens_per_second_per_gpu:3543.514501874257 mfu:16.715362491257814 +Step 15680 | loss:0.7235035300254822 lr:6.0808750812216853e-05 tokens_per_second_per_gpu:4750.3671796361705 mfu:22.408292482560384 +Step 15681 | loss:0.7848145365715027 lr:6.080451708140124e-05 tokens_per_second_per_gpu:5046.046016670962 mfu:23.803060004865127 +Step 15682 | loss:0.776652991771698 lr:6.080028326932343e-05 tokens_per_second_per_gpu:5326.6130682549 mfu:25.126542656861726 +Step 15683 | loss:0.8098904490470886 lr:6.079604937601534e-05 tokens_per_second_per_gpu:5267.140200855273 mfu:24.845998994220174 +Step 15684 | loss:0.8381221890449524 lr:6.079181540150878e-05 tokens_per_second_per_gpu:4805.678543134645 mfu:22.669205621274422 +Step 15685 | loss:0.7213253974914551 lr:6.078758134583561e-05 tokens_per_second_per_gpu:4235.637736634878 mfu:19.980225878024406 +Step 15686 | loss:0.8699285387992859 lr:6.078334720902764e-05 tokens_per_second_per_gpu:3881.444954943396 mfu:18.309438095265627 +Step 15687 | loss:0.8309493660926819 lr:6.0779112991116763e-05 tokens_per_second_per_gpu:5287.32171078788 mfu:24.941198619892937 +Step 15688 | loss:0.7778602242469788 lr:6.07748786921348e-05 tokens_per_second_per_gpu:4454.392155557961 mfu:21.012127795437117 +Step 15689 | loss:0.814192533493042 lr:6.077064431211358e-05 tokens_per_second_per_gpu:5365.496623866798 mfu:25.309963022901734 +Step 15690 | loss:0.7558236718177795 lr:6.076640985108498e-05 tokens_per_second_per_gpu:5338.890508987161 mfu:25.1844574395432 +Step 15691 | loss:0.8030465245246887 lr:6.0762175309080836e-05 tokens_per_second_per_gpu:3753.112107791491 mfu:17.704070159408538 +Step 15692 | loss:0.8836726546287537 lr:6.075794068613301e-05 tokens_per_second_per_gpu:3382.263719659869 mfu:15.954715039332129 +Step 15693 | loss:0.843061625957489 lr:6.0753705982273325e-05 tokens_per_second_per_gpu:4299.0399541970455 mfu:20.27930495579893 +Step 15694 | loss:0.7790748476982117 lr:6.0749471197533646e-05 tokens_per_second_per_gpu:5360.0990421558 mfu:25.28450170904856 +Step 15695 | loss:0.8767417073249817 lr:6.074523633194583e-05 tokens_per_second_per_gpu:4432.387941003162 mfu:20.908330161075995 +Step 15696 | loss:0.7032161951065063 lr:6.074100138554172e-05 tokens_per_second_per_gpu:3915.7587026566493 mfu:18.471301897758103 +Step 15697 | loss:0.8280308842658997 lr:6.073676635835317e-05 tokens_per_second_per_gpu:4772.516196829169 mfu:22.512773175671768 +Step 15698 | loss:0.8314573168754578 lr:6.0732531250412016e-05 tokens_per_second_per_gpu:3489.6953360053753 mfu:16.461488303357733 +Step 15699 | loss:0.8554748296737671 lr:6.072829606175015e-05 tokens_per_second_per_gpu:5316.868853613264 mfu:25.080577533862677 +Step 15700 | loss:0.8328112363815308 lr:6.072406079239937e-05 tokens_per_second_per_gpu:3180.7461843533997 mfu:15.004122442854586 +Step 15701 | loss:0.8048145174980164 lr:6.071982544239159e-05 tokens_per_second_per_gpu:4534.591832689493 mfu:21.390443355942466 +Step 15702 | loss:0.7354689240455627 lr:6.0715590011758604e-05 tokens_per_second_per_gpu:4382.600461623182 mfu:20.67347412622071 +Step 15703 | loss:0.9627882838249207 lr:6.0711354500532325e-05 tokens_per_second_per_gpu:5322.846931213664 mfu:25.108777145868977 +Step 15704 | loss:0.7529031038284302 lr:6.0707118908744565e-05 tokens_per_second_per_gpu:4197.1353742174315 mfu:19.79860366532089 +Step 15705 | loss:0.8324065208435059 lr:6.070288323642721e-05 tokens_per_second_per_gpu:4066.0460962958564 mfu:19.18023317522768 +Step 15706 | loss:0.7819619178771973 lr:6.069864748361209e-05 tokens_per_second_per_gpu:5319.670903943744 mfu:25.093795283350637 +Step 15707 | loss:0.7587379813194275 lr:6.06944116503311e-05 tokens_per_second_per_gpu:4627.262404554636 mfu:21.82758625466031 +Step 15708 | loss:0.7429401278495789 lr:6.069017573661606e-05 tokens_per_second_per_gpu:5187.566065471241 mfu:24.47063422086621 +Step 15709 | loss:0.9584057927131653 lr:6.068593974249883e-05 tokens_per_second_per_gpu:4356.4552338105195 mfu:20.550142625792528 +Step 15710 | loss:0.9662442803382874 lr:6.0681703668011314e-05 tokens_per_second_per_gpu:5310.94194706647 mfu:25.052619304446917 +Step 15711 | loss:0.7904471158981323 lr:6.067746751318531e-05 tokens_per_second_per_gpu:4614.713444655753 mfu:21.76839066974394 +Step 15712 | loss:0.7034952640533447 lr:6.067323127805273e-05 tokens_per_second_per_gpu:5313.467788668913 mfu:25.064534130238663 +Step 15713 | loss:0.727226197719574 lr:6.066899496264542e-05 tokens_per_second_per_gpu:5281.880911172309 mfu:24.91553343981015 +Step 15714 | loss:0.839759886264801 lr:6.0664758566995236e-05 tokens_per_second_per_gpu:4807.86874622998 mfu:22.679537182962303 +Step 15715 | loss:0.7419220209121704 lr:6.066052209113404e-05 tokens_per_second_per_gpu:4124.876804792076 mfu:19.45774766475826 +Step 15716 | loss:0.927385687828064 lr:6.065628553509368e-05 tokens_per_second_per_gpu:4220.271163563869 mfu:19.907739131040735 +Step 15717 | loss:0.6744986176490784 lr:6.0652048898906064e-05 tokens_per_second_per_gpu:5348.703846228526 mfu:25.2307486256387 +Step 15718 | loss:0.8159910440444946 lr:6.0647812182603e-05 tokens_per_second_per_gpu:3242.604718633378 mfu:15.295919703207502 +Step 15719 | loss:0.7093009948730469 lr:6.064357538621641e-05 tokens_per_second_per_gpu:5334.495136936051 mfu:25.163723719650307 +Step 15720 | loss:0.8415170311927795 lr:6.063933850977811e-05 tokens_per_second_per_gpu:5132.193246739201 mfu:24.209431187330097 +Step 15721 | loss:0.74434494972229 lr:6.0635101553320005e-05 tokens_per_second_per_gpu:4937.003951452321 mfu:23.288689978734038 +Step 15722 | loss:0.8161919116973877 lr:6.063086451687394e-05 tokens_per_second_per_gpu:4111.6171510123895 mfu:19.39519961554926 +Step 15723 | loss:0.855849027633667 lr:6.062662740047177e-05 tokens_per_second_per_gpu:4701.3474078249465 mfu:22.177057855291338 +Step 15724 | loss:0.7802065014839172 lr:6.062239020414539e-05 tokens_per_second_per_gpu:5341.892498504911 mfu:25.19861833629065 +Step 15725 | loss:0.681222140789032 lr:6.0618152927926664e-05 tokens_per_second_per_gpu:5261.313216257628 mfu:24.81851211368699 +Step 15726 | loss:0.7347578406333923 lr:6.061391557184745e-05 tokens_per_second_per_gpu:3800.396748857668 mfu:17.927119878909366 +Step 15727 | loss:0.8312439322471619 lr:6.060967813593961e-05 tokens_per_second_per_gpu:4515.888855581306 mfu:21.302218221866205 +Step 15728 | loss:0.7573580145835876 lr:6.0605440620235055e-05 tokens_per_second_per_gpu:5318.987176579392 mfu:25.09057002471707 +Step 15729 | loss:0.8090350031852722 lr:6.060120302476561e-05 tokens_per_second_per_gpu:3321.776654963087 mfu:15.669387234999991 +Step 15730 | loss:0.9008045196533203 lr:6.059696534956316e-05 tokens_per_second_per_gpu:4197.256738323573 mfu:19.79917616051641 +Step 15731 | loss:0.7633611559867859 lr:6.059272759465959e-05 tokens_per_second_per_gpu:5383.723939638247 mfu:25.395944381296417 +Step 15732 | loss:0.7036054134368896 lr:6.0588489760086764e-05 tokens_per_second_per_gpu:4939.198050305447 mfu:23.299039917376103 +Step 15733 | loss:0.9778113961219788 lr:6.058425184587656e-05 tokens_per_second_per_gpu:4133.171303541799 mfu:19.496874230548116 +Step 15734 | loss:0.8608722686767578 lr:6.058001385206082e-05 tokens_per_second_per_gpu:3899.6718290738218 mfu:18.395417370364246 +Step 15735 | loss:0.8438146114349365 lr:6.057577577867148e-05 tokens_per_second_per_gpu:5314.280313895426 mfu:25.06836695036322 +Step 15736 | loss:0.9782078862190247 lr:6.0571537625740374e-05 tokens_per_second_per_gpu:5026.335519418367 mfu:23.710082226367028 +Step 15737 | loss:0.7752198576927185 lr:6.056729939329937e-05 tokens_per_second_per_gpu:3841.364472227343 mfu:18.120371619858496 +Step 15738 | loss:0.9533308744430542 lr:6.0563061081380354e-05 tokens_per_second_per_gpu:3797.1120350422743 mfu:17.911625323412185 +Step 15739 | loss:0.6944987773895264 lr:6.055882269001524e-05 tokens_per_second_per_gpu:5348.004303711701 mfu:25.227448764232573 +Step 15740 | loss:0.6918004155158997 lr:6.055458421923586e-05 tokens_per_second_per_gpu:5338.157773411159 mfu:25.181001000813612 +Step 15741 | loss:0.8348729014396667 lr:6.055034566907408e-05 tokens_per_second_per_gpu:3165.9572375204225 mfu:14.934360457388914 +Step 15742 | loss:0.7371174097061157 lr:6.054610703956183e-05 tokens_per_second_per_gpu:5366.731967005992 mfu:25.31579034726059 +Step 15743 | loss:0.7366846203804016 lr:6.0541868330730956e-05 tokens_per_second_per_gpu:5224.496711872224 mfu:24.644842380957662 +Step 15744 | loss:0.8555288910865784 lr:6.053762954261335e-05 tokens_per_second_per_gpu:5337.77695341484 mfu:25.179204607916382 +Step 15745 | loss:0.8897833228111267 lr:6.0533390675240874e-05 tokens_per_second_per_gpu:4554.307709685624 mfu:21.483446511608705 +Step 15746 | loss:0.722195565700531 lr:6.052915172864544e-05 tokens_per_second_per_gpu:5359.056456461372 mfu:25.27958365444448 +Step 15747 | loss:0.7789967060089111 lr:6.052491270285889e-05 tokens_per_second_per_gpu:3666.5521113435625 mfu:17.295751887505368 +Step 15748 | loss:0.8335297703742981 lr:6.0520673597913155e-05 tokens_per_second_per_gpu:5343.310889885868 mfu:25.205309130437147 +Step 15749 | loss:0.9422677755355835 lr:6.0516434413840086e-05 tokens_per_second_per_gpu:4780.169154820535 mfu:22.54887348424632 +Step 15750 | loss:0.8686455488204956 lr:6.0512195150671555e-05 tokens_per_second_per_gpu:4351.011387201619 mfu:20.524463072522337 +Step 15751 | loss:0.7560898661613464 lr:6.050795580843948e-05 tokens_per_second_per_gpu:4048.031824920594 mfu:19.095256783598938 +Step 15752 | loss:0.7832154631614685 lr:6.0503716387175716e-05 tokens_per_second_per_gpu:4531.262560370379 mfu:21.374738610380778 +Step 15753 | loss:0.7613879442214966 lr:6.0499476886912166e-05 tokens_per_second_per_gpu:2728.349007130036 mfu:12.870087770974283 +Step 15754 | loss:0.7741400599479675 lr:6.049523730768071e-05 tokens_per_second_per_gpu:4757.716136112123 mfu:22.44295876411872 +Step 15755 | loss:0.7309145927429199 lr:6.049099764951324e-05 tokens_per_second_per_gpu:5325.517979525658 mfu:25.121376936484488 +Step 15756 | loss:0.7700766324996948 lr:6.048675791244164e-05 tokens_per_second_per_gpu:5318.333862627013 mfu:25.087488231336845 +Step 15757 | loss:0.7332519888877869 lr:6.0482518096497786e-05 tokens_per_second_per_gpu:3543.468548018778 mfu:16.715145719080986 +Step 15758 | loss:0.928030252456665 lr:6.047827820171358e-05 tokens_per_second_per_gpu:4383.646896771617 mfu:20.678410339356127 +Step 15759 | loss:0.7672415971755981 lr:6.04740382281209e-05 tokens_per_second_per_gpu:4879.728348846769 mfu:23.018511189018163 +Step 15760 | loss:0.7289606928825378 lr:6.046979817575165e-05 tokens_per_second_per_gpu:5245.855375599241 mfu:24.745594841921804 +Step 15761 | loss:0.8514431118965149 lr:6.046555804463771e-05 tokens_per_second_per_gpu:4083.140350553311 mfu:19.26086968914062 +Step 15762 | loss:0.7082573175430298 lr:6.046131783481097e-05 tokens_per_second_per_gpu:4647.491165031953 mfu:21.923008769214245 +Step 15763 | loss:0.867115318775177 lr:6.045707754630332e-05 tokens_per_second_per_gpu:4483.239719727128 mfu:21.148206677525057 +Step 15764 | loss:0.8567848205566406 lr:6.045283717914667e-05 tokens_per_second_per_gpu:5300.975174013665 mfu:25.005604335450005 +Step 15765 | loss:0.9141393899917603 lr:6.044859673337289e-05 tokens_per_second_per_gpu:3729.1887786565653 mfu:17.591219734138246 +Step 15766 | loss:0.8310065865516663 lr:6.044435620901388e-05 tokens_per_second_per_gpu:4738.839302157433 mfu:22.353913517634464 +Step 15767 | loss:0.8310428857803345 lr:6.044011560610153e-05 tokens_per_second_per_gpu:3902.3238877074546 mfu:18.407927583426954 +Step 15768 | loss:0.8626391291618347 lr:6.043587492466774e-05 tokens_per_second_per_gpu:3748.191732543845 mfu:17.680859909862807 +Step 15769 | loss:0.8215771913528442 lr:6.043163416474441e-05 tokens_per_second_per_gpu:5298.738888125174 mfu:24.995055393364275 +Step 15770 | loss:0.8388451337814331 lr:6.0427393326363425e-05 tokens_per_second_per_gpu:5346.6641569794265 mfu:25.221127063443827 +Step 15771 | loss:0.8350957632064819 lr:6.042315240955668e-05 tokens_per_second_per_gpu:4266.559687235674 mfu:20.126089994837155 +Step 15772 | loss:0.7049376964569092 lr:6.0418911414356085e-05 tokens_per_second_per_gpu:4733.401646786276 mfu:22.328263169489045 +Step 15773 | loss:0.8892287611961365 lr:6.041467034079352e-05 tokens_per_second_per_gpu:4939.557905368672 mfu:23.300737415106937 +Step 15774 | loss:0.6778444647789001 lr:6.0410429188900896e-05 tokens_per_second_per_gpu:4759.303400095958 mfu:22.45044615914563 +Step 15775 | loss:0.9429308176040649 lr:6.04061879587101e-05 tokens_per_second_per_gpu:4701.088310994381 mfu:22.17583565133508 +Step 15776 | loss:0.824644923210144 lr:6.040194665025304e-05 tokens_per_second_per_gpu:3706.4202606582294 mfu:17.483816750030346 +Step 15777 | loss:0.8816337585449219 lr:6.03977052635616e-05 tokens_per_second_per_gpu:5357.82742612813 mfu:25.27378610866838 +Step 15778 | loss:0.877091109752655 lr:6.039346379866771e-05 tokens_per_second_per_gpu:5299.138766705234 mfu:24.99694168885274 +Step 15779 | loss:0.90804123878479 lr:6.038922225560324e-05 tokens_per_second_per_gpu:5314.967849638442 mfu:25.07161017376872 +Step 15780 | loss:0.8076745271682739 lr:6.0384980634400116e-05 tokens_per_second_per_gpu:3672.2856622536747 mfu:17.322798025393027 +Step 15781 | loss:0.8521165251731873 lr:6.038073893509021e-05 tokens_per_second_per_gpu:5342.016002778013 mfu:25.199200927018875 +Step 15782 | loss:0.7753990292549133 lr:6.037649715770547e-05 tokens_per_second_per_gpu:5354.093728398106 mfu:25.25617361944133 +Step 15783 | loss:0.8898763656616211 lr:6.0372255302277736e-05 tokens_per_second_per_gpu:5324.184937786967 mfu:25.11508875116271 +Step 15784 | loss:0.7952072620391846 lr:6.036801336883897e-05 tokens_per_second_per_gpu:4586.577109876524 mfu:21.635666778036587 +Step 15785 | loss:0.7622714042663574 lr:6.0363771357421036e-05 tokens_per_second_per_gpu:3666.9444213765305 mfu:17.297602480865873 +Step 15786 | loss:0.7644387483596802 lr:6.035952926805588e-05 tokens_per_second_per_gpu:4831.215057721423 mfu:22.789665717558925 +Step 15787 | loss:0.7054963707923889 lr:6.035528710077537e-05 tokens_per_second_per_gpu:4502.108950620159 mfu:21.23721605907025 +Step 15788 | loss:0.8094651699066162 lr:6.0351044855611405e-05 tokens_per_second_per_gpu:4262.668433887425 mfu:20.107734288877186 +Step 15789 | loss:0.8709686994552612 lr:6.0346802532595935e-05 tokens_per_second_per_gpu:5286.525022169152 mfu:24.937440503749304 +Step 15790 | loss:0.8474724292755127 lr:6.034256013176084e-05 tokens_per_second_per_gpu:5313.037463407671 mfu:25.06250421256032 +Step 15791 | loss:0.735975444316864 lr:6.033831765313802e-05 tokens_per_second_per_gpu:4376.443536228829 mfu:20.64443085865575 +Step 15792 | loss:0.92311692237854 lr:6.03340750967594e-05 tokens_per_second_per_gpu:4990.173141543448 mfu:23.539498119993144 +Step 15793 | loss:0.7209487557411194 lr:6.032983246265689e-05 tokens_per_second_per_gpu:5313.688346878568 mfu:25.065574540944713 +Step 15794 | loss:0.7293911576271057 lr:6.0325589750862385e-05 tokens_per_second_per_gpu:5349.453248135402 mfu:25.23428368229458 +Step 15795 | loss:0.8560996651649475 lr:6.0321346961407786e-05 tokens_per_second_per_gpu:5309.061789729823 mfu:25.04375028150204 +Step 15796 | loss:0.7508699893951416 lr:6.0317104094325037e-05 tokens_per_second_per_gpu:3821.174923475686 mfu:18.025134073704706 +Step 15797 | loss:0.9114519357681274 lr:6.031286114964604e-05 tokens_per_second_per_gpu:4839.369623343919 mfu:22.828132194912595 +Step 15798 | loss:0.7623690962791443 lr:6.030861812740268e-05 tokens_per_second_per_gpu:4898.095723232224 mfu:23.105153227791575 +Step 15799 | loss:0.7605600357055664 lr:6.030437502762689e-05 tokens_per_second_per_gpu:4880.36896623872 mfu:23.021533090556595 +Step 15800 | loss:0.9725134968757629 lr:6.030013185035059e-05 tokens_per_second_per_gpu:5233.923608422667 mfu:24.689310660380503 +Step 15801 | loss:0.8551118969917297 lr:6.0295888595605685e-05 tokens_per_second_per_gpu:5345.919664160577 mfu:25.217615163793738 +Step 15802 | loss:0.6850273013114929 lr:6.029164526342408e-05 tokens_per_second_per_gpu:4388.909764815947 mfu:20.70323618585911 +Step 15803 | loss:0.7571665048599243 lr:6.028740185383771e-05 tokens_per_second_per_gpu:5031.920175300495 mfu:23.736426001003576 +Step 15804 | loss:0.8761186599731445 lr:6.028315836687847e-05 tokens_per_second_per_gpu:5211.297496619868 mfu:24.582579430593952 +Step 15805 | loss:0.7801166772842407 lr:6.027891480257828e-05 tokens_per_second_per_gpu:5335.525638291035 mfu:25.168584770363427 +Step 15806 | loss:0.773701012134552 lr:6.027467116096905e-05 tokens_per_second_per_gpu:5309.078539143795 mfu:25.043829291345485 +Step 15807 | loss:0.8409867882728577 lr:6.027042744208273e-05 tokens_per_second_per_gpu:5322.393493615093 mfu:25.106638203352084 +Step 15808 | loss:0.749518871307373 lr:6.0266183645951204e-05 tokens_per_second_per_gpu:3830.8829704669047 mfu:18.070928587725902 +Step 15809 | loss:0.6751284003257751 lr:6.0261939772606405e-05 tokens_per_second_per_gpu:5324.458432832833 mfu:25.116378873957128 +Step 15810 | loss:0.7322966456413269 lr:6.025769582208024e-05 tokens_per_second_per_gpu:3594.884110908653 mfu:16.957681701632133 +Step 15811 | loss:0.8612156510353088 lr:6.025345179440465e-05 tokens_per_second_per_gpu:3297.316677719542 mfu:15.554005349039736 +Step 15812 | loss:0.7971458435058594 lr:6.0249207689611533e-05 tokens_per_second_per_gpu:4495.750250311284 mfu:21.20722098480714 +Step 15813 | loss:0.8227222561836243 lr:6.02449635077328e-05 tokens_per_second_per_gpu:5371.205542184776 mfu:25.336892964649547 +Step 15814 | loss:0.7789903283119202 lr:6.02407192488004e-05 tokens_per_second_per_gpu:5311.433222785679 mfu:25.05493673583471 +Step 15815 | loss:0.7662215232849121 lr:6.023647491284625e-05 tokens_per_second_per_gpu:5348.686829595072 mfu:25.230668355275355 +Step 15816 | loss:0.821359395980835 lr:6.0232230499902275e-05 tokens_per_second_per_gpu:4884.5121197170565 mfu:23.041077052408784 +Step 15817 | loss:0.830883264541626 lr:6.022798601000037e-05 tokens_per_second_per_gpu:3790.2389122176423 mfu:17.87920363037272 +Step 15818 | loss:0.8075814843177795 lr:6.022374144317248e-05 tokens_per_second_per_gpu:4396.602834864112 mfu:20.739525709849403 +Step 15819 | loss:0.812828540802002 lr:6.021949679945054e-05 tokens_per_second_per_gpu:4577.071276451689 mfu:21.59082613990922 +Step 15820 | loss:0.8422794938087463 lr:6.0215252078866446e-05 tokens_per_second_per_gpu:5149.140368618533 mfu:24.289373652712058 +Step 15821 | loss:0.7704783082008362 lr:6.021100728145215e-05 tokens_per_second_per_gpu:4646.764633611467 mfu:21.91958159654477 +Step 15822 | loss:0.805167019367218 lr:6.020676240723956e-05 tokens_per_second_per_gpu:5367.034437535663 mfu:25.317217152355838 +Step 15823 | loss:0.7486422061920166 lr:6.02025174562606e-05 tokens_per_second_per_gpu:5324.056234687299 mfu:25.114481636682566 +Step 15824 | loss:0.7482255101203918 lr:6.019827242854721e-05 tokens_per_second_per_gpu:5336.939223754816 mfu:25.175252894178637 +Step 15825 | loss:0.9646821618080139 lr:6.019402732413131e-05 tokens_per_second_per_gpu:4370.950694214396 mfu:20.618520185698188 +Step 15826 | loss:0.6341274380683899 lr:6.0189782143044825e-05 tokens_per_second_per_gpu:5336.0715127936655 mfu:25.17115975352813 +Step 15827 | loss:0.7617289423942566 lr:6.0185536885319704e-05 tokens_per_second_per_gpu:3176.394180227383 mfu:14.983593296863695 +Step 15828 | loss:0.9887814521789551 lr:6.018129155098785e-05 tokens_per_second_per_gpu:4277.282756295752 mfu:20.176672541137783 +Step 15829 | loss:0.7013806104660034 lr:6.017704614008121e-05 tokens_per_second_per_gpu:3474.0778855613116 mfu:16.387818125001274 +Step 15830 | loss:0.81392502784729 lr:6.01728006526317e-05 tokens_per_second_per_gpu:5354.45663262959 mfu:25.25788550061909 +Step 15831 | loss:0.8515356183052063 lr:6.016855508867125e-05 tokens_per_second_per_gpu:5368.588626576373 mfu:25.324548527232867 +Step 15832 | loss:0.911754310131073 lr:6.016430944823182e-05 tokens_per_second_per_gpu:4276.380861860098 mfu:20.172418151205743 +Step 15833 | loss:0.7563697099685669 lr:6.016006373134531e-05 tokens_per_second_per_gpu:5307.563060197917 mfu:25.036680518590412 +Step 15834 | loss:0.7403094172477722 lr:6.015581793804367e-05 tokens_per_second_per_gpu:4031.841890082637 mfu:19.018886098680575 +Step 15835 | loss:0.6634121537208557 lr:6.015157206835881e-05 tokens_per_second_per_gpu:4980.79822400667 mfu:23.495275034446298 +Step 15836 | loss:0.9143190979957581 lr:6.01473261223227e-05 tokens_per_second_per_gpu:4734.54161072036 mfu:22.33364057386402 +Step 15837 | loss:0.8279430866241455 lr:6.0143080099967244e-05 tokens_per_second_per_gpu:3247.7843364525534 mfu:15.320352844194481 +Step 15838 | loss:0.9064867496490479 lr:6.0138834001324394e-05 tokens_per_second_per_gpu:4556.7247764223885 mfu:21.494848227800134 +Step 15839 | loss:0.7151849269866943 lr:6.013458782642608e-05 tokens_per_second_per_gpu:5348.08663158881 mfu:25.22783711887548 +Step 15840 | loss:0.8110673427581787 lr:6.013034157530424e-05 tokens_per_second_per_gpu:5269.563602122279 mfu:24.85743059147153 +Step 15841 | loss:0.7364603281021118 lr:6.012609524799081e-05 tokens_per_second_per_gpu:4516.3004839474115 mfu:21.304159943100167 +Step 15842 | loss:0.7539647221565247 lr:6.0121848844517716e-05 tokens_per_second_per_gpu:4349.61408520182 mfu:20.517871760584978 +Step 15843 | loss:0.7601765990257263 lr:6.011760236491691e-05 tokens_per_second_per_gpu:5344.636778413016 mfu:25.211563572839616 +Step 15844 | loss:0.8064873218536377 lr:6.011335580922033e-05 tokens_per_second_per_gpu:5339.884567163396 mfu:25.189146581564312 +Step 15845 | loss:0.9256161451339722 lr:6.010910917745991e-05 tokens_per_second_per_gpu:4535.223808030615 mfu:21.393424491452652 +Step 15846 | loss:0.7438569664955139 lr:6.010486246966759e-05 tokens_per_second_per_gpu:3725.300549501089 mfu:17.572878293811563 +Step 15847 | loss:0.8161831498146057 lr:6.010061568587529e-05 tokens_per_second_per_gpu:4534.065811476397 mfu:21.387962024131824 +Step 15848 | loss:0.7816897034645081 lr:6.0096368826115e-05 tokens_per_second_per_gpu:4587.697310986114 mfu:21.640950957796537 +Step 15849 | loss:0.7055368423461914 lr:6.009212189041861e-05 tokens_per_second_per_gpu:5344.414319007707 mfu:25.210514193869283 +Step 15850 | loss:0.8358588814735413 lr:6.0087874878818096e-05 tokens_per_second_per_gpu:4345.323738526296 mfu:20.49763346330791 +Step 15851 | loss:0.7847362160682678 lr:6.0083627791345385e-05 tokens_per_second_per_gpu:4874.8197646702365 mfu:22.99535655996664 +Step 15852 | loss:0.7078904509544373 lr:6.007938062803242e-05 tokens_per_second_per_gpu:5042.383257594588 mfu:23.785782145370735 +Step 15853 | loss:0.9176809787750244 lr:6.007513338891114e-05 tokens_per_second_per_gpu:5347.486186282794 mfu:25.22500471592047 +Step 15854 | loss:0.7456207275390625 lr:6.007088607401351e-05 tokens_per_second_per_gpu:5049.5068688048295 mfu:23.819385434862845 +Step 15855 | loss:0.8539726734161377 lr:6.006663868337146e-05 tokens_per_second_per_gpu:5273.617073179058 mfu:24.876551506039835 +Step 15856 | loss:0.6394873261451721 lr:6.0062391217016934e-05 tokens_per_second_per_gpu:4982.350363778931 mfu:23.5025967425744 +Step 15857 | loss:0.7196260094642639 lr:6.0058143674981885e-05 tokens_per_second_per_gpu:5343.014194925926 mfu:25.203909569689724 +Step 15858 | loss:0.933774471282959 lr:6.005389605729824e-05 tokens_per_second_per_gpu:5294.3408830865155 mfu:24.974309253219886 +Step 15859 | loss:0.778559148311615 lr:6.004964836399797e-05 tokens_per_second_per_gpu:3763.6274760735387 mfu:17.753672945701787 +Step 15860 | loss:0.7428101301193237 lr:6.0045400595113e-05 tokens_per_second_per_gpu:4550.390597102469 mfu:21.46496882322564 +Step 15861 | loss:0.7558798789978027 lr:6.00411527506753e-05 tokens_per_second_per_gpu:4808.56726116665 mfu:22.682832197097508 +Step 15862 | loss:0.7636786103248596 lr:6.0036904830716815e-05 tokens_per_second_per_gpu:4915.9955592610495 mfu:23.18958980836649 +Step 15863 | loss:0.8651649355888367 lr:6.0032656835269474e-05 tokens_per_second_per_gpu:4671.217453469103 mfu:22.03492971989452 +Step 15864 | loss:0.7758906483650208 lr:6.0028408764365254e-05 tokens_per_second_per_gpu:5349.276378213986 mfu:25.233449356698145 +Step 15865 | loss:0.7293192744255066 lr:6.0024160618036076e-05 tokens_per_second_per_gpu:5329.990821037867 mfu:25.142476093043037 +Step 15866 | loss:0.8335683345794678 lr:6.0019912396313926e-05 tokens_per_second_per_gpu:5334.57745058471 mfu:25.16411200717519 +Step 15867 | loss:0.6601551175117493 lr:6.001566409923072e-05 tokens_per_second_per_gpu:5084.787431222212 mfu:23.985809867270294 +Step 15868 | loss:0.9255646467208862 lr:6.001141572681844e-05 tokens_per_second_per_gpu:3165.664304234326 mfu:14.932978641098826 +Step 15869 | loss:0.7916538715362549 lr:6.0007167279109025e-05 tokens_per_second_per_gpu:4657.981143784658 mfu:21.972491788765687 +Step 15870 | loss:0.7190535664558411 lr:6.000291875613442e-05 tokens_per_second_per_gpu:5369.270348685424 mfu:25.327764326735046 +Step 15871 | loss:0.8180058002471924 lr:5.999867015792657e-05 tokens_per_second_per_gpu:5330.66229385597 mfu:25.145643544891104 +Step 15872 | loss:0.6731138229370117 lr:5.999442148451747e-05 tokens_per_second_per_gpu:3694.264887402931 mfu:17.426477780464573 +Step 15873 | loss:0.7962090969085693 lr:5.9990172735939044e-05 tokens_per_second_per_gpu:5040.849378584811 mfu:23.778546576374872 +Step 15874 | loss:0.7723588347434998 lr:5.998592391222326e-05 tokens_per_second_per_gpu:4423.732452833791 mfu:20.867500746602513 +Step 15875 | loss:0.7654369473457336 lr:5.998167501340206e-05 tokens_per_second_per_gpu:4592.9389340729485 mfu:21.665676588210253 +Step 15876 | loss:0.771809995174408 lr:5.997742603950742e-05 tokens_per_second_per_gpu:4411.211564144094 mfu:20.808437578369315 +Step 15877 | loss:0.7302297949790955 lr:5.997317699057128e-05 tokens_per_second_per_gpu:4798.277822267234 mfu:22.634295158249664 +Step 15878 | loss:0.8149248957633972 lr:5.9968927866625593e-05 tokens_per_second_per_gpu:3393.9224921709983 mfu:16.00971145845884 +Step 15879 | loss:0.8395832777023315 lr:5.9964678667702335e-05 tokens_per_second_per_gpu:5386.658002670412 mfu:25.40978485722191 +Step 15880 | loss:0.7489197254180908 lr:5.996042939383345e-05 tokens_per_second_per_gpu:5132.202275642543 mfu:24.20947377820765 +Step 15881 | loss:0.7203673720359802 lr:5.995618004505092e-05 tokens_per_second_per_gpu:4430.921005306853 mfu:20.901410375111467 +Step 15882 | loss:0.7357752323150635 lr:5.995193062138668e-05 tokens_per_second_per_gpu:5306.415902457828 mfu:25.031269179805115 +Step 15883 | loss:0.8107246160507202 lr:5.994768112287269e-05 tokens_per_second_per_gpu:3370.813834594219 mfu:15.900703977925621 +Step 15884 | loss:0.7882832288742065 lr:5.994343154954094e-05 tokens_per_second_per_gpu:3852.410618392523 mfu:18.172478175986527 +Step 15885 | loss:0.7434275150299072 lr:5.993918190142336e-05 tokens_per_second_per_gpu:5319.874987591134 mfu:25.094757980735636 +Step 15886 | loss:0.7939704656600952 lr:5.9934932178551925e-05 tokens_per_second_per_gpu:3853.0148220113156 mfu:18.175328307544138 +Step 15887 | loss:0.7859954833984375 lr:5.9930682380958614e-05 tokens_per_second_per_gpu:4318.688358861201 mfu:20.371989832963276 +Step 15888 | loss:0.7297141551971436 lr:5.9926432508675354e-05 tokens_per_second_per_gpu:5226.860685189538 mfu:24.655993646430872 +Step 15889 | loss:0.6761955618858337 lr:5.992218256173413e-05 tokens_per_second_per_gpu:4220.658936986927 mfu:19.90956832443827 +Step 15890 | loss:0.7549453377723694 lr:5.991793254016691e-05 tokens_per_second_per_gpu:5303.786389835806 mfu:25.018865320125926 +Step 15891 | loss:0.8039126992225647 lr:5.991368244400566e-05 tokens_per_second_per_gpu:5164.902279248629 mfu:24.36372527441348 +Step 15892 | loss:0.8025575280189514 lr:5.990943227328234e-05 tokens_per_second_per_gpu:5118.034461670903 mfu:24.142641782425095 +Step 15893 | loss:0.9812413454055786 lr:5.9905182028028904e-05 tokens_per_second_per_gpu:2858.96503097286 mfu:13.486225840832548 +Step 15894 | loss:0.6498364210128784 lr:5.990093170827733e-05 tokens_per_second_per_gpu:4202.67149199777 mfu:19.82471847744984 +Step 15895 | loss:0.7209716439247131 lr:5.9896681314059596e-05 tokens_per_second_per_gpu:5355.539277210905 mfu:25.262992519827367 +Step 15896 | loss:0.76548832654953 lr:5.989243084540765e-05 tokens_per_second_per_gpu:5330.405352240984 mfu:25.144431507454456 +Step 15897 | loss:0.8533036708831787 lr:5.988818030235347e-05 tokens_per_second_per_gpu:5347.629087934958 mfu:25.225678807395266 +Step 15898 | loss:0.7516071200370789 lr:5.988392968492903e-05 tokens_per_second_per_gpu:5295.041717977404 mfu:24.97761521097497 +Step 15899 | loss:0.7415885925292969 lr:5.9879678993166285e-05 tokens_per_second_per_gpu:4032.7229761679614 mfu:19.023042332073835 +Step 15900 | loss:0.8930887579917908 lr:5.987542822709723e-05 tokens_per_second_per_gpu:5265.570414784341 mfu:24.838594045490463 +Step 15901 | loss:0.6714825630187988 lr:5.9871177386753794e-05 tokens_per_second_per_gpu:4487.127707490631 mfu:21.166546979164615 +Step 15902 | loss:0.8364740014076233 lr:5.9866926472168005e-05 tokens_per_second_per_gpu:4773.92297036489 mfu:22.519409166460083 +Step 15903 | loss:0.7751321196556091 lr:5.9862675483371776e-05 tokens_per_second_per_gpu:3955.2201555659885 mfu:18.657448303949586 +Step 15904 | loss:0.7835943102836609 lr:5.985842442039712e-05 tokens_per_second_per_gpu:4420.896109836338 mfu:20.85412123275351 +Step 15905 | loss:0.7412389516830444 lr:5.985417328327599e-05 tokens_per_second_per_gpu:3759.3890010728187 mfu:17.733679335964986 +Step 15906 | loss:0.8953027129173279 lr:5.984992207204037e-05 tokens_per_second_per_gpu:4336.317886283707 mfu:20.455151321722777 +Step 15907 | loss:0.7942370772361755 lr:5.9845670786722216e-05 tokens_per_second_per_gpu:4579.049291064098 mfu:21.60015677232004 +Step 15908 | loss:0.7944566011428833 lr:5.984141942735353e-05 tokens_per_second_per_gpu:5228.31446540241 mfu:24.662851375735467 +Step 15909 | loss:0.6523569822311401 lr:5.983716799396627e-05 tokens_per_second_per_gpu:5300.562560699736 mfu:25.00365796804911 +Step 15910 | loss:0.7934399843215942 lr:5.983291648659241e-05 tokens_per_second_per_gpu:3976.4828475148956 mfu:18.757748049661913 +Step 15911 | loss:0.9182764291763306 lr:5.982866490526394e-05 tokens_per_second_per_gpu:3762.4692070366223 mfu:17.748209193033578 +Step 15912 | loss:0.7178760170936584 lr:5.982441325001281e-05 tokens_per_second_per_gpu:5328.724092003446 mfu:25.13650071606141 +Step 15913 | loss:0.7111250758171082 lr:5.9820161520871046e-05 tokens_per_second_per_gpu:3942.369712308267 mfu:18.596830570591212 +Step 15914 | loss:0.8277099132537842 lr:5.9815909717870556e-05 tokens_per_second_per_gpu:3599.7223835498594 mfu:16.98050465917514 +Step 15915 | loss:0.7889696359634399 lr:5.981165784104338e-05 tokens_per_second_per_gpu:4860.741888203142 mfu:22.92894881473795 +Step 15916 | loss:0.9317991137504578 lr:5.9807405890421464e-05 tokens_per_second_per_gpu:3374.0573497240484 mfu:15.916004192193775 +Step 15917 | loss:0.8039384484291077 lr:5.98031538660368e-05 tokens_per_second_per_gpu:5216.048196320397 mfu:24.6049892916343 +Step 15918 | loss:0.8285253643989563 lr:5.979890176792137e-05 tokens_per_second_per_gpu:4636.19567009441 mfu:21.869725992383692 +Step 15919 | loss:0.6996965408325195 lr:5.9794649596107134e-05 tokens_per_second_per_gpu:4652.2977883746125 mfu:21.94568243161608 +Step 15920 | loss:0.7751361727714539 lr:5.979039735062612e-05 tokens_per_second_per_gpu:5018.7812485863615 mfu:23.674447441960947 +Step 15921 | loss:0.8463894128799438 lr:5.9786145031510245e-05 tokens_per_second_per_gpu:4535.97157931922 mfu:21.396951856203938 +Step 15922 | loss:0.8314653038978577 lr:5.978189263879154e-05 tokens_per_second_per_gpu:5014.953254131654 mfu:23.65639013899489 +Step 15923 | loss:0.8997095227241516 lr:5.977764017250195e-05 tokens_per_second_per_gpu:3617.5505210758934 mfu:17.064603025679364 +Step 15924 | loss:0.8450072407722473 lr:5.977338763267352e-05 tokens_per_second_per_gpu:5363.305820168078 mfu:25.299628628065218 +Step 15925 | loss:0.7733601331710815 lr:5.976913501933816e-05 tokens_per_second_per_gpu:3656.8076584654923 mfu:17.249785640704413 +Step 15926 | loss:0.8946829438209534 lr:5.976488233252791e-05 tokens_per_second_per_gpu:5312.393101858154 mfu:25.05946464919183 +Step 15927 | loss:0.9290961623191833 lr:5.9760629572274715e-05 tokens_per_second_per_gpu:4495.0265837458555 mfu:21.203807326147764 +Step 15928 | loss:0.8221619129180908 lr:5.97563767386106e-05 tokens_per_second_per_gpu:3798.4381788612045 mfu:17.917880970069547 +Step 15929 | loss:0.9118111729621887 lr:5.975212383156752e-05 tokens_per_second_per_gpu:5331.635336284146 mfu:25.15023355204297 +Step 15930 | loss:0.7722798585891724 lr:5.974787085117747e-05 tokens_per_second_per_gpu:5306.314058246292 mfu:25.030788763282885 +Step 15931 | loss:0.7049806118011475 lr:5.974361779747245e-05 tokens_per_second_per_gpu:4618.803972047542 mfu:21.787686385366943 +Step 15932 | loss:0.7941374778747559 lr:5.9739364670484434e-05 tokens_per_second_per_gpu:4579.409422617662 mfu:21.60185557430298 +Step 15933 | loss:0.7506294250488281 lr:5.973511147024541e-05 tokens_per_second_per_gpu:4604.331389150346 mfu:21.719416742563606 +Step 15934 | loss:0.7397059798240662 lr:5.973085819678736e-05 tokens_per_second_per_gpu:4877.6438254659715 mfu:23.00867813657458 +Step 15935 | loss:0.7116870284080505 lr:5.9726604850142306e-05 tokens_per_second_per_gpu:5350.7707622609105 mfu:25.240498621215696 +Step 15936 | loss:0.8258015513420105 lr:5.9722351430342206e-05 tokens_per_second_per_gpu:4307.435129081971 mfu:20.318906428095666 +Step 15937 | loss:0.7512019872665405 lr:5.9718097937419045e-05 tokens_per_second_per_gpu:4211.2538869072805 mfu:19.865203098545564 +Step 15938 | loss:0.7223288416862488 lr:5.9713844371404856e-05 tokens_per_second_per_gpu:3978.707736344281 mfu:18.768243229875303 +Step 15939 | loss:0.8119657039642334 lr:5.970959073233159e-05 tokens_per_second_per_gpu:3992.709530688308 mfu:18.834292082748533 +Step 15940 | loss:0.7279084324836731 lr:5.9705337020231254e-05 tokens_per_second_per_gpu:4673.82435709225 mfu:22.047226929068525 +Step 15941 | loss:0.6841052770614624 lr:5.9701083235135835e-05 tokens_per_second_per_gpu:5351.055106071951 mfu:25.24183991948643 +Step 15942 | loss:0.7566417455673218 lr:5.9696829377077343e-05 tokens_per_second_per_gpu:4070.120378305236 mfu:19.19945225861012 +Step 15943 | loss:0.8920733332633972 lr:5.969257544608775e-05 tokens_per_second_per_gpu:4423.520179336397 mfu:20.866499416297707 +Step 15944 | loss:0.8071409463882446 lr:5.968832144219906e-05 tokens_per_second_per_gpu:4625.796023030259 mfu:21.820669082819336 +Step 15945 | loss:0.7542529106140137 lr:5.9684067365443274e-05 tokens_per_second_per_gpu:4052.8935518352414 mfu:19.11819037895148 +Step 15946 | loss:0.8544840812683105 lr:5.967981321585239e-05 tokens_per_second_per_gpu:5360.128096034696 mfu:25.28463876115241 +Step 15947 | loss:0.8643487691879272 lr:5.967555899345838e-05 tokens_per_second_per_gpu:5342.091855510327 mfu:25.19955873729901 +Step 15948 | loss:0.7593478560447693 lr:5.9671304698293254e-05 tokens_per_second_per_gpu:5350.216752489715 mfu:25.237885262601154 +Step 15949 | loss:0.8516243100166321 lr:5.966705033038903e-05 tokens_per_second_per_gpu:5353.220282532572 mfu:25.252053426268375 +Step 15950 | loss:0.8509296178817749 lr:5.9662795889777666e-05 tokens_per_second_per_gpu:5312.022674420078 mfu:25.057717279764987 +Step 15951 | loss:0.7095808386802673 lr:5.9658541376491194e-05 tokens_per_second_per_gpu:3825.256253814628 mfu:18.04438640526026 +Step 15952 | loss:0.988331139087677 lr:5.96542867905616e-05 tokens_per_second_per_gpu:4338.331769391978 mfu:20.464651151948384 +Step 15953 | loss:0.7436966896057129 lr:5.965003213202087e-05 tokens_per_second_per_gpu:5291.449032741952 mfu:24.96066790174403 +Step 15954 | loss:0.7509134411811829 lr:5.9645777400901034e-05 tokens_per_second_per_gpu:3532.6864106909243 mfu:16.664284537683173 +Step 15955 | loss:0.6632548570632935 lr:5.9641522597234056e-05 tokens_per_second_per_gpu:4133.733303355839 mfu:19.499525279558004 +Step 15956 | loss:0.7046672105789185 lr:5.9637267721051984e-05 tokens_per_second_per_gpu:5323.279550838502 mfu:25.11081788645186 +Step 15957 | loss:0.916312575340271 lr:5.963301277238676e-05 tokens_per_second_per_gpu:5014.342317355182 mfu:23.65350824598371 +Step 15958 | loss:0.7941195368766785 lr:5.962875775127045e-05 tokens_per_second_per_gpu:5318.553384298481 mfu:25.08852375251542 +Step 15959 | loss:0.916641354560852 lr:5.962450265773499e-05 tokens_per_second_per_gpu:4323.901060701335 mfu:20.396579036921832 +Step 15960 | loss:0.7943242788314819 lr:5.962024749181245e-05 tokens_per_second_per_gpu:4170.9793648198465 mfu:19.6752213063171 +Step 15961 | loss:0.7572721838951111 lr:5.961599225353478e-05 tokens_per_second_per_gpu:5285.447279321171 mfu:24.932356606853485 +Step 15962 | loss:0.7971662878990173 lr:5.961173694293401e-05 tokens_per_second_per_gpu:5335.174029501734 mfu:25.16692617171365 +Step 15963 | loss:0.9609556198120117 lr:5.9607481560042155e-05 tokens_per_second_per_gpu:5201.4272543954785 mfu:24.536019813984534 +Step 15964 | loss:0.7943527698516846 lr:5.960322610489118e-05 tokens_per_second_per_gpu:4940.130221293182 mfu:23.303437126970017 +Step 15965 | loss:0.8273979425430298 lr:5.9598970577513136e-05 tokens_per_second_per_gpu:5337.720717508359 mfu:25.178939333550677 +Step 15966 | loss:0.7576161623001099 lr:5.959471497793999e-05 tokens_per_second_per_gpu:5327.8461593834245 mfu:25.13235935810202 +Step 15967 | loss:0.7400140762329102 lr:5.959045930620379e-05 tokens_per_second_per_gpu:5335.492444176272 mfu:25.168428188060595 +Step 15968 | loss:0.8364279866218567 lr:5.9586203562336504e-05 tokens_per_second_per_gpu:5022.597182539256 mfu:23.69244785348222 +Step 15969 | loss:0.9114437103271484 lr:5.958194774637017e-05 tokens_per_second_per_gpu:4632.674233344186 mfu:21.853114774413825 +Step 15970 | loss:0.8290108442306519 lr:5.9577691858336784e-05 tokens_per_second_per_gpu:3579.2611279927482 mfu:16.88398542566206 +Step 15971 | loss:0.8511136770248413 lr:5.957343589826835e-05 tokens_per_second_per_gpu:4464.840238997359 mfu:21.061413187647613 +Step 15972 | loss:0.8063936829566956 lr:5.9569179866196886e-05 tokens_per_second_per_gpu:4430.819797648725 mfu:20.900932961320272 +Step 15973 | loss:0.8750948905944824 lr:5.956492376215439e-05 tokens_per_second_per_gpu:4774.784858364216 mfu:22.52347483920616 +Step 15974 | loss:0.8147870898246765 lr:5.9560667586172905e-05 tokens_per_second_per_gpu:4987.551060840597 mfu:23.527129317944695 +Step 15975 | loss:0.8302192091941833 lr:5.955641133828439e-05 tokens_per_second_per_gpu:4891.846426081536 mfu:23.07567422689174 +Step 15976 | loss:0.732613205909729 lr:5.95521550185209e-05 tokens_per_second_per_gpu:4094.259541414718 mfu:19.31332080956346 +Step 15977 | loss:0.6460315585136414 lr:5.9547898626914435e-05 tokens_per_second_per_gpu:4983.10788057766 mfu:23.506170078559702 +Step 15978 | loss:0.8255838751792908 lr:5.9543642163497015e-05 tokens_per_second_per_gpu:3748.6513553479012 mfu:17.68302802904928 +Step 15979 | loss:0.6828213334083557 lr:5.953938562830063e-05 tokens_per_second_per_gpu:4220.970397148544 mfu:19.911037535161203 +Step 15980 | loss:0.7768580913543701 lr:5.9535129021357303e-05 tokens_per_second_per_gpu:5322.559507838801 mfu:25.107421320770268 +Step 15981 | loss:0.873575747013092 lr:5.9530872342699074e-05 tokens_per_second_per_gpu:5337.931951232637 mfu:25.17993575906829 +Step 15982 | loss:0.8754891753196716 lr:5.952661559235792e-05 tokens_per_second_per_gpu:5006.526110598884 mfu:23.616637865131867 +Step 15983 | loss:0.8315176367759705 lr:5.9522358770365884e-05 tokens_per_second_per_gpu:4668.4549477191495 mfu:22.021898508939906 +Step 15984 | loss:0.8570101261138916 lr:5.951810187675496e-05 tokens_per_second_per_gpu:5332.184836880787 mfu:25.15282564011179 +Step 15985 | loss:0.7822608351707458 lr:5.9513844911557206e-05 tokens_per_second_per_gpu:5303.36943816704 mfu:25.016898487965065 +Step 15986 | loss:0.7612553834915161 lr:5.950958787480457e-05 tokens_per_second_per_gpu:5331.182429359115 mfu:25.148097112803708 +Step 15987 | loss:0.8426410555839539 lr:5.950533076652914e-05 tokens_per_second_per_gpu:4877.843326087508 mfu:23.009619215044683 +Step 15988 | loss:0.6464986205101013 lr:5.95010735867629e-05 tokens_per_second_per_gpu:5358.305249550821 mfu:25.276040083278293 +Step 15989 | loss:0.684928834438324 lr:5.949681633553787e-05 tokens_per_second_per_gpu:4137.6923926433465 mfu:19.51820097921746 +Step 15990 | loss:0.8111029863357544 lr:5.9492559012886076e-05 tokens_per_second_per_gpu:5337.706324688383 mfu:25.17887144016318 +Step 15991 | loss:0.8200523853302002 lr:5.9488301618839514e-05 tokens_per_second_per_gpu:4711.900779078389 mfu:22.226839908084635 +Step 15992 | loss:0.7891157269477844 lr:5.948404415343025e-05 tokens_per_second_per_gpu:5326.548403529638 mfu:25.126237622319724 +Step 15993 | loss:0.8634851574897766 lr:5.9479786616690266e-05 tokens_per_second_per_gpu:5208.0094319102045 mfu:24.56706906835678 +Step 15994 | loss:0.8084831237792969 lr:5.947552900865161e-05 tokens_per_second_per_gpu:3669.2777591896097 mfu:17.308609233438652 +Step 15995 | loss:0.7895463705062866 lr:5.9471271329346266e-05 tokens_per_second_per_gpu:4831.028684353657 mfu:22.788786562584594 +Step 15996 | loss:0.6727733016014099 lr:5.946701357880632e-05 tokens_per_second_per_gpu:5300.453288186555 mfu:25.003142510960622 +Step 15997 | loss:0.7670109272003174 lr:5.9462755757063736e-05 tokens_per_second_per_gpu:4945.324530416938 mfu:23.327939569347294 +Step 15998 | loss:0.7176010012626648 lr:5.945849786415055e-05 tokens_per_second_per_gpu:4651.629903401441 mfu:21.942531904244703 +Step 15999 | loss:0.6864349246025085 lr:5.945423990009882e-05 tokens_per_second_per_gpu:5069.282170457945 mfu:23.912668906774968 +Step 16000 | loss:0.7224134802818298 lr:5.944998186494052e-05 tokens_per_second_per_gpu:5120.112368064423 mfu:24.152443621410722 +Step 16001 | loss:0.8370484709739685 lr:5.944572375870771e-05 tokens_per_second_per_gpu:5331.405666298736 mfu:25.14915015953605 +Step 16002 | loss:0.8046025037765503 lr:5.94414655814324e-05 tokens_per_second_per_gpu:5362.900703129329 mfu:25.297717621873346 +Step 16003 | loss:0.7759287357330322 lr:5.943720733314664e-05 tokens_per_second_per_gpu:4070.252334759906 mfu:19.200074719720885 +Step 16004 | loss:0.8132181167602539 lr:5.943294901388241e-05 tokens_per_second_per_gpu:3851.427345610841 mfu:18.167839910511695 +Step 16005 | loss:0.7515964508056641 lr:5.942869062367179e-05 tokens_per_second_per_gpu:3785.8584136607137 mfu:17.858540071289568 +Step 16006 | loss:0.8499863743782043 lr:5.942443216254678e-05 tokens_per_second_per_gpu:3614.6333820110435 mfu:17.05084238299522 +Step 16007 | loss:0.7636774182319641 lr:5.9420173630539414e-05 tokens_per_second_per_gpu:4185.563006366269 mfu:19.744014831717404 +Step 16008 | loss:0.784984827041626 lr:5.9415915027681724e-05 tokens_per_second_per_gpu:3904.717185897787 mfu:18.41921717932953 +Step 16009 | loss:0.6847131252288818 lr:5.9411656354005716e-05 tokens_per_second_per_gpu:5180.997117994497 mfu:24.43964737484039 +Step 16010 | loss:0.7063981294631958 lr:5.940739760954347e-05 tokens_per_second_per_gpu:3712.177120480372 mfu:17.510972840032963 +Step 16011 | loss:0.9440247416496277 lr:5.940313879432695e-05 tokens_per_second_per_gpu:4276.821191047131 mfu:20.174495259108884 +Step 16012 | loss:0.7875827550888062 lr:5.939887990838825e-05 tokens_per_second_per_gpu:3596.239834081594 mfu:16.964076879149122 +Step 16013 | loss:0.7243714332580566 lr:5.939462095175937e-05 tokens_per_second_per_gpu:5106.234900811209 mfu:24.086981240637243 +Step 16014 | loss:0.7237529754638672 lr:5.939036192447235e-05 tokens_per_second_per_gpu:5312.713787720463 mfu:25.06097737911899 +Step 16015 | loss:0.6843719482421875 lr:5.938610282655921e-05 tokens_per_second_per_gpu:4020.910146613769 mfu:18.96731919959016 +Step 16016 | loss:0.730395495891571 lr:5.9381843658051994e-05 tokens_per_second_per_gpu:4967.807206047091 mfu:23.43399418623474 +Step 16017 | loss:0.8351826667785645 lr:5.9377584418982735e-05 tokens_per_second_per_gpu:4734.401625028516 mfu:22.33298023747157 +Step 16018 | loss:0.7164344787597656 lr:5.937332510938347e-05 tokens_per_second_per_gpu:4665.614073771351 mfu:22.008497621825065 +Step 16019 | loss:0.7964280247688293 lr:5.9369065729286245e-05 tokens_per_second_per_gpu:3220.0358439136203 mfu:15.189458470507313 +Step 16020 | loss:0.8443410992622375 lr:5.936480627872305e-05 tokens_per_second_per_gpu:4614.1723612537335 mfu:21.765838287018266 +Step 16021 | loss:0.7553979158401489 lr:5.936054675772599e-05 tokens_per_second_per_gpu:5370.853660496902 mfu:25.33523307869064 +Step 16022 | loss:0.9817075133323669 lr:5.935628716632703e-05 tokens_per_second_per_gpu:3999.265525502765 mfu:18.865217828856384 +Step 16023 | loss:0.9953662157058716 lr:5.935202750455825e-05 tokens_per_second_per_gpu:3792.326783067515 mfu:17.88905247339907 +Step 16024 | loss:0.8410260677337646 lr:5.934776777245169e-05 tokens_per_second_per_gpu:4824.221669293424 mfu:22.756676711148945 +Step 16025 | loss:0.663757860660553 lr:5.934350797003937e-05 tokens_per_second_per_gpu:4114.925676117054 mfu:19.41080649296055 +Step 16026 | loss:0.8969812393188477 lr:5.9339248097353326e-05 tokens_per_second_per_gpu:3636.2615246257906 mfu:17.152866021850627 +Step 16027 | loss:0.7590445280075073 lr:5.93349881544256e-05 tokens_per_second_per_gpu:5062.2524658474185 mfu:23.87950859073583 +Step 16028 | loss:0.8424261808395386 lr:5.933072814128826e-05 tokens_per_second_per_gpu:3979.1252320173658 mfu:18.77021262819735 +Step 16029 | loss:0.8966689109802246 lr:5.932646805797331e-05 tokens_per_second_per_gpu:4839.924509577775 mfu:22.830749687951048 +Step 16030 | loss:0.7232090830802917 lr:5.932220790451279e-05 tokens_per_second_per_gpu:4919.740790391595 mfu:23.207256702611595 +Step 16031 | loss:0.742518961429596 lr:5.931794768093878e-05 tokens_per_second_per_gpu:3651.3605173058477 mfu:17.22409054647625 +Step 16032 | loss:0.8319931626319885 lr:5.931368738728329e-05 tokens_per_second_per_gpu:4322.290332266955 mfu:20.388980956078253 +Step 16033 | loss:0.9082564115524292 lr:5.930942702357837e-05 tokens_per_second_per_gpu:4829.513206226872 mfu:22.781637793693317 +Step 16034 | loss:0.8817200660705566 lr:5.930516658985604e-05 tokens_per_second_per_gpu:3802.7620928613146 mfu:17.938277610143867 +Step 16035 | loss:0.8417935967445374 lr:5.930090608614838e-05 tokens_per_second_per_gpu:4709.927899904354 mfu:22.217533500412813 +Step 16036 | loss:0.9442474842071533 lr:5.9296645512487426e-05 tokens_per_second_per_gpu:4313.450582428213 mfu:20.34728234787206 +Step 16037 | loss:0.8714622855186462 lr:5.9292384868905206e-05 tokens_per_second_per_gpu:3727.448131960711 mfu:17.58300880668895 +Step 16038 | loss:0.729333758354187 lr:5.928812415543377e-05 tokens_per_second_per_gpu:4490.012177471316 mfu:21.180153516204832 +Step 16039 | loss:0.9956019520759583 lr:5.928386337210519e-05 tokens_per_second_per_gpu:4297.697876848434 mfu:20.272974147963527 +Step 16040 | loss:0.8041241765022278 lr:5.927960251895146e-05 tokens_per_second_per_gpu:3014.6934984924696 mfu:14.220823592138919 +Step 16041 | loss:0.8796492218971252 lr:5.927534159600466e-05 tokens_per_second_per_gpu:3363.5844575808514 mfu:15.86660177309466 +Step 16042 | loss:0.7168426513671875 lr:5.9271080603296846e-05 tokens_per_second_per_gpu:4466.312103774504 mfu:21.06835622492743 +Step 16043 | loss:0.9762015342712402 lr:5.926681954086004e-05 tokens_per_second_per_gpu:4468.655913548372 mfu:21.079412375525948 +Step 16044 | loss:0.8483469486236572 lr:5.926255840872631e-05 tokens_per_second_per_gpu:4947.683731893876 mfu:23.339068325235292 +Step 16045 | loss:0.8264588117599487 lr:5.925829720692768e-05 tokens_per_second_per_gpu:4693.136226472243 mfu:22.138324311878634 +Step 16046 | loss:0.7910041213035583 lr:5.925403593549624e-05 tokens_per_second_per_gpu:5330.265962665479 mfu:25.143773983044483 +Step 16047 | loss:0.7392823696136475 lr:5.9249774594464004e-05 tokens_per_second_per_gpu:4878.685370733745 mfu:23.013591283308614 +Step 16048 | loss:0.7664774656295776 lr:5.924551318386303e-05 tokens_per_second_per_gpu:5020.694472977864 mfu:23.683472447845407 +Step 16049 | loss:0.7914597988128662 lr:5.924125170372538e-05 tokens_per_second_per_gpu:4639.945239769594 mfu:21.88741335228409 +Step 16050 | loss:0.7844275236129761 lr:5.9236990154083094e-05 tokens_per_second_per_gpu:4868.787089229772 mfu:22.9668994006334 +Step 16051 | loss:0.7643857002258301 lr:5.923272853496822e-05 tokens_per_second_per_gpu:3614.4749600035652 mfu:17.050095079356364 +Step 16052 | loss:0.8527234792709351 lr:5.922846684641281e-05 tokens_per_second_per_gpu:5326.400697413463 mfu:25.125540867368464 +Step 16053 | loss:0.8003018498420715 lr:5.922420508844894e-05 tokens_per_second_per_gpu:5343.690414181139 mfu:25.207099411291466 +Step 16054 | loss:0.8387776613235474 lr:5.9219943261108636e-05 tokens_per_second_per_gpu:5342.840312693845 mfu:25.203089337533626 +Step 16055 | loss:0.8298526406288147 lr:5.921568136442397e-05 tokens_per_second_per_gpu:4871.950039237758 mfu:22.981819575475992 +Step 16056 | loss:0.7642244696617126 lr:5.921141939842697e-05 tokens_per_second_per_gpu:5330.665169020935 mfu:25.145657107534795 +Step 16057 | loss:0.7002198696136475 lr:5.9207157363149744e-05 tokens_per_second_per_gpu:5362.218268353647 mfu:25.294498460601076 +Step 16058 | loss:0.8200515508651733 lr:5.920289525862428e-05 tokens_per_second_per_gpu:4898.953147717344 mfu:23.109197845379587 +Step 16059 | loss:0.7657755613327026 lr:5.919863308488268e-05 tokens_per_second_per_gpu:4557.278691701931 mfu:21.497461140681427 +Step 16060 | loss:0.6703776121139526 lr:5.919437084195698e-05 tokens_per_second_per_gpu:5324.9018522003325 mfu:25.11847055876877 +Step 16061 | loss:0.8131030201911926 lr:5.919010852987925e-05 tokens_per_second_per_gpu:5302.095449029583 mfu:25.010888863838712 +Step 16062 | loss:0.8168519139289856 lr:5.918584614868155e-05 tokens_per_second_per_gpu:5340.942515325622 mfu:25.194137103550098 +Step 16063 | loss:0.5917648077011108 lr:5.918158369839591e-05 tokens_per_second_per_gpu:5304.091980598035 mfu:25.020306843890516 +Step 16064 | loss:0.9910274147987366 lr:5.9177321179054414e-05 tokens_per_second_per_gpu:4056.2488880873766 mfu:19.134018072532193 +Step 16065 | loss:0.8424060344696045 lr:5.917305859068912e-05 tokens_per_second_per_gpu:4956.330675595952 mfu:23.37985743399871 +Step 16066 | loss:0.7193027138710022 lr:5.916879593333209e-05 tokens_per_second_per_gpu:4977.174274972734 mfu:23.478180248543755 +Step 16067 | loss:0.8448373079299927 lr:5.916453320701536e-05 tokens_per_second_per_gpu:3760.1854557444703 mfu:17.73743634854035 +Step 16068 | loss:0.7479032278060913 lr:5.9160270411771015e-05 tokens_per_second_per_gpu:4739.046770508971 mfu:22.354892181246424 +Step 16069 | loss:0.8179561495780945 lr:5.91560075476311e-05 tokens_per_second_per_gpu:5360.218612563688 mfu:25.28506574306338 +Step 16070 | loss:0.7881664037704468 lr:5.915174461462768e-05 tokens_per_second_per_gpu:4892.546736145022 mfu:23.078977708946418 +Step 16071 | loss:0.7300022840499878 lr:5.9147481612792834e-05 tokens_per_second_per_gpu:4677.50041152907 mfu:22.06456750504657 +Step 16072 | loss:0.8911824822425842 lr:5.9143218542158606e-05 tokens_per_second_per_gpu:4566.665454760684 mfu:21.54174010357649 +Step 16073 | loss:0.6570887565612793 lr:5.9138955402757066e-05 tokens_per_second_per_gpu:4909.715611609727 mfu:23.159966223825645 +Step 16074 | loss:0.6642484664916992 lr:5.9134692194620266e-05 tokens_per_second_per_gpu:5348.7913574073 mfu:25.231161430799922 +Step 16075 | loss:0.8380195498466492 lr:5.91304289177803e-05 tokens_per_second_per_gpu:3612.6706703953046 mfu:17.041583937430648 +Step 16076 | loss:0.8420519232749939 lr:5.9126165572269195e-05 tokens_per_second_per_gpu:5364.015573633717 mfu:25.30297665625907 +Step 16077 | loss:0.6909388899803162 lr:5.9121902158119035e-05 tokens_per_second_per_gpu:4624.918314528043 mfu:21.816528782061873 +Step 16078 | loss:0.8963214159011841 lr:5.9117638675361895e-05 tokens_per_second_per_gpu:4690.731457880981 mfu:22.12698060813782 +Step 16079 | loss:0.6418564319610596 lr:5.9113375124029836e-05 tokens_per_second_per_gpu:4008.0901863941713 mfu:18.906845259923553 +Step 16080 | loss:0.9406999945640564 lr:5.9109111504154915e-05 tokens_per_second_per_gpu:4559.693485500357 mfu:21.50885213503491 +Step 16081 | loss:0.7607023119926453 lr:5.910484781576919e-05 tokens_per_second_per_gpu:4450.649450241532 mfu:20.994472815888145 +Step 16082 | loss:0.970408022403717 lr:5.910058405890475e-05 tokens_per_second_per_gpu:4265.720140560579 mfu:20.122129709929304 +Step 16083 | loss:0.8595982193946838 lr:5.909632023359366e-05 tokens_per_second_per_gpu:3518.9519896362017 mfu:16.599496930234178 +Step 16084 | loss:0.8539747595787048 lr:5.909205633986799e-05 tokens_per_second_per_gpu:4222.620352993038 mfu:19.918820658391176 +Step 16085 | loss:0.802379846572876 lr:5.90877923777598e-05 tokens_per_second_per_gpu:5310.528839179198 mfu:25.05067060405918 +Step 16086 | loss:0.8118543028831482 lr:5.908352834730116e-05 tokens_per_second_per_gpu:4127.997547680875 mfu:19.472468741418368 +Step 16087 | loss:0.8526659607887268 lr:5.9079264248524146e-05 tokens_per_second_per_gpu:4314.68679833416 mfu:20.353113789220593 +Step 16088 | loss:0.8892670273780823 lr:5.907500008146082e-05 tokens_per_second_per_gpu:4843.731719364822 mfu:22.848708946093524 +Step 16089 | loss:0.7821405529975891 lr:5.907073584614328e-05 tokens_per_second_per_gpu:4800.282166213621 mfu:22.643749990621107 +Step 16090 | loss:0.8463497757911682 lr:5.906647154260356e-05 tokens_per_second_per_gpu:5049.77810441667 mfu:23.820664899521464 +Step 16091 | loss:0.846808910369873 lr:5.906220717087376e-05 tokens_per_second_per_gpu:4215.203613822371 mfu:19.88383463429695 +Step 16092 | loss:0.6830887794494629 lr:5.905794273098593e-05 tokens_per_second_per_gpu:5275.142248572372 mfu:24.883746018591854 +Step 16093 | loss:0.7768630385398865 lr:5.9053678222972174e-05 tokens_per_second_per_gpu:4451.438266024864 mfu:20.998193794523974 +Step 16094 | loss:0.7960799932479858 lr:5.904941364686454e-05 tokens_per_second_per_gpu:4200.677532503125 mfu:19.815312630309027 +Step 16095 | loss:0.9437997341156006 lr:5.904514900269511e-05 tokens_per_second_per_gpu:3609.530174920557 mfu:17.026769684452482 +Step 16096 | loss:0.7784611582756042 lr:5.904088429049597e-05 tokens_per_second_per_gpu:5197.59486748018 mfu:24.517941791032438 +Step 16097 | loss:0.8090259432792664 lr:5.9036619510299175e-05 tokens_per_second_per_gpu:5346.656730865658 mfu:25.22109203319815 +Step 16098 | loss:0.8665834665298462 lr:5.9032354662136824e-05 tokens_per_second_per_gpu:5322.7389845759435 mfu:25.10826794316126 +Step 16099 | loss:0.8003389835357666 lr:5.902808974604096e-05 tokens_per_second_per_gpu:4551.521231306163 mfu:21.470302217670554 +Step 16100 | loss:0.7878220677375793 lr:5.90238247620437e-05 tokens_per_second_per_gpu:5320.890887567755 mfu:25.09955015425577 +Step 16101 | loss:0.7885617613792419 lr:5.9019559710177096e-05 tokens_per_second_per_gpu:5238.355543995906 mfu:24.710216856645655 +Step 16102 | loss:0.5563249588012695 lr:5.9015294590473235e-05 tokens_per_second_per_gpu:4334.8859284132 mfu:20.448396532130396 +Step 16103 | loss:0.843193531036377 lr:5.901102940296419e-05 tokens_per_second_per_gpu:4974.634103143314 mfu:23.466197824626377 +Step 16104 | loss:0.736296534538269 lr:5.9006764147682045e-05 tokens_per_second_per_gpu:5335.507100153109 mfu:25.168497322803898 +Step 16105 | loss:0.6578832268714905 lr:5.900249882465888e-05 tokens_per_second_per_gpu:4877.328044446289 mfu:23.007188543626942 +Step 16106 | loss:0.6946312785148621 lr:5.899823343392676e-05 tokens_per_second_per_gpu:5315.822164886499 mfu:25.07564012455502 +Step 16107 | loss:0.8235166668891907 lr:5.8993967975517784e-05 tokens_per_second_per_gpu:5349.297953001616 mfu:25.233551128652433 +Step 16108 | loss:0.8063207268714905 lr:5.8989702449464035e-05 tokens_per_second_per_gpu:3392.625132160528 mfu:16.003591589937006 +Step 16109 | loss:0.8102653622627258 lr:5.898543685579758e-05 tokens_per_second_per_gpu:4771.482665192928 mfu:22.507897830604943 +Step 16110 | loss:0.8481942415237427 lr:5.898117119455051e-05 tokens_per_second_per_gpu:5300.149042711379 mfu:25.001707333144584 +Step 16111 | loss:0.7639267444610596 lr:5.8976905465754905e-05 tokens_per_second_per_gpu:4241.557221158342 mfu:20.008149096489788 +Step 16112 | loss:0.6977518796920776 lr:5.897263966944285e-05 tokens_per_second_per_gpu:3985.7625781071706 mfu:18.801522122150978 +Step 16113 | loss:0.8488909006118774 lr:5.8968373805646425e-05 tokens_per_second_per_gpu:3824.2543311170352 mfu:18.039660164950075 +Step 16114 | loss:0.7965991497039795 lr:5.896410787439772e-05 tokens_per_second_per_gpu:5310.602488993232 mfu:25.051018022797937 +Step 16115 | loss:0.7823728919029236 lr:5.8959841875728814e-05 tokens_per_second_per_gpu:4166.2387988795745 mfu:19.652859247952886 +Step 16116 | loss:0.8677920699119568 lr:5.8955575809671794e-05 tokens_per_second_per_gpu:4216.967062098417 mfu:19.89215312069036 +Step 16117 | loss:0.8177444934844971 lr:5.8951309676258736e-05 tokens_per_second_per_gpu:3895.7810128397373 mfu:18.377063726346368 +Step 16118 | loss:0.7760533690452576 lr:5.894704347552174e-05 tokens_per_second_per_gpu:5314.436358272869 mfu:25.069103038315536 +Step 16119 | loss:0.9770221710205078 lr:5.89427772074929e-05 tokens_per_second_per_gpu:3796.1844685554242 mfu:17.907249834034552 +Step 16120 | loss:0.8696350455284119 lr:5.893851087220429e-05 tokens_per_second_per_gpu:5330.064956760746 mfu:25.142825803896717 +Step 16121 | loss:0.775408148765564 lr:5.893424446968799e-05 tokens_per_second_per_gpu:5357.959600727794 mfu:25.274409598806304 +Step 16122 | loss:0.7456514239311218 lr:5.892997799997611e-05 tokens_per_second_per_gpu:4501.763154714534 mfu:21.23558488078455 +Step 16123 | loss:0.7815166711807251 lr:5.8925711463100716e-05 tokens_per_second_per_gpu:4961.39603070497 mfu:23.403751578286645 +Step 16124 | loss:0.6635156869888306 lr:5.89214448590939e-05 tokens_per_second_per_gpu:3796.146403542632 mfu:17.907070274874652 +Step 16125 | loss:0.9194682836532593 lr:5.891717818798777e-05 tokens_per_second_per_gpu:3954.686457671683 mfu:18.6549307599239 +Step 16126 | loss:0.6276106238365173 lr:5.89129114498144e-05 tokens_per_second_per_gpu:4435.980471763422 mfu:20.925276741621097 +Step 16127 | loss:0.8058913350105286 lr:5.890864464460589e-05 tokens_per_second_per_gpu:4811.890426404968 mfu:22.698508134517112 +Step 16128 | loss:0.7703554034233093 lr:5.890437777239432e-05 tokens_per_second_per_gpu:3407.8667838125643 mfu:16.075489061273586 +Step 16129 | loss:0.7907792329788208 lr:5.8900110833211785e-05 tokens_per_second_per_gpu:4542.900591351731 mfu:21.429637188172656 +Step 16130 | loss:0.7559705972671509 lr:5.8895843827090394e-05 tokens_per_second_per_gpu:4161.362836300031 mfu:19.629858500540173 +Step 16131 | loss:0.7956104874610901 lr:5.889157675406223e-05 tokens_per_second_per_gpu:3546.495266294851 mfu:16.72942326560061 +Step 16132 | loss:0.9387801289558411 lr:5.888730961415937e-05 tokens_per_second_per_gpu:5323.976200448683 mfu:25.11410410152376 +Step 16133 | loss:0.8332138657569885 lr:5.888304240741392e-05 tokens_per_second_per_gpu:4202.364836263231 mfu:19.823271930029158 +Step 16134 | loss:0.757036030292511 lr:5.8878775133857987e-05 tokens_per_second_per_gpu:3928.203804624504 mfu:18.53000756709382 +Step 16135 | loss:0.8777944445610046 lr:5.887450779352364e-05 tokens_per_second_per_gpu:4413.128368842124 mfu:20.817479473170525 +Step 16136 | loss:0.7138384580612183 lr:5.887024038644299e-05 tokens_per_second_per_gpu:5217.851081140508 mfu:24.613493806934628 +Step 16137 | loss:0.7542732357978821 lr:5.8865972912648136e-05 tokens_per_second_per_gpu:4918.736991392373 mfu:23.202521611466494 +Step 16138 | loss:0.7663319706916809 lr:5.886170537217116e-05 tokens_per_second_per_gpu:5320.87632997289 mfu:25.099481483597007 +Step 16139 | loss:0.8096596598625183 lr:5.8857437765044174e-05 tokens_per_second_per_gpu:4467.438733131236 mfu:21.073670727825924 +Step 16140 | loss:0.7331538796424866 lr:5.885317009129926e-05 tokens_per_second_per_gpu:3875.6342658702815 mfu:18.282028083501956 +Step 16141 | loss:0.8613279461860657 lr:5.8848902350968546e-05 tokens_per_second_per_gpu:4856.88301810135 mfu:22.910745866076667 +Step 16142 | loss:0.7375430464744568 lr:5.884463454408409e-05 tokens_per_second_per_gpu:4721.740630605028 mfu:22.27325616658758 +Step 16143 | loss:0.7889009118080139 lr:5.8840366670678005e-05 tokens_per_second_per_gpu:5069.4359135757 mfu:23.9133941392927 +Step 16144 | loss:0.8497840762138367 lr:5.883609873078241e-05 tokens_per_second_per_gpu:4893.299737558784 mfu:23.082529745090422 +Step 16145 | loss:0.8957309722900391 lr:5.8831830724429384e-05 tokens_per_second_per_gpu:4959.460004753395 mfu:23.39461901355298 +Step 16146 | loss:0.7438879609107971 lr:5.882756265165102e-05 tokens_per_second_per_gpu:5347.296458152118 mfu:25.224109735957118 +Step 16147 | loss:0.7253403663635254 lr:5.882329451247943e-05 tokens_per_second_per_gpu:4478.197431960128 mfu:21.124421345825112 +Step 16148 | loss:0.8178529143333435 lr:5.8819026306946734e-05 tokens_per_second_per_gpu:4718.889156504166 mfu:22.259805276742163 +Step 16149 | loss:0.7674713730812073 lr:5.8814758035085e-05 tokens_per_second_per_gpu:5123.399848151222 mfu:24.167951225882092 +Step 16150 | loss:0.7623029351234436 lr:5.881048969692636e-05 tokens_per_second_per_gpu:4374.625840791918 mfu:20.63585647914902 +Step 16151 | loss:0.7053629755973816 lr:5.8806221292502896e-05 tokens_per_second_per_gpu:4674.31215924819 mfu:22.049527975065725 +Step 16152 | loss:0.7470314502716064 lr:5.880195282184673e-05 tokens_per_second_per_gpu:5334.76808908215 mfu:25.165011281493896 +Step 16153 | loss:0.7174896001815796 lr:5.879768428498993e-05 tokens_per_second_per_gpu:4674.425041801859 mfu:22.050060461331135 +Step 16154 | loss:0.7939984798431396 lr:5.8793415681964637e-05 tokens_per_second_per_gpu:4793.283239227779 mfu:22.61073485788881 +Step 16155 | loss:0.8396350145339966 lr:5.8789147012802935e-05 tokens_per_second_per_gpu:3317.118310025799 mfu:15.647413027135201 +Step 16156 | loss:0.7420607805252075 lr:5.8784878277536945e-05 tokens_per_second_per_gpu:4687.659958484807 mfu:22.112491821434233 +Step 16157 | loss:0.8196709156036377 lr:5.878060947619877e-05 tokens_per_second_per_gpu:5361.97189100879 mfu:25.293336256629075 +Step 16158 | loss:0.8167561888694763 lr:5.8776340608820494e-05 tokens_per_second_per_gpu:5037.61595917205 mfu:23.763293985326637 +Step 16159 | loss:0.6845844388008118 lr:5.877207167543426e-05 tokens_per_second_per_gpu:4526.048476623634 mfu:21.350142887732957 +Step 16160 | loss:0.9125744104385376 lr:5.8767802676072136e-05 tokens_per_second_per_gpu:5358.940044432673 mfu:25.27903451904316 +Step 16161 | loss:0.7698315382003784 lr:5.8763533610766264e-05 tokens_per_second_per_gpu:5194.732814722291 mfu:24.504440999856918 +Step 16162 | loss:0.8721996545791626 lr:5.8759264479548726e-05 tokens_per_second_per_gpu:4498.538688574109 mfu:21.22037452384951 +Step 16163 | loss:0.766694962978363 lr:5.875499528245165e-05 tokens_per_second_per_gpu:4538.861854676363 mfu:21.410585778195408 +Step 16164 | loss:0.8583651185035706 lr:5.875072601950713e-05 tokens_per_second_per_gpu:4724.473679591631 mfu:22.286148424963624 +Step 16165 | loss:0.9362873435020447 lr:5.874645669074729e-05 tokens_per_second_per_gpu:4874.142082284255 mfu:22.992159816527426 +Step 16166 | loss:0.8397212624549866 lr:5.874218729620423e-05 tokens_per_second_per_gpu:4945.04337555346 mfu:23.326613313886302 +Step 16167 | loss:0.9330891370773315 lr:5.8737917835910075e-05 tokens_per_second_per_gpu:5148.662390172236 mfu:24.287118946829338 +Step 16168 | loss:0.809895932674408 lr:5.873364830989692e-05 tokens_per_second_per_gpu:3655.0010704714346 mfu:17.241263656900745 +Step 16169 | loss:0.7510225772857666 lr:5.872937871819687e-05 tokens_per_second_per_gpu:5320.509300020645 mfu:25.0977501406905 +Step 16170 | loss:0.8638242483139038 lr:5.872510906084207e-05 tokens_per_second_per_gpu:4525.1903871963 mfu:21.346095133499073 +Step 16171 | loss:0.7765921354293823 lr:5.872083933786457e-05 tokens_per_second_per_gpu:5285.0777871871105 mfu:24.930613649415264 +Step 16172 | loss:0.8196006417274475 lr:5.8716569549296565e-05 tokens_per_second_per_gpu:4715.379858738307 mfu:22.243251320432666 +Step 16173 | loss:0.7346230149269104 lr:5.871229969517011e-05 tokens_per_second_per_gpu:4280.416667887044 mfu:20.191455736814362 +Step 16174 | loss:0.88702392578125 lr:5.8708029775517346e-05 tokens_per_second_per_gpu:5009.358472719218 mfu:23.62999860050429 +Step 16175 | loss:0.8645914196968079 lr:5.870375979037038e-05 tokens_per_second_per_gpu:4754.18408281021 mfu:22.42629746606316 +Step 16176 | loss:0.8583203554153442 lr:5.869948973976131e-05 tokens_per_second_per_gpu:3219.221353598077 mfu:15.185616380722516 +Step 16177 | loss:0.9387241005897522 lr:5.86952196237223e-05 tokens_per_second_per_gpu:4582.034082214773 mfu:21.614236541435655 +Step 16178 | loss:0.8248385190963745 lr:5.8690949442285394e-05 tokens_per_second_per_gpu:4062.141587472145 mfu:19.161814940928334 +Step 16179 | loss:0.7375097870826721 lr:5.8686679195482774e-05 tokens_per_second_per_gpu:5356.753998555942 mfu:25.26872256766474 +Step 16180 | loss:0.738268256187439 lr:5.868240888334653e-05 tokens_per_second_per_gpu:5330.397317107537 mfu:25.14439360435905 +Step 16181 | loss:0.7879208922386169 lr:5.867813850590878e-05 tokens_per_second_per_gpu:4820.749440589213 mfu:22.740297615927876 +Step 16182 | loss:0.7621369361877441 lr:5.867386806320162e-05 tokens_per_second_per_gpu:4438.772094708896 mfu:20.938445303354882 +Step 16183 | loss:0.6629884243011475 lr:5.8669597555257205e-05 tokens_per_second_per_gpu:4200.482452605497 mfu:19.814392405147853 +Step 16184 | loss:0.786111056804657 lr:5.866532698210765e-05 tokens_per_second_per_gpu:3998.9315028123524 mfu:18.863642186835563 +Step 16185 | loss:0.777755856513977 lr:5.866105634378507e-05 tokens_per_second_per_gpu:3432.2106619126744 mfu:16.190323287765704 +Step 16186 | loss:0.8070242404937744 lr:5.8656785640321566e-05 tokens_per_second_per_gpu:5360.501557061602 mfu:25.286400440535353 +Step 16187 | loss:0.869635820388794 lr:5.8652514871749267e-05 tokens_per_second_per_gpu:4634.463964981284 mfu:21.861557243905718 +Step 16188 | loss:0.7498434782028198 lr:5.864824403810032e-05 tokens_per_second_per_gpu:5249.39221951858 mfu:24.762278738137123 +Step 16189 | loss:0.7678117752075195 lr:5.864397313940681e-05 tokens_per_second_per_gpu:3847.1321775454103 mfu:18.14757886991582 +Step 16190 | loss:0.8445773124694824 lr:5.8639702175700884e-05 tokens_per_second_per_gpu:5344.35328676232 mfu:25.210226294354616 +Step 16191 | loss:0.9856476187705994 lr:5.863543114701465e-05 tokens_per_second_per_gpu:3262.057937101944 mfu:15.387683853785049 +Step 16192 | loss:0.8200395703315735 lr:5.863116005338024e-05 tokens_per_second_per_gpu:5324.985286846517 mfu:25.118864134230236 +Step 16193 | loss:0.7596855163574219 lr:5.8626888894829776e-05 tokens_per_second_per_gpu:4102.157300327575 mfu:19.35057588585203 +Step 16194 | loss:0.7796480655670166 lr:5.862261767139536e-05 tokens_per_second_per_gpu:4491.816787040247 mfu:21.18866616743962 +Step 16195 | loss:0.9047719240188599 lr:5.861834638310917e-05 tokens_per_second_per_gpu:4823.418979498766 mfu:22.752890286434543 +Step 16196 | loss:0.7383266687393188 lr:5.8614075030003266e-05 tokens_per_second_per_gpu:4887.0060424140875 mfu:23.052841311277767 +Step 16197 | loss:0.7573135495185852 lr:5.860980361210983e-05 tokens_per_second_per_gpu:3733.737744168254 mfu:17.61267798059043 +Step 16198 | loss:0.8052471280097961 lr:5.860553212946095e-05 tokens_per_second_per_gpu:4875.675123339168 mfu:22.999391432747228 +Step 16199 | loss:0.7035008668899536 lr:5.860126058208877e-05 tokens_per_second_per_gpu:4706.731998498853 mfu:22.202457888206055 +Step 16200 | loss:0.774173378944397 lr:5.859698897002539e-05 tokens_per_second_per_gpu:3173.830518938784 mfu:14.971500069159653 +Step 16201 | loss:0.6973960399627686 lr:5.859271729330298e-05 tokens_per_second_per_gpu:5359.487726209514 mfu:25.28161802742867 +Step 16202 | loss:0.7376687526702881 lr:5.858844555195365e-05 tokens_per_second_per_gpu:5126.0851822002305 mfu:24.180618404756498 +Step 16203 | loss:0.7233040928840637 lr:5.8584173746009516e-05 tokens_per_second_per_gpu:3936.498994846348 mfu:18.569137394675653 +Step 16204 | loss:0.769356369972229 lr:5.857990187550272e-05 tokens_per_second_per_gpu:4753.60505077335 mfu:22.423566073152237 +Step 16205 | loss:0.6420787572860718 lr:5.857562994046537e-05 tokens_per_second_per_gpu:5230.720336240643 mfu:24.674200278963305 +Step 16206 | loss:0.8612645864486694 lr:5.857135794092965e-05 tokens_per_second_per_gpu:3416.7352027651214 mfu:16.117322906581098 +Step 16207 | loss:0.8144983649253845 lr:5.8567085876927616e-05 tokens_per_second_per_gpu:4286.095045432604 mfu:20.21824161252788 +Step 16208 | loss:0.9014237523078918 lr:5.8562813748491455e-05 tokens_per_second_per_gpu:4257.321540286408 mfu:20.08251207948707 +Step 16209 | loss:0.8494957089424133 lr:5.855854155565328e-05 tokens_per_second_per_gpu:4362.66097324566 mfu:20.579416157516416 +Step 16210 | loss:0.7471944689750671 lr:5.855426929844522e-05 tokens_per_second_per_gpu:4650.745709901744 mfu:21.938361012647665 +Step 16211 | loss:0.8469306230545044 lr:5.8549996976899404e-05 tokens_per_second_per_gpu:5009.642346707503 mfu:23.631337682540927 +Step 16212 | loss:0.8966746926307678 lr:5.8545724591047966e-05 tokens_per_second_per_gpu:3729.2430374742626 mfu:17.591475682238837 +Step 16213 | loss:0.8701143860816956 lr:5.854145214092306e-05 tokens_per_second_per_gpu:4188.895927443332 mfu:19.759736789092877 +Step 16214 | loss:0.8402101397514343 lr:5.8537179626556785e-05 tokens_per_second_per_gpu:3767.688174538852 mfu:17.77282795318927 +Step 16215 | loss:0.8180184960365295 lr:5.85329070479813e-05 tokens_per_second_per_gpu:3803.392250986351 mfu:17.941250173535728 +Step 16216 | loss:0.680290699005127 lr:5.8528634405228734e-05 tokens_per_second_per_gpu:5380.296057481885 mfu:25.379774476307855 +Step 16217 | loss:0.769592821598053 lr:5.852436169833122e-05 tokens_per_second_per_gpu:3196.3999309256787 mfu:15.077963836240597 +Step 16218 | loss:0.7341917157173157 lr:5.852008892732088e-05 tokens_per_second_per_gpu:5303.876222359369 mfu:25.01928907546658 +Step 16219 | loss:0.7730786204338074 lr:5.851581609222988e-05 tokens_per_second_per_gpu:3498.8276808187798 mfu:16.504567131981258 +Step 16220 | loss:0.8788090348243713 lr:5.8511543193090344e-05 tokens_per_second_per_gpu:4030.978351872455 mfu:19.01481264160833 +Step 16221 | loss:0.7604612112045288 lr:5.8507270229934384e-05 tokens_per_second_per_gpu:5319.789153601137 mfu:25.094353087160027 +Step 16222 | loss:0.694839596748352 lr:5.8502997202794186e-05 tokens_per_second_per_gpu:4474.798716641039 mfu:21.108389025784344 +Step 16223 | loss:0.6520488858222961 lr:5.849872411170183e-05 tokens_per_second_per_gpu:4932.678132020906 mfu:23.26828435041515 +Step 16224 | loss:0.6744866967201233 lr:5.849445095668952e-05 tokens_per_second_per_gpu:5326.750191051994 mfu:25.127189488491183 +Step 16225 | loss:0.7812819480895996 lr:5.8490177737789334e-05 tokens_per_second_per_gpu:4548.705461024178 mfu:21.457019748830476 +Step 16226 | loss:0.7120723128318787 lr:5.8485904455033444e-05 tokens_per_second_per_gpu:4545.416404117753 mfu:21.4415047062319 +Step 16227 | loss:0.8182594776153564 lr:5.8481631108453994e-05 tokens_per_second_per_gpu:5320.185348935281 mfu:25.09622200814908 +Step 16228 | loss:0.7269494533538818 lr:5.847735769808311e-05 tokens_per_second_per_gpu:4531.980926870126 mfu:21.378127267726043 +Step 16229 | loss:0.8008276224136353 lr:5.847308422395293e-05 tokens_per_second_per_gpu:5327.371766135979 mfu:25.130121564213983 +Step 16230 | loss:0.6783434748649597 lr:5.846881068609561e-05 tokens_per_second_per_gpu:5356.382191339857 mfu:25.266968689589614 +Step 16231 | loss:0.7344127893447876 lr:5.846453708454329e-05 tokens_per_second_per_gpu:3902.4111913940746 mfu:18.40833940981229 +Step 16232 | loss:0.7499580979347229 lr:5.8460263419328084e-05 tokens_per_second_per_gpu:3966.3789278500935 mfu:18.71008613669658 +Step 16233 | loss:0.7155751585960388 lr:5.8455989690482184e-05 tokens_per_second_per_gpu:5340.515197645767 mfu:25.19212137314631 +Step 16234 | loss:0.7503523230552673 lr:5.845171589803768e-05 tokens_per_second_per_gpu:3628.5786597344695 mfu:17.116624637326346 +Step 16235 | loss:0.7576884627342224 lr:5.8447442042026767e-05 tokens_per_second_per_gpu:4343.531008147205 mfu:20.489176848238394 +Step 16236 | loss:0.7402783036231995 lr:5.844316812248154e-05 tokens_per_second_per_gpu:3642.1277472949196 mfu:17.1805379950617 +Step 16237 | loss:0.9292843341827393 lr:5.8438894139434184e-05 tokens_per_second_per_gpu:4562.663476944599 mfu:21.522862091410193 +Step 16238 | loss:0.8384864926338196 lr:5.843462009291685e-05 tokens_per_second_per_gpu:5001.536402196757 mfu:23.593100559266993 +Step 16239 | loss:0.8443634510040283 lr:5.843034598296162e-05 tokens_per_second_per_gpu:4273.992577093332 mfu:20.161152204477514 +Step 16240 | loss:0.7242470383644104 lr:5.8426071809600705e-05 tokens_per_second_per_gpu:4020.5037536194586 mfu:18.965402174498205 +Step 16241 | loss:0.7914525270462036 lr:5.842179757286621e-05 tokens_per_second_per_gpu:4274.255140635222 mfu:20.162390761502568 +Step 16242 | loss:0.7897869348526001 lr:5.841752327279033e-05 tokens_per_second_per_gpu:4004.549176195614 mfu:18.890141710658323 +Step 16243 | loss:0.9136413335800171 lr:5.841324890940516e-05 tokens_per_second_per_gpu:4491.01373913219 mfu:21.184878053443214 +Step 16244 | loss:0.7085216045379639 lr:5.8408974482742886e-05 tokens_per_second_per_gpu:4514.63549939881 mfu:21.29630592690875 +Step 16245 | loss:0.7861117720603943 lr:5.840469999283563e-05 tokens_per_second_per_gpu:4188.3710864768445 mfu:19.757261024706857 +Step 16246 | loss:0.8955303430557251 lr:5.8400425439715565e-05 tokens_per_second_per_gpu:5318.794195636729 mfu:25.08965970068458 +Step 16247 | loss:0.7595878839492798 lr:5.839615082341482e-05 tokens_per_second_per_gpu:3344.5663082167425 mfu:15.77688991771328 +Step 16248 | loss:0.6881348490715027 lr:5.839187614396555e-05 tokens_per_second_per_gpu:5308.349247643852 mfu:25.040389098910293 +Step 16249 | loss:0.743939995765686 lr:5.838760140139993e-05 tokens_per_second_per_gpu:4804.047726169081 mfu:22.661512779401793 +Step 16250 | loss:0.7382392883300781 lr:5.838332659575005e-05 tokens_per_second_per_gpu:5348.284243144386 mfu:25.228769286299144 +Step 16251 | loss:0.8256857991218567 lr:5.837905172704814e-05 tokens_per_second_per_gpu:4495.998728801153 mfu:21.208393100238656 +Step 16252 | loss:0.764388382434845 lr:5.837477679532628e-05 tokens_per_second_per_gpu:5288.886010997865 mfu:24.948577690880576 +Step 16253 | loss:0.7666866183280945 lr:5.837050180061668e-05 tokens_per_second_per_gpu:3898.2140940646036 mfu:18.388540985610526 +Step 16254 | loss:0.7830987572669983 lr:5.836622674295147e-05 tokens_per_second_per_gpu:4622.205069663083 mfu:21.803729943106543 +Step 16255 | loss:0.7815724611282349 lr:5.8361951622362785e-05 tokens_per_second_per_gpu:4438.521177990654 mfu:20.93726168638413 +Step 16256 | loss:0.8206218481063843 lr:5.8357676438882816e-05 tokens_per_second_per_gpu:4494.645604163318 mfu:21.20201018045512 +Step 16257 | loss:0.8165497183799744 lr:5.8353401192543666e-05 tokens_per_second_per_gpu:4499.239595394656 mfu:21.223680820909067 +Step 16258 | loss:0.7927542328834534 lr:5.834912588337754e-05 tokens_per_second_per_gpu:4982.62208927354 mfu:23.50387851809362 +Step 16259 | loss:0.7825757265090942 lr:5.834485051141656e-05 tokens_per_second_per_gpu:3258.435174491464 mfu:15.370594664444376 +Step 16260 | loss:0.8349614143371582 lr:5.834057507669292e-05 tokens_per_second_per_gpu:5276.096791444593 mfu:24.888248760181924 +Step 16261 | loss:0.7523194551467896 lr:5.833629957923872e-05 tokens_per_second_per_gpu:4714.894233558445 mfu:22.24096054360306 +Step 16262 | loss:0.7720414996147156 lr:5.833202401908615e-05 tokens_per_second_per_gpu:4618.93937715073 mfu:21.78832511434118 +Step 16263 | loss:0.9314578771591187 lr:5.832774839626738e-05 tokens_per_second_per_gpu:5062.827466744325 mfu:23.882220968071806 +Step 16264 | loss:0.7859722375869751 lr:5.8323472710814554e-05 tokens_per_second_per_gpu:5155.810474094009 mfu:24.320837678277222 +Step 16265 | loss:0.7991361618041992 lr:5.831919696275982e-05 tokens_per_second_per_gpu:4486.242400525141 mfu:21.162370835159457 +Step 16266 | loss:0.7822484970092773 lr:5.831492115213533e-05 tokens_per_second_per_gpu:4874.592558576111 mfu:22.994284790055293 +Step 16267 | loss:0.8883776068687439 lr:5.831064527897328e-05 tokens_per_second_per_gpu:3909.7862521736033 mfu:18.443128829824445 +Step 16268 | loss:0.7977696657180786 lr:5.830636934330579e-05 tokens_per_second_per_gpu:4984.749679064523 mfu:23.51391472214194 +Step 16269 | loss:0.7475472092628479 lr:5.830209334516505e-05 tokens_per_second_per_gpu:3835.557235167902 mfu:18.092977891832298 +Step 16270 | loss:0.8306816816329956 lr:5.829781728458319e-05 tokens_per_second_per_gpu:4844.799087834359 mfu:22.85374390527617 +Step 16271 | loss:0.7384265661239624 lr:5.829354116159241e-05 tokens_per_second_per_gpu:4146.178439730404 mfu:19.55823111119604 +Step 16272 | loss:0.9039503931999207 lr:5.828926497622483e-05 tokens_per_second_per_gpu:4046.400149376452 mfu:19.087559891664668 +Step 16273 | loss:0.855835497379303 lr:5.828498872851264e-05 tokens_per_second_per_gpu:3666.8815426285482 mfu:17.29730587108342 +Step 16274 | loss:0.8336354494094849 lr:5.828071241848799e-05 tokens_per_second_per_gpu:3468.8425304739617 mfu:16.363122061810557 +Step 16275 | loss:0.8454559445381165 lr:5.827643604618305e-05 tokens_per_second_per_gpu:4416.456667325212 mfu:20.83317962498153 +Step 16276 | loss:0.8066257238388062 lr:5.8272159611629964e-05 tokens_per_second_per_gpu:5345.277933987677 mfu:25.214588013077712 +Step 16277 | loss:0.7823073863983154 lr:5.826788311486091e-05 tokens_per_second_per_gpu:5328.5474896346805 mfu:25.135667652556474 +Step 16278 | loss:0.7271278500556946 lr:5.826360655590808e-05 tokens_per_second_per_gpu:5317.534701296401 mfu:25.083718450989366 +Step 16279 | loss:0.8600374460220337 lr:5.825932993480358e-05 tokens_per_second_per_gpu:4624.615514209063 mfu:21.81510042129418 +Step 16280 | loss:0.7571557760238647 lr:5.8255053251579616e-05 tokens_per_second_per_gpu:5030.527912462426 mfu:23.729858459651865 +Step 16281 | loss:0.712046205997467 lr:5.825077650626834e-05 tokens_per_second_per_gpu:4106.039266599447 mfu:19.368887783088194 +Step 16282 | loss:0.7686401605606079 lr:5.824649969890194e-05 tokens_per_second_per_gpu:4881.781857767788 mfu:23.02819794096336 +Step 16283 | loss:0.7281239032745361 lr:5.824222282951255e-05 tokens_per_second_per_gpu:4260.354146392451 mfu:20.09681739990523 +Step 16284 | loss:0.7641339302062988 lr:5.823794589813234e-05 tokens_per_second_per_gpu:3508.2874201833592 mfu:16.549190336561946 +Step 16285 | loss:0.6850478053092957 lr:5.82336689047935e-05 tokens_per_second_per_gpu:3610.3376442846234 mfu:17.03057865521126 +Step 16286 | loss:0.7848814725875854 lr:5.822939184952818e-05 tokens_per_second_per_gpu:5333.699080911345 mfu:25.159968587560606 +Step 16287 | loss:0.9117092490196228 lr:5.8225114732368556e-05 tokens_per_second_per_gpu:5331.197215147791 mfu:25.148166859891607 +Step 16288 | loss:0.8372767567634583 lr:5.8220837553346795e-05 tokens_per_second_per_gpu:5119.161617855265 mfu:24.14795877045941 +Step 16289 | loss:0.9008350372314453 lr:5.821656031249507e-05 tokens_per_second_per_gpu:4625.198628666964 mfu:21.817851071681687 +Step 16290 | loss:0.8783618807792664 lr:5.821228300984554e-05 tokens_per_second_per_gpu:5356.076690208368 mfu:25.26552758862851 +Step 16291 | loss:0.8650630712509155 lr:5.8208005645430374e-05 tokens_per_second_per_gpu:4390.080256145069 mfu:20.708757593164997 +Step 16292 | loss:0.8098434209823608 lr:5.820372821928177e-05 tokens_per_second_per_gpu:4560.007466028581 mfu:21.510333234757145 +Step 16293 | loss:0.7212179899215698 lr:5.819945073143187e-05 tokens_per_second_per_gpu:3195.2880956731583 mfu:15.072719119655801 +Step 16294 | loss:0.6920511722564697 lr:5.8195173181912856e-05 tokens_per_second_per_gpu:4329.479842845386 mfu:20.4228951041337 +Step 16295 | loss:0.7132510542869568 lr:5.819089557075689e-05 tokens_per_second_per_gpu:3976.8830807111585 mfu:18.75963601793551 +Step 16296 | loss:0.6763722896575928 lr:5.8186617897996165e-05 tokens_per_second_per_gpu:5327.900422148388 mfu:25.132615324822535 +Step 16297 | loss:0.7323474884033203 lr:5.818234016366283e-05 tokens_per_second_per_gpu:5358.475344576369 mfu:25.27684245053501 +Step 16298 | loss:0.847170352935791 lr:5.817806236778908e-05 tokens_per_second_per_gpu:3522.529659614273 mfu:16.616373409934184 +Step 16299 | loss:0.6458114385604858 lr:5.817378451040708e-05 tokens_per_second_per_gpu:5191.319192928903 mfu:24.488338363433574 +Step 16300 | loss:0.8735825419425964 lr:5.816950659154901e-05 tokens_per_second_per_gpu:5069.72105344294 mfu:23.91473919269635 +Step 16301 | loss:0.7610050439834595 lr:5.816522861124702e-05 tokens_per_second_per_gpu:4514.936742381591 mfu:21.29772694145589 +Step 16302 | loss:0.7363178730010986 lr:5.81609505695333e-05 tokens_per_second_per_gpu:5299.181226304598 mfu:24.997141978027294 +Step 16303 | loss:0.8216053247451782 lr:5.815667246644005e-05 tokens_per_second_per_gpu:4472.785750085795 mfu:21.098893519095746 +Step 16304 | loss:0.7928059101104736 lr:5.815239430199941e-05 tokens_per_second_per_gpu:4538.721092425803 mfu:21.409921778643 +Step 16305 | loss:0.7493321895599365 lr:5.814811607624359e-05 tokens_per_second_per_gpu:5278.420811478143 mfu:24.89921155919889 +Step 16306 | loss:0.7715427875518799 lr:5.814383778920474e-05 tokens_per_second_per_gpu:5330.126305840775 mfu:25.143115197974442 +Step 16307 | loss:0.8153606057167053 lr:5.813955944091505e-05 tokens_per_second_per_gpu:4132.488599744409 mfu:19.493653800254553 +Step 16308 | loss:0.8233303427696228 lr:5.8135281031406694e-05 tokens_per_second_per_gpu:4404.886770511374 mfu:20.778602447682704 +Step 16309 | loss:0.7939081192016602 lr:5.8131002560711845e-05 tokens_per_second_per_gpu:3183.0249871991664 mfu:15.014871944681925 +Step 16310 | loss:0.8338740468025208 lr:5.812672402886269e-05 tokens_per_second_per_gpu:3823.0531643060876 mfu:18.03399405616264 +Step 16311 | loss:0.7457870244979858 lr:5.81224454358914e-05 tokens_per_second_per_gpu:5275.749576718493 mfu:24.88661088907797 +Step 16312 | loss:0.8661732077598572 lr:5.8118166781830176e-05 tokens_per_second_per_gpu:4923.11039663155 mfu:23.223151710159577 +Step 16313 | loss:0.8506338000297546 lr:5.8113888066711176e-05 tokens_per_second_per_gpu:3740.314764136002 mfu:17.64370290593362 +Step 16314 | loss:0.7415786385536194 lr:5.8109609290566604e-05 tokens_per_second_per_gpu:3627.7667677797053 mfu:17.112794804452673 +Step 16315 | loss:0.7183800339698792 lr:5.81053304534286e-05 tokens_per_second_per_gpu:4068.9126188954324 mfu:19.193755051409067 +Step 16316 | loss:0.8658044934272766 lr:5.810105155532938e-05 tokens_per_second_per_gpu:4604.550210321755 mfu:21.720448959363885 +Step 16317 | loss:0.7059032917022705 lr:5.809677259630113e-05 tokens_per_second_per_gpu:5338.841615129369 mfu:25.184226798873837 +Step 16318 | loss:0.6296288967132568 lr:5.8092493576376004e-05 tokens_per_second_per_gpu:3180.3767970451154 mfu:15.002379979897436 +Step 16319 | loss:1.0003654956817627 lr:5.808821449558621e-05 tokens_per_second_per_gpu:5333.079768793169 mfu:25.157047186633626 +Step 16320 | loss:0.6753445863723755 lr:5.808393535396391e-05 tokens_per_second_per_gpu:5338.259983506602 mfu:25.181483143272658 +Step 16321 | loss:0.8546387553215027 lr:5.807965615154133e-05 tokens_per_second_per_gpu:4250.41753523439 mfu:20.049944709711973 +Step 16322 | loss:0.8477835059165955 lr:5.8075376888350595e-05 tokens_per_second_per_gpu:4313.0443404523585 mfu:20.345366035159927 +Step 16323 | loss:0.9898154139518738 lr:5.807109756442394e-05 tokens_per_second_per_gpu:5231.3336123701765 mfu:24.67709320710201 +Step 16324 | loss:0.7576831579208374 lr:5.806681817979351e-05 tokens_per_second_per_gpu:5137.303266825983 mfu:24.233536023938775 +Step 16325 | loss:0.774897575378418 lr:5.8062538734491534e-05 tokens_per_second_per_gpu:5064.1924887469695 mfu:23.888660009755053 +Step 16326 | loss:0.8634495139122009 lr:5.805825922855016e-05 tokens_per_second_per_gpu:4531.555528735251 mfu:21.376120592143433 +Step 16327 | loss:0.8137415647506714 lr:5.805397966200159e-05 tokens_per_second_per_gpu:5322.047000973464 mfu:25.10500373844378 +Step 16328 | loss:0.8142501711845398 lr:5.804970003487802e-05 tokens_per_second_per_gpu:4358.575900725062 mfu:20.56014617345151 +Step 16329 | loss:0.6639078259468079 lr:5.804542034721162e-05 tokens_per_second_per_gpu:5305.367815867842 mfu:25.02632517653814 +Step 16330 | loss:0.7054551839828491 lr:5.8041140599034594e-05 tokens_per_second_per_gpu:3626.3145135329232 mfu:17.105944273390477 +Step 16331 | loss:0.8771088719367981 lr:5.803686079037911e-05 tokens_per_second_per_gpu:3538.678824676554 mfu:16.692551776864644 +Step 16332 | loss:0.7307137846946716 lr:5.80325809212774e-05 tokens_per_second_per_gpu:4853.385289406424 mfu:22.894246483048647 +Step 16333 | loss:0.7662776708602905 lr:5.80283009917616e-05 tokens_per_second_per_gpu:4158.810824682561 mfu:19.61782022632249 +Step 16334 | loss:0.8117471933364868 lr:5.802402100186393e-05 tokens_per_second_per_gpu:4539.458308052443 mfu:21.41339935053582 +Step 16335 | loss:0.6781112551689148 lr:5.801974095161657e-05 tokens_per_second_per_gpu:5359.989654432389 mfu:25.28398570849376 +Step 16336 | loss:0.8001730442047119 lr:5.801546084105173e-05 tokens_per_second_per_gpu:4057.2191317520646 mfu:19.138594877439168 +Step 16337 | loss:0.7141435146331787 lr:5.801118067020157e-05 tokens_per_second_per_gpu:5174.452149728208 mfu:24.408773642841574 +Step 16338 | loss:0.8548781275749207 lr:5.800690043909829e-05 tokens_per_second_per_gpu:5207.794395888021 mfu:24.566054706751217 +Step 16339 | loss:0.8475660681724548 lr:5.800262014777412e-05 tokens_per_second_per_gpu:3784.2439013788635 mfu:17.85092414139188 +Step 16340 | loss:0.6751952171325684 lr:5.799833979626119e-05 tokens_per_second_per_gpu:4831.421752315541 mfu:22.79064073122511 +Step 16341 | loss:0.7849989533424377 lr:5.799405938459175e-05 tokens_per_second_per_gpu:4184.2152922197 mfu:19.737657434144293 +Step 16342 | loss:0.8292446732521057 lr:5.7989778912797966e-05 tokens_per_second_per_gpu:5411.229895047326 mfu:25.525694665960962 +Step 16343 | loss:0.7987349033355713 lr:5.798549838091203e-05 tokens_per_second_per_gpu:5175.825837161331 mfu:24.41525356083886 +Step 16344 | loss:0.8516333699226379 lr:5.798121778896615e-05 tokens_per_second_per_gpu:3745.5068221795195 mfu:17.668194729581337 +Step 16345 | loss:0.7689149379730225 lr:5.797693713699248e-05 tokens_per_second_per_gpu:4983.424533359029 mfu:23.50766378375595 +Step 16346 | loss:0.7047597169876099 lr:5.797265642502327e-05 tokens_per_second_per_gpu:4822.726131447849 mfu:22.7496220039663 +Step 16347 | loss:0.8789483308792114 lr:5.796837565309071e-05 tokens_per_second_per_gpu:3878.3109675107967 mfu:18.29465454183266 +Step 16348 | loss:0.7440920472145081 lr:5.7964094821226963e-05 tokens_per_second_per_gpu:5310.713796395689 mfu:25.05154307880633 +Step 16349 | loss:0.7294278740882874 lr:5.795981392946424e-05 tokens_per_second_per_gpu:4110.992847883884 mfu:19.392254671175 +Step 16350 | loss:0.8214326500892639 lr:5.795553297783476e-05 tokens_per_second_per_gpu:5046.025921728765 mfu:23.802965213594234 +Step 16351 | loss:0.7431864142417908 lr:5.795125196637068e-05 tokens_per_second_per_gpu:4098.575158087346 mfu:19.33367830972838 +Step 16352 | loss:0.8277674913406372 lr:5.7946970895104226e-05 tokens_per_second_per_gpu:5121.946749769249 mfu:24.16109671289018 +Step 16353 | loss:0.8035492897033691 lr:5.7942689764067595e-05 tokens_per_second_per_gpu:4622.164719611452 mfu:21.803539605028572 +Step 16354 | loss:0.7530167102813721 lr:5.793840857329298e-05 tokens_per_second_per_gpu:5293.353917301805 mfu:24.96965356721968 +Step 16355 | loss:0.8213469386100769 lr:5.793412732281257e-05 tokens_per_second_per_gpu:5332.189719374636 mfu:25.15284867166832 +Step 16356 | loss:0.8934540748596191 lr:5.7929846012658584e-05 tokens_per_second_per_gpu:4110.906317106025 mfu:19.39184649073217 +Step 16357 | loss:0.8317936658859253 lr:5.7925564642863205e-05 tokens_per_second_per_gpu:4628.759783088704 mfu:21.83464964468463 +Step 16358 | loss:0.8357473015785217 lr:5.792128321345866e-05 tokens_per_second_per_gpu:4180.7859283072585 mfu:19.721480539459485 +Step 16359 | loss:0.8883992433547974 lr:5.791700172447713e-05 tokens_per_second_per_gpu:5277.361912907942 mfu:24.894216553976523 +Step 16360 | loss:0.7114675045013428 lr:5.791272017595082e-05 tokens_per_second_per_gpu:4205.23280149086 mfu:19.836800611333206 +Step 16361 | loss:0.8683096170425415 lr:5.790843856791193e-05 tokens_per_second_per_gpu:4615.760376126942 mfu:21.773329224118466 +Step 16362 | loss:0.6965932846069336 lr:5.790415690039266e-05 tokens_per_second_per_gpu:4868.41949024489 mfu:22.96516537350286 +Step 16363 | loss:0.7251129150390625 lr:5.789987517342521e-05 tokens_per_second_per_gpu:5359.538115800117 mfu:25.2818557237249 +Step 16364 | loss:0.7106946110725403 lr:5.78955933870418e-05 tokens_per_second_per_gpu:4884.858090616326 mfu:23.042709056169357 +Step 16365 | loss:0.6429485082626343 lr:5.789131154127463e-05 tokens_per_second_per_gpu:4587.580630950915 mfu:21.64040055816284 +Step 16366 | loss:0.7336151003837585 lr:5.7887029636155885e-05 tokens_per_second_per_gpu:5315.014379325853 mfu:25.071829662243022 +Step 16367 | loss:1.0587871074676514 lr:5.7882747671717784e-05 tokens_per_second_per_gpu:5368.256288054022 mfu:25.322980829720134 +Step 16368 | loss:0.8884214758872986 lr:5.7878465647992554e-05 tokens_per_second_per_gpu:4563.8287888967725 mfu:21.528359066711317 +Step 16369 | loss:0.8838422894477844 lr:5.7874183565012354e-05 tokens_per_second_per_gpu:5130.12964895604 mfu:24.19969684450084 +Step 16370 | loss:0.6931599974632263 lr:5.786990142280943e-05 tokens_per_second_per_gpu:4942.89139500298 mfu:23.316462054463948 +Step 16371 | loss:0.855074942111969 lr:5.7865619221415967e-05 tokens_per_second_per_gpu:5319.21525863507 mfu:25.091645926688518 +Step 16372 | loss:0.8278362154960632 lr:5.786133696086417e-05 tokens_per_second_per_gpu:5319.88864252277 mfu:25.09482239337794 +Step 16373 | loss:0.7794017195701599 lr:5.785705464118628e-05 tokens_per_second_per_gpu:4402.4333330454665 mfu:20.76702916455642 +Step 16374 | loss:0.7869699001312256 lr:5.785277226241446e-05 tokens_per_second_per_gpu:5073.392018895512 mfu:23.93205576306742 +Step 16375 | loss:0.8056533336639404 lr:5.784848982458093e-05 tokens_per_second_per_gpu:3411.607274280219 mfu:16.093133593002996 +Step 16376 | loss:0.7933512330055237 lr:5.7844207327717925e-05 tokens_per_second_per_gpu:4290.794707965316 mfu:20.240410722539792 +Step 16377 | loss:0.9546242952346802 lr:5.7839924771857635e-05 tokens_per_second_per_gpu:4382.658339246775 mfu:20.673747144846136 +Step 16378 | loss:0.7084898352622986 lr:5.7835642157032265e-05 tokens_per_second_per_gpu:4481.854126744178 mfu:21.14167059901794 +Step 16379 | loss:0.8860155940055847 lr:5.783135948327403e-05 tokens_per_second_per_gpu:5175.359893283359 mfu:24.413055624068242 +Step 16380 | loss:0.8128282427787781 lr:5.782707675061514e-05 tokens_per_second_per_gpu:5308.6440018332605 mfu:25.041779504711982 +Step 16381 | loss:0.6883082389831543 lr:5.7822793959087795e-05 tokens_per_second_per_gpu:4170.766096612846 mfu:19.67421528379737 +Step 16382 | loss:0.7645348310470581 lr:5.781851110872424e-05 tokens_per_second_per_gpu:4272.60667531763 mfu:20.154614669342298 +Step 16383 | loss:0.7884762287139893 lr:5.7814228199556655e-05 tokens_per_second_per_gpu:4795.603356614646 mfu:22.62167924745551 +Step 16384 | loss:0.8938860297203064 lr:5.780994523161727e-05 tokens_per_second_per_gpu:3253.294647617182 mfu:15.346345922114063 +Step 16385 | loss:0.7320389151573181 lr:5.780566220493827e-05 tokens_per_second_per_gpu:5068.528075395226 mfu:23.909111711702238 +Step 16386 | loss:0.7657633423805237 lr:5.780137911955191e-05 tokens_per_second_per_gpu:4539.272152873263 mfu:21.412521224794276 +Step 16387 | loss:0.8361423015594482 lr:5.779709597549037e-05 tokens_per_second_per_gpu:5009.158034195842 mfu:23.629053097791186 +Step 16388 | loss:0.7392011880874634 lr:5.779281277278589e-05 tokens_per_second_per_gpu:4343.925234514474 mfu:20.491036481274758 +Step 16389 | loss:0.8353102803230286 lr:5.778852951147066e-05 tokens_per_second_per_gpu:4700.929086452461 mfu:22.17508456200434 +Step 16390 | loss:0.7514342665672302 lr:5.7784246191576905e-05 tokens_per_second_per_gpu:4912.510889087963 mfu:23.173152024614 +Step 16391 | loss:0.8236421942710876 lr:5.7779962813136845e-05 tokens_per_second_per_gpu:5321.023042928126 mfu:25.10017355363867 +Step 16392 | loss:0.7426591515541077 lr:5.777567937618269e-05 tokens_per_second_per_gpu:5088.383824405885 mfu:24.002774667525628 +Step 16393 | loss:0.8316803574562073 lr:5.777139588074666e-05 tokens_per_second_per_gpu:4310.56201830661 mfu:20.3336564980706 +Step 16394 | loss:0.8281974196434021 lr:5.776711232686096e-05 tokens_per_second_per_gpu:4220.413345190562 mfu:19.908409826032536 +Step 16395 | loss:0.7213162779808044 lr:5.776282871455784e-05 tokens_per_second_per_gpu:5056.941043567306 mfu:23.854453705618454 +Step 16396 | loss:0.8420464396476746 lr:5.7758545043869485e-05 tokens_per_second_per_gpu:3602.3938217156588 mfu:16.993106288797566 +Step 16397 | loss:0.7867386937141418 lr:5.775426131482811e-05 tokens_per_second_per_gpu:3928.638459009153 mfu:18.532057905985898 +Step 16398 | loss:0.7862123847007751 lr:5.774997752746596e-05 tokens_per_second_per_gpu:4359.24437132451 mfu:20.563299463321954 +Step 16399 | loss:0.7467969059944153 lr:5.7745693681815226e-05 tokens_per_second_per_gpu:4538.3083897959405 mfu:21.407974989923613 +Step 16400 | loss:0.8204066753387451 lr:5.7741409777908154e-05 tokens_per_second_per_gpu:4920.509395441007 mfu:23.210882343767214 +Step 16401 | loss:0.8442505598068237 lr:5.773712581577695e-05 tokens_per_second_per_gpu:4409.257923546921 mfu:20.799221922346987 +Step 16402 | loss:0.6793074011802673 lr:5.773284179545383e-05 tokens_per_second_per_gpu:5347.027850550772 mfu:25.222842668072385 +Step 16403 | loss:0.8733530640602112 lr:5.772855771697101e-05 tokens_per_second_per_gpu:5023.045274858467 mfu:23.69456158140437 +Step 16404 | loss:0.8775641322135925 lr:5.772427358036074e-05 tokens_per_second_per_gpu:3985.8969563296396 mfu:18.802156007153656 +Step 16405 | loss:0.8003568649291992 lr:5.771998938565522e-05 tokens_per_second_per_gpu:4746.477439930862 mfu:22.38994391671241 +Step 16406 | loss:0.8034455180168152 lr:5.7715705132886677e-05 tokens_per_second_per_gpu:4608.8296623335145 mfu:21.74063586465316 +Step 16407 | loss:0.8349643349647522 lr:5.7711420822087325e-05 tokens_per_second_per_gpu:5346.54226894994 mfu:25.220552096812497 +Step 16408 | loss:0.7549097537994385 lr:5.7707136453289404e-05 tokens_per_second_per_gpu:3213.5841125505062 mfu:15.159024552888622 +Step 16409 | loss:0.690645158290863 lr:5.770285202652511e-05 tokens_per_second_per_gpu:5344.334718078488 mfu:25.21013870270484 +Step 16410 | loss:0.7621949315071106 lr:5.7698567541826675e-05 tokens_per_second_per_gpu:3060.9975133227103 mfu:14.439247530372747 +Step 16411 | loss:0.9845868349075317 lr:5.7694282999226355e-05 tokens_per_second_per_gpu:3842.8970961345576 mfu:18.12760126832143 +Step 16412 | loss:0.8053097724914551 lr:5.768999839875634e-05 tokens_per_second_per_gpu:4531.311745785891 mfu:21.374970626377642 +Step 16413 | loss:0.9154186248779297 lr:5.7685713740448865e-05 tokens_per_second_per_gpu:3844.552897406324 mfu:18.13541196542918 +Step 16414 | loss:0.728288471698761 lr:5.7681429024336164e-05 tokens_per_second_per_gpu:3726.332610948874 mfu:17.577746703748605 +Step 16415 | loss:0.8163269758224487 lr:5.7677144250450455e-05 tokens_per_second_per_gpu:3090.133572370502 mfu:14.576687292025458 +Step 16416 | loss:0.8733626008033752 lr:5.7672859418823954e-05 tokens_per_second_per_gpu:3692.5167246049627 mfu:17.418231398279225 +Step 16417 | loss:0.7309293746948242 lr:5.7668574529488895e-05 tokens_per_second_per_gpu:4436.836817716633 mfu:20.92931627159008 +Step 16418 | loss:0.880826473236084 lr:5.766428958247751e-05 tokens_per_second_per_gpu:4827.528645087308 mfu:22.77227628019695 +Step 16419 | loss:0.8015015721321106 lr:5.766000457782205e-05 tokens_per_second_per_gpu:4855.212055253892 mfu:22.90286364922955 +Step 16420 | loss:0.7961617112159729 lr:5.76557195155547e-05 tokens_per_second_per_gpu:5365.2972965357985 mfu:25.30902276187266 +Step 16421 | loss:0.8007344603538513 lr:5.765143439570769e-05 tokens_per_second_per_gpu:4427.8420800241975 mfu:20.88688655923436 +Step 16422 | loss:0.8874709010124207 lr:5.764714921831329e-05 tokens_per_second_per_gpu:3630.2344933132063 mfu:17.124435486831587 +Step 16423 | loss:0.8103687167167664 lr:5.7642863983403695e-05 tokens_per_second_per_gpu:4447.029916733996 mfu:20.977398858777004 +Step 16424 | loss:0.7803304195404053 lr:5.7638578691011146e-05 tokens_per_second_per_gpu:5020.736639062507 mfu:23.68367135246094 +Step 16425 | loss:0.7709647417068481 lr:5.7634293341167875e-05 tokens_per_second_per_gpu:4596.4386831121765 mfu:21.682185501547053 +Step 16426 | loss:0.725817084312439 lr:5.763000793390612e-05 tokens_per_second_per_gpu:5328.06870199838 mfu:25.133409129586447 +Step 16427 | loss:0.849299430847168 lr:5.762572246925808e-05 tokens_per_second_per_gpu:5321.004565195484 mfu:25.10008639102117 +Step 16428 | loss:0.8715538382530212 lr:5.7621436947256015e-05 tokens_per_second_per_gpu:4583.666646522977 mfu:21.62193762582165 +Step 16429 | loss:0.7945542335510254 lr:5.761715136793215e-05 tokens_per_second_per_gpu:5321.705589007949 mfu:25.103393239951565 +Step 16430 | loss:0.8340001702308655 lr:5.761286573131872e-05 tokens_per_second_per_gpu:2796.2524283593893 mfu:13.190399794431466 +Step 16431 | loss:0.8171201944351196 lr:5.760858003744796e-05 tokens_per_second_per_gpu:3871.4860358864407 mfu:18.26246017490716 +Step 16432 | loss:0.8736975193023682 lr:5.76042942863521e-05 tokens_per_second_per_gpu:4399.992055878064 mfu:20.755513243632947 +Step 16433 | loss:0.7899569272994995 lr:5.760000847806337e-05 tokens_per_second_per_gpu:4192.003310683004 mfu:19.774394846008665 +Step 16434 | loss:0.785104513168335 lr:5.7595722612614e-05 tokens_per_second_per_gpu:4948.730624825716 mfu:23.344006697813132 +Step 16435 | loss:1.0096559524536133 lr:5.759143669003623e-05 tokens_per_second_per_gpu:5362.2156777663995 mfu:25.294486240359028 +Step 16436 | loss:0.8016321659088135 lr:5.7587150710362304e-05 tokens_per_second_per_gpu:4643.435127832453 mfu:21.9038757497129 +Step 16437 | loss:0.8061116337776184 lr:5.7582864673624434e-05 tokens_per_second_per_gpu:3957.1674259624756 mfu:18.666633910648404 +Step 16438 | loss:0.883196234703064 lr:5.7578578579854894e-05 tokens_per_second_per_gpu:5338.562938697543 mfu:25.182912234597815 +Step 16439 | loss:0.821191132068634 lr:5.757429242908587e-05 tokens_per_second_per_gpu:5349.294638046303 mfu:25.23353549144263 +Step 16440 | loss:0.7118616700172424 lr:5.757000622134965e-05 tokens_per_second_per_gpu:4642.731692457949 mfu:21.900557525032912 +Step 16441 | loss:0.7395730018615723 lr:5.7565719956678444e-05 tokens_per_second_per_gpu:3612.8456411703028 mfu:17.042409304429405 +Step 16442 | loss:0.8973581790924072 lr:5.756143363510449e-05 tokens_per_second_per_gpu:4340.427807458286 mfu:20.474538521128128 +Step 16443 | loss:0.7373435497283936 lr:5.755714725666003e-05 tokens_per_second_per_gpu:5383.17010919381 mfu:25.39333186859679 +Step 16444 | loss:0.7724869847297668 lr:5.7552860821377295e-05 tokens_per_second_per_gpu:5350.6119959214475 mfu:25.239749693304116 +Step 16445 | loss:0.9725475311279297 lr:5.754857432928854e-05 tokens_per_second_per_gpu:3329.5649466064942 mfu:15.70612593550152 +Step 16446 | loss:0.8334187269210815 lr:5.754428778042598e-05 tokens_per_second_per_gpu:3651.3295258226694 mfu:17.22394435436241 +Step 16447 | loss:0.7960149049758911 lr:5.754000117482188e-05 tokens_per_second_per_gpu:4984.040520705335 mfu:23.51056950116676 +Step 16448 | loss:0.9515054225921631 lr:5.753571451250846e-05 tokens_per_second_per_gpu:4313.518522060157 mfu:20.347602830707636 +Step 16449 | loss:0.6470796465873718 lr:5.753142779351798e-05 tokens_per_second_per_gpu:3993.062835203361 mfu:18.835958680426007 +Step 16450 | loss:0.7772648334503174 lr:5.752714101788267e-05 tokens_per_second_per_gpu:5322.755957290986 mfu:25.108348006353953 +Step 16451 | loss:0.7925360798835754 lr:5.752285418563476e-05 tokens_per_second_per_gpu:3524.590857181766 mfu:16.626096430536613 +Step 16452 | loss:0.9474977850914001 lr:5.7518567296806516e-05 tokens_per_second_per_gpu:4618.654732480569 mfu:21.78698239686664 +Step 16453 | loss:0.7464038133621216 lr:5.751428035143016e-05 tokens_per_second_per_gpu:4867.438804541379 mfu:22.960539311717167 +Step 16454 | loss:0.7477893233299255 lr:5.750999334953795e-05 tokens_per_second_per_gpu:5340.6694887416625 mfu:25.192849190570424 +Step 16455 | loss:0.8968942165374756 lr:5.750570629116212e-05 tokens_per_second_per_gpu:4795.6378683858475 mfu:22.62184204536858 +Step 16456 | loss:0.7570656538009644 lr:5.7501419176334916e-05 tokens_per_second_per_gpu:5322.37767936617 mfu:25.106563604841728 +Step 16457 | loss:0.7392131686210632 lr:5.7497132005088564e-05 tokens_per_second_per_gpu:4993.036921986328 mfu:23.553007060953536 +Step 16458 | loss:0.7429410815238953 lr:5.749284477745534e-05 tokens_per_second_per_gpu:3371.6856132961834 mfu:15.904816306803573 +Step 16459 | loss:0.8486084342002869 lr:5.7488557493467476e-05 tokens_per_second_per_gpu:3663.852793833766 mfu:17.283018746260602 +Step 16460 | loss:0.698547899723053 lr:5.748427015315722e-05 tokens_per_second_per_gpu:5290.1263739312635 mfu:24.954428694465154 +Step 16461 | loss:0.6669695973396301 lr:5.747998275655681e-05 tokens_per_second_per_gpu:5304.3985214239565 mfu:25.021752849267646 +Step 16462 | loss:0.7208719253540039 lr:5.7475695303698495e-05 tokens_per_second_per_gpu:5360.306847196199 mfu:25.285481960879313 +Step 16463 | loss:0.7978823781013489 lr:5.7471407794614516e-05 tokens_per_second_per_gpu:4339.506384220987 mfu:20.47019201050677 +Step 16464 | loss:0.7288012504577637 lr:5.7467120229337124e-05 tokens_per_second_per_gpu:4570.674259463952 mfu:21.560650319334414 +Step 16465 | loss:0.77176433801651 lr:5.7462832607898574e-05 tokens_per_second_per_gpu:4401.41171040962 mfu:20.76220999627634 +Step 16466 | loss:0.7962145209312439 lr:5.74585449303311e-05 tokens_per_second_per_gpu:4775.771633794793 mfu:22.52812962727344 +Step 16467 | loss:0.6779671907424927 lr:5.7454257196666975e-05 tokens_per_second_per_gpu:3684.2896140479793 mfu:17.37942271409213 +Step 16468 | loss:0.9165254235267639 lr:5.744996940693842e-05 tokens_per_second_per_gpu:4756.0755170695065 mfu:22.435219684176225 +Step 16469 | loss:0.764691174030304 lr:5.744568156117768e-05 tokens_per_second_per_gpu:5088.7106892466445 mfu:24.004316544748384 +Step 16470 | loss:0.8087875247001648 lr:5.744139365941704e-05 tokens_per_second_per_gpu:4407.440954665694 mfu:20.790650970127384 +Step 16471 | loss:0.6936437487602234 lr:5.743710570168872e-05 tokens_per_second_per_gpu:5203.773005371043 mfu:24.547085121562656 +Step 16472 | loss:0.8855850696563721 lr:5.743281768802498e-05 tokens_per_second_per_gpu:3934.4484909260996 mfu:18.5594648178318 +Step 16473 | loss:0.7842451333999634 lr:5.742852961845807e-05 tokens_per_second_per_gpu:5348.493837604664 mfu:25.229757979128777 +Step 16474 | loss:0.7847177386283875 lr:5.742424149302025e-05 tokens_per_second_per_gpu:3682.2610695751127 mfu:17.369853723708243 +Step 16475 | loss:0.6124386191368103 lr:5.741995331174375e-05 tokens_per_second_per_gpu:4110.106304603752 mfu:19.388072695262736 +Step 16476 | loss:0.9257262945175171 lr:5.741566507466083e-05 tokens_per_second_per_gpu:4223.627148722264 mfu:19.923569885622996 +Step 16477 | loss:0.5931274890899658 lr:5.741137678180377e-05 tokens_per_second_per_gpu:4973.42816332914 mfu:23.46050920076837 +Step 16478 | loss:0.8710145354270935 lr:5.740708843320477e-05 tokens_per_second_per_gpu:3659.177218321792 mfu:17.26096325883518 +Step 16479 | loss:0.793119490146637 lr:5.740280002889613e-05 tokens_per_second_per_gpu:4680.371248413821 mfu:22.07810973244722 +Step 16480 | loss:0.697795033454895 lr:5.7398511568910085e-05 tokens_per_second_per_gpu:4161.303413574959 mfu:19.62957819340745 +Step 16481 | loss:0.9260959029197693 lr:5.73942230532789e-05 tokens_per_second_per_gpu:5323.609881401739 mfu:25.112376112078852 +Step 16482 | loss:0.7209096550941467 lr:5.738993448203479e-05 tokens_per_second_per_gpu:4048.4456913462873 mfu:19.09720906214183 +Step 16483 | loss:0.7199426293373108 lr:5.7385645855210066e-05 tokens_per_second_per_gpu:4559.2361303045345 mfu:21.50669471254337 +Step 16484 | loss:0.7880984544754028 lr:5.7381357172836946e-05 tokens_per_second_per_gpu:5327.680723587041 mfu:25.131578969224222 +Step 16485 | loss:0.7678421139717102 lr:5.73770684349477e-05 tokens_per_second_per_gpu:5214.911257622947 mfu:24.599626157816683 +Step 16486 | loss:0.8842095732688904 lr:5.737277964157459e-05 tokens_per_second_per_gpu:3334.786913338423 mfu:15.730758843535426 +Step 16487 | loss:0.8297227621078491 lr:5.736849079274984e-05 tokens_per_second_per_gpu:5367.289682521507 mfu:25.31842118650335 +Step 16488 | loss:0.8609935641288757 lr:5.7364201888505766e-05 tokens_per_second_per_gpu:4330.382303770754 mfu:20.427152166294377 +Step 16489 | loss:0.7304169535636902 lr:5.735991292887456e-05 tokens_per_second_per_gpu:4815.30109558126 mfu:22.714596842941848 +Step 16490 | loss:0.6998322010040283 lr:5.735562391388851e-05 tokens_per_second_per_gpu:5314.7705295539145 mfu:25.070679381263727 +Step 16491 | loss:0.7318878173828125 lr:5.735133484357989e-05 tokens_per_second_per_gpu:5332.940823222666 mfu:25.156391756671006 +Step 16492 | loss:0.7482113838195801 lr:5.734704571798094e-05 tokens_per_second_per_gpu:5319.093430383194 mfu:25.09107124203837 +Step 16493 | loss:0.8178780674934387 lr:5.734275653712391e-05 tokens_per_second_per_gpu:3679.315327738629 mfu:17.355958156870166 +Step 16494 | loss:0.8105260729789734 lr:5.7338467301041076e-05 tokens_per_second_per_gpu:3272.8543235875977 mfu:15.438612250891358 +Step 16495 | loss:0.8481415510177612 lr:5.733417800976471e-05 tokens_per_second_per_gpu:4071.9395378459776 mfu:19.208033544544417 +Step 16496 | loss:0.8687477707862854 lr:5.7329888663327034e-05 tokens_per_second_per_gpu:3576.6410413664576 mfu:16.871626030012976 +Step 16497 | loss:0.8247313499450684 lr:5.732559926176033e-05 tokens_per_second_per_gpu:5336.501313516371 mfu:25.173187196868696 +Step 16498 | loss:0.8708497285842896 lr:5.732130980509687e-05 tokens_per_second_per_gpu:3670.247323411224 mfu:17.313182833296718 +Step 16499 | loss:0.9421300292015076 lr:5.731702029336891e-05 tokens_per_second_per_gpu:3972.691891268728 mfu:18.73986546224491 +Step 16500 | loss:0.8538386225700378 lr:5.731273072660869e-05 tokens_per_second_per_gpu:3718.9966266645883 mfu:17.543141614231647 +Step 16501 | loss:0.6782373189926147 lr:5.73084411048485e-05 tokens_per_second_per_gpu:5337.154647637754 mfu:25.176269085389556 +Step 16502 | loss:0.9322125911712646 lr:5.730415142812059e-05 tokens_per_second_per_gpu:3462.4770523429406 mfu:16.33309501540397 +Step 16503 | loss:0.7135132551193237 lr:5.7299861696457225e-05 tokens_per_second_per_gpu:5325.5154758667995 mfu:25.12136512629849 +Step 16504 | loss:0.72474604845047 lr:5.729557190989067e-05 tokens_per_second_per_gpu:3757.7723141925317 mfu:17.726053148115447 +Step 16505 | loss:0.7636217474937439 lr:5.7291282068453166e-05 tokens_per_second_per_gpu:3747.597505713099 mfu:17.67805684051663 +Step 16506 | loss:0.7166622877120972 lr:5.728699217217703e-05 tokens_per_second_per_gpu:4417.934949931569 mfu:20.840152936256732 +Step 16507 | loss:0.8011540770530701 lr:5.728270222109448e-05 tokens_per_second_per_gpu:5335.50095262691 mfu:25.16846832387394 +Step 16508 | loss:0.7725781202316284 lr:5.727841221523781e-05 tokens_per_second_per_gpu:5343.253918989433 mfu:25.205040388998427 +Step 16509 | loss:0.7873087525367737 lr:5.727412215463926e-05 tokens_per_second_per_gpu:4989.766377704261 mfu:23.537579345562634 +Step 16510 | loss:0.7540979385375977 lr:5.726983203933112e-05 tokens_per_second_per_gpu:5353.793038903308 mfu:25.254755215791455 +Step 16511 | loss:0.8673641681671143 lr:5.726554186934564e-05 tokens_per_second_per_gpu:4080.9451077393946 mfu:19.25051435424033 +Step 16512 | loss:0.8046594262123108 lr:5.726125164471509e-05 tokens_per_second_per_gpu:4363.052904900449 mfu:20.581264966002493 +Step 16513 | loss:0.715578556060791 lr:5.725696136547174e-05 tokens_per_second_per_gpu:4476.765751708954 mfu:21.117667865810482 +Step 16514 | loss:0.8602833151817322 lr:5.725267103164785e-05 tokens_per_second_per_gpu:4789.048560786079 mfu:22.590759157168673 +Step 16515 | loss:0.7081335186958313 lr:5.7248380643275713e-05 tokens_per_second_per_gpu:5327.989813141429 mfu:25.133036997388384 +Step 16516 | loss:0.7760297060012817 lr:5.724409020038757e-05 tokens_per_second_per_gpu:3680.634565928784 mfu:17.362181228498372 +Step 16517 | loss:0.8565788865089417 lr:5.723979970301571e-05 tokens_per_second_per_gpu:5383.0866086915 mfu:25.39293798248074 +Step 16518 | loss:0.8091260194778442 lr:5.723550915119238e-05 tokens_per_second_per_gpu:5313.646794010693 mfu:25.065378528977295 +Step 16519 | loss:0.785284698009491 lr:5.723121854494986e-05 tokens_per_second_per_gpu:4775.888422107806 mfu:22.528680537672066 +Step 16520 | loss:0.9050986170768738 lr:5.7226927884320435e-05 tokens_per_second_per_gpu:3887.5291214387853 mfu:18.338138146690223 +Step 16521 | loss:0.7811273336410522 lr:5.722263716933636e-05 tokens_per_second_per_gpu:4415.044726434574 mfu:20.82651925889843 +Step 16522 | loss:0.7216746211051941 lr:5.721834640002991e-05 tokens_per_second_per_gpu:5054.787197896463 mfu:23.844293648105122 +Step 16523 | loss:0.7287477254867554 lr:5.7214055576433346e-05 tokens_per_second_per_gpu:4008.7530244736304 mfu:18.909971980236204 +Step 16524 | loss:0.8007266521453857 lr:5.720976469857897e-05 tokens_per_second_per_gpu:5339.563219756906 mfu:25.187630731020377 +Step 16525 | loss:0.711234986782074 lr:5.7205473766499005e-05 tokens_per_second_per_gpu:5360.563128568096 mfu:25.286690883836663 +Step 16526 | loss:0.7480529546737671 lr:5.720118278022577e-05 tokens_per_second_per_gpu:3396.4464193663885 mfu:16.021617253665877 +Step 16527 | loss:0.7193589210510254 lr:5.719689173979151e-05 tokens_per_second_per_gpu:4369.9014533385 mfu:20.613570737467633 +Step 16528 | loss:0.7495567798614502 lr:5.719260064522853e-05 tokens_per_second_per_gpu:4694.079602624709 mfu:22.142774378146385 +Step 16529 | loss:0.6802110075950623 lr:5.718830949656908e-05 tokens_per_second_per_gpu:4655.612373438221 mfu:21.96131789488765 +Step 16530 | loss:0.7515708804130554 lr:5.718401829384541e-05 tokens_per_second_per_gpu:5355.325152763106 mfu:25.261982458273703 +Step 16531 | loss:0.8018898963928223 lr:5.7179727037089856e-05 tokens_per_second_per_gpu:3858.748200356504 mfu:18.20237363141361 +Step 16532 | loss:0.8285233974456787 lr:5.717543572633464e-05 tokens_per_second_per_gpu:4900.282336495765 mfu:23.115467855630385 +Step 16533 | loss:0.8475174903869629 lr:5.717114436161205e-05 tokens_per_second_per_gpu:5303.206898563825 mfu:25.016131760924583 +Step 16534 | loss:0.8550093770027161 lr:5.7166852942954365e-05 tokens_per_second_per_gpu:3888.8247729124496 mfu:18.34424995575243 +Step 16535 | loss:0.8679648637771606 lr:5.716256147039389e-05 tokens_per_second_per_gpu:4373.787586730743 mfu:20.63190229172157 +Step 16536 | loss:0.8554911017417908 lr:5.715826994396285e-05 tokens_per_second_per_gpu:4533.449389419432 mfu:21.385054256116668 +Step 16537 | loss:0.8393161296844482 lr:5.715397836369356e-05 tokens_per_second_per_gpu:4036.477239521692 mfu:19.04075183285676 +Step 16538 | loss:0.7628209590911865 lr:5.714968672961829e-05 tokens_per_second_per_gpu:4171.21550514916 mfu:19.676335220540167 +Step 16539 | loss:0.7698066830635071 lr:5.7145395041769326e-05 tokens_per_second_per_gpu:2872.2447703946327 mfu:13.548868637442233 +Step 16540 | loss:0.7713022232055664 lr:5.714110330017891e-05 tokens_per_second_per_gpu:3731.0122981690183 mfu:17.599821586803 +Step 16541 | loss:0.7124437093734741 lr:5.7136811504879336e-05 tokens_per_second_per_gpu:4386.263292541923 mfu:20.69075232460781 +Step 16542 | loss:0.7710607647895813 lr:5.7132519655902925e-05 tokens_per_second_per_gpu:5328.407509466516 mfu:25.135007342220163 +Step 16543 | loss:0.6950694918632507 lr:5.7128227753281904e-05 tokens_per_second_per_gpu:4623.057077837886 mfu:21.807749011034677 +Step 16544 | loss:0.7615007758140564 lr:5.712393579704858e-05 tokens_per_second_per_gpu:4050.5799568248544 mfu:19.107276756547467 +Step 16545 | loss:0.8869124054908752 lr:5.7119643787235225e-05 tokens_per_second_per_gpu:4048.8354259437097 mfu:19.09904750673345 +Step 16546 | loss:0.7618597745895386 lr:5.711535172387411e-05 tokens_per_second_per_gpu:5358.1364529344955 mfu:25.275243840839412 +Step 16547 | loss:0.726891279220581 lr:5.711105960699754e-05 tokens_per_second_per_gpu:5366.5599149492955 mfu:25.314978748354044 +Step 16548 | loss:0.6858494877815247 lr:5.710676743663776e-05 tokens_per_second_per_gpu:4563.684390899299 mfu:21.527677916720393 +Step 16549 | loss:0.6849772930145264 lr:5.7102475212827114e-05 tokens_per_second_per_gpu:2868.259096727317 mfu:13.530067534727454 +Step 16550 | loss:0.8393096327781677 lr:5.709818293559781e-05 tokens_per_second_per_gpu:4623.864060072348 mfu:21.81155568392 +Step 16551 | loss:0.8764731287956238 lr:5.7093890604982195e-05 tokens_per_second_per_gpu:5344.321966693671 mfu:25.210078552247044 +Step 16552 | loss:0.7242896556854248 lr:5.7089598221012496e-05 tokens_per_second_per_gpu:5231.862087221261 mfu:24.679586113141685 +Step 16553 | loss:0.719618558883667 lr:5.7085305783721043e-05 tokens_per_second_per_gpu:4682.737248451031 mfu:22.089270558303973 +Step 16554 | loss:0.7168598771095276 lr:5.70810132931401e-05 tokens_per_second_per_gpu:5367.963435374744 mfu:25.321599393666165 +Step 16555 | loss:0.9098279476165771 lr:5.707672074930194e-05 tokens_per_second_per_gpu:5024.173937961434 mfu:23.699885677830846 +Step 16556 | loss:0.8016651272773743 lr:5.7072428152238874e-05 tokens_per_second_per_gpu:4457.211185524776 mfu:21.025425640766688 +Step 16557 | loss:0.8033311367034912 lr:5.7068135501983165e-05 tokens_per_second_per_gpu:5334.944443966068 mfu:25.165843177570583 +Step 16558 | loss:0.8365831971168518 lr:5.7063842798567113e-05 tokens_per_second_per_gpu:3670.4358315735176 mfu:17.31407205846107 +Step 16559 | loss:0.7160172462463379 lr:5.705955004202299e-05 tokens_per_second_per_gpu:5339.064811527052 mfu:25.185279654362944 +Step 16560 | loss:0.762028157711029 lr:5.705525723238311e-05 tokens_per_second_per_gpu:4909.114397574651 mfu:23.157130194644502 +Step 16561 | loss:0.9036065340042114 lr:5.705096436967972e-05 tokens_per_second_per_gpu:4774.169464713673 mfu:22.52057192235956 +Step 16562 | loss:0.6828933954238892 lr:5.704667145394512e-05 tokens_per_second_per_gpu:5082.163568298586 mfu:23.973432658182375 +Step 16563 | loss:0.85502690076828 lr:5.7042378485211624e-05 tokens_per_second_per_gpu:3326.370817222562 mfu:15.691058682222844 +Step 16564 | loss:0.8339545726776123 lr:5.703808546351149e-05 tokens_per_second_per_gpu:5258.458427252674 mfu:24.80504558687339 +Step 16565 | loss:0.7050026655197144 lr:5.703379238887703e-05 tokens_per_second_per_gpu:4384.340624343087 mfu:20.68168277067288 +Step 16566 | loss:0.6829625368118286 lr:5.70294992613405e-05 tokens_per_second_per_gpu:5204.795033117024 mfu:24.551906200816365 +Step 16567 | loss:0.7430770993232727 lr:5.7025206080934236e-05 tokens_per_second_per_gpu:4908.01287934497 mfu:23.151934145216707 +Step 16568 | loss:0.7117837071418762 lr:5.7020912847690476e-05 tokens_per_second_per_gpu:5362.242801420698 mfu:25.29461418726413 +Step 16569 | loss:0.9346098899841309 lr:5.701661956164155e-05 tokens_per_second_per_gpu:5130.5226660296175 mfu:24.20155077309253 +Step 16570 | loss:0.7352263331413269 lr:5.701232622281971e-05 tokens_per_second_per_gpu:5336.702905548757 mfu:25.17413814088061 +Step 16571 | loss:0.720329225063324 lr:5.70080328312573e-05 tokens_per_second_per_gpu:3909.384804521702 mfu:18.441235132756393 +Step 16572 | loss:0.6970759630203247 lr:5.7003739386986556e-05 tokens_per_second_per_gpu:5340.6507510889805 mfu:25.192760801865752 +Step 16573 | loss:0.8469672799110413 lr:5.69994458900398e-05 tokens_per_second_per_gpu:4616.521462187214 mfu:21.77691939691944 +Step 16574 | loss:0.8771783709526062 lr:5.699515234044933e-05 tokens_per_second_per_gpu:5351.179814068343 mfu:25.242428188382792 +Step 16575 | loss:0.8611152768135071 lr:5.699085873824743e-05 tokens_per_second_per_gpu:4662.86107970511 mfu:21.99551127910538 +Step 16576 | loss:0.7694615125656128 lr:5.698656508346638e-05 tokens_per_second_per_gpu:4227.535837743308 mfu:19.942007838626374 +Step 16577 | loss:0.7729325294494629 lr:5.698227137613848e-05 tokens_per_second_per_gpu:4222.893355523324 mfu:19.92010845790557 +Step 16578 | loss:0.7543676495552063 lr:5.697797761629603e-05 tokens_per_second_per_gpu:4452.877716972324 mfu:21.004983930238804 +Step 16579 | loss:0.8120435476303101 lr:5.697368380397131e-05 tokens_per_second_per_gpu:3352.565214534415 mfu:15.814622123567888 +Step 16580 | loss:0.7372807264328003 lr:5.696938993919664e-05 tokens_per_second_per_gpu:4621.556207595683 mfu:21.800669150027336 +Step 16581 | loss:0.887793779373169 lr:5.696509602200429e-05 tokens_per_second_per_gpu:4940.712252357875 mfu:23.30618266680691 +Step 16582 | loss:0.747146487236023 lr:5.6960802052426566e-05 tokens_per_second_per_gpu:5367.708463689459 mfu:25.320396648724017 +Step 16583 | loss:0.6710273623466492 lr:5.695650803049577e-05 tokens_per_second_per_gpu:5313.7514285520165 mfu:25.065872107961212 +Step 16584 | loss:0.8347627520561218 lr:5.6952213956244174e-05 tokens_per_second_per_gpu:4966.727455640842 mfu:23.428900819344047 +Step 16585 | loss:0.7474706172943115 lr:5.69479198297041e-05 tokens_per_second_per_gpu:4305.139382506393 mfu:20.308076999803905 +Step 16586 | loss:0.7404281497001648 lr:5.6943625650907825e-05 tokens_per_second_per_gpu:5001.5847047135285 mfu:23.593328410479863 +Step 16587 | loss:0.7859582304954529 lr:5.693933141988765e-05 tokens_per_second_per_gpu:4500.312070843872 mfu:21.2287398705862 +Step 16588 | loss:0.7215695381164551 lr:5.693503713667589e-05 tokens_per_second_per_gpu:5031.00189635168 mfu:23.73209432252747 +Step 16589 | loss:0.7348475456237793 lr:5.693074280130484e-05 tokens_per_second_per_gpu:5209.689627841491 mfu:24.574994839235643 +Step 16590 | loss:0.8858450055122375 lr:5.6926448413806775e-05 tokens_per_second_per_gpu:3639.7960688513303 mfu:17.16953907001793 +Step 16591 | loss:0.7267333269119263 lr:5.692215397421401e-05 tokens_per_second_per_gpu:3956.6366425769143 mfu:18.664130114857727 +Step 16592 | loss:0.6345421671867371 lr:5.6917859482558845e-05 tokens_per_second_per_gpu:3630.9651950235893 mfu:17.127882331470033 +Step 16593 | loss:0.7887529134750366 lr:5.691356493887358e-05 tokens_per_second_per_gpu:3997.083129876651 mfu:18.854923096332918 +Step 16594 | loss:0.7797999382019043 lr:5.69092703431905e-05 tokens_per_second_per_gpu:5410.203954510347 mfu:25.520855129404527 +Step 16595 | loss:0.750199019908905 lr:5.690497569554192e-05 tokens_per_second_per_gpu:3747.0508762442196 mfu:17.67547829604732 +Step 16596 | loss:0.6652708053588867 lr:5.6900680995960144e-05 tokens_per_second_per_gpu:4198.377912282979 mfu:19.804464929374934 +Step 16597 | loss:0.7643811106681824 lr:5.6896386244477454e-05 tokens_per_second_per_gpu:5364.369209759448 mfu:25.304644818200682 +Step 16598 | loss:0.8388874530792236 lr:5.689209144112617e-05 tokens_per_second_per_gpu:5350.168200920861 mfu:25.237656236567116 +Step 16599 | loss:0.7519915699958801 lr:5.688779658593859e-05 tokens_per_second_per_gpu:4408.918233228185 mfu:20.79761954515637 +Step 16600 | loss:0.742195725440979 lr:5.6883501678947005e-05 tokens_per_second_per_gpu:4695.173726339951 mfu:22.147935546387934 +Step 16601 | loss:0.7980010509490967 lr:5.687920672018373e-05 tokens_per_second_per_gpu:3244.635462262996 mfu:15.305499067389 +Step 16602 | loss:0.7088137865066528 lr:5.6874911709681056e-05 tokens_per_second_per_gpu:4730.316385234055 mfu:22.313709464347316 +Step 16603 | loss:0.765468418598175 lr:5.687061664747132e-05 tokens_per_second_per_gpu:3944.7730887631305 mfu:18.60816770738711 +Step 16604 | loss:0.8586544990539551 lr:5.6866321533586764e-05 tokens_per_second_per_gpu:4443.736731634866 mfu:20.96186434278909 +Step 16605 | loss:0.713392972946167 lr:5.686202636805975e-05 tokens_per_second_per_gpu:3243.3106645376197 mfu:15.299249770484602 +Step 16606 | loss:0.6911503672599792 lr:5.685773115092254e-05 tokens_per_second_per_gpu:5078.032545475926 mfu:23.953945918702715 +Step 16607 | loss:0.8280785083770752 lr:5.6853435882207484e-05 tokens_per_second_per_gpu:5329.6371350859645 mfu:25.140807696062865 +Step 16608 | loss:0.771051287651062 lr:5.6849140561946845e-05 tokens_per_second_per_gpu:5325.919596558083 mfu:25.12327143254937 +Step 16609 | loss:0.8322679996490479 lr:5.6844845190172955e-05 tokens_per_second_per_gpu:5254.408971162495 mfu:24.785943611587562 +Step 16610 | loss:0.9499863386154175 lr:5.6840549766918104e-05 tokens_per_second_per_gpu:4404.126463720642 mfu:20.775015950829108 +Step 16611 | loss:0.7084238529205322 lr:5.683625429221462e-05 tokens_per_second_per_gpu:3987.34679187751 mfu:18.808995128800408 +Step 16612 | loss:0.6429184675216675 lr:5.683195876609478e-05 tokens_per_second_per_gpu:5358.278552509593 mfu:25.275914148779837 +Step 16613 | loss:0.7641189098358154 lr:5.682766318859089e-05 tokens_per_second_per_gpu:5005.769550422424 mfu:23.613069041697027 +Step 16614 | loss:0.8713822364807129 lr:5.682336755973532e-05 tokens_per_second_per_gpu:5001.052237861624 mfu:23.590816673491215 +Step 16615 | loss:0.7948523163795471 lr:5.681907187956029e-05 tokens_per_second_per_gpu:5322.434796928195 mfu:25.10683303812706 +Step 16616 | loss:0.8040473461151123 lr:5.681477614809818e-05 tokens_per_second_per_gpu:5103.296979828953 mfu:24.07312256610293 +Step 16617 | loss:0.7843905091285706 lr:5.6810480365381255e-05 tokens_per_second_per_gpu:5321.34590009317 mfu:25.101696525971292 +Step 16618 | loss:0.7453542947769165 lr:5.680618453144184e-05 tokens_per_second_per_gpu:5223.587888854734 mfu:24.640555307722966 +Step 16619 | loss:0.8532277941703796 lr:5.680188864631224e-05 tokens_per_second_per_gpu:5332.14633644313 mfu:25.15264402697851 +Step 16620 | loss:0.7973937392234802 lr:5.679759271002476e-05 tokens_per_second_per_gpu:5318.171306392702 mfu:25.086721425844647 +Step 16621 | loss:0.7417244911193848 lr:5.6793296722611735e-05 tokens_per_second_per_gpu:5171.876463960974 mfu:24.396623693620228 +Step 16622 | loss:0.7135930061340332 lr:5.678900068410544e-05 tokens_per_second_per_gpu:5316.069980994331 mfu:25.076809115417447 +Step 16623 | loss:0.7238227128982544 lr:5.6784704594538216e-05 tokens_per_second_per_gpu:5338.797367259767 mfu:25.18401807412346 +Step 16624 | loss:0.6969108581542969 lr:5.6780408453942356e-05 tokens_per_second_per_gpu:4994.094200776125 mfu:23.557994425395375 +Step 16625 | loss:0.7310659885406494 lr:5.67761122623502e-05 tokens_per_second_per_gpu:4121.594982322117 mfu:19.44226674823081 +Step 16626 | loss:0.7883679270744324 lr:5.677181601979402e-05 tokens_per_second_per_gpu:5343.3027638635895 mfu:25.205270798603618 +Step 16627 | loss:0.762386679649353 lr:5.676751972630615e-05 tokens_per_second_per_gpu:4037.0391997517627 mfu:19.04340269514231 +Step 16628 | loss:0.7104657292366028 lr:5.6763223381918904e-05 tokens_per_second_per_gpu:5385.274543655316 mfu:25.403258845005254 +Step 16629 | loss:0.8031702637672424 lr:5.6758926986664595e-05 tokens_per_second_per_gpu:4608.5845529079415 mfu:21.73947964167275 +Step 16630 | loss:0.7605825066566467 lr:5.675463054057553e-05 tokens_per_second_per_gpu:5329.4560029511595 mfu:25.13995326487866 +Step 16631 | loss:0.8719884157180786 lr:5.6750334043684014e-05 tokens_per_second_per_gpu:5183.768754859007 mfu:24.4527216588203 +Step 16632 | loss:0.7746713161468506 lr:5.6746037496022395e-05 tokens_per_second_per_gpu:4557.497662177657 mfu:21.49849406177564 +Step 16633 | loss:0.779543399810791 lr:5.6741740897622954e-05 tokens_per_second_per_gpu:4662.136454307898 mfu:21.992093097472893 +Step 16634 | loss:0.7143831253051758 lr:5.673744424851802e-05 tokens_per_second_per_gpu:4649.333132935391 mfu:21.931697646086896 +Step 16635 | loss:0.7634692788124084 lr:5.6733147548739916e-05 tokens_per_second_per_gpu:4283.674473535125 mfu:20.20682336189532 +Step 16636 | loss:0.7984771728515625 lr:5.672885079832095e-05 tokens_per_second_per_gpu:4105.387311458297 mfu:19.365812399452366 +Step 16637 | loss:0.8883084058761597 lr:5.672455399729344e-05 tokens_per_second_per_gpu:4046.517747260296 mfu:19.088114620947024 +Step 16638 | loss:0.8745123744010925 lr:5.672025714568968e-05 tokens_per_second_per_gpu:4665.371141663125 mfu:22.00735166962207 +Step 16639 | loss:0.7169148921966553 lr:5.6715960243542024e-05 tokens_per_second_per_gpu:4643.870373117762 mfu:21.905928875984205 +Step 16640 | loss:0.8171640038490295 lr:5.6711663290882776e-05 tokens_per_second_per_gpu:3912.69600407796 mfu:18.456854626012163 +Step 16641 | loss:0.769572377204895 lr:5.6707366287744254e-05 tokens_per_second_per_gpu:5301.383460734061 mfu:25.007530293571012 +Step 16642 | loss:0.739975094795227 lr:5.670306923415876e-05 tokens_per_second_per_gpu:5265.37101442473 mfu:24.837653439972968 +Step 16643 | loss:0.7241538763046265 lr:5.669877213015865e-05 tokens_per_second_per_gpu:5321.495235104849 mfu:25.102400964700557 +Step 16644 | loss:0.8156723976135254 lr:5.66944749757762e-05 tokens_per_second_per_gpu:2748.016038395911 mfu:12.962860512996963 +Step 16645 | loss:0.8937907814979553 lr:5.669017777104376e-05 tokens_per_second_per_gpu:5294.187905013855 mfu:24.973587629550398 +Step 16646 | loss:0.7638493776321411 lr:5.668588051599364e-05 tokens_per_second_per_gpu:5315.587891039964 mfu:25.07453501484967 +Step 16647 | loss:0.99103844165802 lr:5.6681583210658154e-05 tokens_per_second_per_gpu:3779.885879156871 mfu:17.830366607015478 +Step 16648 | loss:0.826870858669281 lr:5.667728585506963e-05 tokens_per_second_per_gpu:5289.427837569346 mfu:24.951133579263395 +Step 16649 | loss:0.8359465003013611 lr:5.667298844926038e-05 tokens_per_second_per_gpu:5327.102963216319 mfu:25.12885357497944 +Step 16650 | loss:0.8195019364356995 lr:5.666869099326275e-05 tokens_per_second_per_gpu:5171.721633190709 mfu:24.395893330460034 +Step 16651 | loss:0.8433051705360413 lr:5.6664393487109024e-05 tokens_per_second_per_gpu:3780.3436229179247 mfu:17.832525862435503 +Step 16652 | loss:0.7326673865318298 lr:5.666009593083156e-05 tokens_per_second_per_gpu:5285.36401907759 mfu:24.931963854078848 +Step 16653 | loss:0.9211950898170471 lr:5.665579832446266e-05 tokens_per_second_per_gpu:4482.880309219416 mfu:21.14651127683842 +Step 16654 | loss:0.7299848198890686 lr:5.665150066803465e-05 tokens_per_second_per_gpu:5388.436049864985 mfu:25.418172209204844 +Step 16655 | loss:0.9108156561851501 lr:5.6647202961579856e-05 tokens_per_second_per_gpu:4005.3549306905256 mfu:18.893942592086713 +Step 16656 | loss:1.0099982023239136 lr:5.664290520513059e-05 tokens_per_second_per_gpu:5312.468440016511 mfu:25.059820032138816 +Step 16657 | loss:0.7214440703392029 lr:5.66386073987192e-05 tokens_per_second_per_gpu:5289.056773029784 mfu:24.949383204520146 +Step 16658 | loss:0.8384301662445068 lr:5.6634309542378e-05 tokens_per_second_per_gpu:3719.6981429218035 mfu:17.546450786107947 +Step 16659 | loss:0.9711047410964966 lr:5.663001163613929e-05 tokens_per_second_per_gpu:5361.056619056393 mfu:25.28901876266812 +Step 16660 | loss:0.8481405377388 lr:5.662571368003543e-05 tokens_per_second_per_gpu:4672.713164773161 mfu:22.04198524531043 +Step 16661 | loss:0.7166373133659363 lr:5.6621415674098744e-05 tokens_per_second_per_gpu:3237.86284130284 mfu:15.273551458791657 +Step 16662 | loss:0.9914262890815735 lr:5.6617117618361526e-05 tokens_per_second_per_gpu:5323.0472787124345 mfu:25.109722219203487 +Step 16663 | loss:0.7810648083686829 lr:5.661281951285613e-05 tokens_per_second_per_gpu:4316.36611211166 mfu:20.361035398829536 +Step 16664 | loss:0.9592115879058838 lr:5.6608521357614874e-05 tokens_per_second_per_gpu:4445.864928664013 mfu:20.971903411283435 +Step 16665 | loss:0.7785055637359619 lr:5.660422315267009e-05 tokens_per_second_per_gpu:5011.891168683375 mfu:23.64194575948971 +Step 16666 | loss:0.8096503615379333 lr:5.659992489805409e-05 tokens_per_second_per_gpu:4347.159186119652 mfu:20.50629158276542 +Step 16667 | loss:0.7762242555618286 lr:5.659562659379921e-05 tokens_per_second_per_gpu:3506.3402317250884 mfu:16.54000511637967 +Step 16668 | loss:0.7086685299873352 lr:5.65913282399378e-05 tokens_per_second_per_gpu:4455.094090049031 mfu:21.015438940193867 +Step 16669 | loss:0.8581704497337341 lr:5.658702983650216e-05 tokens_per_second_per_gpu:5304.8146552664 mfu:25.02371582360221 +Step 16670 | loss:0.8175148963928223 lr:5.658273138352462e-05 tokens_per_second_per_gpu:4841.547959431423 mfu:22.838407777899814 +Step 16671 | loss:0.824183464050293 lr:5.657843288103753e-05 tokens_per_second_per_gpu:3858.853570181497 mfu:18.202870678856048 +Step 16672 | loss:0.8620682954788208 lr:5.6574134329073205e-05 tokens_per_second_per_gpu:5340.309057214755 mfu:25.191148973562044 +Step 16673 | loss:0.723763108253479 lr:5.656983572766398e-05 tokens_per_second_per_gpu:4451.418911403651 mfu:20.99810249547347 +Step 16674 | loss:0.6203730702400208 lr:5.6565537076842165e-05 tokens_per_second_per_gpu:4120.966192783549 mfu:19.439300640694828 +Step 16675 | loss:0.8881428241729736 lr:5.656123837664011e-05 tokens_per_second_per_gpu:4480.205048108039 mfu:21.133891613729443 +Step 16676 | loss:0.8309985399246216 lr:5.6556939627090164e-05 tokens_per_second_per_gpu:4987.58181223855 mfu:23.527274377535573 +Step 16677 | loss:0.7786394357681274 lr:5.655264082822462e-05 tokens_per_second_per_gpu:4351.208657848478 mfu:20.525393631821412 +Step 16678 | loss:0.8033658266067505 lr:5.654834198007584e-05 tokens_per_second_per_gpu:5344.326683341672 mfu:25.210100801480408 +Step 16679 | loss:0.6889066100120544 lr:5.6544043082676145e-05 tokens_per_second_per_gpu:4917.784238669314 mfu:23.19802731431545 +Step 16680 | loss:0.6539592146873474 lr:5.653974413605785e-05 tokens_per_second_per_gpu:4730.927138869593 mfu:22.316590493451468 +Step 16681 | loss:0.7516492009162903 lr:5.653544514025332e-05 tokens_per_second_per_gpu:4789.266072716241 mfu:22.591785198054605 +Step 16682 | loss:0.886425793170929 lr:5.653114609529488e-05 tokens_per_second_per_gpu:4990.054507464845 mfu:23.538938502803315 +Step 16683 | loss:0.8299254179000854 lr:5.6526847001214846e-05 tokens_per_second_per_gpu:4920.313547168669 mfu:23.20995849405144 +Step 16684 | loss:0.8750947117805481 lr:5.652254785804556e-05 tokens_per_second_per_gpu:5328.7960166444855 mfu:25.136839996864804 +Step 16685 | loss:0.6633801460266113 lr:5.651824866581935e-05 tokens_per_second_per_gpu:5108.7772074345885 mfu:24.09897373474196 +Step 16686 | loss:0.9656768441200256 lr:5.651394942456858e-05 tokens_per_second_per_gpu:3702.2601839831877 mfu:17.464192958571903 +Step 16687 | loss:0.8784483671188354 lr:5.650965013432556e-05 tokens_per_second_per_gpu:3777.0920941321447 mfu:17.817187846384964 +Step 16688 | loss:0.8330826759338379 lr:5.650535079512263e-05 tokens_per_second_per_gpu:4651.544781305377 mfu:21.942130368794242 +Step 16689 | loss:0.8189786076545715 lr:5.650105140699212e-05 tokens_per_second_per_gpu:4754.8220412472165 mfu:22.429306824857456 +Step 16690 | loss:0.8124254941940308 lr:5.6496751969966386e-05 tokens_per_second_per_gpu:3627.3091961853315 mfu:17.110636361172247 +Step 16691 | loss:0.7426640391349792 lr:5.649245248407774e-05 tokens_per_second_per_gpu:5362.045822427584 mfu:25.29368500374553 +Step 16692 | loss:0.7068408727645874 lr:5.648815294935853e-05 tokens_per_second_per_gpu:5026.562127430423 mfu:23.71115117502321 +Step 16693 | loss:0.6883442997932434 lr:5.64838533658411e-05 tokens_per_second_per_gpu:3382.4710848653376 mfu:15.955693216386477 +Step 16694 | loss:0.7091212272644043 lr:5.647955373355778e-05 tokens_per_second_per_gpu:4148.006603481453 mfu:19.56685487152664 +Step 16695 | loss:0.7320622801780701 lr:5.6475254052540903e-05 tokens_per_second_per_gpu:4846.936121378443 mfu:22.863824657119135 +Step 16696 | loss:0.8169270157814026 lr:5.647095432282282e-05 tokens_per_second_per_gpu:4353.29568893049 mfu:20.53523851351948 +Step 16697 | loss:0.7713841795921326 lr:5.646665454443587e-05 tokens_per_second_per_gpu:5354.24522489062 mfu:25.25688825424449 +Step 16698 | loss:0.6623886823654175 lr:5.646235471741237e-05 tokens_per_second_per_gpu:5331.268620093946 mfu:25.148503689205562 +Step 16699 | loss:0.7511657476425171 lr:5.645805484178469e-05 tokens_per_second_per_gpu:5310.075335557836 mfu:25.048531350101523 +Step 16700 | loss:0.7717264890670776 lr:5.645375491758515e-05 tokens_per_second_per_gpu:4290.624880507354 mfu:20.239609617445037 +Step 16701 | loss:0.7241979837417603 lr:5.6449454944846106e-05 tokens_per_second_per_gpu:3956.0815558722315 mfu:18.66151167616418 +Step 16702 | loss:0.6643404364585876 lr:5.6445154923599874e-05 tokens_per_second_per_gpu:4161.357423157211 mfu:19.62983296582192 +Step 16703 | loss:0.72884202003479 lr:5.644085485387881e-05 tokens_per_second_per_gpu:5355.290472865425 mfu:25.261818867279338 +Step 16704 | loss:0.8021204471588135 lr:5.6436554735715256e-05 tokens_per_second_per_gpu:4737.445464921836 mfu:22.347338549581 +Step 16705 | loss:0.7403327226638794 lr:5.643225456914156e-05 tokens_per_second_per_gpu:5319.086071009716 mfu:25.091036526618037 +Step 16706 | loss:0.8719390630722046 lr:5.642795435419005e-05 tokens_per_second_per_gpu:3015.8442647977367 mfu:14.226251953141842 +Step 16707 | loss:0.8125596046447754 lr:5.642365409089309e-05 tokens_per_second_per_gpu:3943.7170283002765 mfu:18.60318609000133 +Step 16708 | loss:0.8851991891860962 lr:5.641935377928299e-05 tokens_per_second_per_gpu:4745.9645296936305 mfu:22.387524431611883 +Step 16709 | loss:0.7656658887863159 lr:5.6415053419392114e-05 tokens_per_second_per_gpu:5346.390301400699 mfu:25.21983523995439 +Step 16710 | loss:0.8267214298248291 lr:5.64107530112528e-05 tokens_per_second_per_gpu:5228.681863111854 mfu:24.664584453415426 +Step 16711 | loss:0.8422711491584778 lr:5.640645255489741e-05 tokens_per_second_per_gpu:4942.8233138094 mfu:23.31614090385798 +Step 16712 | loss:0.8385348916053772 lr:5.640215205035826e-05 tokens_per_second_per_gpu:5327.342494123139 mfu:25.1299834831351 +Step 16713 | loss:0.7467666268348694 lr:5.6397851497667707e-05 tokens_per_second_per_gpu:4873.113603484154 mfu:22.987308306550773 +Step 16714 | loss:0.7781420946121216 lr:5.639355089685809e-05 tokens_per_second_per_gpu:5345.458903921372 mfu:25.215441679131608 +Step 16715 | loss:0.8462104797363281 lr:5.638925024796177e-05 tokens_per_second_per_gpu:5093.425475173893 mfu:24.026556994392642 +Step 16716 | loss:0.8677243590354919 lr:5.638494955101109e-05 tokens_per_second_per_gpu:4111.23328925105 mfu:19.393388874127453 +Step 16717 | loss:0.818634569644928 lr:5.638064880603838e-05 tokens_per_second_per_gpu:4240.570359841568 mfu:20.00349390328266 +Step 16718 | loss:0.7884952425956726 lr:5.637634801307601e-05 tokens_per_second_per_gpu:5357.452528228779 mfu:25.272017651312957 +Step 16719 | loss:0.7769890427589417 lr:5.637204717215629e-05 tokens_per_second_per_gpu:4939.817936307208 mfu:23.30196402540986 +Step 16720 | loss:0.8838642239570618 lr:5.636774628331161e-05 tokens_per_second_per_gpu:4949.857508335718 mfu:23.349322399595962 +Step 16721 | loss:0.8653172254562378 lr:5.6363445346574284e-05 tokens_per_second_per_gpu:5343.005930249573 mfu:25.203870583801265 +Step 16722 | loss:0.7147414684295654 lr:5.635914436197668e-05 tokens_per_second_per_gpu:4103.361794553571 mfu:19.35625769062396 +Step 16723 | loss:0.9205825328826904 lr:5.635484332955115e-05 tokens_per_second_per_gpu:4036.906772032754 mfu:19.042778010997246 +Step 16724 | loss:0.9086495041847229 lr:5.635054224933002e-05 tokens_per_second_per_gpu:4547.310467081171 mfu:21.450439324390803 +Step 16725 | loss:0.6802367568016052 lr:5.6346241121345654e-05 tokens_per_second_per_gpu:3917.746257182299 mfu:18.480677531567363 +Step 16726 | loss:0.7165722846984863 lr:5.63419399456304e-05 tokens_per_second_per_gpu:5369.559980920292 mfu:25.329130571403272 +Step 16727 | loss:0.7372212409973145 lr:5.63376387222166e-05 tokens_per_second_per_gpu:4459.953073059108 mfu:21.03835959208039 +Step 16728 | loss:0.8893875479698181 lr:5.633333745113661e-05 tokens_per_second_per_gpu:3242.1017616097142 mfu:15.293547169113562 +Step 16729 | loss:0.7818267941474915 lr:5.632903613242279e-05 tokens_per_second_per_gpu:5342.940795498555 mfu:25.203563332067578 +Step 16730 | loss:0.7142571210861206 lr:5.632473476610748e-05 tokens_per_second_per_gpu:5068.675551543922 mfu:23.90980738185766 +Step 16731 | loss:0.6874421238899231 lr:5.632043335222303e-05 tokens_per_second_per_gpu:3395.065588901692 mfu:16.015103640769933 +Step 16732 | loss:0.8853130340576172 lr:5.631613189080178e-05 tokens_per_second_per_gpu:5351.8164579251 mfu:25.245431346077755 +Step 16733 | loss:0.8445236086845398 lr:5.631183038187612e-05 tokens_per_second_per_gpu:3332.38507292121 mfu:15.719428952491201 +Step 16734 | loss:0.7755633592605591 lr:5.630752882547837e-05 tokens_per_second_per_gpu:4128.134883998222 mfu:19.473116580259283 +Step 16735 | loss:0.8080078959465027 lr:5.6303227221640896e-05 tokens_per_second_per_gpu:5348.014027562333 mfu:25.227494633295066 +Step 16736 | loss:0.7321664690971375 lr:5.6298925570396045e-05 tokens_per_second_per_gpu:4802.029634655452 mfu:22.651993097410088 +Step 16737 | loss:0.7990099191665649 lr:5.6294623871776166e-05 tokens_per_second_per_gpu:3887.1195112848427 mfu:18.336205945707377 +Step 16738 | loss:0.7297602891921997 lr:5.6290322125813624e-05 tokens_per_second_per_gpu:4345.363209203159 mfu:20.497819653225065 +Step 16739 | loss:0.7349193692207336 lr:5.628602033254075e-05 tokens_per_second_per_gpu:5348.226740154716 mfu:25.22849803488585 +Step 16740 | loss:0.7771041989326477 lr:5.628171849198992e-05 tokens_per_second_per_gpu:4065.559861001184 mfu:19.177939520382413 +Step 16741 | loss:0.806797981262207 lr:5.627741660419349e-05 tokens_per_second_per_gpu:4773.893804779618 mfu:22.519271587418192 +Step 16742 | loss:0.7765566110610962 lr:5.627311466918381e-05 tokens_per_second_per_gpu:4084.192705792437 mfu:19.265833828353898 +Step 16743 | loss:0.9523545503616333 lr:5.626881268699323e-05 tokens_per_second_per_gpu:4717.628820131971 mfu:22.25386005503896 +Step 16744 | loss:0.8061243295669556 lr:5.6264510657654104e-05 tokens_per_second_per_gpu:5337.060669463447 mfu:25.17582577430702 +Step 16745 | loss:0.8234601020812988 lr:5.6260208581198815e-05 tokens_per_second_per_gpu:3348.2055253010903 mfu:15.794056725614611 +Step 16746 | loss:0.7193737626075745 lr:5.625590645765968e-05 tokens_per_second_per_gpu:4285.206144633472 mfu:20.214048515796257 +Step 16747 | loss:0.6967407464981079 lr:5.625160428706908e-05 tokens_per_second_per_gpu:4403.825972659398 mfu:20.773598483224152 +Step 16748 | loss:0.7326196432113647 lr:5.624730206945936e-05 tokens_per_second_per_gpu:4659.0310474642265 mfu:21.977444363554252 +Step 16749 | loss:0.7058821320533752 lr:5.62429998048629e-05 tokens_per_second_per_gpu:5188.323796197605 mfu:24.474208565984654 +Step 16750 | loss:0.5884637832641602 lr:5.623869749331202e-05 tokens_per_second_per_gpu:4413.812689525194 mfu:20.820707530589715 +Step 16751 | loss:0.810089111328125 lr:5.623439513483911e-05 tokens_per_second_per_gpu:4242.83450428179 mfu:20.014174258910067 +Step 16752 | loss:0.8376688957214355 lr:5.623009272947654e-05 tokens_per_second_per_gpu:5309.070297406322 mfu:25.04379041366364 +Step 16753 | loss:0.8233393430709839 lr:5.622579027725662e-05 tokens_per_second_per_gpu:5325.650027087479 mfu:25.12199982736324 +Step 16754 | loss:0.6922473311424255 lr:5.6221487778211745e-05 tokens_per_second_per_gpu:5045.68811681626 mfu:23.80137173018649 +Step 16755 | loss:0.8553867340087891 lr:5.621718523237427e-05 tokens_per_second_per_gpu:5289.731438533109 mfu:24.952565716809133 +Step 16756 | loss:0.8766562938690186 lr:5.6212882639776544e-05 tokens_per_second_per_gpu:3704.9573340017473 mfu:17.476915875390695 +Step 16757 | loss:0.7014011144638062 lr:5.620858000045093e-05 tokens_per_second_per_gpu:5360.8750652322415 mfu:25.28816234230332 +Step 16758 | loss:0.7328406572341919 lr:5.620427731442981e-05 tokens_per_second_per_gpu:4071.1330834090804 mfu:19.20422936137026 +Step 16759 | loss:0.7913069128990173 lr:5.619997458174552e-05 tokens_per_second_per_gpu:5231.500619104101 mfu:24.677881006360224 +Step 16760 | loss:0.7191596627235413 lr:5.619567180243044e-05 tokens_per_second_per_gpu:3633.531243511917 mfu:17.139986820002495 +Step 16761 | loss:0.7125997543334961 lr:5.6191368976516913e-05 tokens_per_second_per_gpu:3878.321881445022 mfu:18.29470602472251 +Step 16762 | loss:0.7348426580429077 lr:5.61870661040373e-05 tokens_per_second_per_gpu:3540.499027787805 mfu:16.701137985499088 +Step 16763 | loss:0.8041080236434937 lr:5.618276318502399e-05 tokens_per_second_per_gpu:4322.830262119338 mfu:20.391527897313473 +Step 16764 | loss:0.8797522187232971 lr:5.617846021950931e-05 tokens_per_second_per_gpu:4335.699669885471 mfu:20.45223508949371 +Step 16765 | loss:0.8201874494552612 lr:5.617415720752566e-05 tokens_per_second_per_gpu:4022.8811087335166 mfu:18.976616564189825 +Step 16766 | loss:0.7366070747375488 lr:5.6169854149105385e-05 tokens_per_second_per_gpu:3994.5945338169836 mfu:18.84318396412119 +Step 16767 | loss:0.7994532585144043 lr:5.616555104428085e-05 tokens_per_second_per_gpu:5306.929771966607 mfu:25.033693189953205 +Step 16768 | loss:0.6132528185844421 lr:5.61612478930844e-05 tokens_per_second_per_gpu:5286.148850894199 mfu:24.93566603966391 +Step 16769 | loss:0.7925131320953369 lr:5.615694469554843e-05 tokens_per_second_per_gpu:4099.7806116474385 mfu:19.339364639842923 +Step 16770 | loss:0.5992323160171509 lr:5.615264145170531e-05 tokens_per_second_per_gpu:3951.4748450091784 mfu:18.639781035037238 +Step 16771 | loss:0.7576011419296265 lr:5.614833816158737e-05 tokens_per_second_per_gpu:4220.325532292815 mfu:19.90799559761144 +Step 16772 | loss:0.8825989365577698 lr:5.6144034825227e-05 tokens_per_second_per_gpu:4829.544590351487 mfu:22.78178583796381 +Step 16773 | loss:0.8355451822280884 lr:5.6139731442656564e-05 tokens_per_second_per_gpu:3390.628792765772 mfu:15.994174516401316 +Step 16774 | loss:0.7984253168106079 lr:5.613542801390842e-05 tokens_per_second_per_gpu:5252.394846012918 mfu:24.776442639611822 +Step 16775 | loss:0.7033124566078186 lr:5.6131124539014926e-05 tokens_per_second_per_gpu:5255.966030812181 mfu:24.793288527616905 +Step 16776 | loss:0.7963162660598755 lr:5.612682101800848e-05 tokens_per_second_per_gpu:4479.653962327458 mfu:21.131292048077515 +Step 16777 | loss:0.6375565528869629 lr:5.612251745092142e-05 tokens_per_second_per_gpu:4074.659030164988 mfu:19.220861853806426 +Step 16778 | loss:0.9819129109382629 lr:5.6118213837786136e-05 tokens_per_second_per_gpu:3997.0868696781317 mfu:18.854940737614577 +Step 16779 | loss:0.7574900984764099 lr:5.611391017863498e-05 tokens_per_second_per_gpu:4331.862690798021 mfu:20.434135404482063 +Step 16780 | loss:0.8065612316131592 lr:5.610960647350032e-05 tokens_per_second_per_gpu:4620.584142359215 mfu:21.796083752455484 +Step 16781 | loss:0.9387483596801758 lr:5.6105302722414545e-05 tokens_per_second_per_gpu:5247.177569967988 mfu:24.751831858348808 +Step 16782 | loss:0.7671656608581543 lr:5.610099892540999e-05 tokens_per_second_per_gpu:5066.398085666709 mfu:23.899064186739103 +Step 16783 | loss:0.7176111936569214 lr:5.6096695082519055e-05 tokens_per_second_per_gpu:5329.264410408118 mfu:25.139049490164076 +Step 16784 | loss:0.8261878490447998 lr:5.6092391193774095e-05 tokens_per_second_per_gpu:4414.740016793146 mfu:20.825081891530992 +Step 16785 | loss:0.8905733227729797 lr:5.608808725920749e-05 tokens_per_second_per_gpu:3584.734152146746 mfu:16.90980261439068 +Step 16786 | loss:0.8196595311164856 lr:5.60837832788516e-05 tokens_per_second_per_gpu:3386.043673125587 mfu:15.972545724756486 +Step 16787 | loss:0.6628409028053284 lr:5.607947925273879e-05 tokens_per_second_per_gpu:5368.024218560366 mfu:25.321886118323423 +Step 16788 | loss:0.761605441570282 lr:5.607517518090146e-05 tokens_per_second_per_gpu:5350.785202837525 mfu:25.2405667398794 +Step 16789 | loss:0.7728077173233032 lr:5.607087106337194e-05 tokens_per_second_per_gpu:5318.457723810032 mfu:25.08807250567004 +Step 16790 | loss:0.7870683073997498 lr:5.6066566900182635e-05 tokens_per_second_per_gpu:4675.9807766534 mfu:22.05739912806159 +Step 16791 | loss:0.9854090213775635 lr:5.60622626913659e-05 tokens_per_second_per_gpu:3876.4606468043476 mfu:18.285926263363073 +Step 16792 | loss:0.885732889175415 lr:5.6057958436954126e-05 tokens_per_second_per_gpu:3549.5379908638492 mfu:16.74377631653492 +Step 16793 | loss:0.9517477750778198 lr:5.6053654136979664e-05 tokens_per_second_per_gpu:5339.849092312688 mfu:25.18897924064089 +Step 16794 | loss:0.8077927827835083 lr:5.6049349791474904e-05 tokens_per_second_per_gpu:4489.530120455417 mfu:21.177879571013428 +Step 16795 | loss:0.7625569701194763 lr:5.604504540047221e-05 tokens_per_second_per_gpu:5357.3814561688505 mfu:25.271682392280134 +Step 16796 | loss:0.7178289294242859 lr:5.604074096400396e-05 tokens_per_second_per_gpu:5333.100629789082 mfu:25.157145591510602 +Step 16797 | loss:0.6912334561347961 lr:5.603643648210253e-05 tokens_per_second_per_gpu:5351.384082623245 mfu:25.243391758008073 +Step 16798 | loss:0.700059711933136 lr:5.603213195480027e-05 tokens_per_second_per_gpu:4734.845657183104 mfu:22.335074812904463 +Step 16799 | loss:0.7745475172996521 lr:5.60278273821296e-05 tokens_per_second_per_gpu:4893.24996569962 mfu:23.082294962738786 +Step 16800 | loss:0.6777575612068176 lr:5.602352276412286e-05 tokens_per_second_per_gpu:4574.866185123785 mfu:21.580424347888112 +Step 16801 | loss:0.7653950452804565 lr:5.6019218100812435e-05 tokens_per_second_per_gpu:5331.141332664429 mfu:25.147903252683054 +Step 16802 | loss:0.6555169820785522 lr:5.601491339223071e-05 tokens_per_second_per_gpu:4535.974494840416 mfu:21.39696560921486 +Step 16803 | loss:0.7782996892929077 lr:5.6010608638410044e-05 tokens_per_second_per_gpu:4600.73673654339 mfu:21.702460152905623 +Step 16804 | loss:0.6881730556488037 lr:5.600630383938283e-05 tokens_per_second_per_gpu:4585.9647698397985 mfu:21.63277826560737 +Step 16805 | loss:0.7692052125930786 lr:5.600199899518144e-05 tokens_per_second_per_gpu:4599.047183211166 mfu:21.694490241569994 +Step 16806 | loss:0.749033510684967 lr:5.599769410583825e-05 tokens_per_second_per_gpu:4347.435202361873 mfu:20.507593598473207 +Step 16807 | loss:0.7437052130699158 lr:5.599338917138563e-05 tokens_per_second_per_gpu:4324.694756312305 mfu:20.40032303453721 +Step 16808 | loss:0.8585877418518066 lr:5.598908419185597e-05 tokens_per_second_per_gpu:4208.42439904792 mfu:19.85185592155255 +Step 16809 | loss:0.6400856375694275 lr:5.598477916728163e-05 tokens_per_second_per_gpu:4535.088815017635 mfu:21.392787706378243 +Step 16810 | loss:0.696963906288147 lr:5.598047409769504e-05 tokens_per_second_per_gpu:3584.9470166288897 mfu:16.91080673247138 +Step 16811 | loss:0.6775137782096863 lr:5.5976168983128496e-05 tokens_per_second_per_gpu:5352.242262621714 mfu:25.247439939481456 +Step 16812 | loss:0.8714526295661926 lr:5.5971863823614455e-05 tokens_per_second_per_gpu:3596.5245541371764 mfu:16.965419952229556 +Step 16813 | loss:0.6764460206031799 lr:5.5967558619185246e-05 tokens_per_second_per_gpu:4246.508014481223 mfu:20.03150283328626 +Step 16814 | loss:0.919029951095581 lr:5.596325336987328e-05 tokens_per_second_per_gpu:4746.991900570715 mfu:22.392370715326642 +Step 16815 | loss:0.7192148566246033 lr:5.595894807571092e-05 tokens_per_second_per_gpu:5142.819994397421 mfu:24.259559369144032 +Step 16816 | loss:0.737747073173523 lr:5.5954642736730544e-05 tokens_per_second_per_gpu:4199.864646833398 mfu:19.81147810991233 +Step 16817 | loss:0.8896171450614929 lr:5.595033735296455e-05 tokens_per_second_per_gpu:5190.947167713599 mfu:24.486583457018938 +Step 16818 | loss:0.8337765336036682 lr:5.5946031924445295e-05 tokens_per_second_per_gpu:5342.300998620068 mfu:25.200545301031152 +Step 16819 | loss:0.9379003047943115 lr:5.594172645120519e-05 tokens_per_second_per_gpu:5332.2869622193175 mfu:25.15330738275802 +Step 16820 | loss:0.7348665595054626 lr:5.5937420933276594e-05 tokens_per_second_per_gpu:3929.2739478217513 mfu:18.535055615140493 +Step 16821 | loss:0.79386305809021 lr:5.593311537069191e-05 tokens_per_second_per_gpu:5309.553826580991 mfu:25.046071303279138 +Step 16822 | loss:0.6965035796165466 lr:5.5928809763483505e-05 tokens_per_second_per_gpu:5358.242029338267 mfu:25.275741862749193 +Step 16823 | loss:0.8005748391151428 lr:5.592450411168375e-05 tokens_per_second_per_gpu:4304.760404119216 mfu:20.30628929409121 +Step 16824 | loss:0.7547556757926941 lr:5.5920198415325064e-05 tokens_per_second_per_gpu:4379.5586233447075 mfu:20.65912525607002 +Step 16825 | loss:0.7894262671470642 lr:5.591589267443978e-05 tokens_per_second_per_gpu:5360.601094693557 mfu:25.28686997652815 +Step 16826 | loss:0.8215292692184448 lr:5.591158688906034e-05 tokens_per_second_per_gpu:3920.4121253004273 mfu:18.493252886324573 +Step 16827 | loss:0.8100162744522095 lr:5.590728105921908e-05 tokens_per_second_per_gpu:5360.004417537787 mfu:25.284055348580765 +Step 16828 | loss:0.7930372357368469 lr:5.590297518494844e-05 tokens_per_second_per_gpu:4333.935621352394 mfu:20.443913771585468 +Step 16829 | loss:0.8012688755989075 lr:5.5898669266280734e-05 tokens_per_second_per_gpu:4038.735784626935 mfu:19.051405775465724 +Step 16830 | loss:0.6905407905578613 lr:5.58943633032484e-05 tokens_per_second_per_gpu:2914.4619681080458 mfu:13.748014362053082 +Step 16831 | loss:0.9085922837257385 lr:5.5890057295883805e-05 tokens_per_second_per_gpu:3406.7442021558795 mfu:16.070193652066823 +Step 16832 | loss:0.6970144510269165 lr:5.5885751244219343e-05 tokens_per_second_per_gpu:4900.118074938414 mfu:23.11469300583845 +Step 16833 | loss:0.8131284713745117 lr:5.5881445148287395e-05 tokens_per_second_per_gpu:4095.1609367857795 mfu:19.31757284532267 +Step 16834 | loss:0.8066554665565491 lr:5.587713900812033e-05 tokens_per_second_per_gpu:4382.993049084136 mfu:20.67532602825636 +Step 16835 | loss:0.8954216837882996 lr:5.587283282375058e-05 tokens_per_second_per_gpu:4539.64381660246 mfu:21.41427442601727 +Step 16836 | loss:0.8774611949920654 lr:5.586852659521048e-05 tokens_per_second_per_gpu:5322.006114233634 mfu:25.104810868716008 +Step 16837 | loss:0.7611462473869324 lr:5.586422032253245e-05 tokens_per_second_per_gpu:5309.893345026097 mfu:25.047672869712525 +Step 16838 | loss:0.7365992069244385 lr:5.585991400574887e-05 tokens_per_second_per_gpu:3920.6471001941754 mfu:18.494361302989237 +Step 16839 | loss:0.5890095829963684 lr:5.585560764489214e-05 tokens_per_second_per_gpu:5283.144776825799 mfu:24.92149530973538 +Step 16840 | loss:0.8074455857276917 lr:5.585130123999462e-05 tokens_per_second_per_gpu:4721.884447445774 mfu:22.273934575163057 +Step 16841 | loss:0.7237576842308044 lr:5.5846994791088724e-05 tokens_per_second_per_gpu:3743.6257701664586 mfu:17.659321486295163 +Step 16842 | loss:0.8123286962509155 lr:5.584268829820685e-05 tokens_per_second_per_gpu:3628.020760154273 mfu:17.113992929819076 +Step 16843 | loss:0.7975878119468689 lr:5.5838381761381354e-05 tokens_per_second_per_gpu:4579.789980349336 mfu:21.603650730053133 +Step 16844 | loss:0.8303406238555908 lr:5.583407518064465e-05 tokens_per_second_per_gpu:4287.1877928087915 mfu:20.223396288343285 +Step 16845 | loss:0.6360462307929993 lr:5.5829768556029104e-05 tokens_per_second_per_gpu:5282.385873631635 mfu:24.91791543388627 +Step 16846 | loss:0.7914257645606995 lr:5.5825461887567156e-05 tokens_per_second_per_gpu:5305.58050808723 mfu:25.027328482026025 +Step 16847 | loss:0.6669318079948425 lr:5.582115517529114e-05 tokens_per_second_per_gpu:5320.695934737257 mfu:25.098630528491924 +Step 16848 | loss:0.8892291188240051 lr:5.581684841923349e-05 tokens_per_second_per_gpu:5309.132911688613 mfu:25.044085775916304 +Step 16849 | loss:0.8831495642662048 lr:5.5812541619426575e-05 tokens_per_second_per_gpu:4559.497492109433 mfu:21.50792760077 +Step 16850 | loss:0.7952781319618225 lr:5.58082347759028e-05 tokens_per_second_per_gpu:5356.180662323767 mfu:25.26601804283632 +Step 16851 | loss:0.6985911130905151 lr:5.580392788869454e-05 tokens_per_second_per_gpu:4157.693140549046 mfu:19.612547919568122 +Step 16852 | loss:0.6142949461936951 lr:5.5799620957834196e-05 tokens_per_second_per_gpu:4376.855195801219 mfu:20.64637272709508 +Step 16853 | loss:0.725411593914032 lr:5.579531398335418e-05 tokens_per_second_per_gpu:3822.66964897575 mfu:18.032184948914978 +Step 16854 | loss:0.7008910775184631 lr:5.579100696528684e-05 tokens_per_second_per_gpu:4590.383632709997 mfu:21.653622795701843 +Step 16855 | loss:0.898756206035614 lr:5.578669990366462e-05 tokens_per_second_per_gpu:4070.6138469025705 mfu:19.201780034178167 +Step 16856 | loss:0.7648696303367615 lr:5.578239279851989e-05 tokens_per_second_per_gpu:4841.589983503619 mfu:22.83860601261841 +Step 16857 | loss:0.7773571610450745 lr:5.577808564988504e-05 tokens_per_second_per_gpu:5247.783534494189 mfu:24.7546902964071 +Step 16858 | loss:0.796829342842102 lr:5.577377845779247e-05 tokens_per_second_per_gpu:5326.316763645838 mfu:25.125144937466008 +Step 16859 | loss:0.95527184009552 lr:5.5769471222274574e-05 tokens_per_second_per_gpu:5331.174054228489 mfu:25.148057605883597 +Step 16860 | loss:0.9187210202217102 lr:5.576516394336375e-05 tokens_per_second_per_gpu:4860.720852768445 mfu:22.92884958700347 +Step 16861 | loss:0.7399484515190125 lr:5.5760856621092394e-05 tokens_per_second_per_gpu:5091.271624745911 mfu:24.016396914439078 +Step 16862 | loss:0.8015406131744385 lr:5.5756549255492894e-05 tokens_per_second_per_gpu:5309.756809009184 mfu:25.047028806025295 +Step 16863 | loss:0.7787317037582397 lr:5.575224184659765e-05 tokens_per_second_per_gpu:5173.802926799395 mfu:24.405711147517508 +Step 16864 | loss:0.6345914602279663 lr:5.574793439443908e-05 tokens_per_second_per_gpu:5327.930497966475 mfu:25.132757197587914 +Step 16865 | loss:0.8112780451774597 lr:5.574362689904955e-05 tokens_per_second_per_gpu:4557.77983838395 mfu:21.49982513508495 +Step 16866 | loss:0.7967557311058044 lr:5.573931936046147e-05 tokens_per_second_per_gpu:4685.387911059108 mfu:22.101774186075946 +Step 16867 | loss:0.7764914631843567 lr:5.5735011778707235e-05 tokens_per_second_per_gpu:4819.9228771758435 mfu:22.736398575280752 +Step 16868 | loss:0.7699732780456543 lr:5.5730704153819255e-05 tokens_per_second_per_gpu:4507.523811765752 mfu:21.26275888296459 +Step 16869 | loss:0.7287454605102539 lr:5.5726396485829914e-05 tokens_per_second_per_gpu:4493.459614612444 mfu:21.196415665392962 +Step 16870 | loss:0.7581913471221924 lr:5.57220887747716e-05 tokens_per_second_per_gpu:5373.897512433326 mfu:25.34959144760944 +Step 16871 | loss:0.8020206093788147 lr:5.571778102067675e-05 tokens_per_second_per_gpu:5312.78297379638 mfu:25.061303741643453 +Step 16872 | loss:0.8044882416725159 lr:5.5713473223577715e-05 tokens_per_second_per_gpu:3555.4282878645377 mfu:16.771561852475465 +Step 16873 | loss:0.811485767364502 lr:5.5709165383506936e-05 tokens_per_second_per_gpu:4721.368388991275 mfu:22.271500239384537 +Step 16874 | loss:0.7709033489227295 lr:5.5704857500496784e-05 tokens_per_second_per_gpu:4404.66496122334 mfu:20.77755613542648 +Step 16875 | loss:0.8959411382675171 lr:5.5700549574579684e-05 tokens_per_second_per_gpu:5312.8296362272295 mfu:25.061523856291203 +Step 16876 | loss:0.6932274699211121 lr:5.569624160578801e-05 tokens_per_second_per_gpu:3757.6450299271787 mfu:17.725452726518842 +Step 16877 | loss:0.6768662929534912 lr:5.569193359415418e-05 tokens_per_second_per_gpu:4829.039447806677 mfu:22.779402994393738 +Step 16878 | loss:0.6557266712188721 lr:5.56876255397106e-05 tokens_per_second_per_gpu:3246.9858059969747 mfu:15.316586039792172 +Step 16879 | loss:0.8219906687736511 lr:5.568331744248965e-05 tokens_per_second_per_gpu:4954.661682043611 mfu:23.371984506652655 +Step 16880 | loss:0.7470564842224121 lr:5.5679009302523744e-05 tokens_per_second_per_gpu:4008.0812116025368 mfu:18.90680292430016 +Step 16881 | loss:0.7431426644325256 lr:5.567470111984528e-05 tokens_per_second_per_gpu:5363.434828051467 mfu:25.30023718026218 +Step 16882 | loss:0.8008905053138733 lr:5.567039289448669e-05 tokens_per_second_per_gpu:5273.771165640678 mfu:24.87727838647218 +Step 16883 | loss:0.7922467589378357 lr:5.566608462648033e-05 tokens_per_second_per_gpu:4098.493401340006 mfu:19.33329264920204 +Step 16884 | loss:0.7491768598556519 lr:5.5661776315858624e-05 tokens_per_second_per_gpu:5203.625244921542 mfu:24.546388110311828 +Step 16885 | loss:0.8845639824867249 lr:5.5657467962653985e-05 tokens_per_second_per_gpu:4479.726103190981 mfu:21.13163234883925 +Step 16886 | loss:0.7897993922233582 lr:5.5653159566898815e-05 tokens_per_second_per_gpu:4815.97965805068 mfu:22.717797737884798 +Step 16887 | loss:0.7994614839553833 lr:5.56488511286255e-05 tokens_per_second_per_gpu:5336.209118150295 mfu:25.171808861473245 +Step 16888 | loss:0.7441004514694214 lr:5.564454264786645e-05 tokens_per_second_per_gpu:4163.614927085807 mfu:19.640481997022988 +Step 16889 | loss:0.8082926273345947 lr:5.564023412465409e-05 tokens_per_second_per_gpu:5362.433210823697 mfu:25.295512380905265 +Step 16890 | loss:0.8514584898948669 lr:5.5635925559020795e-05 tokens_per_second_per_gpu:4631.037304694861 mfu:21.845393102686128 +Step 16891 | loss:0.8279796838760376 lr:5.5631616950998986e-05 tokens_per_second_per_gpu:4033.523916409289 mfu:19.026820503846672 +Step 16892 | loss:0.6669637560844421 lr:5.562730830062107e-05 tokens_per_second_per_gpu:4235.7726242063845 mfu:19.98086216571339 +Step 16893 | loss:0.8266217112541199 lr:5.562299960791946e-05 tokens_per_second_per_gpu:3786.201968165585 mfu:17.860160676505487 +Step 16894 | loss:0.8484224677085876 lr:5.561869087292654e-05 tokens_per_second_per_gpu:5024.202372238799 mfu:23.70001980716857 +Step 16895 | loss:0.7262383699417114 lr:5.561438209567472e-05 tokens_per_second_per_gpu:5315.081034260989 mfu:25.07214408494456 +Step 16896 | loss:0.8215786814689636 lr:5.561007327619644e-05 tokens_per_second_per_gpu:4929.4982172070395 mfu:23.253284149688728 +Step 16897 | loss:0.7903821468353271 lr:5.5605764414524064e-05 tokens_per_second_per_gpu:5359.443238316374 mfu:25.281408170446312 +Step 16898 | loss:0.74153071641922 lr:5.5601455510690024e-05 tokens_per_second_per_gpu:4484.593023592167 mfu:21.154590442753893 +Step 16899 | loss:0.7957760691642761 lr:5.559714656472672e-05 tokens_per_second_per_gpu:4474.795393162151 mfu:21.108373348367387 +Step 16900 | loss:0.7259173393249512 lr:5.559283757666658e-05 tokens_per_second_per_gpu:5340.455086097628 mfu:25.19183781671765 +Step 16901 | loss:0.8122512698173523 lr:5.558852854654197e-05 tokens_per_second_per_gpu:4416.66646422209 mfu:20.834169272739146 +Step 16902 | loss:0.6851255893707275 lr:5.5584219474385336e-05 tokens_per_second_per_gpu:4182.978481171487 mfu:19.731823185408146 +Step 16903 | loss:0.8331314325332642 lr:5.557991036022908e-05 tokens_per_second_per_gpu:4318.204725578831 mfu:20.369708452253988 +Step 16904 | loss:0.7237657904624939 lr:5.55756012041056e-05 tokens_per_second_per_gpu:5346.777975722471 mfu:25.221663965874246 +Step 16905 | loss:0.7716571092605591 lr:5.557129200604731e-05 tokens_per_second_per_gpu:4229.472872068336 mfu:19.95114516002967 +Step 16906 | loss:0.6707985401153564 lr:5.556698276608662e-05 tokens_per_second_per_gpu:5353.755472257594 mfu:25.25457800751409 +Step 16907 | loss:0.8353656530380249 lr:5.5562673484255955e-05 tokens_per_second_per_gpu:3997.7177020532536 mfu:18.85791648155913 +Step 16908 | loss:0.7076932787895203 lr:5.55583641605877e-05 tokens_per_second_per_gpu:5333.334667789566 mfu:25.158249588689884 +Step 16909 | loss:0.7789329290390015 lr:5.5554054795114284e-05 tokens_per_second_per_gpu:5073.513320067038 mfu:23.932627961389628 +Step 16910 | loss:0.8162584900856018 lr:5.5549745387868114e-05 tokens_per_second_per_gpu:4709.956221807515 mfu:22.21766709966224 +Step 16911 | loss:0.8014567494392395 lr:5.5545435938881596e-05 tokens_per_second_per_gpu:5334.3014771068865 mfu:25.162810193194368 +Step 16912 | loss:0.7492945790290833 lr:5.5541126448187156e-05 tokens_per_second_per_gpu:3707.6019114421415 mfu:17.489390798388474 +Step 16913 | loss:0.7529426217079163 lr:5.553681691581718e-05 tokens_per_second_per_gpu:3414.535884474107 mfu:16.10694834695999 +Step 16914 | loss:0.7190754413604736 lr:5.5532507341804116e-05 tokens_per_second_per_gpu:3635.131925906984 mfu:17.147537511964604 +Step 16915 | loss:0.7801883816719055 lr:5.552819772618034e-05 tokens_per_second_per_gpu:4055.5397322423855 mfu:19.13067286341691 +Step 16916 | loss:0.8799813985824585 lr:5.5523888068978304e-05 tokens_per_second_per_gpu:4611.840442957856 mfu:21.754838230550323 +Step 16917 | loss:0.8526560068130493 lr:5.551957837023037e-05 tokens_per_second_per_gpu:4485.545363539285 mfu:21.1590827927704 +Step 16918 | loss:0.742440402507782 lr:5.551526862996902e-05 tokens_per_second_per_gpu:2518.390559328133 mfu:11.879677950087602 +Step 16919 | loss:0.8216851949691772 lr:5.5510958848226616e-05 tokens_per_second_per_gpu:5340.2467958987845 mfu:25.190855276311733 +Step 16920 | loss:0.8644578456878662 lr:5.5506649025035586e-05 tokens_per_second_per_gpu:5132.5525529840825 mfu:24.211126096191165 +Step 16921 | loss:0.7290192246437073 lr:5.550233916042834e-05 tokens_per_second_per_gpu:5317.916906571623 mfu:25.08552137848353 +Step 16922 | loss:0.6973719596862793 lr:5.549802925443731e-05 tokens_per_second_per_gpu:4540.493065783049 mfu:21.41828047929896 +Step 16923 | loss:0.8220414519309998 lr:5.549371930709489e-05 tokens_per_second_per_gpu:4142.807696440238 mfu:19.542330739988223 +Step 16924 | loss:0.9149911403656006 lr:5.548940931843351e-05 tokens_per_second_per_gpu:3994.728060268128 mfu:18.843813831173204 +Step 16925 | loss:0.6629712581634521 lr:5.54850992884856e-05 tokens_per_second_per_gpu:4003.2009180933737 mfu:18.8837817471535 +Step 16926 | loss:0.7744188904762268 lr:5.548078921728352e-05 tokens_per_second_per_gpu:5159.588158887164 mfu:24.338657661986023 +Step 16927 | loss:0.7279454469680786 lr:5.547647910485976e-05 tokens_per_second_per_gpu:5318.559687392976 mfu:25.088553485287573 +Step 16928 | loss:0.71635502576828 lr:5.5472168951246683e-05 tokens_per_second_per_gpu:3778.924099116454 mfu:17.825829726468232 +Step 16929 | loss:0.9242445826530457 lr:5.546785875647673e-05 tokens_per_second_per_gpu:4395.333946327787 mfu:20.73354014612885 +Step 16930 | loss:0.7577363848686218 lr:5.546354852058232e-05 tokens_per_second_per_gpu:5311.895180000178 mfu:25.057115866833527 +Step 16931 | loss:0.8155274391174316 lr:5.5459238243595844e-05 tokens_per_second_per_gpu:5242.191226148751 mfu:24.72831042379548 +Step 16932 | loss:0.7209861874580383 lr:5.545492792554976e-05 tokens_per_second_per_gpu:5334.283484469813 mfu:25.162725318855426 +Step 16933 | loss:0.8208128213882446 lr:5.545061756647647e-05 tokens_per_second_per_gpu:4866.196418844237 mfu:22.954678766410886 +Step 16934 | loss:0.7378702759742737 lr:5.544630716640838e-05 tokens_per_second_per_gpu:5342.522482253145 mfu:25.201590077117384 +Step 16935 | loss:0.664939820766449 lr:5.544199672537791e-05 tokens_per_second_per_gpu:5253.154161284546 mfu:24.780024459300634 +Step 16936 | loss:0.7756675481796265 lr:5.543768624341752e-05 tokens_per_second_per_gpu:5330.41908712619 mfu:25.144496297251383 +Step 16937 | loss:0.8886020183563232 lr:5.5433375720559566e-05 tokens_per_second_per_gpu:4193.558904454974 mfu:19.781732847242715 +Step 16938 | loss:0.8935550451278687 lr:5.542906515683651e-05 tokens_per_second_per_gpu:4211.018861709902 mfu:19.864094444590098 +Step 16939 | loss:0.8197216391563416 lr:5.542475455228077e-05 tokens_per_second_per_gpu:4128.803448164609 mfu:19.476270311500944 +Step 16940 | loss:0.7287940382957458 lr:5.542044390692475e-05 tokens_per_second_per_gpu:5354.246029905154 mfu:25.256892051635393 +Step 16941 | loss:0.8028379082679749 lr:5.541613322080088e-05 tokens_per_second_per_gpu:4167.147500999743 mfu:19.65714575089429 +Step 16942 | loss:0.7668418288230896 lr:5.541182249394158e-05 tokens_per_second_per_gpu:4712.545665987134 mfu:22.229881949662236 +Step 16943 | loss:0.8064818978309631 lr:5.540751172637928e-05 tokens_per_second_per_gpu:5356.035412646551 mfu:25.26533287532674 +Step 16944 | loss:0.7803653478622437 lr:5.540320091814638e-05 tokens_per_second_per_gpu:4873.201302318186 mfu:22.98772199691387 +Step 16945 | loss:0.7420931458473206 lr:5.539889006927533e-05 tokens_per_second_per_gpu:4076.295925763878 mfu:19.22858336962958 +Step 16946 | loss:0.7633845806121826 lr:5.5394579179798534e-05 tokens_per_second_per_gpu:4008.490658437123 mfu:18.908734354877733 +Step 16947 | loss:0.7130924463272095 lr:5.539026824974842e-05 tokens_per_second_per_gpu:5354.564435393954 mfu:25.258394024651643 +Step 16948 | loss:0.7439846992492676 lr:5.538595727915742e-05 tokens_per_second_per_gpu:4533.279217156292 mfu:21.384251524517055 +Step 16949 | loss:0.8005800247192383 lr:5.538164626805793e-05 tokens_per_second_per_gpu:4404.01890047062 mfu:20.774508556626568 +Step 16950 | loss:0.8507253527641296 lr:5.5377335216482395e-05 tokens_per_second_per_gpu:4997.982255409396 mfu:23.57633504247104 +Step 16951 | loss:0.7946012616157532 lr:5.5373024124463244e-05 tokens_per_second_per_gpu:4405.504567314275 mfu:20.78151670060899 +Step 16952 | loss:0.6466830372810364 lr:5.53687129920329e-05 tokens_per_second_per_gpu:2852.106124664068 mfu:13.453871209523887 +Step 16953 | loss:0.8301050662994385 lr:5.5364401819223755e-05 tokens_per_second_per_gpu:4419.944169316565 mfu:20.84963076690394 +Step 16954 | loss:0.7064841985702515 lr:5.5360090606068294e-05 tokens_per_second_per_gpu:4457.777621207814 mfu:21.02809761452748 +Step 16955 | loss:0.8054621815681458 lr:5.535577935259888e-05 tokens_per_second_per_gpu:4466.783662422878 mfu:21.070580647528008 +Step 16956 | loss:0.6990789771080017 lr:5.535146805884798e-05 tokens_per_second_per_gpu:5141.660515731059 mfu:24.25408991044734 +Step 16957 | loss:0.8505125045776367 lr:5.5347156724847995e-05 tokens_per_second_per_gpu:5240.909230727957 mfu:24.72226303266462 +Step 16958 | loss:0.8189668655395508 lr:5.534284535063137e-05 tokens_per_second_per_gpu:5339.23297418063 mfu:25.186072906291372 +Step 16959 | loss:0.8930451273918152 lr:5.533853393623052e-05 tokens_per_second_per_gpu:3900.180891918584 mfu:18.397818706658068 +Step 16960 | loss:0.720209002494812 lr:5.5334222481677865e-05 tokens_per_second_per_gpu:5333.7874262325395 mfu:25.160385327513946 +Step 16961 | loss:0.8798696994781494 lr:5.532991098700586e-05 tokens_per_second_per_gpu:2775.7546551794408 mfu:13.093708301060916 +Step 16962 | loss:0.6875067353248596 lr:5.532559945224691e-05 tokens_per_second_per_gpu:5267.718383825396 mfu:24.848726381938427 +Step 16963 | loss:0.7238868474960327 lr:5.532128787743345e-05 tokens_per_second_per_gpu:4275.5094133628245 mfu:20.16830737995966 +Step 16964 | loss:0.793150782585144 lr:5.53169762625979e-05 tokens_per_second_per_gpu:5328.792555953021 mfu:25.136823672192712 +Step 16965 | loss:0.8014809489250183 lr:5.5312664607772694e-05 tokens_per_second_per_gpu:5322.724015372074 mfu:25.108197330872596 +Step 16966 | loss:0.7194241881370544 lr:5.530835291299026e-05 tokens_per_second_per_gpu:3679.5902971880696 mfu:17.3572552346778 +Step 16967 | loss:0.8696228861808777 lr:5.530404117828302e-05 tokens_per_second_per_gpu:5365.303995631079 mfu:25.30905436264808 +Step 16968 | loss:0.8613389730453491 lr:5.529972940368341e-05 tokens_per_second_per_gpu:5325.733182073817 mfu:25.122392083621495 +Step 16969 | loss:0.7103870511054993 lr:5.5295417589223866e-05 tokens_per_second_per_gpu:4523.029715827094 mfu:21.335902878001978 +Step 16970 | loss:0.7408199310302734 lr:5.529110573493681e-05 tokens_per_second_per_gpu:5170.349927994748 mfu:24.38942276301364 +Step 16971 | loss:0.7286293506622314 lr:5.5286793840854657e-05 tokens_per_second_per_gpu:4384.690208323484 mfu:20.683331817953587 +Step 16972 | loss:0.7868744730949402 lr:5.5282481907009886e-05 tokens_per_second_per_gpu:5338.490108638357 mfu:25.182568682782545 +Step 16973 | loss:0.7835087776184082 lr:5.5278169933434866e-05 tokens_per_second_per_gpu:3692.207486408167 mfu:17.416772668943466 +Step 16974 | loss:0.8006064295768738 lr:5.527385792016207e-05 tokens_per_second_per_gpu:3698.745789822121 mfu:17.44761496169668 +Step 16975 | loss:0.8853806257247925 lr:5.52695458672239e-05 tokens_per_second_per_gpu:2779.7547076658593 mfu:13.112577231118681 +Step 16976 | loss:0.9250098466873169 lr:5.526523377465283e-05 tokens_per_second_per_gpu:4224.455137962209 mfu:19.92747565213735 +Step 16977 | loss:0.7258878350257874 lr:5.526092164248124e-05 tokens_per_second_per_gpu:4941.494342356237 mfu:23.3098719187675 +Step 16978 | loss:0.6356348991394043 lr:5.525660947074158e-05 tokens_per_second_per_gpu:5335.327173957552 mfu:25.167648580240883 +Step 16979 | loss:0.6575688123703003 lr:5.5252297259466315e-05 tokens_per_second_per_gpu:5247.530764579147 mfu:24.753497937590858 +Step 16980 | loss:0.8311870098114014 lr:5.524798500868784e-05 tokens_per_second_per_gpu:4295.063029254079 mfu:20.260545122309598 +Step 16981 | loss:0.8216813206672668 lr:5.524367271843861e-05 tokens_per_second_per_gpu:3330.3745192835445 mfu:15.70994482794651 +Step 16982 | loss:0.7389737963676453 lr:5.523936038875104e-05 tokens_per_second_per_gpu:5217.559931762806 mfu:24.612120405644806 +Step 16983 | loss:0.6159935593605042 lr:5.5235048019657575e-05 tokens_per_second_per_gpu:5346.095373138953 mfu:25.21844401302378 +Step 16984 | loss:0.9360143542289734 lr:5.5230735611190645e-05 tokens_per_second_per_gpu:4288.934148338871 mfu:20.231634145337647 +Step 16985 | loss:0.7567044496536255 lr:5.522642316338268e-05 tokens_per_second_per_gpu:4961.554359047545 mfu:23.404498440092173 +Step 16986 | loss:0.821153461933136 lr:5.5222110676266126e-05 tokens_per_second_per_gpu:4029.3371275961026 mfu:19.00707070667504 +Step 16987 | loss:0.7471649646759033 lr:5.52177981498734e-05 tokens_per_second_per_gpu:4859.112117437629 mfu:22.92126090796946 +Step 16988 | loss:0.7446146011352539 lr:5.5213485584236965e-05 tokens_per_second_per_gpu:3510.764257852258 mfu:16.560874002438226 +Step 16989 | loss:0.8529260754585266 lr:5.520917297938922e-05 tokens_per_second_per_gpu:5351.759274577554 mfu:25.245161602470898 +Step 16990 | loss:0.7866566777229309 lr:5.520486033536264e-05 tokens_per_second_per_gpu:3446.5049022099015 mfu:16.257751658096605 +Step 16991 | loss:0.864341139793396 lr:5.520054765218963e-05 tokens_per_second_per_gpu:4889.357507372207 mfu:23.06393357268651 +Step 16992 | loss:0.6741865277290344 lr:5.5196234929902644e-05 tokens_per_second_per_gpu:5306.316006102211 mfu:25.030797951651593 +Step 16993 | loss:0.891076922416687 lr:5.519192216853411e-05 tokens_per_second_per_gpu:3457.3799380935543 mfu:16.30905105783225 +Step 16994 | loss:0.9605509638786316 lr:5.5187609368116465e-05 tokens_per_second_per_gpu:4392.873750158624 mfu:20.721934981046157 +Step 16995 | loss:0.7343098521232605 lr:5.5183296528682146e-05 tokens_per_second_per_gpu:5047.408200459011 mfu:23.809485658205872 +Step 16996 | loss:0.8186593055725098 lr:5.51789836502636e-05 tokens_per_second_per_gpu:5320.996590566961 mfu:25.10004877333781 +Step 16997 | loss:0.8957280516624451 lr:5.517467073289324e-05 tokens_per_second_per_gpu:5319.645887018668 mfu:25.093677274246726 +Step 16998 | loss:0.695995032787323 lr:5.5170357776603545e-05 tokens_per_second_per_gpu:3593.3679638995422 mfu:16.950529777508805 +Step 16999 | loss:0.8775700926780701 lr:5.516604478142692e-05 tokens_per_second_per_gpu:4386.867350635767 mfu:20.69360176969941 +Step 17000 | loss:0.7255353927612305 lr:5.516173174739581e-05 tokens_per_second_per_gpu:4121.094331712976 mfu:19.43990509388769 +Step 17001 | loss:0.7985616326332092 lr:5.515741867454266e-05 tokens_per_second_per_gpu:4460.613801405815 mfu:21.04147636042369 +Step 17002 | loss:0.6427743434906006 lr:5.51531055628999e-05 tokens_per_second_per_gpu:4876.49012697841 mfu:23.003235943967493 +Step 17003 | loss:0.7628577351570129 lr:5.514879241249997e-05 tokens_per_second_per_gpu:4818.482312309039 mfu:22.729603185018327 +Step 17004 | loss:0.6768139600753784 lr:5.5144479223375324e-05 tokens_per_second_per_gpu:4158.169024957635 mfu:19.614792747517686 +Step 17005 | loss:0.8351098299026489 lr:5.51401659955584e-05 tokens_per_second_per_gpu:5022.919592176741 mfu:23.693968714750998 +Step 17006 | loss:0.8279188871383667 lr:5.513585272908163e-05 tokens_per_second_per_gpu:5343.803504129222 mfu:25.20763287587171 +Step 17007 | loss:0.8545475006103516 lr:5.513153942397744e-05 tokens_per_second_per_gpu:4687.181342224685 mfu:22.110234106874884 +Step 17008 | loss:0.818310558795929 lr:5.5127226080278294e-05 tokens_per_second_per_gpu:4800.681574845528 mfu:22.645634069283954 +Step 17009 | loss:0.6832289695739746 lr:5.512291269801665e-05 tokens_per_second_per_gpu:4343.049850530073 mfu:20.48690714566443 +Step 17010 | loss:0.9190198183059692 lr:5.511859927722489e-05 tokens_per_second_per_gpu:5355.521895875422 mfu:25.262910529102328 +Step 17011 | loss:0.9042765498161316 lr:5.5114285817935516e-05 tokens_per_second_per_gpu:5321.343005999242 mfu:25.101682874036452 +Step 17012 | loss:0.8868233561515808 lr:5.510997232018094e-05 tokens_per_second_per_gpu:4174.334399779515 mfu:19.691047578649844 +Step 17013 | loss:0.6803795695304871 lr:5.5105658783993606e-05 tokens_per_second_per_gpu:5253.964769088918 mfu:24.78384823461781 +Step 17014 | loss:0.8504897952079773 lr:5.510134520940596e-05 tokens_per_second_per_gpu:3612.286567817859 mfu:17.03977206003818 +Step 17015 | loss:0.7870394587516785 lr:5.5097031596450457e-05 tokens_per_second_per_gpu:3890.3073249872928 mfu:18.351243406838936 +Step 17016 | loss:0.7615112066268921 lr:5.509271794515952e-05 tokens_per_second_per_gpu:4068.8977910796216 mfu:19.193685106072003 +Step 17017 | loss:0.8587095737457275 lr:5.5088404255565605e-05 tokens_per_second_per_gpu:4839.022482487093 mfu:22.826494672262612 +Step 17018 | loss:0.8486376404762268 lr:5.508409052770115e-05 tokens_per_second_per_gpu:4917.295103544509 mfu:23.19571998047669 +Step 17019 | loss:0.7451909184455872 lr:5.50797767615986e-05 tokens_per_second_per_gpu:4508.700453373024 mfu:21.26830930218185 +Step 17020 | loss:0.793533980846405 lr:5.50754629572904e-05 tokens_per_second_per_gpu:4283.201781776361 mfu:20.204593594219702 +Step 17021 | loss:0.8516339063644409 lr:5.5071149114808996e-05 tokens_per_second_per_gpu:4211.190795863222 mfu:19.864905487326297 +Step 17022 | loss:0.9223997592926025 lr:5.506683523418683e-05 tokens_per_second_per_gpu:4609.698192017545 mfu:21.744732867359176 +Step 17023 | loss:0.6524031162261963 lr:5.506252131545636e-05 tokens_per_second_per_gpu:4905.428177116886 mfu:23.139741663811158 +Step 17024 | loss:0.7805032730102539 lr:5.505820735865001e-05 tokens_per_second_per_gpu:4017.0113472337284 mfu:18.94892789771074 +Step 17025 | loss:0.7955495119094849 lr:5.5053893363800237e-05 tokens_per_second_per_gpu:3674.7453436768146 mfu:17.334400762330905 +Step 17026 | loss:0.718684732913971 lr:5.504957933093949e-05 tokens_per_second_per_gpu:3717.3901566567897 mfu:17.535563621112814 +Step 17027 | loss:0.9425163269042969 lr:5.5045265260100234e-05 tokens_per_second_per_gpu:4130.506905284667 mfu:19.484305809376004 +Step 17028 | loss:0.8824430108070374 lr:5.504095115131487e-05 tokens_per_second_per_gpu:4444.561451182722 mfu:20.965754685607987 +Step 17029 | loss:0.7092557549476624 lr:5.503663700461588e-05 tokens_per_second_per_gpu:3802.75713753605 mfu:17.938254235029127 +Step 17030 | loss:0.952434778213501 lr:5.503232282003569e-05 tokens_per_second_per_gpu:3600.947376899555 mfu:16.986283161811073 +Step 17031 | loss:0.7094729542732239 lr:5.502800859760676e-05 tokens_per_second_per_gpu:5082.859404009953 mfu:23.976715033954704 +Step 17032 | loss:0.9083465933799744 lr:5.502369433736153e-05 tokens_per_second_per_gpu:3681.022915041368 mfu:17.36401313752162 +Step 17033 | loss:0.8367244601249695 lr:5.501938003933247e-05 tokens_per_second_per_gpu:3790.4059535001093 mfu:17.8799915926023 +Step 17034 | loss:0.9039710164070129 lr:5.5015065703552e-05 tokens_per_second_per_gpu:4512.286768925597 mfu:21.285226564531868 +Step 17035 | loss:0.872990071773529 lr:5.501075133005259e-05 tokens_per_second_per_gpu:5196.322184446133 mfu:24.511938328018722 +Step 17036 | loss:0.7123590707778931 lr:5.5006436918866675e-05 tokens_per_second_per_gpu:5325.478672514437 mfu:25.12119151860609 +Step 17037 | loss:0.756354808807373 lr:5.500212247002671e-05 tokens_per_second_per_gpu:3479.8285285658058 mfu:16.414944889214343 +Step 17038 | loss:0.7084424495697021 lr:5.499780798356514e-05 tokens_per_second_per_gpu:5313.501608032161 mfu:25.064693661945277 +Step 17039 | loss:0.7062012553215027 lr:5.499349345951442e-05 tokens_per_second_per_gpu:4574.434988369598 mfu:21.578390319228863 +Step 17040 | loss:0.8425737619400024 lr:5.4989178897907e-05 tokens_per_second_per_gpu:5315.752329300296 mfu:25.075310698179717 +Step 17041 | loss:0.8035306930541992 lr:5.498486429877533e-05 tokens_per_second_per_gpu:5270.94275457226 mfu:24.863936288885764 +Step 17042 | loss:0.9010653495788574 lr:5.498054966215186e-05 tokens_per_second_per_gpu:4192.730931018852 mfu:19.77782715527769 +Step 17043 | loss:0.8056471347808838 lr:5.497623498806904e-05 tokens_per_second_per_gpu:4535.1649585150035 mfu:21.39314688824675 +Step 17044 | loss:0.6939146518707275 lr:5.497192027655931e-05 tokens_per_second_per_gpu:4891.500983217682 mfu:23.07404471396431 +Step 17045 | loss:0.7420900464057922 lr:5.496760552765516e-05 tokens_per_second_per_gpu:4144.116517050972 mfu:19.548504670116877 +Step 17046 | loss:0.8368856310844421 lr:5.496329074138898e-05 tokens_per_second_per_gpu:3778.5971620930513 mfu:17.824287508747634 +Step 17047 | loss:0.8171894550323486 lr:5.4958975917793285e-05 tokens_per_second_per_gpu:5082.46548977908 mfu:23.974856873318977 +Step 17048 | loss:0.8006803393363953 lr:5.495466105690049e-05 tokens_per_second_per_gpu:5342.1443031918525 mfu:25.19980614196126 +Step 17049 | loss:0.7859635949134827 lr:5.495034615874306e-05 tokens_per_second_per_gpu:3805.293189958507 mfu:17.95021722700102 +Step 17050 | loss:0.7604607343673706 lr:5.494603122335343e-05 tokens_per_second_per_gpu:4648.354636080505 mfu:21.927081909473618 +Step 17051 | loss:0.8154698014259338 lr:5.4941716250764084e-05 tokens_per_second_per_gpu:4594.3805874355885 mfu:21.672477113093255 +Step 17052 | loss:1.1015797853469849 lr:5.493740124100746e-05 tokens_per_second_per_gpu:3536.7214088814085 mfu:16.683318312589368 +Step 17053 | loss:0.8324547410011292 lr:5.4933086194116003e-05 tokens_per_second_per_gpu:3888.468547027825 mfu:18.342569577476258 +Step 17054 | loss:0.8468989133834839 lr:5.492877111012218e-05 tokens_per_second_per_gpu:5349.939760523981 mfu:25.23657864423908 +Step 17055 | loss:0.741212010383606 lr:5.4924455989058434e-05 tokens_per_second_per_gpu:4928.625293126949 mfu:23.24916641786674 +Step 17056 | loss:0.820857048034668 lr:5.4920140830957243e-05 tokens_per_second_per_gpu:5332.320467190476 mfu:25.15346543142351 +Step 17057 | loss:0.7089115381240845 lr:5.491582563585103e-05 tokens_per_second_per_gpu:5353.58203709291 mfu:25.253759884251185 +Step 17058 | loss:0.820171594619751 lr:5.491151040377227e-05 tokens_per_second_per_gpu:4133.641568295158 mfu:19.499092549625157 +Step 17059 | loss:0.7753769755363464 lr:5.490719513475341e-05 tokens_per_second_per_gpu:5322.691750139355 mfu:25.10804513026484 +Step 17060 | loss:0.766261100769043 lr:5.490287982882691e-05 tokens_per_second_per_gpu:5326.717212486613 mfu:25.127033922971556 +Step 17061 | loss:0.8826532363891602 lr:5.4898564486025226e-05 tokens_per_second_per_gpu:4697.596311505524 mfu:22.159363293949593 +Step 17062 | loss:0.754249095916748 lr:5.4894249106380814e-05 tokens_per_second_per_gpu:5337.413608707314 mfu:25.177490648937987 +Step 17063 | loss:0.7096607685089111 lr:5.4889933689926145e-05 tokens_per_second_per_gpu:3887.824094668244 mfu:18.3395295857412 +Step 17064 | loss:0.836429238319397 lr:5.4885618236693624e-05 tokens_per_second_per_gpu:5254.344126053804 mfu:24.785637726146152 +Step 17065 | loss:0.781994640827179 lr:5.488130274671578e-05 tokens_per_second_per_gpu:5282.597488347302 mfu:24.918913656605266 +Step 17066 | loss:0.7636404633522034 lr:5.487698722002502e-05 tokens_per_second_per_gpu:5238.345273195553 mfu:24.71016840752801 +Step 17067 | loss:0.9039118885993958 lr:5.4872671656653817e-05 tokens_per_second_per_gpu:4765.093594023321 mfu:22.47775949180996 +Step 17068 | loss:0.8304145932197571 lr:5.486835605663462e-05 tokens_per_second_per_gpu:4549.267648651292 mfu:21.459671683786674 +Step 17069 | loss:0.8893991112709045 lr:5.486404041999991e-05 tokens_per_second_per_gpu:5313.537846407841 mfu:25.064864604546266 +Step 17070 | loss:0.8375392556190491 lr:5.4859724746782124e-05 tokens_per_second_per_gpu:3631.953219958562 mfu:17.132543013662737 +Step 17071 | loss:0.8009835481643677 lr:5.485540903701374e-05 tokens_per_second_per_gpu:4868.122666867895 mfu:22.96376520699028 +Step 17072 | loss:0.7525063157081604 lr:5.4851093290727195e-05 tokens_per_second_per_gpu:4990.780529819149 mfu:23.542363274120753 +Step 17073 | loss:0.873123824596405 lr:5.484677750795495e-05 tokens_per_second_per_gpu:5007.324623356756 mfu:23.620404586051073 +Step 17074 | loss:0.9071277976036072 lr:5.48424616887295e-05 tokens_per_second_per_gpu:4983.050273619392 mfu:23.50589833670879 +Step 17075 | loss:0.7437533736228943 lr:5.483814583308325e-05 tokens_per_second_per_gpu:5368.257475693913 mfu:25.322986432020134 +Step 17076 | loss:0.7410930395126343 lr:5.4833829941048694e-05 tokens_per_second_per_gpu:3977.357514999793 mfu:18.761874005421838 +Step 17077 | loss:0.6320013403892517 lr:5.48295140126583e-05 tokens_per_second_per_gpu:3515.8058143532285 mfu:16.584655884631538 +Step 17078 | loss:0.7886764407157898 lr:5.4825198047944504e-05 tokens_per_second_per_gpu:5301.602433680767 mfu:25.008563226321257 +Step 17079 | loss:0.7530500888824463 lr:5.482088204693978e-05 tokens_per_second_per_gpu:5323.3076576495305 mfu:25.110950471074997 +Step 17080 | loss:0.7250377535820007 lr:5.4816566009676586e-05 tokens_per_second_per_gpu:4313.764294429205 mfu:20.348762180905524 +Step 17081 | loss:0.85616135597229 lr:5.481224993618739e-05 tokens_per_second_per_gpu:4266.251155526787 mfu:20.124634598125404 +Step 17082 | loss:0.7441214919090271 lr:5.4807933826504634e-05 tokens_per_second_per_gpu:3722.6650845710947 mfu:17.560446356080096 +Step 17083 | loss:0.7392939925193787 lr:5.4803617680660814e-05 tokens_per_second_per_gpu:4494.031813123269 mfu:21.199114823395657 +Step 17084 | loss:0.7818296551704407 lr:5.479930149868836e-05 tokens_per_second_per_gpu:5329.316064062353 mfu:25.13929314926412 +Step 17085 | loss:0.8143754601478577 lr:5.479498528061975e-05 tokens_per_second_per_gpu:4392.760807408968 mfu:20.721402210826007 +Step 17086 | loss:0.7304295897483826 lr:5.479066902648743e-05 tokens_per_second_per_gpu:4800.566456719462 mfu:22.645091037442125 +Step 17087 | loss:0.913185179233551 lr:5.478635273632389e-05 tokens_per_second_per_gpu:3705.2461801654363 mfu:17.478278412027052 +Step 17088 | loss:0.751361608505249 lr:5.4782036410161586e-05 tokens_per_second_per_gpu:5146.432259036354 mfu:24.276599037761777 +Step 17089 | loss:0.8666553497314453 lr:5.4777720048032965e-05 tokens_per_second_per_gpu:4650.5147285596695 mfu:21.93727143424762 +Step 17090 | loss:0.804012656211853 lr:5.477340364997051e-05 tokens_per_second_per_gpu:4905.03194316519 mfu:23.137872560656525 +Step 17091 | loss:0.7682130336761475 lr:5.4769087216006664e-05 tokens_per_second_per_gpu:3347.601111929761 mfu:15.791205604618494 +Step 17092 | loss:0.9073834419250488 lr:5.4764770746173935e-05 tokens_per_second_per_gpu:5364.3246178894115 mfu:25.304434470741565 +Step 17093 | loss:0.6965461373329163 lr:5.476045424050472e-05 tokens_per_second_per_gpu:5350.965484690646 mfu:25.24141716013983 +Step 17094 | loss:0.7897048592567444 lr:5.475613769903154e-05 tokens_per_second_per_gpu:4564.2114680611785 mfu:21.53016422962162 +Step 17095 | loss:0.7429224252700806 lr:5.475182112178684e-05 tokens_per_second_per_gpu:4972.307936131676 mfu:23.455224897947463 +Step 17096 | loss:0.6943610906600952 lr:5.474750450880309e-05 tokens_per_second_per_gpu:5350.22859153266 mfu:25.237941109386675 +Step 17097 | loss:0.9250426292419434 lr:5.4743187860112755e-05 tokens_per_second_per_gpu:4476.370617125506 mfu:21.11580394856345 +Step 17098 | loss:0.7268959283828735 lr:5.4738871175748284e-05 tokens_per_second_per_gpu:5045.420945458938 mfu:23.800111437309866 +Step 17099 | loss:0.7561678290367126 lr:5.4734554455742185e-05 tokens_per_second_per_gpu:4337.798095078669 mfu:20.462133719158278 +Step 17100 | loss:0.7095314264297485 lr:5.473023770012686e-05 tokens_per_second_per_gpu:5367.691276783526 mfu:25.32031557515634 +Step 17101 | loss:0.7938204407691956 lr:5.472592090893484e-05 tokens_per_second_per_gpu:5335.0508316796695 mfu:25.166345026567257 +Step 17102 | loss:0.6530554294586182 lr:5.472160408219855e-05 tokens_per_second_per_gpu:4290.2019633363225 mfu:20.23761464499217 +Step 17103 | loss:0.907474935054779 lr:5.47172872199505e-05 tokens_per_second_per_gpu:5330.919854716055 mfu:25.14685850341254 +Step 17104 | loss:0.6497154235839844 lr:5.47129703222231e-05 tokens_per_second_per_gpu:4817.056092179524 mfu:22.722875461329156 +Step 17105 | loss:0.8142147064208984 lr:5.470865338904886e-05 tokens_per_second_per_gpu:5350.720550412948 mfu:25.240261763361833 +Step 17106 | loss:0.9657933115959167 lr:5.4704336420460246e-05 tokens_per_second_per_gpu:4955.190597120407 mfu:23.37447948931212 +Step 17107 | loss:0.7321540117263794 lr:5.470001941648971e-05 tokens_per_second_per_gpu:5379.59605124114 mfu:25.376472427436877 +Step 17108 | loss:0.864237368106842 lr:5.4695702377169735e-05 tokens_per_second_per_gpu:4511.067743365628 mfu:21.279476212978288 +Step 17109 | loss:0.8065198659896851 lr:5.469138530253276e-05 tokens_per_second_per_gpu:5319.301411866844 mfu:25.09205232618205 +Step 17110 | loss:0.5994188189506531 lr:5.4687068192611314e-05 tokens_per_second_per_gpu:4596.964475229391 mfu:21.68466575267334 +Step 17111 | loss:0.8637994527816772 lr:5.46827510474378e-05 tokens_per_second_per_gpu:4628.459973660646 mfu:21.833235392459716 +Step 17112 | loss:0.7101882100105286 lr:5.467843386704473e-05 tokens_per_second_per_gpu:4504.27731487973 mfu:21.247444603243736 +Step 17113 | loss:0.6936941146850586 lr:5.4674116651464555e-05 tokens_per_second_per_gpu:5105.717374535442 mfu:24.08453998089536 +Step 17114 | loss:0.7612296938896179 lr:5.466979940072976e-05 tokens_per_second_per_gpu:5345.740410893754 mfu:25.216769595550375 +Step 17115 | loss:0.8059508800506592 lr:5.466548211487281e-05 tokens_per_second_per_gpu:4261.326533478786 mfu:20.101404315697817 +Step 17116 | loss:0.7467368245124817 lr:5.466116479392617e-05 tokens_per_second_per_gpu:4973.152135553782 mfu:23.459207130656807 +Step 17117 | loss:0.8155195116996765 lr:5.465684743792232e-05 tokens_per_second_per_gpu:4071.263082797529 mfu:19.204842590665756 +Step 17118 | loss:1.0011905431747437 lr:5.465253004689371e-05 tokens_per_second_per_gpu:5318.239780797292 mfu:25.087044431294036 +Step 17119 | loss:0.889956533908844 lr:5.464821262087284e-05 tokens_per_second_per_gpu:4680.333356054726 mfu:22.077930987724265 +Step 17120 | loss:0.8615761399269104 lr:5.464389515989215e-05 tokens_per_second_per_gpu:5348.454603375072 mfu:25.22957290457374 +Step 17121 | loss:0.7823867201805115 lr:5.4639577663984175e-05 tokens_per_second_per_gpu:4699.109783707089 mfu:22.166502600549364 +Step 17122 | loss:0.752923846244812 lr:5.463526013318131e-05 tokens_per_second_per_gpu:5206.37391823768 mfu:24.55935407131608 +Step 17123 | loss:0.8990561962127686 lr:5.463094256751607e-05 tokens_per_second_per_gpu:5335.146386187894 mfu:25.16679577348583 +Step 17124 | loss:0.8581292629241943 lr:5.4626624967020926e-05 tokens_per_second_per_gpu:3670.982033542307 mfu:17.31664858633951 +Step 17125 | loss:0.7012808322906494 lr:5.462230733172834e-05 tokens_per_second_per_gpu:5352.414239758368 mfu:25.24825118497738 +Step 17126 | loss:0.6972755789756775 lr:5.46179896616708e-05 tokens_per_second_per_gpu:5339.455295770094 mfu:25.187121635160853 +Step 17127 | loss:0.8277965188026428 lr:5.461367195688075e-05 tokens_per_second_per_gpu:5325.95919063803 mfu:25.123458204579926 +Step 17128 | loss:0.7437544465065002 lr:5.460935421739071e-05 tokens_per_second_per_gpu:4960.7803165100295 mfu:23.400847149377437 +Step 17129 | loss:0.7379228472709656 lr:5.4605036443233105e-05 tokens_per_second_per_gpu:3942.3234864468914 mfu:18.596612515316156 +Step 17130 | loss:0.8288944959640503 lr:5.460071863444044e-05 tokens_per_second_per_gpu:5327.667536860626 mfu:25.131516765185832 +Step 17131 | loss:0.6671488881111145 lr:5.4596400791045185e-05 tokens_per_second_per_gpu:5308.437996594321 mfu:25.040807742851786 +Step 17132 | loss:0.8572047352790833 lr:5.4592082913079826e-05 tokens_per_second_per_gpu:4332.417648854981 mfu:20.436753236322353 +Step 17133 | loss:0.8545072674751282 lr:5.458776500057682e-05 tokens_per_second_per_gpu:4071.7233678213415 mfu:19.2070138336558 +Step 17134 | loss:0.7228854894638062 lr:5.4583447053568625e-05 tokens_per_second_per_gpu:5370.711540824345 mfu:25.33456267594707 +Step 17135 | loss:0.890036940574646 lr:5.4579129072087776e-05 tokens_per_second_per_gpu:4832.34639125992 mfu:22.795002410885417 +Step 17136 | loss:0.6658555865287781 lr:5.4574811056166675e-05 tokens_per_second_per_gpu:5351.294131175599 mfu:25.24296744168223 +Step 17137 | loss:0.7435877323150635 lr:5.457049300583786e-05 tokens_per_second_per_gpu:5175.370601497836 mfu:24.41310613654294 +Step 17138 | loss:0.7012922763824463 lr:5.456617492113377e-05 tokens_per_second_per_gpu:3828.6817011251765 mfu:18.06054482466569 +Step 17139 | loss:0.7704418897628784 lr:5.456185680208692e-05 tokens_per_second_per_gpu:4865.61507663239 mfu:22.951936476010395 +Step 17140 | loss:0.7636464238166809 lr:5.4557538648729745e-05 tokens_per_second_per_gpu:5360.988517777523 mfu:25.288697517316418 +Step 17141 | loss:0.8396909832954407 lr:5.455322046109473e-05 tokens_per_second_per_gpu:5330.842748773958 mfu:25.146494781528038 +Step 17142 | loss:0.8869960308074951 lr:5.4548902239214384e-05 tokens_per_second_per_gpu:5340.8513806735955 mfu:25.193707205847318 +Step 17143 | loss:0.8155981302261353 lr:5.4544583983121157e-05 tokens_per_second_per_gpu:4986.351727893985 mfu:23.52147185980823 +Step 17144 | loss:0.8729101419448853 lr:5.4540265692847534e-05 tokens_per_second_per_gpu:5200.535324009684 mfu:24.531812426175723 +Step 17145 | loss:0.8411264419555664 lr:5.453594736842599e-05 tokens_per_second_per_gpu:3812.953896037929 mfu:17.98635408462889 +Step 17146 | loss:0.6890006065368652 lr:5.453162900988902e-05 tokens_per_second_per_gpu:5325.888727579756 mfu:25.12312581831181 +Step 17147 | loss:0.7393020987510681 lr:5.452731061726907e-05 tokens_per_second_per_gpu:4253.811581096731 mfu:20.065955003128572 +Step 17148 | loss:0.8008806109428406 lr:5.4522992190598654e-05 tokens_per_second_per_gpu:4628.391230023312 mfu:21.83291111699354 +Step 17149 | loss:0.7577465772628784 lr:5.451867372991023e-05 tokens_per_second_per_gpu:3798.2882542328716 mfu:17.91717375001792 +Step 17150 | loss:0.8392175436019897 lr:5.451435523523628e-05 tokens_per_second_per_gpu:3692.10937042289 mfu:17.4163098391006 +Step 17151 | loss:0.7587392926216125 lr:5.45100367066093e-05 tokens_per_second_per_gpu:5032.96028837422 mfu:23.741332391833925 +Step 17152 | loss:0.850471019744873 lr:5.450571814406175e-05 tokens_per_second_per_gpu:4084.813542418269 mfu:19.268762420643597 +Step 17153 | loss:0.7950040698051453 lr:5.450139954762613e-05 tokens_per_second_per_gpu:3836.230296032404 mfu:18.096152834766183 +Step 17154 | loss:0.9139013886451721 lr:5.449708091733489e-05 tokens_per_second_per_gpu:5251.334099949582 mfu:24.77143891944913 +Step 17155 | loss:0.8975324630737305 lr:5.4492762253220544e-05 tokens_per_second_per_gpu:3815.671392893324 mfu:17.99917298094856 +Step 17156 | loss:0.9569958448410034 lr:5.448844355531555e-05 tokens_per_second_per_gpu:4373.28658211048 mfu:20.629538967447584 +Step 17157 | loss:0.9104515910148621 lr:5.448412482365241e-05 tokens_per_second_per_gpu:5390.880012311277 mfu:25.429700797046998 +Step 17158 | loss:0.9489582777023315 lr:5.4479806058263584e-05 tokens_per_second_per_gpu:3410.075535704832 mfu:16.085908120801346 +Step 17159 | loss:0.821597158908844 lr:5.447548725918158e-05 tokens_per_second_per_gpu:4780.765298533899 mfu:22.551685595854863 +Step 17160 | loss:0.8185504078865051 lr:5.447116842643886e-05 tokens_per_second_per_gpu:5335.375340036355 mfu:25.167875787852573 +Step 17161 | loss:0.6707375645637512 lr:5.4466849560067914e-05 tokens_per_second_per_gpu:4864.629036663102 mfu:22.9472851572395 +Step 17162 | loss:0.7822163701057434 lr:5.446253066010122e-05 tokens_per_second_per_gpu:5377.515783465798 mfu:25.36665944941031 +Step 17163 | loss:0.6944615840911865 lr:5.445821172657124e-05 tokens_per_second_per_gpu:5317.044184739625 mfu:25.0814046007003 +Step 17164 | loss:0.5553491115570068 lr:5.445389275951052e-05 tokens_per_second_per_gpu:5319.41723402619 mfu:25.09259867906898 +Step 17165 | loss:0.8044712543487549 lr:5.4449573758951476e-05 tokens_per_second_per_gpu:3720.505430926696 mfu:17.55025890136456 +Step 17166 | loss:0.9599034190177917 lr:5.444525472492663e-05 tokens_per_second_per_gpu:4681.105544873731 mfu:22.08157353413251 +Step 17167 | loss:0.7804688215255737 lr:5.444093565746846e-05 tokens_per_second_per_gpu:3981.9064243901553 mfu:18.78333198714015 +Step 17168 | loss:0.8609026670455933 lr:5.443661655660944e-05 tokens_per_second_per_gpu:3855.152910574708 mfu:18.18541403609319 +Step 17169 | loss:0.7978872060775757 lr:5.4432297422382075e-05 tokens_per_second_per_gpu:4900.362328361526 mfu:23.1158451909092 +Step 17170 | loss:0.8375226855278015 lr:5.442797825481881e-05 tokens_per_second_per_gpu:3763.802619271499 mfu:17.75449912604908 +Step 17171 | loss:0.6493777632713318 lr:5.4423659053952184e-05 tokens_per_second_per_gpu:4652.389793881819 mfu:21.946116437291334 +Step 17172 | loss:0.757699728012085 lr:5.441933981981463e-05 tokens_per_second_per_gpu:5332.107610302346 mfu:25.152461349165073 +Step 17173 | loss:0.7454759478569031 lr:5.4415020552438664e-05 tokens_per_second_per_gpu:4110.789251128821 mfu:19.391294270544183 +Step 17174 | loss:0.8621519804000854 lr:5.441070125185676e-05 tokens_per_second_per_gpu:5313.463505309485 mfu:25.06451392490142 +Step 17175 | loss:0.8496891260147095 lr:5.4406381918101433e-05 tokens_per_second_per_gpu:5061.651358860161 mfu:23.876673066517508 +Step 17176 | loss:0.7450201511383057 lr:5.440206255120511e-05 tokens_per_second_per_gpu:5338.194347261292 mfu:25.181173525904832 +Step 17177 | loss:0.7431990504264832 lr:5.439774315120034e-05 tokens_per_second_per_gpu:4299.462779371944 mfu:20.28129949429034 +Step 17178 | loss:0.7016569972038269 lr:5.439342371811958e-05 tokens_per_second_per_gpu:2993.7109730292627 mfu:14.121845439540971 +Step 17179 | loss:0.7958739399909973 lr:5.438910425199531e-05 tokens_per_second_per_gpu:5373.537337098144 mfu:25.34789243910114 +Step 17180 | loss:0.7346429228782654 lr:5.438478475286003e-05 tokens_per_second_per_gpu:4206.2162214657055 mfu:19.8414395711433 +Step 17181 | loss:0.761594831943512 lr:5.438046522074622e-05 tokens_per_second_per_gpu:5332.137618891215 mfu:25.152602904798762 +Step 17182 | loss:0.8363879919052124 lr:5.4376145655686394e-05 tokens_per_second_per_gpu:3712.8463506008197 mfu:17.51412971269355 +Step 17183 | loss:0.7740087509155273 lr:5.4371826057712994e-05 tokens_per_second_per_gpu:5035.7017932990075 mfu:23.75426453831312 +Step 17184 | loss:0.757094144821167 lr:5.436750642685854e-05 tokens_per_second_per_gpu:5354.074005104252 mfu:25.25608058129908 +Step 17185 | loss:0.7815366387367249 lr:5.436318676315551e-05 tokens_per_second_per_gpu:4335.5127040998095 mfu:20.45135313998309 +Step 17186 | loss:0.8277832865715027 lr:5.4358867066636396e-05 tokens_per_second_per_gpu:3544.6533423803003 mfu:16.720734596231047 +Step 17187 | loss:0.7965628504753113 lr:5.4354547337333695e-05 tokens_per_second_per_gpu:5370.662608379009 mfu:25.334331853253683 +Step 17188 | loss:0.8105013966560364 lr:5.4350227575279876e-05 tokens_per_second_per_gpu:4241.292365839754 mfu:20.00689972876303 +Step 17189 | loss:0.7797368764877319 lr:5.434590778050746e-05 tokens_per_second_per_gpu:4368.675079104474 mfu:20.607785720964714 +Step 17190 | loss:0.7322120666503906 lr:5.434158795304889e-05 tokens_per_second_per_gpu:4499.536624386386 mfu:21.22508195734143 +Step 17191 | loss:0.7434898614883423 lr:5.43372680929367e-05 tokens_per_second_per_gpu:5101.803401965011 mfu:24.06607710448806 +Step 17192 | loss:0.6868218183517456 lr:5.4332948200203346e-05 tokens_per_second_per_gpu:5360.877419282521 mfu:25.288173446760133 +Step 17193 | loss:0.8037287592887878 lr:5.4328628274881365e-05 tokens_per_second_per_gpu:4982.977906863823 mfu:23.50555697037573 +Step 17194 | loss:0.7689567804336548 lr:5.432430831700319e-05 tokens_per_second_per_gpu:4289.558810516516 mfu:20.234580783407957 +Step 17195 | loss:0.8027390241622925 lr:5.431998832660136e-05 tokens_per_second_per_gpu:3859.225722531805 mfu:18.204626184988186 +Step 17196 | loss:0.7048689126968384 lr:5.431566830370833e-05 tokens_per_second_per_gpu:4856.31019032549 mfu:22.90804373972344 +Step 17197 | loss:0.7212580442428589 lr:5.431134824835663e-05 tokens_per_second_per_gpu:4660.839378144578 mfu:21.9859745679065 +Step 17198 | loss:0.9448813796043396 lr:5.430702816057871e-05 tokens_per_second_per_gpu:4398.140876313096 mfu:20.74678091378119 +Step 17199 | loss:0.6919164657592773 lr:5.430270804040708e-05 tokens_per_second_per_gpu:5384.11128323225 mfu:25.39777154711696 +Step 17200 | loss:0.8219258785247803 lr:5.429838788787426e-05 tokens_per_second_per_gpu:5296.587045292186 mfu:24.984904783578827 +Step 17201 | loss:0.7428585290908813 lr:5.429406770301268e-05 tokens_per_second_per_gpu:4076.0702758056764 mfu:19.227518940276887 +Step 17202 | loss:0.7660862803459167 lr:5.4289747485854905e-05 tokens_per_second_per_gpu:4358.48433042305 mfu:20.559714220713417 +Step 17203 | loss:0.79764324426651 lr:5.428542723643337e-05 tokens_per_second_per_gpu:3814.164472552854 mfu:17.992064580595443 +Step 17204 | loss:0.8801929354667664 lr:5.4281106954780594e-05 tokens_per_second_per_gpu:3244.8018100899594 mfu:15.306283758470654 +Step 17205 | loss:0.7689312100410461 lr:5.4276786640929066e-05 tokens_per_second_per_gpu:4101.972713764764 mfu:19.349705159541536 +Step 17206 | loss:0.7817440629005432 lr:5.427246629491127e-05 tokens_per_second_per_gpu:5354.295662614896 mfu:25.257126177595463 +Step 17207 | loss:0.8070172667503357 lr:5.426814591675974e-05 tokens_per_second_per_gpu:4806.928053879396 mfu:22.675099776645933 +Step 17208 | loss:0.7580961585044861 lr:5.426382550650691e-05 tokens_per_second_per_gpu:3186.5870347325413 mfu:15.031674730644832 +Step 17209 | loss:0.7553607225418091 lr:5.425950506418531e-05 tokens_per_second_per_gpu:4292.318275166159 mfu:20.247597648973542 +Step 17210 | loss:0.7131025195121765 lr:5.4255184589827434e-05 tokens_per_second_per_gpu:5355.116630202595 mfu:25.260998821778152 +Step 17211 | loss:0.8310694098472595 lr:5.425086408346578e-05 tokens_per_second_per_gpu:4988.360961758383 mfu:23.53094975875724 +Step 17212 | loss:0.8194672465324402 lr:5.424654354513282e-05 tokens_per_second_per_gpu:5186.991144078826 mfu:24.467922218566482 +Step 17213 | loss:0.7269012331962585 lr:5.424222297486107e-05 tokens_per_second_per_gpu:5346.945662527184 mfu:25.222454973142398 +Step 17214 | loss:0.8179126381874084 lr:5.423790237268303e-05 tokens_per_second_per_gpu:3886.327684937547 mfu:18.332470765727678 +Step 17215 | loss:0.68406081199646 lr:5.4233581738631165e-05 tokens_per_second_per_gpu:5359.825639189245 mfu:25.28321202060796 +Step 17216 | loss:0.661339521408081 lr:5.4229261072738005e-05 tokens_per_second_per_gpu:4910.013997478743 mfu:23.161373760879673 +Step 17217 | loss:0.8803350925445557 lr:5.422494037503603e-05 tokens_per_second_per_gpu:4940.075252397067 mfu:23.30317782930838 +Step 17218 | loss:0.7814144492149353 lr:5.4220619645557744e-05 tokens_per_second_per_gpu:3920.108176941739 mfu:18.49181911005771 +Step 17219 | loss:0.7599433064460754 lr:5.421629888433563e-05 tokens_per_second_per_gpu:4152.32691688441 mfu:19.587234526969866 +Step 17220 | loss:0.7865166068077087 lr:5.4211978091402196e-05 tokens_per_second_per_gpu:4500.534425892389 mfu:21.229788757287736 +Step 17221 | loss:0.811103880405426 lr:5.4207657266789935e-05 tokens_per_second_per_gpu:5344.360510026555 mfu:25.210260367724434 +Step 17222 | loss:0.7195359468460083 lr:5.420333641053136e-05 tokens_per_second_per_gpu:4677.669479506799 mfu:22.065365027543486 +Step 17223 | loss:0.7827485799789429 lr:5.4199015522658944e-05 tokens_per_second_per_gpu:5345.695491433801 mfu:25.216557702793725 +Step 17224 | loss:0.7504209876060486 lr:5.41946946032052e-05 tokens_per_second_per_gpu:4298.517467461609 mfu:20.27684029672245 +Step 17225 | loss:0.75259929895401 lr:5.419037365220262e-05 tokens_per_second_per_gpu:3442.083080946866 mfu:16.236893172758364 +Step 17226 | loss:0.8551781177520752 lr:5.4186052669683704e-05 tokens_per_second_per_gpu:4978.277166968518 mfu:23.483382778260648 +Step 17227 | loss:0.7147997617721558 lr:5.4181731655680955e-05 tokens_per_second_per_gpu:3379.6705655557 mfu:15.942482688985052 +Step 17228 | loss:0.7039220333099365 lr:5.4177410610226863e-05 tokens_per_second_per_gpu:4359.473783221549 mfu:20.564381638381988 +Step 17229 | loss:0.8628088235855103 lr:5.417308953335395e-05 tokens_per_second_per_gpu:4054.417190265215 mfu:19.125377641385185 +Step 17230 | loss:0.7296968698501587 lr:5.4168768425094673e-05 tokens_per_second_per_gpu:3326.7017486813606 mfu:15.692619742376726 +Step 17231 | loss:0.8218169212341309 lr:5.416444728548158e-05 tokens_per_second_per_gpu:4701.905851017946 mfu:22.17969212711295 +Step 17232 | loss:0.678928554058075 lr:5.416012611454714e-05 tokens_per_second_per_gpu:5368.568489981413 mfu:25.32445353947878 +Step 17233 | loss:0.6777295470237732 lr:5.415580491232386e-05 tokens_per_second_per_gpu:3947.3119584035576 mfu:18.62014398865683 +Step 17234 | loss:0.8770918846130371 lr:5.415148367884424e-05 tokens_per_second_per_gpu:4110.147742309834 mfu:19.38826816399233 +Step 17235 | loss:0.715498685836792 lr:5.414716241414077e-05 tokens_per_second_per_gpu:4799.276702358256 mfu:22.639007046065533 +Step 17236 | loss:0.6321251392364502 lr:5.414284111824598e-05 tokens_per_second_per_gpu:5341.199585149602 mfu:25.195349745770447 +Step 17237 | loss:0.8597236275672913 lr:5.413851979119234e-05 tokens_per_second_per_gpu:4834.161827732494 mfu:22.80356613447195 +Step 17238 | loss:0.8702216744422913 lr:5.413419843301237e-05 tokens_per_second_per_gpu:4255.477478211124 mfu:20.073813324048523 +Step 17239 | loss:0.7979168891906738 lr:5.4129877043738574e-05 tokens_per_second_per_gpu:4554.406040457427 mfu:21.483910354636258 +Step 17240 | loss:0.8951929807662964 lr:5.4125555623403434e-05 tokens_per_second_per_gpu:3873.2411849448563 mfu:18.270739512475018 +Step 17241 | loss:0.6813148260116577 lr:5.412123417203947e-05 tokens_per_second_per_gpu:4549.741846379373 mfu:21.46190855537651 +Step 17242 | loss:0.9220812916755676 lr:5.411691268967918e-05 tokens_per_second_per_gpu:5074.008031146818 mfu:23.93496159795936 +Step 17243 | loss:0.7329557538032532 lr:5.4112591176355054e-05 tokens_per_second_per_gpu:3765.7784074918645 mfu:17.763819256188775 +Step 17244 | loss:0.8094850182533264 lr:5.410826963209962e-05 tokens_per_second_per_gpu:4533.29655414971 mfu:21.384333306073003 +Step 17245 | loss:0.7760300636291504 lr:5.4103948056945354e-05 tokens_per_second_per_gpu:4328.298896012294 mfu:20.4173243764317 +Step 17246 | loss:0.8037174344062805 lr:5.409962645092477e-05 tokens_per_second_per_gpu:5338.778083051464 mfu:25.18392710722267 +Step 17247 | loss:0.7893299460411072 lr:5.409530481407038e-05 tokens_per_second_per_gpu:4578.870232656864 mfu:21.5993121232626 +Step 17248 | loss:0.882255494594574 lr:5.4090983146414674e-05 tokens_per_second_per_gpu:5367.336151219287 mfu:25.31864038727916 +Step 17249 | loss:0.9031928181648254 lr:5.408666144799016e-05 tokens_per_second_per_gpu:4236.217848986087 mfu:19.98296236696243 +Step 17250 | loss:0.7458367347717285 lr:5.4082339718829356e-05 tokens_per_second_per_gpu:5314.449325459294 mfu:25.069164206746333 +Step 17251 | loss:0.6872028112411499 lr:5.4078017958964746e-05 tokens_per_second_per_gpu:5244.64523158811 mfu:24.73988638614944 +Step 17252 | loss:0.6927867531776428 lr:5.407369616842884e-05 tokens_per_second_per_gpu:5345.052952735887 mfu:25.213526738127257 +Step 17253 | loss:0.784479558467865 lr:5.406937434725414e-05 tokens_per_second_per_gpu:4091.475787119649 mfu:19.30018936559616 +Step 17254 | loss:0.7907417416572571 lr:5.4065052495473177e-05 tokens_per_second_per_gpu:4493.214621886097 mfu:21.195259992902713 +Step 17255 | loss:0.7737417221069336 lr:5.406073061311841e-05 tokens_per_second_per_gpu:5328.619493216232 mfu:25.136007305735728 +Step 17256 | loss:0.879851758480072 lr:5.405640870022239e-05 tokens_per_second_per_gpu:2823.7794642531767 mfu:13.320249519339082 +Step 17257 | loss:0.854760468006134 lr:5.40520867568176e-05 tokens_per_second_per_gpu:3843.53867523238 mfu:18.130627706390538 +Step 17258 | loss:0.808890700340271 lr:5.404776478293655e-05 tokens_per_second_per_gpu:4631.593719150848 mfu:21.848017804609118 +Step 17259 | loss:0.8516712784767151 lr:5.404344277861174e-05 tokens_per_second_per_gpu:5029.374202855906 mfu:23.724416214594758 +Step 17260 | loss:0.8109832406044006 lr:5.403912074387567e-05 tokens_per_second_per_gpu:4261.616763835517 mfu:20.102773381808923 +Step 17261 | loss:0.7401283979415894 lr:5.403479867876087e-05 tokens_per_second_per_gpu:5036.202981015253 mfu:23.75662872628209 +Step 17262 | loss:0.82514888048172 lr:5.4030476583299825e-05 tokens_per_second_per_gpu:5287.05010585753 mfu:24.9399174130959 +Step 17263 | loss:0.7529653906822205 lr:5.402615445752507e-05 tokens_per_second_per_gpu:4248.8731299707 mfu:20.042659486580437 +Step 17264 | loss:0.6847206354141235 lr:5.402183230146907e-05 tokens_per_second_per_gpu:3985.0357856799465 mfu:18.798093718268188 +Step 17265 | loss:0.6275317072868347 lr:5.4017510115164385e-05 tokens_per_second_per_gpu:5298.933163056373 mfu:24.99597182136146 +Step 17266 | loss:0.7078941464424133 lr:5.4013187898643455e-05 tokens_per_second_per_gpu:4667.403209368319 mfu:22.016937279693973 +Step 17267 | loss:0.8124531507492065 lr:5.400886565193886e-05 tokens_per_second_per_gpu:2772.804027113121 mfu:13.079789685043023 +Step 17268 | loss:0.7025448679924011 lr:5.400454337508306e-05 tokens_per_second_per_gpu:2962.6779330998856 mfu:13.975457295411507 +Step 17269 | loss:0.8365918397903442 lr:5.400022106810858e-05 tokens_per_second_per_gpu:3821.0571257965553 mfu:18.024578401953892 +Step 17270 | loss:0.9611502885818481 lr:5.399589873104794e-05 tokens_per_second_per_gpu:3803.328187798798 mfu:17.940947976549733 +Step 17271 | loss:0.7428262829780579 lr:5.39915763639336e-05 tokens_per_second_per_gpu:2995.2428319026512 mfu:14.12907147920882 +Step 17272 | loss:0.7776075005531311 lr:5.3987253966798136e-05 tokens_per_second_per_gpu:5146.721512818784 mfu:24.27796349720528 +Step 17273 | loss:0.7908624410629272 lr:5.398293153967402e-05 tokens_per_second_per_gpu:3658.248068615796 mfu:17.25658030114271 +Step 17274 | loss:0.8942846059799194 lr:5.3978609082593766e-05 tokens_per_second_per_gpu:3653.14265382536 mfu:17.232497188502638 +Step 17275 | loss:0.8248379826545715 lr:5.3974286595589886e-05 tokens_per_second_per_gpu:5278.447126478707 mfu:24.89933569154635 +Step 17276 | loss:0.8321413993835449 lr:5.396996407869489e-05 tokens_per_second_per_gpu:3649.509926814795 mfu:17.21536099538666 +Step 17277 | loss:0.6941034197807312 lr:5.396564153194129e-05 tokens_per_second_per_gpu:5362.485207054533 mfu:25.295757655997594 +Step 17278 | loss:0.9298728704452515 lr:5.396131895536159e-05 tokens_per_second_per_gpu:4879.002362731793 mfu:23.01508658864782 +Step 17279 | loss:0.7162323594093323 lr:5.3956996348988295e-05 tokens_per_second_per_gpu:5235.184110007241 mfu:24.695256661418686 +Step 17280 | loss:0.7248867750167847 lr:5.395267371285394e-05 tokens_per_second_per_gpu:4685.788858841457 mfu:22.103665525174033 +Step 17281 | loss:0.7475548386573792 lr:5.394835104699102e-05 tokens_per_second_per_gpu:3049.2318042465067 mfu:14.383746673223325 +Step 17282 | loss:0.768099844455719 lr:5.394402835143204e-05 tokens_per_second_per_gpu:4164.645263893615 mfu:19.645342271539008 +Step 17283 | loss:0.8480618000030518 lr:5.3939705626209535e-05 tokens_per_second_per_gpu:4168.904770908956 mfu:19.66543509287721 +Step 17284 | loss:0.7252310514450073 lr:5.3935382871356e-05 tokens_per_second_per_gpu:5167.714376187741 mfu:24.376990415468484 +Step 17285 | loss:0.7584133744239807 lr:5.393106008690393e-05 tokens_per_second_per_gpu:5345.947203164501 mfu:25.21774506997389 +Step 17286 | loss:0.7858471870422363 lr:5.392673727288586e-05 tokens_per_second_per_gpu:5118.214376709927 mfu:24.143490472360845 +Step 17287 | loss:0.7788317799568176 lr:5.392241442933431e-05 tokens_per_second_per_gpu:4656.613386424145 mfu:21.966039843932677 +Step 17288 | loss:0.8011854290962219 lr:5.391809155628177e-05 tokens_per_second_per_gpu:5120.3408128517885 mfu:24.153521234449425 +Step 17289 | loss:0.6659391522407532 lr:5.391376865376077e-05 tokens_per_second_per_gpu:5266.688368371516 mfu:24.843867623303716 +Step 17290 | loss:0.8369202613830566 lr:5.390944572180381e-05 tokens_per_second_per_gpu:4511.856995007662 mfu:21.283199247633974 +Step 17291 | loss:0.8473405838012695 lr:5.390512276044342e-05 tokens_per_second_per_gpu:5341.247623794555 mfu:25.19557635225493 +Step 17292 | loss:0.8350545167922974 lr:5.3900799769712095e-05 tokens_per_second_per_gpu:3951.0253867105675 mfu:18.637660863557198 +Step 17293 | loss:0.7571110129356384 lr:5.3896476749642355e-05 tokens_per_second_per_gpu:5347.357794945666 mfu:25.224399072077414 +Step 17294 | loss:0.7916603684425354 lr:5.3892153700266725e-05 tokens_per_second_per_gpu:3838.2382203908032 mfu:18.105624556551973 +Step 17295 | loss:0.8579678535461426 lr:5.3887830621617705e-05 tokens_per_second_per_gpu:3739.018161657721 mfu:17.63758661082037 +Step 17296 | loss:0.7807233929634094 lr:5.3883507513727806e-05 tokens_per_second_per_gpu:5034.389166631802 mfu:23.74807265436671 +Step 17297 | loss:0.7569842338562012 lr:5.387918437662957e-05 tokens_per_second_per_gpu:4799.932414046369 mfu:22.642100149974304 +Step 17298 | loss:0.5731973648071289 lr:5.387486121035548e-05 tokens_per_second_per_gpu:4781.583732021918 mfu:22.555546286257 +Step 17299 | loss:0.7231865525245667 lr:5.3870538014938066e-05 tokens_per_second_per_gpu:3746.817236751482 mfu:17.674376178697067 +Step 17300 | loss:0.855333149433136 lr:5.386621479040985e-05 tokens_per_second_per_gpu:3243.1249922173674 mfu:15.298373922471008 +Step 17301 | loss:0.6918208599090576 lr:5.386189153680333e-05 tokens_per_second_per_gpu:5349.5743091510185 mfu:25.234854747760384 +Step 17302 | loss:0.8354716897010803 lr:5.3857568254151047e-05 tokens_per_second_per_gpu:5309.226279686612 mfu:25.04452620869309 +Step 17303 | loss:0.7283954620361328 lr:5.385324494248547e-05 tokens_per_second_per_gpu:5073.43596036786 mfu:23.93226304248968 +Step 17304 | loss:0.7676199674606323 lr:5.384892160183917e-05 tokens_per_second_per_gpu:4729.6068189545595 mfu:22.310362319142442 +Step 17305 | loss:0.7795098423957825 lr:5.3844598232244634e-05 tokens_per_second_per_gpu:4950.766274127649 mfu:23.353609202887935 +Step 17306 | loss:0.8140522837638855 lr:5.384027483373438e-05 tokens_per_second_per_gpu:3897.7975414849393 mfu:18.386576036021047 +Step 17307 | loss:0.7051284313201904 lr:5.383595140634093e-05 tokens_per_second_per_gpu:5317.1808518154085 mfu:25.082049282614577 +Step 17308 | loss:1.0686851739883423 lr:5.383162795009681e-05 tokens_per_second_per_gpu:2734.9444074339303 mfu:12.90119939949908 +Step 17309 | loss:0.729429304599762 lr:5.382730446503452e-05 tokens_per_second_per_gpu:3376.4759286956155 mfu:15.92741304185478 +Step 17310 | loss:0.8361438512802124 lr:5.3822980951186587e-05 tokens_per_second_per_gpu:3902.4926300431293 mfu:18.408723569814743 +Step 17311 | loss:0.7858029007911682 lr:5.381865740858553e-05 tokens_per_second_per_gpu:4777.474809822877 mfu:22.536163799189712 +Step 17312 | loss:0.8476217985153198 lr:5.381433383726385e-05 tokens_per_second_per_gpu:4465.012294727749 mfu:21.06222480388359 +Step 17313 | loss:0.7460705041885376 lr:5.381001023725409e-05 tokens_per_second_per_gpu:4798.668865276123 mfu:22.636139774841407 +Step 17314 | loss:0.7532870769500732 lr:5.380568660858875e-05 tokens_per_second_per_gpu:4302.688293910741 mfu:20.296514796699377 +Step 17315 | loss:0.6895495057106018 lr:5.380136295130036e-05 tokens_per_second_per_gpu:4275.609363048602 mfu:20.168778859680646 +Step 17316 | loss:0.9543850421905518 lr:5.379703926542143e-05 tokens_per_second_per_gpu:5344.19092514371 mfu:25.20946040689789 +Step 17317 | loss:0.7556762099266052 lr:5.379271555098448e-05 tokens_per_second_per_gpu:4645.908042485661 mfu:21.915540910054826 +Step 17318 | loss:0.7894019484519958 lr:5.3788391808022056e-05 tokens_per_second_per_gpu:2946.448876976649 mfu:13.8989020687157 +Step 17319 | loss:0.709648847579956 lr:5.378406803656663e-05 tokens_per_second_per_gpu:4239.680813910976 mfu:19.999297763355873 +Step 17320 | loss:0.6835757493972778 lr:5.3779744236650756e-05 tokens_per_second_per_gpu:4177.124592507966 mfu:19.704209393805918 +Step 17321 | loss:0.8125841617584229 lr:5.377542040830693e-05 tokens_per_second_per_gpu:4800.058105137058 mfu:22.6426930562942 +Step 17322 | loss:0.8109144568443298 lr:5.37710965515677e-05 tokens_per_second_per_gpu:5334.429166206758 mfu:25.163412524464466 +Step 17323 | loss:0.7510148286819458 lr:5.376677266646556e-05 tokens_per_second_per_gpu:3877.977215128154 mfu:18.29308017489971 +Step 17324 | loss:0.7542845606803894 lr:5.3762448753033056e-05 tokens_per_second_per_gpu:4397.682616856274 mfu:20.744619225736233 +Step 17325 | loss:0.8954795002937317 lr:5.375812481130268e-05 tokens_per_second_per_gpu:4947.565127417453 mfu:23.338508847683997 +Step 17326 | loss:0.7679861187934875 lr:5.3753800841306964e-05 tokens_per_second_per_gpu:5326.655280595931 mfu:25.12674177967654 +Step 17327 | loss:0.773891270160675 lr:5.374947684307845e-05 tokens_per_second_per_gpu:5373.9971607584175 mfu:25.350061505760657 +Step 17328 | loss:0.743678092956543 lr:5.374515281664965e-05 tokens_per_second_per_gpu:3453.3662392682813 mfu:16.2901177556658 +Step 17329 | loss:0.7882975339889526 lr:5.3740828762053054e-05 tokens_per_second_per_gpu:3197.954292390905 mfu:15.08529602447348 +Step 17330 | loss:0.7905701994895935 lr:5.373650467932122e-05 tokens_per_second_per_gpu:5327.666325162991 mfu:25.13151104940135 +Step 17331 | loss:0.8473004102706909 lr:5.3732180568486656e-05 tokens_per_second_per_gpu:5301.456427339899 mfu:25.007874489500132 +Step 17332 | loss:0.9029656052589417 lr:5.372785642958188e-05 tokens_per_second_per_gpu:4315.761578046479 mfu:20.358183708499936 +Step 17333 | loss:0.7656788229942322 lr:5.372353226263942e-05 tokens_per_second_per_gpu:5384.821138179805 mfu:25.401120054018318 +Step 17334 | loss:0.7253896594047546 lr:5.3719208067691804e-05 tokens_per_second_per_gpu:5331.589317270862 mfu:25.150016472505655 +Step 17335 | loss:0.649327278137207 lr:5.3714883844771555e-05 tokens_per_second_per_gpu:5325.0022658310945 mfu:25.118944226997403 +Step 17336 | loss:0.7117882370948792 lr:5.371055959391119e-05 tokens_per_second_per_gpu:4030.63278024963 mfu:19.013182521302866 +Step 17337 | loss:0.8095000982284546 lr:5.3706235315143216e-05 tokens_per_second_per_gpu:4078.328283489895 mfu:19.23817034777918 +Step 17338 | loss:0.7449324131011963 lr:5.370191100850019e-05 tokens_per_second_per_gpu:2932.3759580015676 mfu:13.83251770882267 +Step 17339 | loss:0.6959772109985352 lr:5.369758667401461e-05 tokens_per_second_per_gpu:5177.336520152617 mfu:24.422379710257847 +Step 17340 | loss:0.7395921945571899 lr:5.369326231171902e-05 tokens_per_second_per_gpu:4069.072558984794 mfu:19.194509516099178 +Step 17341 | loss:0.7728205919265747 lr:5.368893792164592e-05 tokens_per_second_per_gpu:3201.2794181555328 mfu:15.100981210030188 +Step 17342 | loss:0.9245039224624634 lr:5.3684613503827855e-05 tokens_per_second_per_gpu:5343.276240452183 mfu:25.205145683146654 +Step 17343 | loss:0.6630133390426636 lr:5.3680289058297325e-05 tokens_per_second_per_gpu:4896.378455310218 mfu:23.09705258200735 +Step 17344 | loss:0.8708797097206116 lr:5.367596458508689e-05 tokens_per_second_per_gpu:4509.3240842705645 mfu:21.271251075527953 +Step 17345 | loss:0.709392249584198 lr:5.367164008422906e-05 tokens_per_second_per_gpu:5248.044195630947 mfu:24.755919879462567 +Step 17346 | loss:0.7204450964927673 lr:5.366731555575635e-05 tokens_per_second_per_gpu:4940.084076703305 mfu:23.303219455066486 +Step 17347 | loss:0.7906204462051392 lr:5.366299099970129e-05 tokens_per_second_per_gpu:4614.954088070159 mfu:21.769525825787692 +Step 17348 | loss:0.8573344945907593 lr:5.3658666416096406e-05 tokens_per_second_per_gpu:5288.384020543808 mfu:24.946209716260384 +Step 17349 | loss:0.7766316533088684 lr:5.365434180497423e-05 tokens_per_second_per_gpu:4759.229732812072 mfu:22.45009865799861 +Step 17350 | loss:0.612596869468689 lr:5.3650017166367275e-05 tokens_per_second_per_gpu:5340.118789235736 mfu:25.19025144704194 +Step 17351 | loss:0.8651415109634399 lr:5.364569250030807e-05 tokens_per_second_per_gpu:5331.897005346379 mfu:25.151467889654867 +Step 17352 | loss:0.7050466537475586 lr:5.364136780682916e-05 tokens_per_second_per_gpu:5332.306620680773 mfu:25.15340011507479 +Step 17353 | loss:0.792881190776825 lr:5.3637043085963056e-05 tokens_per_second_per_gpu:4510.723976695756 mfu:21.27785460694386 +Step 17354 | loss:0.7608039379119873 lr:5.363271833774228e-05 tokens_per_second_per_gpu:4777.848081165376 mfu:22.537924583798937 +Step 17355 | loss:0.8944778442382812 lr:5.362839356219937e-05 tokens_per_second_per_gpu:3625.263607573421 mfu:17.100986970676455 +Step 17356 | loss:0.8172339200973511 lr:5.3624068759366866e-05 tokens_per_second_per_gpu:3635.7502290505813 mfu:17.150454153386345 +Step 17357 | loss:0.7605260610580444 lr:5.3619743929277246e-05 tokens_per_second_per_gpu:5218.070681731999 mfu:24.61452970039242 +Step 17358 | loss:0.8360669016838074 lr:5.361541907196309e-05 tokens_per_second_per_gpu:3972.547867752955 mfu:18.739186078748062 +Step 17359 | loss:0.8095428943634033 lr:5.36110941874569e-05 tokens_per_second_per_gpu:3459.7308305042375 mfu:16.320140618437044 +Step 17360 | loss:0.8516318202018738 lr:5.3606769275791215e-05 tokens_per_second_per_gpu:4329.217748540858 mfu:20.421658760581348 +Step 17361 | loss:0.820932149887085 lr:5.360244433699855e-05 tokens_per_second_per_gpu:5382.313762305857 mfu:25.38929233421847 +Step 17362 | loss:0.6661047339439392 lr:5.359811937111144e-05 tokens_per_second_per_gpu:4548.0918713766205 mfu:21.45412534177376 +Step 17363 | loss:0.7410504817962646 lr:5.359379437816243e-05 tokens_per_second_per_gpu:5330.425118287506 mfu:25.14452474726833 +Step 17364 | loss:0.8157928586006165 lr:5.358946935818403e-05 tokens_per_second_per_gpu:4623.577174351879 mfu:21.810202395028796 +Step 17365 | loss:0.8058041334152222 lr:5.3585144311208765e-05 tokens_per_second_per_gpu:5279.05990505783 mfu:24.90222627265488 +Step 17366 | loss:0.815361738204956 lr:5.358081923726917e-05 tokens_per_second_per_gpu:4747.396316179074 mfu:22.39427841275169 +Step 17367 | loss:0.9001187682151794 lr:5.357649413639779e-05 tokens_per_second_per_gpu:4778.801857055425 mfu:22.54242370740324 +Step 17368 | loss:0.7776783108711243 lr:5.357216900862713e-05 tokens_per_second_per_gpu:4813.684694713627 mfu:22.7069720042628 +Step 17369 | loss:0.8237230181694031 lr:5.356784385398973e-05 tokens_per_second_per_gpu:5343.212640841455 mfu:25.204845672931636 +Step 17370 | loss:0.7339670658111572 lr:5.356351867251813e-05 tokens_per_second_per_gpu:5329.7240428397035 mfu:25.141217654765242 +Step 17371 | loss:0.7322375774383545 lr:5.355919346424485e-05 tokens_per_second_per_gpu:4889.067098462151 mfu:23.062563664308907 +Step 17372 | loss:0.7059462666511536 lr:5.3554868229202416e-05 tokens_per_second_per_gpu:4199.327170503579 mfu:19.80894274236165 +Step 17373 | loss:0.7342060208320618 lr:5.3550542967423356e-05 tokens_per_second_per_gpu:4900.103211248657 mfu:23.114622891277865 +Step 17374 | loss:0.6474135518074036 lr:5.3546217678940234e-05 tokens_per_second_per_gpu:5133.266014152108 mfu:24.21449161424977 +Step 17375 | loss:0.866685688495636 lr:5.354189236378554e-05 tokens_per_second_per_gpu:3932.789340873609 mfu:18.55163832395401 +Step 17376 | loss:0.7768243551254272 lr:5.353756702199182e-05 tokens_per_second_per_gpu:5308.239778781532 mfu:25.039872715609416 +Step 17377 | loss:0.7976224422454834 lr:5.353324165359162e-05 tokens_per_second_per_gpu:5342.7063972081105 mfu:25.202457635339083 +Step 17378 | loss:0.845177948474884 lr:5.352891625861745e-05 tokens_per_second_per_gpu:5324.884812609177 mfu:25.118390180110033 +Step 17379 | loss:1.0069737434387207 lr:5.352459083710184e-05 tokens_per_second_per_gpu:4677.983115351124 mfu:22.066844501332962 +Step 17380 | loss:0.8124655485153198 lr:5.352026538907734e-05 tokens_per_second_per_gpu:5372.940056142079 mfu:25.34507496292474 +Step 17381 | loss:0.7679329514503479 lr:5.3515939914576486e-05 tokens_per_second_per_gpu:4108.211413737138 mfu:19.3791341717432 +Step 17382 | loss:0.9245133399963379 lr:5.351161441363179e-05 tokens_per_second_per_gpu:5308.655374059796 mfu:25.04183314944485 +Step 17383 | loss:0.7825935482978821 lr:5.3507288886275796e-05 tokens_per_second_per_gpu:3826.0113544156793 mfu:18.047948343628864 +Step 17384 | loss:0.6850849390029907 lr:5.350296333254102e-05 tokens_per_second_per_gpu:4223.868391362481 mfu:19.92470786831714 +Step 17385 | loss:0.7093287110328674 lr:5.349863775246004e-05 tokens_per_second_per_gpu:4557.0795025707675 mfu:21.496521531564557 +Step 17386 | loss:0.6272374391555786 lr:5.349431214606534e-05 tokens_per_second_per_gpu:3762.2762748954997 mfu:17.747299099206284 +Step 17387 | loss:0.8050668239593506 lr:5.348998651338948e-05 tokens_per_second_per_gpu:4261.481195076027 mfu:20.102133880839993 +Step 17388 | loss:0.8024632930755615 lr:5.348566085446498e-05 tokens_per_second_per_gpu:3862.6691678225957 mfu:18.220869503937198 +Step 17389 | loss:0.8648826479911804 lr:5.3481335169324375e-05 tokens_per_second_per_gpu:5350.21260818621 mfu:25.237865713214426 +Step 17390 | loss:0.7109202742576599 lr:5.3477009458000216e-05 tokens_per_second_per_gpu:5301.47463855101 mfu:25.007960394890095 +Step 17391 | loss:0.6720293760299683 lr:5.3472683720525006e-05 tokens_per_second_per_gpu:4757.594108084111 mfu:22.44238313709046 +Step 17392 | loss:0.8656260371208191 lr:5.3468357956931325e-05 tokens_per_second_per_gpu:3729.6746642537923 mfu:17.593511739401 +Step 17393 | loss:0.7331082224845886 lr:5.3464032167251666e-05 tokens_per_second_per_gpu:5322.3788281836305 mfu:25.106569024009698 +Step 17394 | loss:0.8644323348999023 lr:5.345970635151858e-05 tokens_per_second_per_gpu:5375.962722358806 mfu:25.359333395189076 +Step 17395 | loss:0.8091251254081726 lr:5.3455380509764596e-05 tokens_per_second_per_gpu:5367.3700325932905 mfu:25.31880021150109 +Step 17396 | loss:0.7904798984527588 lr:5.345105464202227e-05 tokens_per_second_per_gpu:4360.116557187504 mfu:20.56741371284795 +Step 17397 | loss:0.8620792031288147 lr:5.3446728748324106e-05 tokens_per_second_per_gpu:5365.052370511 mfu:25.307867404025 +Step 17398 | loss:0.9424169063568115 lr:5.344240282870267e-05 tokens_per_second_per_gpu:4613.272474042018 mfu:21.761593365504012 +Step 17399 | loss:0.8138833045959473 lr:5.3438076883190466e-05 tokens_per_second_per_gpu:4706.667783647802 mfu:22.20215497579747 +Step 17400 | loss:0.8305720686912537 lr:5.343375091182006e-05 tokens_per_second_per_gpu:5373.972349289946 mfu:25.3499444658306 +Step 17401 | loss:0.7378368377685547 lr:5.3429424914623974e-05 tokens_per_second_per_gpu:3451.213336521991 mfu:16.27996214608846 +Step 17402 | loss:0.722602903842926 lr:5.3425098891634726e-05 tokens_per_second_per_gpu:4415.727308790159 mfu:20.829739116330288 +Step 17403 | loss:0.6960430145263672 lr:5.3420772842884904e-05 tokens_per_second_per_gpu:5332.165946998733 mfu:25.152736533315196 +Step 17404 | loss:0.7646875977516174 lr:5.341644676840699e-05 tokens_per_second_per_gpu:4654.257681570756 mfu:21.954927582214548 +Step 17405 | loss:0.7121819257736206 lr:5.341212066823355e-05 tokens_per_second_per_gpu:4244.076566037246 mfu:20.020033276128142 +Step 17406 | loss:0.7294426560401917 lr:5.3407794542397125e-05 tokens_per_second_per_gpu:5001.750880441504 mfu:23.594112289741226 +Step 17407 | loss:0.6602545380592346 lr:5.340346839093023e-05 tokens_per_second_per_gpu:5349.7297260755695 mfu:25.235587875910355 +Step 17408 | loss:0.7095581889152527 lr:5.339914221386543e-05 tokens_per_second_per_gpu:5088.463917082671 mfu:24.003152478347253 +Step 17409 | loss:0.7959359884262085 lr:5.339481601123522e-05 tokens_per_second_per_gpu:4063.2564128443814 mfu:19.167073762418 +Step 17410 | loss:0.6982115507125854 lr:5.3390489783072195e-05 tokens_per_second_per_gpu:4265.289016616267 mfu:20.120096024726653 +Step 17411 | loss:0.886990487575531 lr:5.338616352940884e-05 tokens_per_second_per_gpu:3853.420525298393 mfu:18.177242079169403 +Step 17412 | loss:0.8463783264160156 lr:5.338183725027773e-05 tokens_per_second_per_gpu:5353.5719313074 mfu:25.25371221353671 +Step 17413 | loss:0.7578935027122498 lr:5.337751094571139e-05 tokens_per_second_per_gpu:5208.75310219061 mfu:24.57057708795269 +Step 17414 | loss:0.7811770439147949 lr:5.337318461574236e-05 tokens_per_second_per_gpu:4300.428511307179 mfu:20.285855016599704 +Step 17415 | loss:0.6704992651939392 lr:5.3368858260403165e-05 tokens_per_second_per_gpu:5322.92043509336 mfu:25.10912387621056 +Step 17416 | loss:0.9865183234214783 lr:5.336453187972638e-05 tokens_per_second_per_gpu:5314.342391810453 mfu:25.068659782479983 +Step 17417 | loss:0.6764985918998718 lr:5.336020547374451e-05 tokens_per_second_per_gpu:5183.377217263195 mfu:24.450874709177654 +Step 17418 | loss:0.8622471690177917 lr:5.33558790424901e-05 tokens_per_second_per_gpu:3595.877082763629 mfu:16.962365719290524 +Step 17419 | loss:0.7940700650215149 lr:5.3351552585995704e-05 tokens_per_second_per_gpu:5328.797119095182 mfu:25.136845197312837 +Step 17420 | loss:0.7699491381645203 lr:5.334722610429383e-05 tokens_per_second_per_gpu:3439.7563604042407 mfu:16.22591763497934 +Step 17421 | loss:0.7642648816108704 lr:5.3342899597417086e-05 tokens_per_second_per_gpu:4416.111884521777 mfu:20.831553225671712 +Step 17422 | loss:0.7813162803649902 lr:5.3338573065397934e-05 tokens_per_second_per_gpu:5325.970213105161 mfu:25.12351019943801 +Step 17423 | loss:0.8405375480651855 lr:5.333424650826896e-05 tokens_per_second_per_gpu:5250.39941282813 mfu:24.76702984082317 +Step 17424 | loss:0.8275325894355774 lr:5.332991992606269e-05 tokens_per_second_per_gpu:4059.730748718479 mfu:19.150442603195785 +Step 17425 | loss:0.6383861899375916 lr:5.332559331881166e-05 tokens_per_second_per_gpu:4291.089678976774 mfu:20.241802151128084 +Step 17426 | loss:0.5918876528739929 lr:5.3321266686548435e-05 tokens_per_second_per_gpu:5365.732721793855 mfu:25.311076737106124 +Step 17427 | loss:0.7226012945175171 lr:5.331694002930552e-05 tokens_per_second_per_gpu:3571.5971958863674 mfu:16.847833350314612 +Step 17428 | loss:0.9698007106781006 lr:5.33126133471155e-05 tokens_per_second_per_gpu:3749.4482454821446 mfu:17.686787095775983 +Step 17429 | loss:0.7485325932502747 lr:5.3308286640010866e-05 tokens_per_second_per_gpu:3611.1152250057557 mfu:17.034246636141795 +Step 17430 | loss:0.9665002226829529 lr:5.33039599080242e-05 tokens_per_second_per_gpu:3793.332981528556 mfu:17.89379888321496 +Step 17431 | loss:0.7705621123313904 lr:5.329963315118802e-05 tokens_per_second_per_gpu:3356.3357130289833 mfu:15.832408208280146 +Step 17432 | loss:0.7452437281608582 lr:5.32953063695349e-05 tokens_per_second_per_gpu:5367.508850801262 mfu:25.31945504067292 +Step 17433 | loss:0.6979110836982727 lr:5.329097956309733e-05 tokens_per_second_per_gpu:5341.693602415116 mfu:25.197680109499917 +Step 17434 | loss:0.7264547348022461 lr:5.32866527319079e-05 tokens_per_second_per_gpu:4968.214922905241 mfu:23.435917456218302 +Step 17435 | loss:0.7807906866073608 lr:5.328232587599914e-05 tokens_per_second_per_gpu:5194.169364854124 mfu:24.50178311069432 +Step 17436 | loss:0.7235902547836304 lr:5.327799899540359e-05 tokens_per_second_per_gpu:5169.096381639514 mfu:24.383509570978944 +Step 17437 | loss:0.8634235262870789 lr:5.3273672090153784e-05 tokens_per_second_per_gpu:5017.427366629864 mfu:23.66806094977607 +Step 17438 | loss:0.981404721736908 lr:5.326934516028226e-05 tokens_per_second_per_gpu:5357.675163929406 mfu:25.273067861898387 +Step 17439 | loss:0.8134684562683105 lr:5.3265018205821595e-05 tokens_per_second_per_gpu:5309.783029886834 mfu:25.047152494378917 +Step 17440 | loss:0.8309429287910461 lr:5.326069122680429e-05 tokens_per_second_per_gpu:4353.675717082319 mfu:20.537031171151156 +Step 17441 | loss:0.817380428314209 lr:5.325636422326293e-05 tokens_per_second_per_gpu:5212.6468198830125 mfu:24.588944418644495 +Step 17442 | loss:0.803180992603302 lr:5.3252037195230034e-05 tokens_per_second_per_gpu:3942.916495920277 mfu:18.599409842179966 +Step 17443 | loss:0.7897238731384277 lr:5.324771014273815e-05 tokens_per_second_per_gpu:5126.7307563795775 mfu:24.18366368830679 +Step 17444 | loss:0.7483540773391724 lr:5.324338306581983e-05 tokens_per_second_per_gpu:4444.420738782341 mfu:20.96509092120738 +Step 17445 | loss:0.6270191073417664 lr:5.323905596450759e-05 tokens_per_second_per_gpu:5307.980051875314 mfu:25.038647539478802 +Step 17446 | loss:0.7371177077293396 lr:5.3234728838834024e-05 tokens_per_second_per_gpu:4823.609972056273 mfu:22.753791230914995 +Step 17447 | loss:0.7620657682418823 lr:5.3230401688831635e-05 tokens_per_second_per_gpu:4371.697081962755 mfu:20.622041024054074 +Step 17448 | loss:0.6568976044654846 lr:5.322607451453299e-05 tokens_per_second_per_gpu:4397.172135518744 mfu:20.74221119817001 +Step 17449 | loss:0.7231127619743347 lr:5.3221747315970605e-05 tokens_per_second_per_gpu:5328.303608985545 mfu:25.13451722592417 +Step 17450 | loss:0.7879785299301147 lr:5.3217420093177085e-05 tokens_per_second_per_gpu:5335.483015006403 mfu:25.168383709057537 +Step 17451 | loss:0.8568953275680542 lr:5.3213092846184896e-05 tokens_per_second_per_gpu:5134.649537505478 mfu:24.221017929961043 +Step 17452 | loss:0.8542123436927795 lr:5.3208765575026655e-05 tokens_per_second_per_gpu:3767.5552127028654 mfu:17.772200749523293 +Step 17453 | loss:0.8343306183815002 lr:5.320443827973487e-05 tokens_per_second_per_gpu:4417.405269944272 mfu:20.837654345383932 +Step 17454 | loss:0.9538135528564453 lr:5.3200110960342096e-05 tokens_per_second_per_gpu:4883.546610734026 mfu:23.03652258180301 +Step 17455 | loss:0.8692689538002014 lr:5.319578361688089e-05 tokens_per_second_per_gpu:3570.070101479371 mfu:16.840629785447636 +Step 17456 | loss:0.8076930642127991 lr:5.3191456249383766e-05 tokens_per_second_per_gpu:4393.353720369594 mfu:20.724199082422665 +Step 17457 | loss:0.7555249333381653 lr:5.318712885788332e-05 tokens_per_second_per_gpu:4676.395925058675 mfu:22.05935745392089 +Step 17458 | loss:0.7601264119148254 lr:5.318280144241204e-05 tokens_per_second_per_gpu:3641.5103047927096 mfu:17.177625413432224 +Step 17459 | loss:0.7797280550003052 lr:5.317847400300252e-05 tokens_per_second_per_gpu:4999.1730488709545 mfu:23.581952218399525 +Step 17460 | loss:0.8536075949668884 lr:5.3174146539687295e-05 tokens_per_second_per_gpu:3698.828614154501 mfu:17.448005658203805 +Step 17461 | loss:0.7530913949012756 lr:5.3169819052498905e-05 tokens_per_second_per_gpu:4399.115066229482 mfu:20.751376333830724 +Step 17462 | loss:0.7302552461624146 lr:5.3165491541469904e-05 tokens_per_second_per_gpu:5082.049071542247 mfu:23.972892557447558 +Step 17463 | loss:0.8376516103744507 lr:5.316116400663283e-05 tokens_per_second_per_gpu:5326.88498209873 mfu:25.12782532085636 +Step 17464 | loss:0.7917302846908569 lr:5.315683644802025e-05 tokens_per_second_per_gpu:4291.884819213723 mfu:20.245552963290688 +Step 17465 | loss:0.7928084135055542 lr:5.315250886566468e-05 tokens_per_second_per_gpu:5373.069957885705 mfu:25.345687731612948 +Step 17466 | loss:0.8320370316505432 lr:5.314818125959871e-05 tokens_per_second_per_gpu:5330.960823607161 mfu:25.147051760661896 +Step 17467 | loss:0.8579021096229553 lr:5.314385362985485e-05 tokens_per_second_per_gpu:5359.868503688099 mfu:25.28341421976249 +Step 17468 | loss:0.7753216028213501 lr:5.313952597646568e-05 tokens_per_second_per_gpu:5235.238086899089 mfu:24.695511279626935 +Step 17469 | loss:0.7901424169540405 lr:5.3135198299463715e-05 tokens_per_second_per_gpu:4562.502791942697 mfu:21.52210411284053 +Step 17470 | loss:0.8130425810813904 lr:5.313087059888154e-05 tokens_per_second_per_gpu:4661.617512606768 mfu:21.989645160928077 +Step 17471 | loss:0.9405713677406311 lr:5.312654287475168e-05 tokens_per_second_per_gpu:4860.210358350704 mfu:22.926441497735656 +Step 17472 | loss:0.7465370297431946 lr:5.31222151271067e-05 tokens_per_second_per_gpu:4427.24403793111 mfu:20.884065492644766 +Step 17473 | loss:0.7273388504981995 lr:5.311788735597915e-05 tokens_per_second_per_gpu:4396.1051223447885 mfu:20.73717791497724 +Step 17474 | loss:0.8717746734619141 lr:5.311355956140155e-05 tokens_per_second_per_gpu:3830.979464307058 mfu:18.071383765633453 +Step 17475 | loss:0.7875270247459412 lr:5.3109231743406484e-05 tokens_per_second_per_gpu:5375.043850635021 mfu:25.354998920492303 +Step 17476 | loss:0.7375732660293579 lr:5.310490390202648e-05 tokens_per_second_per_gpu:4583.270833571435 mfu:21.620070508596573 +Step 17477 | loss:0.8482476472854614 lr:5.31005760372941e-05 tokens_per_second_per_gpu:4757.908913674616 mfu:22.44386812877224 +Step 17478 | loss:0.7327874302864075 lr:5.30962481492419e-05 tokens_per_second_per_gpu:4964.857223427116 mfu:23.420078614897385 +Step 17479 | loss:0.8270547389984131 lr:5.3091920237902424e-05 tokens_per_second_per_gpu:5184.710413798838 mfu:24.457163624703067 +Step 17480 | loss:0.7030933499336243 lr:5.308759230330822e-05 tokens_per_second_per_gpu:3800.4373662672538 mfu:17.927311478160426 +Step 17481 | loss:0.8564938306808472 lr:5.3083264345491826e-05 tokens_per_second_per_gpu:3334.8235153635496 mfu:15.730931501532899 +Step 17482 | loss:0.8605435490608215 lr:5.307893636448583e-05 tokens_per_second_per_gpu:4498.439403672161 mfu:21.219906180027273 +Step 17483 | loss:0.7326669692993164 lr:5.3074608360322744e-05 tokens_per_second_per_gpu:4400.365675787326 mfu:20.757275672491065 +Step 17484 | loss:0.8382071256637573 lr:5.307028033303514e-05 tokens_per_second_per_gpu:5367.339049705793 mfu:25.318654059934534 +Step 17485 | loss:0.7278720140457153 lr:5.306595228265555e-05 tokens_per_second_per_gpu:4299.17689494181 mfu:20.27995092865649 +Step 17486 | loss:0.7557442784309387 lr:5.3061624209216575e-05 tokens_per_second_per_gpu:4474.93832272729 mfu:21.10904757151244 +Step 17487 | loss:0.6692901849746704 lr:5.305729611275071e-05 tokens_per_second_per_gpu:3412.7937045009276 mfu:16.098730186779946 +Step 17488 | loss:0.7539358139038086 lr:5.305296799329054e-05 tokens_per_second_per_gpu:5309.072044064033 mfu:25.043798652946077 +Step 17489 | loss:0.8561587333679199 lr:5.3048639850868596e-05 tokens_per_second_per_gpu:4293.530515755025 mfu:20.25331599466079 +Step 17490 | loss:0.7010868787765503 lr:5.304431168551746e-05 tokens_per_second_per_gpu:3519.6674406354027 mfu:16.602871834665937 +Step 17491 | loss:0.9206227660179138 lr:5.3039983497269666e-05 tokens_per_second_per_gpu:4591.0847445493755 mfu:21.656930059870206 +Step 17492 | loss:0.6904361844062805 lr:5.303565528615775e-05 tokens_per_second_per_gpu:4908.007745627095 mfu:23.151909928593586 +Step 17493 | loss:0.6433038115501404 lr:5.3031327052214305e-05 tokens_per_second_per_gpu:5301.069305995428 mfu:25.006048372068555 +Step 17494 | loss:0.7615076899528503 lr:5.3026998795471836e-05 tokens_per_second_per_gpu:3666.2447399871876 mfu:17.294301964374387 +Step 17495 | loss:0.7740278840065002 lr:5.3022670515962945e-05 tokens_per_second_per_gpu:5361.624896391931 mfu:25.291699423817963 +Step 17496 | loss:0.8518649339675903 lr:5.3018342213720163e-05 tokens_per_second_per_gpu:5340.745343773907 mfu:25.193207011699727 +Step 17497 | loss:0.7822707295417786 lr:5.301401388877604e-05 tokens_per_second_per_gpu:4445.104428389516 mfu:20.968316001733903 +Step 17498 | loss:0.705983579158783 lr:5.300968554116314e-05 tokens_per_second_per_gpu:4420.140419845424 mfu:20.850556514132357 +Step 17499 | loss:0.7684937715530396 lr:5.3005357170913984e-05 tokens_per_second_per_gpu:4131.33560164048 mfu:19.488214911476554 +Step 17500 | loss:0.8607975244522095 lr:5.300102877806119e-05 tokens_per_second_per_gpu:5345.850499494372 mfu:25.217288902262545 +Step 17501 | loss:0.7183898091316223 lr:5.2996700362637255e-05 tokens_per_second_per_gpu:5011.444006223928 mfu:23.639836417874715 +Step 17502 | loss:0.7552584409713745 lr:5.299237192467476e-05 tokens_per_second_per_gpu:4280.552355626121 mfu:20.192095799030305 +Step 17503 | loss:0.667759358882904 lr:5.298804346420624e-05 tokens_per_second_per_gpu:4583.50006177994 mfu:21.621151817166457 +Step 17504 | loss:0.8506579995155334 lr:5.298371498126429e-05 tokens_per_second_per_gpu:3808.333282291084 mfu:17.964557861227025 +Step 17505 | loss:0.8642958402633667 lr:5.297938647588141e-05 tokens_per_second_per_gpu:3863.8465134948615 mfu:18.22642324434916 +Step 17506 | loss:0.7916945219039917 lr:5.29750579480902e-05 tokens_per_second_per_gpu:4647.07416767292 mfu:21.921041721524603 +Step 17507 | loss:0.7118942737579346 lr:5.29707293979232e-05 tokens_per_second_per_gpu:5350.524848261524 mfu:25.239338602923336 +Step 17508 | loss:0.7931062579154968 lr:5.296640082541297e-05 tokens_per_second_per_gpu:5302.893372629733 mfu:25.014652805600427 +Step 17509 | loss:0.9276885986328125 lr:5.296207223059205e-05 tokens_per_second_per_gpu:5308.167030551216 mfu:25.03952954979489 +Step 17510 | loss:0.7852838635444641 lr:5.2957743613492994e-05 tokens_per_second_per_gpu:4750.525581138144 mfu:22.409039689471275 +Step 17511 | loss:0.9446325898170471 lr:5.295341497414841e-05 tokens_per_second_per_gpu:4149.158450379837 mfu:19.572288329871665 +Step 17512 | loss:0.7677556872367859 lr:5.294908631259078e-05 tokens_per_second_per_gpu:5312.59522207295 mfu:25.060418084730284 +Step 17513 | loss:0.7861679792404175 lr:5.2944757628852714e-05 tokens_per_second_per_gpu:5311.69789714263 mfu:25.056185249934547 +Step 17514 | loss:0.8063856363296509 lr:5.294042892296674e-05 tokens_per_second_per_gpu:4853.004550256063 mfu:22.8924504715156 +Step 17515 | loss:0.6297574639320374 lr:5.293610019496543e-05 tokens_per_second_per_gpu:3927.2759193983306 mfu:18.525630574168417 +Step 17516 | loss:0.7911839485168457 lr:5.293177144488134e-05 tokens_per_second_per_gpu:5214.871821322692 mfu:24.599440130059854 +Step 17517 | loss:0.7293195128440857 lr:5.2927442672747007e-05 tokens_per_second_per_gpu:3967.8844312078686 mfu:18.71718785290057 +Step 17518 | loss:0.893882155418396 lr:5.292311387859503e-05 tokens_per_second_per_gpu:4634.700016080369 mfu:21.86267073721457 +Step 17519 | loss:0.849320113658905 lr:5.291878506245791e-05 tokens_per_second_per_gpu:4652.516519915026 mfu:21.94671422561166 +Step 17520 | loss:0.6485414505004883 lr:5.2914456224368256e-05 tokens_per_second_per_gpu:5135.103973284761 mfu:24.223161581066663 +Step 17521 | loss:0.7729057669639587 lr:5.291012736435859e-05 tokens_per_second_per_gpu:5350.51523244494 mfu:25.23929324347588 +Step 17522 | loss:0.7147933840751648 lr:5.290579848246151e-05 tokens_per_second_per_gpu:3674.9648249482048 mfu:17.335436092935417 +Step 17523 | loss:0.7646550536155701 lr:5.2901469578709515e-05 tokens_per_second_per_gpu:5329.271515450532 mfu:25.139083005861387 +Step 17524 | loss:0.8289715051651001 lr:5.289714065313522e-05 tokens_per_second_per_gpu:3692.852334931789 mfu:17.41981452945189 +Step 17525 | loss:0.7728666663169861 lr:5.289281170577116e-05 tokens_per_second_per_gpu:3961.990497950134 mfu:18.689385164115127 +Step 17526 | loss:0.8434579968452454 lr:5.288848273664989e-05 tokens_per_second_per_gpu:5001.418728318323 mfu:23.592545471505186 +Step 17527 | loss:0.7582229971885681 lr:5.2884153745803974e-05 tokens_per_second_per_gpu:4667.220911945824 mfu:22.016077351648914 +Step 17528 | loss:0.6433061957359314 lr:5.287982473326596e-05 tokens_per_second_per_gpu:4845.678991427948 mfu:22.85789456065416 +Step 17529 | loss:0.7103477716445923 lr:5.2875495699068446e-05 tokens_per_second_per_gpu:4428.664677065366 mfu:20.890766889827077 +Step 17530 | loss:0.7339074611663818 lr:5.287116664324393e-05 tokens_per_second_per_gpu:5332.857446241487 mfu:25.155998453225205 +Step 17531 | loss:0.6813198328018188 lr:5.286683756582501e-05 tokens_per_second_per_gpu:5248.056075150873 mfu:24.755975917184912 +Step 17532 | loss:0.7999683022499084 lr:5.286250846684424e-05 tokens_per_second_per_gpu:4836.9893854059765 mfu:22.816904206448854 +Step 17533 | loss:0.8536815047264099 lr:5.285817934633419e-05 tokens_per_second_per_gpu:3749.932593710308 mfu:17.68907184900656 +Step 17534 | loss:0.8652207255363464 lr:5.28538502043274e-05 tokens_per_second_per_gpu:4081.0453135467756 mfu:19.25098704213062 +Step 17535 | loss:0.7299469709396362 lr:5.284952104085643e-05 tokens_per_second_per_gpu:4754.532044273118 mfu:22.42793885965205 +Step 17536 | loss:0.840140700340271 lr:5.284519185595386e-05 tokens_per_second_per_gpu:4335.37111007755 mfu:20.450685216821658 +Step 17537 | loss:0.8086817860603333 lr:5.2840862649652234e-05 tokens_per_second_per_gpu:5353.619360241595 mfu:25.253935943911518 +Step 17538 | loss:0.915208101272583 lr:5.283653342198412e-05 tokens_per_second_per_gpu:5318.60209538704 mfu:25.088753531031124 +Step 17539 | loss:0.7531024813652039 lr:5.283220417298207e-05 tokens_per_second_per_gpu:5042.160082020228 mfu:23.784729388108797 +Step 17540 | loss:0.8190168142318726 lr:5.282787490267867e-05 tokens_per_second_per_gpu:5048.314007383824 mfu:23.813758504016818 +Step 17541 | loss:0.922446072101593 lr:5.282354561110645e-05 tokens_per_second_per_gpu:4124.126799373103 mfu:19.454209761232843 +Step 17542 | loss:0.7244868874549866 lr:5.281921629829797e-05 tokens_per_second_per_gpu:5348.754698376844 mfu:25.230988503898587 +Step 17543 | loss:0.7406653761863708 lr:5.2814886964285824e-05 tokens_per_second_per_gpu:3543.0285720210795 mfu:16.713070277232195 +Step 17544 | loss:0.7218022346496582 lr:5.281055760910254e-05 tokens_per_second_per_gpu:5118.248328636836 mfu:24.14365062939287 +Step 17545 | loss:0.7455565333366394 lr:5.2806228232780706e-05 tokens_per_second_per_gpu:5016.167592055329 mfu:23.66211837816828 +Step 17546 | loss:0.7190183401107788 lr:5.280189883535286e-05 tokens_per_second_per_gpu:4300.62747101678 mfu:20.286793543495918 +Step 17547 | loss:0.7749314904212952 lr:5.279756941685159e-05 tokens_per_second_per_gpu:5364.562935616802 mfu:25.305558656122862 +Step 17548 | loss:0.7234069108963013 lr:5.279323997730943e-05 tokens_per_second_per_gpu:3719.0216137560383 mfu:17.543259482605254 +Step 17549 | loss:0.7835855484008789 lr:5.278891051675896e-05 tokens_per_second_per_gpu:3991.408823833969 mfu:18.828156426593143 +Step 17550 | loss:0.8437600135803223 lr:5.278458103523274e-05 tokens_per_second_per_gpu:4251.398315525255 mfu:20.05457121768236 +Step 17551 | loss:0.7053949236869812 lr:5.278025153276333e-05 tokens_per_second_per_gpu:4456.245298839868 mfu:21.02086938847672 +Step 17552 | loss:0.7603325843811035 lr:5.277592200938329e-05 tokens_per_second_per_gpu:5319.175833919516 mfu:25.0914599535787 +Step 17553 | loss:0.8843299746513367 lr:5.277159246512518e-05 tokens_per_second_per_gpu:4866.044031748791 mfu:22.95395993048175 +Step 17554 | loss:0.8370044231414795 lr:5.2767262900021575e-05 tokens_per_second_per_gpu:4625.959538646122 mfu:21.821440413878005 +Step 17555 | loss:0.7867480516433716 lr:5.2762933314105044e-05 tokens_per_second_per_gpu:3382.895701533485 mfu:15.957696205657204 +Step 17556 | loss:0.6793454885482788 lr:5.2758603707408126e-05 tokens_per_second_per_gpu:5208.037812544889 mfu:24.567202944652824 +Step 17557 | loss:0.7648656368255615 lr:5.275427407996339e-05 tokens_per_second_per_gpu:3858.549111302301 mfu:18.201434494376837 +Step 17558 | loss:0.7185202836990356 lr:5.274994443180343e-05 tokens_per_second_per_gpu:4039.1899745695055 mfu:19.053548266917627 +Step 17559 | loss:0.8607614636421204 lr:5.2745614762960784e-05 tokens_per_second_per_gpu:5348.747840517129 mfu:25.23095615420424 +Step 17560 | loss:0.7427317500114441 lr:5.274128507346801e-05 tokens_per_second_per_gpu:5151.7135740592 mfu:24.301511901809803 +Step 17561 | loss:0.8160492777824402 lr:5.273695536335768e-05 tokens_per_second_per_gpu:3941.7080347632536 mfu:18.593709324717555 +Step 17562 | loss:0.8111034035682678 lr:5.273262563266236e-05 tokens_per_second_per_gpu:4482.593018224107 mfu:21.145156076196653 +Step 17563 | loss:0.7591829895973206 lr:5.2728295881414625e-05 tokens_per_second_per_gpu:4410.875544516343 mfu:20.806852516454352 +Step 17564 | loss:0.7346118092536926 lr:5.2723966109647005e-05 tokens_per_second_per_gpu:5329.095993677348 mfu:25.138255039710177 +Step 17565 | loss:0.6800668239593506 lr:5.271963631739211e-05 tokens_per_second_per_gpu:5359.648922023342 mfu:25.282378415585338 +Step 17566 | loss:0.7396860718727112 lr:5.2715306504682474e-05 tokens_per_second_per_gpu:4244.239269720452 mfu:20.020800777161934 +Step 17567 | loss:0.9504616856575012 lr:5.271097667155067e-05 tokens_per_second_per_gpu:5318.536608813911 mfu:25.08844461969251 +Step 17568 | loss:0.8305246233940125 lr:5.270664681802927e-05 tokens_per_second_per_gpu:5146.86807733057 mfu:24.27865486701386 +Step 17569 | loss:0.7420565485954285 lr:5.2702316944150834e-05 tokens_per_second_per_gpu:5320.91669270721 mfu:25.099671881501386 +Step 17570 | loss:0.8306534886360168 lr:5.269798704994793e-05 tokens_per_second_per_gpu:4361.426998485097 mfu:20.573595288033037 +Step 17571 | loss:0.6983309984207153 lr:5.26936571354531e-05 tokens_per_second_per_gpu:4354.554224506938 mfu:20.541175240561905 +Step 17572 | loss:0.9754977226257324 lr:5.268932720069896e-05 tokens_per_second_per_gpu:3605.4791943981395 mfu:17.00766051815969 +Step 17573 | loss:0.6780617237091064 lr:5.2684997245718024e-05 tokens_per_second_per_gpu:5028.645239955892 mfu:23.720977572221493 +Step 17574 | loss:0.7504504919052124 lr:5.268066727054291e-05 tokens_per_second_per_gpu:4437.496328131458 mfu:20.932427294740886 +Step 17575 | loss:0.7468556761741638 lr:5.267633727520613e-05 tokens_per_second_per_gpu:4417.5481623916685 mfu:20.838328393438264 +Step 17576 | loss:0.7820003032684326 lr:5.267200725974029e-05 tokens_per_second_per_gpu:5341.206217716686 mfu:25.19538103272102 +Step 17577 | loss:0.6539474725723267 lr:5.266767722417792e-05 tokens_per_second_per_gpu:5257.614724356342 mfu:24.801065696360837 +Step 17578 | loss:0.7620888352394104 lr:5.2663347168551625e-05 tokens_per_second_per_gpu:5201.035325935038 mfu:24.534171020566742 +Step 17579 | loss:0.7851417660713196 lr:5.265901709289396e-05 tokens_per_second_per_gpu:4338.081635664983 mfu:20.463471228480934 +Step 17580 | loss:0.7258083820343018 lr:5.265468699723748e-05 tokens_per_second_per_gpu:3852.8488178201746 mfu:18.17454523745129 +Step 17581 | loss:0.7877875566482544 lr:5.265035688161477e-05 tokens_per_second_per_gpu:4268.170437136331 mfu:20.13368817694083 +Step 17582 | loss:0.8935863971710205 lr:5.2646026746058375e-05 tokens_per_second_per_gpu:5355.952624621799 mfu:25.264942350088894 +Step 17583 | loss:0.7924297451972961 lr:5.2641696590600875e-05 tokens_per_second_per_gpu:4357.44121202668 mfu:20.55479365326339 +Step 17584 | loss:0.669593334197998 lr:5.263736641527485e-05 tokens_per_second_per_gpu:5313.145947707422 mfu:25.063015951511574 +Step 17585 | loss:0.7633194327354431 lr:5.2633036220112856e-05 tokens_per_second_per_gpu:5357.040644188774 mfu:25.270074724021597 +Step 17586 | loss:0.9787657856941223 lr:5.2628706005147456e-05 tokens_per_second_per_gpu:4747.960471824157 mfu:22.39693963118442 +Step 17587 | loss:0.665874183177948 lr:5.2624375770411216e-05 tokens_per_second_per_gpu:4379.467332852276 mfu:20.658694623241455 +Step 17588 | loss:0.7727208733558655 lr:5.2620045515936724e-05 tokens_per_second_per_gpu:3868.358246951279 mfu:18.247705860844185 +Step 17589 | loss:0.6888476610183716 lr:5.261571524175651e-05 tokens_per_second_per_gpu:4696.941177518822 mfu:22.15627291515694 +Step 17590 | loss:0.7707298398017883 lr:5.26113849479032e-05 tokens_per_second_per_gpu:5337.36232424596 mfu:25.177248731383745 +Step 17591 | loss:0.7953062653541565 lr:5.2607054634409315e-05 tokens_per_second_per_gpu:3640.364361291666 mfu:17.17221980241968 +Step 17592 | loss:0.7923002243041992 lr:5.260272430130744e-05 tokens_per_second_per_gpu:4502.30152866644 mfu:21.238124482571763 +Step 17593 | loss:0.8228430151939392 lr:5.259839394863014e-05 tokens_per_second_per_gpu:3387.09232396347 mfu:15.977492389677062 +Step 17594 | loss:0.7252374291419983 lr:5.2594063576410004e-05 tokens_per_second_per_gpu:4630.18863065395 mfu:21.841389762436428 +Step 17595 | loss:0.8307985663414001 lr:5.258973318467957e-05 tokens_per_second_per_gpu:4936.323144349818 mfu:23.28547849547386 +Step 17596 | loss:0.6862922310829163 lr:5.2585402773471414e-05 tokens_per_second_per_gpu:4023.0378494828897 mfu:18.977355937047346 +Step 17597 | loss:0.895348072052002 lr:5.258107234281813e-05 tokens_per_second_per_gpu:4184.646969801427 mfu:19.73969373094909 +Step 17598 | loss:0.7892612814903259 lr:5.257674189275227e-05 tokens_per_second_per_gpu:3818.82009521299 mfu:18.01402594701444 +Step 17599 | loss:0.7856103777885437 lr:5.25724114233064e-05 tokens_per_second_per_gpu:4937.81043531489 mfu:23.29249430071411 +Step 17600 | loss:0.6425883769989014 lr:5.256808093451309e-05 tokens_per_second_per_gpu:5363.319360496392 mfu:25.299692500104094 +Step 17601 | loss:0.7994850873947144 lr:5.256375042640492e-05 tokens_per_second_per_gpu:5272.347355820688 mfu:24.87056203262428 +Step 17602 | loss:0.732635498046875 lr:5.255941989901446e-05 tokens_per_second_per_gpu:5329.099272345545 mfu:25.138270505747457 +Step 17603 | loss:0.8757553100585938 lr:5.2555089352374276e-05 tokens_per_second_per_gpu:3348.8191196077705 mfu:15.796951154649205 +Step 17604 | loss:0.7517192959785461 lr:5.255075878651694e-05 tokens_per_second_per_gpu:4476.125428302724 mfu:21.114647351052856 +Step 17605 | loss:0.7070631384849548 lr:5.254642820147503e-05 tokens_per_second_per_gpu:5248.668595362323 mfu:24.758865279529047 +Step 17606 | loss:0.7610672116279602 lr:5.254209759728109e-05 tokens_per_second_per_gpu:3884.2882371716923 mfu:18.322850342650472 +Step 17607 | loss:0.8113579154014587 lr:5.2537766973967716e-05 tokens_per_second_per_gpu:4462.493223535041 mfu:21.050341915269726 +Step 17608 | loss:0.8611325621604919 lr:5.2533436331567475e-05 tokens_per_second_per_gpu:5156.054717591079 mfu:24.321989816525136 +Step 17609 | loss:0.8273619413375854 lr:5.252910567011293e-05 tokens_per_second_per_gpu:4024.841874570893 mfu:18.985865831185876 +Step 17610 | loss:0.7699043154716492 lr:5.2524774989636674e-05 tokens_per_second_per_gpu:3862.520641870435 mfu:18.22016888167971 +Step 17611 | loss:0.7497702836990356 lr:5.2520444290171255e-05 tokens_per_second_per_gpu:4246.412146961316 mfu:20.03105060983817 +Step 17612 | loss:0.7833796143531799 lr:5.251611357174924e-05 tokens_per_second_per_gpu:3949.334162791715 mfu:18.629683071779414 +Step 17613 | loss:0.7123172283172607 lr:5.251178283440323e-05 tokens_per_second_per_gpu:4445.406533053287 mfu:20.969741080977283 +Step 17614 | loss:0.8207195401191711 lr:5.250745207816577e-05 tokens_per_second_per_gpu:3650.1354852059285 mfu:17.218311861048903 +Step 17615 | loss:0.7544106841087341 lr:5.250312130306946e-05 tokens_per_second_per_gpu:5300.751037227391 mfu:25.00454704398747 +Step 17616 | loss:0.6416904330253601 lr:5.249879050914684e-05 tokens_per_second_per_gpu:5301.743887988737 mfu:25.00923049042655 +Step 17617 | loss:0.8200963735580444 lr:5.24944596964305e-05 tokens_per_second_per_gpu:3892.186627622868 mfu:18.36010839801304 +Step 17618 | loss:0.6871954202651978 lr:5.2490128864953004e-05 tokens_per_second_per_gpu:5393.411909419276 mfu:25.44164418769162 +Step 17619 | loss:0.7771354913711548 lr:5.2485798014746945e-05 tokens_per_second_per_gpu:4674.947565227694 mfu:22.05252529348249 +Step 17620 | loss:0.9364287853240967 lr:5.2481467145844876e-05 tokens_per_second_per_gpu:4930.738438646374 mfu:23.259134485821665 +Step 17621 | loss:0.6640598773956299 lr:5.247713625827938e-05 tokens_per_second_per_gpu:5058.168801361842 mfu:23.86024524861236 +Step 17622 | loss:0.697860836982727 lr:5.247280535208302e-05 tokens_per_second_per_gpu:5351.005959856201 mfu:25.241608088401527 +Step 17623 | loss:0.666778028011322 lr:5.246847442728838e-05 tokens_per_second_per_gpu:4049.0967167976573 mfu:19.100280060272336 +Step 17624 | loss:0.6959325075149536 lr:5.246414348392803e-05 tokens_per_second_per_gpu:3976.552113122678 mfu:18.758074787351752 +Step 17625 | loss:0.7219405770301819 lr:5.2459812522034525e-05 tokens_per_second_per_gpu:5341.377815574785 mfu:25.196190489095283 +Step 17626 | loss:0.8142727613449097 lr:5.245548154164047e-05 tokens_per_second_per_gpu:4585.17657008324 mfu:21.629060192874935 +Step 17627 | loss:0.8171111941337585 lr:5.245115054277843e-05 tokens_per_second_per_gpu:4143.254576800906 mfu:19.544438750895143 +Step 17628 | loss:0.841772735118866 lr:5.244681952548097e-05 tokens_per_second_per_gpu:5347.506946893721 mfu:25.22510264726434 +Step 17629 | loss:0.7055320143699646 lr:5.244248848978067e-05 tokens_per_second_per_gpu:5297.669626332516 mfu:24.99001150305343 +Step 17630 | loss:0.668315052986145 lr:5.243815743571009e-05 tokens_per_second_per_gpu:5295.3585751300625 mfu:24.979109880225465 +Step 17631 | loss:0.7463269233703613 lr:5.243382636330184e-05 tokens_per_second_per_gpu:4568.928044881304 mfu:21.552413127213136 +Step 17632 | loss:0.8009445667266846 lr:5.2429495272588446e-05 tokens_per_second_per_gpu:4582.073078380674 mfu:21.614420493003642 +Step 17633 | loss:0.8844621181488037 lr:5.2425164163602526e-05 tokens_per_second_per_gpu:5360.953818403013 mfu:25.28853383444653 +Step 17634 | loss:0.7581542730331421 lr:5.242083303637664e-05 tokens_per_second_per_gpu:5314.697608632177 mfu:25.070335400834296 +Step 17635 | loss:0.7177457809448242 lr:5.241650189094335e-05 tokens_per_second_per_gpu:4199.9115134020785 mfu:19.811699187513057 +Step 17636 | loss:0.8582379221916199 lr:5.241217072733523e-05 tokens_per_second_per_gpu:3069.149094057675 mfu:14.477699927502663 +Step 17637 | loss:0.6842249631881714 lr:5.240783954558488e-05 tokens_per_second_per_gpu:3727.4293966317705 mfu:17.582920428945762 +Step 17638 | loss:0.7287753224372864 lr:5.240350834572486e-05 tokens_per_second_per_gpu:3622.6908126180256 mfu:17.088850657907773 +Step 17639 | loss:0.6916234493255615 lr:5.239917712778775e-05 tokens_per_second_per_gpu:5171.414841176563 mfu:24.394446140185188 +Step 17640 | loss:0.7812402248382568 lr:5.239484589180611e-05 tokens_per_second_per_gpu:4663.156651017992 mfu:21.996905539417913 +Step 17641 | loss:0.7028681635856628 lr:5.239051463781255e-05 tokens_per_second_per_gpu:5195.052972064079 mfu:24.505951236662607 +Step 17642 | loss:0.8668704628944397 lr:5.2386183365839605e-05 tokens_per_second_per_gpu:3573.0053868023097 mfu:16.854476027127255 +Step 17643 | loss:0.7970171570777893 lr:5.238185207591987e-05 tokens_per_second_per_gpu:5088.341051520524 mfu:24.002572900527575 +Step 17644 | loss:0.7927767038345337 lr:5.237752076808593e-05 tokens_per_second_per_gpu:5343.419140208333 mfu:25.205819765677624 +Step 17645 | loss:0.9089323282241821 lr:5.2373189442370355e-05 tokens_per_second_per_gpu:4925.913747154719 mfu:23.236375592878282 +Step 17646 | loss:0.978502631187439 lr:5.236885809880572e-05 tokens_per_second_per_gpu:3758.6240008441187 mfu:17.73007069936371 +Step 17647 | loss:0.5778838396072388 lr:5.23645267374246e-05 tokens_per_second_per_gpu:4776.768831282874 mfu:22.532833577964155 +Step 17648 | loss:0.7371009588241577 lr:5.236019535825956e-05 tokens_per_second_per_gpu:4977.180269341507 mfu:23.478208525004 +Step 17649 | loss:0.7649307250976562 lr:5.235586396134321e-05 tokens_per_second_per_gpu:4351.668700593247 mfu:20.52756373193988 +Step 17650 | loss:0.8536887168884277 lr:5.235153254670808e-05 tokens_per_second_per_gpu:3669.1869835648235 mfu:17.308181029328615 +Step 17651 | loss:0.8843078017234802 lr:5.23472011143868e-05 tokens_per_second_per_gpu:3390.7944392373665 mfu:15.994955899070032 +Step 17652 | loss:0.7330405116081238 lr:5.234286966441191e-05 tokens_per_second_per_gpu:4362.190663511213 mfu:20.57719762625591 +Step 17653 | loss:0.6937679052352905 lr:5.2338538196816e-05 tokens_per_second_per_gpu:4230.601098657321 mfu:19.956467197345216 +Step 17654 | loss:0.8223230838775635 lr:5.233420671163164e-05 tokens_per_second_per_gpu:4597.8253679859035 mfu:21.688726730689257 +Step 17655 | loss:0.7444595694541931 lr:5.232987520889141e-05 tokens_per_second_per_gpu:5283.63412589307 mfu:24.92380365277726 +Step 17656 | loss:0.7778079509735107 lr:5.232554368862791e-05 tokens_per_second_per_gpu:4712.23470168474 mfu:22.228415077991862 +Step 17657 | loss:0.7848658561706543 lr:5.232121215087369e-05 tokens_per_second_per_gpu:4908.265260642828 mfu:23.153124670859395 +Step 17658 | loss:0.8495388627052307 lr:5.231688059566133e-05 tokens_per_second_per_gpu:5361.768065142896 mfu:25.292374775243378 +Step 17659 | loss:0.6704836487770081 lr:5.231254902302342e-05 tokens_per_second_per_gpu:3757.3558313426643 mfu:17.724088527453322 +Step 17660 | loss:0.8118471503257751 lr:5.2308217432992544e-05 tokens_per_second_per_gpu:5345.187909980894 mfu:25.21416335447804 +Step 17661 | loss:0.7779458165168762 lr:5.230388582560125e-05 tokens_per_second_per_gpu:4429.471833760005 mfu:20.894574385670822 +Step 17662 | loss:0.7314028143882751 lr:5.229955420088215e-05 tokens_per_second_per_gpu:4357.187942408067 mfu:20.553598937260137 +Step 17663 | loss:0.789077639579773 lr:5.2295222558867806e-05 tokens_per_second_per_gpu:2776.2601193123255 mfu:13.096092661616806 +Step 17664 | loss:0.7250066995620728 lr:5.22908908995908e-05 tokens_per_second_per_gpu:5351.418612645367 mfu:25.24355464201385 +Step 17665 | loss:0.6684554219245911 lr:5.228655922308372e-05 tokens_per_second_per_gpu:5351.6974134817 mfu:25.24486979312716 +Step 17666 | loss:0.7350038290023804 lr:5.2282227529379125e-05 tokens_per_second_per_gpu:4123.9513736729505 mfu:19.45338224827451 +Step 17667 | loss:0.7894701957702637 lr:5.227789581850962e-05 tokens_per_second_per_gpu:5363.070040397953 mfu:25.298516414661364 +Step 17668 | loss:0.6463608145713806 lr:5.227356409050774e-05 tokens_per_second_per_gpu:5354.012299052081 mfu:25.255789503322855 +Step 17669 | loss:0.8558001518249512 lr:5.226923234540613e-05 tokens_per_second_per_gpu:4851.519923017708 mfu:22.885447231528747 +Step 17670 | loss:0.9068524837493896 lr:5.226490058323731e-05 tokens_per_second_per_gpu:5325.456391490754 mfu:25.121086415215846 +Step 17671 | loss:0.8248438239097595 lr:5.2260568804033894e-05 tokens_per_second_per_gpu:4484.060394271287 mfu:21.15207793937137 +Step 17672 | loss:0.8767716884613037 lr:5.225623700782843e-05 tokens_per_second_per_gpu:5329.999042709109 mfu:25.142514876069043 +Step 17673 | loss:0.7650402188301086 lr:5.225190519465355e-05 tokens_per_second_per_gpu:5327.0621245749935 mfu:25.12866093214052 +Step 17674 | loss:0.7009662389755249 lr:5.224757336454179e-05 tokens_per_second_per_gpu:4945.664238588557 mfu:23.32954203075301 +Step 17675 | loss:0.7563613653182983 lr:5.2243241517525754e-05 tokens_per_second_per_gpu:4288.805765500465 mfu:20.231028541584514 +Step 17676 | loss:0.7418201565742493 lr:5.2238909653638e-05 tokens_per_second_per_gpu:4835.6721451904305 mfu:22.810690559607764 +Step 17677 | loss:0.8247268199920654 lr:5.223457777291112e-05 tokens_per_second_per_gpu:4201.38168212772 mfu:19.818634224228553 +Step 17678 | loss:0.7196615934371948 lr:5.2230245875377715e-05 tokens_per_second_per_gpu:5349.040150229086 mfu:25.23233502898872 +Step 17679 | loss:0.6946898698806763 lr:5.2225913961070314e-05 tokens_per_second_per_gpu:3634.967234116708 mfu:17.14676063269055 +Step 17680 | loss:0.7732256650924683 lr:5.222158203002156e-05 tokens_per_second_per_gpu:5327.449730561674 mfu:25.13048933571209 +Step 17681 | loss:0.7760886549949646 lr:5.221725008226399e-05 tokens_per_second_per_gpu:3767.46102500506 mfu:17.771756450082666 +Step 17682 | loss:0.8042086362838745 lr:5.22129181178302e-05 tokens_per_second_per_gpu:4412.473726273225 mfu:20.81439141247949 +Step 17683 | loss:0.7911823987960815 lr:5.220858613675278e-05 tokens_per_second_per_gpu:3369.857270690402 mfu:15.896191702784687 +Step 17684 | loss:0.716999888420105 lr:5.220425413906428e-05 tokens_per_second_per_gpu:5334.198268564054 mfu:25.16232334088863 +Step 17685 | loss:0.8447962999343872 lr:5.219992212479733e-05 tokens_per_second_per_gpu:4702.600416365132 mfu:22.182968510360695 +Step 17686 | loss:0.7183192372322083 lr:5.2195590093984456e-05 tokens_per_second_per_gpu:5362.682316415259 mfu:25.296687454482463 +Step 17687 | loss:0.7097111344337463 lr:5.2191258046658296e-05 tokens_per_second_per_gpu:4621.287069974454 mfu:21.79939958194839 +Step 17688 | loss:0.7793800830841064 lr:5.2186925982851396e-05 tokens_per_second_per_gpu:4560.793527697704 mfu:21.514041221766348 +Step 17689 | loss:0.8010298609733582 lr:5.218259390259634e-05 tokens_per_second_per_gpu:5324.731875775458 mfu:25.117668750971667 +Step 17690 | loss:0.9071931838989258 lr:5.217826180592571e-05 tokens_per_second_per_gpu:5011.933750428013 mfu:23.64214662484436 +Step 17691 | loss:0.7447575926780701 lr:5.2173929692872115e-05 tokens_per_second_per_gpu:4923.768671717573 mfu:23.226256906053653 +Step 17692 | loss:0.6906406283378601 lr:5.216959756346811e-05 tokens_per_second_per_gpu:5343.691875030376 mfu:25.20710630236656 +Step 17693 | loss:0.806919515132904 lr:5.216526541774628e-05 tokens_per_second_per_gpu:5338.670280633889 mfu:25.183418584825983 +Step 17694 | loss:0.818282425403595 lr:5.216093325573922e-05 tokens_per_second_per_gpu:5202.7797906988735 mfu:24.542399958494137 +Step 17695 | loss:0.7744349241256714 lr:5.2156601077479483e-05 tokens_per_second_per_gpu:4916.666625221867 mfu:23.19275534099933 +Step 17696 | loss:0.8823662996292114 lr:5.215226888299971e-05 tokens_per_second_per_gpu:5344.336400083879 mfu:25.21014663701125 +Step 17697 | loss:0.8534101247787476 lr:5.214793667233242e-05 tokens_per_second_per_gpu:4712.198777922375 mfu:22.228245619475558 +Step 17698 | loss:0.7360239624977112 lr:5.2143604445510244e-05 tokens_per_second_per_gpu:3600.992837353431 mfu:16.986497606528356 +Step 17699 | loss:0.8378074169158936 lr:5.213927220256573e-05 tokens_per_second_per_gpu:4063.009491089571 mfu:19.16590899037136 +Step 17700 | loss:0.8026179075241089 lr:5.213493994353148e-05 tokens_per_second_per_gpu:5355.347539329629 mfu:25.26208805952753 +Step 17701 | loss:0.6136758327484131 lr:5.213060766844008e-05 tokens_per_second_per_gpu:4372.81336385276 mfu:20.62730671618656 +Step 17702 | loss:0.7554799914360046 lr:5.2126275377324095e-05 tokens_per_second_per_gpu:4055.697598507402 mfu:19.131417545523753 +Step 17703 | loss:0.799580454826355 lr:5.212194307021615e-05 tokens_per_second_per_gpu:5371.488460383946 mfu:25.33822753806585 +Step 17704 | loss:0.6402723789215088 lr:5.2117610747148774e-05 tokens_per_second_per_gpu:3545.279499554544 mfu:16.723688286455385 +Step 17705 | loss:0.7921733260154724 lr:5.2113278408154584e-05 tokens_per_second_per_gpu:4922.165014619105 mfu:23.218692181908064 +Step 17706 | loss:0.7663230299949646 lr:5.210894605326615e-05 tokens_per_second_per_gpu:5225.862075900676 mfu:24.651283036034993 +Step 17707 | loss:0.8400508761405945 lr:5.210461368251608e-05 tokens_per_second_per_gpu:4822.876383294752 mfu:22.75033076756375 +Step 17708 | loss:0.6710917353630066 lr:5.210028129593693e-05 tokens_per_second_per_gpu:4412.139421451468 mfu:20.812814439596973 +Step 17709 | loss:0.7028440833091736 lr:5.2095948893561295e-05 tokens_per_second_per_gpu:4452.983232294671 mfu:21.005481664017083 +Step 17710 | loss:0.6969852447509766 lr:5.209161647542177e-05 tokens_per_second_per_gpu:5115.659135087613 mfu:24.131436961659215 +Step 17711 | loss:0.7136840224266052 lr:5.208728404155093e-05 tokens_per_second_per_gpu:5324.874350689488 mfu:25.118340829449874 +Step 17712 | loss:0.7284828424453735 lr:5.2082951591981354e-05 tokens_per_second_per_gpu:5356.218713167527 mfu:25.26619753515848 +Step 17713 | loss:0.8309298753738403 lr:5.207861912674563e-05 tokens_per_second_per_gpu:4843.429478876446 mfu:22.84728322614231 +Step 17714 | loss:0.7022322416305542 lr:5.207428664587637e-05 tokens_per_second_per_gpu:4742.031739402167 mfu:22.368972788803834 +Step 17715 | loss:0.7658891081809998 lr:5.206995414940611e-05 tokens_per_second_per_gpu:5307.646527262452 mfu:25.037074246974772 +Step 17716 | loss:0.7412484884262085 lr:5.206562163736747e-05 tokens_per_second_per_gpu:5223.506660531853 mfu:24.640172139864955 +Step 17717 | loss:0.7785840630531311 lr:5.206128910979302e-05 tokens_per_second_per_gpu:5324.372178541148 mfu:25.11597199774675 +Step 17718 | loss:0.7380721569061279 lr:5.205695656671536e-05 tokens_per_second_per_gpu:3586.892348222674 mfu:16.919983193507054 +Step 17719 | loss:0.9259027242660522 lr:5.205262400816706e-05 tokens_per_second_per_gpu:4730.603907280454 mfu:22.3150657548965 +Step 17720 | loss:0.7282965183258057 lr:5.2048291434180716e-05 tokens_per_second_per_gpu:3117.1646185247087 mfu:14.704197348706895 +Step 17721 | loss:0.743438720703125 lr:5.2043958844788923e-05 tokens_per_second_per_gpu:5376.433365422445 mfu:25.361553498820246 +Step 17722 | loss:0.7342553734779358 lr:5.203962624002423e-05 tokens_per_second_per_gpu:4698.803111443674 mfu:22.165055975159152 +Step 17723 | loss:0.792348325252533 lr:5.203529361991927e-05 tokens_per_second_per_gpu:3627.7119361021923 mfu:17.112536154074593 +Step 17724 | loss:0.9335139393806458 lr:5.2030960984506585e-05 tokens_per_second_per_gpu:3390.4611748537804 mfu:15.993383834111889 +Step 17725 | loss:0.6386645436286926 lr:5.2026628333818806e-05 tokens_per_second_per_gpu:4290.6674718064105 mfu:20.23981052786951 +Step 17726 | loss:0.8610529899597168 lr:5.202229566788848e-05 tokens_per_second_per_gpu:4815.226276797586 mfu:22.714243909973856 +Step 17727 | loss:0.8305811285972595 lr:5.201796298674821e-05 tokens_per_second_per_gpu:5339.073344338035 mfu:25.185319905088193 +Step 17728 | loss:0.7142461538314819 lr:5.2013630290430604e-05 tokens_per_second_per_gpu:5327.678856821486 mfu:25.131570163372594 +Step 17729 | loss:0.7760188579559326 lr:5.200929757896821e-05 tokens_per_second_per_gpu:3787.9195437839503 mfu:17.868262773745936 +Step 17730 | loss:0.7802953720092773 lr:5.200496485239363e-05 tokens_per_second_per_gpu:4174.046550417549 mfu:19.68968974409729 +Step 17731 | loss:0.8103241920471191 lr:5.2000632110739453e-05 tokens_per_second_per_gpu:4256.968973499674 mfu:20.080848961799948 +Step 17732 | loss:0.7968800663948059 lr:5.199629935403829e-05 tokens_per_second_per_gpu:5326.622762857465 mfu:25.126588387956097 +Step 17733 | loss:0.7479497790336609 lr:5.199196658232268e-05 tokens_per_second_per_gpu:5325.883322925245 mfu:25.123100323634354 +Step 17734 | loss:0.8392689228057861 lr:5.1987633795625244e-05 tokens_per_second_per_gpu:4911.234288931057 mfu:23.16713008386343 +Step 17735 | loss:0.704433262348175 lr:5.198330099397857e-05 tokens_per_second_per_gpu:5333.404616564707 mfu:25.158579548996716 +Step 17736 | loss:0.7175678610801697 lr:5.197896817741522e-05 tokens_per_second_per_gpu:4667.2593591280975 mfu:22.016258713567467 +Step 17737 | loss:0.7042405009269714 lr:5.1974635345967806e-05 tokens_per_second_per_gpu:5360.903715847314 mfu:25.288297492142917 +Step 17738 | loss:0.6323257088661194 lr:5.19703024996689e-05 tokens_per_second_per_gpu:4930.281925286986 mfu:23.256981034416224 +Step 17739 | loss:0.8745375871658325 lr:5.1965969638551114e-05 tokens_per_second_per_gpu:3824.6223401023162 mfu:18.04139612612245 +Step 17740 | loss:0.7867039442062378 lr:5.196163676264699e-05 tokens_per_second_per_gpu:4710.313434246021 mfu:22.219352131681397 +Step 17741 | loss:0.9545195698738098 lr:5.1957303871989173e-05 tokens_per_second_per_gpu:4490.150741598245 mfu:21.180807146833036 +Step 17742 | loss:0.7364885807037354 lr:5.19529709666102e-05 tokens_per_second_per_gpu:4082.9374150066874 mfu:19.259912407542824 +Step 17743 | loss:0.6965502500534058 lr:5.194863804654272e-05 tokens_per_second_per_gpu:5213.175750089151 mfu:24.59143947267165 +Step 17744 | loss:0.8140714168548584 lr:5.194430511181925e-05 tokens_per_second_per_gpu:5262.830667852179 mfu:24.82567019176243 +Step 17745 | loss:0.6998862028121948 lr:5.193997216247243e-05 tokens_per_second_per_gpu:3870.9547047976043 mfu:18.25995379550669 +Step 17746 | loss:0.7167643308639526 lr:5.1935639198534824e-05 tokens_per_second_per_gpu:3210.2191402282506 mfu:15.143151404321063 +Step 17747 | loss:0.8299873471260071 lr:5.193130622003903e-05 tokens_per_second_per_gpu:4096.754279300827 mfu:19.325088913818416 +Step 17748 | loss:0.7113702893257141 lr:5.192697322701765e-05 tokens_per_second_per_gpu:5338.006732395728 mfu:25.180288514573572 +Step 17749 | loss:0.9447869658470154 lr:5.192264021950324e-05 tokens_per_second_per_gpu:5178.810762852702 mfu:24.429333964605895 +Step 17750 | loss:0.883462131023407 lr:5.1918307197528415e-05 tokens_per_second_per_gpu:4370.689510082434 mfu:20.617288135585113 +Step 17751 | loss:0.8045955300331116 lr:5.191397416112576e-05 tokens_per_second_per_gpu:5346.738988007926 mfu:25.221480054172748 +Step 17752 | loss:0.6437639594078064 lr:5.1909641110327845e-05 tokens_per_second_per_gpu:5361.320146545211 mfu:25.2902618667953 +Step 17753 | loss:0.6425839066505432 lr:5.190530804516729e-05 tokens_per_second_per_gpu:4333.946593652665 mfu:20.443965529797914 +Step 17754 | loss:0.9110843539237976 lr:5.1900974965676675e-05 tokens_per_second_per_gpu:4695.046539975431 mfu:22.147335586606264 +Step 17755 | loss:0.7230682969093323 lr:5.189664187188857e-05 tokens_per_second_per_gpu:4430.052132773698 mfu:20.89731175512763 +Step 17756 | loss:0.8608177900314331 lr:5.189230876383558e-05 tokens_per_second_per_gpu:4171.724488621773 mfu:19.678736182422018 +Step 17757 | loss:0.912280797958374 lr:5.188797564155031e-05 tokens_per_second_per_gpu:4607.695637902262 mfu:21.735286477926422 +Step 17758 | loss:0.7198030352592468 lr:5.1883642505065323e-05 tokens_per_second_per_gpu:4222.42133012783 mfu:19.917881833579184 +Step 17759 | loss:0.8503892421722412 lr:5.1879309354413216e-05 tokens_per_second_per_gpu:3837.3852468469854 mfu:18.10160093481349 +Step 17760 | loss:0.7061244249343872 lr:5.187497618962659e-05 tokens_per_second_per_gpu:5347.57528802344 mfu:25.225425024033424 +Step 17761 | loss:0.7921276092529297 lr:5.187064301073804e-05 tokens_per_second_per_gpu:4891.089535746183 mfu:23.072103846041067 +Step 17762 | loss:0.7517557144165039 lr:5.186630981778012e-05 tokens_per_second_per_gpu:3800.470171253728 mfu:17.92746622487873 +Step 17763 | loss:0.7272950410842896 lr:5.186197661078547e-05 tokens_per_second_per_gpu:5353.719059539575 mfu:25.254406242510544 +Step 17764 | loss:0.8367746472358704 lr:5.1857643389786647e-05 tokens_per_second_per_gpu:4660.656963854342 mfu:21.985114088576367 +Step 17765 | loss:0.7840612530708313 lr:5.1853310154816245e-05 tokens_per_second_per_gpu:4160.247241705381 mfu:19.62459605049715 +Step 17766 | loss:0.7928338646888733 lr:5.1848976905906865e-05 tokens_per_second_per_gpu:4843.031767471904 mfu:22.84540715359052 +Step 17767 | loss:0.8001213669776917 lr:5.1844643643091096e-05 tokens_per_second_per_gpu:3496.5433955497974 mfu:16.493791768628213 +Step 17768 | loss:0.7853370308876038 lr:5.1840310366401536e-05 tokens_per_second_per_gpu:4209.347153576095 mfu:19.85620871210011 +Step 17769 | loss:0.608473539352417 lr:5.183597707587075e-05 tokens_per_second_per_gpu:5091.627138978078 mfu:24.018073935729188 +Step 17770 | loss:0.8216657042503357 lr:5.183164377153135e-05 tokens_per_second_per_gpu:4547.292054295155 mfu:21.450352468137623 +Step 17771 | loss:0.7811123728752136 lr:5.182731045341592e-05 tokens_per_second_per_gpu:3802.5696348527 mfu:17.93736975288051 +Step 17772 | loss:0.6337423324584961 lr:5.1822977121557074e-05 tokens_per_second_per_gpu:4540.926144219128 mfu:21.420323384172747 +Step 17773 | loss:0.6435256600379944 lr:5.1818643775987364e-05 tokens_per_second_per_gpu:4353.906303338758 mfu:20.53811888586485 +Step 17774 | loss:0.7128534317016602 lr:5.181431041673941e-05 tokens_per_second_per_gpu:3810.794922485935 mfu:17.97616983802526 +Step 17775 | loss:0.6742551922798157 lr:5.1809977043845804e-05 tokens_per_second_per_gpu:4986.379410702703 mfu:23.52160244434013 +Step 17776 | loss:0.8259261250495911 lr:5.180564365733911e-05 tokens_per_second_per_gpu:5331.799427678906 mfu:25.151007599148926 +Step 17777 | loss:0.7559884190559387 lr:5.180131025725196e-05 tokens_per_second_per_gpu:4775.314287815093 mfu:22.525972248255922 +Step 17778 | loss:0.7545907497406006 lr:5.17969768436169e-05 tokens_per_second_per_gpu:5349.05574273436 mfu:25.232408581496454 +Step 17779 | loss:0.8584439158439636 lr:5.179264341646658e-05 tokens_per_second_per_gpu:4492.038355604525 mfu:21.189711344160816 +Step 17780 | loss:0.7507836222648621 lr:5.178830997583353e-05 tokens_per_second_per_gpu:3935.203647909293 mfu:18.563027022164587 +Step 17781 | loss:0.7890406250953674 lr:5.1783976521750386e-05 tokens_per_second_per_gpu:5343.814108154715 mfu:25.20768289686919 +Step 17782 | loss:0.7299870848655701 lr:5.1779643054249735e-05 tokens_per_second_per_gpu:5320.176718716982 mfu:25.096181297936898 +Step 17783 | loss:0.765950620174408 lr:5.177530957336415e-05 tokens_per_second_per_gpu:4986.327250161795 mfu:23.521356394169157 +Step 17784 | loss:0.6640161871910095 lr:5.1770976079126244e-05 tokens_per_second_per_gpu:4530.098721100184 mfu:21.36924858197181 +Step 17785 | loss:0.6698113679885864 lr:5.176664257156859e-05 tokens_per_second_per_gpu:4964.808130639029 mfu:23.41984703583988 +Step 17786 | loss:0.688120424747467 lr:5.176230905072381e-05 tokens_per_second_per_gpu:4962.634192701072 mfu:23.409592199674478 +Step 17787 | loss:0.8627976179122925 lr:5.1757975516624456e-05 tokens_per_second_per_gpu:4234.238380445046 mfu:19.97362487612218 +Step 17788 | loss:0.7608543634414673 lr:5.175364196930316e-05 tokens_per_second_per_gpu:4662.070026864838 mfu:21.991779747890284 +Step 17789 | loss:0.7440782785415649 lr:5.17493084087925e-05 tokens_per_second_per_gpu:3535.30012584644 mfu:16.676613878017534 +Step 17790 | loss:0.8094040751457214 lr:5.174497483512506e-05 tokens_per_second_per_gpu:4196.711267551344 mfu:19.796603081816894 +Step 17791 | loss:0.9154930710792542 lr:5.174064124833344e-05 tokens_per_second_per_gpu:5023.770144103842 mfu:23.69798091331025 +Step 17792 | loss:0.8369344472885132 lr:5.1736307648450234e-05 tokens_per_second_per_gpu:3751.083807862496 mfu:17.694502322579815 +Step 17793 | loss:0.6726859211921692 lr:5.173197403550805e-05 tokens_per_second_per_gpu:5032.322220328347 mfu:23.738322515995907 +Step 17794 | loss:0.6289909482002258 lr:5.1727640409539446e-05 tokens_per_second_per_gpu:5092.26879167469 mfu:24.021100720976353 +Step 17795 | loss:0.7063905000686646 lr:5.172330677057706e-05 tokens_per_second_per_gpu:5344.506706986805 mfu:25.210950003729703 +Step 17796 | loss:0.8264838457107544 lr:5.171897311865345e-05 tokens_per_second_per_gpu:4772.162230031637 mfu:22.511103453894847 +Step 17797 | loss:0.7517079710960388 lr:5.171463945380124e-05 tokens_per_second_per_gpu:4344.8771293621385 mfu:20.495526731681196 +Step 17798 | loss:0.6934095025062561 lr:5.1710305776052995e-05 tokens_per_second_per_gpu:4642.997553560091 mfu:21.901811637212294 +Step 17799 | loss:0.7026880979537964 lr:5.170597208544131e-05 tokens_per_second_per_gpu:5330.052182945309 mfu:25.14276554762986 +Step 17800 | loss:0.7828759551048279 lr:5.170163838199882e-05 tokens_per_second_per_gpu:5090.813340877228 mfu:24.014235111239213 +Step 17801 | loss:0.6318549513816833 lr:5.169730466575807e-05 tokens_per_second_per_gpu:4330.093307911128 mfu:20.425788923516745 +Step 17802 | loss:0.7619441151618958 lr:5.1692970936751686e-05 tokens_per_second_per_gpu:3592.6787266812366 mfu:16.94727852795406 +Step 17803 | loss:0.7596465349197388 lr:5.168863719501223e-05 tokens_per_second_per_gpu:4348.4079703810785 mfu:20.512182311191136 +Step 17804 | loss:0.796691358089447 lr:5.168430344057236e-05 tokens_per_second_per_gpu:3709.020152436842 mfu:17.496080883137342 +Step 17805 | loss:0.7254236936569214 lr:5.167996967346459e-05 tokens_per_second_per_gpu:5349.742737544529 mfu:25.23564925322942 +Step 17806 | loss:0.7468987107276917 lr:5.1675635893721565e-05 tokens_per_second_per_gpu:4145.74663266722 mfu:19.556194203605124 +Step 17807 | loss:0.7748886942863464 lr:5.167130210137587e-05 tokens_per_second_per_gpu:5327.905532383355 mfu:25.13263943067277 +Step 17808 | loss:0.788702666759491 lr:5.1666968296460106e-05 tokens_per_second_per_gpu:4704.117265275926 mfu:22.190123745474022 +Step 17809 | loss:1.0094108581542969 lr:5.166263447900686e-05 tokens_per_second_per_gpu:3290.1887372535066 mfu:15.520381637709397 +Step 17810 | loss:0.8106145858764648 lr:5.1658300649048706e-05 tokens_per_second_per_gpu:4678.536763991772 mfu:22.069456156433095 +Step 17811 | loss:0.7242670655250549 lr:5.165396680661829e-05 tokens_per_second_per_gpu:5309.860230021913 mfu:25.04751666058778 +Step 17812 | loss:0.7906209826469421 lr:5.164963295174816e-05 tokens_per_second_per_gpu:5321.035960196413 mfu:25.10023448659709 +Step 17813 | loss:0.8125271797180176 lr:5.164529908447093e-05 tokens_per_second_per_gpu:3657.467510092253 mfu:17.25289827341014 +Step 17814 | loss:0.5381726622581482 lr:5.16409652048192e-05 tokens_per_second_per_gpu:5372.644106323632 mfu:25.343678917136877 +Step 17815 | loss:0.8910372853279114 lr:5.1636631312825576e-05 tokens_per_second_per_gpu:4643.549557865596 mfu:21.904415535703013 +Step 17816 | loss:0.841059148311615 lr:5.1632297408522624e-05 tokens_per_second_per_gpu:4859.3363991699425 mfu:22.92231888316719 +Step 17817 | loss:0.9054616093635559 lr:5.162796349194295e-05 tokens_per_second_per_gpu:5332.167125188181 mfu:25.152742091035844 +Step 17818 | loss:0.7518270015716553 lr:5.162362956311917e-05 tokens_per_second_per_gpu:3733.8876400469258 mfu:17.613385065025 +Step 17819 | loss:0.6863046884536743 lr:5.1619295622083855e-05 tokens_per_second_per_gpu:5009.789462342638 mfu:23.632031652092376 +Step 17820 | loss:0.7014088034629822 lr:5.161496166886961e-05 tokens_per_second_per_gpu:3434.563747551414 mfu:16.201423194201446 +Step 17821 | loss:0.8046250939369202 lr:5.161062770350903e-05 tokens_per_second_per_gpu:3294.5472312405973 mfu:15.540941397482259 +Step 17822 | loss:0.729130744934082 lr:5.1606293726034726e-05 tokens_per_second_per_gpu:4505.831896126565 mfu:21.25477782822397 +Step 17823 | loss:0.7161694765090942 lr:5.1601959736479264e-05 tokens_per_second_per_gpu:5348.048378455168 mfu:25.227656672317437 +Step 17824 | loss:0.7618908882141113 lr:5.159762573487528e-05 tokens_per_second_per_gpu:4061.4297855240548 mfu:19.158457249693257 +Step 17825 | loss:0.6739820837974548 lr:5.159329172125533e-05 tokens_per_second_per_gpu:4048.874014705111 mfu:19.0992295365049 +Step 17826 | loss:0.808035671710968 lr:5.158895769565205e-05 tokens_per_second_per_gpu:4813.962631605825 mfu:22.708283080004218 +Step 17827 | loss:0.7515148520469666 lr:5.1584623658097997e-05 tokens_per_second_per_gpu:5086.110979111134 mfu:23.992053268482895 +Step 17828 | loss:0.8968381285667419 lr:5.158028960862579e-05 tokens_per_second_per_gpu:5351.756708738941 mfu:25.24514949897238 +Step 17829 | loss:0.8424133062362671 lr:5.157595554726803e-05 tokens_per_second_per_gpu:4369.006759524918 mfu:20.609350314098464 +Step 17830 | loss:0.937852680683136 lr:5.157162147405731e-05 tokens_per_second_per_gpu:3484.7278436318356 mfu:16.43805579429015 +Step 17831 | loss:0.7953131198883057 lr:5.156728738902622e-05 tokens_per_second_per_gpu:5318.634719483877 mfu:25.088907424462146 +Step 17832 | loss:0.7858436703681946 lr:5.156295329220736e-05 tokens_per_second_per_gpu:4318.763992968052 mfu:20.3723466119498 +Step 17833 | loss:0.767530083656311 lr:5.1558619183633324e-05 tokens_per_second_per_gpu:5357.7939882069595 mfu:25.27362837628912 +Step 17834 | loss:0.7996047139167786 lr:5.155428506333673e-05 tokens_per_second_per_gpu:5087.792447132935 mfu:23.99998504002976 +Step 17835 | loss:0.6406780481338501 lr:5.154995093135013e-05 tokens_per_second_per_gpu:4057.023931192315 mfu:19.13767408309418 +Step 17836 | loss:0.9001193046569824 lr:5.1545616787706184e-05 tokens_per_second_per_gpu:3977.84344205599 mfu:18.764166206253623 +Step 17837 | loss:0.7082371711730957 lr:5.154128263243744e-05 tokens_per_second_per_gpu:5327.727726748849 mfu:25.131800691158034 +Step 17838 | loss:0.8161816000938416 lr:5.153694846557652e-05 tokens_per_second_per_gpu:5387.452131401117 mfu:25.413530897936983 +Step 17839 | loss:0.7800438404083252 lr:5.1532614287156e-05 tokens_per_second_per_gpu:5279.8135670053825 mfu:24.905781424649472 +Step 17840 | loss:0.7551377415657043 lr:5.1528280097208514e-05 tokens_per_second_per_gpu:5336.076111070153 mfu:25.17118144438287 +Step 17841 | loss:0.8181790113449097 lr:5.152394589576662e-05 tokens_per_second_per_gpu:3570.0350184798385 mfu:16.840464292953065 +Step 17842 | loss:0.9053351283073425 lr:5.151961168286294e-05 tokens_per_second_per_gpu:5048.353373878918 mfu:23.81394420249079 +Step 17843 | loss:0.8621945977210999 lr:5.151527745853007e-05 tokens_per_second_per_gpu:4972.5349517355235 mfu:23.456295769283667 +Step 17844 | loss:0.6478090286254883 lr:5.151094322280061e-05 tokens_per_second_per_gpu:4894.931941957427 mfu:23.090229131721607 +Step 17845 | loss:0.8257381916046143 lr:5.1506608975707146e-05 tokens_per_second_per_gpu:3675.8748684242937 mfu:17.33972892328156 +Step 17846 | loss:0.7695769667625427 lr:5.1502274717282284e-05 tokens_per_second_per_gpu:4099.964443195965 mfu:19.34023180462222 +Step 17847 | loss:0.7724010348320007 lr:5.1497940447558626e-05 tokens_per_second_per_gpu:4909.2895826675185 mfu:23.157956572617216 +Step 17848 | loss:0.943981409072876 lr:5.149360616656875e-05 tokens_per_second_per_gpu:5349.818147720723 mfu:25.23600497589683 +Step 17849 | loss:0.8679561614990234 lr:5.148927187434529e-05 tokens_per_second_per_gpu:4884.770420420698 mfu:23.042295500898508 +Step 17850 | loss:0.7890347242355347 lr:5.1484937570920834e-05 tokens_per_second_per_gpu:4791.969806174266 mfu:22.604539170080603 +Step 17851 | loss:0.8008502125740051 lr:5.148060325632795e-05 tokens_per_second_per_gpu:4235.006328679945 mfu:19.97724741897168 +Step 17852 | loss:0.7004814743995667 lr:5.147626893059928e-05 tokens_per_second_per_gpu:5233.699061271921 mfu:24.688251433158737 +Step 17853 | loss:0.6524276733398438 lr:5.1471934593767376e-05 tokens_per_second_per_gpu:5148.342673634783 mfu:24.285610789372907 +Step 17854 | loss:0.8980363607406616 lr:5.146760024586489e-05 tokens_per_second_per_gpu:3856.2036856220375 mfu:18.19037072127215 +Step 17855 | loss:0.8596352934837341 lr:5.146326588692438e-05 tokens_per_second_per_gpu:4360.886127285041 mfu:20.571043906300226 +Step 17856 | loss:0.8197079300880432 lr:5.145893151697847e-05 tokens_per_second_per_gpu:4337.680229561111 mfu:20.461577727402027 +Step 17857 | loss:0.7955544590950012 lr:5.145459713605975e-05 tokens_per_second_per_gpu:4593.110864708556 mfu:21.666487614352803 +Step 17858 | loss:0.8438709378242493 lr:5.145026274420082e-05 tokens_per_second_per_gpu:3808.076385862555 mfu:17.963346036942603 +Step 17859 | loss:0.7398248314857483 lr:5.144592834143427e-05 tokens_per_second_per_gpu:4670.942478924506 mfu:22.033632618052156 +Step 17860 | loss:0.8431059122085571 lr:5.144159392779271e-05 tokens_per_second_per_gpu:5338.419101405858 mfu:25.18223372955176 +Step 17861 | loss:0.7518807649612427 lr:5.143725950330874e-05 tokens_per_second_per_gpu:5337.696824530392 mfu:25.17882662629705 +Step 17862 | loss:0.7230652570724487 lr:5.143292506801497e-05 tokens_per_second_per_gpu:5281.156217156322 mfu:24.912114934491193 +Step 17863 | loss:0.8273709416389465 lr:5.1428590621943974e-05 tokens_per_second_per_gpu:5331.0927837291465 mfu:25.147674239072025 +Step 17864 | loss:0.8569331765174866 lr:5.1424256165128356e-05 tokens_per_second_per_gpu:3986.1232369700247 mfu:18.803223411541808 +Step 17865 | loss:0.7097402215003967 lr:5.1419921697600746e-05 tokens_per_second_per_gpu:5358.407351296017 mfu:25.276521714630736 +Step 17866 | loss:0.7160555720329285 lr:5.141558721939371e-05 tokens_per_second_per_gpu:4432.281755290877 mfu:20.907829264954 +Step 17867 | loss:0.8067329525947571 lr:5.141125273053987e-05 tokens_per_second_per_gpu:5283.567354797074 mfu:24.92348868212512 +Step 17868 | loss:0.7572729587554932 lr:5.140691823107182e-05 tokens_per_second_per_gpu:4374.814053982812 mfu:20.63674431288295 +Step 17869 | loss:0.7589641809463501 lr:5.1402583721022154e-05 tokens_per_second_per_gpu:5359.861065344437 mfu:25.283379131826326 +Step 17870 | loss:0.8869364261627197 lr:5.1398249200423485e-05 tokens_per_second_per_gpu:5250.346178640771 mfu:24.76677872607871 +Step 17871 | loss:0.7785006761550903 lr:5.139391466930839e-05 tokens_per_second_per_gpu:5310.57316155988 mfu:25.050879680291036 +Step 17872 | loss:0.7832639217376709 lr:5.1389580127709494e-05 tokens_per_second_per_gpu:4123.241372130939 mfu:19.450033049862093 +Step 17873 | loss:0.6335731744766235 lr:5.1385245575659394e-05 tokens_per_second_per_gpu:4211.660844369376 mfu:19.867122786328373 +Step 17874 | loss:0.8390592336654663 lr:5.138091101319068e-05 tokens_per_second_per_gpu:5352.156533842715 mfu:25.24703554220408 +Step 17875 | loss:0.8112075924873352 lr:5.1376576440335946e-05 tokens_per_second_per_gpu:4076.2051954822336 mfu:19.22815537941088 +Step 17876 | loss:0.8453428149223328 lr:5.137224185712781e-05 tokens_per_second_per_gpu:4446.757023969376 mfu:20.97611157704598 +Step 17877 | loss:0.8170157670974731 lr:5.136790726359888e-05 tokens_per_second_per_gpu:5322.775733966167 mfu:25.10844129630503 +Step 17878 | loss:0.7438700199127197 lr:5.1363572659781746e-05 tokens_per_second_per_gpu:5115.0687984721235 mfu:24.128652243122136 +Step 17879 | loss:0.8573620915412903 lr:5.135923804570899e-05 tokens_per_second_per_gpu:5086.622990350232 mfu:23.994468512855846 +Step 17880 | loss:0.7495648264884949 lr:5.135490342141325e-05 tokens_per_second_per_gpu:5319.257707643097 mfu:25.091846165901924 +Step 17881 | loss:0.7689769268035889 lr:5.13505687869271e-05 tokens_per_second_per_gpu:5319.230290135246 mfu:25.09171683283948 +Step 17882 | loss:0.6285079121589661 lr:5.1346234142283144e-05 tokens_per_second_per_gpu:3959.7971426732856 mfu:18.679038733049232 +Step 17883 | loss:0.7500833868980408 lr:5.134189948751399e-05 tokens_per_second_per_gpu:4681.2160833930775 mfu:22.082094963188084 +Step 17884 | loss:0.8689522743225098 lr:5.133756482265225e-05 tokens_per_second_per_gpu:4745.986076273953 mfu:22.38762607050749 +Step 17885 | loss:0.6523525714874268 lr:5.1333230147730515e-05 tokens_per_second_per_gpu:4530.165546335479 mfu:21.36956380800826 +Step 17886 | loss:0.7114440202713013 lr:5.132889546278138e-05 tokens_per_second_per_gpu:5385.387351378064 mfu:25.403790978280437 +Step 17887 | loss:0.7997474074363708 lr:5.132456076783745e-05 tokens_per_second_per_gpu:5328.619620724019 mfu:25.136007907211717 +Step 17888 | loss:0.7737336158752441 lr:5.132022606293133e-05 tokens_per_second_per_gpu:4749.625949588061 mfu:22.404795973956567 +Step 17889 | loss:0.9324893951416016 lr:5.131589134809561e-05 tokens_per_second_per_gpu:4675.134833485109 mfu:22.05340866980397 +Step 17890 | loss:0.7917755246162415 lr:5.1311556623362925e-05 tokens_per_second_per_gpu:4803.251666094647 mfu:22.65775762820883 +Step 17891 | loss:0.8257718086242676 lr:5.130722188876585e-05 tokens_per_second_per_gpu:3746.1325256169016 mfu:17.671146279451488 +Step 17892 | loss:0.7921722531318665 lr:5.130288714433699e-05 tokens_per_second_per_gpu:5368.955074768122 mfu:25.326277125875855 +Step 17893 | loss:0.7319266200065613 lr:5.129855239010893e-05 tokens_per_second_per_gpu:4396.932453877919 mfu:20.741080578976337 +Step 17894 | loss:0.6585416197776794 lr:5.1294217626114314e-05 tokens_per_second_per_gpu:5331.04270350535 mfu:25.147438002111805 +Step 17895 | loss:0.6365272998809814 lr:5.128988285238572e-05 tokens_per_second_per_gpu:5344.832585940037 mfu:25.21248723035278 +Step 17896 | loss:0.7728542685508728 lr:5.128554806895576e-05 tokens_per_second_per_gpu:3575.525413992625 mfu:16.86636342534495 +Step 17897 | loss:0.7604579329490662 lr:5.128121327585702e-05 tokens_per_second_per_gpu:4600.948829843856 mfu:21.703460633190765 +Step 17898 | loss:0.7362873554229736 lr:5.12768784731221e-05 tokens_per_second_per_gpu:4590.8115789020485 mfu:21.655641490904898 +Step 17899 | loss:0.8403424620628357 lr:5.1272543660783635e-05 tokens_per_second_per_gpu:5315.596097446705 mfu:25.074573725870426 +Step 17900 | loss:0.69301438331604 lr:5.126820883887419e-05 tokens_per_second_per_gpu:4322.570877740509 mfu:20.390304336942997 +Step 17901 | loss:0.6272295117378235 lr:5.1263874007426396e-05 tokens_per_second_per_gpu:3853.8063138398957 mfu:18.179061909541982 +Step 17902 | loss:0.8464643359184265 lr:5.125953916647284e-05 tokens_per_second_per_gpu:4602.106019540015 mfu:21.70891929442447 +Step 17903 | loss:0.6774096488952637 lr:5.125520431604613e-05 tokens_per_second_per_gpu:5358.038811683389 mfu:25.274783250398627 +Step 17904 | loss:0.9277359247207642 lr:5.125086945617886e-05 tokens_per_second_per_gpu:3695.385377549456 mfu:17.431763323661063 +Step 17905 | loss:0.645893931388855 lr:5.124653458690365e-05 tokens_per_second_per_gpu:5337.866718243167 mfu:25.179628043927067 +Step 17906 | loss:0.7603892683982849 lr:5.1242199708253103e-05 tokens_per_second_per_gpu:3958.1030426131683 mfu:18.671047374022386 +Step 17907 | loss:0.7737016081809998 lr:5.123786482025978e-05 tokens_per_second_per_gpu:3529.4870283778855 mfu:16.649192505441658 +Step 17908 | loss:0.9029268622398376 lr:5.123352992295635e-05 tokens_per_second_per_gpu:5318.100557308674 mfu:25.086387690344246 +Step 17909 | loss:0.7800354957580566 lr:5.122919501637538e-05 tokens_per_second_per_gpu:3638.0675350180686 mfu:17.16138528100815 +Step 17910 | loss:0.8484736084938049 lr:5.1224860100549474e-05 tokens_per_second_per_gpu:5040.406916108612 mfu:23.776459405382756 +Step 17911 | loss:0.6546740531921387 lr:5.122052517551123e-05 tokens_per_second_per_gpu:5056.578094815148 mfu:23.85274161442908 +Step 17912 | loss:0.8835367560386658 lr:5.121619024129326e-05 tokens_per_second_per_gpu:4051.5194166984425 mfu:19.11170834905957 +Step 17913 | loss:0.6570788025856018 lr:5.121185529792818e-05 tokens_per_second_per_gpu:5375.779099638163 mfu:25.358467215486534 +Step 17914 | loss:0.757222592830658 lr:5.120752034544858e-05 tokens_per_second_per_gpu:4657.204511308265 mfu:21.968828280867378 +Step 17915 | loss:0.8084790706634521 lr:5.1203185383887065e-05 tokens_per_second_per_gpu:4917.8767672855465 mfu:23.19846378758523 +Step 17916 | loss:0.793901801109314 lr:5.119885041327624e-05 tokens_per_second_per_gpu:4236.001123504455 mfu:19.981940035888393 +Step 17917 | loss:0.7401642799377441 lr:5.11945154336487e-05 tokens_per_second_per_gpu:5317.331748835202 mfu:25.082761089602627 +Step 17918 | loss:0.8794779181480408 lr:5.119018044503706e-05 tokens_per_second_per_gpu:4568.688883047956 mfu:21.551284960040014 +Step 17919 | loss:0.839998185634613 lr:5.118584544747392e-05 tokens_per_second_per_gpu:3585.343562541978 mfu:16.912677307201125 +Step 17920 | loss:0.7600027918815613 lr:5.118151044099189e-05 tokens_per_second_per_gpu:4008.8126849172013 mfu:18.910253408727648 +Step 17921 | loss:0.8487779498100281 lr:5.1177175425623566e-05 tokens_per_second_per_gpu:5332.657599443038 mfu:25.155055741779538 +Step 17922 | loss:0.6554638147354126 lr:5.1172840401401557e-05 tokens_per_second_per_gpu:3689.332369655408 mfu:17.403210252675198 +Step 17923 | loss:0.6687232255935669 lr:5.116850536835846e-05 tokens_per_second_per_gpu:4787.272716315497 mfu:22.58238219581062 +Step 17924 | loss:0.7179681062698364 lr:5.1164170326526903e-05 tokens_per_second_per_gpu:4888.596147254212 mfu:23.060342107108497 +Step 17925 | loss:0.9535179138183594 lr:5.115983527593945e-05 tokens_per_second_per_gpu:5318.327370348098 mfu:25.087457606149506 +Step 17926 | loss:0.7443379163742065 lr:5.115550021662874e-05 tokens_per_second_per_gpu:5352.882270242075 mfu:25.25045896462326 +Step 17927 | loss:0.8941771984100342 lr:5.1151165148627364e-05 tokens_per_second_per_gpu:4974.0281493302455 mfu:23.46333943710375 +Step 17928 | loss:0.7699692845344543 lr:5.1146830071967936e-05 tokens_per_second_per_gpu:5346.573961155391 mfu:25.220701594352853 +Step 17929 | loss:0.8066248297691345 lr:5.114249498668303e-05 tokens_per_second_per_gpu:5322.541711880929 mfu:25.10733737420068 +Step 17930 | loss:0.7296482920646667 lr:5.113815989280528e-05 tokens_per_second_per_gpu:4875.73858737622 mfu:22.99969080343808 +Step 17931 | loss:0.9249931573867798 lr:5.1133824790367304e-05 tokens_per_second_per_gpu:4560.959823649125 mfu:21.514825668142212 +Step 17932 | loss:0.6468152403831482 lr:5.1129489679401667e-05 tokens_per_second_per_gpu:4563.310212213512 mfu:21.52591285201807 +Step 17933 | loss:0.7333019375801086 lr:5.1125154559941e-05 tokens_per_second_per_gpu:5355.196956919959 mfu:25.26137773660911 +Step 17934 | loss:0.769218921661377 lr:5.1120819432017905e-05 tokens_per_second_per_gpu:4854.260233812778 mfu:22.89837374509433 +Step 17935 | loss:0.7066009640693665 lr:5.111648429566499e-05 tokens_per_second_per_gpu:5337.784970322594 mfu:25.179242425038087 +Step 17936 | loss:0.6901815533638 lr:5.111214915091483e-05 tokens_per_second_per_gpu:5341.725246896305 mfu:25.19782938191679 +Step 17937 | loss:0.743971586227417 lr:5.110781399780008e-05 tokens_per_second_per_gpu:4656.16955909777 mfu:21.963946234709415 +Step 17938 | loss:0.7868853211402893 lr:5.1103478836353315e-05 tokens_per_second_per_gpu:5244.124131977579 mfu:24.737428270377805 +Step 17939 | loss:0.9306119084358215 lr:5.109914366660714e-05 tokens_per_second_per_gpu:4368.02955497502 mfu:20.604740673508303 +Step 17940 | loss:0.8119925260543823 lr:5.109480848859417e-05 tokens_per_second_per_gpu:3469.0960638390548 mfu:16.364318021950954 +Step 17941 | loss:0.7148842811584473 lr:5.109047330234699e-05 tokens_per_second_per_gpu:5327.000586311185 mfu:25.12837064565062 +Step 17942 | loss:0.8621081709861755 lr:5.108613810789824e-05 tokens_per_second_per_gpu:5208.777772042621 mfu:24.570693459853736 +Step 17943 | loss:0.725715160369873 lr:5.10818029052805e-05 tokens_per_second_per_gpu:4876.553158044369 mfu:23.00353327225983 +Step 17944 | loss:0.7778895497322083 lr:5.107746769452637e-05 tokens_per_second_per_gpu:5329.216075312257 mfu:25.138821485270082 +Step 17945 | loss:0.7983286380767822 lr:5.107313247566849e-05 tokens_per_second_per_gpu:4054.5762771107043 mfu:19.12612808117837 +Step 17946 | loss:0.8867776989936829 lr:5.106879724873943e-05 tokens_per_second_per_gpu:5393.486674004069 mfu:25.44199686499424 +Step 17947 | loss:0.8788013458251953 lr:5.106446201377181e-05 tokens_per_second_per_gpu:3803.916616779219 mfu:17.943723696446717 +Step 17948 | loss:0.6489388942718506 lr:5.1060126770798234e-05 tokens_per_second_per_gpu:4249.167626994779 mfu:20.044048679289393 +Step 17949 | loss:0.7272390723228455 lr:5.1055791519851315e-05 tokens_per_second_per_gpu:5312.887823018517 mfu:25.061798333313398 +Step 17950 | loss:0.8335919380187988 lr:5.1051456260963645e-05 tokens_per_second_per_gpu:5307.027667807754 mfu:25.03415498133865 +Step 17951 | loss:0.7497116327285767 lr:5.104712099416785e-05 tokens_per_second_per_gpu:4715.588794803046 mfu:22.24423690749821 +Step 17952 | loss:0.7774001955986023 lr:5.104278571949652e-05 tokens_per_second_per_gpu:4166.653014022066 mfu:19.654813171452403 +Step 17953 | loss:0.7956233024597168 lr:5.103845043698228e-05 tokens_per_second_per_gpu:3981.0044878965873 mfu:18.779077398813754 +Step 17954 | loss:0.7778893709182739 lr:5.103411514665769e-05 tokens_per_second_per_gpu:3676.3032073395184 mfu:17.341749471026827 +Step 17955 | loss:0.9370089769363403 lr:5.1029779848555404e-05 tokens_per_second_per_gpu:4610.476615258179 mfu:21.748404822597944 +Step 17956 | loss:0.7788524627685547 lr:5.102544454270801e-05 tokens_per_second_per_gpu:5337.105222098603 mfu:25.176035936688514 +Step 17957 | loss:0.7980931401252747 lr:5.1021109229148125e-05 tokens_per_second_per_gpu:4208.5513749181155 mfu:19.852454888396583 +Step 17958 | loss:0.6568171381950378 lr:5.1016773907908347e-05 tokens_per_second_per_gpu:5383.911717258241 mfu:25.396830160368093 +Step 17959 | loss:0.8207475543022156 lr:5.101243857902127e-05 tokens_per_second_per_gpu:4593.0414764577945 mfu:21.666160298135107 +Step 17960 | loss:0.8109942078590393 lr:5.100810324251953e-05 tokens_per_second_per_gpu:3310.182441072628 mfu:15.614695349901824 +Step 17961 | loss:0.8324485421180725 lr:5.1003767898435695e-05 tokens_per_second_per_gpu:5123.164927200005 mfu:24.166843063673554 +Step 17962 | loss:0.8248864412307739 lr:5.09994325468024e-05 tokens_per_second_per_gpu:4323.993319570992 mfu:20.397014237750987 +Step 17963 | loss:0.7903692126274109 lr:5.099509718765226e-05 tokens_per_second_per_gpu:5204.595480028522 mfu:24.550964874850628 +Step 17964 | loss:0.6775153875350952 lr:5.099076182101786e-05 tokens_per_second_per_gpu:3196.528058720014 mfu:15.078568236907333 +Step 17965 | loss:0.7706984877586365 lr:5.098642644693181e-05 tokens_per_second_per_gpu:4478.269471979477 mfu:21.124761170887677 +Step 17966 | loss:0.7716169953346252 lr:5.098209106542672e-05 tokens_per_second_per_gpu:4343.027023276118 mfu:20.486799465612787 +Step 17967 | loss:0.7545256018638611 lr:5.097775567653521e-05 tokens_per_second_per_gpu:5359.72407940383 mfu:25.28273294577227 +Step 17968 | loss:0.7735074162483215 lr:5.097342028028986e-05 tokens_per_second_per_gpu:3718.1018054808615 mfu:17.53892059003577 +Step 17969 | loss:0.8597059845924377 lr:5.0969084876723305e-05 tokens_per_second_per_gpu:4345.021037908265 mfu:20.496205572846605 +Step 17970 | loss:0.9348869323730469 lr:5.0964749465868125e-05 tokens_per_second_per_gpu:5131.02489713972 mfu:24.203919882928346 +Step 17971 | loss:0.7668123841285706 lr:5.096041404775696e-05 tokens_per_second_per_gpu:5344.879540244935 mfu:25.212708721820217 +Step 17972 | loss:0.9431297183036804 lr:5.095607862242238e-05 tokens_per_second_per_gpu:4375.44106250147 mfu:20.639702019044147 +Step 17973 | loss:0.7456812858581543 lr:5.095174318989701e-05 tokens_per_second_per_gpu:5002.549450658729 mfu:23.597879281706007 +Step 17974 | loss:0.7344508171081543 lr:5.0947407750213475e-05 tokens_per_second_per_gpu:4609.822011814005 mfu:21.745316946464914 +Step 17975 | loss:0.7823976278305054 lr:5.094307230340435e-05 tokens_per_second_per_gpu:4025.8516600260145 mfu:18.99062916146485 +Step 17976 | loss:0.7693015933036804 lr:5.0938736849502264e-05 tokens_per_second_per_gpu:5107.452937787907 mfu:24.092726928874864 +Step 17977 | loss:0.7456240653991699 lr:5.093440138853981e-05 tokens_per_second_per_gpu:5162.500444541411 mfu:24.352395410304272 +Step 17978 | loss:0.7089104652404785 lr:5.093006592054962e-05 tokens_per_second_per_gpu:4278.6508401763185 mfu:20.183126026221593 +Step 17979 | loss:0.728753387928009 lr:5.092573044556427e-05 tokens_per_second_per_gpu:3697.425534707095 mfu:17.44138709306035 +Step 17980 | loss:0.975663959980011 lr:5.092139496361639e-05 tokens_per_second_per_gpu:4431.047844763576 mfu:20.902008698469874 +Step 17981 | loss:0.7902874946594238 lr:5.0917059474738584e-05 tokens_per_second_per_gpu:3358.1573636381036 mfu:15.841001245009009 +Step 17982 | loss:0.8072254657745361 lr:5.091272397896346e-05 tokens_per_second_per_gpu:5372.039057954548 mfu:25.340824800748017 +Step 17983 | loss:0.716322660446167 lr:5.09083884763236e-05 tokens_per_second_per_gpu:4441.045218711264 mfu:20.949168017109002 +Step 17984 | loss:0.6805897951126099 lr:5.0904052966851654e-05 tokens_per_second_per_gpu:5351.15279245118 mfu:25.24230072280427 +Step 17985 | loss:0.7240571975708008 lr:5.089971745058021e-05 tokens_per_second_per_gpu:4179.922474307753 mfu:19.717407479623372 +Step 17986 | loss:0.8358482718467712 lr:5.089538192754187e-05 tokens_per_second_per_gpu:4135.7257461730405 mfu:19.508923972272537 +Step 17987 | loss:0.7515913248062134 lr:5.089104639776926e-05 tokens_per_second_per_gpu:5338.5532945460645 mfu:25.18286674148983 +Step 17988 | loss:0.7936587929725647 lr:5.088671086129496e-05 tokens_per_second_per_gpu:3814.835950598219 mfu:17.995232057101397 +Step 17989 | loss:0.797558605670929 lr:5.088237531815161e-05 tokens_per_second_per_gpu:3748.0577544965886 mfu:17.680227912554887 +Step 17990 | loss:0.85468590259552 lr:5.0878039768371786e-05 tokens_per_second_per_gpu:4519.161328906537 mfu:21.317655036882844 +Step 17991 | loss:0.7510267496109009 lr:5.0873704211988126e-05 tokens_per_second_per_gpu:5318.3382472720095 mfu:25.087508914455388 +Step 17992 | loss:0.9478153586387634 lr:5.086936864903322e-05 tokens_per_second_per_gpu:3724.968238336467 mfu:17.57131072534991 +Step 17993 | loss:0.7717829346656799 lr:5.086503307953968e-05 tokens_per_second_per_gpu:4800.418598720712 mfu:22.644393566034843 +Step 17994 | loss:0.6675220727920532 lr:5.0860697503540126e-05 tokens_per_second_per_gpu:4619.165158396137 mfu:21.789390162998014 +Step 17995 | loss:0.8338585495948792 lr:5.085636192106714e-05 tokens_per_second_per_gpu:5028.6484410396415 mfu:23.7209926722797 +Step 17996 | loss:0.6473025679588318 lr:5.085202633215337e-05 tokens_per_second_per_gpu:4852.010615842054 mfu:22.88776191330202 +Step 17997 | loss:0.7509353160858154 lr:5.0847690736831376e-05 tokens_per_second_per_gpu:4424.7843333885785 mfu:20.872462646648774 +Step 17998 | loss:0.8630909323692322 lr:5.08433551351338e-05 tokens_per_second_per_gpu:5032.141147090576 mfu:23.73746836263906 +Step 17999 | loss:0.7640405893325806 lr:5.083901952709323e-05 tokens_per_second_per_gpu:5371.577514197045 mfu:25.33864762009639 +Step 18000 | loss:0.8625101447105408 lr:5.083468391274232e-05 tokens_per_second_per_gpu:4799.165293991405 mfu:22.638481513790865 +Step 18001 | loss:0.782392144203186 lr:5.0830348292113615e-05 tokens_per_second_per_gpu:3759.354520356971 mfu:17.733516684545318 +Step 18002 | loss:0.7096044421195984 lr:5.082601266523977e-05 tokens_per_second_per_gpu:5364.689951841817 mfu:25.306157813327463 +Step 18003 | loss:0.7395393252372742 lr:5.082167703215338e-05 tokens_per_second_per_gpu:4847.245016510724 mfu:22.865281768161047 +Step 18004 | loss:0.7855772972106934 lr:5.0817341392887055e-05 tokens_per_second_per_gpu:5352.757095257876 mfu:25.24986849286574 +Step 18005 | loss:0.7605414986610413 lr:5.081300574747338e-05 tokens_per_second_per_gpu:5176.006346478376 mfu:24.416105054085428 +Step 18006 | loss:0.8903145790100098 lr:5.0808670095945e-05 tokens_per_second_per_gpu:5345.004095346104 mfu:25.21329626948372 +Step 18007 | loss:0.7643003463745117 lr:5.080433443833452e-05 tokens_per_second_per_gpu:5320.8193700715965 mfu:25.099212794024243 +Step 18008 | loss:0.7035707235336304 lr:5.079999877467452e-05 tokens_per_second_per_gpu:4429.044590199933 mfu:20.892559004902694 +Step 18009 | loss:0.6004515290260315 lr:5.0795663104997626e-05 tokens_per_second_per_gpu:5316.689570062261 mfu:25.07973182276367 +Step 18010 | loss:0.7415883541107178 lr:5.079132742933647e-05 tokens_per_second_per_gpu:5345.26519109449 mfu:25.214527902676384 +Step 18011 | loss:0.6995137929916382 lr:5.078699174772362e-05 tokens_per_second_per_gpu:4782.582718703787 mfu:22.560258676880082 +Step 18012 | loss:0.8246159553527832 lr:5.078265606019171e-05 tokens_per_second_per_gpu:4986.2117650432865 mfu:23.520811631161358 +Step 18013 | loss:0.7626349329948425 lr:5.077832036677333e-05 tokens_per_second_per_gpu:3923.8282110982427 mfu:18.50936714587721 +Step 18014 | loss:0.6725568771362305 lr:5.077398466750113e-05 tokens_per_second_per_gpu:5050.042522772167 mfu:23.821912206018766 +Step 18015 | loss:0.668517529964447 lr:5.0769648962407665e-05 tokens_per_second_per_gpu:3503.051001714095 mfu:16.524489257217308 +Step 18016 | loss:0.7693713903427124 lr:5.0765313251525596e-05 tokens_per_second_per_gpu:4700.72759551736 mfu:22.17413409488557 +Step 18017 | loss:0.8923115134239197 lr:5.076097753488749e-05 tokens_per_second_per_gpu:4338.413745372559 mfu:20.46503784663533 +Step 18018 | loss:0.7414999604225159 lr:5.0756641812526e-05 tokens_per_second_per_gpu:4377.347987304865 mfu:20.64869730869668 +Step 18019 | loss:0.7455089688301086 lr:5.0752306084473676e-05 tokens_per_second_per_gpu:4542.77386805026 mfu:21.429039412738387 +Step 18020 | loss:0.6445392966270447 lr:5.074797035076319e-05 tokens_per_second_per_gpu:4919.308129657033 mfu:23.20521576810713 +Step 18021 | loss:0.7937171459197998 lr:5.074363461142711e-05 tokens_per_second_per_gpu:4160.439726698189 mfu:19.625504035049545 +Step 18022 | loss:0.7523027658462524 lr:5.073929886649806e-05 tokens_per_second_per_gpu:4443.360899862831 mfu:20.960091480196887 +Step 18023 | loss:0.7097658514976501 lr:5.073496311600865e-05 tokens_per_second_per_gpu:3346.0753424504856 mfu:15.784008289661635 +Step 18024 | loss:0.8459793925285339 lr:5.073062735999149e-05 tokens_per_second_per_gpu:4394.411912227105 mfu:20.72919075396069 +Step 18025 | loss:0.7971843481063843 lr:5.072629159847919e-05 tokens_per_second_per_gpu:5375.407207437819 mfu:25.356712936526165 +Step 18026 | loss:0.7598509192466736 lr:5.072195583150435e-05 tokens_per_second_per_gpu:5327.759249934271 mfu:25.131949391402078 +Step 18027 | loss:0.7041520476341248 lr:5.07176200590996e-05 tokens_per_second_per_gpu:5204.163855427191 mfu:24.548928827963405 +Step 18028 | loss:0.8045383095741272 lr:5.071328428129753e-05 tokens_per_second_per_gpu:4048.104007189855 mfu:19.095597279678607 +Step 18029 | loss:0.8850913047790527 lr:5.070894849813076e-05 tokens_per_second_per_gpu:3987.2089248477496 mfu:18.808344786498488 +Step 18030 | loss:0.8625057339668274 lr:5.0704612709631906e-05 tokens_per_second_per_gpu:4625.4730902934525 mfu:21.819145754001813 +Step 18031 | loss:0.6154479384422302 lr:5.070027691583356e-05 tokens_per_second_per_gpu:4483.24103996482 mfu:21.148212905311503 +Step 18032 | loss:0.6363124251365662 lr:5.069594111676835e-05 tokens_per_second_per_gpu:5143.322967928193 mfu:24.261931981104816 +Step 18033 | loss:0.6673701405525208 lr:5.0691605312468856e-05 tokens_per_second_per_gpu:5342.79982953138 mfu:25.20289837154914 +Step 18034 | loss:0.749295711517334 lr:5.0687269502967724e-05 tokens_per_second_per_gpu:4546.903074782718 mfu:21.44851758541071 +Step 18035 | loss:0.9621540904045105 lr:5.068293368829755e-05 tokens_per_second_per_gpu:3424.384514262276 mfu:16.153406014020295 +Step 18036 | loss:0.691246747970581 lr:5.067859786849096e-05 tokens_per_second_per_gpu:5348.556840659055 mfu:25.230055175285766 +Step 18037 | loss:0.848146915435791 lr:5.067426204358052e-05 tokens_per_second_per_gpu:5360.300034812026 mfu:25.28544982570086 +Step 18038 | loss:0.7718494534492493 lr:5.066992621359889e-05 tokens_per_second_per_gpu:5325.548221911568 mfu:25.121519594978714 +Step 18039 | loss:0.8476927876472473 lr:5.0665590378578676e-05 tokens_per_second_per_gpu:4642.694640934642 mfu:21.90038274667583 +Step 18040 | loss:0.8602629899978638 lr:5.066125453855244e-05 tokens_per_second_per_gpu:3726.1884361341968 mfu:17.577066606549014 +Step 18041 | loss:0.8686544895172119 lr:5.0656918693552836e-05 tokens_per_second_per_gpu:5315.212521390611 mfu:25.072764332169008 +Step 18042 | loss:0.7468135952949524 lr:5.065258284361245e-05 tokens_per_second_per_gpu:5359.432009922034 mfu:25.281355204194448 +Step 18043 | loss:0.8591925501823425 lr:5.064824698876393e-05 tokens_per_second_per_gpu:5371.36119024666 mfu:25.337627183113753 +Step 18044 | loss:0.6620145440101624 lr:5.064391112903983e-05 tokens_per_second_per_gpu:3345.210545182094 mfu:15.779928893397003 +Step 18045 | loss:0.6840814352035522 lr:5.063957526447282e-05 tokens_per_second_per_gpu:4617.409205410154 mfu:21.781107033166663 +Step 18046 | loss:0.7132622003555298 lr:5.063523939509548e-05 tokens_per_second_per_gpu:4978.206289789869 mfu:23.483048438516064 +Step 18047 | loss:0.8412183523178101 lr:5.063090352094041e-05 tokens_per_second_per_gpu:4896.135655088014 mfu:23.095907251933223 +Step 18048 | loss:0.6833211183547974 lr:5.0626567642040244e-05 tokens_per_second_per_gpu:3951.55932280187 mfu:18.64017953119876 +Step 18049 | loss:0.8310604095458984 lr:5.062223175842757e-05 tokens_per_second_per_gpu:4553.998695972498 mfu:21.481988841200717 +Step 18050 | loss:0.6822171807289124 lr:5.0617895870135034e-05 tokens_per_second_per_gpu:3562.3519518184585 mfu:16.80422195664499 +Step 18051 | loss:0.742922306060791 lr:5.06135599771952e-05 tokens_per_second_per_gpu:4891.161232163748 mfu:23.072442050275388 +Step 18052 | loss:0.9279353022575378 lr:5.0609224079640717e-05 tokens_per_second_per_gpu:5358.916534093829 mfu:25.27892361676356 +Step 18053 | loss:0.680142879486084 lr:5.060488817750417e-05 tokens_per_second_per_gpu:5359.631027408592 mfu:25.282294003634426 +Step 18054 | loss:0.7504333257675171 lr:5.060055227081821e-05 tokens_per_second_per_gpu:5347.398852715854 mfu:25.224592748584573 +Step 18055 | loss:0.7636215090751648 lr:5.059621635961539e-05 tokens_per_second_per_gpu:3492.596911443788 mfu:16.475175529760545 +Step 18056 | loss:0.7572774291038513 lr:5.0591880443928366e-05 tokens_per_second_per_gpu:3657.8598239294265 mfu:17.254748884715728 +Step 18057 | loss:0.6447437405586243 lr:5.058754452378974e-05 tokens_per_second_per_gpu:5351.27447570801 mfu:25.24287472348804 +Step 18058 | loss:0.7763156890869141 lr:5.05832085992321e-05 tokens_per_second_per_gpu:4532.038257026319 mfu:21.378397703854343 +Step 18059 | loss:0.8748987317085266 lr:5.057887267028808e-05 tokens_per_second_per_gpu:5309.903718813557 mfu:25.04772180463792 +Step 18060 | loss:0.8021437525749207 lr:5.0574536736990276e-05 tokens_per_second_per_gpu:3873.63323619687 mfu:18.27258888512161 +Step 18061 | loss:0.5864706635475159 lr:5.0570200799371324e-05 tokens_per_second_per_gpu:4934.31560369018 mfu:23.276008583660595 +Step 18062 | loss:0.6672627329826355 lr:5.056586485746381e-05 tokens_per_second_per_gpu:4204.463964497421 mfu:19.833173876058037 +Step 18063 | loss:0.7630185484886169 lr:5.056152891130035e-05 tokens_per_second_per_gpu:4070.8628524002065 mfu:19.202954635595237 +Step 18064 | loss:0.7200717329978943 lr:5.055719296091357e-05 tokens_per_second_per_gpu:4036.227453164464 mfu:19.03957354799177 +Step 18065 | loss:0.693100094795227 lr:5.055285700633606e-05 tokens_per_second_per_gpu:5330.87518236426 mfu:25.14664777630723 +Step 18066 | loss:0.839688777923584 lr:5.054852104760046e-05 tokens_per_second_per_gpu:4972.878869712585 mfu:23.457918089059923 +Step 18067 | loss:0.7776076793670654 lr:5.054418508473934e-05 tokens_per_second_per_gpu:3700.3482790375306 mfu:17.45517417673862 +Step 18068 | loss:0.8374958634376526 lr:5.053984911778535e-05 tokens_per_second_per_gpu:4872.9612523928545 mfu:22.986589640459336 +Step 18069 | loss:0.6914270520210266 lr:5.053551314677108e-05 tokens_per_second_per_gpu:4139.878546971898 mfu:19.528513442181627 +Step 18070 | loss:0.8132985234260559 lr:5.053117717172915e-05 tokens_per_second_per_gpu:5216.275019556494 mfu:24.606059255539005 +Step 18071 | loss:0.652453601360321 lr:5.052684119269215e-05 tokens_per_second_per_gpu:5331.359439311779 mfu:25.148932098951434 +Step 18072 | loss:0.8214676976203918 lr:5.052250520969275e-05 tokens_per_second_per_gpu:5299.313110797628 mfu:24.99776409968273 +Step 18073 | loss:0.8408384323120117 lr:5.05181692227635e-05 tokens_per_second_per_gpu:4806.743327911688 mfu:22.67422839273875 +Step 18074 | loss:0.7782187461853027 lr:5.051383323193702e-05 tokens_per_second_per_gpu:4389.251829200815 mfu:20.704849761924738 +Step 18075 | loss:0.8322886824607849 lr:5.050949723724596e-05 tokens_per_second_per_gpu:5356.264965900428 mfu:25.266415717191105 +Step 18076 | loss:0.877906858921051 lr:5.0505161238722886e-05 tokens_per_second_per_gpu:3737.1578053552744 mfu:17.628810992732216 +Step 18077 | loss:0.8244740962982178 lr:5.050082523640044e-05 tokens_per_second_per_gpu:4626.215103287603 mfu:21.82264595589569 +Step 18078 | loss:0.8029420375823975 lr:5.04964892303112e-05 tokens_per_second_per_gpu:4087.8262610481556 mfu:19.282973928443575 +Step 18079 | loss:0.6938450932502747 lr:5.049215322048785e-05 tokens_per_second_per_gpu:5335.319591605051 mfu:25.167612812990452 +Step 18080 | loss:0.7097692489624023 lr:5.048781720696291e-05 tokens_per_second_per_gpu:3859.191560160204 mfu:18.204465035252642 +Step 18081 | loss:0.6676906943321228 lr:5.048348118976906e-05 tokens_per_second_per_gpu:5374.420113829516 mfu:25.352056647560257 +Step 18082 | loss:0.7337028980255127 lr:5.047914516893887e-05 tokens_per_second_per_gpu:4827.009494892903 mfu:22.76982736015378 +Step 18083 | loss:0.677904486656189 lr:5.047480914450498e-05 tokens_per_second_per_gpu:4810.7156139834815 mfu:22.69296634388129 +Step 18084 | loss:0.6595788598060608 lr:5.047047311649998e-05 tokens_per_second_per_gpu:4991.907700168068 mfu:23.547680328971698 +Step 18085 | loss:0.763660728931427 lr:5.0466137084956485e-05 tokens_per_second_per_gpu:5357.718292639565 mfu:25.273271307382725 +Step 18086 | loss:0.7961892485618591 lr:5.046180104990713e-05 tokens_per_second_per_gpu:5311.53667324197 mfu:25.0554247292875 +Step 18087 | loss:0.7574458122253418 lr:5.0457465011384496e-05 tokens_per_second_per_gpu:4522.853052501538 mfu:21.335069526953458 +Step 18088 | loss:0.7956886887550354 lr:5.045312896942123e-05 tokens_per_second_per_gpu:4149.2261494343165 mfu:19.572607677862294 +Step 18089 | loss:0.7670689225196838 lr:5.0448792924049894e-05 tokens_per_second_per_gpu:3944.9788794753354 mfu:18.609138457288093 +Step 18090 | loss:0.7214329242706299 lr:5.044445687530316e-05 tokens_per_second_per_gpu:4513.385621979989 mfu:21.2904100418731 +Step 18091 | loss:0.6170473694801331 lr:5.04401208232136e-05 tokens_per_second_per_gpu:4201.505142589137 mfu:19.819216608289587 +Step 18092 | loss:0.8072211742401123 lr:5.0435784767813824e-05 tokens_per_second_per_gpu:5077.107189799618 mfu:23.949580858095608 +Step 18093 | loss:0.7875955700874329 lr:5.0431448709136455e-05 tokens_per_second_per_gpu:4484.224885904839 mfu:21.152853874472 +Step 18094 | loss:0.7192595601081848 lr:5.042711264721412e-05 tokens_per_second_per_gpu:5345.4710500151095 mfu:25.215498974328106 +Step 18095 | loss:0.6987417936325073 lr:5.0422776582079404e-05 tokens_per_second_per_gpu:5358.259160682023 mfu:25.275822674220603 +Step 18096 | loss:0.7934067249298096 lr:5.041844051376492e-05 tokens_per_second_per_gpu:4168.575069632545 mfu:19.66387983570354 +Step 18097 | loss:0.8487052321434021 lr:5.0414104442303326e-05 tokens_per_second_per_gpu:4721.905673600982 mfu:22.27403470255873 +Step 18098 | loss:0.761808454990387 lr:5.040976836772716e-05 tokens_per_second_per_gpu:4303.362971758161 mfu:20.299697367217565 +Step 18099 | loss:0.812239944934845 lr:5.04054322900691e-05 tokens_per_second_per_gpu:5316.556006261079 mfu:25.07910177952516 +Step 18100 | loss:0.6985002160072327 lr:5.040109620936173e-05 tokens_per_second_per_gpu:5097.616015021647 mfu:24.04632448583433 +Step 18101 | loss:0.6851450800895691 lr:5.039676012563767e-05 tokens_per_second_per_gpu:5333.2120480253325 mfu:25.157671170340905 +Step 18102 | loss:0.7918065786361694 lr:5.039242403892951e-05 tokens_per_second_per_gpu:5338.04704253072 mfu:25.18047866435795 +Step 18103 | loss:0.7769443392753601 lr:5.038808794926988e-05 tokens_per_second_per_gpu:4883.632164702396 mfu:23.036926154468826 +Step 18104 | loss:0.7402178049087524 lr:5.038375185669141e-05 tokens_per_second_per_gpu:5341.384153423346 mfu:25.196220385808296 +Step 18105 | loss:0.6503791213035583 lr:5.037941576122667e-05 tokens_per_second_per_gpu:5311.023413584777 mfu:25.053003595913392 +Step 18106 | loss:0.7483675479888916 lr:5.03750796629083e-05 tokens_per_second_per_gpu:2668.7617467363657 mfu:12.589004497062366 +Step 18107 | loss:0.919651210308075 lr:5.037074356176892e-05 tokens_per_second_per_gpu:4652.4681974388395 mfu:21.946486280246823 +Step 18108 | loss:0.7015228867530823 lr:5.036640745784113e-05 tokens_per_second_per_gpu:5357.775551186987 mfu:25.27354140572023 +Step 18109 | loss:0.7796456217765808 lr:5.036207135115752e-05 tokens_per_second_per_gpu:5170.909755097671 mfu:24.392063562974485 +Step 18110 | loss:0.794722855091095 lr:5.0357735241750735e-05 tokens_per_second_per_gpu:5334.147825446736 mfu:25.162085392097755 +Step 18111 | loss:0.9061611294746399 lr:5.035339912965339e-05 tokens_per_second_per_gpu:3448.4433868984324 mfu:16.266895821111998 +Step 18112 | loss:0.781944990158081 lr:5.034906301489808e-05 tokens_per_second_per_gpu:4209.693708514703 mfu:19.857843470873892 +Step 18113 | loss:0.8028182983398438 lr:5.034472689751742e-05 tokens_per_second_per_gpu:4627.723374897053 mfu:21.829760730415142 +Step 18114 | loss:0.7121259570121765 lr:5.034039077754402e-05 tokens_per_second_per_gpu:5159.094871584591 mfu:24.33633074161629 +Step 18115 | loss:0.7096735239028931 lr:5.0336054655010526e-05 tokens_per_second_per_gpu:4043.053922465126 mfu:19.07177516839876 +Step 18116 | loss:0.6732339262962341 lr:5.033171852994949e-05 tokens_per_second_per_gpu:5173.7768341907085 mfu:24.405588064230475 +Step 18117 | loss:0.7270714044570923 lr:5.0327382402393565e-05 tokens_per_second_per_gpu:4579.231688469991 mfu:21.601017172003846 +Step 18118 | loss:0.7656500935554504 lr:5.032304627237536e-05 tokens_per_second_per_gpu:5019.426253346308 mfu:23.677490039462775 +Step 18119 | loss:0.7811625003814697 lr:5.031871013992748e-05 tokens_per_second_per_gpu:4473.831574744481 mfu:21.103826852446854 +Step 18120 | loss:0.7592304348945618 lr:5.0314374005082546e-05 tokens_per_second_per_gpu:4210.347068638904 mfu:19.86092548205448 +Step 18121 | loss:0.7093274593353271 lr:5.031003786787315e-05 tokens_per_second_per_gpu:5348.605597152706 mfu:25.230285167985034 +Step 18122 | loss:0.9091514945030212 lr:5.030570172833193e-05 tokens_per_second_per_gpu:4224.9479280864125 mfu:19.929800227231883 +Step 18123 | loss:0.7873721718788147 lr:5.030136558649149e-05 tokens_per_second_per_gpu:5254.9716477033335 mfu:24.788597852833217 +Step 18124 | loss:0.7290552854537964 lr:5.029702944238443e-05 tokens_per_second_per_gpu:5354.099622275767 mfu:25.256201421867857 +Step 18125 | loss:0.8256460428237915 lr:5.029269329604339e-05 tokens_per_second_per_gpu:3927.2141382895984 mfu:18.525339142137582 +Step 18126 | loss:0.7310308814048767 lr:5.0288357147500956e-05 tokens_per_second_per_gpu:4109.8116121107005 mfu:19.38668258049329 +Step 18127 | loss:0.7422158718109131 lr:5.028402099678976e-05 tokens_per_second_per_gpu:4174.241024655064 mfu:19.690607112256377 +Step 18128 | loss:0.7352228164672852 lr:5.027968484394239e-05 tokens_per_second_per_gpu:5353.432896320082 mfu:25.253056361779716 +Step 18129 | loss:0.6998602151870728 lr:5.027534868899149e-05 tokens_per_second_per_gpu:4609.495339987959 mfu:21.74377597972549 +Step 18130 | loss:0.7724665403366089 lr:5.0271012531969644e-05 tokens_per_second_per_gpu:3770.4504674813315 mfu:17.785858160294723 +Step 18131 | loss:0.8851327300071716 lr:5.026667637290949e-05 tokens_per_second_per_gpu:5352.644060868531 mfu:25.24933529036554 +Step 18132 | loss:0.8456205129623413 lr:5.0262340211843615e-05 tokens_per_second_per_gpu:4406.196895575374 mfu:20.784782531139975 +Step 18133 | loss:0.8518396615982056 lr:5.0258004048804666e-05 tokens_per_second_per_gpu:3949.8010399036016 mfu:18.631885410773634 +Step 18134 | loss:0.7698560953140259 lr:5.025366788382522e-05 tokens_per_second_per_gpu:4795.2707842064965 mfu:22.620110446663563 +Step 18135 | loss:0.7334575057029724 lr:5.024933171693791e-05 tokens_per_second_per_gpu:5370.627462303502 mfu:25.33416606321898 +Step 18136 | loss:0.6988600492477417 lr:5.0244995548175346e-05 tokens_per_second_per_gpu:4182.654400890238 mfu:19.730294443427695 +Step 18137 | loss:0.7426322102546692 lr:5.024065937757013e-05 tokens_per_second_per_gpu:5289.583409209227 mfu:24.951867437231975 +Step 18138 | loss:0.7036402225494385 lr:5.02363232051549e-05 tokens_per_second_per_gpu:5309.564595804157 mfu:25.046122103542228 +Step 18139 | loss:0.767329752445221 lr:5.023198703096223e-05 tokens_per_second_per_gpu:5357.598311633485 mfu:25.27270533650617 +Step 18140 | loss:0.7953102588653564 lr:5.0227650855024774e-05 tokens_per_second_per_gpu:5391.987884064087 mfu:25.434926817127412 +Step 18141 | loss:0.7498171329498291 lr:5.022331467737513e-05 tokens_per_second_per_gpu:5317.450011851513 mfu:25.083318956428318 +Step 18142 | loss:0.677455484867096 lr:5.02189784980459e-05 tokens_per_second_per_gpu:4940.92201815182 mfu:23.307172167846687 +Step 18143 | loss:0.6918251514434814 lr:5.021464231706971e-05 tokens_per_second_per_gpu:4214.45887889357 mfu:19.88032159257291 +Step 18144 | loss:0.7934516072273254 lr:5.021030613447917e-05 tokens_per_second_per_gpu:5084.123545802367 mfu:23.98267820647435 +Step 18145 | loss:0.7394607663154602 lr:5.0205969950306886e-05 tokens_per_second_per_gpu:5335.773773053116 mfu:25.169755264372323 +Step 18146 | loss:0.8318387866020203 lr:5.0201633764585465e-05 tokens_per_second_per_gpu:4239.4267870665735 mfu:19.998099475389377 +Step 18147 | loss:0.7164584994316101 lr:5.0197297577347536e-05 tokens_per_second_per_gpu:5383.894612621239 mfu:25.396749474876998 +Step 18148 | loss:0.5757924914360046 lr:5.0192961388625715e-05 tokens_per_second_per_gpu:4981.378936154478 mfu:23.49801435272721 +Step 18149 | loss:0.9550482034683228 lr:5.01886251984526e-05 tokens_per_second_per_gpu:5287.824889197177 mfu:24.94357219830068 +Step 18150 | loss:0.7865567207336426 lr:5.018428900686081e-05 tokens_per_second_per_gpu:4652.3630690635055 mfu:21.94599037176362 +Step 18151 | loss:0.8580780029296875 lr:5.017995281388297e-05 tokens_per_second_per_gpu:4183.970819981776 mfu:19.73650421688674 +Step 18152 | loss:0.5909789204597473 lr:5.017561661955167e-05 tokens_per_second_per_gpu:5349.921009828246 mfu:25.236490194008102 +Step 18153 | loss:0.745789110660553 lr:5.017128042389953e-05 tokens_per_second_per_gpu:4744.741631248955 mfu:22.38175581100036 +Step 18154 | loss:0.8966683745384216 lr:5.016694422695919e-05 tokens_per_second_per_gpu:4088.164464590935 mfu:19.284569292258336 +Step 18155 | loss:0.7250725626945496 lr:5.016260802876323e-05 tokens_per_second_per_gpu:4831.702591088416 mfu:22.791965495632798 +Step 18156 | loss:0.8251634240150452 lr:5.0158271829344275e-05 tokens_per_second_per_gpu:5339.880388604751 mfu:25.189126870590243 +Step 18157 | loss:0.8260310292243958 lr:5.015393562873493e-05 tokens_per_second_per_gpu:5336.471592497722 mfu:25.173046997752802 +Step 18158 | loss:0.7422195076942444 lr:5.014959942696782e-05 tokens_per_second_per_gpu:5041.421942612983 mfu:23.781247458585252 +Step 18159 | loss:0.8395518064498901 lr:5.0145263224075554e-05 tokens_per_second_per_gpu:3280.599569175739 mfu:15.475147895805785 +Step 18160 | loss:0.7536793947219849 lr:5.014092702009075e-05 tokens_per_second_per_gpu:3940.157903365228 mfu:18.586397090433376 +Step 18161 | loss:0.7997948527336121 lr:5.0136590815046005e-05 tokens_per_second_per_gpu:5017.982385631473 mfu:23.670679069102665 +Step 18162 | loss:0.8846971392631531 lr:5.013225460897395e-05 tokens_per_second_per_gpu:4629.331674519728 mfu:21.83734735413951 +Step 18163 | loss:0.6288492679595947 lr:5.0127918401907194e-05 tokens_per_second_per_gpu:4402.485271937611 mfu:20.767274169172357 +Step 18164 | loss:0.6768900156021118 lr:5.012358219387834e-05 tokens_per_second_per_gpu:5390.031033647469 mfu:25.425696019839116 +Step 18165 | loss:0.7729039192199707 lr:5.011924598492001e-05 tokens_per_second_per_gpu:3265.5082932502955 mfu:15.403959772428287 +Step 18166 | loss:0.7801185250282288 lr:5.011490977506482e-05 tokens_per_second_per_gpu:3747.9075062212287 mfu:17.679519165805015 +Step 18167 | loss:0.7216975092887878 lr:5.011057356434538e-05 tokens_per_second_per_gpu:3497.82742601102 mfu:16.499848759391988 +Step 18168 | loss:0.89042067527771 lr:5.0106237352794284e-05 tokens_per_second_per_gpu:5334.709266740376 mfu:25.16473380647162 +Step 18169 | loss:0.9047752022743225 lr:5.0101901140444193e-05 tokens_per_second_per_gpu:4020.3636675615194 mfu:18.964741364661755 +Step 18170 | loss:0.7701204419136047 lr:5.009756492732767e-05 tokens_per_second_per_gpu:4481.626556118527 mfu:21.140597109549194 +Step 18171 | loss:0.7113955616950989 lr:5.0093228713477346e-05 tokens_per_second_per_gpu:4494.399992015053 mfu:21.200851586046 +Step 18172 | loss:0.8137363195419312 lr:5.008889249892586e-05 tokens_per_second_per_gpu:4576.368938575183 mfu:21.58751309231454 +Step 18173 | loss:0.7297219634056091 lr:5.008455628370579e-05 tokens_per_second_per_gpu:4399.844481881422 mfu:20.754817111912068 +Step 18174 | loss:0.8155258297920227 lr:5.008022006784975e-05 tokens_per_second_per_gpu:4241.0846734765955 mfu:20.005920008450044 +Step 18175 | loss:0.867054283618927 lr:5.007588385139037e-05 tokens_per_second_per_gpu:4089.059174628618 mfu:19.288789792159363 +Step 18176 | loss:0.7663536071777344 lr:5.007154763436026e-05 tokens_per_second_per_gpu:4653.093177662039 mfu:21.949434418592276 +Step 18177 | loss:0.891814649105072 lr:5.006721141679204e-05 tokens_per_second_per_gpu:3477.6281852536576 mfu:16.40456549439366 +Step 18178 | loss:0.7235625982284546 lr:5.00628751987183e-05 tokens_per_second_per_gpu:5385.383037007367 mfu:25.403770626657757 +Step 18179 | loss:0.8123870491981506 lr:5.005853898017168e-05 tokens_per_second_per_gpu:5354.928182579793 mfu:25.260109882188917 +Step 18180 | loss:0.8705298900604248 lr:5.0054202761184774e-05 tokens_per_second_per_gpu:4081.1356768378937 mfu:19.251413301192393 +Step 18181 | loss:0.8316448330879211 lr:5.00498665417902e-05 tokens_per_second_per_gpu:4604.581131490545 mfu:21.720594819792765 +Step 18182 | loss:0.7934712171554565 lr:5.004553032202057e-05 tokens_per_second_per_gpu:4841.1771657800455 mfu:22.836658680982346 +Step 18183 | loss:0.7191749215126038 lr:5.0041194101908514e-05 tokens_per_second_per_gpu:3920.127313569172 mfu:18.491909380794354 +Step 18184 | loss:0.8191718459129333 lr:5.0036857881486633e-05 tokens_per_second_per_gpu:4613.046272247963 mfu:21.76052633304745 +Step 18185 | loss:0.8475558757781982 lr:5.003252166078752e-05 tokens_per_second_per_gpu:5116.037807171704 mfu:24.13322322248799 +Step 18186 | loss:0.8463776707649231 lr:5.002818543984381e-05 tokens_per_second_per_gpu:4999.521447876689 mfu:23.583595675952573 +Step 18187 | loss:0.8092308640480042 lr:5.002384921868813e-05 tokens_per_second_per_gpu:4275.019484820241 mfu:20.165996303433865 +Step 18188 | loss:0.8103898763656616 lr:5.001951299735307e-05 tokens_per_second_per_gpu:5373.123767230277 mfu:25.345941559472113 +Step 18189 | loss:0.7517211437225342 lr:5.001517677587125e-05 tokens_per_second_per_gpu:5265.662074414209 mfu:24.839026419603297 +Step 18190 | loss:0.8230862617492676 lr:5.001084055427528e-05 tokens_per_second_per_gpu:4429.2183409133395 mfu:20.893378616662762 +Step 18191 | loss:0.6714479327201843 lr:5.0006504332597794e-05 tokens_per_second_per_gpu:5352.604430582354 mfu:25.249148347544025 +Step 18192 | loss:0.7171737551689148 lr:5.0002168110871374e-05 tokens_per_second_per_gpu:4117.615459569298 mfu:19.423494660429164 +Step 18193 | loss:0.8342835307121277 lr:4.999783188912864e-05 tokens_per_second_per_gpu:3957.234048038022 mfu:18.66694817834566 +Step 18194 | loss:0.6721192598342896 lr:4.999349566740222e-05 tokens_per_second_per_gpu:3873.9038815330637 mfu:18.27386556534929 +Step 18195 | loss:0.7761701941490173 lr:4.998915944572472e-05 tokens_per_second_per_gpu:4513.138374385052 mfu:21.28924373278299 +Step 18196 | loss:0.8208156824111938 lr:4.9984823224128746e-05 tokens_per_second_per_gpu:4482.708287907812 mfu:21.145699822961724 +Step 18197 | loss:0.7331773638725281 lr:4.9980487002646935e-05 tokens_per_second_per_gpu:4468.353634754879 mfu:21.07798647488296 +Step 18198 | loss:0.7545862197875977 lr:4.997615078131188e-05 tokens_per_second_per_gpu:4340.296322064021 mfu:20.473918282089663 +Step 18199 | loss:0.9206231236457825 lr:4.9971814560156196e-05 tokens_per_second_per_gpu:4196.2987375746225 mfu:19.794657107531826 +Step 18200 | loss:0.7114745378494263 lr:4.996747833921249e-05 tokens_per_second_per_gpu:5039.68812361716 mfu:23.773068738561932 +Step 18201 | loss:0.6673945784568787 lr:4.9963142118513385e-05 tokens_per_second_per_gpu:5196.996799394163 mfu:24.515120601829754 +Step 18202 | loss:0.833847165107727 lr:4.995880589809149e-05 tokens_per_second_per_gpu:3946.632796185733 mfu:18.61694026459833 +Step 18203 | loss:0.8415305614471436 lr:4.995446967797943e-05 tokens_per_second_per_gpu:5322.714090167989 mfu:25.10815051199151 +Step 18204 | loss:0.8666070103645325 lr:4.9950133458209817e-05 tokens_per_second_per_gpu:3641.9670923121384 mfu:17.17978015809736 +Step 18205 | loss:0.7854452133178711 lr:4.9945797238815245e-05 tokens_per_second_per_gpu:5358.314563711369 mfu:25.27608401976276 +Step 18206 | loss:0.7938785552978516 lr:4.994146101982833e-05 tokens_per_second_per_gpu:4572.84532049773 mfu:21.570891584651957 +Step 18207 | loss:0.8372307419776917 lr:4.993712480128171e-05 tokens_per_second_per_gpu:5176.863509037387 mfu:24.420148436123295 +Step 18208 | loss:0.8388918042182922 lr:4.993278858320798e-05 tokens_per_second_per_gpu:3617.8792024623413 mfu:17.066153471858065 +Step 18209 | loss:0.7622449398040771 lr:4.992845236563974e-05 tokens_per_second_per_gpu:4375.932889857041 mfu:20.642022052594776 +Step 18210 | loss:0.8084011077880859 lr:4.9924116148609636e-05 tokens_per_second_per_gpu:3536.057686408334 mfu:16.680187420441147 +Step 18211 | loss:0.7423444986343384 lr:4.991977993215027e-05 tokens_per_second_per_gpu:5389.73210661113 mfu:25.424285930006494 +Step 18212 | loss:0.8547736406326294 lr:4.991544371629424e-05 tokens_per_second_per_gpu:5363.727489345322 mfu:25.301617713518464 +Step 18213 | loss:0.7169710993766785 lr:4.9911107501074154e-05 tokens_per_second_per_gpu:4466.403864183195 mfu:21.068789074431166 +Step 18214 | loss:0.7760955691337585 lr:4.990677128652266e-05 tokens_per_second_per_gpu:4260.330209904195 mfu:20.09670448740614 +Step 18215 | loss:0.8150571584701538 lr:4.9902435072672345e-05 tokens_per_second_per_gpu:3765.3828492862253 mfu:17.761953340643586 +Step 18216 | loss:0.6048624515533447 lr:4.989809885955582e-05 tokens_per_second_per_gpu:5332.017541312028 mfu:25.152036478370388 +Step 18217 | loss:0.9370433688163757 lr:4.9893762647205714e-05 tokens_per_second_per_gpu:4486.2046956282 mfu:21.162192974727468 +Step 18218 | loss:0.894334077835083 lr:4.988942643565465e-05 tokens_per_second_per_gpu:4708.939247366033 mfu:22.212869857708036 +Step 18219 | loss:0.8732719421386719 lr:4.9885090224935204e-05 tokens_per_second_per_gpu:4273.316509173703 mfu:20.15796307675154 +Step 18220 | loss:0.6641796231269836 lr:4.9880754015080005e-05 tokens_per_second_per_gpu:5070.99286103419 mfu:23.920738526097026 +Step 18221 | loss:0.6869708299636841 lr:4.987641780612168e-05 tokens_per_second_per_gpu:4239.834110030229 mfu:20.000020887305404 +Step 18222 | loss:0.803420901298523 lr:4.987208159809282e-05 tokens_per_second_per_gpu:4314.046278503659 mfu:20.35009234789617 +Step 18223 | loss:0.814659059047699 lr:4.986774539102606e-05 tokens_per_second_per_gpu:5055.34085677024 mfu:23.84690535147667 +Step 18224 | loss:0.9035449624061584 lr:4.9863409184954e-05 tokens_per_second_per_gpu:4210.906438095246 mfu:19.863564123218318 +Step 18225 | loss:0.8055173754692078 lr:4.985907297990926e-05 tokens_per_second_per_gpu:4495.980324402037 mfu:21.208306283547913 +Step 18226 | loss:0.5928351879119873 lr:4.985473677592446e-05 tokens_per_second_per_gpu:5322.01975756735 mfu:25.10487522664894 +Step 18227 | loss:0.7713743448257446 lr:4.9850400573032195e-05 tokens_per_second_per_gpu:3996.627082725855 mfu:18.852771844112947 +Step 18228 | loss:0.8811935186386108 lr:4.984606437126509e-05 tokens_per_second_per_gpu:4171.1485299077485 mfu:19.676019286899272 +Step 18229 | loss:0.9491631984710693 lr:4.984172817065574e-05 tokens_per_second_per_gpu:5285.952109052709 mfu:24.934737974829982 +Step 18230 | loss:0.7269385457038879 lr:4.9837391971236784e-05 tokens_per_second_per_gpu:5333.407507624046 mfu:25.158593186616887 +Step 18231 | loss:0.8157747983932495 lr:4.9833055773040824e-05 tokens_per_second_per_gpu:4207.170890451868 mfu:19.845942907634353 +Step 18232 | loss:0.7947688698768616 lr:4.9828719576100465e-05 tokens_per_second_per_gpu:4317.972250696693 mfu:20.368611828570636 +Step 18233 | loss:0.6259413957595825 lr:4.982438338044834e-05 tokens_per_second_per_gpu:4365.275332117582 mfu:20.591748534370893 +Step 18234 | loss:0.9114944934844971 lr:4.982004718611705e-05 tokens_per_second_per_gpu:3914.5376694485203 mfu:18.46554207578561 +Step 18235 | loss:0.8440822958946228 lr:4.9815710993139204e-05 tokens_per_second_per_gpu:4698.597130378169 mfu:22.164084327329167 +Step 18236 | loss:0.7654081583023071 lr:4.9811374801547404e-05 tokens_per_second_per_gpu:5046.237175304198 mfu:23.803961732753127 +Step 18237 | loss:0.6301599144935608 lr:4.98070386113743e-05 tokens_per_second_per_gpu:4522.933546334186 mfu:21.335449230095723 +Step 18238 | loss:0.7705721855163574 lr:4.980270242265246e-05 tokens_per_second_per_gpu:4381.270189920492 mfu:20.667199007631364 +Step 18239 | loss:0.690937340259552 lr:4.979836623541454e-05 tokens_per_second_per_gpu:4481.6689384777965 mfu:21.140797034369044 +Step 18240 | loss:0.766785204410553 lr:4.979403004969314e-05 tokens_per_second_per_gpu:4907.781897614072 mfu:23.15084456498251 +Step 18241 | loss:0.7448595762252808 lr:4.978969386552086e-05 tokens_per_second_per_gpu:5337.016164526394 mfu:25.17561583692562 +Step 18242 | loss:0.7279012203216553 lr:4.97853576829303e-05 tokens_per_second_per_gpu:5318.242153823697 mfu:25.087055625264465 +Step 18243 | loss:0.8048608899116516 lr:4.978102150195411e-05 tokens_per_second_per_gpu:4163.481328008535 mfu:19.639851787381144 +Step 18244 | loss:0.8553210496902466 lr:4.977668532262488e-05 tokens_per_second_per_gpu:5321.908318662851 mfu:25.10434955032298 +Step 18245 | loss:0.897421658039093 lr:4.9772349144975224e-05 tokens_per_second_per_gpu:3845.008963122014 mfu:18.137563305222923 +Step 18246 | loss:0.7351891994476318 lr:4.9768012969037765e-05 tokens_per_second_per_gpu:5055.466143172764 mfu:23.84749634881344 +Step 18247 | loss:0.7209870219230652 lr:4.976367679484512e-05 tokens_per_second_per_gpu:4876.222772926655 mfu:23.00197478929006 +Step 18248 | loss:0.7426509857177734 lr:4.975934062242988e-05 tokens_per_second_per_gpu:5347.766076352259 mfu:25.226325005132306 +Step 18249 | loss:0.812153160572052 lr:4.9755004451824665e-05 tokens_per_second_per_gpu:4030.310278376549 mfu:19.011661224943055 +Step 18250 | loss:0.6513473391532898 lr:4.9750668283062104e-05 tokens_per_second_per_gpu:3861.8225769153382 mfu:18.21687599018436 +Step 18251 | loss:0.9131243824958801 lr:4.97463321161748e-05 tokens_per_second_per_gpu:5071.367820908336 mfu:23.92250727579813 +Step 18252 | loss:0.8147456049919128 lr:4.9741995951195345e-05 tokens_per_second_per_gpu:5341.228638834369 mfu:25.195486796958566 +Step 18253 | loss:0.6974791884422302 lr:4.973765978815639e-05 tokens_per_second_per_gpu:4193.006025767155 mfu:19.779124824141437 +Step 18254 | loss:0.7103599309921265 lr:4.973332362709053e-05 tokens_per_second_per_gpu:4916.450074549759 mfu:23.191733834531636 +Step 18255 | loss:0.6791859865188599 lr:4.9728987468030375e-05 tokens_per_second_per_gpu:4999.904034085118 mfu:23.585400400373672 +Step 18256 | loss:0.8060079216957092 lr:4.972465131100853e-05 tokens_per_second_per_gpu:4618.288364027075 mfu:21.785254174362606 +Step 18257 | loss:0.6686439514160156 lr:4.972031515605762e-05 tokens_per_second_per_gpu:4767.10846230206 mfu:22.487263969252744 +Step 18258 | loss:0.8389835953712463 lr:4.9715979003210254e-05 tokens_per_second_per_gpu:4025.7969686366123 mfu:18.990371172849724 +Step 18259 | loss:0.798241376876831 lr:4.971164285249905e-05 tokens_per_second_per_gpu:5049.8790407050155 mfu:23.821141033215095 +Step 18260 | loss:0.652958333492279 lr:4.970730670395662e-05 tokens_per_second_per_gpu:3887.61680733859 mfu:18.33855177604036 +Step 18261 | loss:0.7348007559776306 lr:4.9702970557615565e-05 tokens_per_second_per_gpu:4790.468296184663 mfu:22.597456291275627 +Step 18262 | loss:0.7653189897537231 lr:4.969863441350853e-05 tokens_per_second_per_gpu:4700.18258979527 mfu:22.171563209906907 +Step 18263 | loss:0.7019778490066528 lr:4.969429827166808e-05 tokens_per_second_per_gpu:4072.136596230992 mfu:19.20896310256816 +Step 18264 | loss:0.7316539287567139 lr:4.968996213212686e-05 tokens_per_second_per_gpu:4723.003845133401 mfu:22.279214965044538 +Step 18265 | loss:0.7427484393119812 lr:4.9685625994917465e-05 tokens_per_second_per_gpu:4932.818286421703 mfu:23.268945482636518 +Step 18266 | loss:0.8543707132339478 lr:4.968128986007253e-05 tokens_per_second_per_gpu:4499.80783407103 mfu:21.22636129969704 +Step 18267 | loss:0.7617329359054565 lr:4.9676953727624656e-05 tokens_per_second_per_gpu:4697.434079255841 mfu:22.158598016748165 +Step 18268 | loss:0.8239293694496155 lr:4.967261759760643e-05 tokens_per_second_per_gpu:5344.442686788009 mfu:25.21064800952901 +Step 18269 | loss:0.7772607803344727 lr:4.966828147005053e-05 tokens_per_second_per_gpu:3938.60946297389 mfu:18.579092832916263 +Step 18270 | loss:0.6957371234893799 lr:4.96639453449895e-05 tokens_per_second_per_gpu:5351.486868213255 mfu:25.24387661517318 +Step 18271 | loss:0.9343125224113464 lr:4.965960922245599e-05 tokens_per_second_per_gpu:4507.307134127825 mfu:21.261736777577468 +Step 18272 | loss:0.7004774212837219 lr:4.965527310248259e-05 tokens_per_second_per_gpu:4584.49802834912 mfu:21.625859395743934 +Step 18273 | loss:0.8279751539230347 lr:4.965093698510193e-05 tokens_per_second_per_gpu:4384.4050775497 mfu:20.681986807445323 +Step 18274 | loss:0.7486816644668579 lr:4.964660087034661e-05 tokens_per_second_per_gpu:3965.2332714277636 mfu:18.704681879883314 +Step 18275 | loss:0.8316600918769836 lr:4.9642264758249257e-05 tokens_per_second_per_gpu:5343.265616879764 mfu:25.205095569942984 +Step 18276 | loss:0.7572398781776428 lr:4.963792864884249e-05 tokens_per_second_per_gpu:3540.8973041721747 mfu:16.70301672315712 +Step 18277 | loss:0.6858112812042236 lr:4.9633592542158905e-05 tokens_per_second_per_gpu:4014.6794017271545 mfu:18.937927712885298 +Step 18278 | loss:0.7323633432388306 lr:4.962925643823109e-05 tokens_per_second_per_gpu:4320.6391219556635 mfu:20.381191915314357 +Step 18279 | loss:0.7578635811805725 lr:4.9624920337091704e-05 tokens_per_second_per_gpu:5264.156539154006 mfu:24.831924552910117 +Step 18280 | loss:0.6211314797401428 lr:4.962058423877335e-05 tokens_per_second_per_gpu:5345.508562269829 mfu:25.215675926033864 +Step 18281 | loss:0.7103482484817505 lr:4.961624814330861e-05 tokens_per_second_per_gpu:4116.9452331469065 mfu:19.420333088042327 +Step 18282 | loss:0.7348730564117432 lr:4.961191205073012e-05 tokens_per_second_per_gpu:3737.421390921681 mfu:17.63005437082129 +Step 18283 | loss:0.7595352530479431 lr:4.960757596107051e-05 tokens_per_second_per_gpu:4933.5852464738255 mfu:23.272563364059256 +Step 18284 | loss:0.8179095387458801 lr:4.960323987436236e-05 tokens_per_second_per_gpu:5350.837232199481 mfu:25.240812171256874 +Step 18285 | loss:0.8262425065040588 lr:4.959890379063828e-05 tokens_per_second_per_gpu:5325.318347124939 mfu:25.120435236389497 +Step 18286 | loss:0.71630859375 lr:4.959456770993091e-05 tokens_per_second_per_gpu:3125.8492838266857 mfu:14.745164396693017 +Step 18287 | loss:0.8116492629051208 lr:4.959023163227284e-05 tokens_per_second_per_gpu:3831.5207796584723 mfu:18.073937242503995 +Step 18288 | loss:0.6242614984512329 lr:4.9585895557696686e-05 tokens_per_second_per_gpu:5335.012767123701 mfu:25.166165469562273 +Step 18289 | loss:0.9310072660446167 lr:4.9581559486235077e-05 tokens_per_second_per_gpu:5104.276327608778 mfu:24.07774231667433 +Step 18290 | loss:0.749110996723175 lr:4.95772234179206e-05 tokens_per_second_per_gpu:5329.614678779547 mfu:25.140701765834052 +Step 18291 | loss:0.881637692451477 lr:4.9572887352785914e-05 tokens_per_second_per_gpu:3646.0970421859392 mfu:17.199261836294188 +Step 18292 | loss:0.7856970429420471 lr:4.9568551290863556e-05 tokens_per_second_per_gpu:3632.0027978896296 mfu:17.132776881222448 +Step 18293 | loss:0.7273327708244324 lr:4.9564215232186195e-05 tokens_per_second_per_gpu:3963.4611633375075 mfu:18.696322543668757 +Step 18294 | loss:0.7364598512649536 lr:4.955987917678642e-05 tokens_per_second_per_gpu:5002.740582141888 mfu:23.59878088152246 +Step 18295 | loss:0.7343926429748535 lr:4.955554312469685e-05 tokens_per_second_per_gpu:4100.115774231696 mfu:19.3409456589375 +Step 18296 | loss:0.7284836769104004 lr:4.9551207075950104e-05 tokens_per_second_per_gpu:4655.304346557426 mfu:21.95986487953613 +Step 18297 | loss:0.6627939343452454 lr:4.9546871030578776e-05 tokens_per_second_per_gpu:3955.692200126112 mfu:18.659675018678794 +Step 18298 | loss:0.7835459113121033 lr:4.954253498861552e-05 tokens_per_second_per_gpu:5318.518889592117 mfu:25.088361035100096 +Step 18299 | loss:0.868018627166748 lr:4.953819895009288e-05 tokens_per_second_per_gpu:4712.836954035889 mfu:22.23125600508708 +Step 18300 | loss:0.808194637298584 lr:4.9533862915043526e-05 tokens_per_second_per_gpu:4431.369647815926 mfu:20.90352669837307 +Step 18301 | loss:0.7360818386077881 lr:4.952952688350003e-05 tokens_per_second_per_gpu:5386.71426668492 mfu:25.41005026417798 +Step 18302 | loss:0.8118600249290466 lr:4.9525190855495034e-05 tokens_per_second_per_gpu:4737.289671998271 mfu:22.34660364777953 +Step 18303 | loss:0.7535825371742249 lr:4.952085483106114e-05 tokens_per_second_per_gpu:5319.835273180731 mfu:25.094570641085788 +Step 18304 | loss:0.7685896158218384 lr:4.9516518810230946e-05 tokens_per_second_per_gpu:4593.134874899208 mfu:21.666600874518668 +Step 18305 | loss:0.8692591190338135 lr:4.95121827930371e-05 tokens_per_second_per_gpu:4076.93559532674 mfu:19.231600800096476 +Step 18306 | loss:0.7767429947853088 lr:4.9507846779512176e-05 tokens_per_second_per_gpu:4866.135972592648 mfu:22.954393631128948 +Step 18307 | loss:0.7073686122894287 lr:4.95035107696888e-05 tokens_per_second_per_gpu:4360.808217746652 mfu:20.570676393714912 +Step 18308 | loss:0.8275527358055115 lr:4.949917476359957e-05 tokens_per_second_per_gpu:5331.741614478706 mfu:25.150734884419634 +Step 18309 | loss:0.6718283295631409 lr:4.9494838761277126e-05 tokens_per_second_per_gpu:4681.0030483933915 mfu:22.081090040747046 +Step 18310 | loss:0.7423837184906006 lr:4.9490502762754045e-05 tokens_per_second_per_gpu:5007.7567922912185 mfu:23.622443200650554 +Step 18311 | loss:0.6392495632171631 lr:4.948616676806298e-05 tokens_per_second_per_gpu:5312.430462514233 mfu:25.059640885780944 +Step 18312 | loss:0.8044631481170654 lr:4.948183077723652e-05 tokens_per_second_per_gpu:4270.275252977684 mfu:20.143616952383105 +Step 18313 | loss:0.8979099988937378 lr:4.947749479030727e-05 tokens_per_second_per_gpu:3160.7056701224233 mfu:14.90958791796276 +Step 18314 | loss:0.6855583190917969 lr:4.947315880730785e-05 tokens_per_second_per_gpu:3901.216924962265 mfu:18.402705851290573 +Step 18315 | loss:0.738192081451416 lr:4.946882282827086e-05 tokens_per_second_per_gpu:4662.298364919489 mfu:21.992856857452566 +Step 18316 | loss:0.7991973161697388 lr:4.946448685322893e-05 tokens_per_second_per_gpu:4049.2299753895013 mfu:19.10090866378632 +Step 18317 | loss:0.6636901497840881 lr:4.946015088221465e-05 tokens_per_second_per_gpu:5386.267789183579 mfu:25.407944153627234 +Step 18318 | loss:0.7658127546310425 lr:4.9455814915260665e-05 tokens_per_second_per_gpu:4962.745758393197 mfu:23.410118474079304 +Step 18319 | loss:0.8301099538803101 lr:4.9451478952399555e-05 tokens_per_second_per_gpu:5327.324624222961 mfu:25.129899187767037 +Step 18320 | loss:0.8415508270263672 lr:4.944714299366395e-05 tokens_per_second_per_gpu:3276.144922651816 mfu:15.454134568111371 +Step 18321 | loss:0.7802343964576721 lr:4.944280703908644e-05 tokens_per_second_per_gpu:5397.551700495591 mfu:25.46117228851979 +Step 18322 | loss:0.7450587153434753 lr:4.9438471088699656e-05 tokens_per_second_per_gpu:4330.298902258742 mfu:20.42675874713245 +Step 18323 | loss:0.7564133405685425 lr:4.94341351425362e-05 tokens_per_second_per_gpu:5300.7561183529315 mfu:25.004571012523574 +Step 18324 | loss:0.6988203525543213 lr:4.942979920062869e-05 tokens_per_second_per_gpu:5263.458255161355 mfu:24.828630628177244 +Step 18325 | loss:0.7054900527000427 lr:4.942546326300972e-05 tokens_per_second_per_gpu:5334.207863021533 mfu:25.16236859958154 +Step 18326 | loss:0.778685986995697 lr:4.942112732971192e-05 tokens_per_second_per_gpu:4545.555366947375 mfu:21.442160217608727 +Step 18327 | loss:0.7168351411819458 lr:4.9416791400767926e-05 tokens_per_second_per_gpu:4583.034394000218 mfu:21.61895518279839 +Step 18328 | loss:0.8396317958831787 lr:4.941245547621028e-05 tokens_per_second_per_gpu:4703.920970530571 mfu:22.189197789668846 +Step 18329 | loss:0.8573825359344482 lr:4.9408119556071645e-05 tokens_per_second_per_gpu:4759.909777081079 mfu:22.45330654284303 +Step 18330 | loss:0.8669472932815552 lr:4.940378364038462e-05 tokens_per_second_per_gpu:4021.690418415721 mfu:18.970999874808065 +Step 18331 | loss:0.7013827562332153 lr:4.93994477291818e-05 tokens_per_second_per_gpu:4271.635957153696 mfu:20.150035626141424 +Step 18332 | loss:0.8320223093032837 lr:4.939511182249583e-05 tokens_per_second_per_gpu:3874.2584790693363 mfu:18.275538262428466 +Step 18333 | loss:0.8033964037895203 lr:4.9390775920359275e-05 tokens_per_second_per_gpu:4622.601293537626 mfu:21.80559899872553 +Step 18334 | loss:0.7725799083709717 lr:4.938644002280482e-05 tokens_per_second_per_gpu:4609.561149499538 mfu:21.744086414419876 +Step 18335 | loss:0.7359727621078491 lr:4.9382104129864984e-05 tokens_per_second_per_gpu:5200.677707186573 mfu:24.532484071914183 +Step 18336 | loss:0.7680492401123047 lr:4.9377768241572434e-05 tokens_per_second_per_gpu:5324.397009499543 mfu:25.116089129614096 +Step 18337 | loss:0.6910549402236938 lr:4.937343235795977e-05 tokens_per_second_per_gpu:5232.06474996261 mfu:24.68054210787046 +Step 18338 | loss:0.7071656584739685 lr:4.936909647905959e-05 tokens_per_second_per_gpu:4408.367662773922 mfu:20.795022410386604 +Step 18339 | loss:0.7694644927978516 lr:4.9364760604904537e-05 tokens_per_second_per_gpu:3964.5194699827266 mfu:18.701314756679988 +Step 18340 | loss:0.7878080606460571 lr:4.936042473552718e-05 tokens_per_second_per_gpu:4629.746592250035 mfu:21.839304591865634 +Step 18341 | loss:0.6855930089950562 lr:4.935608887096017e-05 tokens_per_second_per_gpu:3953.418825536591 mfu:18.648951122862325 +Step 18342 | loss:0.8202611804008484 lr:4.935175301123609e-05 tokens_per_second_per_gpu:3355.1555257864757 mfu:15.82684106369653 +Step 18343 | loss:0.7850592136383057 lr:4.934741715638756e-05 tokens_per_second_per_gpu:3747.096477902672 mfu:17.675693406850687 +Step 18344 | loss:0.7992791533470154 lr:4.9343081306447176e-05 tokens_per_second_per_gpu:3748.901089149359 mfu:17.684206065999735 +Step 18345 | loss:0.7048102617263794 lr:4.933874546144757e-05 tokens_per_second_per_gpu:5001.596358251644 mfu:23.593383382207453 +Step 18346 | loss:0.9434517621994019 lr:4.933440962142134e-05 tokens_per_second_per_gpu:4446.837848075845 mfu:20.97649283814621 +Step 18347 | loss:0.7573850750923157 lr:4.9330073786401106e-05 tokens_per_second_per_gpu:5310.523095342653 mfu:25.0506435094022 +Step 18348 | loss:0.673006534576416 lr:4.932573795641947e-05 tokens_per_second_per_gpu:5321.616168899796 mfu:25.10297143004476 +Step 18349 | loss:0.7410249710083008 lr:4.932140213150906e-05 tokens_per_second_per_gpu:5217.762449172157 mfu:24.613075714817747 +Step 18350 | loss:0.592825710773468 lr:4.9317066311702456e-05 tokens_per_second_per_gpu:3604.142483533898 mfu:17.001355024946673 +Step 18351 | loss:0.8543439507484436 lr:4.931273049703228e-05 tokens_per_second_per_gpu:5359.476272306554 mfu:25.28156399741417 +Step 18352 | loss:0.8759387731552124 lr:4.930839468753115e-05 tokens_per_second_per_gpu:4099.921990697158 mfu:19.340031548942203 +Step 18353 | loss:0.7794078588485718 lr:4.9304058883231666e-05 tokens_per_second_per_gpu:3891.0817757721784 mfu:18.354896623326173 +Step 18354 | loss:0.7744185924530029 lr:4.9299723084166446e-05 tokens_per_second_per_gpu:4144.42254722115 mfu:19.54994826664374 +Step 18355 | loss:0.8057910203933716 lr:4.9295387290368105e-05 tokens_per_second_per_gpu:5022.835972320081 mfu:23.69357426562011 +Step 18356 | loss:0.7447059154510498 lr:4.9291051501869256e-05 tokens_per_second_per_gpu:4279.243806663718 mfu:20.185923150313172 +Step 18357 | loss:0.7542306184768677 lr:4.928671571870248e-05 tokens_per_second_per_gpu:3155.507549707274 mfu:14.885067497073704 +Step 18358 | loss:0.6543956398963928 lr:4.928237994090042e-05 tokens_per_second_per_gpu:5317.603413484185 mfu:25.08404257810265 +Step 18359 | loss:0.8226478695869446 lr:4.9278044168495655e-05 tokens_per_second_per_gpu:5325.643878071451 mfu:25.1219708214055 +Step 18360 | loss:0.7752782106399536 lr:4.927370840152082e-05 tokens_per_second_per_gpu:5058.422684309662 mfu:23.861442857794124 +Step 18361 | loss:0.6735641956329346 lr:4.9269372640008516e-05 tokens_per_second_per_gpu:5169.075837254235 mfu:24.383412659608346 +Step 18362 | loss:0.7983506321907043 lr:4.926503688399134e-05 tokens_per_second_per_gpu:4733.4371413248555 mfu:22.328430603283515 +Step 18363 | loss:0.708361029624939 lr:4.9260701133501964e-05 tokens_per_second_per_gpu:4125.294710155013 mfu:19.459718995657237 +Step 18364 | loss:0.7982963919639587 lr:4.9256365388572904e-05 tokens_per_second_per_gpu:4108.0097808132705 mfu:19.378183034839193 +Step 18365 | loss:0.780939519405365 lr:4.925202964923683e-05 tokens_per_second_per_gpu:5335.261609812344 mfu:25.167339302981542 +Step 18366 | loss:0.8964361548423767 lr:4.9247693915526336e-05 tokens_per_second_per_gpu:3394.489918929909 mfu:16.012388107293607 +Step 18367 | loss:0.8888316750526428 lr:4.924335818747402e-05 tokens_per_second_per_gpu:4931.263841301928 mfu:23.26161289979085 +Step 18368 | loss:0.6942281126976013 lr:4.9239022465112514e-05 tokens_per_second_per_gpu:4634.9360573701515 mfu:21.863784184251273 +Step 18369 | loss:0.6270363926887512 lr:4.923468674847441e-05 tokens_per_second_per_gpu:5354.871711839912 mfu:25.259843500073995 +Step 18370 | loss:0.6382768154144287 lr:4.923035103759235e-05 tokens_per_second_per_gpu:4601.393462829382 mfu:21.705558042845702 +Step 18371 | loss:0.8874801397323608 lr:4.922601533249889e-05 tokens_per_second_per_gpu:4256.75663136068 mfu:20.079847307701012 +Step 18372 | loss:0.7426095604896545 lr:4.922167963322667e-05 tokens_per_second_per_gpu:4440.860646747726 mfu:20.948297359665766 +Step 18373 | loss:0.9094250798225403 lr:4.92173439398083e-05 tokens_per_second_per_gpu:3843.4731061735374 mfu:18.130318405952742 +Step 18374 | loss:0.8240759968757629 lr:4.9213008252276396e-05 tokens_per_second_per_gpu:5353.937899761928 mfu:25.255438549177374 +Step 18375 | loss:0.7042300701141357 lr:4.920867257066355e-05 tokens_per_second_per_gpu:5304.913313612659 mfu:25.024181211854362 +Step 18376 | loss:0.7449077367782593 lr:4.9204336895002365e-05 tokens_per_second_per_gpu:4111.968678563633 mfu:19.39685783098526 +Step 18377 | loss:0.7128872275352478 lr:4.9200001225325495e-05 tokens_per_second_per_gpu:4290.007979825992 mfu:20.236699591676942 +Step 18378 | loss:0.7769442796707153 lr:4.91956655616655e-05 tokens_per_second_per_gpu:4746.024754436389 mfu:22.38780852199888 +Step 18379 | loss:0.9064990282058716 lr:4.919132990405501e-05 tokens_per_second_per_gpu:4770.287188458408 mfu:22.502258562878847 +Step 18380 | loss:0.734264612197876 lr:4.918699425252662e-05 tokens_per_second_per_gpu:5233.690819443013 mfu:24.68821255504558 +Step 18381 | loss:0.6691802740097046 lr:4.918265860711296e-05 tokens_per_second_per_gpu:4593.589879938445 mfu:21.668747210920618 +Step 18382 | loss:0.6566071510314941 lr:4.917832296784663e-05 tokens_per_second_per_gpu:4710.684183393036 mfu:22.221101018664193 +Step 18383 | loss:0.7477827072143555 lr:4.9173987334760226e-05 tokens_per_second_per_gpu:4062.158065863676 mfu:19.161892672312693 +Step 18384 | loss:0.849227249622345 lr:4.9169651707886377e-05 tokens_per_second_per_gpu:4094.3109751627608 mfu:19.313563431328486 +Step 18385 | loss:0.8294356465339661 lr:4.91653160872577e-05 tokens_per_second_per_gpu:5345.552752527485 mfu:25.215884379018863 +Step 18386 | loss:0.7010143399238586 lr:4.916098047290677e-05 tokens_per_second_per_gpu:3800.976606586913 mfu:17.929855166752144 +Step 18387 | loss:0.9701082110404968 lr:4.9156644864866206e-05 tokens_per_second_per_gpu:5347.793123925857 mfu:25.226452593151762 +Step 18388 | loss:0.7162264585494995 lr:4.9152309263168636e-05 tokens_per_second_per_gpu:4983.65026701536 mfu:23.508728607927 +Step 18389 | loss:0.7012097239494324 lr:4.914797366784664e-05 tokens_per_second_per_gpu:4149.727253457049 mfu:19.574971471034686 +Step 18390 | loss:0.7811844944953918 lr:4.9143638078932865e-05 tokens_per_second_per_gpu:4757.37555763669 mfu:22.441352197341423 +Step 18391 | loss:0.7953691482543945 lr:4.913930249645988e-05 tokens_per_second_per_gpu:4408.136131037903 mfu:20.793930235684417 +Step 18392 | loss:0.6560356020927429 lr:4.913496692046033e-05 tokens_per_second_per_gpu:5331.675698909041 mfu:25.150423949431172 +Step 18393 | loss:0.7196019291877747 lr:4.9130631350966785e-05 tokens_per_second_per_gpu:5346.422674231255 mfu:25.219987948119496 +Step 18394 | loss:0.6840258240699768 lr:4.9126295788011885e-05 tokens_per_second_per_gpu:5321.982820584563 mfu:25.104700988599035 +Step 18395 | loss:0.7091354727745056 lr:4.912196023162822e-05 tokens_per_second_per_gpu:4875.7722382654865 mfu:22.999849540424187 +Step 18396 | loss:0.7254035472869873 lr:4.91176246818484e-05 tokens_per_second_per_gpu:4892.587531937865 mfu:23.079170149661735 +Step 18397 | loss:0.8201027512550354 lr:4.9113289138705046e-05 tokens_per_second_per_gpu:5283.473091455427 mfu:24.92304402585959 +Step 18398 | loss:0.7765747308731079 lr:4.910895360223075e-05 tokens_per_second_per_gpu:4472.442365438101 mfu:21.09727371512511 +Step 18399 | loss:0.6989132165908813 lr:4.9104618072458146e-05 tokens_per_second_per_gpu:3939.2768823290894 mfu:18.582241163888224 +Step 18400 | loss:0.7696466445922852 lr:4.91002825494198e-05 tokens_per_second_per_gpu:4063.938665971498 mfu:19.17029206682299 +Step 18401 | loss:0.7803584337234497 lr:4.909594703314836e-05 tokens_per_second_per_gpu:4628.259151619217 mfu:21.832288080626522 +Step 18402 | loss:0.81479811668396 lr:4.9091611523676405e-05 tokens_per_second_per_gpu:5300.7802503651055 mfu:25.004684847342347 +Step 18403 | loss:0.838363766670227 lr:4.908727602103655e-05 tokens_per_second_per_gpu:5098.107103598998 mfu:24.048641034442042 +Step 18404 | loss:0.738013744354248 lr:4.908294052526143e-05 tokens_per_second_per_gpu:3535.4656664776458 mfu:16.677394761419073 +Step 18405 | loss:0.757701575756073 lr:4.9078605036383604e-05 tokens_per_second_per_gpu:4395.077060241412 mfu:20.732328370630135 +Step 18406 | loss:0.7537669539451599 lr:4.907426955443575e-05 tokens_per_second_per_gpu:4396.787316386911 mfu:20.74039594066824 +Step 18407 | loss:0.7055230736732483 lr:4.90699340794504e-05 tokens_per_second_per_gpu:4705.4408976646755 mfu:22.196367545286726 +Step 18408 | loss:0.7864469885826111 lr:4.90655986114602e-05 tokens_per_second_per_gpu:5182.434671709608 mfu:24.44642856098653 +Step 18409 | loss:0.7590463161468506 lr:4.906126315049775e-05 tokens_per_second_per_gpu:4253.815009899037 mfu:20.065971177374063 +Step 18410 | loss:0.6958239674568176 lr:4.905692769659566e-05 tokens_per_second_per_gpu:5341.199038652755 mfu:25.19534716785158 +Step 18411 | loss:0.758743166923523 lr:4.905259224978654e-05 tokens_per_second_per_gpu:4365.922453328558 mfu:20.594801115533325 +Step 18412 | loss:0.6551426649093628 lr:4.904825681010299e-05 tokens_per_second_per_gpu:5349.759406243027 mfu:25.23572788232421 +Step 18413 | loss:0.8063440322875977 lr:4.9043921377577625e-05 tokens_per_second_per_gpu:5209.594424544335 mfu:24.574545749039608 +Step 18414 | loss:0.7313572764396667 lr:4.903958595224307e-05 tokens_per_second_per_gpu:4378.7816637158385 mfu:20.65546020493766 +Step 18415 | loss:0.7229794859886169 lr:4.903525053413189e-05 tokens_per_second_per_gpu:4296.592132679104 mfu:20.267758164057387 +Step 18416 | loss:0.8589633703231812 lr:4.9030915123276714e-05 tokens_per_second_per_gpu:4457.655954491817 mfu:21.02752369186964 +Step 18417 | loss:0.8192949891090393 lr:4.9026579719710145e-05 tokens_per_second_per_gpu:5335.687949591521 mfu:25.169350420460983 +Step 18418 | loss:0.7681626677513123 lr:4.9022244323464806e-05 tokens_per_second_per_gpu:4118.926441944723 mfu:19.4296787879707 +Step 18419 | loss:0.7932851314544678 lr:4.901790893457327e-05 tokens_per_second_per_gpu:5000.279326629443 mfu:23.587170719336765 +Step 18420 | loss:0.7075815796852112 lr:4.901357355306819e-05 tokens_per_second_per_gpu:3951.528013111786 mfu:18.640031838048657 +Step 18421 | loss:0.9187759160995483 lr:4.900923817898216e-05 tokens_per_second_per_gpu:3365.647235546669 mfu:15.876332248705634 +Step 18422 | loss:0.8270455002784729 lr:4.9004902812347756e-05 tokens_per_second_per_gpu:3982.2178562922522 mfu:18.784801064558103 +Step 18423 | loss:0.7383325695991516 lr:4.90005674531976e-05 tokens_per_second_per_gpu:4480.678196802647 mfu:21.136123536849404 +Step 18424 | loss:0.8048855662345886 lr:4.899623210156431e-05 tokens_per_second_per_gpu:4347.552272800166 mfu:20.50814583970491 +Step 18425 | loss:0.8019142150878906 lr:4.899189675748048e-05 tokens_per_second_per_gpu:4000.8869426160086 mfu:18.872866330021843 +Step 18426 | loss:0.76592618227005 lr:4.8987561420978734e-05 tokens_per_second_per_gpu:4932.837965717725 mfu:23.269038313233484 +Step 18427 | loss:0.9334537982940674 lr:4.8983226092091665e-05 tokens_per_second_per_gpu:4952.075563693725 mfu:23.359785345158947 +Step 18428 | loss:0.7896713018417358 lr:4.8978890770851894e-05 tokens_per_second_per_gpu:5316.432188652977 mfu:25.078517710742283 +Step 18429 | loss:0.7764574289321899 lr:4.8974555457291994e-05 tokens_per_second_per_gpu:4784.5493538174405 mfu:22.569535630253736 +Step 18430 | loss:0.7438380122184753 lr:4.897022015144461e-05 tokens_per_second_per_gpu:5325.202715973473 mfu:25.11988978451932 +Step 18431 | loss:0.7952507734298706 lr:4.896588485334232e-05 tokens_per_second_per_gpu:4576.807471166959 mfu:21.58958172537132 +Step 18432 | loss:0.6091500520706177 lr:4.896154956301774e-05 tokens_per_second_per_gpu:3747.7105742512226 mfu:17.678590204102438 +Step 18433 | loss:0.7651985883712769 lr:4.895721428050348e-05 tokens_per_second_per_gpu:3912.491775251285 mfu:18.455891243791704 +Step 18434 | loss:0.734027624130249 lr:4.895287900583216e-05 tokens_per_second_per_gpu:3593.2515708029173 mfu:16.949980731413316 +Step 18435 | loss:0.836061954498291 lr:4.894854373903637e-05 tokens_per_second_per_gpu:4091.275059272733 mfu:19.29924249809456 +Step 18436 | loss:0.6931359767913818 lr:4.89442084801487e-05 tokens_per_second_per_gpu:4100.286065045067 mfu:19.341748949758777 +Step 18437 | loss:0.7032700181007385 lr:4.893987322920178e-05 tokens_per_second_per_gpu:4752.72817193674 mfu:22.419429686068995 +Step 18438 | loss:0.8601404428482056 lr:4.893553798622821e-05 tokens_per_second_per_gpu:5354.035803856846 mfu:25.255900379497263 +Step 18439 | loss:0.7161076664924622 lr:4.893120275126058e-05 tokens_per_second_per_gpu:3742.9901174377737 mfu:17.65632300392042 +Step 18440 | loss:0.7976035475730896 lr:4.892686752433152e-05 tokens_per_second_per_gpu:5320.158882377664 mfu:25.09609716088114 +Step 18441 | loss:0.7702328562736511 lr:4.892253230547362e-05 tokens_per_second_per_gpu:5344.256046543909 mfu:25.209767595653485 +Step 18442 | loss:0.8655421137809753 lr:4.891819709471951e-05 tokens_per_second_per_gpu:4043.5214366943474 mfu:19.073980512784146 +Step 18443 | loss:0.8029405474662781 lr:4.891386189210178e-05 tokens_per_second_per_gpu:4247.974250254308 mfu:20.038419317593522 +Step 18444 | loss:0.6799450516700745 lr:4.890952669765302e-05 tokens_per_second_per_gpu:4569.490538562506 mfu:21.555066505878685 +Step 18445 | loss:0.7456490993499756 lr:4.890519151140585e-05 tokens_per_second_per_gpu:4008.186778547698 mfu:18.90730090159206 +Step 18446 | loss:0.8684602975845337 lr:4.890085633339287e-05 tokens_per_second_per_gpu:3345.888168012526 mfu:15.783125355902527 +Step 18447 | loss:0.7632031440734863 lr:4.88965211636467e-05 tokens_per_second_per_gpu:4430.9500356873305 mfu:20.901547316369275 +Step 18448 | loss:0.7822780609130859 lr:4.889218600219992e-05 tokens_per_second_per_gpu:5358.6151133258745 mfu:25.277501763573433 +Step 18449 | loss:0.7900786399841309 lr:4.888785084908516e-05 tokens_per_second_per_gpu:3932.577614826772 mfu:18.550639576066892 +Step 18450 | loss:0.6705430746078491 lr:4.888351570433503e-05 tokens_per_second_per_gpu:4165.927491562467 mfu:19.651390758223584 +Step 18451 | loss:0.7959632873535156 lr:4.8879180567982107e-05 tokens_per_second_per_gpu:5334.836630774518 mfu:25.165334604351216 +Step 18452 | loss:0.8614412546157837 lr:4.8874845440059005e-05 tokens_per_second_per_gpu:3883.1549219781946 mfu:18.317504301519232 +Step 18453 | loss:0.7851576805114746 lr:4.887051032059834e-05 tokens_per_second_per_gpu:4392.710218382526 mfu:20.721163573756908 +Step 18454 | loss:0.7507335543632507 lr:4.8866175209632714e-05 tokens_per_second_per_gpu:4447.9890355409825 mfu:20.981923185832102 +Step 18455 | loss:0.7382259368896484 lr:4.886184010719471e-05 tokens_per_second_per_gpu:4845.972775054481 mfu:22.859280388145173 +Step 18456 | loss:0.6803332567214966 lr:4.8857505013316975e-05 tokens_per_second_per_gpu:3941.7002117768097 mfu:18.59367242235576 +Step 18457 | loss:0.7879486680030823 lr:4.8853169928032096e-05 tokens_per_second_per_gpu:5332.172426239055 mfu:25.15276709699727 +Step 18458 | loss:0.8630723357200623 lr:4.884883485137266e-05 tokens_per_second_per_gpu:4125.159598093924 mfu:19.459081649012585 +Step 18459 | loss:0.8859665989875793 lr:4.884449978337127e-05 tokens_per_second_per_gpu:5343.117212171049 mfu:25.20439551961152 +Step 18460 | loss:0.7097249627113342 lr:4.8840164724060564e-05 tokens_per_second_per_gpu:3774.565174897855 mfu:17.805267937219938 +Step 18461 | loss:0.744583785533905 lr:4.883582967347311e-05 tokens_per_second_per_gpu:3580.7904485612285 mfu:16.891199491712577 +Step 18462 | loss:0.8505041003227234 lr:4.883149463164154e-05 tokens_per_second_per_gpu:4454.881382105842 mfu:21.014435560534128 +Step 18463 | loss:0.763922929763794 lr:4.882715959859845e-05 tokens_per_second_per_gpu:5006.487254931381 mfu:23.616454576318795 +Step 18464 | loss:0.7132718563079834 lr:4.882282457437645e-05 tokens_per_second_per_gpu:5022.9269250982825 mfu:23.694003305393032 +Step 18465 | loss:0.6414888501167297 lr:4.881848955900812e-05 tokens_per_second_per_gpu:5316.426422399418 mfu:25.078490510340416 +Step 18466 | loss:0.7436213493347168 lr:4.8814154552526095e-05 tokens_per_second_per_gpu:5350.5712761039185 mfu:25.239557610977318 +Step 18467 | loss:0.9002858400344849 lr:4.8809819554962956e-05 tokens_per_second_per_gpu:3849.2326220908813 mfu:18.1574870252096 +Step 18468 | loss:0.7366568446159363 lr:4.880548456635131e-05 tokens_per_second_per_gpu:5079.904108555092 mfu:23.962774401068728 +Step 18469 | loss:0.8396415114402771 lr:4.880114958672377e-05 tokens_per_second_per_gpu:5049.912249650587 mfu:23.821297685477386 +Step 18470 | loss:0.8361560702323914 lr:4.8796814616112947e-05 tokens_per_second_per_gpu:3490.8252742203294 mfu:16.46681841470501 +Step 18471 | loss:0.8651013970375061 lr:4.8792479654551446e-05 tokens_per_second_per_gpu:3630.6094818060806 mfu:17.12620437153211 +Step 18472 | loss:0.7095127105712891 lr:4.878814470207183e-05 tokens_per_second_per_gpu:4705.9706035582785 mfu:22.198866258363996 +Step 18473 | loss:0.768017590045929 lr:4.878380975870675e-05 tokens_per_second_per_gpu:5259.063207337954 mfu:24.80789843772178 +Step 18474 | loss:0.8073722720146179 lr:4.8779474824488784e-05 tokens_per_second_per_gpu:4280.324879285291 mfu:20.191022754319143 +Step 18475 | loss:0.7238238453865051 lr:4.877513989945054e-05 tokens_per_second_per_gpu:5324.115436883229 mfu:25.114760903541704 +Step 18476 | loss:0.7406195998191833 lr:4.8770804983624635e-05 tokens_per_second_per_gpu:5201.439878377398 mfu:24.53607936346112 +Step 18477 | loss:0.7996435761451721 lr:4.876647007704365e-05 tokens_per_second_per_gpu:4553.3418630777605 mfu:21.478890448368986 +Step 18478 | loss:0.9143487811088562 lr:4.876213517974021e-05 tokens_per_second_per_gpu:4709.977098328868 mfu:22.217765577775364 +Step 18479 | loss:0.7947052121162415 lr:4.875780029174692e-05 tokens_per_second_per_gpu:3236.177281193113 mfu:15.265600384167863 +Step 18480 | loss:0.7421013712882996 lr:4.875346541309637e-05 tokens_per_second_per_gpu:4700.869888239588 mfu:22.174805313933962 +Step 18481 | loss:0.7915338277816772 lr:4.8749130543821145e-05 tokens_per_second_per_gpu:5367.129323631719 mfu:25.317664746260174 +Step 18482 | loss:0.741956353187561 lr:4.8744795683953885e-05 tokens_per_second_per_gpu:4287.461250659998 mfu:20.224686235684146 +Step 18483 | loss:0.7423803806304932 lr:4.874046083352717e-05 tokens_per_second_per_gpu:4468.618446846506 mfu:21.07923563870077 +Step 18484 | loss:0.7803131341934204 lr:4.873612599257361e-05 tokens_per_second_per_gpu:4512.539181795483 mfu:21.286417239105198 +Step 18485 | loss:0.6922755241394043 lr:4.8731791161125814e-05 tokens_per_second_per_gpu:4750.513251527517 mfu:22.4089815285943 +Step 18486 | loss:0.8121118545532227 lr:4.872745633921639e-05 tokens_per_second_per_gpu:5365.165315988739 mfu:25.308400187113982 +Step 18487 | loss:0.7050731182098389 lr:4.872312152687791e-05 tokens_per_second_per_gpu:4134.940658309256 mfu:19.505220578869267 +Step 18488 | loss:0.8491600751876831 lr:4.8718786724142995e-05 tokens_per_second_per_gpu:2923.3756290718025 mfu:13.790061621646506 +Step 18489 | loss:0.8649708032608032 lr:4.871445193104426e-05 tokens_per_second_per_gpu:3540.2341093634523 mfu:16.699888320091674 +Step 18490 | loss:0.6159479022026062 lr:4.871011714761429e-05 tokens_per_second_per_gpu:5337.577281876332 mfu:25.17826272320176 +Step 18491 | loss:0.811525821685791 lr:4.8705782373885684e-05 tokens_per_second_per_gpu:4287.061529730353 mfu:20.22280068385959 +Step 18492 | loss:0.8803043961524963 lr:4.870144760989107e-05 tokens_per_second_per_gpu:5322.700034562052 mfu:25.10808420930017 +Step 18493 | loss:0.7088342905044556 lr:4.869711285566304e-05 tokens_per_second_per_gpu:3923.9391460028146 mfu:18.50989044474952 +Step 18494 | loss:0.7758362889289856 lr:4.869277811123418e-05 tokens_per_second_per_gpu:5315.06616109131 mfu:25.07207392566556 +Step 18495 | loss:0.8012508153915405 lr:4.868844337663709e-05 tokens_per_second_per_gpu:5314.7413551648415 mfu:25.070541760692805 +Step 18496 | loss:0.7031165361404419 lr:4.868410865190439e-05 tokens_per_second_per_gpu:5290.244559160471 mfu:24.954986194355804 +Step 18497 | loss:0.9078532457351685 lr:4.867977393706868e-05 tokens_per_second_per_gpu:5361.465026139191 mfu:25.29094528855911 +Step 18498 | loss:0.7139527797698975 lr:4.867543923216256e-05 tokens_per_second_per_gpu:4564.464749240766 mfu:21.531359000159956 +Step 18499 | loss:0.6927680969238281 lr:4.867110453721864e-05 tokens_per_second_per_gpu:5136.093440163976 mfu:24.227829065156378 +Step 18500 | loss:0.7806055545806885 lr:4.866676985226951e-05 tokens_per_second_per_gpu:5076.514394232376 mfu:23.946784540263575 +Step 18501 | loss:0.7323285341262817 lr:4.866243517734776e-05 tokens_per_second_per_gpu:5275.976176791192 mfu:24.887679800282854 +Step 18502 | loss:0.738239049911499 lr:4.865810051248602e-05 tokens_per_second_per_gpu:5062.38355918162 mfu:23.88012698036009 +Step 18503 | loss:0.8210119605064392 lr:4.865376585771687e-05 tokens_per_second_per_gpu:4110.378836171217 mfu:19.389358273165147 +Step 18504 | loss:0.903226912021637 lr:4.8649431213072914e-05 tokens_per_second_per_gpu:4173.436104951778 mfu:19.686810168682513 +Step 18505 | loss:0.7195033431053162 lr:4.864509657858676e-05 tokens_per_second_per_gpu:5339.101563085878 mfu:25.18545301773639 +Step 18506 | loss:0.7044165730476379 lr:4.864076195429101e-05 tokens_per_second_per_gpu:5316.10825673398 mfu:25.076989668611887 +Step 18507 | loss:0.7080056667327881 lr:4.863642734021826e-05 tokens_per_second_per_gpu:4610.533567476376 mfu:21.748673475928236 +Step 18508 | loss:0.7314819097518921 lr:4.863209273640113e-05 tokens_per_second_per_gpu:4124.5133604960565 mfu:19.456033236003943 +Step 18509 | loss:0.9176910519599915 lr:4.86277581428722e-05 tokens_per_second_per_gpu:5352.008841816948 mfu:25.24633885371968 +Step 18510 | loss:0.692976713180542 lr:4.8623423559664066e-05 tokens_per_second_per_gpu:4559.513311633564 mfu:21.50800222416441 +Step 18511 | loss:0.6882319450378418 lr:4.861908898680934e-05 tokens_per_second_per_gpu:5382.988113648311 mfu:25.392473364557027 +Step 18512 | loss:0.7652708888053894 lr:4.861475442434062e-05 tokens_per_second_per_gpu:4180.226326184243 mfu:19.718840800767254 +Step 18513 | loss:0.837945282459259 lr:4.861041987229051e-05 tokens_per_second_per_gpu:4681.716871643037 mfu:22.084457266806428 +Step 18514 | loss:0.8324360251426697 lr:4.860608533069161e-05 tokens_per_second_per_gpu:5323.6564683958395 mfu:25.11259587087859 +Step 18515 | loss:0.8071228265762329 lr:4.860175079957653e-05 tokens_per_second_per_gpu:3688.0303197657677 mfu:17.397068261193148 +Step 18516 | loss:0.7679201364517212 lr:4.859741627897786e-05 tokens_per_second_per_gpu:4690.867721924004 mfu:22.127623388877804 +Step 18517 | loss:0.6801843643188477 lr:4.859308176892819e-05 tokens_per_second_per_gpu:5373.590643693578 mfu:25.348143895406842 +Step 18518 | loss:0.7446046471595764 lr:4.8588747269460133e-05 tokens_per_second_per_gpu:5349.495901343027 mfu:25.23448488475234 +Step 18519 | loss:0.6625407934188843 lr:4.8584412780606294e-05 tokens_per_second_per_gpu:5007.609361808245 mfu:23.621747745908138 +Step 18520 | loss:0.8213818073272705 lr:4.858007830239926e-05 tokens_per_second_per_gpu:4038.586536004945 mfu:19.050701744251356 +Step 18521 | loss:0.7023048400878906 lr:4.857574383487164e-05 tokens_per_second_per_gpu:4257.140086480321 mfu:20.081656130924543 +Step 18522 | loss:0.7498549222946167 lr:4.857140937805605e-05 tokens_per_second_per_gpu:3254.0701001454527 mfu:15.350003863995761 +Step 18523 | loss:0.7090862989425659 lr:4.856707493198506e-05 tokens_per_second_per_gpu:3775.53420901948 mfu:17.809839036505917 +Step 18524 | loss:0.7348125576972961 lr:4.856274049669127e-05 tokens_per_second_per_gpu:5347.378728478217 mfu:25.224497819122092 +Step 18525 | loss:0.7334024906158447 lr:4.8558406072207305e-05 tokens_per_second_per_gpu:4923.611024673072 mfu:23.225513258047386 +Step 18526 | loss:0.6698253154754639 lr:4.855407165856575e-05 tokens_per_second_per_gpu:4861.570401935955 mfu:22.932857055374644 +Step 18527 | loss:0.8297628164291382 lr:4.854973725579919e-05 tokens_per_second_per_gpu:3850.7640451026487 mfu:18.164711008843817 +Step 18528 | loss:0.6924991607666016 lr:4.854540286394026e-05 tokens_per_second_per_gpu:4279.220344828369 mfu:20.18581247683283 +Step 18529 | loss:0.8428823351860046 lr:4.854106848302155e-05 tokens_per_second_per_gpu:4254.843482429005 mfu:20.07082265777343 +Step 18530 | loss:0.634156346321106 lr:4.853673411307564e-05 tokens_per_second_per_gpu:5299.820545479375 mfu:25.000157755653817 +Step 18531 | loss:0.723150372505188 lr:4.8532399754135124e-05 tokens_per_second_per_gpu:4232.864044726858 mfu:19.967141900053633 +Step 18532 | loss:0.7866675853729248 lr:4.852806540623263e-05 tokens_per_second_per_gpu:5343.745402720365 mfu:25.207358801613005 +Step 18533 | loss:0.6197278499603271 lr:4.852373106940073e-05 tokens_per_second_per_gpu:5021.462114947509 mfu:23.68709354598945 +Step 18534 | loss:0.8627763986587524 lr:4.8519396743672055e-05 tokens_per_second_per_gpu:3495.6376261219425 mfu:16.48951909969682 +Step 18535 | loss:0.5353420972824097 lr:4.8515062429079185e-05 tokens_per_second_per_gpu:5334.885549990141 mfu:25.16556536463779 +Step 18536 | loss:0.7017948627471924 lr:4.85107281256547e-05 tokens_per_second_per_gpu:4238.507305640416 mfu:19.99376212462249 +Step 18537 | loss:0.7201516628265381 lr:4.8506393833431266e-05 tokens_per_second_per_gpu:4635.04728560305 mfu:21.86430886680335 +Step 18538 | loss:0.9084007740020752 lr:4.850205955244139e-05 tokens_per_second_per_gpu:4354.275497812966 mfu:20.539860439190758 +Step 18539 | loss:0.8234084844589233 lr:4.8497725282717734e-05 tokens_per_second_per_gpu:3200.5861424426994 mfu:15.097710910207718 +Step 18540 | loss:0.6366643309593201 lr:4.849339102429286e-05 tokens_per_second_per_gpu:4571.059800648233 mfu:21.56246898288078 +Step 18541 | loss:0.7885952591896057 lr:4.84890567771994e-05 tokens_per_second_per_gpu:4951.970068088202 mfu:23.359287704388294 +Step 18542 | loss:0.6628626585006714 lr:4.848472254146993e-05 tokens_per_second_per_gpu:4738.381705223633 mfu:22.35175495482364 +Step 18543 | loss:0.631913959980011 lr:4.8480388317137055e-05 tokens_per_second_per_gpu:4178.231745843448 mfu:19.709432024989226 +Step 18544 | loss:0.7421590089797974 lr:4.847605410423339e-05 tokens_per_second_per_gpu:3689.714654628979 mfu:17.405013556120075 +Step 18545 | loss:0.6764482855796814 lr:4.8471719902791504e-05 tokens_per_second_per_gpu:4392.231424718125 mfu:20.7189050223513 +Step 18546 | loss:0.7307153940200806 lr:4.846738571284401e-05 tokens_per_second_per_gpu:5030.380078699873 mfu:23.729161102570863 +Step 18547 | loss:0.8899504542350769 lr:4.846305153442349e-05 tokens_per_second_per_gpu:5229.74329596781 mfu:24.66959141329609 +Step 18548 | loss:0.7644678950309753 lr:4.8458717367562565e-05 tokens_per_second_per_gpu:4934.670377015483 mfu:23.277682109966616 +Step 18549 | loss:0.8309890627861023 lr:4.845438321229382e-05 tokens_per_second_per_gpu:5371.809302164797 mfu:25.339741003487383 +Step 18550 | loss:0.8004253506660461 lr:4.845004906864986e-05 tokens_per_second_per_gpu:5312.686777938416 mfu:25.060849969368732 +Step 18551 | loss:0.7176420092582703 lr:4.844571493666329e-05 tokens_per_second_per_gpu:5324.203484343693 mfu:25.115176238435158 +Step 18552 | loss:0.7175378203392029 lr:4.8441380816366694e-05 tokens_per_second_per_gpu:3861.8644089676477 mfu:18.217073319112387 +Step 18553 | loss:0.8497927188873291 lr:4.8437046707792647e-05 tokens_per_second_per_gpu:5354.360049459349 mfu:25.257429901326276 +Step 18554 | loss:0.5934663414955139 lr:4.84327126109738e-05 tokens_per_second_per_gpu:5097.101366879275 mfu:24.04389680273873 +Step 18555 | loss:0.7156901955604553 lr:4.8428378525942705e-05 tokens_per_second_per_gpu:4873.891987493677 mfu:22.990980076729702 +Step 18556 | loss:0.7103418707847595 lr:4.842404445273197e-05 tokens_per_second_per_gpu:4630.844730219196 mfu:21.844484696027827 +Step 18557 | loss:0.6888055801391602 lr:4.841971039137421e-05 tokens_per_second_per_gpu:5206.17766030026 mfu:24.55842828914021 +Step 18558 | loss:0.8138522505760193 lr:4.841537634190202e-05 tokens_per_second_per_gpu:5331.177626257257 mfu:25.14807445575275 +Step 18559 | loss:0.759082019329071 lr:4.8411042304347973e-05 tokens_per_second_per_gpu:4417.002194878931 mfu:20.83575297152902 +Step 18560 | loss:0.7144247889518738 lr:4.840670827874467e-05 tokens_per_second_per_gpu:4552.823859318748 mfu:21.476446936257172 +Step 18561 | loss:0.6371477842330933 lr:4.840237426512473e-05 tokens_per_second_per_gpu:4845.5804960700525 mfu:22.85742994124593 +Step 18562 | loss:0.7031094431877136 lr:4.839804026352074e-05 tokens_per_second_per_gpu:5335.914125737495 mfu:25.170417331931176 +Step 18563 | loss:0.8020097613334656 lr:4.839370627396528e-05 tokens_per_second_per_gpu:5346.022533991809 mfu:25.218100418339063 +Step 18564 | loss:0.689470112323761 lr:4.838937229649097e-05 tokens_per_second_per_gpu:5285.26579871183 mfu:24.931500531855438 +Step 18565 | loss:0.8060157299041748 lr:4.838503833113038e-05 tokens_per_second_per_gpu:5037.797742624576 mfu:23.764151488887656 +Step 18566 | loss:0.7135003805160522 lr:4.838070437791617e-05 tokens_per_second_per_gpu:5349.510253704157 mfu:25.23455258728855 +Step 18567 | loss:0.8315514326095581 lr:4.837637043688085e-05 tokens_per_second_per_gpu:4781.711201224844 mfu:22.55614758023569 +Step 18568 | loss:0.792212188243866 lr:4.837203650805706e-05 tokens_per_second_per_gpu:5328.548260895431 mfu:25.135671290725025 +Step 18569 | loss:0.7991572022438049 lr:4.836770259147738e-05 tokens_per_second_per_gpu:4558.974795259386 mfu:21.50546195054705 +Step 18570 | loss:0.7923322319984436 lr:4.8363368687174436e-05 tokens_per_second_per_gpu:3607.2649132283 mfu:17.016084058556608 +Step 18571 | loss:0.6655932664871216 lr:4.8359034795180804e-05 tokens_per_second_per_gpu:5321.692327204554 mfu:25.103330681762248 +Step 18572 | loss:0.8206024765968323 lr:4.835470091552906e-05 tokens_per_second_per_gpu:5306.548923639953 mfu:25.03189666341695 +Step 18573 | loss:0.6535612940788269 lr:4.835036704825186e-05 tokens_per_second_per_gpu:5363.499601679305 mfu:25.30054272851621 +Step 18574 | loss:0.8123573660850525 lr:4.8346033193381726e-05 tokens_per_second_per_gpu:3288.6529395263406 mfu:15.513137017796243 +Step 18575 | loss:0.752341628074646 lr:4.8341699350951306e-05 tokens_per_second_per_gpu:4755.514430537266 mfu:22.43257294327241 +Step 18576 | loss:0.7070099115371704 lr:4.8337365520993154e-05 tokens_per_second_per_gpu:4776.762080068501 mfu:22.53280173133408 +Step 18577 | loss:0.7338141202926636 lr:4.8333031703539905e-05 tokens_per_second_per_gpu:4386.791947683672 mfu:20.693246081109308 +Step 18578 | loss:0.6982240676879883 lr:4.8328697898624134e-05 tokens_per_second_per_gpu:3913.6300343094476 mfu:18.46126060597608 +Step 18579 | loss:0.7164101004600525 lr:4.832436410627843e-05 tokens_per_second_per_gpu:5228.3367970824675 mfu:24.662956718080476 +Step 18580 | loss:0.6666162014007568 lr:4.832003032653542e-05 tokens_per_second_per_gpu:4005.027643312446 mfu:18.892398721683325 +Step 18581 | loss:0.6931424140930176 lr:4.831569655942767e-05 tokens_per_second_per_gpu:4258.154885101929 mfu:20.08644310916517 +Step 18582 | loss:0.7395963668823242 lr:4.831136280498777e-05 tokens_per_second_per_gpu:5331.683713658871 mfu:25.150461756373577 +Step 18583 | loss:0.7992854714393616 lr:4.8307029063248326e-05 tokens_per_second_per_gpu:5331.630209196749 mfu:25.15020936669694 +Step 18584 | loss:0.7589079141616821 lr:4.8302695334241946e-05 tokens_per_second_per_gpu:3868.014185220941 mfu:18.246082862959195 +Step 18585 | loss:0.7460656762123108 lr:4.8298361618001186e-05 tokens_per_second_per_gpu:3303.14885427219 mfu:15.581516720910503 +Step 18586 | loss:0.7743824124336243 lr:4.829402791455868e-05 tokens_per_second_per_gpu:4233.925625403707 mfu:19.972149557230672 +Step 18587 | loss:0.7774850726127625 lr:4.8289694223947024e-05 tokens_per_second_per_gpu:5365.913524577498 mfu:25.311929614684704 +Step 18588 | loss:0.8671131134033203 lr:4.828536054619879e-05 tokens_per_second_per_gpu:4611.313628002665 mfu:21.752353154522787 +Step 18589 | loss:0.7990610003471375 lr:4.828102688134656e-05 tokens_per_second_per_gpu:4019.3692726902586 mfu:18.96005063439276 +Step 18590 | loss:0.8714396357536316 lr:4.827669322942295e-05 tokens_per_second_per_gpu:3740.6552994883873 mfu:17.64530926929214 +Step 18591 | loss:0.79032963514328 lr:4.827235959046056e-05 tokens_per_second_per_gpu:5368.196733092217 mfu:25.322699898804178 +Step 18592 | loss:0.682634174823761 lr:4.8268025964491956e-05 tokens_per_second_per_gpu:4259.564249170262 mfu:20.0930913199378 +Step 18593 | loss:0.7544560432434082 lr:4.8263692351549764e-05 tokens_per_second_per_gpu:4521.4606703925565 mfu:21.328501422980505 +Step 18594 | loss:0.7676366567611694 lr:4.825935875166658e-05 tokens_per_second_per_gpu:5317.681247623767 mfu:25.084409735018934 +Step 18595 | loss:0.753510057926178 lr:4.825502516487497e-05 tokens_per_second_per_gpu:4632.258035698507 mfu:21.851151499108514 +Step 18596 | loss:0.9117364287376404 lr:4.825069159120752e-05 tokens_per_second_per_gpu:4349.090814544693 mfu:20.515403403616133 +Step 18597 | loss:0.6755087971687317 lr:4.824635803069685e-05 tokens_per_second_per_gpu:4410.633978702274 mfu:20.80571300929401 +Step 18598 | loss:0.6782311797142029 lr:4.8242024483375556e-05 tokens_per_second_per_gpu:5372.555008938137 mfu:25.34325862956792 +Step 18599 | loss:0.6402031779289246 lr:4.823769094927621e-05 tokens_per_second_per_gpu:3536.446279227184 mfu:16.682020479068434 +Step 18600 | loss:0.5898181200027466 lr:4.823335742843141e-05 tokens_per_second_per_gpu:5311.546425301874 mfu:25.055470731417937 +Step 18601 | loss:0.76165372133255 lr:4.8229023920873754e-05 tokens_per_second_per_gpu:4658.003987060173 mfu:21.972599544393756 +Step 18602 | loss:0.8907480835914612 lr:4.8224690426635864e-05 tokens_per_second_per_gpu:3241.3420750290707 mfu:15.28996359789677 +Step 18603 | loss:0.7475290894508362 lr:4.8220356945750276e-05 tokens_per_second_per_gpu:5305.688164722926 mfu:25.027836316744718 +Step 18604 | loss:0.7442291975021362 lr:4.8216023478249625e-05 tokens_per_second_per_gpu:3673.576521422925 mfu:17.328887228337667 +Step 18605 | loss:0.8715041875839233 lr:4.821169002416648e-05 tokens_per_second_per_gpu:5323.260168281337 mfu:25.11072645562271 +Step 18606 | loss:0.7160471677780151 lr:4.8207356583533433e-05 tokens_per_second_per_gpu:4866.242893330993 mfu:22.95489799449428 +Step 18607 | loss:0.8350726962089539 lr:4.82030231563831e-05 tokens_per_second_per_gpu:5334.10217709172 mfu:25.161870061018735 +Step 18608 | loss:0.7124651670455933 lr:4.8198689742748046e-05 tokens_per_second_per_gpu:5357.096935561685 mfu:25.270340260031915 +Step 18609 | loss:0.8089370131492615 lr:4.819435634266091e-05 tokens_per_second_per_gpu:5335.382401511939 mfu:25.16790909803772 +Step 18610 | loss:0.7440975904464722 lr:4.819002295615422e-05 tokens_per_second_per_gpu:5335.557525290003 mfu:25.168735186778044 +Step 18611 | loss:0.5339632034301758 lr:4.8185689583260604e-05 tokens_per_second_per_gpu:5263.923039274133 mfu:24.830823094137564 +Step 18612 | loss:0.7383884787559509 lr:4.818135622401264e-05 tokens_per_second_per_gpu:5306.301073110222 mfu:25.03072751018055 +Step 18613 | loss:0.7208290696144104 lr:4.817702287844294e-05 tokens_per_second_per_gpu:4617.745015505496 mfu:21.7826911066815 +Step 18614 | loss:0.7395319938659668 lr:4.817268954658408e-05 tokens_per_second_per_gpu:3972.982149758227 mfu:18.741234661063082 +Step 18615 | loss:0.6839938163757324 lr:4.8168356228468656e-05 tokens_per_second_per_gpu:3352.8357057998596 mfu:15.81589807701756 +Step 18616 | loss:0.873259961605072 lr:4.8164022924129265e-05 tokens_per_second_per_gpu:3994.8701948183098 mfu:18.84448430409707 +Step 18617 | loss:0.7001133561134338 lr:4.815968963359849e-05 tokens_per_second_per_gpu:4461.020055732661 mfu:21.043392731397553 +Step 18618 | loss:0.7266120910644531 lr:4.815535635690892e-05 tokens_per_second_per_gpu:4974.528508041379 mfu:23.46569971451417 +Step 18619 | loss:0.8670960664749146 lr:4.815102309409314e-05 tokens_per_second_per_gpu:5223.363143500351 mfu:24.639495145540824 +Step 18620 | loss:0.7825208902359009 lr:4.814668984518376e-05 tokens_per_second_per_gpu:5056.815521366678 mfu:23.853861595981797 +Step 18621 | loss:0.8416221737861633 lr:4.8142356610213365e-05 tokens_per_second_per_gpu:5282.154492993437 mfu:24.916823971938285 +Step 18622 | loss:0.6383599042892456 lr:4.8138023389214535e-05 tokens_per_second_per_gpu:4261.669971730014 mfu:20.103024372525468 +Step 18623 | loss:0.8144466280937195 lr:4.813369018221989e-05 tokens_per_second_per_gpu:5059.352760520144 mfu:23.865830185966335 +Step 18624 | loss:0.7175688147544861 lr:4.812935698926198e-05 tokens_per_second_per_gpu:5350.369014128077 mfu:25.238603506727816 +Step 18625 | loss:0.6515551805496216 lr:4.812502381037342e-05 tokens_per_second_per_gpu:4168.4657549222065 mfu:19.663364179563857 +Step 18626 | loss:0.5971298217773438 lr:4.812069064558679e-05 tokens_per_second_per_gpu:5200.248468077472 mfu:24.530459277800812 +Step 18627 | loss:0.7048790454864502 lr:4.811635749493469e-05 tokens_per_second_per_gpu:5317.466376975657 mfu:25.083396153511085 +Step 18628 | loss:0.7524438500404358 lr:4.8112024358449695e-05 tokens_per_second_per_gpu:4929.56857389625 mfu:23.253616034195826 +Step 18629 | loss:0.7505003809928894 lr:4.8107691236164416e-05 tokens_per_second_per_gpu:4117.342345637622 mfu:19.42220633541541 +Step 18630 | loss:0.7576838731765747 lr:4.810335812811144e-05 tokens_per_second_per_gpu:5353.189977795524 mfu:25.2519104736532 +Step 18631 | loss:0.935952365398407 lr:4.809902503432335e-05 tokens_per_second_per_gpu:4499.779285145016 mfu:21.226226629542083 +Step 18632 | loss:0.7288550734519958 lr:4.809469195483272e-05 tokens_per_second_per_gpu:5258.506785198006 mfu:24.805273699552238 +Step 18633 | loss:0.7940294146537781 lr:4.809035888967216e-05 tokens_per_second_per_gpu:5312.405827003146 mfu:25.059524675871895 +Step 18634 | loss:0.7549666166305542 lr:4.808602583887426e-05 tokens_per_second_per_gpu:5341.8271994978495 mfu:25.19831030973318 +Step 18635 | loss:0.7289740443229675 lr:4.808169280247159e-05 tokens_per_second_per_gpu:4993.5471286161 mfu:23.5554137926764 +Step 18636 | loss:0.7340359091758728 lr:4.8077359780496764e-05 tokens_per_second_per_gpu:5037.808835387593 mfu:23.764203815343425 +Step 18637 | loss:0.7415199875831604 lr:4.8073026772982355e-05 tokens_per_second_per_gpu:3886.058942055547 mfu:18.331203059701018 +Step 18638 | loss:0.7034849524497986 lr:4.8068693779960986e-05 tokens_per_second_per_gpu:4856.4266245995095 mfu:22.908592980059726 +Step 18639 | loss:0.8731345534324646 lr:4.806436080146519e-05 tokens_per_second_per_gpu:4661.691810929801 mfu:21.989995638794426 +Step 18640 | loss:0.8240126967430115 lr:4.806002783752758e-05 tokens_per_second_per_gpu:4239.839168097505 mfu:20.00004474707174 +Step 18641 | loss:0.8532438278198242 lr:4.805569488818076e-05 tokens_per_second_per_gpu:5330.146471617134 mfu:25.143210323382164 +Step 18642 | loss:0.8958013653755188 lr:4.8051361953457296e-05 tokens_per_second_per_gpu:4760.228440528361 mfu:22.454809732693448 +Step 18643 | loss:0.8950632214546204 lr:4.80470290333898e-05 tokens_per_second_per_gpu:5345.689487718382 mfu:25.216529382243756 +Step 18644 | loss:0.7074363827705383 lr:4.8042696128010825e-05 tokens_per_second_per_gpu:3074.4143579650263 mfu:14.502537075700296 +Step 18645 | loss:0.7840081453323364 lr:4.803836323735302e-05 tokens_per_second_per_gpu:3473.163534961635 mfu:16.383504977218298 +Step 18646 | loss:0.7683595418930054 lr:4.803403036144891e-05 tokens_per_second_per_gpu:4516.489478060456 mfu:21.305051460577115 +Step 18647 | loss:0.7757525444030762 lr:4.802969750033112e-05 tokens_per_second_per_gpu:5318.556715932282 mfu:25.088539468400494 +Step 18648 | loss:0.6910257339477539 lr:4.8025364654032205e-05 tokens_per_second_per_gpu:5347.408922205888 mfu:25.2246402480871 +Step 18649 | loss:0.6249172687530518 lr:4.802103182258479e-05 tokens_per_second_per_gpu:3892.0096149725496 mfu:18.35927339914001 +Step 18650 | loss:0.7002891302108765 lr:4.801669900602145e-05 tokens_per_second_per_gpu:5096.157004297419 mfu:24.039442083315244 +Step 18651 | loss:0.759064257144928 lr:4.8012366204374754e-05 tokens_per_second_per_gpu:4266.698944576408 mfu:20.126746895473154 +Step 18652 | loss:0.8325769901275635 lr:4.800803341767733e-05 tokens_per_second_per_gpu:3650.591055429496 mfu:17.22046086352177 +Step 18653 | loss:0.6833438277244568 lr:4.800370064596173e-05 tokens_per_second_per_gpu:2865.939024086931 mfu:13.519123356237545 +Step 18654 | loss:0.657457172870636 lr:4.799936788926055e-05 tokens_per_second_per_gpu:5000.767000201081 mfu:23.58947115877987 +Step 18655 | loss:0.7821159958839417 lr:4.799503514760637e-05 tokens_per_second_per_gpu:4811.8617504081685 mfu:22.69837286494758 +Step 18656 | loss:0.7020372748374939 lr:4.79907024210318e-05 tokens_per_second_per_gpu:4978.1737938239785 mfu:23.482895149500578 +Step 18657 | loss:0.7930091619491577 lr:4.7986369709569414e-05 tokens_per_second_per_gpu:5283.116361247117 mfu:24.92136126860254 +Step 18658 | loss:0.7768027782440186 lr:4.798203701325179e-05 tokens_per_second_per_gpu:5357.599731148061 mfu:25.272712032598626 +Step 18659 | loss:0.7458815574645996 lr:4.7977704332111526e-05 tokens_per_second_per_gpu:4365.03736865875 mfu:20.590626020134827 +Step 18660 | loss:0.7429455518722534 lr:4.797337166618122e-05 tokens_per_second_per_gpu:4868.939637136015 mfu:22.967618995134508 +Step 18661 | loss:0.742322564125061 lr:4.7969039015493426e-05 tokens_per_second_per_gpu:5322.454769282591 mfu:25.106927251130383 +Step 18662 | loss:0.6529499888420105 lr:4.796470638008075e-05 tokens_per_second_per_gpu:4161.316542659046 mfu:19.62964012553714 +Step 18663 | loss:0.9357975721359253 lr:4.796037375997578e-05 tokens_per_second_per_gpu:4663.324760767712 mfu:21.997698541789823 +Step 18664 | loss:0.7440230846405029 lr:4.795604115521109e-05 tokens_per_second_per_gpu:5320.197991708232 mfu:25.09628164626617 +Step 18665 | loss:0.7454107999801636 lr:4.795170856581929e-05 tokens_per_second_per_gpu:5271.620796273999 mfu:24.86713472727871 +Step 18666 | loss:0.6754301190376282 lr:4.7947375991832944e-05 tokens_per_second_per_gpu:4338.15962634619 mfu:20.463839123830766 +Step 18667 | loss:0.7819318175315857 lr:4.7943043433284663e-05 tokens_per_second_per_gpu:5183.450841000224 mfu:24.45122200490672 +Step 18668 | loss:0.6893708109855652 lr:4.793871089020699e-05 tokens_per_second_per_gpu:3986.071049158612 mfu:18.802977232731216 +Step 18669 | loss:0.6860877275466919 lr:4.793437836263255e-05 tokens_per_second_per_gpu:5132.321940838563 mfu:24.21003825935426 +Step 18670 | loss:0.8383563756942749 lr:4.793004585059391e-05 tokens_per_second_per_gpu:4232.326429942875 mfu:19.964605879391044 +Step 18671 | loss:0.7774680256843567 lr:4.792571335412365e-05 tokens_per_second_per_gpu:5035.959511942621 mfu:23.75548024112638 +Step 18672 | loss:0.8468611836433411 lr:4.792138087325437e-05 tokens_per_second_per_gpu:4689.022857406517 mfu:22.118920848182473 +Step 18673 | loss:0.7345966100692749 lr:4.791704840801866e-05 tokens_per_second_per_gpu:5140.888885064798 mfu:24.25044999694082 +Step 18674 | loss:0.7307750582695007 lr:4.7912715958449094e-05 tokens_per_second_per_gpu:3686.2183073435635 mfu:17.38852068943641 +Step 18675 | loss:0.6714329123497009 lr:4.790838352457824e-05 tokens_per_second_per_gpu:5096.672640014185 mfu:24.04187442496775 +Step 18676 | loss:0.6966036558151245 lr:4.7904051106438716e-05 tokens_per_second_per_gpu:3939.7916954265656 mfu:18.58466962510519 +Step 18677 | loss:0.6761898398399353 lr:4.7899718704063084e-05 tokens_per_second_per_gpu:5218.440381884517 mfu:24.616273639089197 +Step 18678 | loss:0.7491191029548645 lr:4.789538631748393e-05 tokens_per_second_per_gpu:5337.342868147556 mfu:25.177156953648023 +Step 18679 | loss:0.7098020315170288 lr:4.789105394673385e-05 tokens_per_second_per_gpu:3612.868472823915 mfu:17.042517005234977 +Step 18680 | loss:0.7698560953140259 lr:4.788672159184542e-05 tokens_per_second_per_gpu:4589.856567258642 mfu:21.65113653804116 +Step 18681 | loss:0.8402242660522461 lr:4.788238925285126e-05 tokens_per_second_per_gpu:5305.057683479031 mfu:25.024862229146073 +Step 18682 | loss:0.6977605223655701 lr:4.7878056929783876e-05 tokens_per_second_per_gpu:5374.449063747055 mfu:25.35219320926066 +Step 18683 | loss:0.8441109657287598 lr:4.7873724622675916e-05 tokens_per_second_per_gpu:4477.289497780568 mfu:21.120138465390585 +Step 18684 | loss:0.7384837865829468 lr:4.7869392331559925e-05 tokens_per_second_per_gpu:5333.313053421594 mfu:25.15814763002837 +Step 18685 | loss:0.8777837157249451 lr:4.786506005646852e-05 tokens_per_second_per_gpu:5346.538286327029 mfu:25.220533310100716 +Step 18686 | loss:0.7066262364387512 lr:4.786072779743428e-05 tokens_per_second_per_gpu:5335.210068982102 mfu:25.167096176091516 +Step 18687 | loss:0.748111367225647 lr:4.785639555448976e-05 tokens_per_second_per_gpu:3778.964384072939 mfu:17.826019757481298 +Step 18688 | loss:0.9050277471542358 lr:4.785206332766759e-05 tokens_per_second_per_gpu:3017.381779936281 mfu:14.233504674377489 +Step 18689 | loss:0.7164274454116821 lr:4.784773111700031e-05 tokens_per_second_per_gpu:4304.630042891116 mfu:20.305674357936343 +Step 18690 | loss:0.7961603403091431 lr:4.784339892252052e-05 tokens_per_second_per_gpu:4507.854530694862 mfu:21.264318940579667 +Step 18691 | loss:0.7468826770782471 lr:4.7839066744260794e-05 tokens_per_second_per_gpu:5095.108544543012 mfu:24.034496319768024 +Step 18692 | loss:0.7329060435295105 lr:4.783473458225373e-05 tokens_per_second_per_gpu:5359.484862320991 mfu:25.281604517977875 +Step 18693 | loss:0.8763633370399475 lr:4.783040243653191e-05 tokens_per_second_per_gpu:4747.269958932324 mfu:22.393682364059995 +Step 18694 | loss:0.8537868857383728 lr:4.782607030712789e-05 tokens_per_second_per_gpu:5322.530159306699 mfu:25.107282878736964 +Step 18695 | loss:0.8907440304756165 lr:4.782173819407428e-05 tokens_per_second_per_gpu:5278.4898094362015 mfu:24.899537034339318 +Step 18696 | loss:0.7018235921859741 lr:4.781740609740368e-05 tokens_per_second_per_gpu:4717.547476517373 mfu:22.25347634333025 +Step 18697 | loss:0.760303795337677 lr:4.781307401714863e-05 tokens_per_second_per_gpu:4741.117045771941 mfu:22.36465802288723 +Step 18698 | loss:0.751442551612854 lr:4.7808741953341715e-05 tokens_per_second_per_gpu:3791.10182779905 mfu:17.8832741503988 +Step 18699 | loss:0.7088924646377563 lr:4.780440990601555e-05 tokens_per_second_per_gpu:5162.8831207371795 mfu:24.354200559211005 +Step 18700 | loss:0.6885236501693726 lr:4.780007787520268e-05 tokens_per_second_per_gpu:5341.314157358447 mfu:25.195890202427602 +Step 18701 | loss:0.743584394454956 lr:4.779574586093572e-05 tokens_per_second_per_gpu:5321.969597649317 mfu:25.104638613757412 +Step 18702 | loss:0.820947527885437 lr:4.779141386324724e-05 tokens_per_second_per_gpu:4031.0674133987472 mfu:19.015232760023316 +Step 18703 | loss:0.677824079990387 lr:4.778708188216982e-05 tokens_per_second_per_gpu:5121.702402682143 mfu:24.159944085990382 +Step 18704 | loss:0.8499457836151123 lr:4.778274991773602e-05 tokens_per_second_per_gpu:5317.924771690887 mfu:25.085558479593026 +Step 18705 | loss:0.7768462300300598 lr:4.7778417969978454e-05 tokens_per_second_per_gpu:4048.3841280956062 mfu:19.09691865778459 +Step 18706 | loss:0.8234443664550781 lr:4.777408603892969e-05 tokens_per_second_per_gpu:4399.471051020841 mfu:20.753055574828828 +Step 18707 | loss:0.6664560437202454 lr:4.7769754124622296e-05 tokens_per_second_per_gpu:4747.134139054953 mfu:22.393041678525176 +Step 18708 | loss:0.7398908734321594 lr:4.776542222708888e-05 tokens_per_second_per_gpu:3891.4889242074305 mfu:18.356817211961943 +Step 18709 | loss:0.7962694764137268 lr:4.776109034636201e-05 tokens_per_second_per_gpu:4449.151589768584 mfu:20.987407152475903 +Step 18710 | loss:0.6189144849777222 lr:4.775675848247427e-05 tokens_per_second_per_gpu:3038.3950871481925 mfu:14.332628029735984 +Step 18711 | loss:0.7539739012718201 lr:4.7752426635458217e-05 tokens_per_second_per_gpu:4472.302355557018 mfu:21.09661326462789 +Step 18712 | loss:0.6923167109489441 lr:4.774809480534646e-05 tokens_per_second_per_gpu:4975.76749610502 mfu:23.471544232603524 +Step 18713 | loss:0.6811931133270264 lr:4.774376299217157e-05 tokens_per_second_per_gpu:4523.9013417768565 mfu:21.340014486321728 +Step 18714 | loss:0.6867318749427795 lr:4.773943119596612e-05 tokens_per_second_per_gpu:5332.222584722297 mfu:25.15300370312064 +Step 18715 | loss:0.8102191090583801 lr:4.77350994167627e-05 tokens_per_second_per_gpu:4134.483311745146 mfu:19.50306319709499 +Step 18716 | loss:0.7013194561004639 lr:4.7730767654593875e-05 tokens_per_second_per_gpu:5006.070196628519 mfu:23.61448724114671 +Step 18717 | loss:0.889991283416748 lr:4.7726435909492276e-05 tokens_per_second_per_gpu:5065.781970792002 mfu:23.89615786775507 +Step 18718 | loss:0.6480562090873718 lr:4.772210418149041e-05 tokens_per_second_per_gpu:4373.74762152044 mfu:20.6317137690058 +Step 18719 | loss:0.7247359156608582 lr:4.771777247062089e-05 tokens_per_second_per_gpu:5096.8539288054035 mfu:24.04272959512699 +Step 18720 | loss:0.8107278943061829 lr:4.771344077691629e-05 tokens_per_second_per_gpu:4783.206976593847 mfu:22.563203407856875 +Step 18721 | loss:0.736778974533081 lr:4.770910910040921e-05 tokens_per_second_per_gpu:4321.655199382381 mfu:20.385984925896874 +Step 18722 | loss:0.7485417127609253 lr:4.7704777441132206e-05 tokens_per_second_per_gpu:5114.811023704993 mfu:24.127436275564673 +Step 18723 | loss:0.7878707051277161 lr:4.770044579911785e-05 tokens_per_second_per_gpu:5330.040164956394 mfu:25.142708856725665 +Step 18724 | loss:0.8438246846199036 lr:4.769611417439875e-05 tokens_per_second_per_gpu:4749.394698396341 mfu:22.4037051226297 +Step 18725 | loss:0.751202404499054 lr:4.769178256700748e-05 tokens_per_second_per_gpu:5168.454095136921 mfu:24.380479795960834 +Step 18726 | loss:0.7513117790222168 lr:4.7687450976976594e-05 tokens_per_second_per_gpu:5345.456808652693 mfu:25.21543179539176 +Step 18727 | loss:0.7540674805641174 lr:4.7683119404338675e-05 tokens_per_second_per_gpu:3657.301237492375 mfu:17.252113937187485 +Step 18728 | loss:0.7607242465019226 lr:4.7678787849126324e-05 tokens_per_second_per_gpu:4197.7498630417795 mfu:19.801502313947942 +Step 18729 | loss:0.8373299837112427 lr:4.767445631137211e-05 tokens_per_second_per_gpu:5351.277322689776 mfu:25.24288815318678 +Step 18730 | loss:0.8213163614273071 lr:4.767012479110858e-05 tokens_per_second_per_gpu:3961.0389981642647 mfu:18.684896777282578 +Step 18731 | loss:0.6046417951583862 lr:4.766579328836836e-05 tokens_per_second_per_gpu:3510.626492272425 mfu:16.560224138692917 +Step 18732 | loss:0.7346960306167603 lr:4.766146180318402e-05 tokens_per_second_per_gpu:4228.515983527025 mfu:19.946631353518285 +Step 18733 | loss:0.698724091053009 lr:4.7657130335588115e-05 tokens_per_second_per_gpu:4281.028447073818 mfu:20.194341603618664 +Step 18734 | loss:0.6192576289176941 lr:4.765279888561322e-05 tokens_per_second_per_gpu:3434.355832512938 mfu:16.200442423490994 +Step 18735 | loss:0.7079494595527649 lr:4.7648467453291924e-05 tokens_per_second_per_gpu:3775.2019523295667 mfu:17.80827172501076 +Step 18736 | loss:0.8640621304512024 lr:4.76441360386568e-05 tokens_per_second_per_gpu:3853.5232307252745 mfu:18.177726558191353 +Step 18737 | loss:0.7696188688278198 lr:4.763980464174044e-05 tokens_per_second_per_gpu:5317.88840011936 mfu:25.085386908684505 +Step 18738 | loss:0.7337573766708374 lr:4.763547326257542e-05 tokens_per_second_per_gpu:4569.314530502395 mfu:21.55423624582907 +Step 18739 | loss:0.7170678973197937 lr:4.7631141901194305e-05 tokens_per_second_per_gpu:3653.1468129999075 mfu:17.232516808038604 +Step 18740 | loss:0.802947998046875 lr:4.762681055762966e-05 tokens_per_second_per_gpu:4465.793807982582 mfu:21.06591133524655 +Step 18741 | loss:0.7488447427749634 lr:4.762247923191408e-05 tokens_per_second_per_gpu:4198.614283789012 mfu:19.805579934098027 +Step 18742 | loss:0.627617359161377 lr:4.761814792408014e-05 tokens_per_second_per_gpu:5356.386958420596 mfu:25.266991176722808 +Step 18743 | loss:0.6311196088790894 lr:4.76138166341604e-05 tokens_per_second_per_gpu:4868.084556459537 mfu:22.963585433691758 +Step 18744 | loss:0.7334741353988647 lr:4.7609485362187464e-05 tokens_per_second_per_gpu:4249.1020224685135 mfu:20.04373921154571 +Step 18745 | loss:0.8301267027854919 lr:4.7605154108193894e-05 tokens_per_second_per_gpu:5364.550602818608 mfu:25.305500480209595 +Step 18746 | loss:0.6691625714302063 lr:4.7600822872212276e-05 tokens_per_second_per_gpu:3931.4184217977986 mfu:18.5451714647705 +Step 18747 | loss:0.8163554668426514 lr:4.7596491654275156e-05 tokens_per_second_per_gpu:4438.310204746517 mfu:20.93626648959579 +Step 18748 | loss:0.8252802491188049 lr:4.7592160454415136e-05 tokens_per_second_per_gpu:5055.684935889497 mfu:23.848528431388157 +Step 18749 | loss:0.7659953832626343 lr:4.758782927266478e-05 tokens_per_second_per_gpu:5355.723336189174 mfu:25.26386075743156 +Step 18750 | loss:0.7378055453300476 lr:4.758349810905667e-05 tokens_per_second_per_gpu:5318.75240689474 mfu:25.089462576058725 +Step 18751 | loss:0.7998803853988647 lr:4.757916696362338e-05 tokens_per_second_per_gpu:3402.3168391059553 mfu:16.049309025174303 +Step 18752 | loss:0.8149003386497498 lr:4.757483583639747e-05 tokens_per_second_per_gpu:4451.838729412618 mfu:21.000082848650106 +Step 18753 | loss:0.7384467124938965 lr:4.757050472741155e-05 tokens_per_second_per_gpu:5359.050576153197 mfu:25.27955591602753 +Step 18754 | loss:0.7226250767707825 lr:4.756617363669818e-05 tokens_per_second_per_gpu:5342.801752341965 mfu:25.202907441774734 +Step 18755 | loss:0.6761897802352905 lr:4.756184256428992e-05 tokens_per_second_per_gpu:5308.504261507757 mfu:25.041120325754427 +Step 18756 | loss:0.6574880480766296 lr:4.755751151021934e-05 tokens_per_second_per_gpu:4876.628126041456 mfu:23.00388690909295 +Step 18757 | loss:0.8790302872657776 lr:4.755318047451904e-05 tokens_per_second_per_gpu:4149.509406075871 mfu:19.573943847769073 +Step 18758 | loss:0.7874636650085449 lr:4.754884945722158e-05 tokens_per_second_per_gpu:4976.639359969877 mfu:23.475656963209367 +Step 18759 | loss:0.7520085573196411 lr:4.754451845835952e-05 tokens_per_second_per_gpu:4854.64642829514 mfu:22.900195490359202 +Step 18760 | loss:0.752504289150238 lr:4.754018747796547e-05 tokens_per_second_per_gpu:5112.215689362852 mfu:24.11519364066309 +Step 18761 | loss:0.9514310956001282 lr:4.7535856516071994e-05 tokens_per_second_per_gpu:4649.344695997563 mfu:21.93175219102402 +Step 18762 | loss:0.8041048645973206 lr:4.753152557271164e-05 tokens_per_second_per_gpu:5314.810161807945 mfu:25.07086633336795 +Step 18763 | loss:0.7666195034980774 lr:4.752719464791699e-05 tokens_per_second_per_gpu:5343.892439357975 mfu:25.208052398519648 +Step 18764 | loss:0.6597546935081482 lr:4.7522863741720634e-05 tokens_per_second_per_gpu:4523.424933577918 mfu:21.337767187563713 +Step 18765 | loss:0.7945215106010437 lr:4.7518532854155136e-05 tokens_per_second_per_gpu:5305.290807345192 mfu:25.025961914197786 +Step 18766 | loss:0.8916643857955933 lr:4.751420198525306e-05 tokens_per_second_per_gpu:5349.012611878142 mfu:25.23220512588879 +Step 18767 | loss:0.6730805039405823 lr:4.7509871135046994e-05 tokens_per_second_per_gpu:5227.63472543911 mfu:24.659644926352986 +Step 18768 | loss:0.6797429323196411 lr:4.750554030356952e-05 tokens_per_second_per_gpu:5322.095536176515 mfu:25.105232687277535 +Step 18769 | loss:0.8784596920013428 lr:4.7501209490853184e-05 tokens_per_second_per_gpu:5115.431259342884 mfu:24.130362032891718 +Step 18770 | loss:0.7490214705467224 lr:4.749687869693056e-05 tokens_per_second_per_gpu:4017.2079613741885 mfu:18.94985536015712 +Step 18771 | loss:0.6888527274131775 lr:4.749254792183424e-05 tokens_per_second_per_gpu:3749.3257464467088 mfu:17.68620924692534 +Step 18772 | loss:0.7217453122138977 lr:4.7488217165596774e-05 tokens_per_second_per_gpu:3935.2282927441447 mfu:18.563143276055424 +Step 18773 | loss:0.6550173759460449 lr:4.748388642825076e-05 tokens_per_second_per_gpu:5320.370630257238 mfu:25.097096011757007 +Step 18774 | loss:0.8252627849578857 lr:4.747955570982876e-05 tokens_per_second_per_gpu:4804.565925541711 mfu:22.66395721425642 +Step 18775 | loss:0.8074291944503784 lr:4.747522501036335e-05 tokens_per_second_per_gpu:3378.109036725082 mfu:15.935116691066735 +Step 18776 | loss:0.6600444912910461 lr:4.7470894329887075e-05 tokens_per_second_per_gpu:4854.366207793759 mfu:22.89887364244364 +Step 18777 | loss:0.6262610554695129 lr:4.746656366843254e-05 tokens_per_second_per_gpu:5372.308786579812 mfu:25.342097156693832 +Step 18778 | loss:0.6592859625816345 lr:4.74622330260323e-05 tokens_per_second_per_gpu:4126.847945908645 mfu:19.46704587371647 +Step 18779 | loss:0.8464550971984863 lr:4.745790240271892e-05 tokens_per_second_per_gpu:3467.234034305524 mfu:16.355534510946292 +Step 18780 | loss:0.766430139541626 lr:4.7453571798524986e-05 tokens_per_second_per_gpu:5341.392526607135 mfu:25.196259883544837 +Step 18781 | loss:0.7960241436958313 lr:4.744924121348307e-05 tokens_per_second_per_gpu:4269.0682953073065 mfu:20.13792352712372 +Step 18782 | loss:0.8704925775527954 lr:4.744491064762572e-05 tokens_per_second_per_gpu:4379.9220526898835 mfu:20.660839614298652 +Step 18783 | loss:0.6604330539703369 lr:4.744058010098555e-05 tokens_per_second_per_gpu:4477.355620515856 mfu:21.120450377614638 +Step 18784 | loss:0.7731737494468689 lr:4.7436249573595085e-05 tokens_per_second_per_gpu:4136.581926563765 mfu:19.512962721255022 +Step 18785 | loss:0.8345506191253662 lr:4.7431919065486924e-05 tokens_per_second_per_gpu:4508.123274782077 mfu:21.265586652291532 +Step 18786 | loss:0.7666481733322144 lr:4.742758857669361e-05 tokens_per_second_per_gpu:4728.313820420032 mfu:22.304263024446783 +Step 18787 | loss:0.6619324684143066 lr:4.742325810724775e-05 tokens_per_second_per_gpu:5361.03960051779 mfu:25.288938483317867 +Step 18788 | loss:0.647146463394165 lr:4.741892765718187e-05 tokens_per_second_per_gpu:3286.346583602511 mfu:15.502257543398642 +Step 18789 | loss:0.6247645020484924 lr:4.7414597226528584e-05 tokens_per_second_per_gpu:5358.331524151363 mfu:25.27616402505196 +Step 18790 | loss:0.7805100679397583 lr:4.7410266815320457e-05 tokens_per_second_per_gpu:4804.198852014778 mfu:22.662225665800673 +Step 18791 | loss:0.6517555713653564 lr:4.740593642359003e-05 tokens_per_second_per_gpu:5329.321627444242 mfu:25.139319392685707 +Step 18792 | loss:0.875817060470581 lr:4.7401606051369865e-05 tokens_per_second_per_gpu:3786.0791076919622 mfu:17.85958112268925 +Step 18793 | loss:0.7823922038078308 lr:4.7397275698692575e-05 tokens_per_second_per_gpu:4508.891341093731 mfu:21.26920975212923 +Step 18794 | loss:0.7120566368103027 lr:4.7392945365590704e-05 tokens_per_second_per_gpu:3969.038744217151 mfu:18.722632944311464 +Step 18795 | loss:0.8840016722679138 lr:4.7388615052096805e-05 tokens_per_second_per_gpu:4461.222921958081 mfu:21.04434968599541 +Step 18796 | loss:0.7138645648956299 lr:4.738428475824349e-05 tokens_per_second_per_gpu:5343.629794511598 mfu:25.206813457967456 +Step 18797 | loss:0.7295367121696472 lr:4.73799544840633e-05 tokens_per_second_per_gpu:5081.017356388638 mfu:23.96802578103923 +Step 18798 | loss:0.9097700715065002 lr:4.73756242295888e-05 tokens_per_second_per_gpu:5345.278500809089 mfu:25.21459068687102 +Step 18799 | loss:0.6904030442237854 lr:4.737129399485256e-05 tokens_per_second_per_gpu:5331.648806044505 mfu:25.15029709120076 +Step 18800 | loss:0.7432923913002014 lr:4.7366963779887156e-05 tokens_per_second_per_gpu:5315.373149457619 mfu:25.073522042167053 +Step 18801 | loss:0.717257022857666 lr:4.736263358472516e-05 tokens_per_second_per_gpu:5001.151590066741 mfu:23.591285334793987 +Step 18802 | loss:0.8325119614601135 lr:4.735830340939912e-05 tokens_per_second_per_gpu:5330.031375160868 mfu:25.142667393760476 +Step 18803 | loss:0.8166372179985046 lr:4.735397325394163e-05 tokens_per_second_per_gpu:5341.372763794802 mfu:25.196166658987185 +Step 18804 | loss:0.7191485166549683 lr:4.734964311838526e-05 tokens_per_second_per_gpu:5105.928207942074 mfu:24.08553451804633 +Step 18805 | loss:0.748023509979248 lr:4.7345313002762545e-05 tokens_per_second_per_gpu:3979.741484249621 mfu:18.773119595120374 +Step 18806 | loss:0.7500160336494446 lr:4.734098290710606e-05 tokens_per_second_per_gpu:4692.293801111128 mfu:22.134350447717665 +Step 18807 | loss:0.6121936440467834 lr:4.733665283144839e-05 tokens_per_second_per_gpu:4027.0893105654804 mfu:18.996467370224465 +Step 18808 | loss:0.7477741837501526 lr:4.733232277582208e-05 tokens_per_second_per_gpu:5333.918744053361 mfu:25.161004776079892 +Step 18809 | loss:0.6397125720977783 lr:4.732799274025973e-05 tokens_per_second_per_gpu:4660.29901361821 mfu:21.983425576239924 +Step 18810 | loss:0.7178579568862915 lr:4.732366272479388e-05 tokens_per_second_per_gpu:5380.660654677774 mfu:25.381494343487557 +Step 18811 | loss:0.8384395837783813 lr:4.7319332729457124e-05 tokens_per_second_per_gpu:4621.218510044479 mfu:21.799076173061163 +Step 18812 | loss:0.7629963159561157 lr:4.731500275428199e-05 tokens_per_second_per_gpu:4660.70740396577 mfu:21.985352023187943 +Step 18813 | loss:0.6843712329864502 lr:4.731067279930106e-05 tokens_per_second_per_gpu:5358.159416247936 mfu:25.27535216270686 +Step 18814 | loss:0.8252551555633545 lr:4.7306342864546905e-05 tokens_per_second_per_gpu:5192.95519220488 mfu:24.49605565114862 +Step 18815 | loss:0.7153338193893433 lr:4.7302012950052084e-05 tokens_per_second_per_gpu:5355.643230093071 mfu:25.263482883308733 +Step 18816 | loss:0.7225631475448608 lr:4.729768305584917e-05 tokens_per_second_per_gpu:4600.434577636838 mfu:21.701034817789786 +Step 18817 | loss:0.790886402130127 lr:4.729335318197074e-05 tokens_per_second_per_gpu:5242.7224588095805 mfu:24.730816338894254 +Step 18818 | loss:0.6899029612541199 lr:4.728902332844933e-05 tokens_per_second_per_gpu:4055.315451597322 mfu:19.12961489334792 +Step 18819 | loss:0.673382580280304 lr:4.728469349531754e-05 tokens_per_second_per_gpu:5368.220095594661 mfu:25.32281010371426 +Step 18820 | loss:0.8455435037612915 lr:4.728036368260791e-05 tokens_per_second_per_gpu:4201.473848570748 mfu:19.81906898906553 +Step 18821 | loss:0.7967739701271057 lr:4.727603389035301e-05 tokens_per_second_per_gpu:5354.768941246509 mfu:25.25935871365044 +Step 18822 | loss:0.7877000570297241 lr:4.727170411858539e-05 tokens_per_second_per_gpu:5361.338774626673 mfu:25.290349738633836 +Step 18823 | loss:0.7019757032394409 lr:4.726737436733765e-05 tokens_per_second_per_gpu:4692.070056320107 mfu:22.133295005363625 +Step 18824 | loss:0.7387413382530212 lr:4.726304463664232e-05 tokens_per_second_per_gpu:5275.785572335933 mfu:24.88678068654692 +Step 18825 | loss:0.726955771446228 lr:4.725871492653199e-05 tokens_per_second_per_gpu:5216.759677567341 mfu:24.608345470067352 +Step 18826 | loss:0.9050301909446716 lr:4.7254385237039235e-05 tokens_per_second_per_gpu:5350.7414256549855 mfu:25.240360235440207 +Step 18827 | loss:0.5998880863189697 lr:4.7250055568196586e-05 tokens_per_second_per_gpu:5342.290855885908 mfu:25.20049745602359 +Step 18828 | loss:0.731474757194519 lr:4.724572592003661e-05 tokens_per_second_per_gpu:4292.671555455841 mfu:20.249264132375806 +Step 18829 | loss:0.8356450200080872 lr:4.7241396292591886e-05 tokens_per_second_per_gpu:4179.435314312947 mfu:19.715109462809814 +Step 18830 | loss:0.8252742290496826 lr:4.7237066685894974e-05 tokens_per_second_per_gpu:4916.9606777760755 mfu:23.194142437068066 +Step 18831 | loss:0.695235013961792 lr:4.7232737099978416e-05 tokens_per_second_per_gpu:5326.4125377270775 mfu:25.12559672014795 +Step 18832 | loss:0.8648269772529602 lr:4.722840753487482e-05 tokens_per_second_per_gpu:4485.808314821883 mfu:21.160323178834528 +Step 18833 | loss:0.8328412771224976 lr:4.722407799061673e-05 tokens_per_second_per_gpu:4046.526374915556 mfu:19.088155319068917 +Step 18834 | loss:0.7433115839958191 lr:4.72197484672367e-05 tokens_per_second_per_gpu:4659.664787642117 mfu:21.980433824100583 +Step 18835 | loss:0.7660777568817139 lr:4.721541896476728e-05 tokens_per_second_per_gpu:5312.859228571568 mfu:25.06166344842842 +Step 18836 | loss:0.7230005860328674 lr:4.7211089483241054e-05 tokens_per_second_per_gpu:5358.533505588261 mfu:25.277116805951394 +Step 18837 | loss:0.7264114618301392 lr:4.720676002269059e-05 tokens_per_second_per_gpu:5337.177078471899 mfu:25.176374895461315 +Step 18838 | loss:0.7395030856132507 lr:4.720243058314842e-05 tokens_per_second_per_gpu:5315.502431428833 mfu:25.074131887283052 +Step 18839 | loss:0.7180885076522827 lr:4.7198101164647146e-05 tokens_per_second_per_gpu:5116.0096282610675 mfu:24.1330902977587 +Step 18840 | loss:0.6695223450660706 lr:4.719377176721931e-05 tokens_per_second_per_gpu:4149.599735322846 mfu:19.57436994623881 +Step 18841 | loss:0.8318796157836914 lr:4.718944239089747e-05 tokens_per_second_per_gpu:4200.0304883126455 mfu:19.812260412465413 +Step 18842 | loss:0.7126895785331726 lr:4.718511303571419e-05 tokens_per_second_per_gpu:4431.685625236264 mfu:20.905017217776113 +Step 18843 | loss:0.8276336789131165 lr:4.7180783701702044e-05 tokens_per_second_per_gpu:5353.391535688907 mfu:25.252861256625614 +Step 18844 | loss:0.905193567276001 lr:4.717645438889356e-05 tokens_per_second_per_gpu:5231.267959678376 mfu:24.676783512153285 +Step 18845 | loss:0.7845709323883057 lr:4.717212509732134e-05 tokens_per_second_per_gpu:5301.290280138668 mfu:25.00709074480429 +Step 18846 | loss:0.7974694967269897 lr:4.716779582701793e-05 tokens_per_second_per_gpu:3484.2821468337074 mfu:16.435953366449667 +Step 18847 | loss:0.7654516696929932 lr:4.7163466578015876e-05 tokens_per_second_per_gpu:3301.2565857992663 mfu:15.57259056766943 +Step 18848 | loss:0.7543438673019409 lr:4.715913735034779e-05 tokens_per_second_per_gpu:3768.86467454001 mfu:17.778377704426333 +Step 18849 | loss:0.9492664337158203 lr:4.715480814404615e-05 tokens_per_second_per_gpu:4126.927084209131 mfu:19.467419182582248 +Step 18850 | loss:0.7707737684249878 lr:4.715047895914359e-05 tokens_per_second_per_gpu:4081.231891782773 mfu:19.25186716350335 +Step 18851 | loss:0.9028006196022034 lr:4.714614979567261e-05 tokens_per_second_per_gpu:3180.8106235453574 mfu:15.00442641351751 +Step 18852 | loss:0.8104930520057678 lr:4.7141820653665826e-05 tokens_per_second_per_gpu:4200.122360430937 mfu:19.812693788922395 +Step 18853 | loss:0.7714638710021973 lr:4.7137491533155764e-05 tokens_per_second_per_gpu:4543.231385927334 mfu:21.431197602625144 +Step 18854 | loss:0.8210989236831665 lr:4.713316243417499e-05 tokens_per_second_per_gpu:4538.915145707904 mfu:21.410837161083574 +Step 18855 | loss:0.7901018261909485 lr:4.7128833356756087e-05 tokens_per_second_per_gpu:4476.828537478412 mfu:21.117964036997375 +Step 18856 | loss:0.7533367872238159 lr:4.7124504300931585e-05 tokens_per_second_per_gpu:5076.1855091050775 mfu:23.94523313300452 +Step 18857 | loss:0.5981903672218323 lr:4.712017526673405e-05 tokens_per_second_per_gpu:5293.56511509597 mfu:24.970649823248817 +Step 18858 | loss:0.7273935079574585 lr:4.711584625419603e-05 tokens_per_second_per_gpu:4167.675668230922 mfu:19.659637205838383 +Step 18859 | loss:0.8502693772315979 lr:4.711151726335012e-05 tokens_per_second_per_gpu:4536.755950385789 mfu:21.400651868352448 +Step 18860 | loss:0.9230694770812988 lr:4.7107188294228836e-05 tokens_per_second_per_gpu:4872.057946148764 mfu:22.982328590786025 +Step 18861 | loss:0.875760018825531 lr:4.7102859346864776e-05 tokens_per_second_per_gpu:5365.779273188868 mfu:25.311296327978887 +Step 18862 | loss:0.9653149247169495 lr:4.709853042129049e-05 tokens_per_second_per_gpu:4314.640487065819 mfu:20.352895331066314 +Step 18863 | loss:0.7004356384277344 lr:4.7094201517538516e-05 tokens_per_second_per_gpu:4439.39344407004 mfu:20.94137631430453 +Step 18864 | loss:0.6964772939682007 lr:4.7089872635641416e-05 tokens_per_second_per_gpu:4809.1785883335715 mfu:22.68571593164705 +Step 18865 | loss:0.6545958518981934 lr:4.708554377563175e-05 tokens_per_second_per_gpu:5311.451585993887 mfu:25.055023358220915 +Step 18866 | loss:0.7757622003555298 lr:4.7081214937542095e-05 tokens_per_second_per_gpu:5134.401833666402 mfu:24.21984946868949 +Step 18867 | loss:0.7008974552154541 lr:4.7076886121404987e-05 tokens_per_second_per_gpu:3999.498121732463 mfu:18.866315024956883 +Step 18868 | loss:0.8030585646629333 lr:4.707255732725299e-05 tokens_per_second_per_gpu:3819.4053660934246 mfu:18.01678676961496 +Step 18869 | loss:0.7504494190216064 lr:4.706822855511868e-05 tokens_per_second_per_gpu:5337.867600941317 mfu:25.179632207764836 +Step 18870 | loss:0.7364374399185181 lr:4.7063899805034586e-05 tokens_per_second_per_gpu:4909.366906459061 mfu:23.15832132213472 +Step 18871 | loss:0.8506680727005005 lr:4.705957107703327e-05 tokens_per_second_per_gpu:4306.168096684671 mfu:20.31292962009958 +Step 18872 | loss:0.7446441054344177 lr:4.7055242371147304e-05 tokens_per_second_per_gpu:5331.562488294163 mfu:25.14988991564507 +Step 18873 | loss:0.6508899331092834 lr:4.705091368740923e-05 tokens_per_second_per_gpu:5347.375544308143 mfu:25.224482798848573 +Step 18874 | loss:0.7086775302886963 lr:4.7046585025851605e-05 tokens_per_second_per_gpu:5299.089560158811 mfu:24.99670957317776 +Step 18875 | loss:0.7551887631416321 lr:4.7042256386507e-05 tokens_per_second_per_gpu:5342.544098320887 mfu:25.201692043797003 +Step 18876 | loss:0.746837854385376 lr:4.7037927769407954e-05 tokens_per_second_per_gpu:4304.562230885576 mfu:20.30535447713629 +Step 18877 | loss:0.7582041621208191 lr:4.7033599174587064e-05 tokens_per_second_per_gpu:4618.3738686449815 mfu:21.78565751423386 +Step 18878 | loss:0.7908624410629272 lr:4.702927060207681e-05 tokens_per_second_per_gpu:3770.678723475461 mfu:17.786934882763518 +Step 18879 | loss:0.6451800465583801 lr:4.702494205190982e-05 tokens_per_second_per_gpu:5369.755806147466 mfu:25.33005431241108 +Step 18880 | loss:0.9232358336448669 lr:4.70206135241186e-05 tokens_per_second_per_gpu:5287.1175592852205 mfu:24.940235602423055 +Step 18881 | loss:0.6732917428016663 lr:4.701628501873573e-05 tokens_per_second_per_gpu:4177.717354142702 mfu:19.707005551572532 +Step 18882 | loss:0.7912022471427917 lr:4.701195653579376e-05 tokens_per_second_per_gpu:3550.94722435989 mfu:16.75042391137577 +Step 18883 | loss:0.8496772646903992 lr:4.700762807532525e-05 tokens_per_second_per_gpu:4470.432137351577 mfu:21.087791126260697 +Step 18884 | loss:0.7720895409584045 lr:4.700329963736278e-05 tokens_per_second_per_gpu:5249.766136346129 mfu:24.764042567609465 +Step 18885 | loss:0.86109858751297 lr:4.699897122193883e-05 tokens_per_second_per_gpu:3985.065417105577 mfu:18.798233494758527 +Step 18886 | loss:0.9414538145065308 lr:4.699464282908602e-05 tokens_per_second_per_gpu:3959.18115710766 mfu:18.676133024038204 +Step 18887 | loss:0.7025890946388245 lr:4.6990314458836875e-05 tokens_per_second_per_gpu:4881.532818842792 mfu:23.027023181863733 +Step 18888 | loss:0.8381964564323425 lr:4.698598611122397e-05 tokens_per_second_per_gpu:4665.487369913066 mfu:22.007899938107624 +Step 18889 | loss:0.7309902906417847 lr:4.698165778627985e-05 tokens_per_second_per_gpu:4009.669811066032 mfu:18.914296619012482 +Step 18890 | loss:0.7437577843666077 lr:4.697732948403705e-05 tokens_per_second_per_gpu:3914.310811579697 mfu:18.46447194851243 +Step 18891 | loss:0.6042092442512512 lr:4.697300120452817e-05 tokens_per_second_per_gpu:3296.6702421418086 mfu:15.550956002126282 +Step 18892 | loss:0.6832991242408752 lr:4.696867294778572e-05 tokens_per_second_per_gpu:5341.797415881937 mfu:25.198169815335312 +Step 18893 | loss:0.7766556143760681 lr:4.696434471384226e-05 tokens_per_second_per_gpu:5321.27303201567 mfu:25.101352794817146 +Step 18894 | loss:0.7455157041549683 lr:4.696001650273035e-05 tokens_per_second_per_gpu:3581.1558439368423 mfu:16.892923124042227 +Step 18895 | loss:0.7400256991386414 lr:4.6955688314482546e-05 tokens_per_second_per_gpu:4660.6366189525215 mfu:21.98501811820328 +Step 18896 | loss:0.7753874659538269 lr:4.695136014913141e-05 tokens_per_second_per_gpu:5025.045806027135 mfu:23.703998428252792 +Step 18897 | loss:0.6848522424697876 lr:4.694703200670947e-05 tokens_per_second_per_gpu:4052.866816191326 mfu:19.118064262357557 +Step 18898 | loss:0.7949303388595581 lr:4.694270388724931e-05 tokens_per_second_per_gpu:3725.5305132060844 mfu:17.57396307184388 +Step 18899 | loss:0.6792638897895813 lr:4.693837579078345e-05 tokens_per_second_per_gpu:5370.983038307803 mfu:25.335843375898786 +Step 18900 | loss:0.782580554485321 lr:4.693404771734446e-05 tokens_per_second_per_gpu:5299.945525224198 mfu:25.00074730643389 +Step 18901 | loss:0.6710090041160583 lr:4.692971966696488e-05 tokens_per_second_per_gpu:4040.5290258713876 mfu:19.059864800374314 +Step 18902 | loss:0.7305240631103516 lr:4.6925391639677274e-05 tokens_per_second_per_gpu:2775.730035166711 mfu:13.093592164260262 +Step 18903 | loss:0.6354764699935913 lr:4.692106363551418e-05 tokens_per_second_per_gpu:5358.690718344051 mfu:25.277858405343718 +Step 18904 | loss:0.7093508839607239 lr:4.691673565450818e-05 tokens_per_second_per_gpu:4639.35082678874 mfu:21.884609404836088 +Step 18905 | loss:0.8494173288345337 lr:4.69124076966918e-05 tokens_per_second_per_gpu:5366.9652985023 mfu:25.316891011739134 +Step 18906 | loss:0.7133170366287231 lr:4.6908079762097594e-05 tokens_per_second_per_gpu:5320.978932201137 mfu:25.09996547581339 +Step 18907 | loss:0.7553321123123169 lr:4.6903751850758106e-05 tokens_per_second_per_gpu:4716.887353645917 mfu:22.250362431117015 +Step 18908 | loss:0.8499645590782166 lr:4.6899423962705904e-05 tokens_per_second_per_gpu:3539.1384870818106 mfu:16.694720082856733 +Step 18909 | loss:0.8860002160072327 lr:4.689509609797353e-05 tokens_per_second_per_gpu:5356.689209967714 mfu:25.268416948839995 +Step 18910 | loss:0.712120771408081 lr:4.689076825659352e-05 tokens_per_second_per_gpu:4926.901260719826 mfu:23.241033862852078 +Step 18911 | loss:0.7462159991264343 lr:4.688644043859845e-05 tokens_per_second_per_gpu:3708.5195525939007 mfu:17.49371946826705 +Step 18912 | loss:0.7197461128234863 lr:4.688211264402086e-05 tokens_per_second_per_gpu:4714.867091492864 mfu:22.2408325098488 +Step 18913 | loss:0.7528176307678223 lr:4.6877784872893314e-05 tokens_per_second_per_gpu:5343.24851798299 mfu:25.205014911529528 +Step 18914 | loss:0.5871723294258118 lr:4.687345712524832e-05 tokens_per_second_per_gpu:4399.485404621642 mfu:20.753123283212787 +Step 18915 | loss:0.8417376279830933 lr:4.686912940111847e-05 tokens_per_second_per_gpu:4689.532729687203 mfu:22.12132600272365 +Step 18916 | loss:0.8553715348243713 lr:4.686480170053629e-05 tokens_per_second_per_gpu:4031.2895339935603 mfu:19.01628054076673 +Step 18917 | loss:0.8257009983062744 lr:4.6860474023534335e-05 tokens_per_second_per_gpu:4299.634118347068 mfu:20.282107729469445 +Step 18918 | loss:0.6904315948486328 lr:4.685614637014516e-05 tokens_per_second_per_gpu:5263.691248229537 mfu:24.82972969623258 +Step 18919 | loss:0.8440364003181458 lr:4.68518187404013e-05 tokens_per_second_per_gpu:5324.911297196625 mfu:25.11851511242776 +Step 18920 | loss:0.7905214428901672 lr:4.684749113433534e-05 tokens_per_second_per_gpu:4503.056017279147 mfu:21.24168353408693 +Step 18921 | loss:0.669906735420227 lr:4.684316355197977e-05 tokens_per_second_per_gpu:3733.5652069713256 mfu:17.61186409319464 +Step 18922 | loss:0.700250506401062 lr:4.6838835993367185e-05 tokens_per_second_per_gpu:5325.780868745314 mfu:25.122617029787133 +Step 18923 | loss:0.7859627604484558 lr:4.683450845853011e-05 tokens_per_second_per_gpu:5336.239179397401 mfu:25.171950665504824 +Step 18924 | loss:0.7848196029663086 lr:4.683018094750111e-05 tokens_per_second_per_gpu:4741.961267584724 mfu:22.368640361217224 +Step 18925 | loss:0.725284218788147 lr:4.6825853460312716e-05 tokens_per_second_per_gpu:4177.347869470834 mfu:19.705262629334747 +Step 18926 | loss:0.6884201169013977 lr:4.682152599699748e-05 tokens_per_second_per_gpu:4183.113205155234 mfu:19.73245870142576 +Step 18927 | loss:0.7436926364898682 lr:4.681719855758797e-05 tokens_per_second_per_gpu:4986.032980748961 mfu:23.519968275141277 +Step 18928 | loss:0.8636457920074463 lr:4.681287114211671e-05 tokens_per_second_per_gpu:5357.1146468631105 mfu:25.270423807262596 +Step 18929 | loss:0.7132871150970459 lr:4.6808543750616246e-05 tokens_per_second_per_gpu:4861.317412763306 mfu:22.93166366228286 +Step 18930 | loss:0.6658862829208374 lr:4.6804216383119126e-05 tokens_per_second_per_gpu:5286.1983761640895 mfu:24.935899658811866 +Step 18931 | loss:0.659371554851532 lr:4.679988903965791e-05 tokens_per_second_per_gpu:5338.916481732338 mfu:25.1845799574136 +Step 18932 | loss:0.7745904922485352 lr:4.679556172026513e-05 tokens_per_second_per_gpu:5101.594996009068 mfu:24.065094018036152 +Step 18933 | loss:0.8517160415649414 lr:4.679123442497334e-05 tokens_per_second_per_gpu:4615.616969957582 mfu:21.772652752749977 +Step 18934 | loss:0.7770099639892578 lr:4.678690715381511e-05 tokens_per_second_per_gpu:5340.512183767923 mfu:25.192107156170298 +Step 18935 | loss:0.6468623876571655 lr:4.678257990682294e-05 tokens_per_second_per_gpu:5180.222777887082 mfu:24.43599468043767 +Step 18936 | loss:0.7398588061332703 lr:4.6778252684029406e-05 tokens_per_second_per_gpu:3596.2908238414025 mfu:16.964317406545888 +Step 18937 | loss:0.776294469833374 lr:4.677392548546703e-05 tokens_per_second_per_gpu:4457.543962969943 mfu:21.026995408753233 +Step 18938 | loss:0.7825298309326172 lr:4.6769598311168376e-05 tokens_per_second_per_gpu:5324.698717136496 mfu:25.117512336014236 +Step 18939 | loss:0.657309889793396 lr:4.676527116116598e-05 tokens_per_second_per_gpu:5323.465265216597 mfu:25.111693932859474 +Step 18940 | loss:0.8794064521789551 lr:4.6760944035492404e-05 tokens_per_second_per_gpu:5316.890047259771 mfu:25.080677507909257 +Step 18941 | loss:0.7127968072891235 lr:4.6756616934180185e-05 tokens_per_second_per_gpu:3511.4353730395796 mfu:16.564039767280878 +Step 18942 | loss:0.7594155073165894 lr:4.6752289857261866e-05 tokens_per_second_per_gpu:4021.984556850775 mfu:18.972387375991406 +Step 18943 | loss:0.7431728839874268 lr:4.674796280476997e-05 tokens_per_second_per_gpu:4864.721312873825 mfu:22.947720439869425 +Step 18944 | loss:0.8527962565422058 lr:4.674363577673708e-05 tokens_per_second_per_gpu:4810.46600281217 mfu:22.691788885398157 +Step 18945 | loss:0.9196454286575317 lr:4.673930877319571e-05 tokens_per_second_per_gpu:5335.258902817987 mfu:25.167326533627296 +Step 18946 | loss:0.7164014577865601 lr:4.673498179417841e-05 tokens_per_second_per_gpu:5183.170732909697 mfu:24.449900687252338 +Step 18947 | loss:0.6945558190345764 lr:4.6730654839717744e-05 tokens_per_second_per_gpu:4918.41336716935 mfu:23.200995020790508 +Step 18948 | loss:0.7247949242591858 lr:4.672632790984623e-05 tokens_per_second_per_gpu:4575.537859437355 mfu:21.58359275022468 +Step 18949 | loss:0.7976065278053284 lr:4.672200100459644e-05 tokens_per_second_per_gpu:4165.89577519997 mfu:19.651241146730314 +Step 18950 | loss:0.9124330878257751 lr:4.671767412400087e-05 tokens_per_second_per_gpu:5354.173041818393 mfu:25.25654775437709 +Step 18951 | loss:0.7266392111778259 lr:4.671334726809211e-05 tokens_per_second_per_gpu:5339.2967059567045 mfu:25.186373539952882 +Step 18952 | loss:0.77708899974823 lr:4.670902043690268e-05 tokens_per_second_per_gpu:5332.761622918866 mfu:25.155546438263254 +Step 18953 | loss:0.8933865427970886 lr:4.6704693630465116e-05 tokens_per_second_per_gpu:5354.461640740845 mfu:25.25790912473435 +Step 18954 | loss:0.6113260388374329 lr:4.6700366848811985e-05 tokens_per_second_per_gpu:3828.4040553466443 mfu:18.0592351221571 +Step 18955 | loss:0.7307493090629578 lr:4.66960400919758e-05 tokens_per_second_per_gpu:4251.211707317166 mfu:20.053690954926157 +Step 18956 | loss:0.7793722748756409 lr:4.669171335998915e-05 tokens_per_second_per_gpu:4336.880483320021 mfu:20.457805187932024 +Step 18957 | loss:0.6015957593917847 lr:4.668738665288452e-05 tokens_per_second_per_gpu:5378.832002051483 mfu:25.372868277049033 +Step 18958 | loss:0.6847601532936096 lr:4.6683059970694494e-05 tokens_per_second_per_gpu:3642.4541228198327 mfu:17.182077564098126 +Step 18959 | loss:0.8502785563468933 lr:4.667873331345158e-05 tokens_per_second_per_gpu:4758.487201104738 mfu:22.44659600924614 +Step 18960 | loss:0.8516431450843811 lr:4.6674406681188346e-05 tokens_per_second_per_gpu:4349.358200517151 mfu:20.51666470886897 +Step 18961 | loss:0.702573299407959 lr:4.667008007393732e-05 tokens_per_second_per_gpu:4749.143756620734 mfu:22.402521387459082 +Step 18962 | loss:0.7131896018981934 lr:4.666575349173104e-05 tokens_per_second_per_gpu:5004.790650270604 mfu:23.6084514026633 +Step 18963 | loss:0.7367777824401855 lr:4.666142693460208e-05 tokens_per_second_per_gpu:5367.507203308267 mfu:25.319447269167373 +Step 18964 | loss:0.8056892156600952 lr:4.665710040258294e-05 tokens_per_second_per_gpu:5330.020691149344 mfu:25.14261699545524 +Step 18965 | loss:0.8023242950439453 lr:4.665277389570617e-05 tokens_per_second_per_gpu:4543.592309582829 mfu:21.432900141088044 +Step 18966 | loss:0.8257021307945251 lr:4.664844741400431e-05 tokens_per_second_per_gpu:3746.912002310835 mfu:17.674823204009208 +Step 18967 | loss:0.8619588613510132 lr:4.664412095750991e-05 tokens_per_second_per_gpu:4329.093537734266 mfu:20.421072837014098 +Step 18968 | loss:0.806540310382843 lr:4.663979452625551e-05 tokens_per_second_per_gpu:4304.480306135887 mfu:20.304968024114203 +Step 18969 | loss:0.90256267786026 lr:4.663546812027363e-05 tokens_per_second_per_gpu:4994.782343042387 mfu:23.56124050987451 +Step 18970 | loss:0.7176757454872131 lr:4.663114173959683e-05 tokens_per_second_per_gpu:5343.484033428507 mfu:25.206125878069113 +Step 18971 | loss:0.8019164204597473 lr:4.662681538425766e-05 tokens_per_second_per_gpu:5004.0133079500865 mfu:23.604784546309148 +Step 18972 | loss:0.5653390884399414 lr:4.662248905428863e-05 tokens_per_second_per_gpu:5186.694869194972 mfu:24.466524639389633 +Step 18973 | loss:0.6212775707244873 lr:4.6618162749722275e-05 tokens_per_second_per_gpu:4337.52389755157 mfu:20.46084028264003 +Step 18974 | loss:0.731934130191803 lr:4.661383647059117e-05 tokens_per_second_per_gpu:4976.041002536962 mfu:23.472834409108003 +Step 18975 | loss:0.7578437328338623 lr:4.660951021692782e-05 tokens_per_second_per_gpu:5208.146763282413 mfu:24.56771688387064 +Step 18976 | loss:0.710599422454834 lr:4.660518398876478e-05 tokens_per_second_per_gpu:5285.237381171867 mfu:24.9313664814709 +Step 18977 | loss:0.6710848212242126 lr:4.660085778613459e-05 tokens_per_second_per_gpu:4133.484448505627 mfu:19.498351388770537 +Step 18978 | loss:0.6842246055603027 lr:4.6596531609069786e-05 tokens_per_second_per_gpu:5332.252890374369 mfu:25.15314666005214 +Step 18979 | loss:0.732323944568634 lr:4.6592205457602887e-05 tokens_per_second_per_gpu:5341.497239841766 mfu:25.196753833738512 +Step 18980 | loss:0.6571762561798096 lr:4.658787933176646e-05 tokens_per_second_per_gpu:3600.559370411373 mfu:16.984452869005583 +Step 18981 | loss:0.5191436409950256 lr:4.658355323159302e-05 tokens_per_second_per_gpu:5272.396979838125 mfu:24.87079611758126 +Step 18982 | loss:0.8797274827957153 lr:4.657922715711511e-05 tokens_per_second_per_gpu:4000.1080527621234 mfu:18.869192173688752 +Step 18983 | loss:0.8206925392150879 lr:4.657490110836527e-05 tokens_per_second_per_gpu:3599.6333935047955 mfu:16.980084877949306 +Step 18984 | loss:0.6145454049110413 lr:4.657057508537604e-05 tokens_per_second_per_gpu:4448.692284236535 mfu:20.985240529915618 +Step 18985 | loss:0.6593955755233765 lr:4.656624908817997e-05 tokens_per_second_per_gpu:5328.631094835876 mfu:25.136062032555028 +Step 18986 | loss:0.7528350949287415 lr:4.6561923116809545e-05 tokens_per_second_per_gpu:4688.435228548193 mfu:22.11614890259794 +Step 18987 | loss:0.8101678490638733 lr:4.655759717129735e-05 tokens_per_second_per_gpu:4608.217642262064 mfu:21.73774886155474 +Step 18988 | loss:0.809097170829773 lr:4.6553271251675906e-05 tokens_per_second_per_gpu:5363.6077574370365 mfu:25.301052917678707 +Step 18989 | loss:0.8042426109313965 lr:4.654894535797774e-05 tokens_per_second_per_gpu:5331.969463313056 mfu:25.151809686246285 +Step 18990 | loss:0.8195116519927979 lr:4.654461949023541e-05 tokens_per_second_per_gpu:5124.991948654232 mfu:24.175461435595146 +Step 18991 | loss:0.8450823426246643 lr:4.654029364848142e-05 tokens_per_second_per_gpu:5320.785573360532 mfu:25.099053369171838 +Step 18992 | loss:0.8132566809654236 lr:4.653596783274836e-05 tokens_per_second_per_gpu:5196.241265052888 mfu:24.51155661743451 +Step 18993 | loss:0.7834866642951965 lr:4.653164204306869e-05 tokens_per_second_per_gpu:4366.313965629968 mfu:20.596647945857967 +Step 18994 | loss:0.7831835746765137 lr:4.6527316279475005e-05 tokens_per_second_per_gpu:4167.334167894019 mfu:19.658026290483242 +Step 18995 | loss:0.6868728995323181 lr:4.6522990541999796e-05 tokens_per_second_per_gpu:4210.701182661356 mfu:19.862595898316474 +Step 18996 | loss:0.7635921239852905 lr:4.651866483067564e-05 tokens_per_second_per_gpu:5149.635627628887 mfu:24.291709873963796 +Step 18997 | loss:0.718268632888794 lr:4.6514339145535035e-05 tokens_per_second_per_gpu:4664.987359196487 mfu:22.005541302246897 +Step 18998 | loss:0.7486860156059265 lr:4.6510013486610524e-05 tokens_per_second_per_gpu:3844.9329329158327 mfu:18.137204657768276 +Step 18999 | loss:0.8289487957954407 lr:4.650568785393466e-05 tokens_per_second_per_gpu:5278.300424975584 mfu:24.898643675526262 +Step 19000 | loss:0.8482836484909058 lr:4.650136224753998e-05 tokens_per_second_per_gpu:5266.882446028677 mfu:24.84478312072561 +Step 19001 | loss:0.6701778173446655 lr:4.6497036667458985e-05 tokens_per_second_per_gpu:5310.453056162831 mfu:25.050313122641064 +Step 19002 | loss:0.7055256366729736 lr:4.6492711113724215e-05 tokens_per_second_per_gpu:4183.038851896797 mfu:19.732107964419736 +Step 19003 | loss:0.7432039976119995 lr:4.648838558636822e-05 tokens_per_second_per_gpu:5330.465279757092 mfu:25.14471419577262 +Step 19004 | loss:0.6415043473243713 lr:4.6484060085423525e-05 tokens_per_second_per_gpu:5331.556343391286 mfu:25.14986092908977 +Step 19005 | loss:0.9403715133666992 lr:4.647973461092266e-05 tokens_per_second_per_gpu:5301.559468920677 mfu:25.008360554217372 +Step 19006 | loss:0.7939723134040833 lr:4.647540916289816e-05 tokens_per_second_per_gpu:4478.413275950869 mfu:21.125439518756224 +Step 19007 | loss:0.7463141679763794 lr:4.647108374138258e-05 tokens_per_second_per_gpu:5342.699236826748 mfu:25.202423858598507 +Step 19008 | loss:0.649819552898407 lr:4.646675834640841e-05 tokens_per_second_per_gpu:4569.806123374518 mfu:21.556555173281264 +Step 19009 | loss:0.7939158082008362 lr:4.6462432978008187e-05 tokens_per_second_per_gpu:5350.11446150072 mfu:25.23740273855342 +Step 19010 | loss:0.8100387454032898 lr:4.645810763621448e-05 tokens_per_second_per_gpu:5335.276178763997 mfu:25.16740802721221 +Step 19011 | loss:0.6469484567642212 lr:4.645378232105977e-05 tokens_per_second_per_gpu:5332.515190642243 mfu:25.154383975168745 +Step 19012 | loss:0.8103153109550476 lr:4.6449457032576635e-05 tokens_per_second_per_gpu:4175.950566306693 mfu:19.69867131190499 +Step 19013 | loss:0.7488770484924316 lr:4.644513177079759e-05 tokens_per_second_per_gpu:4226.997831170518 mfu:19.93946996982914 +Step 19014 | loss:0.6387341022491455 lr:4.644080653575517e-05 tokens_per_second_per_gpu:4697.85619736497 mfu:22.1605892198897 +Step 19015 | loss:0.7348034977912903 lr:4.643648132748188e-05 tokens_per_second_per_gpu:5336.397370717601 mfu:25.17269688095243 +Step 19016 | loss:0.8399035930633545 lr:4.643215614601028e-05 tokens_per_second_per_gpu:4363.622466825564 mfu:20.58395168678054 +Step 19017 | loss:0.8248564600944519 lr:4.642783099137288e-05 tokens_per_second_per_gpu:4665.237117175497 mfu:22.006719453246994 +Step 19018 | loss:0.7512335777282715 lr:4.642350586360222e-05 tokens_per_second_per_gpu:4747.725029478273 mfu:22.395829009468283 +Step 19019 | loss:0.5226696729660034 lr:4.641918076273083e-05 tokens_per_second_per_gpu:4190.164554650241 mfu:19.765721120077995 +Step 19020 | loss:0.7331416606903076 lr:4.641485568879124e-05 tokens_per_second_per_gpu:4677.212804082534 mfu:22.063210811650364 +Step 19021 | loss:0.7946421504020691 lr:4.6410530641815995e-05 tokens_per_second_per_gpu:5341.797263412478 mfu:25.19816909611086 +Step 19022 | loss:0.8132121562957764 lr:4.6406205621837586e-05 tokens_per_second_per_gpu:5344.365867821136 mfu:25.210285641355604 +Step 19023 | loss:0.7165569067001343 lr:4.640188062888857e-05 tokens_per_second_per_gpu:4664.534049911402 mfu:22.00340296500661 +Step 19024 | loss:0.812716007232666 lr:4.639755566300146e-05 tokens_per_second_per_gpu:4004.385486692311 mfu:18.88936955934293 +Step 19025 | loss:0.694025456905365 lr:4.639323072420879e-05 tokens_per_second_per_gpu:3357.3952056458957 mfu:15.837406015722202 +Step 19026 | loss:0.6904807090759277 lr:4.638890581254311e-05 tokens_per_second_per_gpu:5361.821269221589 mfu:25.29262574796012 +Step 19027 | loss:0.7141067385673523 lr:4.6384580928036905e-05 tokens_per_second_per_gpu:3671.5822719964294 mfu:17.31948001353885 +Step 19028 | loss:0.7948521971702576 lr:4.638025607072277e-05 tokens_per_second_per_gpu:5324.610177875396 mfu:25.117094681213274 +Step 19029 | loss:0.6636161804199219 lr:4.637593124063316e-05 tokens_per_second_per_gpu:4263.527739404765 mfu:20.111787779615117 +Step 19030 | loss:0.8853185176849365 lr:4.637160643780064e-05 tokens_per_second_per_gpu:5322.756941289836 mfu:25.108352648044416 +Step 19031 | loss:0.7516043186187744 lr:4.6367281662257725e-05 tokens_per_second_per_gpu:5225.080630794571 mfu:24.647596826141214 +Step 19032 | loss:0.7447916865348816 lr:4.6362956914036956e-05 tokens_per_second_per_gpu:5345.18947419004 mfu:25.21417073311946 +Step 19033 | loss:0.9032319188117981 lr:4.6358632193170845e-05 tokens_per_second_per_gpu:3946.7255496881735 mfu:18.617377798694722 +Step 19034 | loss:0.7781805992126465 lr:4.6354307499691926e-05 tokens_per_second_per_gpu:4737.3411766115305 mfu:22.346846603827867 +Step 19035 | loss:0.7571740746498108 lr:4.634998283363273e-05 tokens_per_second_per_gpu:5284.042704105833 mfu:24.925730985917493 +Step 19036 | loss:0.6969454288482666 lr:4.634565819502579e-05 tokens_per_second_per_gpu:4991.367022234858 mfu:23.54512985891158 +Step 19037 | loss:0.6521902680397034 lr:4.634133358390362e-05 tokens_per_second_per_gpu:4091.248046751756 mfu:19.2991150754243 +Step 19038 | loss:0.7124740481376648 lr:4.633700900029872e-05 tokens_per_second_per_gpu:4977.61916848304 mfu:23.480278887138528 +Step 19039 | loss:0.7712479829788208 lr:4.633268444424367e-05 tokens_per_second_per_gpu:3610.348113046372 mfu:17.030628038146588 +Step 19040 | loss:0.7040891647338867 lr:4.6328359915770955e-05 tokens_per_second_per_gpu:4642.519900623511 mfu:21.899558466814455 +Step 19041 | loss:0.7615323662757874 lr:4.632403541491311e-05 tokens_per_second_per_gpu:5327.567096903276 mfu:25.131042972770192 +Step 19042 | loss:0.7060747146606445 lr:4.631971094170267e-05 tokens_per_second_per_gpu:5171.123678611705 mfu:24.393072676689297 +Step 19043 | loss:0.6417633891105652 lr:4.6315386496172163e-05 tokens_per_second_per_gpu:5152.263078767585 mfu:24.304104009274425 +Step 19044 | loss:0.6743846535682678 lr:4.63110620783541e-05 tokens_per_second_per_gpu:5326.543281382819 mfu:25.126213460279256 +Step 19045 | loss:0.7423336505889893 lr:4.6306737688281e-05 tokens_per_second_per_gpu:5120.712456313318 mfu:24.155274340066374 +Step 19046 | loss:0.6917704343795776 lr:4.63024133259854e-05 tokens_per_second_per_gpu:4220.369891376072 mfu:19.9082048469756 +Step 19047 | loss:0.8167513012886047 lr:4.6298088991499814e-05 tokens_per_second_per_gpu:4026.8260323498316 mfu:18.995225441961356 +Step 19048 | loss:0.7064199447631836 lr:4.629376468485678e-05 tokens_per_second_per_gpu:4385.285182633847 mfu:20.686138413288656 +Step 19049 | loss:0.885997474193573 lr:4.6289440406088825e-05 tokens_per_second_per_gpu:5199.4075770074105 mfu:24.526492651152523 +Step 19050 | loss:0.7117895483970642 lr:4.628511615522846e-05 tokens_per_second_per_gpu:4005.0178151507603 mfu:18.8923523605678 +Step 19051 | loss:0.6942687630653381 lr:4.628079193230821e-05 tokens_per_second_per_gpu:4386.917179791084 mfu:20.69383682232677 +Step 19052 | loss:0.7820848822593689 lr:4.6276467737360584e-05 tokens_per_second_per_gpu:4192.91613353288 mfu:19.77870078713517 +Step 19053 | loss:0.6828718781471252 lr:4.6272143570418134e-05 tokens_per_second_per_gpu:4916.318888505517 mfu:23.1911150075783 +Step 19054 | loss:0.8849661946296692 lr:4.6267819431513355e-05 tokens_per_second_per_gpu:4698.772832647616 mfu:22.164913144912052 +Step 19055 | loss:0.8401700258255005 lr:4.626349532067879e-05 tokens_per_second_per_gpu:5308.920194593201 mfu:25.043082353083967 +Step 19056 | loss:0.8052963614463806 lr:4.625917123794695e-05 tokens_per_second_per_gpu:4460.069686516434 mfu:21.0389096776549 +Step 19057 | loss:0.7276429533958435 lr:4.625484718335038e-05 tokens_per_second_per_gpu:5318.975589576471 mfu:25.090515366847345 +Step 19058 | loss:0.6715927720069885 lr:4.625052315692156e-05 tokens_per_second_per_gpu:5387.109845188527 mfu:25.411916275472144 +Step 19059 | loss:0.8245025277137756 lr:4.624619915869304e-05 tokens_per_second_per_gpu:4723.593573416017 mfu:22.281996813971656 +Step 19060 | loss:0.8041634559631348 lr:4.624187518869734e-05 tokens_per_second_per_gpu:4657.962255207416 mfu:21.972402688124113 +Step 19061 | loss:0.7207255363464355 lr:4.623755124696696e-05 tokens_per_second_per_gpu:3878.8042757449375 mfu:18.296981560940164 +Step 19062 | loss:0.7667314410209656 lr:4.6233227333534446e-05 tokens_per_second_per_gpu:4165.831227923056 mfu:19.65093666621221 +Step 19063 | loss:0.7608040571212769 lr:4.62289034484323e-05 tokens_per_second_per_gpu:5371.076963737877 mfu:25.336286438177776 +Step 19064 | loss:0.7670415043830872 lr:4.622457959169307e-05 tokens_per_second_per_gpu:4300.330971880378 mfu:20.285394906481542 +Step 19065 | loss:0.6752038598060608 lr:4.622025576334926e-05 tokens_per_second_per_gpu:5312.750058848458 mfu:25.061148476218236 +Step 19066 | loss:0.7327662110328674 lr:4.621593196343339e-05 tokens_per_second_per_gpu:5363.078306768386 mfu:25.298555408541084 +Step 19067 | loss:0.8607542514801025 lr:4.621160819197796e-05 tokens_per_second_per_gpu:5356.559918023553 mfu:25.267807056679906 +Step 19068 | loss:0.8211444616317749 lr:4.620728444901552e-05 tokens_per_second_per_gpu:4323.638151921851 mfu:20.395338851352143 +Step 19069 | loss:0.8021722435951233 lr:4.620296073457858e-05 tokens_per_second_per_gpu:3991.4273678340837 mfu:18.828243901805617 +Step 19070 | loss:0.7384393215179443 lr:4.619863704869965e-05 tokens_per_second_per_gpu:3626.4964410699604 mfu:17.10680245662272 +Step 19071 | loss:0.7672584652900696 lr:4.6194313391411257e-05 tokens_per_second_per_gpu:5275.176193277488 mfu:24.883906141557446 +Step 19072 | loss:0.8010462522506714 lr:4.6189989762745925e-05 tokens_per_second_per_gpu:4022.634037521281 mfu:18.97545108712701 +Step 19073 | loss:0.6921969652175903 lr:4.6185666162736165e-05 tokens_per_second_per_gpu:4822.316917139585 mfu:22.74769167025506 +Step 19074 | loss:0.7940671443939209 lr:4.618134259141448e-05 tokens_per_second_per_gpu:5297.515876531139 mfu:24.989286239007914 +Step 19075 | loss:0.7752940654754639 lr:4.6177019048813425e-05 tokens_per_second_per_gpu:4232.001522659559 mfu:19.963073236300744 +Step 19076 | loss:0.8705703616142273 lr:4.61726955349655e-05 tokens_per_second_per_gpu:3696.1082179509654 mfu:17.435173085164273 +Step 19077 | loss:0.7344604730606079 lr:4.616837204990319e-05 tokens_per_second_per_gpu:5003.7423745704555 mfu:23.603506507331424 +Step 19078 | loss:0.6918625235557556 lr:4.616404859365907e-05 tokens_per_second_per_gpu:5238.340873580174 mfu:24.710147653791616 +Step 19079 | loss:0.7404078245162964 lr:4.6159725166265635e-05 tokens_per_second_per_gpu:5273.908245953283 mfu:24.877925017695052 +Step 19080 | loss:0.7316262722015381 lr:4.6155401767755384e-05 tokens_per_second_per_gpu:3594.1117424948948 mfu:16.954038308044797 +Step 19081 | loss:0.8919388651847839 lr:4.615107839816084e-05 tokens_per_second_per_gpu:5273.730184303091 mfu:24.877085070510645 +Step 19082 | loss:0.6610308289527893 lr:4.6146755057514534e-05 tokens_per_second_per_gpu:5344.741779645456 mfu:25.21205888156809 +Step 19083 | loss:0.8458839654922485 lr:4.614243174584897e-05 tokens_per_second_per_gpu:3919.464282988284 mfu:18.488781752419623 +Step 19084 | loss:0.7023531198501587 lr:4.613810846319668e-05 tokens_per_second_per_gpu:4454.027001577033 mfu:21.010405301807385 +Step 19085 | loss:0.723409116268158 lr:4.6133785209590165e-05 tokens_per_second_per_gpu:4671.502973882358 mfu:22.036276568398648 +Step 19086 | loss:0.7534900903701782 lr:4.612946198506195e-05 tokens_per_second_per_gpu:5339.489122633116 mfu:25.187281202245185 +Step 19087 | loss:0.7688522338867188 lr:4.612513878964454e-05 tokens_per_second_per_gpu:3604.6568769251426 mfu:17.003781506337734 +Step 19088 | loss:0.6255954504013062 lr:4.612081562337045e-05 tokens_per_second_per_gpu:5337.287530370649 mfu:25.176895915912752 +Step 19089 | loss:0.8126263618469238 lr:4.6116492486272206e-05 tokens_per_second_per_gpu:4014.6724831212396 mfu:18.937895076640764 +Step 19090 | loss:0.8973973989486694 lr:4.61121693783823e-05 tokens_per_second_per_gpu:3862.803187195091 mfu:18.22150169618332 +Step 19091 | loss:0.6826013326644897 lr:4.6107846299733286e-05 tokens_per_second_per_gpu:4386.771692155 mfu:20.693150532324612 +Step 19092 | loss:0.7310135364532471 lr:4.6103523250357657e-05 tokens_per_second_per_gpu:3783.942478325964 mfu:17.849502277423277 +Step 19093 | loss:0.6873472332954407 lr:4.609920023028791e-05 tokens_per_second_per_gpu:5253.273156608141 mfu:24.780585780544733 +Step 19094 | loss:0.6786638498306274 lr:4.60948772395566e-05 tokens_per_second_per_gpu:5124.120824651517 mfu:24.17135219504526 +Step 19095 | loss:0.7253159284591675 lr:4.6090554278196204e-05 tokens_per_second_per_gpu:3832.143225585531 mfu:18.076873426142157 +Step 19096 | loss:0.8619898557662964 lr:4.608623134623925e-05 tokens_per_second_per_gpu:5340.185876631871 mfu:25.1905679097358 +Step 19097 | loss:0.6894755363464355 lr:4.6081908443718235e-05 tokens_per_second_per_gpu:4283.026470761275 mfu:20.203766622250395 +Step 19098 | loss:0.8215869069099426 lr:4.607758557066571e-05 tokens_per_second_per_gpu:5306.585600275745 mfu:25.032069673365662 +Step 19099 | loss:0.6051505208015442 lr:4.607326272711414e-05 tokens_per_second_per_gpu:5319.569148682822 mfu:25.09331528642363 +Step 19100 | loss:0.7702750563621521 lr:4.606893991309608e-05 tokens_per_second_per_gpu:5302.05363463785 mfu:25.01069161821864 +Step 19101 | loss:0.6914733648300171 lr:4.606461712864403e-05 tokens_per_second_per_gpu:4658.91153526944 mfu:21.976880604140558 +Step 19102 | loss:0.7225441932678223 lr:4.606029437379048e-05 tokens_per_second_per_gpu:5318.054575222928 mfu:25.086170785000434 +Step 19103 | loss:0.7661318778991699 lr:4.605597164856797e-05 tokens_per_second_per_gpu:5341.929681550236 mfu:25.198793735059528 +Step 19104 | loss:0.8365271687507629 lr:4.605164895300898e-05 tokens_per_second_per_gpu:4234.801147739596 mfu:19.976279545468575 +Step 19105 | loss:0.7027114033699036 lr:4.6047326287146066e-05 tokens_per_second_per_gpu:5340.9261625841455 mfu:25.194059964878488 +Step 19106 | loss:0.80661940574646 lr:4.6043003651011696e-05 tokens_per_second_per_gpu:5363.7341452723 mfu:25.301649110661636 +Step 19107 | loss:0.6783691644668579 lr:4.603868104463842e-05 tokens_per_second_per_gpu:4756.513390850259 mfu:22.437285209509007 +Step 19108 | loss:0.7998482584953308 lr:4.603435846805873e-05 tokens_per_second_per_gpu:4516.926584029593 mfu:21.307113364011503 +Step 19109 | loss:0.8591249585151672 lr:4.6030035921305127e-05 tokens_per_second_per_gpu:5313.984739270801 mfu:25.06697267442863 +Step 19110 | loss:0.8317081332206726 lr:4.602571340441012e-05 tokens_per_second_per_gpu:5018.1943369831715 mfu:23.671678879791255 +Step 19111 | loss:0.6519308686256409 lr:4.6021390917406246e-05 tokens_per_second_per_gpu:3788.620462479116 mfu:17.87156912682009 +Step 19112 | loss:0.7460602521896362 lr:4.601706846032599e-05 tokens_per_second_per_gpu:4306.477460939289 mfu:20.314388944071595 +Step 19113 | loss:0.8092141151428223 lr:4.601274603320186e-05 tokens_per_second_per_gpu:5330.3206397340755 mfu:25.144031904105923 +Step 19114 | loss:0.7786465287208557 lr:4.6008423636066396e-05 tokens_per_second_per_gpu:5313.303303960023 mfu:25.06375822780284 +Step 19115 | loss:0.7399904727935791 lr:4.6004101268952095e-05 tokens_per_second_per_gpu:4457.942723020529 mfu:21.028876427947328 +Step 19116 | loss:0.8877273797988892 lr:4.599977893189144e-05 tokens_per_second_per_gpu:5183.17870091911 mfu:24.449938273712224 +Step 19117 | loss:0.7288684844970703 lr:4.599545662491696e-05 tokens_per_second_per_gpu:5224.316678107721 mfu:24.64399313097357 +Step 19118 | loss:0.8904485702514648 lr:4.599113434806116e-05 tokens_per_second_per_gpu:4735.69032581797 mfu:22.339059258970455 +Step 19119 | loss:0.602436363697052 lr:4.598681210135655e-05 tokens_per_second_per_gpu:5333.813270108805 mfu:25.160507237487714 +Step 19120 | loss:0.7566646337509155 lr:4.5982489884835634e-05 tokens_per_second_per_gpu:4273.369950704916 mfu:20.158215169572337 +Step 19121 | loss:0.7506098747253418 lr:4.597816769853093e-05 tokens_per_second_per_gpu:4637.695673185113 mfu:21.876801762889457 +Step 19122 | loss:0.7933638095855713 lr:4.597384554247496e-05 tokens_per_second_per_gpu:4947.780123659948 mfu:23.33952302164198 +Step 19123 | loss:0.7638960480690002 lr:4.5969523416700187e-05 tokens_per_second_per_gpu:3570.5165419338064 mfu:16.842735721241805 +Step 19124 | loss:0.8377088904380798 lr:4.5965201321239144e-05 tokens_per_second_per_gpu:5175.049433987289 mfu:24.411591134598865 +Step 19125 | loss:0.7078859806060791 lr:4.5960879256124344e-05 tokens_per_second_per_gpu:5273.05619853187 mfu:24.873905764633964 +Step 19126 | loss:0.7786343693733215 lr:4.595655722138827e-05 tokens_per_second_per_gpu:3573.0163203339503 mfu:16.85452760246146 +Step 19127 | loss:0.7781251072883606 lr:4.595223521706346e-05 tokens_per_second_per_gpu:5371.729316120272 mfu:25.33936369566833 +Step 19128 | loss:0.8306214213371277 lr:4.5947913243182403e-05 tokens_per_second_per_gpu:3848.0425946066302 mfu:18.151873462526762 +Step 19129 | loss:0.754005491733551 lr:4.5943591299777605e-05 tokens_per_second_per_gpu:4116.723242831764 mfu:19.419285921850346 +Step 19130 | loss:0.9402173161506653 lr:4.593926938688159e-05 tokens_per_second_per_gpu:3746.8243521918544 mfu:17.674409743443324 +Step 19131 | loss:0.6265972256660461 lr:4.5934947504526835e-05 tokens_per_second_per_gpu:4443.371503936126 mfu:20.960141501419862 +Step 19132 | loss:0.7958246469497681 lr:4.593062565274586e-05 tokens_per_second_per_gpu:5343.549843641843 mfu:25.206436316073795 +Step 19133 | loss:0.7580833435058594 lr:4.592630383157116e-05 tokens_per_second_per_gpu:4377.237479653521 mfu:20.64817602525074 +Step 19134 | loss:0.8351393938064575 lr:4.592198204103526e-05 tokens_per_second_per_gpu:4344.263475102213 mfu:20.49263201983692 +Step 19135 | loss:0.6623545289039612 lr:4.591766028117064e-05 tokens_per_second_per_gpu:4966.768334340303 mfu:23.429093651144033 +Step 19136 | loss:0.6714355945587158 lr:4.591333855200983e-05 tokens_per_second_per_gpu:5340.545973764715 mfu:25.192266549350332 +Step 19137 | loss:0.7249627709388733 lr:4.590901685358534e-05 tokens_per_second_per_gpu:4748.404409892993 mfu:22.399033762798762 +Step 19138 | loss:0.732937753200531 lr:4.5904695185929635e-05 tokens_per_second_per_gpu:4599.067156713071 mfu:21.694584459986313 +Step 19139 | loss:0.8208088278770447 lr:4.590037354907525e-05 tokens_per_second_per_gpu:3793.3271030163446 mfu:17.89377115326988 +Step 19140 | loss:0.7687187790870667 lr:4.589605194305466e-05 tokens_per_second_per_gpu:5351.96519440397 mfu:25.246132961425648 +Step 19141 | loss:0.6382229328155518 lr:4.58917303679004e-05 tokens_per_second_per_gpu:3646.7804566428736 mfu:17.202485618890076 +Step 19142 | loss:0.791304886341095 lr:4.5887408823644944e-05 tokens_per_second_per_gpu:3604.0239139582013 mfu:17.000795712028054 +Step 19143 | loss:0.7690695524215698 lr:4.5883087310320826e-05 tokens_per_second_per_gpu:4952.060395130027 mfu:23.359713792455963 +Step 19144 | loss:0.8167742490768433 lr:4.5878765827960546e-05 tokens_per_second_per_gpu:4542.567088543125 mfu:21.428063998523008 +Step 19145 | loss:0.7107206583023071 lr:4.5874444376596584e-05 tokens_per_second_per_gpu:4390.4385779948825 mfu:20.710447858466576 +Step 19146 | loss:0.7724275588989258 lr:4.5870122956261444e-05 tokens_per_second_per_gpu:3747.6676432139234 mfu:17.678387691074708 +Step 19147 | loss:0.7675480842590332 lr:4.586580156698764e-05 tokens_per_second_per_gpu:4629.343529582703 mfu:21.837403276494246 +Step 19148 | loss:0.8395246863365173 lr:4.5861480208807675e-05 tokens_per_second_per_gpu:3603.7669545292974 mfu:16.999583590560118 +Step 19149 | loss:0.7875975370407104 lr:4.5857158881754024e-05 tokens_per_second_per_gpu:5353.510194469937 mfu:25.253420990341006 +Step 19150 | loss:0.8098015785217285 lr:4.585283758585923e-05 tokens_per_second_per_gpu:4566.514574621785 mfu:21.541028376218495 +Step 19151 | loss:0.7105826735496521 lr:4.5848516321155785e-05 tokens_per_second_per_gpu:3780.089124627115 mfu:17.83132535057587 +Step 19152 | loss:0.7327703237533569 lr:4.584419508767617e-05 tokens_per_second_per_gpu:4407.111645782835 mfu:20.78909756394105 +Step 19153 | loss:0.7484419345855713 lr:4.583987388545288e-05 tokens_per_second_per_gpu:3947.8094838175334 mfu:18.622490900921143 +Step 19154 | loss:0.7792428135871887 lr:4.583555271451844e-05 tokens_per_second_per_gpu:5321.9508543841475 mfu:25.10455019857768 +Step 19155 | loss:0.8016100525856018 lr:4.583123157490533e-05 tokens_per_second_per_gpu:4415.655110588359 mfu:20.82939854509411 +Step 19156 | loss:0.7986910939216614 lr:4.5826910466646064e-05 tokens_per_second_per_gpu:3891.037417256642 mfu:18.354687376640044 +Step 19157 | loss:0.9148320555686951 lr:4.5822589389773135e-05 tokens_per_second_per_gpu:3430.351774277175 mfu:16.181554597630118 +Step 19158 | loss:0.8646200895309448 lr:4.581826834431904e-05 tokens_per_second_per_gpu:5098.618808714731 mfu:24.051054834778853 +Step 19159 | loss:0.7375779747962952 lr:4.5813947330316314e-05 tokens_per_second_per_gpu:4290.865131891119 mfu:20.240742924213368 +Step 19160 | loss:0.8584005236625671 lr:4.5809626347797394e-05 tokens_per_second_per_gpu:4892.277210494059 mfu:23.077706310465004 +Step 19161 | loss:0.9357558488845825 lr:4.580530539679482e-05 tokens_per_second_per_gpu:3615.1039385470317 mfu:17.053062078460698 +Step 19162 | loss:0.7982240319252014 lr:4.580098447734106e-05 tokens_per_second_per_gpu:5353.349936228102 mfu:25.252665024871394 +Step 19163 | loss:0.753085732460022 lr:4.579666358946865e-05 tokens_per_second_per_gpu:5334.288818843802 mfu:25.162750482007674 +Step 19164 | loss:0.7125101089477539 lr:4.579234273321007e-05 tokens_per_second_per_gpu:4507.1056385223765 mfu:21.260786288427873 +Step 19165 | loss:0.7230919003486633 lr:4.57880219085978e-05 tokens_per_second_per_gpu:4380.012108365655 mfu:20.66126442228635 +Step 19166 | loss:0.7453135848045349 lr:4.5783701115664385e-05 tokens_per_second_per_gpu:4626.692069420535 mfu:21.824895886523546 +Step 19167 | loss:0.7510438561439514 lr:4.577938035444228e-05 tokens_per_second_per_gpu:4340.500515004934 mfu:20.47488149503085 +Step 19168 | loss:0.6842538714408875 lr:4.577505962496399e-05 tokens_per_second_per_gpu:5379.191096345511 mfu:25.374562186101766 +Step 19169 | loss:0.7626570463180542 lr:4.5770738927262e-05 tokens_per_second_per_gpu:4427.706975948509 mfu:20.886249250258203 +Step 19170 | loss:0.695019543170929 lr:4.576641826136884e-05 tokens_per_second_per_gpu:4213.046575949227 mfu:19.87365951862072 +Step 19171 | loss:0.7963865995407104 lr:4.576209762731699e-05 tokens_per_second_per_gpu:4101.827913394082 mfu:19.34902211148754 +Step 19172 | loss:0.702517032623291 lr:4.575777702513893e-05 tokens_per_second_per_gpu:5246.570283149917 mfu:24.748967182814148 +Step 19173 | loss:0.693489670753479 lr:4.57534564548672e-05 tokens_per_second_per_gpu:3348.6121992624207 mfu:15.795975076075937 +Step 19174 | loss:0.6847789883613586 lr:4.5749135916534245e-05 tokens_per_second_per_gpu:4637.688802520079 mfu:21.876769352790234 +Step 19175 | loss:0.6573309302330017 lr:4.574481541017258e-05 tokens_per_second_per_gpu:4615.623873990404 mfu:21.77268532025075 +Step 19176 | loss:0.7648906111717224 lr:4.5740494935814696e-05 tokens_per_second_per_gpu:4714.916475103884 mfu:22.241065460767693 +Step 19177 | loss:0.7022067308425903 lr:4.57361744934931e-05 tokens_per_second_per_gpu:5327.719861426782 mfu:25.131763589091882 +Step 19178 | loss:0.9544957280158997 lr:4.5731854083240275e-05 tokens_per_second_per_gpu:5058.750272663168 mfu:23.86298814794995 +Step 19179 | loss:0.7726221084594727 lr:4.572753370508873e-05 tokens_per_second_per_gpu:3797.2507619582243 mfu:17.912279721943975 +Step 19180 | loss:0.8054963946342468 lr:4.5723213359070946e-05 tokens_per_second_per_gpu:3120.569848911833 mfu:14.720260401435176 +Step 19181 | loss:0.7980296611785889 lr:4.5718893045219424e-05 tokens_per_second_per_gpu:4485.97011794787 mfu:21.16108643178627 +Step 19182 | loss:0.8634037375450134 lr:4.571457276356664e-05 tokens_per_second_per_gpu:5133.207618511108 mfu:24.21421615204785 +Step 19183 | loss:0.7233977913856506 lr:4.571025251414511e-05 tokens_per_second_per_gpu:5360.313136948195 mfu:25.285511630712623 +Step 19184 | loss:0.6954682469367981 lr:4.570593229698732e-05 tokens_per_second_per_gpu:4556.434728209626 mfu:21.493480020893205 +Step 19185 | loss:0.8689793348312378 lr:4.570161211212575e-05 tokens_per_second_per_gpu:3294.2917810511663 mfu:15.539736395354417 +Step 19186 | loss:0.8800726532936096 lr:4.5697291959592915e-05 tokens_per_second_per_gpu:5289.493839128362 mfu:24.951444919878302 +Step 19187 | loss:0.7479926347732544 lr:4.56929718394213e-05 tokens_per_second_per_gpu:4786.447241732349 mfu:22.578488291361424 +Step 19188 | loss:0.8022549748420715 lr:4.56886517516434e-05 tokens_per_second_per_gpu:5373.486135727828 mfu:25.3476509135015 +Step 19189 | loss:0.7154977321624756 lr:4.5684331696291675e-05 tokens_per_second_per_gpu:3954.359987916415 mfu:18.653390746386815 +Step 19190 | loss:0.8849712014198303 lr:4.5680011673398654e-05 tokens_per_second_per_gpu:4227.835777429005 mfu:19.943422705298797 +Step 19191 | loss:0.6587271690368652 lr:4.567569168299682e-05 tokens_per_second_per_gpu:5357.281095098727 mfu:25.271208971988997 +Step 19192 | loss:0.7682757377624512 lr:4.567137172511865e-05 tokens_per_second_per_gpu:4340.385700943944 mfu:20.474339897515865 +Step 19193 | loss:0.6310640573501587 lr:4.566705179979665e-05 tokens_per_second_per_gpu:5373.705357322375 mfu:25.34868501916604 +Step 19194 | loss:0.6659570336341858 lr:4.566273190706331e-05 tokens_per_second_per_gpu:4546.027123086325 mfu:21.444385571806055 +Step 19195 | loss:0.6710656881332397 lr:4.5658412046951134e-05 tokens_per_second_per_gpu:5362.178200339449 mfu:25.294309452941796 +Step 19196 | loss:0.863543689250946 lr:4.565409221949257e-05 tokens_per_second_per_gpu:4613.034731411935 mfu:21.76047189295484 +Step 19197 | loss:0.8189570903778076 lr:4.5649772424720136e-05 tokens_per_second_per_gpu:4039.5940680725876 mfu:19.055454444917125 +Step 19198 | loss:0.6576413512229919 lr:4.564545266266632e-05 tokens_per_second_per_gpu:5335.470938532899 mfu:25.16832674227156 +Step 19199 | loss:0.7878388166427612 lr:4.564113293336361e-05 tokens_per_second_per_gpu:5346.546982784235 mfu:25.22057433277313 +Step 19200 | loss:0.754382848739624 lr:4.56368132368445e-05 tokens_per_second_per_gpu:3783.9513000481566 mfu:17.849543890992 +Step 19201 | loss:0.8855203986167908 lr:4.5632493573141464e-05 tokens_per_second_per_gpu:4111.885536668267 mfu:19.39646563647935 +Step 19202 | loss:0.7209973931312561 lr:4.5628173942287024e-05 tokens_per_second_per_gpu:3284.4191631470035 mfu:15.493165572258173 +Step 19203 | loss:0.7341477274894714 lr:4.562385434431363e-05 tokens_per_second_per_gpu:4227.99851545347 mfu:19.94419036832617 +Step 19204 | loss:0.6840335726737976 lr:4.561953477925379e-05 tokens_per_second_per_gpu:5315.715898485106 mfu:25.075138847808727 +Step 19205 | loss:0.7634604573249817 lr:4.561521524713997e-05 tokens_per_second_per_gpu:4982.94101015213 mfu:23.505382922291446 +Step 19206 | loss:0.7397194504737854 lr:4.56108957480047e-05 tokens_per_second_per_gpu:5369.717272619239 mfu:25.329872543183935 +Step 19207 | loss:0.6370387673377991 lr:4.5606576281880434e-05 tokens_per_second_per_gpu:5187.402100432632 mfu:24.46986076980377 +Step 19208 | loss:0.6850961446762085 lr:4.560225684879966e-05 tokens_per_second_per_gpu:5340.219377193503 mfu:25.190725937600806 +Step 19209 | loss:0.7100234627723694 lr:4.559793744879489e-05 tokens_per_second_per_gpu:5160.559112669319 mfu:24.343237816638243 +Step 19210 | loss:0.7529265284538269 lr:4.559361808189859e-05 tokens_per_second_per_gpu:5359.028664690943 mfu:25.279452555921598 +Step 19211 | loss:0.7745354175567627 lr:4.558929874814325e-05 tokens_per_second_per_gpu:3909.6305387988577 mfu:18.442394303268326 +Step 19212 | loss:0.6267595291137695 lr:4.558497944756134e-05 tokens_per_second_per_gpu:4895.8103826628185 mfu:23.09437288640638 +Step 19213 | loss:0.7641711235046387 lr:4.558066018018538e-05 tokens_per_second_per_gpu:5347.069379443435 mfu:25.223038566944634 +Step 19214 | loss:0.6766656637191772 lr:4.557634094604783e-05 tokens_per_second_per_gpu:4528.510648891101 mfu:21.361757374408505 +Step 19215 | loss:0.8026951551437378 lr:4.5572021745181194e-05 tokens_per_second_per_gpu:3654.465682911806 mfu:17.23873814243544 +Step 19216 | loss:0.7990597486495972 lr:4.5567702577617936e-05 tokens_per_second_per_gpu:5298.395470997417 mfu:24.99343543617964 +Step 19217 | loss:0.6651251912117004 lr:4.5563383443390575e-05 tokens_per_second_per_gpu:5331.168819785209 mfu:25.148032914121558 +Step 19218 | loss:0.7861267924308777 lr:4.555906434253155e-05 tokens_per_second_per_gpu:5097.291446577618 mfu:24.044793441106755 +Step 19219 | loss:0.78970867395401 lr:4.5554745275073376e-05 tokens_per_second_per_gpu:5172.954546646407 mfu:24.401709193587283 +Step 19220 | loss:0.5898967981338501 lr:4.5550426241048536e-05 tokens_per_second_per_gpu:4985.994841625568 mfu:23.519788366389033 +Step 19221 | loss:0.8728576302528381 lr:4.5546107240489486e-05 tokens_per_second_per_gpu:3044.2445113729814 mfu:14.360220761818997 +Step 19222 | loss:0.6734367609024048 lr:4.554178827342875e-05 tokens_per_second_per_gpu:4925.95707700685 mfu:23.23657998718362 +Step 19223 | loss:0.6842511892318726 lr:4.5537469339898795e-05 tokens_per_second_per_gpu:5357.8855828665555 mfu:25.274060443926096 +Step 19224 | loss:0.793220579624176 lr:4.553315043993212e-05 tokens_per_second_per_gpu:4006.557946057436 mfu:18.899617420828015 +Step 19225 | loss:0.7367203831672668 lr:4.552883157356115e-05 tokens_per_second_per_gpu:5374.097302002818 mfu:25.350533889096347 +Step 19226 | loss:0.6466994285583496 lr:4.552451274081843e-05 tokens_per_second_per_gpu:5067.19756983373 mfu:23.902835489961003 +Step 19227 | loss:0.798291802406311 lr:4.552019394173642e-05 tokens_per_second_per_gpu:4668.7302434041185 mfu:22.02319712565619 +Step 19228 | loss:0.9450502395629883 lr:4.551587517634759e-05 tokens_per_second_per_gpu:5351.827177394883 mfu:25.24548191164565 +Step 19229 | loss:0.6334018111228943 lr:4.5511556444684457e-05 tokens_per_second_per_gpu:5354.265588247617 mfu:25.25698431167379 +Step 19230 | loss:0.745816707611084 lr:4.550723774677946e-05 tokens_per_second_per_gpu:4417.896282579402 mfu:20.839970535759235 +Step 19231 | loss:0.6783266067504883 lr:4.550291908266513e-05 tokens_per_second_per_gpu:5309.143602342935 mfu:25.044136205556757 +Step 19232 | loss:0.7394876480102539 lr:4.5498600452373895e-05 tokens_per_second_per_gpu:4640.229508836057 mfu:21.8887542979719 +Step 19233 | loss:0.7632239460945129 lr:4.549428185593826e-05 tokens_per_second_per_gpu:4658.147204245568 mfu:21.973275124292634 +Step 19234 | loss:0.727825403213501 lr:4.548996329339071e-05 tokens_per_second_per_gpu:4058.9567150912903 mfu:19.146791354512583 +Step 19235 | loss:0.820546567440033 lr:4.5485644764763724e-05 tokens_per_second_per_gpu:3902.503498826997 mfu:18.408774839722653 +Step 19236 | loss:0.6659092903137207 lr:4.548132627008978e-05 tokens_per_second_per_gpu:4892.243590872166 mfu:23.07754772097244 +Step 19237 | loss:0.7920375466346741 lr:4.547700780940135e-05 tokens_per_second_per_gpu:5336.439055160829 mfu:25.172893513583237 +Step 19238 | loss:0.7621263861656189 lr:4.547268938273094e-05 tokens_per_second_per_gpu:4523.900124030904 mfu:21.3400087420063 +Step 19239 | loss:0.585229754447937 lr:4.5468370990111006e-05 tokens_per_second_per_gpu:5391.18613622604 mfu:25.43114483578306 +Step 19240 | loss:0.8327098488807678 lr:4.546405263157402e-05 tokens_per_second_per_gpu:5325.547415830405 mfu:25.12151579255633 +Step 19241 | loss:0.6902269124984741 lr:4.545973430715248e-05 tokens_per_second_per_gpu:5317.839002840886 mfu:25.085153893293896 +Step 19242 | loss:0.719505250453949 lr:4.545541601687885e-05 tokens_per_second_per_gpu:5370.272935022072 mfu:25.332493697543878 +Step 19243 | loss:0.7763761878013611 lr:4.545109776078562e-05 tokens_per_second_per_gpu:5360.597290379583 mfu:25.286852030930053 +Step 19244 | loss:0.9988098740577698 lr:4.544677953890526e-05 tokens_per_second_per_gpu:4423.886834068467 mfu:20.86822898922663 +Step 19245 | loss:0.6714330315589905 lr:4.544246135127027e-05 tokens_per_second_per_gpu:4239.980290662598 mfu:20.00071044628928 +Step 19246 | loss:0.7132164239883423 lr:4.54381431979131e-05 tokens_per_second_per_gpu:3513.0767493763756 mfu:16.571782419509223 +Step 19247 | loss:0.7646744251251221 lr:4.5433825078866235e-05 tokens_per_second_per_gpu:3651.7332826107872 mfu:17.225848944019972 +Step 19248 | loss:0.7408355474472046 lr:4.542950699416215e-05 tokens_per_second_per_gpu:4175.902164364327 mfu:19.698442991684626 +Step 19249 | loss:0.7605862617492676 lr:4.542518894383333e-05 tokens_per_second_per_gpu:5371.914743892536 mfu:25.340238390107338 +Step 19250 | loss:0.6757081747055054 lr:4.542087092791224e-05 tokens_per_second_per_gpu:5296.474878508945 mfu:24.98437567372465 +Step 19251 | loss:0.6969057321548462 lr:4.541655294643137e-05 tokens_per_second_per_gpu:3533.301422430516 mfu:16.66718565299065 +Step 19252 | loss:0.5986685752868652 lr:4.54122349994232e-05 tokens_per_second_per_gpu:4403.779103117414 mfu:20.773377391597847 +Step 19253 | loss:0.7729530334472656 lr:4.54079170869202e-05 tokens_per_second_per_gpu:4352.273198560009 mfu:20.530415251987083 +Step 19254 | loss:0.7816729545593262 lr:4.540359920895482e-05 tokens_per_second_per_gpu:4288.783549700477 mfu:20.230923745865564 +Step 19255 | loss:0.7987307906150818 lr:4.5399281365559564e-05 tokens_per_second_per_gpu:5087.826392604551 mfu:24.000145166611066 +Step 19256 | loss:0.8230199217796326 lr:4.5394963556766906e-05 tokens_per_second_per_gpu:4270.154943065196 mfu:20.143049429998907 +Step 19257 | loss:0.844934344291687 lr:4.5390645782609304e-05 tokens_per_second_per_gpu:5022.4441559059915 mfu:23.691726000744076 +Step 19258 | loss:0.728452205657959 lr:4.5386328043119255e-05 tokens_per_second_per_gpu:3744.3385513759777 mfu:17.66268379687348 +Step 19259 | loss:0.7405859231948853 lr:4.538201033832922e-05 tokens_per_second_per_gpu:3167.8666717039137 mfu:14.943367584215506 +Step 19260 | loss:0.7903469800949097 lr:4.537769266827168e-05 tokens_per_second_per_gpu:4629.894249833383 mfu:21.840001117879265 +Step 19261 | loss:0.8551168441772461 lr:4.5373375032979085e-05 tokens_per_second_per_gpu:4152.9797826296335 mfu:19.590314206080468 +Step 19262 | loss:0.8048484921455383 lr:4.536905743248394e-05 tokens_per_second_per_gpu:4987.140259070267 mfu:23.525191495904046 +Step 19263 | loss:0.8148447275161743 lr:4.53647398668187e-05 tokens_per_second_per_gpu:4774.4289220653445 mfu:22.52179582695539 +Step 19264 | loss:0.6884327530860901 lr:4.536042233601584e-05 tokens_per_second_per_gpu:4590.6695916977515 mfu:21.654971713036716 +Step 19265 | loss:0.7893345355987549 lr:4.535610484010784e-05 tokens_per_second_per_gpu:5335.756537579905 mfu:25.16967396170451 +Step 19266 | loss:0.7798071503639221 lr:4.5351787379127156e-05 tokens_per_second_per_gpu:5345.610332850482 mfu:25.216155995226657 +Step 19267 | loss:0.8411855101585388 lr:4.534746995310631e-05 tokens_per_second_per_gpu:4590.694755878646 mfu:21.655090416771447 +Step 19268 | loss:0.6816025376319885 lr:4.53431525620777e-05 tokens_per_second_per_gpu:3740.6947684543793 mfu:17.645495451138828 +Step 19269 | loss:0.6236225366592407 lr:4.533883520607385e-05 tokens_per_second_per_gpu:5329.982677344412 mfu:25.14243767785155 +Step 19270 | loss:0.7900913953781128 lr:4.53345178851272e-05 tokens_per_second_per_gpu:4546.9367858814485 mfu:21.448676606415127 +Step 19271 | loss:0.6887249946594238 lr:4.533020059927024e-05 tokens_per_second_per_gpu:5365.486000120427 mfu:25.309912908877507 +Step 19272 | loss:0.8482714891433716 lr:4.532588334853545e-05 tokens_per_second_per_gpu:5216.39663888521 mfu:24.606632954662956 +Step 19273 | loss:0.6918697357177734 lr:4.5321566132955276e-05 tokens_per_second_per_gpu:5338.558439817236 mfu:25.18289101261182 +Step 19274 | loss:0.8220997452735901 lr:4.531724895256221e-05 tokens_per_second_per_gpu:3407.4764267923792 mfu:16.073647680607596 +Step 19275 | loss:0.8538452386856079 lr:4.531293180738871e-05 tokens_per_second_per_gpu:4677.396851976155 mfu:22.064078996966387 +Step 19276 | loss:0.82432621717453 lr:4.530861469746725e-05 tokens_per_second_per_gpu:5362.116699595101 mfu:25.294019343437597 +Step 19277 | loss:0.7119824886322021 lr:4.530429762283028e-05 tokens_per_second_per_gpu:4248.226433866314 mfu:20.039608910718318 +Step 19278 | loss:0.7853911519050598 lr:4.5299980583510306e-05 tokens_per_second_per_gpu:5097.450760435581 mfu:24.045544951756515 +Step 19279 | loss:0.6571832299232483 lr:4.529566357953977e-05 tokens_per_second_per_gpu:5325.313253135878 mfu:25.120411207173973 +Step 19280 | loss:0.7707200050354004 lr:4.529134661095114e-05 tokens_per_second_per_gpu:5311.99092695456 mfu:25.05756752155346 +Step 19281 | loss:0.6359800696372986 lr:4.528702967777691e-05 tokens_per_second_per_gpu:4479.385426113831 mfu:21.130025316940834 +Step 19282 | loss:0.6828922033309937 lr:4.528271278004953e-05 tokens_per_second_per_gpu:4406.2483529773435 mfu:20.785025264484606 +Step 19283 | loss:1.0184651613235474 lr:4.527839591780146e-05 tokens_per_second_per_gpu:3447.027808402695 mfu:16.26021829582501 +Step 19284 | loss:0.9391613602638245 lr:4.527407909106517e-05 tokens_per_second_per_gpu:4085.0110727194124 mfu:19.269694204776236 +Step 19285 | loss:0.6346779465675354 lr:4.526976229987315e-05 tokens_per_second_per_gpu:5349.272328081005 mfu:25.23343025152985 +Step 19286 | loss:0.8276352286338806 lr:4.5265445544257834e-05 tokens_per_second_per_gpu:5240.507962956151 mfu:24.72037018412167 +Step 19287 | loss:0.704446017742157 lr:4.526112882425172e-05 tokens_per_second_per_gpu:3639.758228475074 mfu:17.169360570506893 +Step 19288 | loss:0.7820661664009094 lr:4.5256812139887264e-05 tokens_per_second_per_gpu:5336.0226770290965 mfu:25.170929386894418 +Step 19289 | loss:0.8301301598548889 lr:4.5252495491196934e-05 tokens_per_second_per_gpu:4199.806755859812 mfu:19.811205028312703 +Step 19290 | loss:0.699146032333374 lr:4.524817887821316e-05 tokens_per_second_per_gpu:3381.255776703516 mfu:15.949960400434211 +Step 19291 | loss:0.8147072792053223 lr:4.524386230096847e-05 tokens_per_second_per_gpu:5058.562374955824 mfu:23.862101802405757 +Step 19292 | loss:0.8024065494537354 lr:4.5239545759495286e-05 tokens_per_second_per_gpu:5186.7173122965105 mfu:24.466630507328777 +Step 19293 | loss:0.7073818445205688 lr:4.523522925382608e-05 tokens_per_second_per_gpu:5335.121919222197 mfu:25.166680358634142 +Step 19294 | loss:0.7107769250869751 lr:4.5230912783993334e-05 tokens_per_second_per_gpu:5372.853052612345 mfu:25.344664552430654 +Step 19295 | loss:0.7563464641571045 lr:4.52265963500295e-05 tokens_per_second_per_gpu:4499.005286571221 mfu:21.222575546211722 +Step 19296 | loss:0.6944888830184937 lr:4.522227995196705e-05 tokens_per_second_per_gpu:4658.590698996092 mfu:21.975367164698863 +Step 19297 | loss:0.6477305293083191 lr:4.5217963589838425e-05 tokens_per_second_per_gpu:5357.066457780719 mfu:25.270196491139053 +Step 19298 | loss:0.7696994543075562 lr:4.521364726367612e-05 tokens_per_second_per_gpu:3837.8708631020772 mfu:18.103891669543437 +Step 19299 | loss:0.7069529891014099 lr:4.520933097351258e-05 tokens_per_second_per_gpu:3866.0373591270863 mfu:18.236757836993174 +Step 19300 | loss:0.9156289100646973 lr:4.520501471938026e-05 tokens_per_second_per_gpu:3292.2131558102255 mfu:15.52993116544286 +Step 19301 | loss:0.7764521241188049 lr:4.5200698501311656e-05 tokens_per_second_per_gpu:3701.540235479349 mfu:17.460796838643418 +Step 19302 | loss:0.6799744963645935 lr:4.519638231933919e-05 tokens_per_second_per_gpu:4006.420503017172 mfu:18.89896907855689 +Step 19303 | loss:0.6027447581291199 lr:4.5192066173495384e-05 tokens_per_second_per_gpu:5337.7507795933625 mfu:25.179081141534763 +Step 19304 | loss:0.670233428478241 lr:4.518775006381263e-05 tokens_per_second_per_gpu:5057.389346651085 mfu:23.856568427753036 +Step 19305 | loss:0.710701048374176 lr:4.518343399032343e-05 tokens_per_second_per_gpu:5347.524338011872 mfu:25.225184684135886 +Step 19306 | loss:0.7054209113121033 lr:4.5179117953060224e-05 tokens_per_second_per_gpu:4087.672006950917 mfu:19.282246285548435 +Step 19307 | loss:0.8720323443412781 lr:4.51748019520555e-05 tokens_per_second_per_gpu:4085.391518008309 mfu:19.27148883011462 +Step 19308 | loss:0.8204529881477356 lr:4.517048598734171e-05 tokens_per_second_per_gpu:5349.839648787374 mfu:25.236106400096684 +Step 19309 | loss:0.7163748145103455 lr:4.51661700589513e-05 tokens_per_second_per_gpu:4546.328906502188 mfu:21.445809135668238 +Step 19310 | loss:0.8252152800559998 lr:4.516185416691675e-05 tokens_per_second_per_gpu:4572.715505612304 mfu:21.570279225688697 +Step 19311 | loss:0.7851797342300415 lr:4.515753831127052e-05 tokens_per_second_per_gpu:4173.439040943979 mfu:19.68682401825866 +Step 19312 | loss:0.8427968621253967 lr:4.515322249204506e-05 tokens_per_second_per_gpu:4734.286468525658 mfu:22.332437024599866 +Step 19313 | loss:0.7745749950408936 lr:4.514890670927282e-05 tokens_per_second_per_gpu:4870.718804144946 mfu:22.97601162947283 +Step 19314 | loss:0.7789940237998962 lr:4.514459096298627e-05 tokens_per_second_per_gpu:4573.163619916646 mfu:21.572393057318465 +Step 19315 | loss:0.7637759447097778 lr:4.514027525321788e-05 tokens_per_second_per_gpu:4874.5614915051465 mfu:22.994138241380966 +Step 19316 | loss:0.8196807503700256 lr:4.513595958000009e-05 tokens_per_second_per_gpu:4873.952446661443 mfu:22.99126527293945 +Step 19317 | loss:0.6786160469055176 lr:4.513164394336538e-05 tokens_per_second_per_gpu:5337.947482283397 mfu:25.180009021684594 +Step 19318 | loss:0.6937498450279236 lr:4.51273283433462e-05 tokens_per_second_per_gpu:3740.517738491937 mfu:17.64466037060156 +Step 19319 | loss:0.7202578186988831 lr:4.5123012779975e-05 tokens_per_second_per_gpu:4194.363741419491 mfu:19.785529400523686 +Step 19320 | loss:0.6028014421463013 lr:4.5118697253284234e-05 tokens_per_second_per_gpu:4246.284165815003 mfu:20.03044690093518 +Step 19321 | loss:0.7749999165534973 lr:4.511438176330638e-05 tokens_per_second_per_gpu:5061.894003244993 mfu:23.87781766148016 +Step 19322 | loss:0.8096782565116882 lr:4.5110066310073873e-05 tokens_per_second_per_gpu:4129.4038921029805 mfu:19.47910270800454 +Step 19323 | loss:0.7189326286315918 lr:4.510575089361919e-05 tokens_per_second_per_gpu:4752.981711871572 mfu:22.420625677199972 +Step 19324 | loss:0.7920052409172058 lr:4.510143551397478e-05 tokens_per_second_per_gpu:5296.855389986078 mfu:24.986170611284237 +Step 19325 | loss:0.7877408266067505 lr:4.509712017117311e-05 tokens_per_second_per_gpu:4266.645097744067 mfu:20.126492890777783 +Step 19326 | loss:0.7939579486846924 lr:4.5092804865246616e-05 tokens_per_second_per_gpu:5310.684370040607 mfu:25.05140426966875 +Step 19327 | loss:0.647513747215271 lr:4.508848959622774e-05 tokens_per_second_per_gpu:5335.8462878736 mfu:25.17009732915285 +Step 19328 | loss:0.9140951633453369 lr:4.5084174364148985e-05 tokens_per_second_per_gpu:5052.925760296694 mfu:23.83551292935301 +Step 19329 | loss:0.9211541414260864 lr:4.507985916904277e-05 tokens_per_second_per_gpu:4502.119814645493 mfu:21.23726730653134 +Step 19330 | loss:0.8105995655059814 lr:4.5075544010941564e-05 tokens_per_second_per_gpu:4627.077878328608 mfu:21.82671581296871 +Step 19331 | loss:0.7089030742645264 lr:4.5071228889877825e-05 tokens_per_second_per_gpu:4533.236144829021 mfu:21.38404834500042 +Step 19332 | loss:0.6006755828857422 lr:4.5066913805884015e-05 tokens_per_second_per_gpu:3763.263541822181 mfu:17.75195620574455 +Step 19333 | loss:0.7241471409797668 lr:4.506259875899255e-05 tokens_per_second_per_gpu:5369.951703029007 mfu:25.330978391424765 +Step 19334 | loss:0.6331084966659546 lr:4.505828374923592e-05 tokens_per_second_per_gpu:3753.818316592731 mfu:17.707401466815632 +Step 19335 | loss:0.5949199199676514 lr:4.5053968776646574e-05 tokens_per_second_per_gpu:5384.900647577073 mfu:25.401495113410935 +Step 19336 | loss:0.8481706380844116 lr:4.504965384125695e-05 tokens_per_second_per_gpu:4563.745098120344 mfu:21.527964283039783 +Step 19337 | loss:0.8036940097808838 lr:4.504533894309952e-05 tokens_per_second_per_gpu:5031.327747834486 mfu:23.733631419567864 +Step 19338 | loss:0.9163353443145752 lr:4.504102408220672e-05 tokens_per_second_per_gpu:5355.2660066332965 mfu:25.261703455888018 +Step 19339 | loss:0.7598397731781006 lr:4.503670925861103e-05 tokens_per_second_per_gpu:5320.572973610438 mfu:25.098050499877388 +Step 19340 | loss:0.8525229692459106 lr:4.503239447234486e-05 tokens_per_second_per_gpu:4768.766851936709 mfu:22.49508687610983 +Step 19341 | loss:0.7013174295425415 lr:4.5028079723440695e-05 tokens_per_second_per_gpu:5376.198511695861 mfu:25.360445653721776 +Step 19342 | loss:0.6413262486457825 lr:4.5023765011930966e-05 tokens_per_second_per_gpu:5320.503738931154 mfu:25.097723908082546 +Step 19343 | loss:0.6755993962287903 lr:4.501945033784815e-05 tokens_per_second_per_gpu:5331.564377171966 mfu:25.14989882580394 +Step 19344 | loss:0.7189461588859558 lr:4.501513570122468e-05 tokens_per_second_per_gpu:3359.560716223241 mfu:15.847621098589743 +Step 19345 | loss:0.7718953490257263 lr:4.5010821102093e-05 tokens_per_second_per_gpu:4388.493439555547 mfu:20.70127230857339 +Step 19346 | loss:0.6371105909347534 lr:4.5006506540485584e-05 tokens_per_second_per_gpu:4825.48561288139 mfu:22.762638948704392 +Step 19347 | loss:0.7398638725280762 lr:4.500219201643488e-05 tokens_per_second_per_gpu:4904.520121075899 mfu:23.135458208535486 +Step 19348 | loss:0.7578487992286682 lr:4.499787752997331e-05 tokens_per_second_per_gpu:5300.381994061485 mfu:25.002806204408635 +Step 19349 | loss:0.8273496031761169 lr:4.499356308113333e-05 tokens_per_second_per_gpu:3680.109193369238 mfu:17.35970295649719 +Step 19350 | loss:0.8268951177597046 lr:4.498924866994742e-05 tokens_per_second_per_gpu:5204.223174060534 mfu:24.549208644077684 +Step 19351 | loss:0.9131749868392944 lr:4.498493429644801e-05 tokens_per_second_per_gpu:5329.52237223703 mfu:25.14026634012388 +Step 19352 | loss:0.7493221759796143 lr:4.498061996066753e-05 tokens_per_second_per_gpu:5281.6086788931925 mfu:24.91424927370194 +Step 19353 | loss:0.9162529706954956 lr:4.497630566263847e-05 tokens_per_second_per_gpu:4136.602766833728 mfu:19.51306102836416 +Step 19354 | loss:0.5955559611320496 lr:4.497199140239326e-05 tokens_per_second_per_gpu:5315.127878830605 mfu:25.07236505877195 +Step 19355 | loss:0.8010294437408447 lr:4.496767717996434e-05 tokens_per_second_per_gpu:4201.041492088079 mfu:19.817029489769165 +Step 19356 | loss:0.8417658805847168 lr:4.496336299538414e-05 tokens_per_second_per_gpu:4233.74479305855 mfu:19.971296540205387 +Step 19357 | loss:0.7591347694396973 lr:4.495904884868514e-05 tokens_per_second_per_gpu:5359.78813474399 mfu:25.283035105740773 +Step 19358 | loss:0.6646777391433716 lr:4.495473473989977e-05 tokens_per_second_per_gpu:4694.347302358566 mfu:22.14403716346527 +Step 19359 | loss:0.6574955582618713 lr:4.49504206690605e-05 tokens_per_second_per_gpu:4927.220684631687 mfu:23.242540639943666 +Step 19360 | loss:0.7125817537307739 lr:4.494610663619976e-05 tokens_per_second_per_gpu:5331.538504994601 mfu:25.14977678232906 +Step 19361 | loss:0.8131169676780701 lr:4.4941792641350005e-05 tokens_per_second_per_gpu:4255.709884620537 mfu:20.074909624734346 +Step 19362 | loss:0.7409586906433105 lr:4.493747868454366e-05 tokens_per_second_per_gpu:3410.683218369475 mfu:16.088774663611822 +Step 19363 | loss:0.8442975878715515 lr:4.4933164765813174e-05 tokens_per_second_per_gpu:5310.840858945535 mfu:25.052142454533055 +Step 19364 | loss:0.6036626696586609 lr:4.4928850885191015e-05 tokens_per_second_per_gpu:5329.997285304028 mfu:25.142506586089436 +Step 19365 | loss:0.6873253583908081 lr:4.4924537042709605e-05 tokens_per_second_per_gpu:5362.220651369237 mfu:25.2945097016922 +Step 19366 | loss:0.822708785533905 lr:4.492022323840141e-05 tokens_per_second_per_gpu:4449.840142068443 mfu:20.990655171153474 +Step 19367 | loss:0.7718130350112915 lr:4.491590947229886e-05 tokens_per_second_per_gpu:5346.810302871109 mfu:25.221816458549952 +Step 19368 | loss:0.8198198676109314 lr:4.491159574443442e-05 tokens_per_second_per_gpu:5090.903027340473 mfu:24.014658177588423 +Step 19369 | loss:0.6470394730567932 lr:4.490728205484049e-05 tokens_per_second_per_gpu:4163.003185994755 mfu:19.637596309922706 +Step 19370 | loss:0.8040962219238281 lr:4.490296840354956e-05 tokens_per_second_per_gpu:4153.444148574506 mfu:19.59250469947143 +Step 19371 | loss:0.701805591583252 lr:4.489865479059405e-05 tokens_per_second_per_gpu:5323.050793150312 mfu:25.10973879740658 +Step 19372 | loss:0.7282162308692932 lr:4.48943412160064e-05 tokens_per_second_per_gpu:4601.859004609461 mfu:21.707754082851956 +Step 19373 | loss:0.926464319229126 lr:4.489002767981907e-05 tokens_per_second_per_gpu:5350.720761980031 mfu:25.24026276135986 +Step 19374 | loss:0.6354732513427734 lr:4.488571418206448e-05 tokens_per_second_per_gpu:4458.425934740667 mfu:21.031155820075636 +Step 19375 | loss:0.6769700646400452 lr:4.4881400722775105e-05 tokens_per_second_per_gpu:4327.404293974127 mfu:20.41310438598282 +Step 19376 | loss:0.6667538285255432 lr:4.4877087301983375e-05 tokens_per_second_per_gpu:5363.5526602493355 mfu:25.300793014843773 +Step 19377 | loss:0.7745663523674011 lr:4.487277391972171e-05 tokens_per_second_per_gpu:5329.876467644979 mfu:25.141936668578083 +Step 19378 | loss:0.7581020593643188 lr:4.486846057602258e-05 tokens_per_second_per_gpu:4225.630951789103 mfu:19.933022166573334 +Step 19379 | loss:0.6871451139450073 lr:4.486414727091839e-05 tokens_per_second_per_gpu:5286.410393773047 mfu:24.936899782047256 +Step 19380 | loss:0.7205191850662231 lr:4.485983400444162e-05 tokens_per_second_per_gpu:5351.411835857341 mfu:25.243522674748494 +Step 19381 | loss:0.8157022595405579 lr:4.4855520776624674e-05 tokens_per_second_per_gpu:5031.713190280313 mfu:23.73544961734809 +Step 19382 | loss:0.6927962303161621 lr:4.485120758750003e-05 tokens_per_second_per_gpu:5348.503533622567 mfu:25.22980371689954 +Step 19383 | loss:0.7377926111221313 lr:4.484689443710012e-05 tokens_per_second_per_gpu:4646.12016488059 mfu:21.91654152758352 +Step 19384 | loss:0.843536913394928 lr:4.4842581325457365e-05 tokens_per_second_per_gpu:4478.717608334813 mfu:21.126875106535742 +Step 19385 | loss:0.7578191757202148 lr:4.4838268252604205e-05 tokens_per_second_per_gpu:4086.4006699125894 mfu:19.27624917182618 +Step 19386 | loss:0.7382375597953796 lr:4.48339552185731e-05 tokens_per_second_per_gpu:4168.940363504136 mfu:19.665602989221394 +Step 19387 | loss:0.7629166841506958 lr:4.482964222339647e-05 tokens_per_second_per_gpu:4320.6787868841475 mfu:20.381379021549552 +Step 19388 | loss:0.7231069803237915 lr:4.482532926710675e-05 tokens_per_second_per_gpu:3639.084893300632 mfu:17.16618433360622 +Step 19389 | loss:0.6407667398452759 lr:4.482101634973641e-05 tokens_per_second_per_gpu:4341.727996140919 mfu:20.48067173297917 +Step 19390 | loss:0.7311222553253174 lr:4.481670347131787e-05 tokens_per_second_per_gpu:3215.271144619676 mfu:15.166982570996248 +Step 19391 | loss:0.7530774474143982 lr:4.4812390631883554e-05 tokens_per_second_per_gpu:4292.3297828120485 mfu:20.24765193250261 +Step 19392 | loss:0.7427926659584045 lr:4.48080778314659e-05 tokens_per_second_per_gpu:3882.548898724479 mfu:18.314645586433226 +Step 19393 | loss:0.7668911218643188 lr:4.480376507009737e-05 tokens_per_second_per_gpu:4245.061751947084 mfu:20.024680566154895 +Step 19394 | loss:0.7958632707595825 lr:4.479945234781038e-05 tokens_per_second_per_gpu:4945.78813192106 mfu:23.330126456740814 +Step 19395 | loss:0.8293203115463257 lr:4.4795139664637366e-05 tokens_per_second_per_gpu:4885.514418261357 mfu:23.04580506565201 +Step 19396 | loss:0.8303630352020264 lr:4.4790827020610784e-05 tokens_per_second_per_gpu:5319.76202178214 mfu:25.09422510174066 +Step 19397 | loss:0.9219370484352112 lr:4.478651441576306e-05 tokens_per_second_per_gpu:4197.777454955456 mfu:19.80163246971242 +Step 19398 | loss:0.8016105890274048 lr:4.478220185012661e-05 tokens_per_second_per_gpu:3350.273595792986 mfu:15.80381216697409 +Step 19399 | loss:0.7149759531021118 lr:4.477788932373389e-05 tokens_per_second_per_gpu:4808.166966240908 mfu:22.680943937635426 +Step 19400 | loss:0.6214343309402466 lr:4.477357683661734e-05 tokens_per_second_per_gpu:4132.8643673046145 mfu:19.4954263599486 +Step 19401 | loss:0.7812070250511169 lr:4.476926438880936e-05 tokens_per_second_per_gpu:5294.78998826254 mfu:24.976427758960437 +Step 19402 | loss:0.8074787259101868 lr:4.476495198034244e-05 tokens_per_second_per_gpu:4960.831852135753 mfu:23.401090251716862 +Step 19403 | loss:0.6168817281723022 lr:4.476063961124897e-05 tokens_per_second_per_gpu:5043.732454156586 mfu:23.792146535750344 +Step 19404 | loss:0.8290883898735046 lr:4.475632728156139e-05 tokens_per_second_per_gpu:4876.621198189981 mfu:23.003854229235532 +Step 19405 | loss:0.8064409494400024 lr:4.475201499131217e-05 tokens_per_second_per_gpu:5327.970088064948 mfu:25.132943950837177 +Step 19406 | loss:0.8123290538787842 lr:4.4747702740533696e-05 tokens_per_second_per_gpu:3474.9080085592764 mfu:16.39173395681626 +Step 19407 | loss:0.7614201307296753 lr:4.474339052925842e-05 tokens_per_second_per_gpu:3698.5737844230785 mfu:17.44680358288225 +Step 19408 | loss:0.9072403311729431 lr:4.473907835751877e-05 tokens_per_second_per_gpu:4899.889707299283 mfu:23.113615756721135 +Step 19409 | loss:0.6837100982666016 lr:4.473476622534719e-05 tokens_per_second_per_gpu:5125.471968437188 mfu:24.17772577081117 +Step 19410 | loss:0.6757751107215881 lr:4.4730454132776096e-05 tokens_per_second_per_gpu:5094.7806136981035 mfu:24.03294941402182 +Step 19411 | loss:0.6681764125823975 lr:4.472614207983794e-05 tokens_per_second_per_gpu:4145.403704759576 mfu:19.554576554155233 +Step 19412 | loss:0.7489696145057678 lr:4.472183006656515e-05 tokens_per_second_per_gpu:4309.756293939953 mfu:20.329855758762324 +Step 19413 | loss:0.7386667728424072 lr:4.471751809299014e-05 tokens_per_second_per_gpu:4739.307996517386 mfu:22.356124428897864 +Step 19414 | loss:0.8648378252983093 lr:4.471320615914535e-05 tokens_per_second_per_gpu:4734.355087804347 mfu:22.332760713445097 +Step 19415 | loss:0.7013965845108032 lr:4.47088942650632e-05 tokens_per_second_per_gpu:5319.193359695909 mfu:25.09154262565615 +Step 19416 | loss:0.6827486157417297 lr:4.4704582410776146e-05 tokens_per_second_per_gpu:4615.77852678521 mfu:21.773414843870334 +Step 19417 | loss:0.7561487555503845 lr:4.470027059631659e-05 tokens_per_second_per_gpu:4425.086599130253 mfu:20.873888485724102 +Step 19418 | loss:0.6749695539474487 lr:4.469595882171699e-05 tokens_per_second_per_gpu:4189.290173953956 mfu:19.76159651714888 +Step 19419 | loss:0.5755060911178589 lr:4.4691647087009765e-05 tokens_per_second_per_gpu:5350.972644460018 mfu:25.24145093399354 +Step 19420 | loss:0.756218433380127 lr:4.468733539222733e-05 tokens_per_second_per_gpu:4712.9755416701855 mfu:22.23190974660355 +Step 19421 | loss:0.7656257152557373 lr:4.468302373740211e-05 tokens_per_second_per_gpu:4721.991580633923 mfu:22.274439940690367 +Step 19422 | loss:0.7859150767326355 lr:4.4678712122566564e-05 tokens_per_second_per_gpu:5347.453192652617 mfu:25.224849079337634 +Step 19423 | loss:0.753864586353302 lr:4.46744005477531e-05 tokens_per_second_per_gpu:3688.6307330705886 mfu:17.39990051319306 +Step 19424 | loss:0.7678456902503967 lr:4.467008901299414e-05 tokens_per_second_per_gpu:4073.7125363292616 mfu:19.216397080894804 +Step 19425 | loss:0.7841680645942688 lr:4.4665777518322126e-05 tokens_per_second_per_gpu:4601.802164607816 mfu:21.70748595886629 +Step 19426 | loss:0.733985424041748 lr:4.46614660637695e-05 tokens_per_second_per_gpu:3986.738095669643 mfu:18.806123804933616 +Step 19427 | loss:0.725703775882721 lr:4.465715464936865e-05 tokens_per_second_per_gpu:4210.9401899887 mfu:19.863723336658406 +Step 19428 | loss:0.7907551527023315 lr:4.465284327515201e-05 tokens_per_second_per_gpu:4005.930648735625 mfu:18.896658352333073 +Step 19429 | loss:0.5619114637374878 lr:4.464853194115204e-05 tokens_per_second_per_gpu:4582.86885518228 mfu:21.61817430795034 +Step 19430 | loss:0.6871209144592285 lr:4.4644220647401136e-05 tokens_per_second_per_gpu:3576.1025019248646 mfu:16.86908564758268 +Step 19431 | loss:0.7644931674003601 lr:4.4639909393931724e-05 tokens_per_second_per_gpu:5379.698418169435 mfu:25.376955309703245 +Step 19432 | loss:0.9689809083938599 lr:4.463559818077624e-05 tokens_per_second_per_gpu:4670.525542462961 mfu:22.031665857626358 +Step 19433 | loss:0.7054604887962341 lr:4.46312870079671e-05 tokens_per_second_per_gpu:4957.137828762037 mfu:23.383664913197663 +Step 19434 | loss:0.6565348505973816 lr:4.4626975875536775e-05 tokens_per_second_per_gpu:4824.211521259659 mfu:22.756628841142245 +Step 19435 | loss:0.7462915182113647 lr:4.462266478351761e-05 tokens_per_second_per_gpu:5363.304793323121 mfu:25.299623784262362 +Step 19436 | loss:0.7576236724853516 lr:4.4618353731942086e-05 tokens_per_second_per_gpu:5329.410187844703 mfu:25.139737147204638 +Step 19437 | loss:0.6752265691757202 lr:4.4614042720842594e-05 tokens_per_second_per_gpu:5081.998641184602 mfu:23.972654668846246 +Step 19438 | loss:0.7758753299713135 lr:4.4609731750251585e-05 tokens_per_second_per_gpu:5300.072516142455 mfu:25.001346344262192 +Step 19439 | loss:0.7338348031044006 lr:4.460542082020148e-05 tokens_per_second_per_gpu:5307.079618155992 mfu:25.034400039994928 +Step 19440 | loss:1.0111721754074097 lr:4.4601109930724675e-05 tokens_per_second_per_gpu:3508.5127360510455 mfu:16.55025318995267 +Step 19441 | loss:0.7013981938362122 lr:4.459679908185363e-05 tokens_per_second_per_gpu:3715.372372057399 mfu:17.526045386887937 +Step 19442 | loss:0.6577861309051514 lr:4.4592488273620744e-05 tokens_per_second_per_gpu:4649.923308371563 mfu:21.93448160431363 +Step 19443 | loss:0.882412314414978 lr:4.4588177506058437e-05 tokens_per_second_per_gpu:5338.570066935377 mfu:25.18294585971188 +Step 19444 | loss:0.7702866196632385 lr:4.458386677919913e-05 tokens_per_second_per_gpu:4602.047548846302 mfu:21.708643478186094 +Step 19445 | loss:0.8022052645683289 lr:4.457955609307526e-05 tokens_per_second_per_gpu:4798.7010363572335 mfu:22.63629153131996 +Step 19446 | loss:0.6912083029747009 lr:4.457524544771925e-05 tokens_per_second_per_gpu:5324.321029978405 mfu:25.11573072124945 +Step 19447 | loss:0.704645574092865 lr:4.4570934843163494e-05 tokens_per_second_per_gpu:4266.1544704831185 mfu:20.124178518278246 +Step 19448 | loss:0.7564252614974976 lr:4.456662427944045e-05 tokens_per_second_per_gpu:4755.641414828014 mfu:22.433171949837625 +Step 19449 | loss:0.7550104856491089 lr:4.456231375658251e-05 tokens_per_second_per_gpu:3834.9401924656563 mfu:18.090067196127467 +Step 19450 | loss:0.729137122631073 lr:4.455800327462209e-05 tokens_per_second_per_gpu:4853.440634833714 mfu:22.89450755687208 +Step 19451 | loss:0.7417007088661194 lr:4.455369283359163e-05 tokens_per_second_per_gpu:3229.9612811236016 mfu:15.236278451281006 +Step 19452 | loss:0.7485237121582031 lr:4.4549382433523546e-05 tokens_per_second_per_gpu:5331.997246634477 mfu:25.15194074491375 +Step 19453 | loss:0.6891981959342957 lr:4.454507207445024e-05 tokens_per_second_per_gpu:5327.4786509860405 mfu:25.13062575828816 +Step 19454 | loss:0.7418355941772461 lr:4.454076175640415e-05 tokens_per_second_per_gpu:3813.7000098190933 mfu:17.98987363063478 +Step 19455 | loss:0.660176157951355 lr:4.45364514794177e-05 tokens_per_second_per_gpu:4741.000812465 mfu:22.36410973054695 +Step 19456 | loss:0.7749961614608765 lr:4.4532141243523286e-05 tokens_per_second_per_gpu:4383.3543462484295 mfu:20.677030328623985 +Step 19457 | loss:0.6572751998901367 lr:4.452783104875333e-05 tokens_per_second_per_gpu:4537.133168343285 mfu:21.40243126981798 +Step 19458 | loss:0.7586421966552734 lr:4.452352089514026e-05 tokens_per_second_per_gpu:5249.627369429956 mfu:24.7633879803898 +Step 19459 | loss:0.7606564164161682 lr:4.451921078271649e-05 tokens_per_second_per_gpu:4962.753542482691 mfu:23.41015519295755 +Step 19460 | loss:0.7191081643104553 lr:4.451490071151442e-05 tokens_per_second_per_gpu:5312.555264935939 mfu:25.060229600097614 +Step 19461 | loss:0.7681241035461426 lr:4.45105906815665e-05 tokens_per_second_per_gpu:4828.29852035146 mfu:22.77590791317222 +Step 19462 | loss:0.7074644565582275 lr:4.450628069290513e-05 tokens_per_second_per_gpu:5379.467626774224 mfu:25.37586662731496 +Step 19463 | loss:0.837670624256134 lr:4.4501970745562716e-05 tokens_per_second_per_gpu:4073.5305942507016 mfu:19.215538829067718 +Step 19464 | loss:0.6847435832023621 lr:4.449766083957167e-05 tokens_per_second_per_gpu:5350.860803530956 mfu:25.240923361249127 +Step 19465 | loss:0.8329920172691345 lr:4.449335097496443e-05 tokens_per_second_per_gpu:5034.769213451498 mfu:23.74986540005789 +Step 19466 | loss:0.8031992316246033 lr:4.44890411517734e-05 tokens_per_second_per_gpu:4906.30887259687 mfu:23.143896054652387 +Step 19467 | loss:0.811526894569397 lr:4.448473137003099e-05 tokens_per_second_per_gpu:4554.240092767562 mfu:21.48312755106957 +Step 19468 | loss:0.711439311504364 lr:4.4480421629769626e-05 tokens_per_second_per_gpu:3086.596154107877 mfu:14.560000686534863 +Step 19469 | loss:0.7517699599266052 lr:4.44761119310217e-05 tokens_per_second_per_gpu:5332.268784372803 mfu:25.153221634754548 +Step 19470 | loss:0.7557167410850525 lr:4.447180227381968e-05 tokens_per_second_per_gpu:4837.213455457003 mfu:22.81796118310984 +Step 19471 | loss:0.6459861397743225 lr:4.44674926581959e-05 tokens_per_second_per_gpu:3259.0678770502686 mfu:15.373579230364447 +Step 19472 | loss:0.8097516298294067 lr:4.4463183084182835e-05 tokens_per_second_per_gpu:5365.895226399319 mfu:25.311843299056243 +Step 19473 | loss:0.7747782468795776 lr:4.4458873551812856e-05 tokens_per_second_per_gpu:4423.835028373645 mfu:20.86798461292519 +Step 19474 | loss:0.7126720547676086 lr:4.4454564061118416e-05 tokens_per_second_per_gpu:4599.114357751831 mfu:21.694807115339476 +Step 19475 | loss:0.555966854095459 lr:4.4450254612131905e-05 tokens_per_second_per_gpu:5345.651479119771 mfu:25.21635008919917 +Step 19476 | loss:0.7874143123626709 lr:4.444594520488572e-05 tokens_per_second_per_gpu:4330.745399182701 mfu:20.428864949302994 +Step 19477 | loss:0.9574663639068604 lr:4.4441635839412314e-05 tokens_per_second_per_gpu:4822.727757155212 mfu:22.749629672705307 +Step 19478 | loss:0.8689555525779724 lr:4.4437326515744064e-05 tokens_per_second_per_gpu:3661.6134563637434 mfu:17.272455409344136 +Step 19479 | loss:0.7676572799682617 lr:4.443301723391339e-05 tokens_per_second_per_gpu:5328.968182034094 mfu:25.13765213038203 +Step 19480 | loss:0.7293863296508789 lr:4.44287079939527e-05 tokens_per_second_per_gpu:5167.204743861782 mfu:24.374586392834694 +Step 19481 | loss:0.7147552371025085 lr:4.442439879589441e-05 tokens_per_second_per_gpu:3751.116346986404 mfu:17.694655815179036 +Step 19482 | loss:0.9925180077552795 lr:4.442008963977093e-05 tokens_per_second_per_gpu:4239.533709557967 mfu:19.998603847024143 +Step 19483 | loss:0.8569542765617371 lr:4.441578052561466e-05 tokens_per_second_per_gpu:4597.356050797127 mfu:21.686512881436027 +Step 19484 | loss:0.7073818445205688 lr:4.4411471453458045e-05 tokens_per_second_per_gpu:5334.718733618607 mfu:25.164778463351453 +Step 19485 | loss:0.5717666149139404 lr:4.440716242333344e-05 tokens_per_second_per_gpu:5337.452373094686 mfu:25.177673507167107 +Step 19486 | loss:0.5521727800369263 lr:4.440285343527329e-05 tokens_per_second_per_gpu:5350.42771203972 mfu:25.238880394791938 +Step 19487 | loss:0.672866702079773 lr:4.439854448930998e-05 tokens_per_second_per_gpu:5339.232629022975 mfu:25.186071278123826 +Step 19488 | loss:0.9682894349098206 lr:4.439423558547595e-05 tokens_per_second_per_gpu:5336.4209380256125 mfu:25.17280805196532 +Step 19489 | loss:0.5387320518493652 lr:4.438992672380356e-05 tokens_per_second_per_gpu:5315.443936809406 mfu:25.073855958183007 +Step 19490 | loss:0.7577542066574097 lr:4.438561790432528e-05 tokens_per_second_per_gpu:5319.972763145199 mfu:25.09521920470639 +Step 19491 | loss:0.936147928237915 lr:4.438130912707348e-05 tokens_per_second_per_gpu:3079.5405269111307 mfu:14.526718089233679 +Step 19492 | loss:0.6272584199905396 lr:4.4377000392080566e-05 tokens_per_second_per_gpu:5293.41309159031 mfu:24.96993270243473 +Step 19493 | loss:0.7768820524215698 lr:4.4372691699378935e-05 tokens_per_second_per_gpu:3950.175309256705 mfu:18.633650903169162 +Step 19494 | loss:0.8268533945083618 lr:4.436838304900102e-05 tokens_per_second_per_gpu:3168.0925709487747 mfu:14.944433189495902 +Step 19495 | loss:0.7953170537948608 lr:4.4364074440979216e-05 tokens_per_second_per_gpu:5333.592589203472 mfu:25.15946624800521 +Step 19496 | loss:0.8397349119186401 lr:4.4359765875345924e-05 tokens_per_second_per_gpu:3624.53150220487 mfu:17.097533504742945 +Step 19497 | loss:0.8325597047805786 lr:4.435545735213355e-05 tokens_per_second_per_gpu:4146.6721648151215 mfu:19.560560096658843 +Step 19498 | loss:0.7526501417160034 lr:4.435114887137451e-05 tokens_per_second_per_gpu:3702.959781884977 mfu:17.467493081238356 +Step 19499 | loss:0.6317026019096375 lr:4.434684043310121e-05 tokens_per_second_per_gpu:4447.1037171209455 mfu:20.97774698779389 +Step 19500 | loss:0.8450836539268494 lr:4.434253203734602e-05 tokens_per_second_per_gpu:3923.9362946705346 mfu:18.509876994528668 +Step 19501 | loss:0.7777738571166992 lr:4.433822368414138e-05 tokens_per_second_per_gpu:3860.0859088562124 mfu:18.208683830643338 +Step 19502 | loss:0.6136397123336792 lr:4.433391537351968e-05 tokens_per_second_per_gpu:5365.801082772375 mfu:25.31139920750539 +Step 19503 | loss:0.6681567430496216 lr:4.432960710551332e-05 tokens_per_second_per_gpu:4465.04496835926 mfu:21.062378930978127 +Step 19504 | loss:0.7748351693153381 lr:4.4325298880154716e-05 tokens_per_second_per_gpu:5344.433812082612 mfu:25.210606146029473 +Step 19505 | loss:0.6819924712181091 lr:4.432099069747625e-05 tokens_per_second_per_gpu:4584.99101061121 mfu:21.628184877185657 +Step 19506 | loss:0.7271193861961365 lr:4.431668255751038e-05 tokens_per_second_per_gpu:5311.621901138027 mfu:25.05582676381458 +Step 19507 | loss:0.7286815047264099 lr:4.431237446028942e-05 tokens_per_second_per_gpu:3081.142138537114 mfu:14.534273164536565 +Step 19508 | loss:0.7486523985862732 lr:4.430806640584583e-05 tokens_per_second_per_gpu:5352.753564286038 mfu:25.249851836669137 +Step 19509 | loss:0.8176488280296326 lr:4.4303758394212e-05 tokens_per_second_per_gpu:5338.164690924143 mfu:25.181033631902608 +Step 19510 | loss:0.8156245946884155 lr:4.429945042542033e-05 tokens_per_second_per_gpu:4413.77534113636 mfu:20.82053135186729 +Step 19511 | loss:0.684989869594574 lr:4.429514249950322e-05 tokens_per_second_per_gpu:5073.664358646971 mfu:23.933340436141005 +Step 19512 | loss:0.6901569366455078 lr:4.429083461649307e-05 tokens_per_second_per_gpu:4152.314967585095 mfu:19.587178160086232 +Step 19513 | loss:0.6330767273902893 lr:4.42865267764223e-05 tokens_per_second_per_gpu:4586.236464276619 mfu:21.6340598946217 +Step 19514 | loss:0.7065661549568176 lr:4.4282218979323275e-05 tokens_per_second_per_gpu:5329.6648973358015 mfu:25.14093865533207 +Step 19515 | loss:0.8740007281303406 lr:4.4277911225228414e-05 tokens_per_second_per_gpu:3666.4652202145785 mfu:17.295342007224797 +Step 19516 | loss:0.7971082925796509 lr:4.42736035141701e-05 tokens_per_second_per_gpu:4726.941445912709 mfu:22.2977892997445 +Step 19517 | loss:0.7593331933021545 lr:4.426929584618076e-05 tokens_per_second_per_gpu:4072.986011523605 mfu:19.212969939428607 +Step 19518 | loss:0.6820594072341919 lr:4.426498822129277e-05 tokens_per_second_per_gpu:5037.715964666147 mfu:23.763765728304985 +Step 19519 | loss:0.7409107089042664 lr:4.4260680639538525e-05 tokens_per_second_per_gpu:5116.554194242664 mfu:24.13565910840643 +Step 19520 | loss:0.7602707743644714 lr:4.425637310095046e-05 tokens_per_second_per_gpu:4584.589807675294 mfu:21.626292334484663 +Step 19521 | loss:0.7510308027267456 lr:4.425206560556093e-05 tokens_per_second_per_gpu:4670.015440216394 mfu:22.029259618296887 +Step 19522 | loss:0.8420255780220032 lr:4.4247758153402354e-05 tokens_per_second_per_gpu:5314.815854055762 mfu:25.0708931846721 +Step 19523 | loss:0.7045758366584778 lr:4.424345074450711e-05 tokens_per_second_per_gpu:5356.743720943377 mfu:25.268674086412727 +Step 19524 | loss:0.6892738938331604 lr:4.4239143378907625e-05 tokens_per_second_per_gpu:4291.273294851621 mfu:20.24266829853784 +Step 19525 | loss:0.6899523735046387 lr:4.423483605663625e-05 tokens_per_second_per_gpu:5297.86650022102 mfu:24.990940190775657 +Step 19526 | loss:0.7260422706604004 lr:4.423052877772543e-05 tokens_per_second_per_gpu:4241.61910235294 mfu:20.008441000642726 +Step 19527 | loss:0.7324128150939941 lr:4.422622154220754e-05 tokens_per_second_per_gpu:4046.386395341368 mfu:19.087495011534493 +Step 19528 | loss:0.6310570240020752 lr:4.4221914350114986e-05 tokens_per_second_per_gpu:4232.677660621336 mfu:19.96626269442748 +Step 19529 | loss:0.7745667099952698 lr:4.421760720148013e-05 tokens_per_second_per_gpu:3286.52558246043 mfu:15.5031019115518 +Step 19530 | loss:0.6830676198005676 lr:4.4213300096335394e-05 tokens_per_second_per_gpu:5326.957955564905 mfu:25.128169549146868 +Step 19531 | loss:0.6792109608650208 lr:4.420899303471317e-05 tokens_per_second_per_gpu:3180.6135642703584 mfu:15.00349685129556 +Step 19532 | loss:0.7238894701004028 lr:4.420468601664583e-05 tokens_per_second_per_gpu:4404.990920647334 mfu:20.77909374164387 +Step 19533 | loss:0.7108640074729919 lr:4.420037904216581e-05 tokens_per_second_per_gpu:5347.414212970086 mfu:25.224665205524513 +Step 19534 | loss:0.6682590842247009 lr:4.419607211130547e-05 tokens_per_second_per_gpu:5358.78320391363 mfu:25.278294675555085 +Step 19535 | loss:0.8744745254516602 lr:4.419176522409723e-05 tokens_per_second_per_gpu:4182.1380392627925 mfu:19.72785867753066 +Step 19536 | loss:0.68170166015625 lr:4.418745838057343e-05 tokens_per_second_per_gpu:5354.101251771465 mfu:25.256209108477087 +Step 19537 | loss:0.9153906106948853 lr:4.418315158076652e-05 tokens_per_second_per_gpu:5330.880834617635 mfu:25.146674438950775 +Step 19538 | loss:0.7662934064865112 lr:4.4178844824708864e-05 tokens_per_second_per_gpu:5324.32945771656 mfu:25.115770476328244 +Step 19539 | loss:0.8381295800209045 lr:4.417453811243285e-05 tokens_per_second_per_gpu:3488.5027736035754 mfu:16.455862783036316 +Step 19540 | loss:0.7082730531692505 lr:4.417023144397089e-05 tokens_per_second_per_gpu:4067.3846178510894 mfu:19.186547209779043 +Step 19541 | loss:0.7769359350204468 lr:4.4165924819355355e-05 tokens_per_second_per_gpu:4182.052567393572 mfu:19.72745549214054 +Step 19542 | loss:0.8046947121620178 lr:4.416161823861867e-05 tokens_per_second_per_gpu:4529.076388087463 mfu:21.364426062722806 +Step 19543 | loss:0.7837912440299988 lr:4.415731170179317e-05 tokens_per_second_per_gpu:5378.364257160085 mfu:25.3706618445908 +Step 19544 | loss:0.7080449461936951 lr:4.415300520891129e-05 tokens_per_second_per_gpu:4787.886096700969 mfu:22.58527561574238 +Step 19545 | loss:0.7935761213302612 lr:4.414869876000538e-05 tokens_per_second_per_gpu:3282.998313439608 mfu:15.486463181766428 +Step 19546 | loss:0.6990473866462708 lr:4.414439235510787e-05 tokens_per_second_per_gpu:4515.46866676122 mfu:21.300236119510252 +Step 19547 | loss:0.8188989758491516 lr:4.4140085994251136e-05 tokens_per_second_per_gpu:5376.556053262943 mfu:25.36213223829658 +Step 19548 | loss:0.628839373588562 lr:4.413577967746755e-05 tokens_per_second_per_gpu:5323.8516180894185 mfu:25.113516425279172 +Step 19549 | loss:0.794882595539093 lr:4.413147340478954e-05 tokens_per_second_per_gpu:4159.047567886896 mfu:19.618936984409892 +Step 19550 | loss:0.7263136506080627 lr:4.412716717624945e-05 tokens_per_second_per_gpu:4052.6630871066773 mfu:19.1171032375055 +Step 19551 | loss:0.782980740070343 lr:4.412286099187968e-05 tokens_per_second_per_gpu:3791.0631955646813 mfu:17.883091915557944 +Step 19552 | loss:0.6296914219856262 lr:4.4118554851712624e-05 tokens_per_second_per_gpu:4713.531347518662 mfu:22.234531577621066 +Step 19553 | loss:0.7873276472091675 lr:4.4114248755780675e-05 tokens_per_second_per_gpu:4204.15849136908 mfu:19.831732907192563 +Step 19554 | loss:0.7391046285629272 lr:4.41099427041162e-05 tokens_per_second_per_gpu:5201.722622732474 mfu:24.537413116825174 +Step 19555 | loss:0.8405702114105225 lr:4.41056366967516e-05 tokens_per_second_per_gpu:4622.78437705598 mfu:21.806462634918876 +Step 19556 | loss:0.6354280710220337 lr:4.4101330733719264e-05 tokens_per_second_per_gpu:5336.271724008936 mfu:25.172104183990115 +Step 19557 | loss:0.7141367793083191 lr:4.409702481505159e-05 tokens_per_second_per_gpu:3352.1021397728714 mfu:15.812437720908985 +Step 19558 | loss:0.7327446341514587 lr:4.409271894078093e-05 tokens_per_second_per_gpu:3777.877431737517 mfu:17.820892417861916 +Step 19559 | loss:0.6335741877555847 lr:4.408841311093967e-05 tokens_per_second_per_gpu:5346.679054768333 mfu:25.221197338855923 +Step 19560 | loss:0.6409309506416321 lr:4.408410732556022e-05 tokens_per_second_per_gpu:4442.222499330448 mfu:20.95472145065448 +Step 19561 | loss:0.7129468321800232 lr:4.407980158467495e-05 tokens_per_second_per_gpu:4809.393081238541 mfu:22.686727731276054 +Step 19562 | loss:0.6643190383911133 lr:4.407549588831626e-05 tokens_per_second_per_gpu:5046.898359554914 mfu:23.807080651673424 +Step 19563 | loss:0.8471485376358032 lr:4.4071190236516514e-05 tokens_per_second_per_gpu:4848.176980151635 mfu:22.869677999664614 +Step 19564 | loss:0.6430947184562683 lr:4.4066884629308115e-05 tokens_per_second_per_gpu:5206.301151224027 mfu:24.559010816897345 +Step 19565 | loss:0.9187615513801575 lr:4.406257906672341e-05 tokens_per_second_per_gpu:4679.034214088627 mfu:22.0718027134137 +Step 19566 | loss:0.7593127489089966 lr:4.405827354879482e-05 tokens_per_second_per_gpu:4429.744085976531 mfu:20.895858645827197 +Step 19567 | loss:0.6443256139755249 lr:4.405396807555472e-05 tokens_per_second_per_gpu:5351.569291952519 mfu:25.24426542201423 +Step 19568 | loss:0.7301774621009827 lr:4.404966264703545e-05 tokens_per_second_per_gpu:4369.390585585698 mfu:20.61116088711464 +Step 19569 | loss:0.6889160871505737 lr:4.404535726326946e-05 tokens_per_second_per_gpu:5351.610604895078 mfu:25.244460302212993 +Step 19570 | loss:0.6869999766349792 lr:4.404105192428909e-05 tokens_per_second_per_gpu:4693.356010916634 mfu:22.13936107260192 +Step 19571 | loss:0.7049779295921326 lr:4.403674663012674e-05 tokens_per_second_per_gpu:4998.910915542365 mfu:23.58071569076401 +Step 19572 | loss:0.7076208591461182 lr:4.403244138081476e-05 tokens_per_second_per_gpu:4351.306265340164 mfu:20.525854063013252 +Step 19573 | loss:0.7715967297554016 lr:4.4028136176385556e-05 tokens_per_second_per_gpu:3430.3103787487053 mfu:16.18135932785952 +Step 19574 | loss:0.7266025543212891 lr:4.402383101687151e-05 tokens_per_second_per_gpu:4363.078075424749 mfu:20.58138369966015 +Step 19575 | loss:0.6405757069587708 lr:4.4019525902304974e-05 tokens_per_second_per_gpu:4415.708279273349 mfu:20.829649350852723 +Step 19576 | loss:0.8145525455474854 lr:4.401522083271836e-05 tokens_per_second_per_gpu:5332.907936087685 mfu:25.156236622444172 +Step 19577 | loss:0.8057168126106262 lr:4.401091580814403e-05 tokens_per_second_per_gpu:4309.566277578893 mfu:20.328959419167074 +Step 19578 | loss:0.8473165035247803 lr:4.400661082861439e-05 tokens_per_second_per_gpu:4924.768770944274 mfu:23.230974544740526 +Step 19579 | loss:0.7753825187683105 lr:4.400230589416176e-05 tokens_per_second_per_gpu:5365.522353733702 mfu:25.310084395073893 +Step 19580 | loss:0.7800372838973999 lr:4.399800100481858e-05 tokens_per_second_per_gpu:4208.731173551412 mfu:19.85330302922665 +Step 19581 | loss:0.6333909630775452 lr:4.3993696160617174e-05 tokens_per_second_per_gpu:4112.136755013889 mfu:19.397650676277916 +Step 19582 | loss:0.6295472383499146 lr:4.398939136158996e-05 tokens_per_second_per_gpu:4318.516630068244 mfu:20.371179758946944 +Step 19583 | loss:0.7091952562332153 lr:4.39850866077693e-05 tokens_per_second_per_gpu:4819.797271334052 mfu:22.735806071094455 +Step 19584 | loss:0.8020426630973816 lr:4.398078189918756e-05 tokens_per_second_per_gpu:5342.468850888492 mfu:25.20133708881986 +Step 19585 | loss:0.6834069490432739 lr:4.3976477235877167e-05 tokens_per_second_per_gpu:4951.767160077767 mfu:23.358330552683395 +Step 19586 | loss:0.6960437893867493 lr:4.3972172617870415e-05 tokens_per_second_per_gpu:4290.457604708565 mfu:20.238820548961954 +Step 19587 | loss:0.7853626012802124 lr:4.396786804519974e-05 tokens_per_second_per_gpu:5194.5851069986975 mfu:24.50374423732319 +Step 19588 | loss:0.9097509980201721 lr:4.3963563517897485e-05 tokens_per_second_per_gpu:3687.4833125049868 mfu:17.394487934614983 +Step 19589 | loss:0.751677393913269 lr:4.395925903599605e-05 tokens_per_second_per_gpu:4965.7776953511875 mfu:23.42442063801193 +Step 19590 | loss:0.7747945189476013 lr:4.3954954599527795e-05 tokens_per_second_per_gpu:5146.218348826469 mfu:24.275589986804906 +Step 19591 | loss:0.7703870534896851 lr:4.3950650208525094e-05 tokens_per_second_per_gpu:4624.3492617584925 mfu:21.81384446305749 +Step 19592 | loss:0.8890451192855835 lr:4.3946345863020334e-05 tokens_per_second_per_gpu:5334.827165095632 mfu:25.165289953128898 +Step 19593 | loss:0.6757907867431641 lr:4.3942041563045886e-05 tokens_per_second_per_gpu:5076.576204478451 mfu:23.947076109740227 +Step 19594 | loss:0.7475743889808655 lr:4.3937737308634106e-05 tokens_per_second_per_gpu:4969.727543730741 mfu:23.44305274673191 +Step 19595 | loss:0.7947767972946167 lr:4.393343309981737e-05 tokens_per_second_per_gpu:5352.681756925707 mfu:25.249513109098856 +Step 19596 | loss:0.7520633935928345 lr:4.392912893662807e-05 tokens_per_second_per_gpu:4323.864662923713 mfu:20.396407342394685 +Step 19597 | loss:0.7102947235107422 lr:4.3924824819098555e-05 tokens_per_second_per_gpu:5365.6594807563615 mfu:25.31073124663606 +Step 19598 | loss:0.7358481287956238 lr:4.392052074726121e-05 tokens_per_second_per_gpu:4636.122646503285 mfu:21.869381527645245 +Step 19599 | loss:0.7993635535240173 lr:4.391621672114841e-05 tokens_per_second_per_gpu:5179.365203922895 mfu:24.431949357731394 +Step 19600 | loss:0.8010091185569763 lr:4.3911912740792534e-05 tokens_per_second_per_gpu:4932.9724397969085 mfu:23.26967265040864 +Step 19601 | loss:0.5803737044334412 lr:4.3907608806225924e-05 tokens_per_second_per_gpu:4682.724203124855 mfu:22.089209021274637 +Step 19602 | loss:0.6167025566101074 lr:4.3903304917480957e-05 tokens_per_second_per_gpu:4184.462251939909 mfu:19.73882238528019 +Step 19603 | loss:0.8055350184440613 lr:4.389900107459002e-05 tokens_per_second_per_gpu:3967.309787308617 mfu:18.714477159583442 +Step 19604 | loss:0.7423122525215149 lr:4.389469727758547e-05 tokens_per_second_per_gpu:4926.483537080619 mfu:23.239063389174934 +Step 19605 | loss:0.7739978432655334 lr:4.3890393526499685e-05 tokens_per_second_per_gpu:5267.594431189459 mfu:24.84814167620621 +Step 19606 | loss:0.865375816822052 lr:4.3886089821365026e-05 tokens_per_second_per_gpu:3214.8807186977483 mfu:15.165140865308985 +Step 19607 | loss:0.5824652910232544 lr:4.388178616221389e-05 tokens_per_second_per_gpu:5351.563001540331 mfu:25.244235749066682 +Step 19608 | loss:0.8367404341697693 lr:4.387748254907858e-05 tokens_per_second_per_gpu:4653.466168989631 mfu:21.95119388232341 +Step 19609 | loss:0.6490079164505005 lr:4.387317898199154e-05 tokens_per_second_per_gpu:4642.955144121961 mfu:21.90161158465684 +Step 19610 | loss:0.6966756582260132 lr:4.3868875460985085e-05 tokens_per_second_per_gpu:5349.829046280048 mfu:25.23605638626065 +Step 19611 | loss:0.7088873386383057 lr:4.3864571986091594e-05 tokens_per_second_per_gpu:5349.671412761406 mfu:25.235312802058168 +Step 19612 | loss:0.6879181861877441 lr:4.3860268557343454e-05 tokens_per_second_per_gpu:5297.081606245367 mfu:24.987237711975776 +Step 19613 | loss:0.5937851071357727 lr:4.3855965174773e-05 tokens_per_second_per_gpu:4988.320665547637 mfu:23.53075967465591 +Step 19614 | loss:0.7310189604759216 lr:4.385166183841265e-05 tokens_per_second_per_gpu:5327.098940137938 mfu:25.128834597432313 +Step 19615 | loss:0.6666799187660217 lr:4.384735854829471e-05 tokens_per_second_per_gpu:5309.410287116233 mfu:25.045394203133515 +Step 19616 | loss:0.869140625 lr:4.384305530445158e-05 tokens_per_second_per_gpu:3452.167848263217 mfu:16.284464740828323 +Step 19617 | loss:0.7631986141204834 lr:4.383875210691561e-05 tokens_per_second_per_gpu:3965.309355568587 mfu:18.70504078175687 +Step 19618 | loss:0.7962327599525452 lr:4.3834448955719166e-05 tokens_per_second_per_gpu:5305.956739709144 mfu:25.02910323077833 +Step 19619 | loss:0.7189618349075317 lr:4.383014585089463e-05 tokens_per_second_per_gpu:5378.180707113551 mfu:25.36979600770443 +Step 19620 | loss:0.7412681579589844 lr:4.382584279247433e-05 tokens_per_second_per_gpu:5343.373470985415 mfu:25.205604336161148 +Step 19621 | loss:0.8163553476333618 lr:4.382153978049068e-05 tokens_per_second_per_gpu:5342.691339822924 mfu:25.2023866070841 +Step 19622 | loss:0.7091239094734192 lr:4.381723681497603e-05 tokens_per_second_per_gpu:5349.534408724501 mfu:25.234666530640787 +Step 19623 | loss:0.865632951259613 lr:4.381293389596271e-05 tokens_per_second_per_gpu:5343.441514987984 mfu:25.20592531133077 +Step 19624 | loss:0.7869077920913696 lr:4.3808631023483105e-05 tokens_per_second_per_gpu:4150.634525407936 mfu:19.579251227623576 +Step 19625 | loss:0.7679997682571411 lr:4.3804328197569574e-05 tokens_per_second_per_gpu:4325.732234086704 mfu:20.405216994211013 +Step 19626 | loss:0.6567266583442688 lr:4.380002541825449e-05 tokens_per_second_per_gpu:4713.037377365872 mfu:22.232201436130357 +Step 19627 | loss:0.6373504400253296 lr:4.37957226855702e-05 tokens_per_second_per_gpu:5336.585209190945 mfu:25.17358294707974 +Step 19628 | loss:0.7396892309188843 lr:4.3791419999549064e-05 tokens_per_second_per_gpu:5324.668922761408 mfu:25.117371790863498 +Step 19629 | loss:0.9030335545539856 lr:4.3787117360223474e-05 tokens_per_second_per_gpu:3182.9424915684203 mfu:15.014482798716491 +Step 19630 | loss:0.7606342434883118 lr:4.378281476762576e-05 tokens_per_second_per_gpu:4280.245967982543 mfu:20.190650516240424 +Step 19631 | loss:0.7279202342033386 lr:4.377851222178827e-05 tokens_per_second_per_gpu:5351.769497254678 mfu:25.245209824583057 +Step 19632 | loss:0.8435706496238708 lr:4.3774209722743395e-05 tokens_per_second_per_gpu:4052.6936257320403 mfu:19.117247293411765 +Step 19633 | loss:0.7883361577987671 lr:4.3769907270523473e-05 tokens_per_second_per_gpu:4118.79765921367 mfu:19.429071297855728 +Step 19634 | loss:0.7982637286186218 lr:4.3765604865160886e-05 tokens_per_second_per_gpu:4474.441809501592 mfu:21.10670543391293 +Step 19635 | loss:0.7652633190155029 lr:4.3761302506687976e-05 tokens_per_second_per_gpu:4826.643304697823 mfu:22.7680999785251 +Step 19636 | loss:0.6785573959350586 lr:4.375700019513712e-05 tokens_per_second_per_gpu:4370.286050362106 mfu:20.615384947247 +Step 19637 | loss:0.8157554864883423 lr:4.3752697930540655e-05 tokens_per_second_per_gpu:4395.579815530933 mfu:20.734699953109704 +Step 19638 | loss:0.7262782454490662 lr:4.374839571293094e-05 tokens_per_second_per_gpu:5335.919066853617 mfu:25.17044064001896 +Step 19639 | loss:0.665738582611084 lr:4.374409354234033e-05 tokens_per_second_per_gpu:3673.765197830468 mfu:17.32977724714356 +Step 19640 | loss:0.6980667114257812 lr:4.37397914188012e-05 tokens_per_second_per_gpu:3007.70244150674 mfu:14.187845583539913 +Step 19641 | loss:0.7063653469085693 lr:4.3735489342345894e-05 tokens_per_second_per_gpu:5223.43575210202 mfu:24.639837652703203 +Step 19642 | loss:0.6370660662651062 lr:4.373118731300678e-05 tokens_per_second_per_gpu:3141.0569465440653 mfu:14.816901536425712 +Step 19643 | loss:0.7759711742401123 lr:4.3726885330816214e-05 tokens_per_second_per_gpu:4740.500114209521 mfu:22.36174785144792 +Step 19644 | loss:0.7385779023170471 lr:4.372258339580652e-05 tokens_per_second_per_gpu:4430.625619476334 mfu:20.90001698975089 +Step 19645 | loss:0.8291095495223999 lr:4.371828150801009e-05 tokens_per_second_per_gpu:5334.383986968657 mfu:25.16319940628997 +Step 19646 | loss:0.5847476124763489 lr:4.371397966745927e-05 tokens_per_second_per_gpu:4238.013734316715 mfu:19.99143386447664 +Step 19647 | loss:0.6567790508270264 lr:4.3709677874186394e-05 tokens_per_second_per_gpu:5356.300481992191 mfu:25.26658325265562 +Step 19648 | loss:0.6374852657318115 lr:4.370537612822385e-05 tokens_per_second_per_gpu:4582.215569780354 mfu:21.615092649247476 +Step 19649 | loss:0.8375621438026428 lr:4.370107442960396e-05 tokens_per_second_per_gpu:5362.0498183810305 mfu:25.293703853339714 +Step 19650 | loss:0.6964893937110901 lr:4.369677277835911e-05 tokens_per_second_per_gpu:5355.2840856705225 mfu:25.261788737791214 +Step 19651 | loss:0.713241457939148 lr:4.3692471174521635e-05 tokens_per_second_per_gpu:5029.703814146284 mfu:23.72597104728895 +Step 19652 | loss:0.8035678267478943 lr:4.3688169618123894e-05 tokens_per_second_per_gpu:5355.663137777417 mfu:25.263576791252397 +Step 19653 | loss:0.8043884038925171 lr:4.368386810919822e-05 tokens_per_second_per_gpu:5312.025993065042 mfu:25.0577329343795 +Step 19654 | loss:0.8600918650627136 lr:4.367956664777698e-05 tokens_per_second_per_gpu:4650.963847449747 mfu:21.939390004679698 +Step 19655 | loss:0.609070897102356 lr:4.367526523389253e-05 tokens_per_second_per_gpu:4106.458792697769 mfu:19.370866759272513 +Step 19656 | loss:0.7647792100906372 lr:4.367096386757721e-05 tokens_per_second_per_gpu:5339.206978910712 mfu:25.185950282167923 +Step 19657 | loss:0.7377996444702148 lr:4.366666254886339e-05 tokens_per_second_per_gpu:4161.897217790993 mfu:19.632379269207927 +Step 19658 | loss:0.8290543556213379 lr:4.3662361277783415e-05 tokens_per_second_per_gpu:5369.308969209238 mfu:25.327946506335216 +Step 19659 | loss:0.8369892239570618 lr:4.365806005436962e-05 tokens_per_second_per_gpu:5175.751625713955 mfu:24.41490349277981 +Step 19660 | loss:0.8742943406105042 lr:4.365375887865436e-05 tokens_per_second_per_gpu:4721.892268557266 mfu:22.27397146868039 +Step 19661 | loss:0.7188785076141357 lr:4.364945775066999e-05 tokens_per_second_per_gpu:4685.766984675456 mfu:22.103562341000888 +Step 19662 | loss:0.7504155039787292 lr:4.3645156670448865e-05 tokens_per_second_per_gpu:4987.602694905752 mfu:23.527372884639718 +Step 19663 | loss:0.6767704486846924 lr:4.364085563802332e-05 tokens_per_second_per_gpu:5101.229650065833 mfu:24.063370618887465 +Step 19664 | loss:0.7820780873298645 lr:4.3636554653425714e-05 tokens_per_second_per_gpu:4949.690430343529 mfu:23.348534264200367 +Step 19665 | loss:0.8418276309967041 lr:4.363225371668841e-05 tokens_per_second_per_gpu:4455.675082558383 mfu:21.018179580988026 +Step 19666 | loss:0.7805163860321045 lr:4.362795282784372e-05 tokens_per_second_per_gpu:5324.6747407529565 mfu:25.117399235322285 +Step 19667 | loss:0.7107951045036316 lr:4.362365198692401e-05 tokens_per_second_per_gpu:5319.148891701316 mfu:25.09133286253863 +Step 19668 | loss:0.6726080775260925 lr:4.3619351193961624e-05 tokens_per_second_per_gpu:3078.6141911096893 mfu:14.522348405209096 +Step 19669 | loss:0.7217382788658142 lr:4.361505044898893e-05 tokens_per_second_per_gpu:5247.57540610948 mfu:24.75370851930607 +Step 19670 | loss:0.7447213530540466 lr:4.361074975203823e-05 tokens_per_second_per_gpu:3805.2406970253774 mfu:17.94996960887923 +Step 19671 | loss:0.843926727771759 lr:4.360644910314191e-05 tokens_per_second_per_gpu:5343.463842887213 mfu:25.206030635840968 +Step 19672 | loss:0.7356614470481873 lr:4.360214850233232e-05 tokens_per_second_per_gpu:4466.075397036143 mfu:21.067239638856247 +Step 19673 | loss:0.8012707829475403 lr:4.3597847949641766e-05 tokens_per_second_per_gpu:3939.5521103269953 mfu:18.583539461313084 +Step 19674 | loss:0.8106929063796997 lr:4.359354744510261e-05 tokens_per_second_per_gpu:5014.3804347244995 mfu:23.653688052118273 +Step 19675 | loss:0.7230793237686157 lr:4.358924698874721e-05 tokens_per_second_per_gpu:4582.378694974894 mfu:21.61586213862215 +Step 19676 | loss:0.629571795463562 lr:4.358494658060789e-05 tokens_per_second_per_gpu:5249.880106954231 mfu:24.764580186413273 +Step 19677 | loss:0.7019758224487305 lr:4.3580646220717026e-05 tokens_per_second_per_gpu:5293.857713560643 mfu:24.972030060129125 +Step 19678 | loss:0.7593258619308472 lr:4.357634590910693e-05 tokens_per_second_per_gpu:4971.72101179798 mfu:23.452456275725805 +Step 19679 | loss:0.6488909721374512 lr:4.357204564580997e-05 tokens_per_second_per_gpu:3547.2917502990476 mfu:16.733180416542588 +Step 19680 | loss:0.6603962779045105 lr:4.3567745430858456e-05 tokens_per_second_per_gpu:4789.645008444084 mfu:22.593572702535944 +Step 19681 | loss:0.7019577026367188 lr:4.356344526428475e-05 tokens_per_second_per_gpu:3427.53208820073 mfu:16.168253657319475 +Step 19682 | loss:0.8728554844856262 lr:4.3559145146121204e-05 tokens_per_second_per_gpu:5326.616656753649 mfu:25.126559584422587 +Step 19683 | loss:0.783849835395813 lr:4.355484507640014e-05 tokens_per_second_per_gpu:4047.9576568929947 mfu:19.094906920358383 +Step 19684 | loss:0.7065190672874451 lr:4.3550545055153906e-05 tokens_per_second_per_gpu:3350.461097454763 mfu:15.804696644303778 +Step 19685 | loss:0.7300594449043274 lr:4.354624508241486e-05 tokens_per_second_per_gpu:5352.305676785064 mfu:25.24773907490955 +Step 19686 | loss:0.6937717199325562 lr:4.3541945158215306e-05 tokens_per_second_per_gpu:3863.9006102730436 mfu:18.22667842808171 +Step 19687 | loss:0.7320659160614014 lr:4.353764528258764e-05 tokens_per_second_per_gpu:5358.428510161622 mfu:25.276621524609872 +Step 19688 | loss:0.6911107301712036 lr:4.353334545556414e-05 tokens_per_second_per_gpu:4330.667090893703 mfu:20.428495555743023 +Step 19689 | loss:0.6631293892860413 lr:4.352904567717719e-05 tokens_per_second_per_gpu:5108.302811280013 mfu:24.096735927140497 +Step 19690 | loss:0.6845223903656006 lr:4.35247459474591e-05 tokens_per_second_per_gpu:4914.44415130586 mfu:23.182271552344368 +Step 19691 | loss:0.7933663725852966 lr:4.352044626644223e-05 tokens_per_second_per_gpu:4623.426306648247 mfu:21.80949072633008 +Step 19692 | loss:0.7695016860961914 lr:4.35161466341589e-05 tokens_per_second_per_gpu:3669.837466922703 mfu:17.311249470311637 +Step 19693 | loss:0.7428566813468933 lr:4.351184705064147e-05 tokens_per_second_per_gpu:4429.216049001109 mfu:20.89336780532173 +Step 19694 | loss:0.864279568195343 lr:4.350754751592228e-05 tokens_per_second_per_gpu:4406.916864012786 mfu:20.78817874509853 +Step 19695 | loss:0.692995011806488 lr:4.350324803003364e-05 tokens_per_second_per_gpu:4246.087966750429 mfu:20.02952139647257 +Step 19696 | loss:0.6515038013458252 lr:4.3498948593007884e-05 tokens_per_second_per_gpu:4531.438534791426 mfu:21.375568711749185 +Step 19697 | loss:0.7903547883033752 lr:4.349464920487738e-05 tokens_per_second_per_gpu:4877.5772885391525 mfu:23.0083642705384 +Step 19698 | loss:0.6434110999107361 lr:4.349034986567445e-05 tokens_per_second_per_gpu:4045.6778365236287 mfu:19.084152618698617 +Step 19699 | loss:0.7051047682762146 lr:4.348605057543142e-05 tokens_per_second_per_gpu:3987.5370524846116 mfu:18.809892620545916 +Step 19700 | loss:0.7581342458724976 lr:4.348175133418064e-05 tokens_per_second_per_gpu:5292.522716705847 mfu:24.965732652191104 +Step 19701 | loss:0.7208321690559387 lr:4.347745214195446e-05 tokens_per_second_per_gpu:5255.545818065388 mfu:24.791306312394592 +Step 19702 | loss:0.8391119241714478 lr:4.3473152998785186e-05 tokens_per_second_per_gpu:5313.094901400568 mfu:25.062775157372748 +Step 19703 | loss:0.7485337257385254 lr:4.3468853904705135e-05 tokens_per_second_per_gpu:5341.667910810718 mfu:25.197558917818533 +Step 19704 | loss:0.8469116687774658 lr:4.3464554859746686e-05 tokens_per_second_per_gpu:5333.94576817779 mfu:25.16113225348562 +Step 19705 | loss:0.7140337824821472 lr:4.3460255863942156e-05 tokens_per_second_per_gpu:5257.880167211938 mfu:24.802317835599904 +Step 19706 | loss:0.762101411819458 lr:4.3455956917323866e-05 tokens_per_second_per_gpu:5108.005334142999 mfu:24.09533267673037 +Step 19707 | loss:0.6656920909881592 lr:4.345165801992417e-05 tokens_per_second_per_gpu:5317.469033359403 mfu:25.08340868412644 +Step 19708 | loss:0.7149167060852051 lr:4.344735917177539e-05 tokens_per_second_per_gpu:4916.2856664310875 mfu:23.190958293384945 +Step 19709 | loss:0.8064901232719421 lr:4.344306037290986e-05 tokens_per_second_per_gpu:4706.1488874357265 mfu:22.199707253832283 +Step 19710 | loss:0.8155801892280579 lr:4.343876162335989e-05 tokens_per_second_per_gpu:4055.8523462757967 mfu:19.13214751714996 +Step 19711 | loss:0.858814537525177 lr:4.3434462923157846e-05 tokens_per_second_per_gpu:5329.369911177062 mfu:25.13954715529147 +Step 19712 | loss:0.9379670023918152 lr:4.343016427233603e-05 tokens_per_second_per_gpu:4337.407750405121 mfu:20.460292396733486 +Step 19713 | loss:0.7076154947280884 lr:4.34258656709268e-05 tokens_per_second_per_gpu:4389.911456670751 mfu:20.707961337244477 +Step 19714 | loss:0.7492643594741821 lr:4.342156711896248e-05 tokens_per_second_per_gpu:4010.1361953933647 mfu:18.916496633457882 +Step 19715 | loss:0.7110753059387207 lr:4.341726861647537e-05 tokens_per_second_per_gpu:4281.425899816107 mfu:20.196216456016472 +Step 19716 | loss:0.6997173428535461 lr:4.3412970163497854e-05 tokens_per_second_per_gpu:4371.313694365949 mfu:20.620232519347276 +Step 19717 | loss:0.7750603556632996 lr:4.340867176006222e-05 tokens_per_second_per_gpu:5033.331796482505 mfu:23.743084858966522 +Step 19718 | loss:0.5820741057395935 lr:4.3404373406200796e-05 tokens_per_second_per_gpu:3571.401551261034 mfu:16.8469104612365 +Step 19719 | loss:0.7487172484397888 lr:4.3400075101945914e-05 tokens_per_second_per_gpu:3487.1042032831892 mfu:16.44926548821435 +Step 19720 | loss:0.7582200765609741 lr:4.339577684732993e-05 tokens_per_second_per_gpu:5166.896935680774 mfu:24.37313440912739 +Step 19721 | loss:0.7588615417480469 lr:4.3391478642385145e-05 tokens_per_second_per_gpu:5296.965107062356 mfu:24.986688165452705 +Step 19722 | loss:0.7568171620368958 lr:4.3387180487143876e-05 tokens_per_second_per_gpu:4895.9966616950705 mfu:23.095251596383992 +Step 19723 | loss:0.544293224811554 lr:4.338288238163849e-05 tokens_per_second_per_gpu:5332.72374782335 mfu:25.155367774975552 +Step 19724 | loss:0.5664299130439758 lr:4.337858432590128e-05 tokens_per_second_per_gpu:4341.37219984978 mfu:20.478993381168763 +Step 19725 | loss:0.8002696633338928 lr:4.3374286319964586e-05 tokens_per_second_per_gpu:5372.607510111605 mfu:25.34350628656079 +Step 19726 | loss:0.7731311321258545 lr:4.3369988363860716e-05 tokens_per_second_per_gpu:5366.878444154195 mfu:25.316481304960238 +Step 19727 | loss:0.8359218239784241 lr:4.336569045762202e-05 tokens_per_second_per_gpu:5268.315449128572 mfu:24.851542840843358 +Step 19728 | loss:0.8020707964897156 lr:4.3361392601280804e-05 tokens_per_second_per_gpu:4327.307567097469 mfu:20.41264810880232 +Step 19729 | loss:0.8264729380607605 lr:4.3357094794869415e-05 tokens_per_second_per_gpu:3956.975241698408 mfu:18.66572734468478 +Step 19730 | loss:0.8819887042045593 lr:4.335279703842017e-05 tokens_per_second_per_gpu:5239.183543055241 mfu:24.714122669479746 +Step 19731 | loss:0.7680438756942749 lr:4.3348499331965376e-05 tokens_per_second_per_gpu:4501.329199063398 mfu:21.233537837937675 +Step 19732 | loss:0.7868151664733887 lr:4.334420167553736e-05 tokens_per_second_per_gpu:5372.118237904244 mfu:25.341198306080223 +Step 19733 | loss:0.6638734936714172 lr:4.333990406916845e-05 tokens_per_second_per_gpu:5341.962214503584 mfu:25.198947198551167 +Step 19734 | loss:0.7785353660583496 lr:4.333560651289099e-05 tokens_per_second_per_gpu:3482.3279473616317 mfu:16.426735074119073 +Step 19735 | loss:0.8082897663116455 lr:4.333130900673726e-05 tokens_per_second_per_gpu:4773.677170741481 mfu:22.518249687698724 +Step 19736 | loss:0.6706920266151428 lr:4.332701155073962e-05 tokens_per_second_per_gpu:4215.405228595827 mfu:19.884785685582365 +Step 19737 | loss:0.7938079833984375 lr:4.332271414493039e-05 tokens_per_second_per_gpu:4209.174714165162 mfu:19.855395285977544 +Step 19738 | loss:0.6157544851303101 lr:4.3318416789341864e-05 tokens_per_second_per_gpu:4154.717531039722 mfu:19.59851146182131 +Step 19739 | loss:0.7861327528953552 lr:4.3314119484006374e-05 tokens_per_second_per_gpu:5347.014070570394 mfu:25.222777665553835 +Step 19740 | loss:0.7766515612602234 lr:4.3309822228956255e-05 tokens_per_second_per_gpu:5319.354271889232 mfu:25.092301675926503 +Step 19741 | loss:0.6658534407615662 lr:4.330552502422381e-05 tokens_per_second_per_gpu:5320.174899038063 mfu:25.09617271420097 +Step 19742 | loss:0.8056036829948425 lr:4.330122786984136e-05 tokens_per_second_per_gpu:5353.175545148987 mfu:25.251842392397023 +Step 19743 | loss:0.5987640619277954 lr:4.329693076584124e-05 tokens_per_second_per_gpu:4447.0580133934445 mfu:20.977531395513406 +Step 19744 | loss:0.811663031578064 lr:4.329263371225575e-05 tokens_per_second_per_gpu:5315.9807106977905 mfu:25.076388012197583 +Step 19745 | loss:0.7401796579360962 lr:4.328833670911724e-05 tokens_per_second_per_gpu:5341.217019592861 mfu:25.195431987013926 +Step 19746 | loss:0.6821495890617371 lr:4.328403975645798e-05 tokens_per_second_per_gpu:4906.206230879497 mfu:23.143411876159053 +Step 19747 | loss:0.8591764569282532 lr:4.327974285431033e-05 tokens_per_second_per_gpu:4895.2000794463 mfu:23.091493982005595 +Step 19748 | loss:0.7247530817985535 lr:4.327544600270658e-05 tokens_per_second_per_gpu:5327.587590199886 mfu:25.131139643146813 +Step 19749 | loss:0.7924479246139526 lr:4.327114920167906e-05 tokens_per_second_per_gpu:5355.807581345727 mfu:25.26425815620873 +Step 19750 | loss:0.7554475665092468 lr:4.3266852451260095e-05 tokens_per_second_per_gpu:4720.613863266638 mfu:22.26794101280606 +Step 19751 | loss:0.8087409138679504 lr:4.326255575148198e-05 tokens_per_second_per_gpu:5040.345504127738 mfu:23.77616971459101 +Step 19752 | loss:0.8711804151535034 lr:4.325825910237706e-05 tokens_per_second_per_gpu:4288.361462018102 mfu:20.228932686252218 +Step 19753 | loss:0.6472544074058533 lr:4.325396250397762e-05 tokens_per_second_per_gpu:4613.883202622606 mfu:21.764474276419687 +Step 19754 | loss:0.7881105542182922 lr:4.3249665956316e-05 tokens_per_second_per_gpu:3937.9062381059416 mfu:18.575775601231616 +Step 19755 | loss:0.7694342732429504 lr:4.3245369459424483e-05 tokens_per_second_per_gpu:4244.399823696298 mfu:20.021558137659333 +Step 19756 | loss:0.7122296690940857 lr:4.324107301333542e-05 tokens_per_second_per_gpu:4071.9414969624545 mfu:19.208042786031097 +Step 19757 | loss:0.6939201354980469 lr:4.323677661808111e-05 tokens_per_second_per_gpu:4427.532111511249 mfu:20.8854243848727 +Step 19758 | loss:0.7406409978866577 lr:4.323248027369385e-05 tokens_per_second_per_gpu:3866.873721803089 mfu:18.24070310243407 +Step 19759 | loss:0.7140946984291077 lr:4.3228183980205996e-05 tokens_per_second_per_gpu:3899.729057416295 mfu:18.395687326219853 +Step 19760 | loss:0.5891134738922119 lr:4.322388773764982e-05 tokens_per_second_per_gpu:5072.061120505526 mfu:23.925777688287177 +Step 19761 | loss:0.7181256413459778 lr:4.321959154605765e-05 tokens_per_second_per_gpu:4197.998240396431 mfu:19.80267395230738 +Step 19762 | loss:0.7039538025856018 lr:4.321529540546179e-05 tokens_per_second_per_gpu:4511.743155476227 mfu:21.282662247141168 +Step 19763 | loss:0.6282683610916138 lr:4.321099931589456e-05 tokens_per_second_per_gpu:4404.227806140146 mfu:20.775494000312705 +Step 19764 | loss:0.6691939234733582 lr:4.320670327738827e-05 tokens_per_second_per_gpu:5214.460414778088 mfu:24.597499455195763 +Step 19765 | loss:0.7513014674186707 lr:4.3202407289975235e-05 tokens_per_second_per_gpu:3555.481448829216 mfu:16.771812621816046 +Step 19766 | loss:0.7036909461021423 lr:4.319811135368778e-05 tokens_per_second_per_gpu:5309.347886926108 mfu:25.045099850790127 +Step 19767 | loss:0.7632387280464172 lr:4.319381546855818e-05 tokens_per_second_per_gpu:5324.299649748229 mfu:25.115629867055862 +Step 19768 | loss:0.7418915033340454 lr:4.318951963461876e-05 tokens_per_second_per_gpu:4853.150767894609 mfu:22.893140205063922 +Step 19769 | loss:0.7856771349906921 lr:4.318522385190183e-05 tokens_per_second_per_gpu:5362.1733641060755 mfu:25.294286639603826 +Step 19770 | loss:0.7329418659210205 lr:4.318092812043971e-05 tokens_per_second_per_gpu:3923.3355935450522 mfu:18.50704338482959 +Step 19771 | loss:0.7163877487182617 lr:4.317663244026469e-05 tokens_per_second_per_gpu:5370.418644813725 mfu:25.333181035491762 +Step 19772 | loss:0.7437474727630615 lr:4.31723368114091e-05 tokens_per_second_per_gpu:4927.83074121983 mfu:23.245418380956323 +Step 19773 | loss:0.7655771970748901 lr:4.316804123390523e-05 tokens_per_second_per_gpu:4057.4309025888497 mfu:19.139593836608114 +Step 19774 | loss:0.7637251019477844 lr:4.316374570778541e-05 tokens_per_second_per_gpu:4597.943262177629 mfu:21.689282857706452 +Step 19775 | loss:0.7068198919296265 lr:4.31594502330819e-05 tokens_per_second_per_gpu:4117.156645781678 mfu:19.42133035751126 +Step 19776 | loss:0.7034890651702881 lr:4.3155154809827056e-05 tokens_per_second_per_gpu:3805.246360408687 mfu:17.949996324024575 +Step 19777 | loss:0.8198694586753845 lr:4.315085943805317e-05 tokens_per_second_per_gpu:5327.565180202444 mfu:25.131033931365344 +Step 19778 | loss:0.7441319227218628 lr:4.314656411779252e-05 tokens_per_second_per_gpu:3990.5074993619055 mfu:18.823904725276606 +Step 19779 | loss:0.7020925283432007 lr:4.314226884907746e-05 tokens_per_second_per_gpu:3958.8118540978976 mfu:18.674390958731564 +Step 19780 | loss:0.6541063785552979 lr:4.313797363194025e-05 tokens_per_second_per_gpu:5045.397808655254 mfu:23.80000229705934 +Step 19781 | loss:0.7792747616767883 lr:4.3133678466413254e-05 tokens_per_second_per_gpu:3583.9940649432037 mfu:16.906311496779676 +Step 19782 | loss:0.6624940037727356 lr:4.312938335252871e-05 tokens_per_second_per_gpu:4080.1812457246833 mfu:19.24691108678864 +Step 19783 | loss:0.8082074522972107 lr:4.3125088290318956e-05 tokens_per_second_per_gpu:4533.417490706969 mfu:21.3849037844475 +Step 19784 | loss:0.8008134961128235 lr:4.312079327981627e-05 tokens_per_second_per_gpu:4475.556401534254 mfu:21.111963154700007 +Step 19785 | loss:0.7483175992965698 lr:4.3116498321053e-05 tokens_per_second_per_gpu:3260.488910394931 mfu:15.380282487104507 +Step 19786 | loss:0.7784563302993774 lr:4.311220341406142e-05 tokens_per_second_per_gpu:5174.321757496307 mfu:24.40815856043649 +Step 19787 | loss:0.7102149128913879 lr:4.3107908558873826e-05 tokens_per_second_per_gpu:3659.53159340828 mfu:17.262634906580768 +Step 19788 | loss:0.6903843879699707 lr:4.310361375552255e-05 tokens_per_second_per_gpu:4231.239579222135 mfu:19.95947901910546 +Step 19789 | loss:0.7600081562995911 lr:4.3099319004039874e-05 tokens_per_second_per_gpu:3914.542687657663 mfu:18.46556574753433 +Step 19790 | loss:0.7971662282943726 lr:4.309502430445809e-05 tokens_per_second_per_gpu:4792.998808123697 mfu:22.60939314784191 +Step 19791 | loss:0.6898777484893799 lr:4.309072965680951e-05 tokens_per_second_per_gpu:5064.957894068798 mfu:23.89227055724973 +Step 19792 | loss:0.6698285937309265 lr:4.3086435061126435e-05 tokens_per_second_per_gpu:5314.98585190956 mfu:25.071695093553085 +Step 19793 | loss:0.7737588882446289 lr:4.308214051744117e-05 tokens_per_second_per_gpu:5173.845962383905 mfu:24.40591415371227 +Step 19794 | loss:0.8418765664100647 lr:4.307784602578599e-05 tokens_per_second_per_gpu:4257.158715035133 mfu:20.081744004995954 +Step 19795 | loss:0.7052077054977417 lr:4.307355158619324e-05 tokens_per_second_per_gpu:5028.414743275078 mfu:23.719890280051306 +Step 19796 | loss:0.7517687678337097 lr:4.306925719869518e-05 tokens_per_second_per_gpu:4184.21500342636 mfu:19.737656071856833 +Step 19797 | loss:0.7708490490913391 lr:4.306496286332412e-05 tokens_per_second_per_gpu:4852.438887195804 mfu:22.889782142348288 +Step 19798 | loss:0.8140186071395874 lr:4.306066858011235e-05 tokens_per_second_per_gpu:5153.475478700721 mfu:24.309823106615788 +Step 19799 | loss:0.7981769442558289 lr:4.305637434909219e-05 tokens_per_second_per_gpu:5361.1977282562275 mfu:25.289684398839455 +Step 19800 | loss:0.6826037168502808 lr:4.305208017029591e-05 tokens_per_second_per_gpu:5114.168768485676 mfu:24.124406648115116 +Step 19801 | loss:0.7502896189689636 lr:4.304778604375583e-05 tokens_per_second_per_gpu:4292.485709905515 mfu:20.248387467206395 +Step 19802 | loss:0.6334593296051025 lr:4.304349196950426e-05 tokens_per_second_per_gpu:5321.963676088269 mfu:25.104610680743622 +Step 19803 | loss:0.7018848061561584 lr:4.303919794757346e-05 tokens_per_second_per_gpu:4432.63056663465 mfu:20.909474667575925 +Step 19804 | loss:0.6990495920181274 lr:4.3034903977995725e-05 tokens_per_second_per_gpu:5164.631820434906 mfu:24.362449474044116 +Step 19805 | loss:0.6971327662467957 lr:4.303061006080338e-05 tokens_per_second_per_gpu:5347.8808442855825 mfu:25.226866385055224 +Step 19806 | loss:0.6590802669525146 lr:4.30263161960287e-05 tokens_per_second_per_gpu:5354.798097103617 mfu:25.25949624680292 +Step 19807 | loss:0.6520469188690186 lr:4.3022022383703976e-05 tokens_per_second_per_gpu:4326.123013656911 mfu:20.40706036811755 +Step 19808 | loss:0.7620059251785278 lr:4.301772862386153e-05 tokens_per_second_per_gpu:4171.487576212226 mfu:19.677618626164755 +Step 19809 | loss:0.746587336063385 lr:4.301343491653364e-05 tokens_per_second_per_gpu:3764.1995309781 mfu:17.756371426289235 +Step 19810 | loss:0.7445773482322693 lr:4.3009141261752595e-05 tokens_per_second_per_gpu:5327.209738324456 mfu:25.12935725138224 +Step 19811 | loss:0.7831258177757263 lr:4.3004847659550675e-05 tokens_per_second_per_gpu:4511.516696902095 mfu:21.28159400340917 +Step 19812 | loss:0.7122017741203308 lr:4.30005541099602e-05 tokens_per_second_per_gpu:5318.4715467260385 mfu:25.088137710723277 +Step 19813 | loss:0.8012029528617859 lr:4.299626061301345e-05 tokens_per_second_per_gpu:4035.4355634888448 mfu:19.035838069281564 +Step 19814 | loss:0.7402300238609314 lr:4.2991967168742706e-05 tokens_per_second_per_gpu:5327.714743480606 mfu:25.131739446866558 +Step 19815 | loss:0.7705245018005371 lr:4.298767377718029e-05 tokens_per_second_per_gpu:5314.394434092016 mfu:25.06890527480145 +Step 19816 | loss:0.7666211128234863 lr:4.2983380438358454e-05 tokens_per_second_per_gpu:5345.925252961843 mfu:25.217641527122858 +Step 19817 | loss:0.6859833598136902 lr:4.297908715230955e-05 tokens_per_second_per_gpu:3755.329606776233 mfu:17.714530480463907 +Step 19818 | loss:0.7715780735015869 lr:4.297479391906579e-05 tokens_per_second_per_gpu:4653.510464816696 mfu:21.951402833297323 +Step 19819 | loss:0.658526599407196 lr:4.297050073865951e-05 tokens_per_second_per_gpu:5364.584233097402 mfu:25.305659119972578 +Step 19820 | loss:0.829454779624939 lr:4.2966207611122983e-05 tokens_per_second_per_gpu:5249.870702672645 mfu:24.764535824812494 +Step 19821 | loss:0.6045826077461243 lr:4.2961914536488515e-05 tokens_per_second_per_gpu:4821.582413958735 mfu:22.744226893431822 +Step 19822 | loss:0.6995425820350647 lr:4.295762151478839e-05 tokens_per_second_per_gpu:4808.561653385428 mfu:22.6828057442367 +Step 19823 | loss:0.6768108010292053 lr:4.295332854605487e-05 tokens_per_second_per_gpu:5370.5248910566415 mfu:25.33368221714708 +Step 19824 | loss:0.6446424126625061 lr:4.2949035630320314e-05 tokens_per_second_per_gpu:4757.487961087924 mfu:22.441882423598905 +Step 19825 | loss:0.7485648989677429 lr:4.294474276761692e-05 tokens_per_second_per_gpu:4558.224815875796 mfu:21.501924169834975 +Step 19826 | loss:0.7869437336921692 lr:4.2940449957977025e-05 tokens_per_second_per_gpu:4745.70065350008 mfu:22.386279682584988 +Step 19827 | loss:0.7068119645118713 lr:4.29361572014329e-05 tokens_per_second_per_gpu:4302.435363613585 mfu:20.29532168133333 +Step 19828 | loss:0.8950884938240051 lr:4.2931864498016846e-05 tokens_per_second_per_gpu:5337.399097711909 mfu:25.177422198096828 +Step 19829 | loss:0.5584287643432617 lr:4.292757184776114e-05 tokens_per_second_per_gpu:5304.414051478566 mfu:25.02182610718494 +Step 19830 | loss:0.8210095167160034 lr:4.2923279250698054e-05 tokens_per_second_per_gpu:3859.9060588205325 mfu:18.207835447339455 +Step 19831 | loss:0.765501081943512 lr:4.291898670685992e-05 tokens_per_second_per_gpu:5311.842235724668 mfu:25.056866119652724 +Step 19832 | loss:0.9676982760429382 lr:4.2914694216278975e-05 tokens_per_second_per_gpu:3732.548649340577 mfu:17.607068817407782 +Step 19833 | loss:0.747490406036377 lr:4.291040177898752e-05 tokens_per_second_per_gpu:5247.13116376464 mfu:24.7516129523699 +Step 19834 | loss:0.6461158990859985 lr:4.2906109395017823e-05 tokens_per_second_per_gpu:3981.561417828261 mfu:18.781704532323502 +Step 19835 | loss:0.8028410077095032 lr:4.2901817064402195e-05 tokens_per_second_per_gpu:4444.694424845677 mfu:20.966381945063794 +Step 19836 | loss:0.8165761828422546 lr:4.289752478717289e-05 tokens_per_second_per_gpu:4917.178590034359 mfu:23.195170366370053 +Step 19837 | loss:0.6145431399345398 lr:4.289323256336223e-05 tokens_per_second_per_gpu:5274.6965716506475 mfu:24.881643684511754 +Step 19838 | loss:0.8015223741531372 lr:4.288894039300247e-05 tokens_per_second_per_gpu:4403.4166867927925 mfu:20.771667811960008 +Step 19839 | loss:0.701603889465332 lr:4.288464827612591e-05 tokens_per_second_per_gpu:3521.757191400083 mfu:16.612729545571177 +Step 19840 | loss:0.7787157893180847 lr:4.2880356212764786e-05 tokens_per_second_per_gpu:3869.73043143903 mfu:18.254178689191722 +Step 19841 | loss:0.842009425163269 lr:4.2876064202951435e-05 tokens_per_second_per_gpu:4641.918121937331 mfu:21.896719774077717 +Step 19842 | loss:0.8913244605064392 lr:4.287177224671811e-05 tokens_per_second_per_gpu:4721.161662258721 mfu:22.270525074115863 +Step 19843 | loss:0.7635471224784851 lr:4.286748034409708e-05 tokens_per_second_per_gpu:4159.500808113883 mfu:19.621074995891348 +Step 19844 | loss:0.7804147005081177 lr:4.2863188495120655e-05 tokens_per_second_per_gpu:5351.918476976614 mfu:25.245912587349995 +Step 19845 | loss:0.7270203232765198 lr:4.28588966998211e-05 tokens_per_second_per_gpu:4681.069588731836 mfu:22.08140392287644 +Step 19846 | loss:0.7619348168373108 lr:4.2854604958230706e-05 tokens_per_second_per_gpu:5305.295952774879 mfu:25.025986186067524 +Step 19847 | loss:0.7624087929725647 lr:4.285031327038171e-05 tokens_per_second_per_gpu:4257.140934616917 mfu:20.081660131729574 +Step 19848 | loss:0.6958211660385132 lr:4.2846021636306444e-05 tokens_per_second_per_gpu:4715.830795543234 mfu:22.24537846627926 +Step 19849 | loss:0.6745590567588806 lr:4.2841730056037154e-05 tokens_per_second_per_gpu:3904.81535633834 mfu:18.419680266047223 +Step 19850 | loss:0.7205231189727783 lr:4.2837438529606125e-05 tokens_per_second_per_gpu:4290.19340546972 mfu:20.237574276075286 +Step 19851 | loss:0.8108119368553162 lr:4.283314705704563e-05 tokens_per_second_per_gpu:3638.509945783391 mfu:17.163472208071234 +Step 19852 | loss:0.7988252639770508 lr:4.282885563838795e-05 tokens_per_second_per_gpu:4338.862392256469 mfu:20.467154190533783 +Step 19853 | loss:0.8020350933074951 lr:4.282456427366539e-05 tokens_per_second_per_gpu:5170.314925859392 mfu:24.38925765196919 +Step 19854 | loss:0.7874613404273987 lr:4.282027296291017e-05 tokens_per_second_per_gpu:3658.8017478733 mfu:17.25919210066778 +Step 19855 | loss:0.6873412132263184 lr:4.2815981706154596e-05 tokens_per_second_per_gpu:4040.2984012087486 mfu:19.05877690449202 +Step 19856 | loss:0.8730193972587585 lr:4.281169050343094e-05 tokens_per_second_per_gpu:4477.140448828793 mfu:21.119435376054685 +Step 19857 | loss:0.7536372542381287 lr:4.280739935477148e-05 tokens_per_second_per_gpu:3952.3153146199334 mfu:18.643745673590075 +Step 19858 | loss:0.8519102931022644 lr:4.2803108260208486e-05 tokens_per_second_per_gpu:3711.6704254136225 mfu:17.508582672952013 +Step 19859 | loss:0.756645143032074 lr:4.2798817219774225e-05 tokens_per_second_per_gpu:4974.076231500368 mfu:23.463566248903884 +Step 19860 | loss:0.6538747549057007 lr:4.2794526233501006e-05 tokens_per_second_per_gpu:4200.774760987002 mfu:19.815771273656363 +Step 19861 | loss:0.7205131649971008 lr:4.2790235301421057e-05 tokens_per_second_per_gpu:5309.2239401618235 mfu:25.0445151727555 +Step 19862 | loss:0.7236116528511047 lr:4.2785944423566666e-05 tokens_per_second_per_gpu:5051.0320071806445 mfu:23.826579772797004 +Step 19863 | loss:0.6921645998954773 lr:4.27816535999701e-05 tokens_per_second_per_gpu:5348.784553120212 mfu:25.231129333816796 +Step 19864 | loss:0.6447679400444031 lr:4.277736283066365e-05 tokens_per_second_per_gpu:4954.653487987506 mfu:23.371945853891955 +Step 19865 | loss:0.7453320622444153 lr:4.2773072115679576e-05 tokens_per_second_per_gpu:5385.1136572376745 mfu:25.402499916323187 +Step 19866 | loss:0.6087790131568909 lr:4.276878145505013e-05 tokens_per_second_per_gpu:3867.636181897353 mfu:18.244299756787814 +Step 19867 | loss:0.686687171459198 lr:4.276449084880762e-05 tokens_per_second_per_gpu:3731.120645790028 mfu:17.600332681017314 +Step 19868 | loss:0.7884448766708374 lr:4.276020029698432e-05 tokens_per_second_per_gpu:5345.937373111129 mfu:25.217698699934953 +Step 19869 | loss:0.7687026858329773 lr:4.275590979961245e-05 tokens_per_second_per_gpu:5328.933813603241 mfu:25.13749000862984 +Step 19870 | loss:0.8345701694488525 lr:4.27516193567243e-05 tokens_per_second_per_gpu:5364.894834982298 mfu:25.307124282057778 +Step 19871 | loss:0.903332531452179 lr:4.2747328968352155e-05 tokens_per_second_per_gpu:4209.245467347183 mfu:19.855729040808875 +Step 19872 | loss:0.7120344638824463 lr:4.274303863452826e-05 tokens_per_second_per_gpu:3808.503100562079 mfu:17.965358922985292 +Step 19873 | loss:0.715170681476593 lr:4.2738748355284916e-05 tokens_per_second_per_gpu:4983.240349847671 mfu:23.50679495870793 +Step 19874 | loss:0.7482008934020996 lr:4.273445813065438e-05 tokens_per_second_per_gpu:5343.6948811359825 mfu:25.2071204826796 +Step 19875 | loss:0.8021466732025146 lr:4.2730167960668905e-05 tokens_per_second_per_gpu:3469.879097782491 mfu:16.36801172666151 +Step 19876 | loss:0.803432822227478 lr:4.272587784536076e-05 tokens_per_second_per_gpu:5117.366539139489 mfu:24.13949107788508 +Step 19877 | loss:0.8344010710716248 lr:4.272158778476221e-05 tokens_per_second_per_gpu:4496.165573627419 mfu:21.209180135750817 +Step 19878 | loss:0.6145962476730347 lr:4.271729777890553e-05 tokens_per_second_per_gpu:5332.4131955650855 mfu:25.153902846987634 +Step 19879 | loss:0.7361627221107483 lr:4.2713007827822974e-05 tokens_per_second_per_gpu:5354.167053265985 mfu:25.25651950535363 +Step 19880 | loss:0.7581033706665039 lr:4.2708717931546825e-05 tokens_per_second_per_gpu:5049.845494978245 mfu:23.82098279229843 +Step 19881 | loss:0.7518348693847656 lr:4.2704428090109344e-05 tokens_per_second_per_gpu:4324.632857898898 mfu:20.40003104916021 +Step 19882 | loss:0.6904518008232117 lr:4.27001383035428e-05 tokens_per_second_per_gpu:5367.341073735514 mfu:25.318663607628068 +Step 19883 | loss:0.5461960434913635 lr:4.269584857187943e-05 tokens_per_second_per_gpu:3806.7903676221886 mfu:17.957279669485455 +Step 19884 | loss:0.6362373232841492 lr:4.2691558895151515e-05 tokens_per_second_per_gpu:4559.523857212989 mfu:21.508051969461825 +Step 19885 | loss:0.7322925329208374 lr:4.2687269273391324e-05 tokens_per_second_per_gpu:3882.062669349684 mfu:18.31235195951303 +Step 19886 | loss:0.7483596205711365 lr:4.2682979706631104e-05 tokens_per_second_per_gpu:3804.52850947763 mfu:17.946610098704777 +Step 19887 | loss:0.8192206621170044 lr:4.2678690194903136e-05 tokens_per_second_per_gpu:4892.778171037185 mfu:23.080069426819612 +Step 19888 | loss:0.7137888669967651 lr:4.267440073823966e-05 tokens_per_second_per_gpu:3896.1880947230934 mfu:18.378984001045488 +Step 19889 | loss:0.6797982454299927 lr:4.267011133667299e-05 tokens_per_second_per_gpu:4114.443017544943 mfu:19.408529710125865 +Step 19890 | loss:0.755279004573822 lr:4.266582199023531e-05 tokens_per_second_per_gpu:4729.2596894684875 mfu:22.308724850130243 +Step 19891 | loss:0.7672846913337708 lr:4.2661532698958936e-05 tokens_per_second_per_gpu:4428.31084191204 mfu:20.88909778903834 +Step 19892 | loss:0.8561673164367676 lr:4.26572434628761e-05 tokens_per_second_per_gpu:4947.9805553593615 mfu:23.34046849216529 +Step 19893 | loss:0.7198336720466614 lr:4.265295428201907e-05 tokens_per_second_per_gpu:4158.582780448713 mfu:19.61674450276307 +Step 19894 | loss:0.7680936455726624 lr:4.264866515642012e-05 tokens_per_second_per_gpu:4631.599549575994 mfu:21.848045307719303 +Step 19895 | loss:0.9316532611846924 lr:4.264437608611148e-05 tokens_per_second_per_gpu:5301.6351282035275 mfu:25.00871745196338 +Step 19896 | loss:0.7051621079444885 lr:4.264008707112547e-05 tokens_per_second_per_gpu:5350.806946036916 mfu:25.240669306260703 +Step 19897 | loss:0.7613003253936768 lr:4.263579811149426e-05 tokens_per_second_per_gpu:5016.243399050114 mfu:23.662475972696686 +Step 19898 | loss:0.74252849817276 lr:4.263150920725016e-05 tokens_per_second_per_gpu:4938.695554457669 mfu:23.296669558727494 +Step 19899 | loss:0.6380071043968201 lr:4.2627220358425426e-05 tokens_per_second_per_gpu:5356.0912377123 mfu:25.26559621168662 +Step 19900 | loss:0.8540810346603394 lr:4.2622931565052304e-05 tokens_per_second_per_gpu:4390.761336058152 mfu:20.71197036331922 +Step 19901 | loss:0.7037386894226074 lr:4.2618642827163066e-05 tokens_per_second_per_gpu:3868.507375450097 mfu:18.248409325417054 +Step 19902 | loss:0.6476664543151855 lr:4.261435414478994e-05 tokens_per_second_per_gpu:4838.637149300597 mfu:22.82467698987729 +Step 19903 | loss:0.8313484787940979 lr:4.2610065517965206e-05 tokens_per_second_per_gpu:5120.037924060649 mfu:24.15209245634346 +Step 19904 | loss:0.6070886850357056 lr:4.2605776946721135e-05 tokens_per_second_per_gpu:5212.98633716806 mfu:24.590545979605686 +Step 19905 | loss:0.5885352492332458 lr:4.260148843108993e-05 tokens_per_second_per_gpu:5332.388916908709 mfu:25.153788320423143 +Step 19906 | loss:0.7317751049995422 lr:4.259719997110388e-05 tokens_per_second_per_gpu:4434.8790598235755 mfu:20.920081193580582 +Step 19907 | loss:0.7096647024154663 lr:4.2592911566795234e-05 tokens_per_second_per_gpu:4126.058532980113 mfu:19.46332207824485 +Step 19908 | loss:0.8690410852432251 lr:4.258862321819625e-05 tokens_per_second_per_gpu:5144.777324519428 mfu:24.26879242928449 +Step 19909 | loss:0.7896257042884827 lr:4.258433492533917e-05 tokens_per_second_per_gpu:4444.020430046941 mfu:20.96320259660279 +Step 19910 | loss:0.6854103207588196 lr:4.258004668825626e-05 tokens_per_second_per_gpu:4560.309761155734 mfu:21.511759212448794 +Step 19911 | loss:0.7486419081687927 lr:4.257575850697977e-05 tokens_per_second_per_gpu:4228.360607034454 mfu:19.945898416092867 +Step 19912 | loss:0.6536450982093811 lr:4.257147038154195e-05 tokens_per_second_per_gpu:4539.660164622357 mfu:21.414351542416384 +Step 19913 | loss:0.8245639204978943 lr:4.256718231197503e-05 tokens_per_second_per_gpu:5021.853473159564 mfu:23.688939649448308 +Step 19914 | loss:0.8708690404891968 lr:4.256289429831129e-05 tokens_per_second_per_gpu:4562.722815557118 mfu:21.52314200176966 +Step 19915 | loss:0.6152352690696716 lr:4.255860634058296e-05 tokens_per_second_per_gpu:5376.380122357392 mfu:25.361302342198194 +Step 19916 | loss:0.661038339138031 lr:4.2554318438822314e-05 tokens_per_second_per_gpu:4254.614626535318 mfu:20.069743105476068 +Step 19917 | loss:0.6762785911560059 lr:4.2550030593061597e-05 tokens_per_second_per_gpu:4250.657161909686 mfu:20.05107506962387 +Step 19918 | loss:0.6957787275314331 lr:4.254574280333306e-05 tokens_per_second_per_gpu:5252.730386434611 mfu:24.77802544103005 +Step 19919 | loss:0.8244421482086182 lr:4.25414550696689e-05 tokens_per_second_per_gpu:4707.151385795295 mfu:22.204436209638228 +Step 19920 | loss:0.6697196364402771 lr:4.253716739210144e-05 tokens_per_second_per_gpu:5317.862713548627 mfu:25.085265740747772 +Step 19921 | loss:0.8152341842651367 lr:4.2532879770662895e-05 tokens_per_second_per_gpu:5329.3224007649305 mfu:25.139323040571337 +Step 19922 | loss:0.7659034729003906 lr:4.2528592205385496e-05 tokens_per_second_per_gpu:4700.143040859026 mfu:22.171376650826897 +Step 19923 | loss:0.7486931085586548 lr:4.252430469630152e-05 tokens_per_second_per_gpu:4615.294611496759 mfu:21.77113213289025 +Step 19924 | loss:0.8232045769691467 lr:4.25200172434432e-05 tokens_per_second_per_gpu:5334.886651726308 mfu:25.16557056171526 +Step 19925 | loss:0.7608141303062439 lr:4.251572984684281e-05 tokens_per_second_per_gpu:5377.148999073861 mfu:25.36492926485367 +Step 19926 | loss:0.8004727959632874 lr:4.2511442506532535e-05 tokens_per_second_per_gpu:4648.923539269715 mfu:21.929765522882015 +Step 19927 | loss:0.7350078225135803 lr:4.250715522254467e-05 tokens_per_second_per_gpu:5356.3784370455005 mfu:25.266950979942585 +Step 19928 | loss:0.7379458546638489 lr:4.250286799491144e-05 tokens_per_second_per_gpu:4831.390725543191 mfu:22.790494372646226 +Step 19929 | loss:0.630512535572052 lr:4.24985808236651e-05 tokens_per_second_per_gpu:5343.518381015113 mfu:25.206287901495433 +Step 19930 | loss:0.8511496186256409 lr:4.2494293708837894e-05 tokens_per_second_per_gpu:5184.856288924452 mfu:24.45785174255945 +Step 19931 | loss:0.8357868790626526 lr:4.2490006650462056e-05 tokens_per_second_per_gpu:3576.936296580904 mfu:16.873018799235254 +Step 19932 | loss:0.7204462885856628 lr:4.2485719648569845e-05 tokens_per_second_per_gpu:5355.18111830107 mfu:25.26130302314146 +Step 19933 | loss:0.7130536437034607 lr:4.24814327031935e-05 tokens_per_second_per_gpu:3865.8807740660386 mfu:18.236019198544046 +Step 19934 | loss:0.6264187693595886 lr:4.2477145814365255e-05 tokens_per_second_per_gpu:4553.350536158275 mfu:21.478931360769554 +Step 19935 | loss:0.5349507927894592 lr:4.2472858982117345e-05 tokens_per_second_per_gpu:4906.75901364906 mfu:23.14601944679755 +Step 19936 | loss:0.6812695264816284 lr:4.246857220648203e-05 tokens_per_second_per_gpu:3675.4658425482266 mfu:17.337799478437137 +Step 19937 | loss:0.6375654935836792 lr:4.246428548749155e-05 tokens_per_second_per_gpu:4198.841425341657 mfu:19.806651399556195 +Step 19938 | loss:0.6338279843330383 lr:4.2459998825178126e-05 tokens_per_second_per_gpu:4533.812298173493 mfu:21.386766158628195 +Step 19939 | loss:0.8090223670005798 lr:4.245571221957402e-05 tokens_per_second_per_gpu:3720.0093563110295 mfu:17.547918832763404 +Step 19940 | loss:0.7291805744171143 lr:4.2451425670711486e-05 tokens_per_second_per_gpu:4328.502997073527 mfu:20.41828715596077 +Step 19941 | loss:0.8674285411834717 lr:4.2447139178622724e-05 tokens_per_second_per_gpu:4004.5977649704487 mfu:18.890370912199316 +Step 19942 | loss:0.6763466000556946 lr:4.244285274333998e-05 tokens_per_second_per_gpu:4926.812505822913 mfu:23.240615190860193 +Step 19943 | loss:0.6980845928192139 lr:4.243856636489553e-05 tokens_per_second_per_gpu:5098.015136045895 mfu:24.048207207802772 +Step 19944 | loss:0.8017441034317017 lr:4.2434280043321575e-05 tokens_per_second_per_gpu:4233.291207746626 mfu:19.96915690089941 +Step 19945 | loss:0.8245552778244019 lr:4.242999377865035e-05 tokens_per_second_per_gpu:3557.8730593719456 mfu:16.78309425679694 +Step 19946 | loss:0.6000089049339294 lr:4.242570757091413e-05 tokens_per_second_per_gpu:5033.098287023787 mfu:23.741983355008927 +Step 19947 | loss:0.8246463537216187 lr:4.242142142014513e-05 tokens_per_second_per_gpu:5325.796650018028 mfu:25.12269147274309 +Step 19948 | loss:0.7720051407814026 lr:4.241713532637558e-05 tokens_per_second_per_gpu:4028.210144942998 mfu:19.001754537217277 +Step 19949 | loss:0.7199845910072327 lr:4.2412849289637714e-05 tokens_per_second_per_gpu:4210.9039302374495 mfu:19.863552293225204 +Step 19950 | loss:0.755102813243866 lr:4.2408563309963784e-05 tokens_per_second_per_gpu:4948.77884066794 mfu:23.344234140167373 +Step 19951 | loss:0.7290023565292358 lr:4.240427738738601e-05 tokens_per_second_per_gpu:3925.6314716804977 mfu:18.51787343371169 +Step 19952 | loss:0.7195953130722046 lr:4.239999152193664e-05 tokens_per_second_per_gpu:4160.756672375013 mfu:19.626999121883635 +Step 19953 | loss:0.6344829201698303 lr:4.2395705713647916e-05 tokens_per_second_per_gpu:5393.098558977956 mfu:25.440166060196812 +Step 19954 | loss:0.6409976482391357 lr:4.239141996255206e-05 tokens_per_second_per_gpu:5327.5518730447975 mfu:25.13097115923229 +Step 19955 | loss:0.6978217363357544 lr:4.238713426868128e-05 tokens_per_second_per_gpu:5019.41099852903 mfu:23.677418079886934 +Step 19956 | loss:0.6272165775299072 lr:4.238284863206786e-05 tokens_per_second_per_gpu:5309.587658636998 mfu:25.046230894859665 +Step 19957 | loss:0.8753659725189209 lr:4.2378563052743996e-05 tokens_per_second_per_gpu:4784.7535003445255 mfu:22.570498624252984 +Step 19958 | loss:0.7605074644088745 lr:4.237427753074192e-05 tokens_per_second_per_gpu:4846.237270306551 mfu:22.860528057376733 +Step 19959 | loss:0.6254076957702637 lr:4.23699920660939e-05 tokens_per_second_per_gpu:3944.248346820331 mfu:18.605692410112717 +Step 19960 | loss:0.5858164429664612 lr:4.236570665883213e-05 tokens_per_second_per_gpu:4601.5065994568495 mfu:21.706091727620567 +Step 19961 | loss:0.7567452788352966 lr:4.2361421308988845e-05 tokens_per_second_per_gpu:5330.362389188875 mfu:25.1442288434074 +Step 19962 | loss:0.7763590216636658 lr:4.2357136016596316e-05 tokens_per_second_per_gpu:3844.908586084646 mfu:18.13708980961157 +Step 19963 | loss:0.7623133063316345 lr:4.235285078168672e-05 tokens_per_second_per_gpu:4337.9152348894795 mfu:20.462686287631985 +Step 19964 | loss:0.7893404960632324 lr:4.234856560429232e-05 tokens_per_second_per_gpu:3721.172712753195 mfu:17.553406583590174 +Step 19965 | loss:0.6393988728523254 lr:4.2344280484445314e-05 tokens_per_second_per_gpu:3797.4023011647732 mfu:17.912994558236523 +Step 19966 | loss:0.7462901473045349 lr:4.233999542217797e-05 tokens_per_second_per_gpu:5201.024405200934 mfu:24.534119505600703 +Step 19967 | loss:0.6925681829452515 lr:4.233571041752248e-05 tokens_per_second_per_gpu:5328.598573951554 mfu:25.135908625995047 +Step 19968 | loss:0.6432029008865356 lr:4.2331425470511096e-05 tokens_per_second_per_gpu:5347.432097618556 mfu:25.224749570462784 +Step 19969 | loss:0.7338714599609375 lr:4.2327140581176064e-05 tokens_per_second_per_gpu:4930.153777903561 mfu:23.25637654134441 +Step 19970 | loss:0.7663682699203491 lr:4.232285574954957e-05 tokens_per_second_per_gpu:5378.807558705491 mfu:25.372752973615523 +Step 19971 | loss:0.7486300468444824 lr:4.231857097566385e-05 tokens_per_second_per_gpu:5323.476253199782 mfu:25.111745765050898 +Step 19972 | loss:0.7998453378677368 lr:4.231428625955114e-05 tokens_per_second_per_gpu:5291.4482233522995 mfu:24.96066408371495 +Step 19973 | loss:0.7822303175926208 lr:4.2310001601243674e-05 tokens_per_second_per_gpu:4304.347495350915 mfu:20.304341532981635 +Step 19974 | loss:0.714682400226593 lr:4.230571700077365e-05 tokens_per_second_per_gpu:4994.4052123096235 mfu:23.559461519862925 +Step 19975 | loss:0.7427957653999329 lr:4.230143245817332e-05 tokens_per_second_per_gpu:5345.144474441532 mfu:25.213958461628 +Step 19976 | loss:0.5257539749145508 lr:4.2297147973474916e-05 tokens_per_second_per_gpu:3998.4050778881588 mfu:18.861158950650786 +Step 19977 | loss:0.7348470091819763 lr:4.229286354671063e-05 tokens_per_second_per_gpu:4220.970129169194 mfu:19.911036271056883 +Step 19978 | loss:0.607913076877594 lr:4.228857917791269e-05 tokens_per_second_per_gpu:4626.868075155477 mfu:21.825726135604942 +Step 19979 | loss:0.8286564350128174 lr:4.228429486711334e-05 tokens_per_second_per_gpu:5356.98664753956 mfu:25.269820012614474 +Step 19980 | loss:0.7477434277534485 lr:4.2280010614344786e-05 tokens_per_second_per_gpu:4749.811544744443 mfu:22.405671457975018 +Step 19981 | loss:0.7879764437675476 lr:4.227572641963926e-05 tokens_per_second_per_gpu:4643.029945439999 mfu:21.901964435236447 +Step 19982 | loss:0.5910371541976929 lr:4.227144228302898e-05 tokens_per_second_per_gpu:4996.714826269444 mfu:23.570356362972174 +Step 19983 | loss:0.7883027791976929 lr:4.226715820454619e-05 tokens_per_second_per_gpu:3981.4595430515883 mfu:18.78122397162031 +Step 19984 | loss:0.777660071849823 lr:4.226287418422308e-05 tokens_per_second_per_gpu:4012.2495245502423 mfu:18.92646556766201 +Step 19985 | loss:0.7525541186332703 lr:4.225859022209186e-05 tokens_per_second_per_gpu:4349.295822286924 mfu:20.516370460114164 +Step 19986 | loss:0.6984186172485352 lr:4.225430631818479e-05 tokens_per_second_per_gpu:4851.044977128104 mfu:22.883206830733364 +Step 19987 | loss:0.7412649989128113 lr:4.225002247253405e-05 tokens_per_second_per_gpu:3464.386992106904 mfu:16.342104527139913 +Step 19988 | loss:0.7446659207344055 lr:4.2245738685171895e-05 tokens_per_second_per_gpu:5097.965463347879 mfu:24.047972893211195 +Step 19989 | loss:0.7513558268547058 lr:4.224145495613053e-05 tokens_per_second_per_gpu:3109.0670220100747 mfu:14.665999604354937 +Step 19990 | loss:0.7769260406494141 lr:4.2237171285442165e-05 tokens_per_second_per_gpu:5210.616970143763 mfu:24.579369271096375 +Step 19991 | loss:0.6399537324905396 lr:4.2232887673139046e-05 tokens_per_second_per_gpu:4449.75429197869 mfu:20.990250201632573 +Step 19992 | loss:0.7900785803794861 lr:4.222860411925336e-05 tokens_per_second_per_gpu:4985.1030003184 mfu:23.5155813987793 +Step 19993 | loss:0.7202380299568176 lr:4.2224320623817326e-05 tokens_per_second_per_gpu:3912.5194791197805 mfu:18.456021927666168 +Step 19994 | loss:0.6021374464035034 lr:4.222003718686316e-05 tokens_per_second_per_gpu:4626.100107310459 mfu:21.822103500251167 +Step 19995 | loss:0.7373087406158447 lr:4.22157538084231e-05 tokens_per_second_per_gpu:4184.36984398324 mfu:19.738386481182157 +Step 19996 | loss:0.7552580833435059 lr:4.2211470488529354e-05 tokens_per_second_per_gpu:5335.531913712929 mfu:25.168614372599198 +Step 19997 | loss:0.676847517490387 lr:4.220718722721412e-05 tokens_per_second_per_gpu:5062.898768884423 mfu:23.88255731243205 +Step 19998 | loss:0.7888399362564087 lr:4.220290402450964e-05 tokens_per_second_per_gpu:3435.821066452101 mfu:16.207354181976715 +Step 19999 | loss:0.679995059967041 lr:4.2198620880448107e-05 tokens_per_second_per_gpu:3824.7799076047027 mfu:18.042139398914994 +Step 20000 | loss:0.7372646331787109 lr:4.219433779506174e-05 tokens_per_second_per_gpu:5343.74534428521 mfu:25.20735852596441 +Step 20001 | loss:0.6272469758987427 lr:4.219005476838275e-05 tokens_per_second_per_gpu:5348.531663513463 mfu:25.229936410394345 +Step 20002 | loss:0.7656605243682861 lr:4.2185771800443356e-05 tokens_per_second_per_gpu:3358.791883349749 mfu:15.843994382749067 +Step 20003 | loss:0.7078700065612793 lr:4.2181488891275765e-05 tokens_per_second_per_gpu:4890.0100872942385 mfu:23.06701190352035 +Step 20004 | loss:0.8060266971588135 lr:4.2177206040912196e-05 tokens_per_second_per_gpu:4906.774510318257 mfu:23.146092547230154 +Step 20005 | loss:0.7289488911628723 lr:4.2172923249384884e-05 tokens_per_second_per_gpu:4672.531305902396 mfu:22.0411273859887 +Step 20006 | loss:0.8191889524459839 lr:4.2168640516725996e-05 tokens_per_second_per_gpu:3886.0993390844214 mfu:18.331393619378666 +Step 20007 | loss:0.7987629771232605 lr:4.216435784296775e-05 tokens_per_second_per_gpu:5326.739586318872 mfu:25.127139464155693 +Step 20008 | loss:0.9086584448814392 lr:4.216007522814238e-05 tokens_per_second_per_gpu:4277.095321353383 mfu:20.175788378534932 +Step 20009 | loss:0.7414511442184448 lr:4.215579267228209e-05 tokens_per_second_per_gpu:4475.19933100477 mfu:21.110278792089918 +Step 20010 | loss:0.7659124732017517 lr:4.2151510175419065e-05 tokens_per_second_per_gpu:5345.866892151197 mfu:25.217366229221664 +Step 20011 | loss:0.623790979385376 lr:4.2147227737585546e-05 tokens_per_second_per_gpu:3925.2263881705267 mfu:18.515962585680782 +Step 20012 | loss:0.6968865990638733 lr:4.2142945358813743e-05 tokens_per_second_per_gpu:5329.188916668342 mfu:25.13869337331299 +Step 20013 | loss:0.736792802810669 lr:4.213866303913584e-05 tokens_per_second_per_gpu:5343.35506329094 mfu:25.205517503925634 +Step 20014 | loss:0.5993547439575195 lr:4.2134380778584045e-05 tokens_per_second_per_gpu:5310.650190563448 mfu:25.05124303924337 +Step 20015 | loss:0.7684754133224487 lr:4.213009857719058e-05 tokens_per_second_per_gpu:5328.711366580149 mfu:25.13644068806854 +Step 20016 | loss:0.6164354681968689 lr:4.212581643498765e-05 tokens_per_second_per_gpu:5330.874360131201 mfu:25.146643897693604 +Step 20017 | loss:0.6246978640556335 lr:4.212153435200746e-05 tokens_per_second_per_gpu:4226.382801691105 mfu:19.936568770839898 +Step 20018 | loss:0.772861659526825 lr:4.2117252328282214e-05 tokens_per_second_per_gpu:5047.066722876505 mfu:23.807874850187112 +Step 20019 | loss:0.9350106716156006 lr:4.2112970363844134e-05 tokens_per_second_per_gpu:3961.5549212406063 mfu:18.687330474459234 +Step 20020 | loss:0.7741801738739014 lr:4.2108688458725396e-05 tokens_per_second_per_gpu:4584.174957064758 mfu:21.624335413373952 +Step 20021 | loss:0.7261283993721008 lr:4.210440661295821e-05 tokens_per_second_per_gpu:5307.192900946987 mfu:25.034934414248088 +Step 20022 | loss:0.8130648732185364 lr:4.2100124826574804e-05 tokens_per_second_per_gpu:4540.819234525221 mfu:21.419819072905913 +Step 20023 | loss:0.5479406118392944 lr:4.2095843099607356e-05 tokens_per_second_per_gpu:3974.0135109525304 mfu:18.746099767784976 +Step 20024 | loss:0.7482883930206299 lr:4.209156143208809e-05 tokens_per_second_per_gpu:4855.338120285352 mfu:22.903458319491854 +Step 20025 | loss:0.7524515986442566 lr:4.20872798240492e-05 tokens_per_second_per_gpu:5334.169385120742 mfu:25.162187092758504 +Step 20026 | loss:0.6275328397750854 lr:4.2082998275522875e-05 tokens_per_second_per_gpu:5345.697916639768 mfu:25.216569142904053 +Step 20027 | loss:0.7454283237457275 lr:4.207871678654135e-05 tokens_per_second_per_gpu:4959.321443418581 mfu:23.39396539609566 +Step 20028 | loss:0.7764239311218262 lr:4.20744353571368e-05 tokens_per_second_per_gpu:5339.130853213946 mfu:25.185591184267892 +Step 20029 | loss:0.7736161947250366 lr:4.2070153987341435e-05 tokens_per_second_per_gpu:4820.943313842946 mfu:22.741212149143923 +Step 20030 | loss:0.8447633981704712 lr:4.2065872677187435e-05 tokens_per_second_per_gpu:4399.022421679182 mfu:20.75093931368017 +Step 20031 | loss:0.5164974331855774 lr:4.2061591426707036e-05 tokens_per_second_per_gpu:4052.762168809806 mfu:19.117570622804244 +Step 20032 | loss:0.6465063095092773 lr:4.205731023593242e-05 tokens_per_second_per_gpu:5349.109314506501 mfu:25.232661288685925 +Step 20033 | loss:0.7035796642303467 lr:4.205302910489577e-05 tokens_per_second_per_gpu:5344.833711274787 mfu:25.212492538748798 +Step 20034 | loss:0.8064347505569458 lr:4.2048748033629337e-05 tokens_per_second_per_gpu:4298.05353517315 mfu:20.274651848962776 +Step 20035 | loss:0.7332738637924194 lr:4.204446702216526e-05 tokens_per_second_per_gpu:5363.662572040596 mfu:25.301311487516035 +Step 20036 | loss:0.6239223480224609 lr:4.204018607053577e-05 tokens_per_second_per_gpu:4966.092899122252 mfu:23.42590750797935 +Step 20037 | loss:0.6260330080986023 lr:4.203590517877304e-05 tokens_per_second_per_gpu:4747.252956383533 mfu:22.393602160136417 +Step 20038 | loss:0.7520172595977783 lr:4.20316243469093e-05 tokens_per_second_per_gpu:5370.224738428917 mfu:25.332266345990778 +Step 20039 | loss:0.9861942529678345 lr:4.202734357497672e-05 tokens_per_second_per_gpu:5234.6372576579115 mfu:24.692677065584267 +Step 20040 | loss:0.6921062469482422 lr:4.202306286300751e-05 tokens_per_second_per_gpu:4028.1697997078654 mfu:19.00156422185955 +Step 20041 | loss:0.8001896142959595 lr:4.201878221103388e-05 tokens_per_second_per_gpu:4683.784168219855 mfu:22.094209057476444 +Step 20042 | loss:0.877799391746521 lr:4.2014501619088e-05 tokens_per_second_per_gpu:3113.8380658465985 mfu:14.688505432155837 +Step 20043 | loss:0.7485586404800415 lr:4.201022108720205e-05 tokens_per_second_per_gpu:5259.187059591045 mfu:24.808482669930985 +Step 20044 | loss:0.7377362847328186 lr:4.2005940615408264e-05 tokens_per_second_per_gpu:4364.4860979790465 mfu:20.58802558228219 +Step 20045 | loss:0.5953672528266907 lr:4.200166020373881e-05 tokens_per_second_per_gpu:4941.819556832365 mfu:23.31140601093871 +Step 20046 | loss:0.7187101244926453 lr:4.199737985222589e-05 tokens_per_second_per_gpu:5362.051148732497 mfu:25.29371012883456 +Step 20047 | loss:0.6425426006317139 lr:4.199309956090171e-05 tokens_per_second_per_gpu:3881.034744982538 mfu:18.307503064891222 +Step 20048 | loss:0.7035484910011292 lr:4.198881932979845e-05 tokens_per_second_per_gpu:4403.0670378866125 mfu:20.770018458413304 +Step 20049 | loss:0.7810078263282776 lr:4.19845391589483e-05 tokens_per_second_per_gpu:3631.3492671438858 mfu:17.129694065190932 +Step 20050 | loss:0.7221631407737732 lr:4.198025904838343e-05 tokens_per_second_per_gpu:4791.705115753817 mfu:22.603290580206505 +Step 20051 | loss:0.6180261373519897 lr:4.197597899813608e-05 tokens_per_second_per_gpu:5334.852703577444 mfu:25.16541042250492 +Step 20052 | loss:0.7299906611442566 lr:4.197169900823841e-05 tokens_per_second_per_gpu:4473.471116651598 mfu:21.102126510122233 +Step 20053 | loss:0.6882398128509521 lr:4.196741907872261e-05 tokens_per_second_per_gpu:5175.026684038942 mfu:24.411483819210968 +Step 20054 | loss:0.6793684959411621 lr:4.1963139209620886e-05 tokens_per_second_per_gpu:4570.705360858233 mfu:21.560797029917683 +Step 20055 | loss:0.8314752578735352 lr:4.19588594009654e-05 tokens_per_second_per_gpu:5333.75713386065 mfu:25.16024243322733 +Step 20056 | loss:0.7381336092948914 lr:4.195457965278839e-05 tokens_per_second_per_gpu:5117.703195446584 mfu:24.141079143125136 +Step 20057 | loss:0.6866516470909119 lr:4.195029996512199e-05 tokens_per_second_per_gpu:4064.234512579089 mfu:19.171687625747637 +Step 20058 | loss:0.7254113554954529 lr:4.1946020337998424e-05 tokens_per_second_per_gpu:5392.6527566153745 mfu:25.43806313439079 +Step 20059 | loss:0.660686731338501 lr:4.194174077144985e-05 tokens_per_second_per_gpu:5320.719334597127 mfu:25.098740909623363 +Step 20060 | loss:0.7296750545501709 lr:4.193746126550848e-05 tokens_per_second_per_gpu:4599.069872406267 mfu:21.694597270374462 +Step 20061 | loss:0.8588283061981201 lr:4.1933181820206494e-05 tokens_per_second_per_gpu:5319.330165079614 mfu:25.09218795999249 +Step 20062 | loss:0.6273255348205566 lr:4.1928902435576064e-05 tokens_per_second_per_gpu:3828.797958769085 mfu:18.061093231807607 +Step 20063 | loss:0.6931100487709045 lr:4.1924623111649416e-05 tokens_per_second_per_gpu:5349.974799052691 mfu:25.236743926957196 +Step 20064 | loss:0.6794835925102234 lr:4.192034384845869e-05 tokens_per_second_per_gpu:3281.6837525519168 mfu:15.480262173771974 +Step 20065 | loss:0.6297386884689331 lr:4.191606464603609e-05 tokens_per_second_per_gpu:4643.243293736501 mfu:21.90297083555202 +Step 20066 | loss:0.6654593348503113 lr:4.19117855044138e-05 tokens_per_second_per_gpu:5293.909973347655 mfu:24.9722765784609 +Step 20067 | loss:0.7303638458251953 lr:4.190750642362401e-05 tokens_per_second_per_gpu:4796.739143021003 mfu:22.62703694571944 +Step 20068 | loss:0.8407264351844788 lr:4.190322740369889e-05 tokens_per_second_per_gpu:3436.691879756231 mfu:16.211461956908366 +Step 20069 | loss:0.8326996564865112 lr:4.189894844467062e-05 tokens_per_second_per_gpu:4517.7228243202935 mfu:21.31086936531534 +Step 20070 | loss:0.6777841448783875 lr:4.1894669546571414e-05 tokens_per_second_per_gpu:4752.865328223474 mfu:22.420076675674796 +Step 20071 | loss:0.7720475792884827 lr:4.189039070943342e-05 tokens_per_second_per_gpu:4710.08832144628 mfu:22.218290236196818 +Step 20072 | loss:0.7070298194885254 lr:4.188611193328884e-05 tokens_per_second_per_gpu:4841.029965803615 mfu:22.835964313578618 +Step 20073 | loss:0.7414218187332153 lr:4.188183321816983e-05 tokens_per_second_per_gpu:5320.091552320378 mfu:25.09577955351323 +Step 20074 | loss:0.7563596963882446 lr:4.18775545641086e-05 tokens_per_second_per_gpu:4093.028199802047 mfu:19.30751236108795 +Step 20075 | loss:0.8448745012283325 lr:4.187327597113732e-05 tokens_per_second_per_gpu:5319.537577965255 mfu:25.093166361962346 +Step 20076 | loss:0.757083535194397 lr:4.186899743928816e-05 tokens_per_second_per_gpu:4690.523995759861 mfu:22.12600197391581 +Step 20077 | loss:0.6885489821434021 lr:4.186471896859333e-05 tokens_per_second_per_gpu:3286.3138212079666 mfu:15.502102997593733 +Step 20078 | loss:0.7920334339141846 lr:4.186044055908497e-05 tokens_per_second_per_gpu:5344.011808892089 mfu:25.208615484978605 +Step 20079 | loss:0.7191316485404968 lr:4.1856162210795266e-05 tokens_per_second_per_gpu:5334.086961147002 mfu:25.16179828481139 +Step 20080 | loss:0.6204520463943481 lr:4.185188392375642e-05 tokens_per_second_per_gpu:5316.031125654363 mfu:25.07662582814941 +Step 20081 | loss:0.6498346924781799 lr:4.184760569800059e-05 tokens_per_second_per_gpu:5364.804685615261 mfu:25.3066990321124 +Step 20082 | loss:0.7711234092712402 lr:4.184332753355995e-05 tokens_per_second_per_gpu:4161.269798081978 mfu:19.62941962339166 +Step 20083 | loss:0.8096892833709717 lr:4.1839049430466695e-05 tokens_per_second_per_gpu:4993.411214163763 mfu:23.55477266101533 +Step 20084 | loss:0.7642876505851746 lr:4.183477138875299e-05 tokens_per_second_per_gpu:5337.984639583671 mfu:25.180184299009678 +Step 20085 | loss:0.5878644585609436 lr:4.183049340845102e-05 tokens_per_second_per_gpu:5320.658789230135 mfu:25.098455306797153 +Step 20086 | loss:0.9717355370521545 lr:4.1826215489592935e-05 tokens_per_second_per_gpu:5359.158733768973 mfu:25.280066113954742 +Step 20087 | loss:0.6240982413291931 lr:4.182193763221093e-05 tokens_per_second_per_gpu:5327.991572010871 mfu:25.133045294275636 +Step 20088 | loss:0.8864206075668335 lr:4.181765983633718e-05 tokens_per_second_per_gpu:5143.4495968706 mfu:24.262529311431273 +Step 20089 | loss:0.8588243722915649 lr:4.181338210200385e-05 tokens_per_second_per_gpu:5354.221631879877 mfu:25.256776961987434 +Step 20090 | loss:0.8603896498680115 lr:4.180910442924312e-05 tokens_per_second_per_gpu:4621.6029618478815 mfu:21.800889697812053 +Step 20091 | loss:0.7953194379806519 lr:4.180482681808715e-05 tokens_per_second_per_gpu:3350.1709771066953 mfu:15.803328097122305 +Step 20092 | loss:0.7214964032173157 lr:4.1800549268568155e-05 tokens_per_second_per_gpu:4944.615996225384 mfu:23.32459729267734 +Step 20093 | loss:0.6152662038803101 lr:4.179627178071824e-05 tokens_per_second_per_gpu:3862.2139889453856 mfu:18.21872234751187 +Step 20094 | loss:0.7351201772689819 lr:4.179199435456964e-05 tokens_per_second_per_gpu:5344.191900849029 mfu:25.209465009466346 +Step 20095 | loss:0.7160642147064209 lr:4.178771699015447e-05 tokens_per_second_per_gpu:5332.059811881761 mfu:25.15223587585992 +Step 20096 | loss:0.7816236615180969 lr:4.178343968750494e-05 tokens_per_second_per_gpu:4145.952093911172 mfu:19.557163399348145 +Step 20097 | loss:0.8115909099578857 lr:4.177916244665322e-05 tokens_per_second_per_gpu:4797.275263027663 mfu:22.629565915262358 +Step 20098 | loss:0.6194361448287964 lr:4.177488526763145e-05 tokens_per_second_per_gpu:4341.524399044519 mfu:20.47971133073821 +Step 20099 | loss:0.8378958702087402 lr:4.1770608150471844e-05 tokens_per_second_per_gpu:4450.585496364305 mfu:20.99417113453761 +Step 20100 | loss:0.771466851234436 lr:4.1766331095206515e-05 tokens_per_second_per_gpu:3047.911700799699 mfu:14.377519520031859 +Step 20101 | loss:0.7519184350967407 lr:4.176205410186768e-05 tokens_per_second_per_gpu:4296.829793560711 mfu:20.26887925098505 +Step 20102 | loss:0.48122721910476685 lr:4.1757777170487463e-05 tokens_per_second_per_gpu:5316.976393551477 mfu:25.081084818099434 +Step 20103 | loss:0.8751018047332764 lr:4.175350030109808e-05 tokens_per_second_per_gpu:5311.099248156128 mfu:25.05336132052517 +Step 20104 | loss:0.7164547443389893 lr:4.174922349373166e-05 tokens_per_second_per_gpu:4447.28199561846 mfu:20.978587957884113 +Step 20105 | loss:0.7360367774963379 lr:4.174494674842038e-05 tokens_per_second_per_gpu:5341.305589814326 mfu:25.19584978786021 +Step 20106 | loss:0.7880522012710571 lr:4.1740670065196434e-05 tokens_per_second_per_gpu:4131.332182235049 mfu:19.488198781557728 +Step 20107 | loss:0.6370495557785034 lr:4.173639344409195e-05 tokens_per_second_per_gpu:4546.728964973165 mfu:21.44769627973285 +Step 20108 | loss:0.628048300743103 lr:4.17321168851391e-05 tokens_per_second_per_gpu:3651.5773260501187 mfu:17.225113270314363 +Step 20109 | loss:0.653657853603363 lr:4.172784038837004e-05 tokens_per_second_per_gpu:5158.295259798027 mfu:24.33255883639123 +Step 20110 | loss:0.8028451204299927 lr:4.172356395381697e-05 tokens_per_second_per_gpu:4653.572106020873 mfu:21.951693605374505 +Step 20111 | loss:0.7935591340065002 lr:4.1719287581512015e-05 tokens_per_second_per_gpu:5346.747347113243 mfu:25.221519485498742 +Step 20112 | loss:0.6551761031150818 lr:4.171501127148736e-05 tokens_per_second_per_gpu:3663.1315220753686 mfu:17.27961638431459 +Step 20113 | loss:0.8626749515533447 lr:4.171073502377518e-05 tokens_per_second_per_gpu:3293.459506522452 mfu:15.535810414401794 +Step 20114 | loss:0.631357729434967 lr:4.1706458838407614e-05 tokens_per_second_per_gpu:4490.983350695047 mfu:21.184734706000413 +Step 20115 | loss:0.730954647064209 lr:4.170218271541682e-05 tokens_per_second_per_gpu:5053.653890159385 mfu:23.838947642147087 +Step 20116 | loss:0.6607010364532471 lr:4.169790665483496e-05 tokens_per_second_per_gpu:4729.025345211773 mfu:22.30761940828807 +Step 20117 | loss:0.6622979044914246 lr:4.1693630656694214e-05 tokens_per_second_per_gpu:4510.063329139147 mfu:21.274738219701238 +Step 20118 | loss:0.827180027961731 lr:4.168935472102672e-05 tokens_per_second_per_gpu:5083.620968035743 mfu:23.980307461400578 +Step 20119 | loss:0.7955223321914673 lr:4.1685078847864664e-05 tokens_per_second_per_gpu:5322.63546769068 mfu:25.10777963635167 +Step 20120 | loss:0.7280798554420471 lr:4.168080303724019e-05 tokens_per_second_per_gpu:3961.19241478405 mfu:18.685620469653678 +Step 20121 | loss:0.7897077202796936 lr:4.167652728918547e-05 tokens_per_second_per_gpu:3028.8841984027836 mfu:14.287763544798938 +Step 20122 | loss:0.7596442699432373 lr:4.167225160373262e-05 tokens_per_second_per_gpu:5277.770255968858 mfu:24.896142777865375 +Step 20123 | loss:0.6536878943443298 lr:4.1667975980913857e-05 tokens_per_second_per_gpu:4051.454952414751 mfu:19.111404260034654 +Step 20124 | loss:0.6428517699241638 lr:4.16637004207613e-05 tokens_per_second_per_gpu:5341.9606637614725 mfu:25.19893988343604 +Step 20125 | loss:0.8462791442871094 lr:4.16594249233071e-05 tokens_per_second_per_gpu:5316.64049391271 mfu:25.079500322192978 +Step 20126 | loss:0.6528245806694031 lr:4.1655149488583445e-05 tokens_per_second_per_gpu:4882.751223330756 mfu:23.032770603714702 +Step 20127 | loss:0.8876916766166687 lr:4.1650874116622464e-05 tokens_per_second_per_gpu:4777.502353853328 mfu:22.536293729080842 +Step 20128 | loss:0.7555795907974243 lr:4.164659880745635e-05 tokens_per_second_per_gpu:3923.341667002805 mfu:18.507072034366047 +Step 20129 | loss:0.7270841002464294 lr:4.1642323561117215e-05 tokens_per_second_per_gpu:5326.179579310051 mfu:25.124497815548036 +Step 20130 | loss:0.8196021318435669 lr:4.163804837763723e-05 tokens_per_second_per_gpu:4175.931646913327 mfu:19.698582065898503 +Step 20131 | loss:0.6421365737915039 lr:4.163377325704854e-05 tokens_per_second_per_gpu:3864.04251051698 mfu:18.227347795743032 +Step 20132 | loss:0.7524504661560059 lr:4.162949819938332e-05 tokens_per_second_per_gpu:5161.708476501227 mfu:24.348659561935044 +Step 20133 | loss:0.8479682803153992 lr:4.162522320467371e-05 tokens_per_second_per_gpu:5172.963993282035 mfu:24.4017537549793 +Step 20134 | loss:0.69217848777771 lr:4.1620948272951866e-05 tokens_per_second_per_gpu:5307.111976169125 mfu:25.034552678263726 +Step 20135 | loss:0.5984289050102234 lr:4.161667340424996e-05 tokens_per_second_per_gpu:3263.4879931921473 mfu:15.394429672355267 +Step 20136 | loss:0.7973319888114929 lr:4.16123985986001e-05 tokens_per_second_per_gpu:4553.520059138783 mfu:21.479731029592216 +Step 20137 | loss:0.691314697265625 lr:4.160812385603446e-05 tokens_per_second_per_gpu:5314.052017575942 mfu:25.067290037672844 +Step 20138 | loss:0.7881203889846802 lr:4.1603849176585184e-05 tokens_per_second_per_gpu:4951.588295588292 mfu:23.35748681836911 +Step 20139 | loss:0.6837831735610962 lr:4.159957456028445e-05 tokens_per_second_per_gpu:5324.378390690534 mfu:25.116001301515304 +Step 20140 | loss:0.7606580853462219 lr:4.159530000716438e-05 tokens_per_second_per_gpu:3865.833232256638 mfu:18.235794935717635 +Step 20141 | loss:0.7078492641448975 lr:4.159102551725712e-05 tokens_per_second_per_gpu:5359.229879114365 mfu:25.2804017186876 +Step 20142 | loss:0.6429225206375122 lr:4.158675109059485e-05 tokens_per_second_per_gpu:4310.264666939016 mfu:20.332253840936282 +Step 20143 | loss:0.7306628823280334 lr:4.158247672720969e-05 tokens_per_second_per_gpu:4036.6786526089495 mfu:19.041701932753973 +Step 20144 | loss:0.7905290126800537 lr:4.157820242713379e-05 tokens_per_second_per_gpu:5373.553760478779 mfu:25.34796991098971 +Step 20145 | loss:0.7676562070846558 lr:4.157392819039931e-05 tokens_per_second_per_gpu:4576.742496171338 mfu:21.5892752272311 +Step 20146 | loss:0.6501022577285767 lr:4.156965401703839e-05 tokens_per_second_per_gpu:4736.486549273368 mfu:22.342815180859297 +Step 20147 | loss:0.6612159013748169 lr:4.156537990708317e-05 tokens_per_second_per_gpu:3357.0356485108114 mfu:15.835709923369235 +Step 20148 | loss:0.6356359124183655 lr:4.156110586056581e-05 tokens_per_second_per_gpu:5341.791784249824 mfu:25.198143249965778 +Step 20149 | loss:0.8788785934448242 lr:4.1556831877518456e-05 tokens_per_second_per_gpu:4339.823201425779 mfu:20.47168649131637 +Step 20150 | loss:0.6308602094650269 lr:4.155255795797325e-05 tokens_per_second_per_gpu:3725.1767346536585 mfu:17.572294238051242 +Step 20151 | loss:0.7701796293258667 lr:4.154828410196233e-05 tokens_per_second_per_gpu:5325.131463751069 mfu:25.11955367562936 +Step 20152 | loss:0.7397733926773071 lr:4.1544010309517834e-05 tokens_per_second_per_gpu:4778.589004083368 mfu:22.54141964361759 +Step 20153 | loss:0.7343605160713196 lr:4.153973658067192e-05 tokens_per_second_per_gpu:5336.261691563235 mfu:25.172056859232036 +Step 20154 | loss:0.6961770057678223 lr:4.153546291545672e-05 tokens_per_second_per_gpu:3856.5987257816305 mfu:18.19223419310575 +Step 20155 | loss:0.9211323261260986 lr:4.1531189313904396e-05 tokens_per_second_per_gpu:3852.283212745839 mfu:18.171877181813475 +Step 20156 | loss:0.7392210364341736 lr:4.152691577604707e-05 tokens_per_second_per_gpu:2787.5670438568313 mfu:13.149429353853497 +Step 20157 | loss:0.7083926200866699 lr:4.152264230191691e-05 tokens_per_second_per_gpu:4256.365888385037 mfu:20.078004106417275 +Step 20158 | loss:0.7566105127334595 lr:4.151836889154601e-05 tokens_per_second_per_gpu:3943.8525225835324 mfu:18.60382523965316 +Step 20159 | loss:0.6632529497146606 lr:4.151409554496656e-05 tokens_per_second_per_gpu:4007.807431404393 mfu:18.905511456393796 +Step 20160 | loss:0.7192116379737854 lr:4.150982226221067e-05 tokens_per_second_per_gpu:5004.219618867828 mfu:23.605757750108406 +Step 20161 | loss:0.8743177652359009 lr:4.150554904331049e-05 tokens_per_second_per_gpu:4087.575623940548 mfu:19.28179163044389 +Step 20162 | loss:0.5459503531455994 lr:4.150127588829816e-05 tokens_per_second_per_gpu:5379.933299733306 mfu:25.378063286114948 +Step 20163 | loss:0.726955235004425 lr:4.149700279720582e-05 tokens_per_second_per_gpu:3722.0567832280926 mfu:17.557576894858244 +Step 20164 | loss:0.6660932898521423 lr:4.149272977006563e-05 tokens_per_second_per_gpu:3920.66641558921 mfu:18.49445241700316 +Step 20165 | loss:0.6242649555206299 lr:4.1488456806909675e-05 tokens_per_second_per_gpu:4775.694732465594 mfu:22.527766870582887 +Step 20166 | loss:0.7653554081916809 lr:4.1484183907770134e-05 tokens_per_second_per_gpu:4345.485720707025 mfu:20.498397560891423 +Step 20167 | loss:0.7816147208213806 lr:4.147991107267913e-05 tokens_per_second_per_gpu:5364.436017982809 mfu:25.30495996398894 +Step 20168 | loss:0.6736072897911072 lr:4.1475638301668796e-05 tokens_per_second_per_gpu:5052.195825559367 mfu:23.832069702657325 +Step 20169 | loss:0.6970577836036682 lr:4.147136559477128e-05 tokens_per_second_per_gpu:3934.5161558671575 mfu:18.559784004903594 +Step 20170 | loss:0.6604811549186707 lr:4.14670929520187e-05 tokens_per_second_per_gpu:4945.127968277975 mfu:23.32701235220085 +Step 20171 | loss:0.7683975100517273 lr:4.146282037344324e-05 tokens_per_second_per_gpu:3904.8444048012457 mfu:18.41981729260293 +Step 20172 | loss:0.8587453961372375 lr:4.145854785907696e-05 tokens_per_second_per_gpu:3992.1773953204856 mfu:18.83178190942686 +Step 20173 | loss:0.5942767262458801 lr:4.1454275408952046e-05 tokens_per_second_per_gpu:4152.207724050623 mfu:19.5866722740365 +Step 20174 | loss:0.6666846871376038 lr:4.14500030231006e-05 tokens_per_second_per_gpu:5288.83102443966 mfu:24.948318309903847 +Step 20175 | loss:0.5795871019363403 lr:4.144573070155479e-05 tokens_per_second_per_gpu:4697.320170036279 mfu:22.158060687524575 +Step 20176 | loss:0.6237483620643616 lr:4.144145844434673e-05 tokens_per_second_per_gpu:4947.417909264993 mfu:23.337814394540832 +Step 20177 | loss:0.7511646151542664 lr:4.143718625150854e-05 tokens_per_second_per_gpu:5331.997146438172 mfu:25.151940272270686 +Step 20178 | loss:0.7588670253753662 lr:4.143291412307238e-05 tokens_per_second_per_gpu:4315.926934610582 mfu:20.358963723625546 +Step 20179 | loss:0.6907092332839966 lr:4.142864205907038e-05 tokens_per_second_per_gpu:5345.764830775884 mfu:25.216884788300842 +Step 20180 | loss:0.6011685132980347 lr:4.142437005953463e-05 tokens_per_second_per_gpu:4304.354859282967 mfu:20.304376269905536 +Step 20181 | loss:0.77998286485672 lr:4.142009812449728e-05 tokens_per_second_per_gpu:4567.448828314383 mfu:21.545435410286597 +Step 20182 | loss:0.5734992027282715 lr:4.1415826253990495e-05 tokens_per_second_per_gpu:3603.4616659546114 mfu:16.998143492265722 +Step 20183 | loss:0.822779655456543 lr:4.141155444804636e-05 tokens_per_second_per_gpu:5326.331356355803 mfu:25.125213773768685 +Step 20184 | loss:0.7477941513061523 lr:4.140728270669701e-05 tokens_per_second_per_gpu:5162.845221000177 mfu:24.354021779685205 +Step 20185 | loss:0.585479736328125 lr:4.1403011029974605e-05 tokens_per_second_per_gpu:4925.538167067957 mfu:23.234603917528453 +Step 20186 | loss:0.946155309677124 lr:4.139873941791125e-05 tokens_per_second_per_gpu:5367.817637669385 mfu:25.32091164101433 +Step 20187 | loss:0.7275896668434143 lr:4.1394467870539075e-05 tokens_per_second_per_gpu:5324.792889022787 mfu:25.11795656086925 +Step 20188 | loss:0.6896244287490845 lr:4.1390196387890184e-05 tokens_per_second_per_gpu:3784.4992940548773 mfu:17.85212887221912 +Step 20189 | loss:0.7558351755142212 lr:4.138592496999674e-05 tokens_per_second_per_gpu:4331.904981179023 mfu:20.434334895424737 +Step 20190 | loss:0.7364152669906616 lr:4.138165361689084e-05 tokens_per_second_per_gpu:4464.890674815848 mfu:21.06165110200866 +Step 20191 | loss:0.8021902441978455 lr:4.137738232860463e-05 tokens_per_second_per_gpu:3666.401582370729 mfu:17.295041816657623 +Step 20192 | loss:0.8387469053268433 lr:4.137311110517023e-05 tokens_per_second_per_gpu:4144.352516157508 mfu:19.549617918168007 +Step 20193 | loss:0.6989243030548096 lr:4.136883994661977e-05 tokens_per_second_per_gpu:3794.8090071798574 mfu:17.90076154804809 +Step 20194 | loss:0.7059661149978638 lr:4.136456885298536e-05 tokens_per_second_per_gpu:4409.918085338366 mfu:20.802336018152452 +Step 20195 | loss:0.8203367590904236 lr:4.136029782429913e-05 tokens_per_second_per_gpu:4788.279493810116 mfu:22.587131337026694 +Step 20196 | loss:0.7871256470680237 lr:4.13560268605932e-05 tokens_per_second_per_gpu:5348.999873941655 mfu:25.232145038869064 +Step 20197 | loss:0.8469884395599365 lr:4.1351755961899685e-05 tokens_per_second_per_gpu:4402.145695262988 mfu:20.765672328060404 +Step 20198 | loss:0.7885393500328064 lr:4.134748512825073e-05 tokens_per_second_per_gpu:5353.759185343417 mfu:25.254595522773432 +Step 20199 | loss:0.7385594844818115 lr:4.134321435967843e-05 tokens_per_second_per_gpu:4820.661886017399 mfu:22.73988460607428 +Step 20200 | loss:0.6872929930686951 lr:4.133894365621496e-05 tokens_per_second_per_gpu:4388.060603891718 mfu:20.699230548897727 +Step 20201 | loss:0.7125242948532104 lr:4.133467301789235e-05 tokens_per_second_per_gpu:5362.907724848226 mfu:25.297750744519412 +Step 20202 | loss:0.6039853692054749 lr:4.13304024447428e-05 tokens_per_second_per_gpu:3752.600229378706 mfu:17.70165554159997 +Step 20203 | loss:0.739568829536438 lr:4.132613193679839e-05 tokens_per_second_per_gpu:5268.851329965759 mfu:24.854070682183067 +Step 20204 | loss:0.6999196410179138 lr:4.132186149409124e-05 tokens_per_second_per_gpu:3733.6867663653684 mfu:17.61243750959651 +Step 20205 | loss:0.7762099504470825 lr:4.131759111665349e-05 tokens_per_second_per_gpu:4490.695105780663 mfu:21.183375005559608 +Step 20206 | loss:0.7132231593132019 lr:4.131332080451723e-05 tokens_per_second_per_gpu:5355.866561706796 mfu:25.264536376634975 +Step 20207 | loss:0.7884496450424194 lr:4.1309050557714604e-05 tokens_per_second_per_gpu:4085.6401202609477 mfu:19.272661529356235 +Step 20208 | loss:0.6479330062866211 lr:4.130478037627773e-05 tokens_per_second_per_gpu:5334.740739548944 mfu:25.16488226907945 +Step 20209 | loss:0.7171890735626221 lr:4.13005102602387e-05 tokens_per_second_per_gpu:5048.230985751918 mfu:23.813366876814165 +Step 20210 | loss:0.5858826041221619 lr:4.129624020962963e-05 tokens_per_second_per_gpu:5319.139705162092 mfu:25.091289528065722 +Step 20211 | loss:0.6953814625740051 lr:4.1291970224482666e-05 tokens_per_second_per_gpu:5321.24323250697 mfu:25.10121222545029 +Step 20212 | loss:0.737933337688446 lr:4.128770030482989e-05 tokens_per_second_per_gpu:4571.080639640578 mfu:21.562567283963165 +Step 20213 | loss:0.7661613821983337 lr:4.128343045070343e-05 tokens_per_second_per_gpu:4340.478493869766 mfu:20.474777617579075 +Step 20214 | loss:0.7103446125984192 lr:4.127916066213542e-05 tokens_per_second_per_gpu:5345.61264265151 mfu:25.2161668909522 +Step 20215 | loss:0.8198274374008179 lr:4.1274890939157964e-05 tokens_per_second_per_gpu:4125.460746821696 mfu:19.46050221894276 +Step 20216 | loss:0.77076655626297 lr:4.127062128180315e-05 tokens_per_second_per_gpu:5363.7771638916265 mfu:25.30185203682873 +Step 20217 | loss:0.7697160243988037 lr:4.12663516901031e-05 tokens_per_second_per_gpu:4449.389144671904 mfu:20.98852773948591 +Step 20218 | loss:0.6290246248245239 lr:4.1262082164089936e-05 tokens_per_second_per_gpu:5340.181096197919 mfu:25.190545359613225 +Step 20219 | loss:0.6782249212265015 lr:4.1257812703795776e-05 tokens_per_second_per_gpu:5293.489681662724 mfu:24.970293990873927 +Step 20220 | loss:0.6391843557357788 lr:4.125354330925271e-05 tokens_per_second_per_gpu:4511.0774212413335 mfu:21.279521865169205 +Step 20221 | loss:0.7456178665161133 lr:4.124927398049287e-05 tokens_per_second_per_gpu:3800.681171227365 mfu:17.928461547754686 +Step 20222 | loss:0.7874037027359009 lr:4.124500471754837e-05 tokens_per_second_per_gpu:5251.381264354598 mfu:24.771661401994667 +Step 20223 | loss:0.7136387825012207 lr:4.124073552045129e-05 tokens_per_second_per_gpu:5219.845453544685 mfu:24.62290160184016 +Step 20224 | loss:0.8033588528633118 lr:4.1236466389233755e-05 tokens_per_second_per_gpu:3689.896442896113 mfu:17.405871082392423 +Step 20225 | loss:0.7108917832374573 lr:4.1232197323927876e-05 tokens_per_second_per_gpu:4646.229461236971 mfu:21.91705709714445 +Step 20226 | loss:0.6782630085945129 lr:4.122792832456575e-05 tokens_per_second_per_gpu:5344.001341483252 mfu:25.2085661084252 +Step 20227 | loss:0.7154491543769836 lr:4.122365939117951e-05 tokens_per_second_per_gpu:3642.925474100977 mfu:17.184301008512104 +Step 20228 | loss:0.7122502326965332 lr:4.1219390523801244e-05 tokens_per_second_per_gpu:5365.031367093604 mfu:25.30776832732159 +Step 20229 | loss:0.80558842420578 lr:4.121512172246307e-05 tokens_per_second_per_gpu:3302.1987882570756 mfu:15.577035097419232 +Step 20230 | loss:0.7343881726264954 lr:4.121085298719708e-05 tokens_per_second_per_gpu:5342.012688731059 mfu:25.199185294093958 +Step 20231 | loss:0.7124208211898804 lr:4.120658431803538e-05 tokens_per_second_per_gpu:3915.5052660048186 mfu:18.470106393830665 +Step 20232 | loss:0.7380913496017456 lr:4.120231571501009e-05 tokens_per_second_per_gpu:4893.831165742052 mfu:23.085036582501893 +Step 20233 | loss:0.6710834503173828 lr:4.1198047178153285e-05 tokens_per_second_per_gpu:3490.919306636391 mfu:16.467261981655156 +Step 20234 | loss:0.7925316095352173 lr:4.119377870749711e-05 tokens_per_second_per_gpu:5339.424979499087 mfu:25.186978628138025 +Step 20235 | loss:0.7413088083267212 lr:4.118951030307365e-05 tokens_per_second_per_gpu:3692.6564127716606 mfu:17.4188903311948 +Step 20236 | loss:0.7156250476837158 lr:4.1185241964915016e-05 tokens_per_second_per_gpu:3916.585745288811 mfu:18.47520319896203 +Step 20237 | loss:0.6656649112701416 lr:4.1180973693053285e-05 tokens_per_second_per_gpu:4611.433323264156 mfu:21.752917777493355 +Step 20238 | loss:0.6099410057067871 lr:4.1176705487520576e-05 tokens_per_second_per_gpu:5008.661467465783 mfu:23.62671070780248 +Step 20239 | loss:0.665017306804657 lr:4.1172437348348995e-05 tokens_per_second_per_gpu:4035.2852543856284 mfu:19.035129035596317 +Step 20240 | loss:0.7055217623710632 lr:4.1168169275570635e-05 tokens_per_second_per_gpu:4031.58518045868 mfu:19.017675155585188 +Step 20241 | loss:0.7155022025108337 lr:4.116390126921761e-05 tokens_per_second_per_gpu:3674.5479924093 mfu:17.33346982273044 +Step 20242 | loss:0.7759149074554443 lr:4.115963332932199e-05 tokens_per_second_per_gpu:5022.18445257155 mfu:23.690500935805556 +Step 20243 | loss:0.691815972328186 lr:4.115536545591592e-05 tokens_per_second_per_gpu:4931.95952026592 mfu:23.264894536159073 +Step 20244 | loss:0.8591108322143555 lr:4.115109764903148e-05 tokens_per_second_per_gpu:4830.101462684859 mfu:22.784412699772542 +Step 20245 | loss:0.7446092367172241 lr:4.1146829908700744e-05 tokens_per_second_per_gpu:5355.364228300912 mfu:25.262166784252507 +Step 20246 | loss:0.8707539439201355 lr:4.114256223495583e-05 tokens_per_second_per_gpu:3031.6609991497726 mfu:14.300862187693546 +Step 20247 | loss:0.8566112518310547 lr:4.1138294627828846e-05 tokens_per_second_per_gpu:3530.188173478395 mfu:16.652499926508455 +Step 20248 | loss:0.8025036454200745 lr:4.1134027087351875e-05 tokens_per_second_per_gpu:4899.7628523887515 mfu:23.11301736046448 +Step 20249 | loss:0.7240798473358154 lr:4.112975961355701e-05 tokens_per_second_per_gpu:4469.213538330688 mfu:21.08204278676473 +Step 20250 | loss:0.8363876938819885 lr:4.1125492206476366e-05 tokens_per_second_per_gpu:3331.288614679595 mfu:15.714256771890437 +Step 20251 | loss:0.7880522012710571 lr:4.112122486614204e-05 tokens_per_second_per_gpu:4628.08483648983 mfu:21.83146580641998 +Step 20252 | loss:0.5769526362419128 lr:4.1116957592586095e-05 tokens_per_second_per_gpu:4121.102676205792 mfu:19.43994445628402 +Step 20253 | loss:0.7380311489105225 lr:4.111269038584064e-05 tokens_per_second_per_gpu:4304.812721228618 mfu:20.30653608282248 +Step 20254 | loss:0.7602011561393738 lr:4.110842324593779e-05 tokens_per_second_per_gpu:4399.4185729041665 mfu:20.752808026598785 +Step 20255 | loss:0.7075515389442444 lr:4.110415617290961e-05 tokens_per_second_per_gpu:3133.197745964617 mfu:14.779828346374565 +Step 20256 | loss:0.5599368214607239 lr:4.1099889166788207e-05 tokens_per_second_per_gpu:5175.801506724168 mfu:24.415138790015558 +Step 20257 | loss:0.617089569568634 lr:4.109562222760568e-05 tokens_per_second_per_gpu:4367.6323928469965 mfu:20.602867191987567 +Step 20258 | loss:0.7171539664268494 lr:4.1091355355394135e-05 tokens_per_second_per_gpu:5102.821498112777 mfu:24.070879637722182 +Step 20259 | loss:0.6699848771095276 lr:4.108708855018562e-05 tokens_per_second_per_gpu:4439.98351509112 mfu:20.944159779987466 +Step 20260 | loss:0.7100106477737427 lr:4.1082821812012245e-05 tokens_per_second_per_gpu:4645.155965119873 mfu:21.911993232802775 +Step 20261 | loss:0.9076551795005798 lr:4.107855514090611e-05 tokens_per_second_per_gpu:4207.193866285266 mfu:19.846051288560574 +Step 20262 | loss:0.7783086895942688 lr:4.1074288536899296e-05 tokens_per_second_per_gpu:4253.383649533367 mfu:20.06393637693153 +Step 20263 | loss:0.7101647257804871 lr:4.1070022000023905e-05 tokens_per_second_per_gpu:4513.371352207942 mfu:21.29034272892368 +Step 20264 | loss:0.6921153664588928 lr:4.106575553031202e-05 tokens_per_second_per_gpu:5062.627684926354 mfu:23.881278563150147 +Step 20265 | loss:0.5922216773033142 lr:4.106148912779573e-05 tokens_per_second_per_gpu:3711.6269678761732 mfu:17.508377676333247 +Step 20266 | loss:0.7880244851112366 lr:4.105722279250711e-05 tokens_per_second_per_gpu:4866.911999338769 mfu:22.95805428169422 +Step 20267 | loss:0.713762640953064 lr:4.105295652447826e-05 tokens_per_second_per_gpu:5114.799840802852 mfu:24.127383523907273 +Step 20268 | loss:0.7281956076622009 lr:4.1048690323741276e-05 tokens_per_second_per_gpu:4604.552319027162 mfu:21.720458906487075 +Step 20269 | loss:0.8173320293426514 lr:4.104442419032821e-05 tokens_per_second_per_gpu:5212.822149365301 mfu:24.589771477726742 +Step 20270 | loss:0.7193471789360046 lr:4.104015812427119e-05 tokens_per_second_per_gpu:4335.472226704894 mfu:20.451162201205115 +Step 20271 | loss:0.6951441168785095 lr:4.103589212560229e-05 tokens_per_second_per_gpu:5382.262133372809 mfu:25.389048791732478 +Step 20272 | loss:0.8547011017799377 lr:4.103162619435357e-05 tokens_per_second_per_gpu:3339.3312864905024 mfu:15.752195427043796 +Step 20273 | loss:0.6577941179275513 lr:4.1027360330557166e-05 tokens_per_second_per_gpu:4357.717553310036 mfu:20.556097202245667 +Step 20274 | loss:0.7027184963226318 lr:4.102309453424511e-05 tokens_per_second_per_gpu:4519.79817352686 mfu:21.320659141613802 +Step 20275 | loss:0.5549413561820984 lr:4.10188288054495e-05 tokens_per_second_per_gpu:3515.938511100126 mfu:16.585281837827054 +Step 20276 | loss:0.6337770819664001 lr:4.101456314420243e-05 tokens_per_second_per_gpu:3742.0177047394027 mfu:17.651735967311485 +Step 20277 | loss:0.6911271810531616 lr:4.101029755053598e-05 tokens_per_second_per_gpu:2889.0654813609463 mfu:13.628214800979562 +Step 20278 | loss:0.6461753249168396 lr:4.100603202448221e-05 tokens_per_second_per_gpu:4903.25629140545 mfu:23.1294965083484 +Step 20279 | loss:0.7548713684082031 lr:4.100176656607324e-05 tokens_per_second_per_gpu:4289.039583750685 mfu:20.23213150216431 +Step 20280 | loss:0.7556905746459961 lr:4.0997501175341144e-05 tokens_per_second_per_gpu:5211.292292754046 mfu:24.582554883070998 +Step 20281 | loss:0.6884937882423401 lr:4.0993235852317974e-05 tokens_per_second_per_gpu:3879.1406909104126 mfu:18.298568488674057 +Step 20282 | loss:0.8020565509796143 lr:4.098897059703582e-05 tokens_per_second_per_gpu:5214.835064299827 mfu:24.599266740911602 +Step 20283 | loss:0.6408382654190063 lr:4.098470540952678e-05 tokens_per_second_per_gpu:5338.7513136376565 mfu:25.183800831330405 +Step 20284 | loss:0.6253947615623474 lr:4.098044028982292e-05 tokens_per_second_per_gpu:5344.67296858295 mfu:25.21173428804587 +Step 20285 | loss:0.792833149433136 lr:4.09761752379563e-05 tokens_per_second_per_gpu:5329.994044287283 mfu:25.142491297660474 +Step 20286 | loss:0.668532133102417 lr:4.097191025395904e-05 tokens_per_second_per_gpu:4165.0302368135835 mfu:19.647158254485653 +Step 20287 | loss:0.7223806381225586 lr:4.09676453378632e-05 tokens_per_second_per_gpu:3207.126434267082 mfu:15.128562582632185 +Step 20288 | loss:0.6907174587249756 lr:4.096338048970084e-05 tokens_per_second_per_gpu:4508.760671461018 mfu:21.268593361177185 +Step 20289 | loss:0.7971470355987549 lr:4.095911570950404e-05 tokens_per_second_per_gpu:3947.2149752093496 mfu:18.619686502382713 +Step 20290 | loss:0.7024861574172974 lr:4.095485099730489e-05 tokens_per_second_per_gpu:4522.587702013076 mfu:21.333817823425917 +Step 20291 | loss:0.7261380553245544 lr:4.095058635313547e-05 tokens_per_second_per_gpu:5339.452660686318 mfu:25.18710920502109 +Step 20292 | loss:0.7857897877693176 lr:4.094632177702783e-05 tokens_per_second_per_gpu:5280.20822788938 mfu:24.90764310737538 +Step 20293 | loss:0.7015093564987183 lr:4.0942057269014064e-05 tokens_per_second_per_gpu:4257.295104492361 mfu:20.082387377336136 +Step 20294 | loss:0.791958212852478 lr:4.093779282912626e-05 tokens_per_second_per_gpu:4445.631305388536 mfu:20.970801370432763 +Step 20295 | loss:0.7281737923622131 lr:4.0933528457396464e-05 tokens_per_second_per_gpu:4546.228192376272 mfu:21.44533404995269 +Step 20296 | loss:0.7068281769752502 lr:4.092926415385674e-05 tokens_per_second_per_gpu:3357.5953752295945 mfu:15.838350249800962 +Step 20297 | loss:0.7181510925292969 lr:4.092499991853919e-05 tokens_per_second_per_gpu:5366.281062874518 mfu:25.31366335554025 +Step 20298 | loss:0.7217473387718201 lr:4.092073575147586e-05 tokens_per_second_per_gpu:4996.805278869962 mfu:23.570783043321626 +Step 20299 | loss:0.7190045714378357 lr:4.091647165269885e-05 tokens_per_second_per_gpu:4892.839331883594 mfu:23.080357932967242 +Step 20300 | loss:0.7285270094871521 lr:4.091220762224021e-05 tokens_per_second_per_gpu:5172.2756495629665 mfu:24.39850672021291 +Step 20301 | loss:0.7511064410209656 lr:4.090794366013201e-05 tokens_per_second_per_gpu:5018.752528796823 mfu:23.674311965813594 +Step 20302 | loss:0.8749024271965027 lr:4.090367976640635e-05 tokens_per_second_per_gpu:5111.295109040643 mfu:24.110851106216032 +Step 20303 | loss:0.8886315822601318 lr:4.089941594109526e-05 tokens_per_second_per_gpu:4954.8413057823345 mfu:23.372831822475185 +Step 20304 | loss:0.66546231508255 lr:4.089515218423082e-05 tokens_per_second_per_gpu:5160.338433694391 mfu:24.34219683626179 +Step 20305 | loss:0.6486415266990662 lr:4.0890888495845104e-05 tokens_per_second_per_gpu:5345.2352343283355 mfu:25.214386591499284 +Step 20306 | loss:0.787803053855896 lr:4.0886624875970176e-05 tokens_per_second_per_gpu:3337.9905655639586 mfu:15.745871017682711 +Step 20307 | loss:0.7895880937576294 lr:4.0882361324638117e-05 tokens_per_second_per_gpu:4504.807250974383 mfu:21.249944402218638 +Step 20308 | loss:0.7291845679283142 lr:4.087809784188096e-05 tokens_per_second_per_gpu:5311.667348367013 mfu:25.056041146147802 +Step 20309 | loss:0.7279778718948364 lr:4.0873834427730816e-05 tokens_per_second_per_gpu:5184.604396179505 mfu:24.456663521504197 +Step 20310 | loss:0.6154544949531555 lr:4.086957108221972e-05 tokens_per_second_per_gpu:4696.814539966727 mfu:22.15567554421711 +Step 20311 | loss:0.8668624758720398 lr:4.086530780537975e-05 tokens_per_second_per_gpu:5346.750437908803 mfu:25.22153406530874 +Step 20312 | loss:0.6006549000740051 lr:4.0861044597242946e-05 tokens_per_second_per_gpu:4189.69304358308 mfu:19.76349692192611 +Step 20313 | loss:0.7017771005630493 lr:4.0856781457841406e-05 tokens_per_second_per_gpu:3024.9609859442467 mfu:14.269257082262952 +Step 20314 | loss:0.8766334056854248 lr:4.085251838720717e-05 tokens_per_second_per_gpu:4193.462488110047 mfu:19.78127803489337 +Step 20315 | loss:0.7016313076019287 lr:4.084825538537232e-05 tokens_per_second_per_gpu:4619.441365940783 mfu:21.790693081111975 +Step 20316 | loss:0.7707355618476868 lr:4.084399245236892e-05 tokens_per_second_per_gpu:4937.12608192961 mfu:23.28926608903302 +Step 20317 | loss:0.7562680244445801 lr:4.083972958822901e-05 tokens_per_second_per_gpu:5191.794808230495 mfu:24.490581921959517 +Step 20318 | loss:0.6680707335472107 lr:4.083546679298465e-05 tokens_per_second_per_gpu:3879.34241336409 mfu:18.299520047905457 +Step 20319 | loss:0.6467856168746948 lr:4.0831204066667924e-05 tokens_per_second_per_gpu:5229.872401798607 mfu:24.670200427527686 +Step 20320 | loss:0.706133246421814 lr:4.082694140931089e-05 tokens_per_second_per_gpu:4434.086299329337 mfu:20.916341607070372 +Step 20321 | loss:0.6734381318092346 lr:4.0822678820945584e-05 tokens_per_second_per_gpu:5378.724887472682 mfu:25.372362999305192 +Step 20322 | loss:0.6961439847946167 lr:4.081841630160409e-05 tokens_per_second_per_gpu:4514.002789225332 mfu:21.29332132506897 +Step 20323 | loss:0.6037140488624573 lr:4.081415385131847e-05 tokens_per_second_per_gpu:4344.286575758917 mfu:20.492740989575967 +Step 20324 | loss:0.71744704246521 lr:4.080989147012077e-05 tokens_per_second_per_gpu:4171.486324290986 mfu:19.67761272063867 +Step 20325 | loss:0.6918385624885559 lr:4.0805629158043026e-05 tokens_per_second_per_gpu:3585.304529677904 mfu:16.912493182521594 +Step 20326 | loss:0.7101390361785889 lr:4.080136691511733e-05 tokens_per_second_per_gpu:4609.844107250796 mfu:21.745421174410126 +Step 20327 | loss:0.7576766610145569 lr:4.079710474137573e-05 tokens_per_second_per_gpu:4672.9292284693865 mfu:22.043004454629788 +Step 20328 | loss:0.714102029800415 lr:4.079284263685027e-05 tokens_per_second_per_gpu:5346.7760646741945 mfu:25.22165495113347 +Step 20329 | loss:0.7749108076095581 lr:4.078858060157302e-05 tokens_per_second_per_gpu:5311.806240034309 mfu:25.0566963218398 +Step 20330 | loss:0.6908925175666809 lr:4.0784318635576055e-05 tokens_per_second_per_gpu:4790.804303803159 mfu:22.599041296540886 +Step 20331 | loss:0.7833744883537292 lr:4.078005673889138e-05 tokens_per_second_per_gpu:4081.054587355228 mfu:19.251030788267364 +Step 20332 | loss:0.5482261776924133 lr:4.0775794911551075e-05 tokens_per_second_per_gpu:5303.258030742379 mfu:25.016372960134866 +Step 20333 | loss:0.9011874198913574 lr:4.0771533153587197e-05 tokens_per_second_per_gpu:5306.422122396 mfu:25.031298520314706 +Step 20334 | loss:0.700783908367157 lr:4.076727146503179e-05 tokens_per_second_per_gpu:5332.928965183719 mfu:25.156335820278102 +Step 20335 | loss:0.7883201241493225 lr:4.0763009845916924e-05 tokens_per_second_per_gpu:2795.2452103138417 mfu:13.1856485750615 +Step 20336 | loss:0.7797659635543823 lr:4.075874829627463e-05 tokens_per_second_per_gpu:4609.711654880545 mfu:21.74479637398095 +Step 20337 | loss:0.7722052931785583 lr:4.0754486816136966e-05 tokens_per_second_per_gpu:4792.84659578582 mfu:22.608675136274037 +Step 20338 | loss:0.5176592469215393 lr:4.075022540553601e-05 tokens_per_second_per_gpu:5272.879266440802 mfu:24.87307114577267 +Step 20339 | loss:0.7762251496315002 lr:4.0745964064503775e-05 tokens_per_second_per_gpu:4906.786396646524 mfu:23.146148617068608 +Step 20340 | loss:0.67902672290802 lr:4.074170279307232e-05 tokens_per_second_per_gpu:4546.563822162871 mfu:21.446917272920423 +Step 20341 | loss:0.6719931960105896 lr:4.073744159127369e-05 tokens_per_second_per_gpu:3945.6342263447873 mfu:18.612229840285007 +Step 20342 | loss:0.8462718725204468 lr:4.0733180459139966e-05 tokens_per_second_per_gpu:4572.780668814481 mfu:21.570586611631356 +Step 20343 | loss:0.6846284866333008 lr:4.072891939670317e-05 tokens_per_second_per_gpu:4960.621063082276 mfu:23.400095923787287 +Step 20344 | loss:0.8165087699890137 lr:4.072465840399534e-05 tokens_per_second_per_gpu:5366.1154904166615 mfu:25.312882322006963 +Step 20345 | loss:0.6874584555625916 lr:4.072039748104855e-05 tokens_per_second_per_gpu:5243.848838859821 mfu:24.73612966577149 +Step 20346 | loss:0.6672831177711487 lr:4.071613662789484e-05 tokens_per_second_per_gpu:4906.5346535985545 mfu:23.14496110215954 +Step 20347 | loss:0.8387091755867004 lr:4.071187584456624e-05 tokens_per_second_per_gpu:5343.527377063252 mfu:25.206330337389392 +Step 20348 | loss:0.6584128737449646 lr:4.07076151310948e-05 tokens_per_second_per_gpu:4597.22485429642 mfu:21.685894005157763 +Step 20349 | loss:0.604623556137085 lr:4.070335448751259e-05 tokens_per_second_per_gpu:4176.002307665041 mfu:19.698915384720305 +Step 20350 | loss:0.782965898513794 lr:4.0699093913851614e-05 tokens_per_second_per_gpu:3590.7272010808006 mfu:16.9380728487322 +Step 20351 | loss:0.7131943702697754 lr:4.0694833410143964e-05 tokens_per_second_per_gpu:3258.1636962919833 mfu:15.369314055458469 +Step 20352 | loss:0.7261420488357544 lr:4.069057297642166e-05 tokens_per_second_per_gpu:5313.935722673489 mfu:25.066741454776015 +Step 20353 | loss:0.8109543919563293 lr:4.068631261271673e-05 tokens_per_second_per_gpu:5317.94884708581 mfu:25.085672047338566 +Step 20354 | loss:0.6895676255226135 lr:4.0682052319061235e-05 tokens_per_second_per_gpu:5288.2475384832 mfu:24.945565907094252 +Step 20355 | loss:0.688640296459198 lr:4.0677792095487214e-05 tokens_per_second_per_gpu:3820.2414769972615 mfu:18.020730847403588 +Step 20356 | loss:0.793159544467926 lr:4.0673531942026704e-05 tokens_per_second_per_gpu:4472.728465481979 mfu:21.098623297845783 +Step 20357 | loss:0.7355490326881409 lr:4.066927185871175e-05 tokens_per_second_per_gpu:4680.0406254037935 mfu:22.07655012729882 +Step 20358 | loss:0.7406095266342163 lr:4.06650118455744e-05 tokens_per_second_per_gpu:3758.267019071652 mfu:17.728386755435615 +Step 20359 | loss:0.6902194023132324 lr:4.066075190264669e-05 tokens_per_second_per_gpu:5319.201280670641 mfu:25.091579990245414 +Step 20360 | loss:0.7464199066162109 lr:4.065649202996066e-05 tokens_per_second_per_gpu:4798.153340033216 mfu:22.633707954312943 +Step 20361 | loss:0.7760200500488281 lr:4.0652232227548325e-05 tokens_per_second_per_gpu:4427.143371424739 mfu:20.88359063155873 +Step 20362 | loss:0.6929174661636353 lr:4.064797249544175e-05 tokens_per_second_per_gpu:3628.3049814393357 mfu:17.115333650113868 +Step 20363 | loss:0.6776502132415771 lr:4.064371283367298e-05 tokens_per_second_per_gpu:4928.907016637397 mfu:23.250495355732504 +Step 20364 | loss:0.7609307169914246 lr:4.063945324227403e-05 tokens_per_second_per_gpu:4489.156808507038 mfu:21.176118594859336 +Step 20365 | loss:0.813924252986908 lr:4.063519372127695e-05 tokens_per_second_per_gpu:5355.9986509906785 mfu:25.265159464323805 +Step 20366 | loss:0.657072901725769 lr:4.063093427071376e-05 tokens_per_second_per_gpu:4462.906621357696 mfu:21.052291983332132 +Step 20367 | loss:0.8232056498527527 lr:4.0626674890616544e-05 tokens_per_second_per_gpu:3705.956718221919 mfu:17.481630141269775 +Step 20368 | loss:0.6486582159996033 lr:4.062241558101727e-05 tokens_per_second_per_gpu:4716.635445814016 mfu:22.24917413889403 +Step 20369 | loss:0.7436940670013428 lr:4.061815634194802e-05 tokens_per_second_per_gpu:4187.50236931826 mfu:19.75316313765138 +Step 20370 | loss:0.8215829133987427 lr:4.06138971734408e-05 tokens_per_second_per_gpu:4245.145720765351 mfu:20.025076661397218 +Step 20371 | loss:0.74564129114151 lr:4.060963807552766e-05 tokens_per_second_per_gpu:3984.628069891018 mfu:18.796170453328322 +Step 20372 | loss:0.6577214002609253 lr:4.060537904824064e-05 tokens_per_second_per_gpu:4774.713507171462 mfu:22.52313826345617 +Step 20373 | loss:0.6845307350158691 lr:4.060112009161175e-05 tokens_per_second_per_gpu:3815.1420752601866 mfu:17.99667609936215 +Step 20374 | loss:0.7774703502655029 lr:4.059686120567306e-05 tokens_per_second_per_gpu:3719.869240095978 mfu:17.547257880670717 +Step 20375 | loss:0.7771707773208618 lr:4.0592602390456554e-05 tokens_per_second_per_gpu:3102.398763367495 mfu:14.634544290615645 +Step 20376 | loss:0.7922075986862183 lr:4.0588343645994295e-05 tokens_per_second_per_gpu:3732.343649509119 mfu:17.606101798226216 +Step 20377 | loss:0.6132344603538513 lr:4.0584084972318295e-05 tokens_per_second_per_gpu:5299.322864317241 mfu:24.997810108698868 +Step 20378 | loss:0.7818664908409119 lr:4.05798263694606e-05 tokens_per_second_per_gpu:5359.3416391433375 mfu:25.280928909812495 +Step 20379 | loss:0.5136002898216248 lr:4.057556783745323e-05 tokens_per_second_per_gpu:3568.471567858039 mfu:16.833089229617855 +Step 20380 | loss:0.6765324473381042 lr:4.0571309376328214e-05 tokens_per_second_per_gpu:3612.016390332472 mfu:17.038497586742576 +Step 20381 | loss:0.6736544966697693 lr:4.05670509861176e-05 tokens_per_second_per_gpu:5390.87065562619 mfu:25.429656659967026 +Step 20382 | loss:0.7979316711425781 lr:4.056279266685339e-05 tokens_per_second_per_gpu:4588.91066310522 mfu:21.64667454676158 +Step 20383 | loss:0.816195011138916 lr:4.055853441856762e-05 tokens_per_second_per_gpu:5052.13348960059 mfu:23.831775653304167 +Step 20384 | loss:0.8182507157325745 lr:4.0554276241292304e-05 tokens_per_second_per_gpu:3284.005388237047 mfu:15.491213725410628 +Step 20385 | loss:0.8542962074279785 lr:4.05500181350595e-05 tokens_per_second_per_gpu:3461.5681342824782 mfu:16.328807493835356 +Step 20386 | loss:0.8202552199363708 lr:4.05457600999012e-05 tokens_per_second_per_gpu:4772.809356650228 mfu:22.51415606056593 +Step 20387 | loss:0.8316314816474915 lr:4.054150213584945e-05 tokens_per_second_per_gpu:5361.883364588192 mfu:25.292918662398954 +Step 20388 | loss:0.7199305295944214 lr:4.053724424293628e-05 tokens_per_second_per_gpu:3852.5921160928388 mfu:18.173334331605616 +Step 20389 | loss:0.7549012899398804 lr:4.05329864211937e-05 tokens_per_second_per_gpu:4589.529271134386 mfu:21.64959262638057 +Step 20390 | loss:0.738633394241333 lr:4.052872867065374e-05 tokens_per_second_per_gpu:3454.1597397583555 mfu:16.293860832862624 +Step 20391 | loss:0.7342345714569092 lr:4.05244709913484e-05 tokens_per_second_per_gpu:4722.258920726148 mfu:22.275701029518313 +Step 20392 | loss:0.7415270805358887 lr:4.052021338330974e-05 tokens_per_second_per_gpu:4306.730418828863 mfu:20.31558218959578 +Step 20393 | loss:0.6499254107475281 lr:4.0515955846569756e-05 tokens_per_second_per_gpu:5184.434552929562 mfu:24.455862341915974 +Step 20394 | loss:0.7328565120697021 lr:4.051169838116048e-05 tokens_per_second_per_gpu:5340.144713955597 mfu:25.19037373836874 +Step 20395 | loss:0.7453130483627319 lr:4.0507440987113935e-05 tokens_per_second_per_gpu:4581.725572554539 mfu:21.61278124873116 +Step 20396 | loss:0.728442370891571 lr:4.0503183664462153e-05 tokens_per_second_per_gpu:4609.0011814640375 mfu:21.741444949656017 +Step 20397 | loss:0.7465645670890808 lr:4.0498926413237116e-05 tokens_per_second_per_gpu:3365.1852821673338 mfu:15.874153135797934 +Step 20398 | loss:0.755355715751648 lr:4.0494669233470875e-05 tokens_per_second_per_gpu:5152.426301693476 mfu:24.304873959664555 +Step 20399 | loss:0.59805828332901 lr:4.0490412125195434e-05 tokens_per_second_per_gpu:5210.802978580566 mfu:24.580246704629086 +Step 20400 | loss:0.6222432255744934 lr:4.048615508844281e-05 tokens_per_second_per_gpu:4303.376018227943 mfu:20.29975890964149 +Step 20401 | loss:0.7254343032836914 lr:4.0481898123245046e-05 tokens_per_second_per_gpu:5006.000187228535 mfu:23.614156994862146 +Step 20402 | loss:0.9272446036338806 lr:4.0477641229634114e-05 tokens_per_second_per_gpu:4872.637069851081 mfu:22.98506041609855 +Step 20403 | loss:0.7033112645149231 lr:4.04733844076421e-05 tokens_per_second_per_gpu:4351.417741010094 mfu:20.52637991276802 +Step 20404 | loss:0.6337406039237976 lr:4.0469127657300944e-05 tokens_per_second_per_gpu:4247.474908556755 mfu:20.03606383760968 +Step 20405 | loss:0.6923514604568481 lr:4.046487097864271e-05 tokens_per_second_per_gpu:3662.5829440332127 mfu:17.277028648092674 +Step 20406 | loss:0.8097389936447144 lr:4.046061437169938e-05 tokens_per_second_per_gpu:5347.835505185519 mfu:25.226652512784575 +Step 20407 | loss:0.6796097755432129 lr:4.0456357836503e-05 tokens_per_second_per_gpu:4762.363774042469 mfu:22.464882465204962 +Step 20408 | loss:0.7542681694030762 lr:4.045210137308557e-05 tokens_per_second_per_gpu:4960.7017257339485 mfu:23.400476423277695 +Step 20409 | loss:0.8165068626403809 lr:4.0447844981479095e-05 tokens_per_second_per_gpu:5329.424232431347 mfu:25.139803397916097 +Step 20410 | loss:0.7431619763374329 lr:4.0443588661715624e-05 tokens_per_second_per_gpu:3901.7859403328257 mfu:18.405389993877545 +Step 20411 | loss:0.6868336796760559 lr:4.043933241382712e-05 tokens_per_second_per_gpu:4730.242415480946 mfu:22.31336053640136 +Step 20412 | loss:0.6397649049758911 lr:4.043507623784562e-05 tokens_per_second_per_gpu:5201.134718019636 mfu:24.53463986998783 +Step 20413 | loss:0.6603280305862427 lr:4.043082013380312e-05 tokens_per_second_per_gpu:3201.344618294227 mfu:15.101288770208699 +Step 20414 | loss:0.5721166133880615 lr:4.042656410173166e-05 tokens_per_second_per_gpu:4161.114666051922 mfu:19.628687839137633 +Step 20415 | loss:0.7605000138282776 lr:4.042230814166323e-05 tokens_per_second_per_gpu:4509.337402663377 mfu:21.2713139006592 +Step 20416 | loss:0.6866538524627686 lr:4.041805225362983e-05 tokens_per_second_per_gpu:5352.235817561433 mfu:25.247409537032468 +Step 20417 | loss:0.7289354801177979 lr:4.04137964376635e-05 tokens_per_second_per_gpu:4959.869304848563 mfu:23.39654975193568 +Step 20418 | loss:0.5765364766120911 lr:4.0409540693796226e-05 tokens_per_second_per_gpu:2678.7469343904045 mfu:12.636106330875156 +Step 20419 | loss:0.6557124853134155 lr:4.040528502206002e-05 tokens_per_second_per_gpu:4126.946339020506 mfu:19.467510010812713 +Step 20420 | loss:0.5343092083930969 lr:4.0401029422486876e-05 tokens_per_second_per_gpu:3865.052098761712 mfu:18.232110195746948 +Step 20421 | loss:0.701691746711731 lr:4.039677389510882e-05 tokens_per_second_per_gpu:4376.800961759507 mfu:20.646116895867046 +Step 20422 | loss:0.7536360025405884 lr:4.039251843995785e-05 tokens_per_second_per_gpu:4175.49620915315 mfu:19.696528031690402 +Step 20423 | loss:0.7866056561470032 lr:4.038826305706599e-05 tokens_per_second_per_gpu:4099.323904504904 mfu:19.337210274329315 +Step 20424 | loss:0.88530433177948 lr:4.038400774646522e-05 tokens_per_second_per_gpu:5338.302596095002 mfu:25.181684154122596 +Step 20425 | loss:0.5973645448684692 lr:4.0379752508187565e-05 tokens_per_second_per_gpu:3864.4152635989267 mfu:18.229106135628808 +Step 20426 | loss:0.7263115644454956 lr:4.037549734226502e-05 tokens_per_second_per_gpu:5341.311050140708 mfu:25.195875545151385 +Step 20427 | loss:0.734697163105011 lr:4.037124224872957e-05 tokens_per_second_per_gpu:4593.493820040424 mfu:21.668294079992144 +Step 20428 | loss:0.6342867016792297 lr:4.036698722761324e-05 tokens_per_second_per_gpu:5147.790121021042 mfu:24.283004304418693 +Step 20429 | loss:0.8143284916877747 lr:4.036273227894802e-05 tokens_per_second_per_gpu:5315.362911176269 mfu:25.0734737464471 +Step 20430 | loss:0.8054599165916443 lr:4.0358477402765935e-05 tokens_per_second_per_gpu:4947.487955399519 mfu:23.338144814108492 +Step 20431 | loss:0.8365476131439209 lr:4.035422259909897e-05 tokens_per_second_per_gpu:4126.213759488567 mfu:19.4640543081696 +Step 20432 | loss:0.6857601404190063 lr:4.0349967867979145e-05 tokens_per_second_per_gpu:4370.787507505982 mfu:20.617750406152073 +Step 20433 | loss:0.7697368264198303 lr:4.0345713209438415e-05 tokens_per_second_per_gpu:5370.37715573564 mfu:25.332985324431668 +Step 20434 | loss:0.6447778344154358 lr:4.034145862350882e-05 tokens_per_second_per_gpu:3722.7508819597506 mfu:17.56085107700093 +Step 20435 | loss:0.7619659900665283 lr:4.0337204110222346e-05 tokens_per_second_per_gpu:4311.885523732187 mfu:20.339899698976357 +Step 20436 | loss:0.696595311164856 lr:4.033294966961098e-05 tokens_per_second_per_gpu:5357.34134946595 mfu:25.271493202119643 +Step 20437 | loss:0.7519062757492065 lr:4.032869530170675e-05 tokens_per_second_per_gpu:5334.408356584709 mfu:25.16331436192677 +Step 20438 | loss:0.8241204619407654 lr:4.032444100654162e-05 tokens_per_second_per_gpu:4707.243841994912 mfu:22.204872341306476 +Step 20439 | loss:0.5984313488006592 lr:4.032018678414764e-05 tokens_per_second_per_gpu:4450.079198687916 mfu:20.99178284201479 +Step 20440 | loss:0.7998617887496948 lr:4.031593263455674e-05 tokens_per_second_per_gpu:3342.410411677419 mfu:15.76672018590341 +Step 20441 | loss:0.7655316591262817 lr:4.0311678557800946e-05 tokens_per_second_per_gpu:5375.803328195355 mfu:25.35858150572463 +Step 20442 | loss:0.7547873854637146 lr:4.030742455391227e-05 tokens_per_second_per_gpu:4533.1975425280225 mfu:21.38386625136038 +Step 20443 | loss:0.6718617677688599 lr:4.030317062292266e-05 tokens_per_second_per_gpu:4905.0238291543155 mfu:23.13783428548284 +Step 20444 | loss:0.5836769342422485 lr:4.029891676486417e-05 tokens_per_second_per_gpu:4326.211844629221 mfu:20.407479398970057 +Step 20445 | loss:0.7516722679138184 lr:4.029466297976875e-05 tokens_per_second_per_gpu:3491.3828467092703 mfu:16.46944857926702 +Step 20446 | loss:0.6769755482673645 lr:4.0290409267668436e-05 tokens_per_second_per_gpu:3790.5685386253454 mfu:17.88075853437792 +Step 20447 | loss:0.7335858941078186 lr:4.028615562859516e-05 tokens_per_second_per_gpu:3580.025085750495 mfu:16.887589144749988 +Step 20448 | loss:0.6271198987960815 lr:4.028190206258096e-05 tokens_per_second_per_gpu:5256.894031048298 mfu:24.797666063063833 +Step 20449 | loss:0.6777599453926086 lr:4.0277648569657806e-05 tokens_per_second_per_gpu:4379.950403670526 mfu:20.66097335071152 +Step 20450 | loss:0.8540610074996948 lr:4.0273395149857705e-05 tokens_per_second_per_gpu:4071.595178815404 mfu:19.206409144243494 +Step 20451 | loss:0.6206130981445312 lr:4.026914180321264e-05 tokens_per_second_per_gpu:3603.321743839059 mfu:16.997483455773484 +Step 20452 | loss:0.6322998404502869 lr:4.0264888529754594e-05 tokens_per_second_per_gpu:5307.308660730855 mfu:25.035480472899252 +Step 20453 | loss:0.5391076803207397 lr:4.0260635329515584e-05 tokens_per_second_per_gpu:4927.317398717302 mfu:23.242996856786636 +Step 20454 | loss:0.6915448307991028 lr:4.0256382202527564e-05 tokens_per_second_per_gpu:4399.055885053391 mfu:20.751097166125753 +Step 20455 | loss:0.6623595356941223 lr:4.025212914882254e-05 tokens_per_second_per_gpu:5361.347977084351 mfu:25.290393148196806 +Step 20456 | loss:0.7899923920631409 lr:4.024787616843249e-05 tokens_per_second_per_gpu:5360.921092926178 mfu:25.288379462788757 +Step 20457 | loss:0.8777220845222473 lr:4.0243623261389406e-05 tokens_per_second_per_gpu:5317.041118551133 mfu:25.081390136966036 +Step 20458 | loss:0.799089789390564 lr:4.023937042772529e-05 tokens_per_second_per_gpu:3960.382405360899 mfu:18.68179951700282 +Step 20459 | loss:0.7371007204055786 lr:4.0235117667472097e-05 tokens_per_second_per_gpu:4414.912380605728 mfu:20.825894961043236 +Step 20460 | loss:0.6983031630516052 lr:4.0230864980661844e-05 tokens_per_second_per_gpu:5227.558993473403 mfu:24.659287685749536 +Step 20461 | loss:0.6629601716995239 lr:4.022661236732651e-05 tokens_per_second_per_gpu:3673.8845048660564 mfu:17.33034003878631 +Step 20462 | loss:0.7519707679748535 lr:4.0222359827498054e-05 tokens_per_second_per_gpu:4551.048986381997 mfu:21.468074557789965 +Step 20463 | loss:0.7975314855575562 lr:4.021810736120848e-05 tokens_per_second_per_gpu:5130.395366302029 mfu:24.200950278557922 +Step 20464 | loss:0.6915829181671143 lr:4.021385496848977e-05 tokens_per_second_per_gpu:4595.976541970169 mfu:21.68000550293079 +Step 20465 | loss:0.7265753149986267 lr:4.02096026493739e-05 tokens_per_second_per_gpu:5325.641076033722 mfu:25.121957603715458 +Step 20466 | loss:0.8441768288612366 lr:4.0205350403892864e-05 tokens_per_second_per_gpu:4700.056632814274 mfu:22.17096904933707 +Step 20467 | loss:0.776552140712738 lr:4.0201098232078644e-05 tokens_per_second_per_gpu:3406.0465927408886 mfu:16.06690290943187 +Step 20468 | loss:0.6106162071228027 lr:4.0196846133963214e-05 tokens_per_second_per_gpu:5302.468424471671 mfu:25.01264825263522 +Step 20469 | loss:0.5850824117660522 lr:4.019259410957855e-05 tokens_per_second_per_gpu:4330.575376285618 mfu:20.428062922288557 +Step 20470 | loss:0.6641061902046204 lr:4.018834215895664e-05 tokens_per_second_per_gpu:4728.926060745495 mfu:22.307151066520962 +Step 20471 | loss:0.6380762457847595 lr:4.0184090282129456e-05 tokens_per_second_per_gpu:5334.280024734472 mfu:25.16270899869353 +Step 20472 | loss:0.7723731994628906 lr:4.017983847912897e-05 tokens_per_second_per_gpu:4165.3238544514625 mfu:19.648543298979863 +Step 20473 | loss:0.688357949256897 lr:4.017558674998719e-05 tokens_per_second_per_gpu:4810.262686415995 mfu:22.690829807268045 +Step 20474 | loss:0.6778582334518433 lr:4.0171335094736077e-05 tokens_per_second_per_gpu:4254.9861684099415 mfu:20.071495731890245 +Step 20475 | loss:0.7954665422439575 lr:4.0167083513407605e-05 tokens_per_second_per_gpu:5143.771812141908 mfu:24.26404925584163 +Step 20476 | loss:0.7223337888717651 lr:4.016283200603374e-05 tokens_per_second_per_gpu:5246.945346265183 mfu:24.750736419521225 +Step 20477 | loss:0.6224759221076965 lr:4.015858057264648e-05 tokens_per_second_per_gpu:4335.204036200628 mfu:20.449897100838474 +Step 20478 | loss:0.7275394201278687 lr:4.015432921327779e-05 tokens_per_second_per_gpu:5332.777078634176 mfu:25.15561934550939 +Step 20479 | loss:0.8487781882286072 lr:4.015007792795964e-05 tokens_per_second_per_gpu:5360.309377141411 mfu:25.285493895062533 +Step 20480 | loss:0.6927841901779175 lr:4.0145826716724025e-05 tokens_per_second_per_gpu:4472.691550448236 mfu:21.09844916333326 +Step 20481 | loss:0.7438494563102722 lr:4.014157557960288e-05 tokens_per_second_per_gpu:4847.785831545827 mfu:22.867832884954186 +Step 20482 | loss:0.8053454160690308 lr:4.013732451662824e-05 tokens_per_second_per_gpu:4951.42046147604 mfu:23.356695116226167 +Step 20483 | loss:0.8405385613441467 lr:4.013307352783201e-05 tokens_per_second_per_gpu:5294.214392147231 mfu:24.9737125738777 +Step 20484 | loss:0.6385177373886108 lr:4.012882261324621e-05 tokens_per_second_per_gpu:3651.0972366033043 mfu:17.222848606482035 +Step 20485 | loss:0.9391164183616638 lr:4.012457177290278e-05 tokens_per_second_per_gpu:3587.5748297195114 mfu:16.923202575170986 +Step 20486 | loss:0.6015414595603943 lr:4.012032100683372e-05 tokens_per_second_per_gpu:5338.38764511875 mfu:25.182085344878477 +Step 20487 | loss:0.6166432499885559 lr:4.011607031507098e-05 tokens_per_second_per_gpu:5339.85857527328 mfu:25.18902397338396 +Step 20488 | loss:0.7557916641235352 lr:4.011181969764653e-05 tokens_per_second_per_gpu:3910.2033645426254 mfu:18.44509642003583 +Step 20489 | loss:0.6915445923805237 lr:4.010756915459235e-05 tokens_per_second_per_gpu:4442.842570705586 mfu:20.95764643312609 +Step 20490 | loss:0.8270502090454102 lr:4.010331868594042e-05 tokens_per_second_per_gpu:5253.0268563971695 mfu:24.77942394042643 +Step 20491 | loss:0.6955538392066956 lr:4.009906829172268e-05 tokens_per_second_per_gpu:5345.235274930936 mfu:25.21438678302868 +Step 20492 | loss:0.7241777777671814 lr:4.009481797197111e-05 tokens_per_second_per_gpu:4221.880809308642 mfu:19.915332104652713 +Step 20493 | loss:0.7716714143753052 lr:4.009056772671768e-05 tokens_per_second_per_gpu:4200.473746905443 mfu:19.814351338875348 +Step 20494 | loss:0.6139273047447205 lr:4.0086317555994354e-05 tokens_per_second_per_gpu:4576.57210160785 mfu:21.588471447002814 +Step 20495 | loss:0.7050293684005737 lr:4.008206745983308e-05 tokens_per_second_per_gpu:3804.044296815689 mfu:17.944325984963204 +Step 20496 | loss:0.7971532344818115 lr:4.007781743826587e-05 tokens_per_second_per_gpu:4922.104053011004 mfu:23.21840461560153 +Step 20497 | loss:0.88749098777771 lr:4.007356749132466e-05 tokens_per_second_per_gpu:4128.757804855831 mfu:19.47605500422588 +Step 20498 | loss:0.7953502535820007 lr:4.006931761904141e-05 tokens_per_second_per_gpu:3887.85134824281 mfu:18.33965814550237 +Step 20499 | loss:0.753653883934021 lr:4.006506782144808e-05 tokens_per_second_per_gpu:4344.02103238663 mfu:20.49148837618252 +Step 20500 | loss:0.7837831377983093 lr:4.006081809857665e-05 tokens_per_second_per_gpu:4405.732746551216 mfu:20.782593061001034 +Step 20501 | loss:0.7901551127433777 lr:4.0056568450459065e-05 tokens_per_second_per_gpu:3385.47644320254 mfu:15.969870004430069 +Step 20502 | loss:0.708407998085022 lr:4.005231887712731e-05 tokens_per_second_per_gpu:4726.805095299322 mfu:22.29714611063734 +Step 20503 | loss:0.7100769281387329 lr:4.004806937861333e-05 tokens_per_second_per_gpu:5347.064069858989 mfu:25.223013520728895 +Step 20504 | loss:0.7715094685554504 lr:4.00438199549491e-05 tokens_per_second_per_gpu:4239.411461383957 mfu:19.998027181529647 +Step 20505 | loss:0.7268712520599365 lr:4.003957060616655e-05 tokens_per_second_per_gpu:5340.049312616768 mfu:25.189923713976334 +Step 20506 | loss:0.7959862351417542 lr:4.003532133229768e-05 tokens_per_second_per_gpu:4443.14611559571 mfu:20.959078306162905 +Step 20507 | loss:0.7168628573417664 lr:4.0031072133374425e-05 tokens_per_second_per_gpu:3578.5791675312503 mfu:16.880768501810724 +Step 20508 | loss:0.7786837816238403 lr:4.002682300942874e-05 tokens_per_second_per_gpu:3763.8595482658566 mfu:17.75476766982855 +Step 20509 | loss:0.7251870036125183 lr:4.0022573960492595e-05 tokens_per_second_per_gpu:4232.2834128199 mfu:19.964402960282495 +Step 20510 | loss:0.8797374963760376 lr:4.0018324986597944e-05 tokens_per_second_per_gpu:4575.657462827585 mfu:21.58415693982284 +Step 20511 | loss:0.8047598600387573 lr:4.0014076087776765e-05 tokens_per_second_per_gpu:4067.497584518276 mfu:19.187080092822264 +Step 20512 | loss:0.6740916967391968 lr:4.000982726406096e-05 tokens_per_second_per_gpu:4305.81500926676 mfu:20.31126404650637 +Step 20513 | loss:0.7393323183059692 lr:4.000557851548254e-05 tokens_per_second_per_gpu:5357.617241819631 mfu:25.27279463342405 +Step 20514 | loss:0.8083606362342834 lr:4.000132984207344e-05 tokens_per_second_per_gpu:4113.982798823225 mfu:19.406358779894102 +Step 20515 | loss:0.7035973072052002 lr:3.9997081243865595e-05 tokens_per_second_per_gpu:5372.301077421818 mfu:25.342060791280257 +Step 20516 | loss:0.713919997215271 lr:3.9992832720890993e-05 tokens_per_second_per_gpu:4289.504262747821 mfu:20.23432347227621 +Step 20517 | loss:0.6651254892349243 lr:3.9988584273181564e-05 tokens_per_second_per_gpu:5341.359424871113 mfu:25.196103737008286 +Step 20518 | loss:0.736007571220398 lr:3.998433590076928e-05 tokens_per_second_per_gpu:3816.4856732522476 mfu:18.00301408033216 +Step 20519 | loss:0.832660436630249 lr:3.998008760368609e-05 tokens_per_second_per_gpu:3682.26183179648 mfu:17.36985731923648 +Step 20520 | loss:0.7643248438835144 lr:3.997583938196393e-05 tokens_per_second_per_gpu:4110.639462350792 mfu:19.390587691321617 +Step 20521 | loss:0.6772893071174622 lr:3.997159123563475e-05 tokens_per_second_per_gpu:3817.5688775363046 mfu:18.008123739753767 +Step 20522 | loss:0.6572105288505554 lr:3.996734316473053e-05 tokens_per_second_per_gpu:4019.756495645554 mfu:18.961877231139937 +Step 20523 | loss:0.8671562075614929 lr:3.99630951692832e-05 tokens_per_second_per_gpu:4492.691335355096 mfu:21.192791561054094 +Step 20524 | loss:0.7083290815353394 lr:3.99588472493247e-05 tokens_per_second_per_gpu:4190.173450885426 mfu:19.765763085137213 +Step 20525 | loss:0.6072940230369568 lr:3.9954599404887e-05 tokens_per_second_per_gpu:3619.7723088675516 mfu:17.0750835777687 +Step 20526 | loss:0.6821813583374023 lr:3.995035163600205e-05 tokens_per_second_per_gpu:5305.686532268165 mfu:25.02782861617709 +Step 20527 | loss:0.6712244153022766 lr:3.9946103942701777e-05 tokens_per_second_per_gpu:4792.436606023883 mfu:22.606741144615516 +Step 20528 | loss:0.6640689373016357 lr:3.9941856325018134e-05 tokens_per_second_per_gpu:4958.15695370921 mfu:23.388472299454122 +Step 20529 | loss:0.7946126461029053 lr:3.993760878298308e-05 tokens_per_second_per_gpu:5357.799493719251 mfu:25.273654346729934 +Step 20530 | loss:0.7270177602767944 lr:3.9933361316628545e-05 tokens_per_second_per_gpu:4812.608954902195 mfu:22.701897556032137 +Step 20531 | loss:0.7881703972816467 lr:3.992911392598649e-05 tokens_per_second_per_gpu:4279.01124653302 mfu:20.184826124497825 +Step 20532 | loss:0.8149431347846985 lr:3.992486661108886e-05 tokens_per_second_per_gpu:4440.898083082855 mfu:20.94847395324587 +Step 20533 | loss:0.7050677537918091 lr:3.99206193719676e-05 tokens_per_second_per_gpu:5073.814440537744 mfu:23.934048398026395 +Step 20534 | loss:0.6901707053184509 lr:3.991637220865463e-05 tokens_per_second_per_gpu:5298.224741422337 mfu:24.992630075644517 +Step 20535 | loss:0.6306965351104736 lr:3.9912125121181916e-05 tokens_per_second_per_gpu:5266.700232855169 mfu:24.8439235900974 +Step 20536 | loss:0.8819095492362976 lr:3.9907878109581395e-05 tokens_per_second_per_gpu:4538.870105110395 mfu:21.41062469690041 +Step 20537 | loss:0.7254831194877625 lr:3.990363117388501e-05 tokens_per_second_per_gpu:4374.382622444154 mfu:20.63470917670529 +Step 20538 | loss:0.6470361948013306 lr:3.989938431412471e-05 tokens_per_second_per_gpu:5374.15766106761 mfu:25.350818613103293 +Step 20539 | loss:0.8629505634307861 lr:3.989513753033243e-05 tokens_per_second_per_gpu:4812.882003980856 mfu:22.703185575123154 +Step 20540 | loss:0.6135523319244385 lr:3.9890890822540116e-05 tokens_per_second_per_gpu:5333.214444132799 mfu:25.157682473188643 +Step 20541 | loss:0.6889902353286743 lr:3.988664419077968e-05 tokens_per_second_per_gpu:3663.4872258508776 mfu:17.28129429971299 +Step 20542 | loss:0.7289050817489624 lr:3.98823976350831e-05 tokens_per_second_per_gpu:4005.1638573826544 mfu:18.893041266693015 +Step 20543 | loss:0.7403581738471985 lr:3.9878151155482296e-05 tokens_per_second_per_gpu:5355.5167700252105 mfu:25.262886349592325 +Step 20544 | loss:0.666090190410614 lr:3.98739047520092e-05 tokens_per_second_per_gpu:3446.1809354256516 mfu:16.256223451500865 +Step 20545 | loss:0.8872410655021667 lr:3.986965842469577e-05 tokens_per_second_per_gpu:4715.544659851493 mfu:22.244028715401534 +Step 20546 | loss:0.61353999376297 lr:3.986541217357392e-05 tokens_per_second_per_gpu:3610.509420203734 mfu:17.0313889515297 +Step 20547 | loss:0.7024082541465759 lr:3.9861165998675624e-05 tokens_per_second_per_gpu:4936.891427893273 mfu:23.28815918590751 +Step 20548 | loss:0.7181318998336792 lr:3.985691990003276e-05 tokens_per_second_per_gpu:4026.7214845803296 mfu:18.994732272294925 +Step 20549 | loss:0.6157389879226685 lr:3.985267387767732e-05 tokens_per_second_per_gpu:4075.7488480414763 mfu:19.226002710671924 +Step 20550 | loss:0.8079702258110046 lr:3.98484279316412e-05 tokens_per_second_per_gpu:3987.4486367143154 mfu:18.809475548272143 +Step 20551 | loss:0.6260033249855042 lr:3.9844182061956345e-05 tokens_per_second_per_gpu:4650.88700861794 mfu:21.93902754280001 +Step 20552 | loss:0.678342342376709 lr:3.98399362686547e-05 tokens_per_second_per_gpu:3616.908897778826 mfu:17.061576379114875 +Step 20553 | loss:0.6178860068321228 lr:3.983569055176819e-05 tokens_per_second_per_gpu:4285.630922909987 mfu:20.21605226740123 +Step 20554 | loss:0.6807360649108887 lr:3.9831444911328745e-05 tokens_per_second_per_gpu:4097.484168856885 mfu:19.328531927386337 +Step 20555 | loss:0.7267361879348755 lr:3.982719934736832e-05 tokens_per_second_per_gpu:5355.926112448751 mfu:25.264817287645187 +Step 20556 | loss:0.7007288932800293 lr:3.982295385991882e-05 tokens_per_second_per_gpu:5115.838936890575 mfu:24.132285117441132 +Step 20557 | loss:0.8228041529655457 lr:3.981870844901216e-05 tokens_per_second_per_gpu:5163.032637708562 mfu:24.354905856275245 +Step 20558 | loss:0.6687502264976501 lr:3.981446311468031e-05 tokens_per_second_per_gpu:5335.618034770493 mfu:25.169020620321497 +Step 20559 | loss:0.8384581804275513 lr:3.981021785695518e-05 tokens_per_second_per_gpu:4864.684360325035 mfu:22.947546128392027 +Step 20560 | loss:0.8402674198150635 lr:3.980597267586869e-05 tokens_per_second_per_gpu:3945.3178224393523 mfu:18.610737309078996 +Step 20561 | loss:0.859331488609314 lr:3.980172757145279e-05 tokens_per_second_per_gpu:5242.88495888132 mfu:24.731582879458017 +Step 20562 | loss:0.7871069312095642 lr:3.9797482543739414e-05 tokens_per_second_per_gpu:4150.031629411999 mfu:19.576407264346077 +Step 20563 | loss:0.6354201436042786 lr:3.979323759276047e-05 tokens_per_second_per_gpu:4551.165681646798 mfu:21.4686250292643 +Step 20564 | loss:0.7410531044006348 lr:3.978899271854787e-05 tokens_per_second_per_gpu:3259.2752363345976 mfu:15.374557379487767 +Step 20565 | loss:0.6253707408905029 lr:3.9784747921133565e-05 tokens_per_second_per_gpu:5208.723354365244 mfu:24.570436762384933 +Step 20566 | loss:0.7265686988830566 lr:3.978050320054948e-05 tokens_per_second_per_gpu:4313.594220037372 mfu:20.347959910981565 +Step 20567 | loss:0.6777625679969788 lr:3.977625855682752e-05 tokens_per_second_per_gpu:5153.793200234468 mfu:24.311321853299354 +Step 20568 | loss:0.8418616056442261 lr:3.9772013989999634e-05 tokens_per_second_per_gpu:3912.9805611175025 mfu:18.458196930118124 +Step 20569 | loss:0.823654294013977 lr:3.976776950009775e-05 tokens_per_second_per_gpu:4012.99029246015 mfu:18.929959896281083 +Step 20570 | loss:0.8614828586578369 lr:3.976352508715376e-05 tokens_per_second_per_gpu:5337.823741724001 mfu:25.179425316353605 +Step 20571 | loss:0.6920873522758484 lr:3.97592807511996e-05 tokens_per_second_per_gpu:5341.768748532642 mfu:25.198034586557537 +Step 20572 | loss:0.6359240412712097 lr:3.975503649226721e-05 tokens_per_second_per_gpu:4126.170110687603 mfu:19.463848409328197 +Step 20573 | loss:0.5964660048484802 lr:3.9750792310388485e-05 tokens_per_second_per_gpu:4054.0430556783826 mfu:19.123612784705152 +Step 20574 | loss:0.6798911690711975 lr:3.974654820559537e-05 tokens_per_second_per_gpu:3664.8851808785857 mfu:17.28788869209441 +Step 20575 | loss:0.8686037063598633 lr:3.9742304177919763e-05 tokens_per_second_per_gpu:4853.950170544242 mfu:22.896911123754393 +Step 20576 | loss:0.7051576972007751 lr:3.973806022739362e-05 tokens_per_second_per_gpu:5340.875635921147 mfu:25.193821621988384 +Step 20577 | loss:0.6655217409133911 lr:3.973381635404881e-05 tokens_per_second_per_gpu:4420.307911248641 mfu:20.851346599658314 +Step 20578 | loss:0.8207062482833862 lr:3.972957255791728e-05 tokens_per_second_per_gpu:3556.9966269959286 mfu:16.778959975744485 +Step 20579 | loss:0.6697314381599426 lr:3.972532883903095e-05 tokens_per_second_per_gpu:5349.548250500244 mfu:25.234731824658617 +Step 20580 | loss:0.845361053943634 lr:3.972108519742173e-05 tokens_per_second_per_gpu:5323.914649680019 mfu:25.11381375604633 +Step 20581 | loss:0.6868942379951477 lr:3.9716841633121544e-05 tokens_per_second_per_gpu:4596.264278612498 mfu:21.681362805765595 +Step 20582 | loss:0.7656159996986389 lr:3.97125981461623e-05 tokens_per_second_per_gpu:3486.373973343863 mfu:16.445820869000475 +Step 20583 | loss:0.7603049874305725 lr:3.9708354736575916e-05 tokens_per_second_per_gpu:5356.13550200808 mfu:25.26580501392208 +Step 20584 | loss:0.7589504718780518 lr:3.970411140439433e-05 tokens_per_second_per_gpu:5360.2592572321655 mfu:25.285257470899285 +Step 20585 | loss:0.6295098662376404 lr:3.969986814964942e-05 tokens_per_second_per_gpu:5298.146135484068 mfu:24.992259278022146 +Step 20586 | loss:0.7492226362228394 lr:3.969562497237312e-05 tokens_per_second_per_gpu:4161.460482709045 mfu:19.630319115311693 +Step 20587 | loss:0.8038785457611084 lr:3.969138187259732e-05 tokens_per_second_per_gpu:3832.3001323170056 mfu:18.077613581965572 +Step 20588 | loss:0.7828201055526733 lr:3.968713885035397e-05 tokens_per_second_per_gpu:5288.127936157806 mfu:24.94500172251911 +Step 20589 | loss:0.6912909746170044 lr:3.9682895905674955e-05 tokens_per_second_per_gpu:4693.7152343935095 mfu:22.141055591032185 +Step 20590 | loss:0.7846190929412842 lr:3.9678653038592205e-05 tokens_per_second_per_gpu:5095.256919696045 mfu:24.03519623068029 +Step 20591 | loss:0.7768223881721497 lr:3.967441024913764e-05 tokens_per_second_per_gpu:5279.6022358608325 mfu:24.90478453958382 +Step 20592 | loss:0.7839056849479675 lr:3.9670167537343134e-05 tokens_per_second_per_gpu:5285.456385462267 mfu:24.932399562074703 +Step 20593 | loss:0.5895739197731018 lr:3.9665924903240605e-05 tokens_per_second_per_gpu:4720.6540942869115 mfu:22.26813078939281 +Step 20594 | loss:0.6743721961975098 lr:3.966168234686199e-05 tokens_per_second_per_gpu:4725.594487906745 mfu:22.291435469015635 +Step 20595 | loss:0.7348150014877319 lr:3.9657439868239173e-05 tokens_per_second_per_gpu:4772.845168820454 mfu:22.51432499268307 +Step 20596 | loss:0.6434154510498047 lr:3.965319746740407e-05 tokens_per_second_per_gpu:4006.5002997938705 mfu:18.89934549356731 +Step 20597 | loss:0.7508872747421265 lr:3.964895514438859e-05 tokens_per_second_per_gpu:4396.808149647922 mfu:20.74049421471495 +Step 20598 | loss:0.7620792984962463 lr:3.9644712899224654e-05 tokens_per_second_per_gpu:4933.695649884241 mfu:23.273084155782954 +Step 20599 | loss:0.7479560375213623 lr:3.964047073194415e-05 tokens_per_second_per_gpu:4089.5915190147093 mfu:19.291300951455874 +Step 20600 | loss:0.7097516655921936 lr:3.963622864257897e-05 tokens_per_second_per_gpu:4368.809872882463 mfu:20.608421566213675 +Step 20601 | loss:0.6681619882583618 lr:3.963198663116104e-05 tokens_per_second_per_gpu:4909.072103744799 mfu:23.156930687433018 +Step 20602 | loss:0.609993040561676 lr:3.962774469772227e-05 tokens_per_second_per_gpu:5326.789743964326 mfu:25.127376066327074 +Step 20603 | loss:0.7873860597610474 lr:3.9623502842294547e-05 tokens_per_second_per_gpu:3072.9574204261367 mfu:14.495664452750177 +Step 20604 | loss:0.6904216408729553 lr:3.961926106490979e-05 tokens_per_second_per_gpu:4882.782607807804 mfu:23.03291864964768 +Step 20605 | loss:0.681736409664154 lr:3.96150193655999e-05 tokens_per_second_per_gpu:3976.9717009025917 mfu:18.760054054498664 +Step 20606 | loss:0.807174563407898 lr:3.961077774439677e-05 tokens_per_second_per_gpu:4769.823826464879 mfu:22.500072805297663 +Step 20607 | loss:0.694262683391571 lr:3.96065362013323e-05 tokens_per_second_per_gpu:5331.846691390335 mfu:25.151230550139758 +Step 20608 | loss:0.727919340133667 lr:3.9602294736438405e-05 tokens_per_second_per_gpu:3846.3967583977337 mfu:18.144109772320785 +Step 20609 | loss:0.8077672719955444 lr:3.9598053349746966e-05 tokens_per_second_per_gpu:4796.595728431348 mfu:22.626360434630985 +Step 20610 | loss:0.7350453734397888 lr:3.959381204128991e-05 tokens_per_second_per_gpu:4161.344340484903 mfu:19.629771252624504 +Step 20611 | loss:0.647701621055603 lr:3.958957081109912e-05 tokens_per_second_per_gpu:4215.931366832714 mfu:19.887267569414202 +Step 20612 | loss:0.6460085511207581 lr:3.958532965920648e-05 tokens_per_second_per_gpu:5360.156895451799 mfu:25.284774612916575 +Step 20613 | loss:0.7082422971725464 lr:3.9581088585643926e-05 tokens_per_second_per_gpu:4282.56381309908 mfu:20.201584187120286 +Step 20614 | loss:0.8052912354469299 lr:3.9576847590443336e-05 tokens_per_second_per_gpu:3528.3127023330385 mfu:16.643653009127434 +Step 20615 | loss:0.6981452703475952 lr:3.957260667363659e-05 tokens_per_second_per_gpu:3750.7729690525393 mfu:17.69303604287868 +Step 20616 | loss:0.7323752045631409 lr:3.9568365835255594e-05 tokens_per_second_per_gpu:4794.728889153463 mfu:22.617554235241574 +Step 20617 | loss:0.597172737121582 lr:3.956412507533226e-05 tokens_per_second_per_gpu:5371.829536892349 mfu:25.33983645414964 +Step 20618 | loss:0.5761362910270691 lr:3.955988439389848e-05 tokens_per_second_per_gpu:5304.838845727269 mfu:25.023829934133445 +Step 20619 | loss:0.7205919027328491 lr:3.955564379098612e-05 tokens_per_second_per_gpu:4131.790435198453 mfu:19.490360438972125 +Step 20620 | loss:0.6134737133979797 lr:3.9551403266627124e-05 tokens_per_second_per_gpu:5378.294799607156 mfu:25.370334201462953 +Step 20621 | loss:0.8591243624687195 lr:3.954716282085334e-05 tokens_per_second_per_gpu:3967.000261492057 mfu:18.713017073496175 +Step 20622 | loss:0.6457725167274475 lr:3.954292245369669e-05 tokens_per_second_per_gpu:5382.1366870804395 mfu:25.388457040168024 +Step 20623 | loss:0.6925885677337646 lr:3.9538682165189035e-05 tokens_per_second_per_gpu:5016.985060983912 mfu:23.665974518578988 +Step 20624 | loss:0.7493019104003906 lr:3.95344419553623e-05 tokens_per_second_per_gpu:4044.8131313495965 mfu:19.08007365685879 +Step 20625 | loss:0.7716982960700989 lr:3.953020182424835e-05 tokens_per_second_per_gpu:5347.603047280523 mfu:25.2255559691853 +Step 20626 | loss:0.718268871307373 lr:3.95259617718791e-05 tokens_per_second_per_gpu:3976.559008078909 mfu:18.758107312036696 +Step 20627 | loss:0.6947275400161743 lr:3.952172179828644e-05 tokens_per_second_per_gpu:3239.901482049766 mfu:15.283168075022884 +Step 20628 | loss:0.6606980562210083 lr:3.951748190350223e-05 tokens_per_second_per_gpu:4721.309925400371 mfu:22.271224456651325 +Step 20629 | loss:0.840096652507782 lr:3.951324208755838e-05 tokens_per_second_per_gpu:3475.2748571150255 mfu:16.393464444044547 +Step 20630 | loss:0.8109616041183472 lr:3.950900235048677e-05 tokens_per_second_per_gpu:4040.704424061696 mfu:19.060692183564015 +Step 20631 | loss:0.6824649572372437 lr:3.9504762692319294e-05 tokens_per_second_per_gpu:5330.1024428956325 mfu:25.143002632390765 +Step 20632 | loss:0.7631652355194092 lr:3.950052311308783e-05 tokens_per_second_per_gpu:4840.1097975818 mfu:22.83162372307943 +Step 20633 | loss:0.6749656200408936 lr:3.949628361282429e-05 tokens_per_second_per_gpu:4000.198504435279 mfu:18.869618849663677 +Step 20634 | loss:0.6778719425201416 lr:3.949204419156054e-05 tokens_per_second_per_gpu:4243.2463634533415 mfu:20.016117068892697 +Step 20635 | loss:0.7060160040855408 lr:3.9487804849328457e-05 tokens_per_second_per_gpu:4793.778470015507 mfu:22.61307094600961 +Step 20636 | loss:0.8097246885299683 lr:3.948356558615993e-05 tokens_per_second_per_gpu:3809.498597075105 mfu:17.970054849886456 +Step 20637 | loss:0.8425312042236328 lr:3.9479326402086856e-05 tokens_per_second_per_gpu:3044.093928534346 mfu:14.359510436877072 +Step 20638 | loss:0.6848838925361633 lr:3.947508729714111e-05 tokens_per_second_per_gpu:5320.920514939135 mfu:25.099689911621518 +Step 20639 | loss:0.6719720959663391 lr:3.9470848271354567e-05 tokens_per_second_per_gpu:4691.305231365767 mfu:22.129687195561406 +Step 20640 | loss:0.6922262907028198 lr:3.946660932475913e-05 tokens_per_second_per_gpu:4521.2262324576195 mfu:21.327395539242172 +Step 20641 | loss:0.6128811240196228 lr:3.9462370457386646e-05 tokens_per_second_per_gpu:5371.017206791551 mfu:25.33600455446637 +Step 20642 | loss:0.680583119392395 lr:3.945813166926906e-05 tokens_per_second_per_gpu:4731.568225931853 mfu:22.319614610504875 +Step 20643 | loss:0.6639183759689331 lr:3.945389296043818e-05 tokens_per_second_per_gpu:5310.483927144632 mfu:25.05045874632935 +Step 20644 | loss:0.664652407169342 lr:3.944965433092593e-05 tokens_per_second_per_gpu:4013.29779632479 mfu:18.931410444476157 +Step 20645 | loss:0.5425050854682922 lr:3.9445415780764154e-05 tokens_per_second_per_gpu:4716.185490551487 mfu:22.247051623150398 +Step 20646 | loss:0.6651123762130737 lr:3.944117730998477e-05 tokens_per_second_per_gpu:5346.752698319385 mfu:25.221544728051118 +Step 20647 | loss:0.6802605986595154 lr:3.9436938918619644e-05 tokens_per_second_per_gpu:4763.8023515876575 mfu:22.471668480932255 +Step 20648 | loss:0.6714351773262024 lr:3.943270060670063e-05 tokens_per_second_per_gpu:4269.938776449706 mfu:20.14202973519182 +Step 20649 | loss:0.7434872984886169 lr:3.942846237425966e-05 tokens_per_second_per_gpu:4638.739581166839 mfu:21.881726054948327 +Step 20650 | loss:0.6410970687866211 lr:3.942422422132853e-05 tokens_per_second_per_gpu:3661.5799789241655 mfu:17.272297490549807 +Step 20651 | loss:0.8407521843910217 lr:3.9419986147939184e-05 tokens_per_second_per_gpu:3752.249256407409 mfu:17.699999942212198 +Step 20652 | loss:0.7354092001914978 lr:3.941574815412345e-05 tokens_per_second_per_gpu:5269.150888993922 mfu:24.85548375322895 +Step 20653 | loss:0.738859236240387 lr:3.941151023991325e-05 tokens_per_second_per_gpu:3382.406517451306 mfu:15.955388640878155 +Step 20654 | loss:0.665722668170929 lr:3.940727240534042e-05 tokens_per_second_per_gpu:4289.875041729144 mfu:20.23607249999253 +Step 20655 | loss:0.7234376072883606 lr:3.940303465043684e-05 tokens_per_second_per_gpu:3140.0602248688274 mfu:14.81219983022536 +Step 20656 | loss:0.5827693343162537 lr:3.939879697523441e-05 tokens_per_second_per_gpu:4461.915912644368 mfu:21.047618641299028 +Step 20657 | loss:0.6505833864212036 lr:3.9394559379764964e-05 tokens_per_second_per_gpu:5335.423793396873 mfu:25.168104350621135 +Step 20658 | loss:0.6774829030036926 lr:3.9390321864060395e-05 tokens_per_second_per_gpu:4632.317237873817 mfu:21.851430765870383 +Step 20659 | loss:0.7124637365341187 lr:3.938608442815256e-05 tokens_per_second_per_gpu:5133.934707343844 mfu:24.217645954126013 +Step 20660 | loss:0.6457852721214294 lr:3.938184707207335e-05 tokens_per_second_per_gpu:3927.554592763147 mfu:18.526945123976834 +Step 20661 | loss:0.6437689065933228 lr:3.937760979585461e-05 tokens_per_second_per_gpu:5161.799161049831 mfu:24.349087336423192 +Step 20662 | loss:0.6510979533195496 lr:3.937337259952823e-05 tokens_per_second_per_gpu:4146.167775345464 mfu:19.55818080547275 +Step 20663 | loss:0.7354115843772888 lr:3.936913548312608e-05 tokens_per_second_per_gpu:4450.789781363837 mfu:20.995134781735022 +Step 20664 | loss:0.8567864894866943 lr:3.936489844668002e-05 tokens_per_second_per_gpu:5084.27018845807 mfu:23.98336994490112 +Step 20665 | loss:0.6594059467315674 lr:3.936066149022191e-05 tokens_per_second_per_gpu:4556.790816391689 mfu:21.495159749602905 +Step 20666 | loss:0.6098675727844238 lr:3.93564246137836e-05 tokens_per_second_per_gpu:5362.172468447726 mfu:25.294282414630576 +Step 20667 | loss:0.5769930481910706 lr:3.9352187817397e-05 tokens_per_second_per_gpu:4124.691972179859 mfu:19.456875777790735 +Step 20668 | loss:0.7046271562576294 lr:3.934795110109395e-05 tokens_per_second_per_gpu:3295.264739619965 mfu:15.544326006927289 +Step 20669 | loss:0.6948286294937134 lr:3.934371446490632e-05 tokens_per_second_per_gpu:5331.894192007832 mfu:25.151454618656935 +Step 20670 | loss:0.7510719895362854 lr:3.933947790886598e-05 tokens_per_second_per_gpu:5367.482903722891 mfu:25.319332643877217 +Step 20671 | loss:0.7601426243782043 lr:3.933524143300479e-05 tokens_per_second_per_gpu:3529.43287816441 mfu:16.648937069645715 +Step 20672 | loss:0.7891048192977905 lr:3.933100503735459e-05 tokens_per_second_per_gpu:5187.519823404895 mfu:24.47041608914954 +Step 20673 | loss:0.7933284640312195 lr:3.932676872194727e-05 tokens_per_second_per_gpu:5317.236722104674 mfu:25.082312832301497 +Step 20674 | loss:0.8035611510276794 lr:3.932253248681469e-05 tokens_per_second_per_gpu:5321.894416698341 mfu:25.104283972384465 +Step 20675 | loss:0.6590580344200134 lr:3.93182963319887e-05 tokens_per_second_per_gpu:5134.153385032074 mfu:24.218677494091658 +Step 20676 | loss:0.7824957370758057 lr:3.931406025750117e-05 tokens_per_second_per_gpu:4048.0200077272534 mfu:19.09520103988173 +Step 20677 | loss:0.7368083000183105 lr:3.9309824263383946e-05 tokens_per_second_per_gpu:5105.556486285254 mfu:24.083781043568862 +Step 20678 | loss:0.7295113801956177 lr:3.930558834966893e-05 tokens_per_second_per_gpu:4307.792535887106 mfu:20.320592376975373 +Step 20679 | loss:0.7473301887512207 lr:3.9301352516387915e-05 tokens_per_second_per_gpu:4495.599930263882 mfu:21.206511899496302 +Step 20680 | loss:0.7099813222885132 lr:3.92971167635728e-05 tokens_per_second_per_gpu:3930.22879773447 mfu:18.53955980509298 +Step 20681 | loss:0.6567994952201843 lr:3.9292881091255446e-05 tokens_per_second_per_gpu:4729.323280281231 mfu:22.309024818843668 +Step 20682 | loss:0.7798943519592285 lr:3.928864549946769e-05 tokens_per_second_per_gpu:3650.588288373599 mfu:17.22044781084699 +Step 20683 | loss:0.7279477715492249 lr:3.92844099882414e-05 tokens_per_second_per_gpu:4786.032398857913 mfu:22.576531406743218 +Step 20684 | loss:0.7288503646850586 lr:3.9280174557608416e-05 tokens_per_second_per_gpu:4398.374086913741 mfu:20.747881007974264 +Step 20685 | loss:0.6835744976997375 lr:3.927593920760065e-05 tokens_per_second_per_gpu:5372.157081504193 mfu:25.341381537968587 +Step 20686 | loss:0.7171752452850342 lr:3.927170393824987e-05 tokens_per_second_per_gpu:3897.3581063377296 mfu:18.384503145456296 +Step 20687 | loss:0.5412802696228027 lr:3.9267468749587996e-05 tokens_per_second_per_gpu:3542.0867345691504 mfu:16.708627469277406 +Step 20688 | loss:0.772925615310669 lr:3.926323364164684e-05 tokens_per_second_per_gpu:4544.755729622828 mfu:21.438388191916644 +Step 20689 | loss:0.8442180156707764 lr:3.9258998614458285e-05 tokens_per_second_per_gpu:4108.00407765176 mfu:19.378156132053284 +Step 20690 | loss:0.8113856315612793 lr:3.925476366805418e-05 tokens_per_second_per_gpu:4174.649932006878 mfu:19.692535998001667 +Step 20691 | loss:0.766173779964447 lr:3.925052880246635e-05 tokens_per_second_per_gpu:4315.708910585485 mfu:20.357935267100423 +Step 20692 | loss:0.8743982911109924 lr:3.9246294017726686e-05 tokens_per_second_per_gpu:4760.829616090082 mfu:22.45764558038921 +Step 20693 | loss:0.5995559692382812 lr:3.924205931386701e-05 tokens_per_second_per_gpu:4210.947887621457 mfu:19.863759647705475 +Step 20694 | loss:0.7444787621498108 lr:3.9237824690919176e-05 tokens_per_second_per_gpu:3876.461061585154 mfu:18.285928219954908 +Step 20695 | loss:0.6799732446670532 lr:3.923359014891503e-05 tokens_per_second_per_gpu:4706.427144988334 mfu:22.201019842183566 +Step 20696 | loss:0.9210025668144226 lr:3.922935568788643e-05 tokens_per_second_per_gpu:5357.548190198095 mfu:25.27246890514384 +Step 20697 | loss:0.6590474247932434 lr:3.922512130786521e-05 tokens_per_second_per_gpu:4201.688175502491 mfu:19.82008000577051 +Step 20698 | loss:0.8638989925384521 lr:3.922088700888324e-05 tokens_per_second_per_gpu:4676.479910725952 mfu:22.05975362864287 +Step 20699 | loss:0.8880529999732971 lr:3.921665279097237e-05 tokens_per_second_per_gpu:5078.590209530426 mfu:23.95657651519509 +Step 20700 | loss:0.633796751499176 lr:3.921241865416442e-05 tokens_per_second_per_gpu:5351.481046830478 mfu:25.243849154717388 +Step 20701 | loss:0.8580361604690552 lr:3.920818459849124e-05 tokens_per_second_per_gpu:3836.968916435493 mfu:18.09963703322919 +Step 20702 | loss:0.6855160593986511 lr:3.920395062398467e-05 tokens_per_second_per_gpu:3968.7857235338693 mfu:18.721439402578643 +Step 20703 | loss:0.6859022974967957 lr:3.9199716730676574e-05 tokens_per_second_per_gpu:5364.4959050310235 mfu:25.305242461413147 +Step 20704 | loss:0.8425206542015076 lr:3.919548291859878e-05 tokens_per_second_per_gpu:3594.149666153355 mfu:16.9542172004122 +Step 20705 | loss:0.6570317149162292 lr:3.9191249187783145e-05 tokens_per_second_per_gpu:4901.65129838307 mfu:23.121925482421787 +Step 20706 | loss:0.80641770362854 lr:3.918701553826151e-05 tokens_per_second_per_gpu:5146.692985425623 mfu:24.27782892862447 +Step 20707 | loss:0.6919334530830383 lr:3.918278197006572e-05 tokens_per_second_per_gpu:4094.655050497317 mfu:19.315186493386893 +Step 20708 | loss:0.5994741916656494 lr:3.9178548483227576e-05 tokens_per_second_per_gpu:5296.839245718775 mfu:24.986094456020872 +Step 20709 | loss:0.6999292969703674 lr:3.917431507777897e-05 tokens_per_second_per_gpu:5329.680808203058 mfu:25.1410137096076 +Step 20710 | loss:0.7010311484336853 lr:3.917008175375172e-05 tokens_per_second_per_gpu:4582.317851171922 mfu:21.615575128022503 +Step 20711 | loss:0.5752218961715698 lr:3.9165848511177656e-05 tokens_per_second_per_gpu:4709.647149885374 mfu:22.21620915467239 +Step 20712 | loss:0.6966099739074707 lr:3.916161535008864e-05 tokens_per_second_per_gpu:3784.514889448406 mfu:17.85220243835124 +Step 20713 | loss:0.6752634644508362 lr:3.915738227051648e-05 tokens_per_second_per_gpu:3700.083708960145 mfu:17.453926154543282 +Step 20714 | loss:0.8098102807998657 lr:3.915314927249307e-05 tokens_per_second_per_gpu:4862.269469970255 mfu:22.936154678565746 +Step 20715 | loss:0.6801052689552307 lr:3.914891635605018e-05 tokens_per_second_per_gpu:4684.030614210153 mfu:22.095371585260704 +Step 20716 | loss:0.7737764120101929 lr:3.914468352121968e-05 tokens_per_second_per_gpu:4431.694228498609 mfu:20.905057800832463 +Step 20717 | loss:0.7674441933631897 lr:3.91404507680334e-05 tokens_per_second_per_gpu:5344.606098015256 mfu:25.211418848168766 +Step 20718 | loss:0.846638023853302 lr:3.913621809652317e-05 tokens_per_second_per_gpu:5321.227431530954 mfu:25.101137689550498 +Step 20719 | loss:0.5771493315696716 lr:3.9131985506720835e-05 tokens_per_second_per_gpu:5210.16479986679 mfu:24.577236306751693 +Step 20720 | loss:0.8198699355125427 lr:3.912775299865821e-05 tokens_per_second_per_gpu:4613.804629810239 mfu:21.764103635057847 +Step 20721 | loss:0.8763787746429443 lr:3.912352057236718e-05 tokens_per_second_per_gpu:5363.968282511636 mfu:25.302753575967515 +Step 20722 | loss:0.6986539363861084 lr:3.911928822787951e-05 tokens_per_second_per_gpu:4677.119847837268 mfu:22.062772321181477 +Step 20723 | loss:0.7604140639305115 lr:3.911505596522707e-05 tokens_per_second_per_gpu:4696.543922780302 mfu:22.154398996776692 +Step 20724 | loss:0.74058598279953 lr:3.9110823784441666e-05 tokens_per_second_per_gpu:5282.564206130351 mfu:24.918756658709373 +Step 20725 | loss:0.7004632949829102 lr:3.910659168555516e-05 tokens_per_second_per_gpu:3535.413049667356 mfu:16.677146558947584 +Step 20726 | loss:0.67974454164505 lr:3.9102359668599366e-05 tokens_per_second_per_gpu:4188.323300226023 mfu:19.757035608808557 +Step 20727 | loss:0.630311131477356 lr:3.909812773360611e-05 tokens_per_second_per_gpu:5372.853976456225 mfu:25.344668910359857 +Step 20728 | loss:0.5906847715377808 lr:3.909389588060725e-05 tokens_per_second_per_gpu:5366.749879599466 mfu:25.31587484402021 +Step 20729 | loss:0.7185357809066772 lr:3.908966410963458e-05 tokens_per_second_per_gpu:4460.773762258302 mfu:21.042230923057 +Step 20730 | loss:0.7311264872550964 lr:3.908543242071993e-05 tokens_per_second_per_gpu:5387.287265957086 mfu:25.412753199508536 +Step 20731 | loss:0.7964656352996826 lr:3.908120081389513e-05 tokens_per_second_per_gpu:5353.639679737097 mfu:25.254031794438664 +Step 20732 | loss:0.7772998213768005 lr:3.9076969289192024e-05 tokens_per_second_per_gpu:4104.0675967094585 mfu:19.359587079815547 +Step 20733 | loss:0.7280954122543335 lr:3.907273784664242e-05 tokens_per_second_per_gpu:5331.393729169669 mfu:25.149093850061558 +Step 20734 | loss:0.6620684266090393 lr:3.906850648627814e-05 tokens_per_second_per_gpu:3964.0228920673835 mfu:18.698972313928383 +Step 20735 | loss:0.8087894916534424 lr:3.906427520813103e-05 tokens_per_second_per_gpu:4174.807958886178 mfu:19.69328143775349 +Step 20736 | loss:0.8111847043037415 lr:3.906004401223291e-05 tokens_per_second_per_gpu:3665.507625893542 mfu:17.29082487143003 +Step 20737 | loss:0.681233286857605 lr:3.905581289861559e-05 tokens_per_second_per_gpu:4393.101036527683 mfu:20.723007129628066 +Step 20738 | loss:0.6571052670478821 lr:3.905158186731088e-05 tokens_per_second_per_gpu:4967.378970129681 mfu:23.431974124347732 +Step 20739 | loss:0.8319395780563354 lr:3.904735091835064e-05 tokens_per_second_per_gpu:3897.9215423028672 mfu:18.387160969035936 +Step 20740 | loss:0.6457939743995667 lr:3.904312005176665e-05 tokens_per_second_per_gpu:3842.7440956704363 mfu:18.1268795390276 +Step 20741 | loss:0.7548813819885254 lr:3.903888926759077e-05 tokens_per_second_per_gpu:4672.614071777242 mfu:22.041517806741666 +Step 20742 | loss:0.8158466219902039 lr:3.9034658565854796e-05 tokens_per_second_per_gpu:4810.382674682094 mfu:22.691395812391345 +Step 20743 | loss:0.8907956480979919 lr:3.9030427946590576e-05 tokens_per_second_per_gpu:3850.7915411916165 mfu:18.164840712586795 +Step 20744 | loss:0.7858936190605164 lr:3.902619740982988e-05 tokens_per_second_per_gpu:5331.711173418934 mfu:25.150591288746913 +Step 20745 | loss:0.7685269117355347 lr:3.902196695560456e-05 tokens_per_second_per_gpu:4245.782557382831 mfu:20.028080728377404 +Step 20746 | loss:0.8855022192001343 lr:3.901773658394643e-05 tokens_per_second_per_gpu:5368.128890926866 mfu:25.32237987573515 +Step 20747 | loss:0.6816098690032959 lr:3.9013506294887295e-05 tokens_per_second_per_gpu:3649.7821632590117 mfu:17.216645181142358 +Step 20748 | loss:0.7190325260162354 lr:3.900927608845899e-05 tokens_per_second_per_gpu:4009.5428464839506 mfu:18.913697705416414 +Step 20749 | loss:0.6913558840751648 lr:3.900504596469332e-05 tokens_per_second_per_gpu:5138.2088795998525 mfu:24.23780795390659 +Step 20750 | loss:0.5923203229904175 lr:3.9000815923622116e-05 tokens_per_second_per_gpu:4771.554337820864 mfu:22.508235922619527 +Step 20751 | loss:0.6948351860046387 lr:3.899658596527715e-05 tokens_per_second_per_gpu:4314.1864123083 mfu:20.350753382961955 +Step 20752 | loss:0.6452639102935791 lr:3.899235608969029e-05 tokens_per_second_per_gpu:5340.763749705813 mfu:25.193293835620903 +Step 20753 | loss:0.6167789101600647 lr:3.898812629689332e-05 tokens_per_second_per_gpu:4687.360182448868 mfu:22.111077726725643 +Step 20754 | loss:0.666788637638092 lr:3.898389658691804e-05 tokens_per_second_per_gpu:4397.5412736429935 mfu:20.74395248568355 +Step 20755 | loss:0.6755480170249939 lr:3.89796669597963e-05 tokens_per_second_per_gpu:4020.7454797340934 mfu:18.966542437823666 +Step 20756 | loss:0.8761340379714966 lr:3.897543741555987e-05 tokens_per_second_per_gpu:3435.26870253228 mfu:16.204748587123916 +Step 20757 | loss:0.6526839733123779 lr:3.8971207954240616e-05 tokens_per_second_per_gpu:3837.5710048704723 mfu:18.102477187103865 +Step 20758 | loss:0.665145218372345 lr:3.8966978575870285e-05 tokens_per_second_per_gpu:5250.852318162714 mfu:24.769166272560025 +Step 20759 | loss:0.7331212162971497 lr:3.896274928048072e-05 tokens_per_second_per_gpu:5372.487522531517 mfu:25.342940284673524 +Step 20760 | loss:0.7438572645187378 lr:3.8958520068103724e-05 tokens_per_second_per_gpu:5352.5326337373735 mfu:25.248809669576445 +Step 20761 | loss:0.7101022601127625 lr:3.895429093877112e-05 tokens_per_second_per_gpu:4921.702816762729 mfu:23.21651191576045 +Step 20762 | loss:0.6106287837028503 lr:3.89500618925147e-05 tokens_per_second_per_gpu:4536.393600267733 mfu:21.398942601022274 +Step 20763 | loss:0.7532925009727478 lr:3.894583292936626e-05 tokens_per_second_per_gpu:5016.098622448096 mfu:23.661793036763534 +Step 20764 | loss:0.6704873442649841 lr:3.894160404935765e-05 tokens_per_second_per_gpu:4453.7246246278055 mfu:21.008978938147052 +Step 20765 | loss:0.6689882278442383 lr:3.893737525252062e-05 tokens_per_second_per_gpu:4806.395079407775 mfu:22.672585645128468 +Step 20766 | loss:0.6919475197792053 lr:3.8933146538887024e-05 tokens_per_second_per_gpu:5337.490925903552 mfu:25.177855367344314 +Step 20767 | loss:0.647435188293457 lr:3.892891790848862e-05 tokens_per_second_per_gpu:4495.824443287221 mfu:21.20757096573326 +Step 20768 | loss:0.6630178093910217 lr:3.892468936135725e-05 tokens_per_second_per_gpu:4868.448679192327 mfu:22.96530306274807 +Step 20769 | loss:0.8453606963157654 lr:3.8920460897524715e-05 tokens_per_second_per_gpu:5312.22760411318 mfu:25.05868396809196 +Step 20770 | loss:0.7214326858520508 lr:3.8916232517022785e-05 tokens_per_second_per_gpu:4511.345899163239 mfu:21.28078832133389 +Step 20771 | loss:0.6865280866622925 lr:3.8912004219883296e-05 tokens_per_second_per_gpu:5343.922095372999 mfu:25.208192291002355 +Step 20772 | loss:0.8390166163444519 lr:3.890777600613805e-05 tokens_per_second_per_gpu:3962.4442736947813 mfu:18.691525701724753 +Step 20773 | loss:0.7357926368713379 lr:3.890354787581883e-05 tokens_per_second_per_gpu:4612.157519151671 mfu:21.756333933055327 +Step 20774 | loss:0.6480919718742371 lr:3.889931982895743e-05 tokens_per_second_per_gpu:4207.969898081751 mfu:19.849711962949275 +Step 20775 | loss:0.8153197765350342 lr:3.889509186558567e-05 tokens_per_second_per_gpu:3989.3835135133645 mfu:18.81860269225728 +Step 20776 | loss:0.6539557576179504 lr:3.889086398573533e-05 tokens_per_second_per_gpu:4160.501470204382 mfu:19.625795289702786 +Step 20777 | loss:0.6275160312652588 lr:3.888663618943823e-05 tokens_per_second_per_gpu:4779.919654575125 mfu:22.547696549019587 +Step 20778 | loss:0.6499025821685791 lr:3.8882408476726156e-05 tokens_per_second_per_gpu:4866.461070146366 mfu:22.9559271717572 +Step 20779 | loss:0.7651735544204712 lr:3.887818084763092e-05 tokens_per_second_per_gpu:4965.544654500122 mfu:23.423321344556587 +Step 20780 | loss:0.7402747869491577 lr:3.887395330218429e-05 tokens_per_second_per_gpu:3687.581640402217 mfu:17.394951764082684 +Step 20781 | loss:0.7924962043762207 lr:3.8869725840418074e-05 tokens_per_second_per_gpu:3957.739793328712 mfu:18.669333865170884 +Step 20782 | loss:0.6782947182655334 lr:3.886549846236408e-05 tokens_per_second_per_gpu:3542.2848050203943 mfu:16.709561801390226 +Step 20783 | loss:0.6235531568527222 lr:3.886127116805408e-05 tokens_per_second_per_gpu:3685.85920056618 mfu:17.386826721497556 +Step 20784 | loss:0.7027390003204346 lr:3.885704395751989e-05 tokens_per_second_per_gpu:3481.0677955175943 mfu:16.420790722866673 +Step 20785 | loss:0.7793043851852417 lr:3.885281683079329e-05 tokens_per_second_per_gpu:4708.486874780172 mfu:22.210735939037807 +Step 20786 | loss:0.7338409423828125 lr:3.88485897879061e-05 tokens_per_second_per_gpu:4900.218627922766 mfu:23.115167331421752 +Step 20787 | loss:0.674071729183197 lr:3.884436282889006e-05 tokens_per_second_per_gpu:4874.113089838351 mfu:22.992023054213476 +Step 20788 | loss:0.6783833503723145 lr:3.8840135953777e-05 tokens_per_second_per_gpu:3821.831329330855 mfu:18.028230452117928 +Step 20789 | loss:0.8091340065002441 lr:3.8835909162598704e-05 tokens_per_second_per_gpu:4468.121618754516 mfu:21.07689201582324 +Step 20790 | loss:0.8531992435455322 lr:3.883168245538695e-05 tokens_per_second_per_gpu:3778.1909007291665 mfu:17.822371104578853 +Step 20791 | loss:0.6588452458381653 lr:3.882745583217354e-05 tokens_per_second_per_gpu:5364.252480115339 mfu:25.30409418455329 +Step 20792 | loss:0.7912646532058716 lr:3.882322929299026e-05 tokens_per_second_per_gpu:3762.6978717627453 mfu:17.749287843560925 +Step 20793 | loss:0.8616596460342407 lr:3.8819002837868926e-05 tokens_per_second_per_gpu:4357.209302085159 mfu:20.55369969450132 +Step 20794 | loss:0.7324446439743042 lr:3.881477646684127e-05 tokens_per_second_per_gpu:5288.644430447014 mfu:24.94743811420432 +Step 20795 | loss:0.6015248894691467 lr:3.881055017993911e-05 tokens_per_second_per_gpu:4123.096470901183 mfu:19.449349526038635 +Step 20796 | loss:0.672773003578186 lr:3.880632397719423e-05 tokens_per_second_per_gpu:4317.478142990389 mfu:20.366281038216535 +Step 20797 | loss:0.6877455115318298 lr:3.880209785863842e-05 tokens_per_second_per_gpu:5338.455252780524 mfu:25.18240426175412 +Step 20798 | loss:0.8053362965583801 lr:3.879787182430346e-05 tokens_per_second_per_gpu:4148.873891331499 mfu:19.570946016289877 +Step 20799 | loss:0.68778395652771 lr:3.879364587422113e-05 tokens_per_second_per_gpu:5238.105924725508 mfu:24.709039359957746 +Step 20800 | loss:0.8081476092338562 lr:3.878942000842322e-05 tokens_per_second_per_gpu:4348.418323530621 mfu:20.512231148763956 +Step 20801 | loss:0.6570088863372803 lr:3.878519422694153e-05 tokens_per_second_per_gpu:4960.1473699001235 mfu:23.39786143222668 +Step 20802 | loss:0.6005465984344482 lr:3.878096852980781e-05 tokens_per_second_per_gpu:5200.916114631267 mfu:24.533608680507314 +Step 20803 | loss:0.8504457473754883 lr:3.877674291705386e-05 tokens_per_second_per_gpu:3843.6372754264344 mfu:18.13109282032882 +Step 20804 | loss:0.7722464799880981 lr:3.877251738871146e-05 tokens_per_second_per_gpu:5328.716318576816 mfu:25.136464047481716 +Step 20805 | loss:0.67448490858078 lr:3.8768291944812386e-05 tokens_per_second_per_gpu:5308.278705435634 mfu:25.040056339278387 +Step 20806 | loss:0.6489537954330444 lr:3.876406658538842e-05 tokens_per_second_per_gpu:4180.143601994142 mfu:19.718450576648344 +Step 20807 | loss:0.7178069353103638 lr:3.875984131047135e-05 tokens_per_second_per_gpu:5351.291925523804 mfu:25.242957037246388 +Step 20808 | loss:0.7785505056381226 lr:3.875561612009295e-05 tokens_per_second_per_gpu:5340.856039077936 mfu:25.193729180335392 +Step 20809 | loss:0.7382484078407288 lr:3.875139101428499e-05 tokens_per_second_per_gpu:4924.571041597727 mfu:23.23004182167676 +Step 20810 | loss:0.8120055794715881 lr:3.874716599307925e-05 tokens_per_second_per_gpu:5240.663647727355 mfu:24.72110457575 +Step 20811 | loss:0.7531436681747437 lr:3.874294105650751e-05 tokens_per_second_per_gpu:4258.365162738895 mfu:20.0874350246558 +Step 20812 | loss:0.6617141366004944 lr:3.8738716204601546e-05 tokens_per_second_per_gpu:4306.992430110279 mfu:20.31681814151396 +Step 20813 | loss:0.7303169369697571 lr:3.873449143739314e-05 tokens_per_second_per_gpu:4810.07210268636 mfu:22.689930791298423 +Step 20814 | loss:0.6592293381690979 lr:3.873026675491406e-05 tokens_per_second_per_gpu:4674.6813939826625 mfu:22.051269718306177 +Step 20815 | loss:0.7688481211662292 lr:3.87260421571961e-05 tokens_per_second_per_gpu:5012.978096514659 mfu:23.64707298351889 +Step 20816 | loss:0.7051196098327637 lr:3.8721817644270985e-05 tokens_per_second_per_gpu:3346.1793258192115 mfu:15.784498796953308 +Step 20817 | loss:0.6936320662498474 lr:3.871759321617053e-05 tokens_per_second_per_gpu:4373.501516129259 mfu:20.630552847885692 +Step 20818 | loss:0.629830002784729 lr:3.87133688729265e-05 tokens_per_second_per_gpu:4694.091079640485 mfu:22.14282851718798 +Step 20819 | loss:0.7015150785446167 lr:3.870914461457065e-05 tokens_per_second_per_gpu:3313.8207399433186 mfu:15.63185782643267 +Step 20820 | loss:0.7568038105964661 lr:3.870492044113477e-05 tokens_per_second_per_gpu:4876.139506647279 mfu:23.001582008043545 +Step 20821 | loss:0.6677132248878479 lr:3.870069635265064e-05 tokens_per_second_per_gpu:5152.885655567416 mfu:24.30704081026178 +Step 20822 | loss:0.7054693102836609 lr:3.869647234915002e-05 tokens_per_second_per_gpu:5275.167330476287 mfu:24.88386433421203 +Step 20823 | loss:0.7248874306678772 lr:3.869224843066464e-05 tokens_per_second_per_gpu:5325.498740265566 mfu:25.12128618161208 +Step 20824 | loss:0.6642741560935974 lr:3.868802459722633e-05 tokens_per_second_per_gpu:5267.255697660334 mfu:24.84654381235543 +Step 20825 | loss:0.6329737901687622 lr:3.868380084886681e-05 tokens_per_second_per_gpu:5366.974331699271 mfu:25.31693362287047 +Step 20826 | loss:0.7330007553100586 lr:3.867957718561787e-05 tokens_per_second_per_gpu:4129.384242503956 mfu:19.479010017493373 +Step 20827 | loss:0.6536526083946228 lr:3.867535360751129e-05 tokens_per_second_per_gpu:4864.485579116097 mfu:22.94660844351389 +Step 20828 | loss:0.7779584527015686 lr:3.86711301145788e-05 tokens_per_second_per_gpu:3620.7510829157986 mfu:17.079700621950355 +Step 20829 | loss:0.7266855835914612 lr:3.8666906706852204e-05 tokens_per_second_per_gpu:3236.16927223633 mfu:15.26556260455225 +Step 20830 | loss:0.7124608755111694 lr:3.8662683384363255e-05 tokens_per_second_per_gpu:5341.640167723772 mfu:25.197428048943948 +Step 20831 | loss:0.6955832242965698 lr:3.86584601471437e-05 tokens_per_second_per_gpu:5328.561849022981 mfu:25.135735388241052 +Step 20832 | loss:0.6899872422218323 lr:3.8654236995225304e-05 tokens_per_second_per_gpu:4243.333275829096 mfu:20.016527049397908 +Step 20833 | loss:0.6293532848358154 lr:3.865001392863985e-05 tokens_per_second_per_gpu:4577.587998741423 mfu:21.5932636071118 +Step 20834 | loss:0.6349613070487976 lr:3.864579094741909e-05 tokens_per_second_per_gpu:5341.033679032717 mfu:25.1945671383106 +Step 20835 | loss:0.6200142502784729 lr:3.8641568051594774e-05 tokens_per_second_per_gpu:5140.476024217255 mfu:24.24850246188164 +Step 20836 | loss:0.7239780426025391 lr:3.863734524119868e-05 tokens_per_second_per_gpu:4745.947055982338 mfu:22.387442005134396 +Step 20837 | loss:0.7787951827049255 lr:3.8633122516262585e-05 tokens_per_second_per_gpu:5393.769209095138 mfu:25.44332963122427 +Step 20838 | loss:0.6532449126243591 lr:3.8628899876818206e-05 tokens_per_second_per_gpu:3619.796041847639 mfu:17.07519553028504 +Step 20839 | loss:0.7454155087471008 lr:3.862467732289731e-05 tokens_per_second_per_gpu:5366.564593752285 mfu:25.31500081906603 +Step 20840 | loss:0.7135331034660339 lr:3.8620454854531684e-05 tokens_per_second_per_gpu:4849.3737936592925 mfu:22.875323573177678 +Step 20841 | loss:0.6796097159385681 lr:3.861623247175307e-05 tokens_per_second_per_gpu:4052.02228223938 mfu:19.114080451613876 +Step 20842 | loss:0.741830587387085 lr:3.8612010174593206e-05 tokens_per_second_per_gpu:4830.6447015163985 mfu:22.786975250027016 +Step 20843 | loss:0.7289479970932007 lr:3.860778796308388e-05 tokens_per_second_per_gpu:5356.280489659066 mfu:25.266488945409225 +Step 20844 | loss:0.6265658736228943 lr:3.860356583725685e-05 tokens_per_second_per_gpu:5147.508436298488 mfu:24.28167554952204 +Step 20845 | loss:0.8768872022628784 lr:3.859934379714384e-05 tokens_per_second_per_gpu:5308.983763912054 mfu:25.043382220407043 +Step 20846 | loss:0.5221172571182251 lr:3.85951218427766e-05 tokens_per_second_per_gpu:4862.091979381397 mfu:22.93531742517509 +Step 20847 | loss:0.6939701437950134 lr:3.859089997418693e-05 tokens_per_second_per_gpu:4372.8091643248645 mfu:20.627286906296963 +Step 20848 | loss:0.6211948990821838 lr:3.858667819140654e-05 tokens_per_second_per_gpu:4624.391844858728 mfu:21.814045334806725 +Step 20849 | loss:0.6999911665916443 lr:3.858245649446721e-05 tokens_per_second_per_gpu:4637.262825989492 mfu:21.874759948816365 +Step 20850 | loss:0.5925247669219971 lr:3.857823488340069e-05 tokens_per_second_per_gpu:4961.333884037612 mfu:23.403458421853404 +Step 20851 | loss:0.6090667247772217 lr:3.8574013358238723e-05 tokens_per_second_per_gpu:5321.546050904139 mfu:25.10264067149588 +Step 20852 | loss:0.8129281401634216 lr:3.856979191901304e-05 tokens_per_second_per_gpu:4524.389732216222 mfu:21.34231830735222 +Step 20853 | loss:0.8000139594078064 lr:3.856557056575542e-05 tokens_per_second_per_gpu:4231.529854587291 mfu:19.960848297528987 +Step 20854 | loss:0.7138699889183044 lr:3.856134929849761e-05 tokens_per_second_per_gpu:5352.139902308708 mfu:25.246957088420537 +Step 20855 | loss:0.9201133251190186 lr:3.855712811727133e-05 tokens_per_second_per_gpu:3426.775309360381 mfu:16.164683802409762 +Step 20856 | loss:0.8029054403305054 lr:3.8552907022108374e-05 tokens_per_second_per_gpu:5340.006010799215 mfu:25.189719451914886 +Step 20857 | loss:0.7600432634353638 lr:3.854868601304046e-05 tokens_per_second_per_gpu:5351.993647284798 mfu:25.24626717851907 +Step 20858 | loss:0.7130906581878662 lr:3.854446509009932e-05 tokens_per_second_per_gpu:5106.3624466091605 mfu:24.08758289592774 +Step 20859 | loss:0.5966958403587341 lr:3.854024425331675e-05 tokens_per_second_per_gpu:5250.016128729861 mfu:24.765221824336624 +Step 20860 | loss:0.6268712878227234 lr:3.853602350272446e-05 tokens_per_second_per_gpu:4812.162814908583 mfu:22.699793037562934 +Step 20861 | loss:0.6592803001403809 lr:3.8531802838354195e-05 tokens_per_second_per_gpu:4947.997784741376 mfu:23.340549766099887 +Step 20862 | loss:0.7311016321182251 lr:3.8527582260237684e-05 tokens_per_second_per_gpu:4832.4278917577785 mfu:22.795386862639237 +Step 20863 | loss:0.8080299496650696 lr:3.8523361768406705e-05 tokens_per_second_per_gpu:5161.631713318995 mfu:24.348297456907307 +Step 20864 | loss:0.8853445053100586 lr:3.8519141362892985e-05 tokens_per_second_per_gpu:4279.222428702813 mfu:20.18582230682413 +Step 20865 | loss:0.86252760887146 lr:3.851492104372827e-05 tokens_per_second_per_gpu:3652.323698645078 mfu:17.228634037189085 +Step 20866 | loss:0.6865944862365723 lr:3.851070081094432e-05 tokens_per_second_per_gpu:4240.701043077141 mfu:20.004110358401718 +Step 20867 | loss:0.7581354975700378 lr:3.850648066457283e-05 tokens_per_second_per_gpu:4744.673412119554 mfu:22.38143400972759 +Step 20868 | loss:0.5257465243339539 lr:3.850226060464557e-05 tokens_per_second_per_gpu:5059.546122715054 mfu:23.866742308430712 +Step 20869 | loss:0.5999682545661926 lr:3.8498040631194274e-05 tokens_per_second_per_gpu:3424.547431192609 mfu:16.154174520977172 +Step 20870 | loss:0.756250262260437 lr:3.8493820744250685e-05 tokens_per_second_per_gpu:3729.897144845202 mfu:17.59456121830978 +Step 20871 | loss:0.9166224002838135 lr:3.848960094384653e-05 tokens_per_second_per_gpu:5321.422100561996 mfu:25.102055976583884 +Step 20872 | loss:0.7714425921440125 lr:3.848538123001356e-05 tokens_per_second_per_gpu:4239.235865586677 mfu:19.997198866194143 +Step 20873 | loss:0.6155256032943726 lr:3.848116160278353e-05 tokens_per_second_per_gpu:5320.027732805643 mfu:25.095478505973496 +Step 20874 | loss:0.7065587043762207 lr:3.8476942062188135e-05 tokens_per_second_per_gpu:4508.983096952347 mfu:21.269642580169485 +Step 20875 | loss:0.8755663633346558 lr:3.847272260825912e-05 tokens_per_second_per_gpu:4344.041798330945 mfu:20.49158633268489 +Step 20876 | loss:0.748782753944397 lr:3.8468503241028234e-05 tokens_per_second_per_gpu:4720.0442971402035 mfu:22.265254272209667 +Step 20877 | loss:0.6181894540786743 lr:3.846428396052721e-05 tokens_per_second_per_gpu:3798.0849803435267 mfu:17.91621487239978 +Step 20878 | loss:0.6940197944641113 lr:3.8460064766787766e-05 tokens_per_second_per_gpu:4440.313984164669 mfu:20.945718658991236 +Step 20879 | loss:0.7160793542861938 lr:3.845584565984166e-05 tokens_per_second_per_gpu:3435.8400642701044 mfu:16.2074437979256 +Step 20880 | loss:0.6437574028968811 lr:3.845162663972063e-05 tokens_per_second_per_gpu:5335.534726168274 mfu:25.168627639430916 +Step 20881 | loss:0.7744702696800232 lr:3.844740770645637e-05 tokens_per_second_per_gpu:3846.676061606712 mfu:18.145427293211576 +Step 20882 | loss:0.7335307002067566 lr:3.844318886008062e-05 tokens_per_second_per_gpu:4413.283387831088 mfu:20.81821072419043 +Step 20883 | loss:0.6584743857383728 lr:3.843897010062513e-05 tokens_per_second_per_gpu:4728.158516786523 mfu:22.303530430709987 +Step 20884 | loss:0.7160120606422424 lr:3.8434751428121615e-05 tokens_per_second_per_gpu:4499.882772593085 mfu:21.226714797491404 +Step 20885 | loss:0.8255428671836853 lr:3.8430532842601825e-05 tokens_per_second_per_gpu:4547.087147954405 mfu:21.449385889967676 +Step 20886 | loss:0.8617746233940125 lr:3.842631434409747e-05 tokens_per_second_per_gpu:4858.81005235478 mfu:22.919836015436044 +Step 20887 | loss:0.7552651166915894 lr:3.842209593264029e-05 tokens_per_second_per_gpu:5324.839367785638 mfu:25.118175809123716 +Step 20888 | loss:0.7447726130485535 lr:3.841787760826199e-05 tokens_per_second_per_gpu:4643.476330132605 mfu:21.90407010799255 +Step 20889 | loss:0.872624933719635 lr:3.841365937099431e-05 tokens_per_second_per_gpu:5313.65716852448 mfu:25.065427467328902 +Step 20890 | loss:0.7283716797828674 lr:3.840944122086898e-05 tokens_per_second_per_gpu:5354.45758438187 mfu:25.257889990196983 +Step 20891 | loss:0.8538092970848083 lr:3.840522315791771e-05 tokens_per_second_per_gpu:3914.0733551166713 mfu:18.463351825862084 +Step 20892 | loss:0.7732400298118591 lr:3.840100518217225e-05 tokens_per_second_per_gpu:3436.617442093792 mfu:16.211110821754264 +Step 20893 | loss:0.794744610786438 lr:3.839678729366431e-05 tokens_per_second_per_gpu:4260.571381732038 mfu:20.097842136066234 +Step 20894 | loss:0.7609188556671143 lr:3.83925694924256e-05 tokens_per_second_per_gpu:3328.414184720959 mfu:15.700697595347227 +Step 20895 | loss:0.6439482569694519 lr:3.838835177848788e-05 tokens_per_second_per_gpu:4604.104768054463 mfu:21.718347732188786 +Step 20896 | loss:0.6807750463485718 lr:3.838413415188284e-05 tokens_per_second_per_gpu:4184.001205504607 mfu:19.736647550582237 +Step 20897 | loss:0.8657233715057373 lr:3.8379916612642206e-05 tokens_per_second_per_gpu:4195.911530447179 mfu:19.79283058544733 +Step 20898 | loss:0.6964291334152222 lr:3.837569916079768e-05 tokens_per_second_per_gpu:4905.117240083716 mfu:23.138274920774364 +Step 20899 | loss:0.8192834258079529 lr:3.837148179638103e-05 tokens_per_second_per_gpu:3167.0549331526177 mfu:14.939538474972165 +Step 20900 | loss:0.7170366048812866 lr:3.8367264519423937e-05 tokens_per_second_per_gpu:5344.014329972337 mfu:25.208627377344275 +Step 20901 | loss:0.6797906756401062 lr:3.8363047329958134e-05 tokens_per_second_per_gpu:4222.330148086939 mfu:19.91745171233504 +Step 20902 | loss:0.653107225894928 lr:3.835883022801536e-05 tokens_per_second_per_gpu:4522.311606476554 mfu:21.332515433673386 +Step 20903 | loss:0.7354245781898499 lr:3.835461321362729e-05 tokens_per_second_per_gpu:4286.355328949759 mfu:20.219469414287246 +Step 20904 | loss:0.798858106136322 lr:3.835039628682565e-05 tokens_per_second_per_gpu:4818.33579922996 mfu:22.7289120578266 +Step 20905 | loss:0.8285627961158752 lr:3.834617944764218e-05 tokens_per_second_per_gpu:4048.634773128832 mfu:19.09810099317077 +Step 20906 | loss:0.7299601435661316 lr:3.834196269610857e-05 tokens_per_second_per_gpu:4229.869473274041 mfu:19.953015995584124 +Step 20907 | loss:0.8713226914405823 lr:3.833774603225655e-05 tokens_per_second_per_gpu:3029.017599142364 mfu:14.288392818848052 +Step 20908 | loss:0.9273546934127808 lr:3.833352945611783e-05 tokens_per_second_per_gpu:5295.5651860154885 mfu:24.980084499023473 +Step 20909 | loss:0.7006670236587524 lr:3.832931296772414e-05 tokens_per_second_per_gpu:4038.173862834547 mfu:19.048755094497285 +Step 20910 | loss:0.6536594033241272 lr:3.832509656710717e-05 tokens_per_second_per_gpu:5237.5167047926825 mfu:24.70625990900332 +Step 20911 | loss:0.7640746831893921 lr:3.832088025429862e-05 tokens_per_second_per_gpu:4695.535391793337 mfu:22.149641584038047 +Step 20912 | loss:0.9095328450202942 lr:3.831666402933024e-05 tokens_per_second_per_gpu:5358.33210680118 mfu:25.276166773510557 +Step 20913 | loss:0.7140284776687622 lr:3.831244789223371e-05 tokens_per_second_per_gpu:4485.5131918887255 mfu:21.158931033605658 +Step 20914 | loss:0.7570833563804626 lr:3.8308231843040744e-05 tokens_per_second_per_gpu:3877.9452873323416 mfu:18.29292956603942 +Step 20915 | loss:0.5815375447273254 lr:3.8304015881783076e-05 tokens_per_second_per_gpu:5356.178687471522 mfu:25.266008727121335 +Step 20916 | loss:0.758213460445404 lr:3.8299800008492395e-05 tokens_per_second_per_gpu:5320.118624754626 mfu:25.095907258804615 +Step 20917 | loss:0.6763688325881958 lr:3.829558422320041e-05 tokens_per_second_per_gpu:4543.035235916019 mfu:21.43027232955513 +Step 20918 | loss:0.6567946076393127 lr:3.8291368525938806e-05 tokens_per_second_per_gpu:5322.518128514999 mfu:25.10722612743985 +Step 20919 | loss:0.7568642497062683 lr:3.828715291673934e-05 tokens_per_second_per_gpu:4927.047796027565 mfu:23.241725094900183 +Step 20920 | loss:0.6992008090019226 lr:3.8282937395633674e-05 tokens_per_second_per_gpu:5309.201029054966 mfu:25.044407097155414 +Step 20921 | loss:0.7675061225891113 lr:3.827872196265354e-05 tokens_per_second_per_gpu:4785.9505338955605 mfu:22.576145235748196 +Step 20922 | loss:0.7500283718109131 lr:3.827450661783063e-05 tokens_per_second_per_gpu:5047.207081929827 mfu:23.808536947789356 +Step 20923 | loss:0.722524106502533 lr:3.8270291361196655e-05 tokens_per_second_per_gpu:5314.073384609052 mfu:25.067390829613615 +Step 20924 | loss:0.6206037402153015 lr:3.826607619278334e-05 tokens_per_second_per_gpu:4071.2033742086487 mfu:19.20456093506467 +Step 20925 | loss:0.6567089557647705 lr:3.826186111262233e-05 tokens_per_second_per_gpu:5368.335140292923 mfu:25.32335278918461 +Step 20926 | loss:0.43805262446403503 lr:3.825764612074537e-05 tokens_per_second_per_gpu:4015.5370121713613 mfu:18.941973207674053 +Step 20927 | loss:0.8690948486328125 lr:3.825343121718414e-05 tokens_per_second_per_gpu:3580.9111026942464 mfu:16.891768637843743 +Step 20928 | loss:0.7097210884094238 lr:3.8249216401970364e-05 tokens_per_second_per_gpu:3623.042489380692 mfu:17.090509577199516 +Step 20929 | loss:0.7065267562866211 lr:3.8245001675135735e-05 tokens_per_second_per_gpu:5346.628342826591 mfu:25.220958121974373 +Step 20930 | loss:0.5887770652770996 lr:3.824078703671193e-05 tokens_per_second_per_gpu:4263.697434027969 mfu:20.112588258105664 +Step 20931 | loss:0.6848328113555908 lr:3.82365724867307e-05 tokens_per_second_per_gpu:3858.392623087811 mfu:18.200696312769445 +Step 20932 | loss:0.7334046959877014 lr:3.8232358025223694e-05 tokens_per_second_per_gpu:3382.986198679881 mfu:15.95812309613716 +Step 20933 | loss:0.6843083500862122 lr:3.822814365222262e-05 tokens_per_second_per_gpu:5268.868350276846 mfu:24.854150969894427 +Step 20934 | loss:0.8579741716384888 lr:3.822392936775917e-05 tokens_per_second_per_gpu:5067.011416442622 mfu:23.901957372654106 +Step 20935 | loss:0.7195311784744263 lr:3.821971517186506e-05 tokens_per_second_per_gpu:4809.319377111858 mfu:22.686380056335267 +Step 20936 | loss:0.6715715527534485 lr:3.821550106457197e-05 tokens_per_second_per_gpu:3819.329524187708 mfu:18.01642901040573 +Step 20937 | loss:0.8329828381538391 lr:3.82112870459116e-05 tokens_per_second_per_gpu:5347.305506528797 mfu:25.224152418693723 +Step 20938 | loss:0.6092072129249573 lr:3.820707311591567e-05 tokens_per_second_per_gpu:5272.534261544096 mfu:24.871443698811007 +Step 20939 | loss:0.7082565426826477 lr:3.820285927461583e-05 tokens_per_second_per_gpu:4556.959368414216 mfu:21.495954838251105 +Step 20940 | loss:0.5971344113349915 lr:3.819864552204379e-05 tokens_per_second_per_gpu:5380.399472934644 mfu:25.380262304643015 +Step 20941 | loss:0.5942929983139038 lr:3.819443185823123e-05 tokens_per_second_per_gpu:3327.3272153284797 mfu:15.6955701752668 +Step 20942 | loss:0.6565731167793274 lr:3.8190218283209865e-05 tokens_per_second_per_gpu:5014.206076631781 mfu:23.652865575246363 +Step 20943 | loss:0.8833552002906799 lr:3.818600479701137e-05 tokens_per_second_per_gpu:5323.003158577123 mfu:25.109514096997554 +Step 20944 | loss:0.6880923509597778 lr:3.818179139966745e-05 tokens_per_second_per_gpu:5324.022062351892 mfu:25.114320439946045 +Step 20945 | loss:0.8389040231704712 lr:3.8177578091209794e-05 tokens_per_second_per_gpu:4182.096166788494 mfu:19.72766115792521 +Step 20946 | loss:0.8031062483787537 lr:3.817336487167007e-05 tokens_per_second_per_gpu:5348.2204608498 mfu:25.228468414333197 +Step 20947 | loss:0.6536921262741089 lr:3.816915174107997e-05 tokens_per_second_per_gpu:4874.140296248714 mfu:22.992151391493053 +Step 20948 | loss:0.7366724610328674 lr:3.81649386994712e-05 tokens_per_second_per_gpu:4878.070689801512 mfu:23.01069172847589 +Step 20949 | loss:0.6243467926979065 lr:3.816072574687543e-05 tokens_per_second_per_gpu:3481.5205257480166 mfu:16.422926328607318 +Step 20950 | loss:0.5610294938087463 lr:3.8156512883324344e-05 tokens_per_second_per_gpu:4280.637090298039 mfu:20.192495506934964 +Step 20951 | loss:0.6655133962631226 lr:3.8152300108849653e-05 tokens_per_second_per_gpu:4367.606363385271 mfu:20.60274440657553 +Step 20952 | loss:0.7218271493911743 lr:3.814808742348301e-05 tokens_per_second_per_gpu:4637.991964623251 mfu:21.878199420155685 +Step 20953 | loss:0.6992453336715698 lr:3.8143874827256146e-05 tokens_per_second_per_gpu:5356.670064514864 mfu:25.26832663647235 +Step 20954 | loss:0.6633283495903015 lr:3.813966232020067e-05 tokens_per_second_per_gpu:4203.860730971053 mfu:19.830328320591907 +Step 20955 | loss:0.7630259394645691 lr:3.8135449902348334e-05 tokens_per_second_per_gpu:5348.156901336708 mfu:25.228168593265572 +Step 20956 | loss:0.8429583311080933 lr:3.813123757373079e-05 tokens_per_second_per_gpu:2972.9043651142056 mfu:14.023697120032274 +Step 20957 | loss:0.709922194480896 lr:3.8127025334379705e-05 tokens_per_second_per_gpu:4252.642180991119 mfu:20.06043874331017 +Step 20958 | loss:0.6593528985977173 lr:3.812281318432679e-05 tokens_per_second_per_gpu:4374.718650807474 mfu:20.63629427982742 +Step 20959 | loss:0.5971723794937134 lr:3.811860112360371e-05 tokens_per_second_per_gpu:3518.5074191169956 mfu:16.59739981524353 +Step 20960 | loss:0.8102729916572571 lr:3.811438915224217e-05 tokens_per_second_per_gpu:5341.461363337064 mfu:25.196584598144668 +Step 20961 | loss:0.7227532267570496 lr:3.81101772702738e-05 tokens_per_second_per_gpu:5339.139049192477 mfu:25.185629846097005 +Step 20962 | loss:0.6255342960357666 lr:3.810596547773031e-05 tokens_per_second_per_gpu:5333.1675959335535 mfu:25.157461482239675 +Step 20963 | loss:0.6712514162063599 lr:3.810175377464336e-05 tokens_per_second_per_gpu:5324.8774549306645 mfu:25.118355472685142 +Step 20964 | loss:0.7662945985794067 lr:3.8097542161044654e-05 tokens_per_second_per_gpu:5358.060245062725 mfu:25.27488435530587 +Step 20965 | loss:0.6258585453033447 lr:3.809333063696585e-05 tokens_per_second_per_gpu:5175.437270361686 mfu:24.413420624948603 +Step 20966 | loss:0.6704129576683044 lr:3.8089119202438604e-05 tokens_per_second_per_gpu:5315.325729282133 mfu:25.07329835310857 +Step 20967 | loss:0.7456022500991821 lr:3.808490785749464e-05 tokens_per_second_per_gpu:5303.948709105619 mfu:25.019631007816155 +Step 20968 | loss:0.7479616403579712 lr:3.808069660216561e-05 tokens_per_second_per_gpu:4139.087771525472 mfu:19.524783219480636 +Step 20969 | loss:0.6888112425804138 lr:3.8076485436483165e-05 tokens_per_second_per_gpu:5376.850381652641 mfu:25.36352063552839 +Step 20970 | loss:0.635981559753418 lr:3.807227436047899e-05 tokens_per_second_per_gpu:4411.614291740454 mfu:20.810337313153756 +Step 20971 | loss:0.7231633067131042 lr:3.806806337418477e-05 tokens_per_second_per_gpu:4276.8360920867335 mfu:20.17456554985507 +Step 20972 | loss:0.7332014441490173 lr:3.8063852477632164e-05 tokens_per_second_per_gpu:5021.76870395772 mfu:23.688539778660093 +Step 20973 | loss:0.6484520435333252 lr:3.805964167085284e-05 tokens_per_second_per_gpu:4922.8054936809085 mfu:23.22171343091959 +Step 20974 | loss:0.8031845092773438 lr:3.805543095387849e-05 tokens_per_second_per_gpu:5365.8337582005715 mfu:25.311553343075197 +Step 20975 | loss:0.7382146716117859 lr:3.8051220326740764e-05 tokens_per_second_per_gpu:3423.393563741783 mfu:16.14873153134154 +Step 20976 | loss:0.7408409118652344 lr:3.804700978947134e-05 tokens_per_second_per_gpu:4128.409840634544 mfu:19.474413597625254 +Step 20977 | loss:0.6589469909667969 lr:3.804279934210186e-05 tokens_per_second_per_gpu:5374.011612058717 mfu:25.350129675009818 +Step 20978 | loss:0.6490971446037292 lr:3.803858898466402e-05 tokens_per_second_per_gpu:4975.704594883511 mfu:23.47124751680961 +Step 20979 | loss:0.8289352059364319 lr:3.803437871718948e-05 tokens_per_second_per_gpu:3792.232199533219 mfu:17.888606306730182 +Step 20980 | loss:0.6981291174888611 lr:3.803016853970991e-05 tokens_per_second_per_gpu:3720.227193848783 mfu:17.548946409595906 +Step 20981 | loss:0.8370215892791748 lr:3.8025958452256976e-05 tokens_per_second_per_gpu:4601.158701503379 mfu:21.704450633616535 +Step 20982 | loss:0.7399377226829529 lr:3.802174845486233e-05 tokens_per_second_per_gpu:5332.540988246806 mfu:25.15450566687099 +Step 20983 | loss:0.6860740184783936 lr:3.801753854755763e-05 tokens_per_second_per_gpu:3593.283997480572 mfu:16.950133693584373 +Step 20984 | loss:0.7752938866615295 lr:3.801332873037456e-05 tokens_per_second_per_gpu:5105.2742565138615 mfu:24.082449717583835 +Step 20985 | loss:0.618061363697052 lr:3.8009119003344784e-05 tokens_per_second_per_gpu:4548.644047793212 mfu:21.456730052141836 +Step 20986 | loss:0.7108170390129089 lr:3.800490936649994e-05 tokens_per_second_per_gpu:4988.504323512955 mfu:23.531626020613594 +Step 20987 | loss:0.7674547433853149 lr:3.8000699819871705e-05 tokens_per_second_per_gpu:3742.5153809190724 mfu:17.654083590763303 +Step 20988 | loss:0.7958259582519531 lr:3.7996490363491735e-05 tokens_per_second_per_gpu:4325.574136262435 mfu:20.404471219799664 +Step 20989 | loss:0.6927764415740967 lr:3.7992280997391715e-05 tokens_per_second_per_gpu:5332.786871664889 mfu:25.155665540906128 +Step 20990 | loss:0.7623175382614136 lr:3.7988071721603255e-05 tokens_per_second_per_gpu:5316.8888822395575 mfu:25.08067201231014 +Step 20991 | loss:0.7243945598602295 lr:3.798386253615806e-05 tokens_per_second_per_gpu:5348.413547277459 mfu:25.229379235956404 +Step 20992 | loss:0.7837485671043396 lr:3.797965344108776e-05 tokens_per_second_per_gpu:3938.4800425191806 mfu:18.57848233454999 +Step 20993 | loss:0.7086598873138428 lr:3.797544443642401e-05 tokens_per_second_per_gpu:5358.434253637547 mfu:25.276648617565748 +Step 20994 | loss:0.6080378293991089 lr:3.797123552219849e-05 tokens_per_second_per_gpu:4722.5904437170575 mfu:22.277264880030092 +Step 20995 | loss:0.68232661485672 lr:3.796702669844283e-05 tokens_per_second_per_gpu:4587.933679181245 mfu:21.642065946901546 +Step 20996 | loss:0.8450823426246643 lr:3.796281796518873e-05 tokens_per_second_per_gpu:5333.173804372063 mfu:25.157490768503397 +Step 20997 | loss:0.7585430145263672 lr:3.795860932246778e-05 tokens_per_second_per_gpu:4306.030007598642 mfu:20.312278230320356 +Step 20998 | loss:0.6539113521575928 lr:3.795440077031167e-05 tokens_per_second_per_gpu:3368.2900470354266 mfu:15.888798841408239 +Step 20999 | loss:0.7139655351638794 lr:3.795019230875205e-05 tokens_per_second_per_gpu:4832.0159559612675 mfu:22.7934436912036 +Step 21000 | loss:0.7788257598876953 lr:3.7945983937820574e-05 tokens_per_second_per_gpu:3764.7239939722244 mfu:17.758845407715135 +Step 21001 | loss:0.7135725617408752 lr:3.794177565754889e-05 tokens_per_second_per_gpu:4824.587410394622 mfu:22.75840197432516 +Step 21002 | loss:0.656639039516449 lr:3.7937567467968645e-05 tokens_per_second_per_gpu:5355.665275294338 mfu:25.263586874284407 +Step 21003 | loss:0.6027774214744568 lr:3.7933359369111506e-05 tokens_per_second_per_gpu:5363.621667231601 mfu:25.301118532552927 +Step 21004 | loss:0.7909837365150452 lr:3.792915136100911e-05 tokens_per_second_per_gpu:3487.3897221156976 mfu:16.45061232926183 +Step 21005 | loss:0.7377405166625977 lr:3.792494344369311e-05 tokens_per_second_per_gpu:2995.8696109579046 mfu:14.132028102951983 +Step 21006 | loss:0.7941897511482239 lr:3.792073561719513e-05 tokens_per_second_per_gpu:5348.928642151012 mfu:25.23180902635921 +Step 21007 | loss:0.7254738211631775 lr:3.791652788154685e-05 tokens_per_second_per_gpu:5352.752039845762 mfu:25.249844645624254 +Step 21008 | loss:0.7322768568992615 lr:3.7912320236779916e-05 tokens_per_second_per_gpu:3824.1631433402213 mfu:18.039230016648602 +Step 21009 | loss:0.8127204179763794 lr:3.790811268292594e-05 tokens_per_second_per_gpu:3752.9082138789086 mfu:17.703108357035 +Step 21010 | loss:0.7363379001617432 lr:3.790390522001662e-05 tokens_per_second_per_gpu:4149.353431251729 mfu:19.573208087911542 +Step 21011 | loss:0.8326204419136047 lr:3.7899697848083564e-05 tokens_per_second_per_gpu:4103.709022464517 mfu:19.35789562392308 +Step 21012 | loss:0.6708765625953674 lr:3.789549056715842e-05 tokens_per_second_per_gpu:5369.308262901124 mfu:25.327943174559334 +Step 21013 | loss:0.7675183415412903 lr:3.7891283377272826e-05 tokens_per_second_per_gpu:4934.017455270623 mfu:23.274602166696177 +Step 21014 | loss:0.6595489978790283 lr:3.788707627845845e-05 tokens_per_second_per_gpu:5306.689889181626 mfu:25.032561621928227 +Step 21015 | loss:0.7280002236366272 lr:3.7882869270746903e-05 tokens_per_second_per_gpu:5347.558499640302 mfu:25.225345830365768 +Step 21016 | loss:0.7613086700439453 lr:3.7878662354169856e-05 tokens_per_second_per_gpu:4543.759049653188 mfu:21.433686682448716 +Step 21017 | loss:0.7085803747177124 lr:3.7874455528758937e-05 tokens_per_second_per_gpu:5057.699209890399 mfu:23.858030105521 +Step 21018 | loss:0.7363402247428894 lr:3.787024879454579e-05 tokens_per_second_per_gpu:4484.55802773952 mfu:21.15442536134605 +Step 21019 | loss:0.7434049844741821 lr:3.786604215156203e-05 tokens_per_second_per_gpu:4432.660258446379 mfu:20.909614728917795 +Step 21020 | loss:0.8644335865974426 lr:3.786183559983933e-05 tokens_per_second_per_gpu:3318.9039938792343 mfu:15.655836402540936 +Step 21021 | loss:0.9271423816680908 lr:3.785762913940931e-05 tokens_per_second_per_gpu:3715.622058164685 mfu:17.527223198856845 +Step 21022 | loss:0.6969141960144043 lr:3.7853422770303606e-05 tokens_per_second_per_gpu:3244.705630804863 mfu:15.305830064372977 +Step 21023 | loss:0.6702420115470886 lr:3.784921649255386e-05 tokens_per_second_per_gpu:3552.096729071608 mfu:16.75584632122667 +Step 21024 | loss:0.69590824842453 lr:3.784501030619171e-05 tokens_per_second_per_gpu:5328.251045406898 mfu:25.13426927455542 +Step 21025 | loss:0.7001659870147705 lr:3.784080421124881e-05 tokens_per_second_per_gpu:5113.864208899549 mfu:24.122969988584128 +Step 21026 | loss:0.7618669271469116 lr:3.7836598207756735e-05 tokens_per_second_per_gpu:5367.113717275434 mfu:25.31759112841486 +Step 21027 | loss:0.702114999294281 lr:3.7832392295747176e-05 tokens_per_second_per_gpu:4847.59401784192 mfu:22.866928066986063 +Step 21028 | loss:0.566093385219574 lr:3.782818647525175e-05 tokens_per_second_per_gpu:4148.130393194555 mfu:19.567438808724045 +Step 21029 | loss:0.7682105898857117 lr:3.7823980746302076e-05 tokens_per_second_per_gpu:3957.1989724065247 mfu:18.666782720607365 +Step 21030 | loss:0.7797044515609741 lr:3.78197751089298e-05 tokens_per_second_per_gpu:4489.303154227118 mfu:21.176808932590113 +Step 21031 | loss:0.6253659725189209 lr:3.7815569563166544e-05 tokens_per_second_per_gpu:4517.755618623655 mfu:21.311024061639582 +Step 21032 | loss:0.6202019453048706 lr:3.7811364109043976e-05 tokens_per_second_per_gpu:4612.49150501077 mfu:21.75790940133656 +Step 21033 | loss:0.7289531230926514 lr:3.7807158746593655e-05 tokens_per_second_per_gpu:5367.788711651725 mfu:25.320775192053738 +Step 21034 | loss:0.6418213844299316 lr:3.780295347584727e-05 tokens_per_second_per_gpu:4821.463345693597 mfu:22.743665228110025 +Step 21035 | loss:0.6365995407104492 lr:3.7798748296836414e-05 tokens_per_second_per_gpu:5081.37809973513 mfu:23.969727468953792 +Step 21036 | loss:0.6624184846878052 lr:3.7794543209592736e-05 tokens_per_second_per_gpu:4120.166004949253 mfu:19.43552601815433 +Step 21037 | loss:0.7691642045974731 lr:3.779033821414786e-05 tokens_per_second_per_gpu:4956.715593767694 mfu:23.38167315868863 +Step 21038 | loss:0.9132134914398193 lr:3.7786133310533396e-05 tokens_per_second_per_gpu:3962.3360618684187 mfu:18.69101524807696 +Step 21039 | loss:1.1401227712631226 lr:3.7781928498781004e-05 tokens_per_second_per_gpu:5347.618252654826 mfu:25.225627695530196 +Step 21040 | loss:0.6826913952827454 lr:3.7777723778922275e-05 tokens_per_second_per_gpu:4757.637154152019 mfu:22.442586192737203 +Step 21041 | loss:0.5575045943260193 lr:3.7773519150988846e-05 tokens_per_second_per_gpu:4838.058321900077 mfu:22.8219465622709 +Step 21042 | loss:0.6500944495201111 lr:3.776931461501232e-05 tokens_per_second_per_gpu:5327.052717113219 mfu:25.128616555538247 +Step 21043 | loss:0.8248298764228821 lr:3.7765110171024364e-05 tokens_per_second_per_gpu:3943.7371270548806 mfu:18.603280899256 +Step 21044 | loss:0.6729695200920105 lr:3.776090581905657e-05 tokens_per_second_per_gpu:5313.32076377869 mfu:25.063840588746444 +Step 21045 | loss:0.48504501581192017 lr:3.7756701559140557e-05 tokens_per_second_per_gpu:3394.8827845653636 mfu:16.014241321525837 +Step 21046 | loss:0.768907904624939 lr:3.7752497391307964e-05 tokens_per_second_per_gpu:3682.364209387322 mfu:17.3703402517999 +Step 21047 | loss:0.7295653820037842 lr:3.77482933155904e-05 tokens_per_second_per_gpu:5311.695562028521 mfu:25.056174234802885 +Step 21048 | loss:0.7149417996406555 lr:3.774408933201949e-05 tokens_per_second_per_gpu:5322.225939170677 mfu:25.105847820450037 +Step 21049 | loss:0.6765959858894348 lr:3.773988544062683e-05 tokens_per_second_per_gpu:4382.5534517786655 mfu:20.673252372763418 +Step 21050 | loss:0.8800303936004639 lr:3.773568164144407e-05 tokens_per_second_per_gpu:4274.437563115153 mfu:20.163251279464887 +Step 21051 | loss:0.6800359487533569 lr:3.77314779345028e-05 tokens_per_second_per_gpu:5370.671758960519 mfu:25.334375018107917 +Step 21052 | loss:0.7317286133766174 lr:3.772727431983466e-05 tokens_per_second_per_gpu:5332.478584657687 mfu:25.154211298493966 +Step 21053 | loss:0.6026546359062195 lr:3.772307079747126e-05 tokens_per_second_per_gpu:5327.343458994794 mfu:25.129988034599318 +Step 21054 | loss:0.6784175634384155 lr:3.771886736744422e-05 tokens_per_second_per_gpu:5298.337054483569 mfu:24.993159875516934 +Step 21055 | loss:0.6711155772209167 lr:3.7714664029785115e-05 tokens_per_second_per_gpu:3692.979396374632 mfu:17.4204138999567 +Step 21056 | loss:0.6640682816505432 lr:3.771046078452562e-05 tokens_per_second_per_gpu:4015.2117566354914 mfu:18.94043892181704 +Step 21057 | loss:0.745864987373352 lr:3.77062576316973e-05 tokens_per_second_per_gpu:4241.474470775815 mfu:20.007758748816737 +Step 21058 | loss:0.7151705026626587 lr:3.770205457133178e-05 tokens_per_second_per_gpu:4405.026412092893 mfu:20.77926116084804 +Step 21059 | loss:0.6590245366096497 lr:3.7697851603460685e-05 tokens_per_second_per_gpu:4251.435156079327 mfu:20.05474500086156 +Step 21060 | loss:0.5809616446495056 lr:3.769364872811562e-05 tokens_per_second_per_gpu:4667.756145365512 mfu:22.01860213900982 +Step 21061 | loss:0.7697238326072693 lr:3.768944594532821e-05 tokens_per_second_per_gpu:5032.178110719794 mfu:23.737642726384728 +Step 21062 | loss:0.6214364171028137 lr:3.768524325513002e-05 tokens_per_second_per_gpu:5330.703793447817 mfu:25.145839305546403 +Step 21063 | loss:0.7962329983711243 lr:3.76810406575527e-05 tokens_per_second_per_gpu:4364.217299243876 mfu:20.586757612786602 +Step 21064 | loss:0.6519016623497009 lr:3.7676838152627845e-05 tokens_per_second_per_gpu:4093.9724170828226 mfu:19.311966395101454 +Step 21065 | loss:0.6867189407348633 lr:3.7672635740387045e-05 tokens_per_second_per_gpu:5348.974973988208 mfu:25.232027581540287 +Step 21066 | loss:0.7023140788078308 lr:3.7668433420861944e-05 tokens_per_second_per_gpu:5326.833289233604 mfu:25.127581476791846 +Step 21067 | loss:0.6991384625434875 lr:3.766423119408412e-05 tokens_per_second_per_gpu:4552.018953106515 mfu:21.47265005632291 +Step 21068 | loss:0.724608302116394 lr:3.766002906008521e-05 tokens_per_second_per_gpu:5360.6681537601735 mfu:25.28718630558684 +Step 21069 | loss:0.8377905488014221 lr:3.765582701889677e-05 tokens_per_second_per_gpu:5254.155038277191 mfu:24.784745766842168 +Step 21070 | loss:0.5834788084030151 lr:3.765162507055044e-05 tokens_per_second_per_gpu:4871.368152659023 mfu:22.979074717204007 +Step 21071 | loss:0.6777122616767883 lr:3.76474232150778e-05 tokens_per_second_per_gpu:5324.69545516237 mfu:25.117496948725734 +Step 21072 | loss:0.6870813965797424 lr:3.7643221452510476e-05 tokens_per_second_per_gpu:5060.938078557862 mfu:23.8733084016334 +Step 21073 | loss:0.6067088842391968 lr:3.763901978288006e-05 tokens_per_second_per_gpu:5332.23245927668 mfu:25.15305028307844 +Step 21074 | loss:0.6763626337051392 lr:3.763481820621815e-05 tokens_per_second_per_gpu:4715.522322731074 mfu:22.243923347393416 +Step 21075 | loss:0.7401880621910095 lr:3.7630616722556355e-05 tokens_per_second_per_gpu:3450.3726191897317 mfu:16.27599633899103 +Step 21076 | loss:0.6680678725242615 lr:3.762641533192628e-05 tokens_per_second_per_gpu:5355.622076913995 mfu:25.26338310015393 +Step 21077 | loss:0.7350721955299377 lr:3.762221403435951e-05 tokens_per_second_per_gpu:4447.7398968137195 mfu:20.98074795594815 +Step 21078 | loss:0.6677030920982361 lr:3.761801282988763e-05 tokens_per_second_per_gpu:5299.610197897062 mfu:24.999165510219047 +Step 21079 | loss:0.7844327688217163 lr:3.761381171854227e-05 tokens_per_second_per_gpu:4708.581927765487 mfu:22.21118432018725 +Step 21080 | loss:0.7236862778663635 lr:3.760961070035501e-05 tokens_per_second_per_gpu:4611.062101157845 mfu:21.751166659480866 +Step 21081 | loss:0.6489266753196716 lr:3.760540977535744e-05 tokens_per_second_per_gpu:3952.828011830778 mfu:18.646164153808794 +Step 21082 | loss:0.765962541103363 lr:3.760120894358118e-05 tokens_per_second_per_gpu:5029.64755555659 mfu:23.725705665922664 +Step 21083 | loss:0.6625165939331055 lr:3.759700820505781e-05 tokens_per_second_per_gpu:4184.241506630458 mfu:19.737781091992478 +Step 21084 | loss:0.826837956905365 lr:3.7592807559818924e-05 tokens_per_second_per_gpu:3997.654281786941 mfu:18.857617317342317 +Step 21085 | loss:0.8302170634269714 lr:3.7588607007896096e-05 tokens_per_second_per_gpu:4784.7874913347905 mfu:22.570658965553523 +Step 21086 | loss:0.6545707583427429 lr:3.7584406549320946e-05 tokens_per_second_per_gpu:4971.344738671501 mfu:23.450681331189383 +Step 21087 | loss:0.7279106974601746 lr:3.758020618412506e-05 tokens_per_second_per_gpu:4434.9571366092005 mfu:20.920449495099653 +Step 21088 | loss:0.5898118019104004 lr:3.7576005912340036e-05 tokens_per_second_per_gpu:4506.934754561075 mfu:21.259980199626607 +Step 21089 | loss:0.793294370174408 lr:3.757180573399745e-05 tokens_per_second_per_gpu:3053.275646226149 mfu:14.402822165791886 +Step 21090 | loss:0.9438067674636841 lr:3.756760564912891e-05 tokens_per_second_per_gpu:4853.176435647582 mfu:22.893261284234068 +Step 21091 | loss:0.7923604846000671 lr:3.7563405657765984e-05 tokens_per_second_per_gpu:5158.7465572427345 mfu:24.334687683437203 +Step 21092 | loss:0.676076352596283 lr:3.755920575994027e-05 tokens_per_second_per_gpu:5367.261499075978 mfu:25.3182882403822 +Step 21093 | loss:0.7868461012840271 lr:3.7555005955683363e-05 tokens_per_second_per_gpu:5319.266243203469 mfu:25.091886429596514 +Step 21094 | loss:0.6680278182029724 lr:3.7550806245026834e-05 tokens_per_second_per_gpu:4171.6931700742325 mfu:19.678588447489773 +Step 21095 | loss:0.858866810798645 lr:3.754660662800228e-05 tokens_per_second_per_gpu:4814.898610368135 mfu:22.712698251520482 +Step 21096 | loss:0.646297812461853 lr:3.75424071046413e-05 tokens_per_second_per_gpu:4188.789089053707 mfu:19.759232814180514 +Step 21097 | loss:0.7490442395210266 lr:3.753820767497547e-05 tokens_per_second_per_gpu:4351.9157021171695 mfu:20.528728880271025 +Step 21098 | loss:0.6843915581703186 lr:3.7534008339036354e-05 tokens_per_second_per_gpu:5354.843705924311 mfu:25.2597113913918 +Step 21099 | loss:0.6956157684326172 lr:3.7529809096855553e-05 tokens_per_second_per_gpu:4997.718946965615 mfu:23.575092971616673 +Step 21100 | loss:0.7279237508773804 lr:3.752560994846466e-05 tokens_per_second_per_gpu:5360.614131329544 mfu:25.286931472564383 +Step 21101 | loss:0.6545585989952087 lr:3.7521410893895225e-05 tokens_per_second_per_gpu:5360.638495014195 mfu:25.287046400221755 +Step 21102 | loss:0.6696878671646118 lr:3.751721193317887e-05 tokens_per_second_per_gpu:3716.824736451014 mfu:17.532896437531534 +Step 21103 | loss:0.7774429321289062 lr:3.751301306634714e-05 tokens_per_second_per_gpu:4918.814322176967 mfu:23.20288639397142 +Step 21104 | loss:0.6034417748451233 lr:3.750881429343167e-05 tokens_per_second_per_gpu:3686.5712515698992 mfu:17.39018558757053 +Step 21105 | loss:0.8022616505622864 lr:3.7504615614463966e-05 tokens_per_second_per_gpu:5201.36666270137 mfu:24.53573399262541 +Step 21106 | loss:0.6564822196960449 lr:3.750041702947565e-05 tokens_per_second_per_gpu:5347.400660182459 mfu:25.224601274712935 +Step 21107 | loss:0.6737722754478455 lr:3.749621853849829e-05 tokens_per_second_per_gpu:4793.426403332005 mfu:22.611410187394764 +Step 21108 | loss:0.6824541091918945 lr:3.7492020141563465e-05 tokens_per_second_per_gpu:4955.805274855557 mfu:23.377379029063658 +Step 21109 | loss:0.7918699979782104 lr:3.7487821838702766e-05 tokens_per_second_per_gpu:4559.492295455005 mfu:21.50790308726442 +Step 21110 | loss:0.7261011004447937 lr:3.748362362994774e-05 tokens_per_second_per_gpu:5317.65612081771 mfu:25.084291207587686 +Step 21111 | loss:0.7081772089004517 lr:3.747942551532998e-05 tokens_per_second_per_gpu:4131.77934136961 mfu:19.490308107488666 +Step 21112 | loss:0.7084338665008545 lr:3.7475227494881084e-05 tokens_per_second_per_gpu:5162.204835509086 mfu:24.351000972062693 +Step 21113 | loss:0.8688089847564697 lr:3.747102956863258e-05 tokens_per_second_per_gpu:4837.572656365099 mfu:22.819655595079322 +Step 21114 | loss:0.7460610270500183 lr:3.7466831736616046e-05 tokens_per_second_per_gpu:4051.1811074597936 mfu:19.11011248665935 +Step 21115 | loss:0.8144115209579468 lr:3.746263399886309e-05 tokens_per_second_per_gpu:5337.8109879364365 mfu:25.179365154561648 +Step 21116 | loss:0.6115738749504089 lr:3.745843635540526e-05 tokens_per_second_per_gpu:5341.327625794464 mfu:25.195953735338254 +Step 21117 | loss:0.7627862691879272 lr:3.7454238806274124e-05 tokens_per_second_per_gpu:4119.7043549087675 mfu:19.43334833614774 +Step 21118 | loss:0.7333769798278809 lr:3.7450041351501266e-05 tokens_per_second_per_gpu:5316.749101195646 mfu:25.080012641276905 +Step 21119 | loss:0.600867509841919 lr:3.744584399111826e-05 tokens_per_second_per_gpu:5314.824884366281 mfu:25.070935782187547 +Step 21120 | loss:0.7311491370201111 lr:3.7441646725156646e-05 tokens_per_second_per_gpu:5335.149139989513 mfu:25.166808763637917 +Step 21121 | loss:0.5997105836868286 lr:3.7437449553648004e-05 tokens_per_second_per_gpu:5271.659765855869 mfu:24.86731855344529 +Step 21122 | loss:0.9134327173233032 lr:3.7433252476623916e-05 tokens_per_second_per_gpu:4306.3917865689855 mfu:20.313984803449323 +Step 21123 | loss:0.7459466457366943 lr:3.7429055494115925e-05 tokens_per_second_per_gpu:5356.123206430063 mfu:25.26574701358273 +Step 21124 | loss:0.8175953030586243 lr:3.7424858606155635e-05 tokens_per_second_per_gpu:4783.595895836055 mfu:22.565038006278836 +Step 21125 | loss:0.6733881235122681 lr:3.742066181277458e-05 tokens_per_second_per_gpu:4398.653600175203 mfu:20.74919951971846 +Step 21126 | loss:0.8607121109962463 lr:3.741646511400434e-05 tokens_per_second_per_gpu:4771.340154675857 mfu:22.507225584181118 +Step 21127 | loss:0.634451150894165 lr:3.7412268509876456e-05 tokens_per_second_per_gpu:5010.917344953076 mfu:23.63735206680243 +Step 21128 | loss:0.6483544707298279 lr:3.740807200042251e-05 tokens_per_second_per_gpu:5376.103357234602 mfu:25.35999679389299 +Step 21129 | loss:0.7882506847381592 lr:3.7403875585674075e-05 tokens_per_second_per_gpu:5331.406837405561 mfu:25.149155683846754 +Step 21130 | loss:0.6528711915016174 lr:3.739967926566268e-05 tokens_per_second_per_gpu:3622.1269570207965 mfu:17.08619085485196 +Step 21131 | loss:0.7465351819992065 lr:3.739548304041991e-05 tokens_per_second_per_gpu:5355.902025228113 mfu:25.264703664115732 +Step 21132 | loss:0.630186915397644 lr:3.739128690997733e-05 tokens_per_second_per_gpu:4726.142869832072 mfu:22.294022280121013 +Step 21133 | loss:0.886199414730072 lr:3.738709087436649e-05 tokens_per_second_per_gpu:4078.227189695374 mfu:19.237693471102048 +Step 21134 | loss:0.6820229887962341 lr:3.738289493361894e-05 tokens_per_second_per_gpu:5268.681830446486 mfu:24.853271124031053 +Step 21135 | loss:0.7457260489463806 lr:3.737869908776625e-05 tokens_per_second_per_gpu:5327.179209282082 mfu:25.129213240680567 +Step 21136 | loss:0.7715201377868652 lr:3.7374503336839986e-05 tokens_per_second_per_gpu:4214.205147039792 mfu:19.879124696127523 +Step 21137 | loss:0.613328754901886 lr:3.737030768087166e-05 tokens_per_second_per_gpu:5338.944791178228 mfu:25.184713497899956 +Step 21138 | loss:0.8280472755432129 lr:3.736611211989289e-05 tokens_per_second_per_gpu:4104.086025912167 mfu:19.359674013509068 +Step 21139 | loss:0.925632894039154 lr:3.7361916653935195e-05 tokens_per_second_per_gpu:5311.027625668936 mfu:25.05302346503303 +Step 21140 | loss:0.6491108536720276 lr:3.735772128303013e-05 tokens_per_second_per_gpu:3626.0916413453624 mfu:17.104892947255838 +Step 21141 | loss:0.5610767006874084 lr:3.7353526007209286e-05 tokens_per_second_per_gpu:5364.864963731945 mfu:25.3069833742733 +Step 21142 | loss:0.7147943377494812 lr:3.734933082650417e-05 tokens_per_second_per_gpu:4882.150524242631 mfu:23.029937003626184 +Step 21143 | loss:0.8012101650238037 lr:3.734513574094634e-05 tokens_per_second_per_gpu:5355.428917999403 mfu:25.26247193659749 +Step 21144 | loss:0.808915376663208 lr:3.734094075056736e-05 tokens_per_second_per_gpu:5127.038121285408 mfu:24.185113581009457 +Step 21145 | loss:0.759880006313324 lr:3.7336745855398794e-05 tokens_per_second_per_gpu:4765.268524555783 mfu:22.478584668977806 +Step 21146 | loss:0.6845818161964417 lr:3.733255105547216e-05 tokens_per_second_per_gpu:3114.022978071499 mfu:14.689377694669973 +Step 21147 | loss:0.7064557075500488 lr:3.732835635081904e-05 tokens_per_second_per_gpu:4137.995712237371 mfu:19.51963178949413 +Step 21148 | loss:0.683779776096344 lr:3.7324161741470976e-05 tokens_per_second_per_gpu:5186.0913586537945 mfu:24.463677777197383 +Step 21149 | loss:0.714309573173523 lr:3.73199672274595e-05 tokens_per_second_per_gpu:4460.0797245228505 mfu:21.03895702864383 +Step 21150 | loss:0.7288458943367004 lr:3.731577280881616e-05 tokens_per_second_per_gpu:5286.291608723046 mfu:24.936339452699578 +Step 21151 | loss:0.6654671430587769 lr:3.7311578485572516e-05 tokens_per_second_per_gpu:3941.0352712189388 mfu:18.590535784293724 +Step 21152 | loss:0.7592280507087708 lr:3.730738425776011e-05 tokens_per_second_per_gpu:5354.753447436878 mfu:25.25928562670689 +Step 21153 | loss:0.5580977201461792 lr:3.7303190125410483e-05 tokens_per_second_per_gpu:4341.547884136309 mfu:20.479822113923152 +Step 21154 | loss:0.6314038038253784 lr:3.729899608855519e-05 tokens_per_second_per_gpu:4292.817780742406 mfu:20.249953902001135 +Step 21155 | loss:0.6166173815727234 lr:3.7294802147225784e-05 tokens_per_second_per_gpu:4146.422440937121 mfu:19.559382106519024 +Step 21156 | loss:0.7161149382591248 lr:3.729060830145377e-05 tokens_per_second_per_gpu:4356.947553924137 mfu:20.55246498376693 +Step 21157 | loss:0.7451303601264954 lr:3.728641455127071e-05 tokens_per_second_per_gpu:4936.997006748694 mfu:23.288657219382138 +Step 21158 | loss:0.7604041695594788 lr:3.728222089670815e-05 tokens_per_second_per_gpu:5318.6302187529445 mfu:25.08888619374643 +Step 21159 | loss:0.7702574133872986 lr:3.727802733779763e-05 tokens_per_second_per_gpu:4872.734072360618 mfu:22.985517993486365 +Step 21160 | loss:0.624135434627533 lr:3.72738338745707e-05 tokens_per_second_per_gpu:5099.963393938332 mfu:24.057397472688468 +Step 21161 | loss:0.6177617311477661 lr:3.726964050705888e-05 tokens_per_second_per_gpu:5333.844321580016 mfu:25.16065371257534 +Step 21162 | loss:0.7169550061225891 lr:3.726544723529374e-05 tokens_per_second_per_gpu:4058.5427606575618 mfu:19.144838660820778 +Step 21163 | loss:0.7130692601203918 lr:3.7261254059306776e-05 tokens_per_second_per_gpu:5330.574698111661 mfu:25.145230340819822 +Step 21164 | loss:0.7085323929786682 lr:3.725706097912954e-05 tokens_per_second_per_gpu:4584.4258880159 mfu:21.62551909748372 +Step 21165 | loss:0.692322313785553 lr:3.725286799479358e-05 tokens_per_second_per_gpu:3679.844248030481 mfu:17.35845316412994 +Step 21166 | loss:0.7576082348823547 lr:3.7248675106330424e-05 tokens_per_second_per_gpu:4587.621556419621 mfu:21.64059361058133 +Step 21167 | loss:0.6815107464790344 lr:3.7244482313771615e-05 tokens_per_second_per_gpu:5295.349051633965 mfu:24.9790649562695 +Step 21168 | loss:0.6862372756004333 lr:3.724028961714868e-05 tokens_per_second_per_gpu:5357.238506896813 mfu:25.271008076174272 +Step 21169 | loss:0.6078587174415588 lr:3.7236097016493154e-05 tokens_per_second_per_gpu:5342.2561077417195 mfu:25.200333543098807 +Step 21170 | loss:0.7243011593818665 lr:3.723190451183659e-05 tokens_per_second_per_gpu:3778.2629329510573 mfu:17.822710892859476 +Step 21171 | loss:0.8184400796890259 lr:3.7227712103210486e-05 tokens_per_second_per_gpu:5172.97836184201 mfu:24.401821533928235 +Step 21172 | loss:0.7771188020706177 lr:3.722351979064639e-05 tokens_per_second_per_gpu:5203.060866887083 mfu:24.54372584283046 +Step 21173 | loss:0.8578310608863831 lr:3.721932757417582e-05 tokens_per_second_per_gpu:5258.139853744362 mfu:24.80354282128044 +Step 21174 | loss:0.5874539017677307 lr:3.721513545383033e-05 tokens_per_second_per_gpu:3546.822522647583 mfu:16.730966989652135 +Step 21175 | loss:0.8831309080123901 lr:3.721094342964142e-05 tokens_per_second_per_gpu:4253.298921391453 mfu:20.063536699830014 +Step 21176 | loss:0.7592971920967102 lr:3.720675150164066e-05 tokens_per_second_per_gpu:3657.976194334886 mfu:17.25529782377312 +Step 21177 | loss:0.7226081490516663 lr:3.720255966985956e-05 tokens_per_second_per_gpu:3520.4782218480655 mfu:16.606696427979607 +Step 21178 | loss:0.7843612432479858 lr:3.719836793432962e-05 tokens_per_second_per_gpu:5315.06849114079 mfu:25.072084916906505 +Step 21179 | loss:0.5430905818939209 lr:3.7194176295082404e-05 tokens_per_second_per_gpu:3715.9957927350706 mfu:17.528986168591047 +Step 21180 | loss:0.6564934849739075 lr:3.7189984752149405e-05 tokens_per_second_per_gpu:3742.9511213946644 mfu:17.65613905293166 +Step 21181 | loss:0.637271523475647 lr:3.7185793305562175e-05 tokens_per_second_per_gpu:4773.792845668299 mfu:22.5187953460647 +Step 21182 | loss:0.7087048292160034 lr:3.718160195535222e-05 tokens_per_second_per_gpu:5041.634635396526 mfu:23.782250766734357 +Step 21183 | loss:0.5973698496818542 lr:3.717741070155108e-05 tokens_per_second_per_gpu:4423.481288615858 mfu:20.866315962133484 +Step 21184 | loss:0.7288404107093811 lr:3.717321954419029e-05 tokens_per_second_per_gpu:4259.334135459055 mfu:20.09200583430059 +Step 21185 | loss:0.6797997951507568 lr:3.716902848330133e-05 tokens_per_second_per_gpu:5322.177454723207 mfu:25.105619111039026 +Step 21186 | loss:0.7540764808654785 lr:3.716483751891574e-05 tokens_per_second_per_gpu:4189.116556108448 mfu:19.760777532149362 +Step 21187 | loss:0.7449642419815063 lr:3.7160646651065055e-05 tokens_per_second_per_gpu:4589.420746246136 mfu:21.64908069596644 +Step 21188 | loss:0.6855119466781616 lr:3.7156455879780785e-05 tokens_per_second_per_gpu:5333.14247130657 mfu:25.15734296508749 +Step 21189 | loss:0.6654341220855713 lr:3.7152265205094436e-05 tokens_per_second_per_gpu:5079.370785562935 mfu:23.960258625520442 +Step 21190 | loss:0.8267351984977722 lr:3.714807462703755e-05 tokens_per_second_per_gpu:5190.769977219182 mfu:24.485747619224956 +Step 21191 | loss:0.7670917510986328 lr:3.714388414564165e-05 tokens_per_second_per_gpu:4981.492222422838 mfu:23.498548743383697 +Step 21192 | loss:0.6661009788513184 lr:3.713969376093822e-05 tokens_per_second_per_gpu:4421.208640393023 mfu:20.855595492712602 +Step 21193 | loss:0.5370895266532898 lr:3.713550347295879e-05 tokens_per_second_per_gpu:4374.348993150676 mfu:20.634550541590205 +Step 21194 | loss:0.757224440574646 lr:3.713131328173489e-05 tokens_per_second_per_gpu:4568.929666854707 mfu:21.552420778338416 +Step 21195 | loss:0.9292436838150024 lr:3.712712318729801e-05 tokens_per_second_per_gpu:4632.882013689857 mfu:21.854094909755332 +Step 21196 | loss:0.6415342688560486 lr:3.712293318967969e-05 tokens_per_second_per_gpu:4113.518749386899 mfu:19.404169779527958 +Step 21197 | loss:0.7220950126647949 lr:3.711874328891144e-05 tokens_per_second_per_gpu:4284.979927307422 mfu:20.212981410072626 +Step 21198 | loss:0.7433567643165588 lr:3.711455348502477e-05 tokens_per_second_per_gpu:3054.417513569222 mfu:14.408208548871817 +Step 21199 | loss:0.6653295755386353 lr:3.7110363778051175e-05 tokens_per_second_per_gpu:5373.7746931021675 mfu:25.349012087869227 +Step 21200 | loss:0.7391570210456848 lr:3.7106174168022175e-05 tokens_per_second_per_gpu:4376.091716363957 mfu:20.64277126432649 +Step 21201 | loss:0.6647348999977112 lr:3.710198465496929e-05 tokens_per_second_per_gpu:5383.958571437597 mfu:25.397051179526265 +Step 21202 | loss:0.63584303855896 lr:3.7097795238924015e-05 tokens_per_second_per_gpu:5192.196801136561 mfu:24.492478191084494 +Step 21203 | loss:0.6592861413955688 lr:3.7093605919917876e-05 tokens_per_second_per_gpu:5341.2037009068945 mfu:25.19536916049982 +Step 21204 | loss:0.686834990978241 lr:3.708941669798238e-05 tokens_per_second_per_gpu:5329.250210066555 mfu:25.138982504730095 +Step 21205 | loss:0.5848922729492188 lr:3.708522757314901e-05 tokens_per_second_per_gpu:4643.952688514756 mfu:21.906317171756278 +Step 21206 | loss:0.7454655766487122 lr:3.708103854544932e-05 tokens_per_second_per_gpu:4953.141480150217 mfu:23.364813454948322 +Step 21207 | loss:0.6678568720817566 lr:3.707684961491477e-05 tokens_per_second_per_gpu:4509.405900412996 mfu:21.271637016231075 +Step 21208 | loss:0.6961547136306763 lr:3.707266078157688e-05 tokens_per_second_per_gpu:4182.035901172084 mfu:19.72737687473023 +Step 21209 | loss:0.6159195899963379 lr:3.706847204546716e-05 tokens_per_second_per_gpu:4968.597667696659 mfu:23.437722928702417 +Step 21210 | loss:0.7218642234802246 lr:3.706428340661712e-05 tokens_per_second_per_gpu:5341.815747964794 mfu:25.198256290897927 +Step 21211 | loss:0.8538833260536194 lr:3.706009486505824e-05 tokens_per_second_per_gpu:4901.63059901549 mfu:23.12182783997324 +Step 21212 | loss:0.7965065240859985 lr:3.705590642082204e-05 tokens_per_second_per_gpu:4753.4941066779165 mfu:22.42304273092507 +Step 21213 | loss:0.889695405960083 lr:3.7051718073940046e-05 tokens_per_second_per_gpu:3913.5687713353586 mfu:18.460971618074762 +Step 21214 | loss:0.664147675037384 lr:3.704752982444372e-05 tokens_per_second_per_gpu:3654.557648300879 mfu:17.23917195886661 +Step 21215 | loss:0.7229176163673401 lr:3.704334167236456e-05 tokens_per_second_per_gpu:4937.899676869134 mfu:23.29291526835166 +Step 21216 | loss:0.7608813047409058 lr:3.703915361773408e-05 tokens_per_second_per_gpu:3772.6094580638496 mfu:17.796042487234235 +Step 21217 | loss:0.7468913197517395 lr:3.7034965660583795e-05 tokens_per_second_per_gpu:5347.691096277523 mfu:25.225971311326855 +Step 21218 | loss:0.8252922296524048 lr:3.703077780094517e-05 tokens_per_second_per_gpu:5144.70052540056 mfu:24.268430154737512 +Step 21219 | loss:0.7777242660522461 lr:3.702659003884973e-05 tokens_per_second_per_gpu:4280.106445152187 mfu:20.189992363244585 +Step 21220 | loss:0.6871011257171631 lr:3.702240237432898e-05 tokens_per_second_per_gpu:4955.265578046514 mfu:23.37483318713363 +Step 21221 | loss:0.5828402638435364 lr:3.701821480741438e-05 tokens_per_second_per_gpu:4612.542632714632 mfu:21.758150579438965 +Step 21222 | loss:0.7792075276374817 lr:3.701402733813744e-05 tokens_per_second_per_gpu:3884.544639429514 mfu:18.324059835847354 +Step 21223 | loss:0.7079558968544006 lr:3.700983996652966e-05 tokens_per_second_per_gpu:5354.295660985282 mfu:25.25712616990829 +Step 21224 | loss:0.8014016151428223 lr:3.7005652692622526e-05 tokens_per_second_per_gpu:4773.857774920963 mfu:22.51910162842759 +Step 21225 | loss:0.6425972580909729 lr:3.7001465516447534e-05 tokens_per_second_per_gpu:5276.109120386269 mfu:24.888306917903346 +Step 21226 | loss:0.7547242045402527 lr:3.699727843803619e-05 tokens_per_second_per_gpu:4780.914932570974 mfu:22.552391445137946 +Step 21227 | loss:0.7434131503105164 lr:3.699309145741998e-05 tokens_per_second_per_gpu:4415.471810367765 mfu:20.828533886679327 +Step 21228 | loss:0.6625385284423828 lr:3.698890457463038e-05 tokens_per_second_per_gpu:3396.998664024644 mfu:16.02422228594152 +Step 21229 | loss:0.6328278183937073 lr:3.6984717789698885e-05 tokens_per_second_per_gpu:4333.759782122579 mfu:20.443084307937166 +Step 21230 | loss:0.7778005003929138 lr:3.698053110265699e-05 tokens_per_second_per_gpu:5104.9983257827 mfu:24.08114810524631 +Step 21231 | loss:0.6385661959648132 lr:3.697634451353619e-05 tokens_per_second_per_gpu:4523.239044112296 mfu:21.33689031523836 +Step 21232 | loss:0.7864347696304321 lr:3.697215802236795e-05 tokens_per_second_per_gpu:5391.463732793729 mfu:25.432454306155705 +Step 21233 | loss:0.5878013372421265 lr:3.696797162918378e-05 tokens_per_second_per_gpu:3768.7694715628154 mfu:17.777928615739604 +Step 21234 | loss:0.7341238260269165 lr:3.696378533401514e-05 tokens_per_second_per_gpu:3035.3632458208217 mfu:14.318326316909305 +Step 21235 | loss:0.6484057903289795 lr:3.6959599136893584e-05 tokens_per_second_per_gpu:4237.317200724457 mfu:19.988148196681067 +Step 21236 | loss:0.6564895510673523 lr:3.6955413037850505e-05 tokens_per_second_per_gpu:4572.635664317337 mfu:21.569902600678006 +Step 21237 | loss:0.7780084609985352 lr:3.695122703691744e-05 tokens_per_second_per_gpu:3908.769823052404 mfu:18.43833416023906 +Step 21238 | loss:0.6805835366249084 lr:3.6947041134125854e-05 tokens_per_second_per_gpu:3898.8813898080825 mfu:18.39168873348491 +Step 21239 | loss:0.5517210960388184 lr:3.694285532950724e-05 tokens_per_second_per_gpu:3868.790591456736 mfu:18.249745303642005 +Step 21240 | loss:0.7879232168197632 lr:3.6938669623093084e-05 tokens_per_second_per_gpu:4321.345626473648 mfu:20.384524617667793 +Step 21241 | loss:0.8240733742713928 lr:3.693448401491485e-05 tokens_per_second_per_gpu:4332.720332063867 mfu:20.43818104466175 +Step 21242 | loss:0.9306624531745911 lr:3.693029850500405e-05 tokens_per_second_per_gpu:5317.010946995013 mfu:25.08124781258802 +Step 21243 | loss:0.6974766850471497 lr:3.692611309339213e-05 tokens_per_second_per_gpu:4405.3909808236995 mfu:20.780980893753092 +Step 21244 | loss:0.7937739491462708 lr:3.6921927780110585e-05 tokens_per_second_per_gpu:3508.776530100748 mfu:16.551497551493025 +Step 21245 | loss:0.706498920917511 lr:3.6917742565190875e-05 tokens_per_second_per_gpu:5325.662687914283 mfu:25.122059550643435 +Step 21246 | loss:0.7348511815071106 lr:3.69135574486645e-05 tokens_per_second_per_gpu:5325.4548664379945 mfu:25.12107922128178 +Step 21247 | loss:0.7335144877433777 lr:3.690937243056294e-05 tokens_per_second_per_gpu:4619.821805695631 mfu:21.79248768034531 +Step 21248 | loss:0.7382350564002991 lr:3.690518751091764e-05 tokens_per_second_per_gpu:4506.987017723013 mfu:21.260226733878483 +Step 21249 | loss:0.6219853758811951 lr:3.690100268976012e-05 tokens_per_second_per_gpu:5359.47096091243 mfu:25.28153894266187 +Step 21250 | loss:0.5925594568252563 lr:3.6896817967121825e-05 tokens_per_second_per_gpu:5069.419522140587 mfu:23.913316818096607 +Step 21251 | loss:0.670215368270874 lr:3.6892633343034225e-05 tokens_per_second_per_gpu:5360.248788141447 mfu:25.285208086412155 +Step 21252 | loss:0.7234622240066528 lr:3.68884488175288e-05 tokens_per_second_per_gpu:4314.307281898538 mfu:20.351323545441584 +Step 21253 | loss:0.6832377314567566 lr:3.688426439063702e-05 tokens_per_second_per_gpu:4206.6157983188805 mfu:19.84332444333457 +Step 21254 | loss:0.8067761063575745 lr:3.6880080062390365e-05 tokens_per_second_per_gpu:5356.444481604024 mfu:25.267262523393505 +Step 21255 | loss:0.8035280108451843 lr:3.6875895832820314e-05 tokens_per_second_per_gpu:4729.282812563811 mfu:22.308833925716094 +Step 21256 | loss:0.6421353816986084 lr:3.687171170195833e-05 tokens_per_second_per_gpu:4843.38760308913 mfu:22.84708569090879 +Step 21257 | loss:0.736890435218811 lr:3.686752766983586e-05 tokens_per_second_per_gpu:3891.1524406469034 mfu:18.355229961596923 +Step 21258 | loss:0.6015599370002747 lr:3.686334373648439e-05 tokens_per_second_per_gpu:3862.363918216888 mfu:18.21942958946612 +Step 21259 | loss:0.8635421395301819 lr:3.68591599019354e-05 tokens_per_second_per_gpu:4934.24095354147 mfu:23.275656446172498 +Step 21260 | loss:0.6925990581512451 lr:3.685497616622034e-05 tokens_per_second_per_gpu:5379.960316648247 mfu:25.378190729512287 +Step 21261 | loss:0.6598414182662964 lr:3.685079252937067e-05 tokens_per_second_per_gpu:4307.438119664029 mfu:20.318920535181476 +Step 21262 | loss:0.6763503551483154 lr:3.684660899141788e-05 tokens_per_second_per_gpu:4007.2879142817005 mfu:18.903060805487293 +Step 21263 | loss:0.7074731588363647 lr:3.6842425552393424e-05 tokens_per_second_per_gpu:4963.39764786122 mfu:23.413193547923782 +Step 21264 | loss:0.7209660410881042 lr:3.6838242212328775e-05 tokens_per_second_per_gpu:4653.342705895243 mfu:21.950611485842206 +Step 21265 | loss:0.7610644698143005 lr:3.6834058971255356e-05 tokens_per_second_per_gpu:4004.0360042489356 mfu:18.887720991029678 +Step 21266 | loss:0.7191062569618225 lr:3.682987582920468e-05 tokens_per_second_per_gpu:4724.974080927067 mfu:22.28850890344015 +Step 21267 | loss:0.7213726043701172 lr:3.682569278620818e-05 tokens_per_second_per_gpu:3418.57454059015 mfu:16.12599937692515 +Step 21268 | loss:0.6138957142829895 lr:3.682150984229732e-05 tokens_per_second_per_gpu:3514.569348251097 mfu:16.578823263064912 +Step 21269 | loss:0.643531084060669 lr:3.6817326997503575e-05 tokens_per_second_per_gpu:3843.05993451801 mfu:18.12836940475927 +Step 21270 | loss:0.6767734289169312 lr:3.681314425185839e-05 tokens_per_second_per_gpu:4622.586196541278 mfu:21.80552778361794 +Step 21271 | loss:0.6295291185379028 lr:3.680896160539326e-05 tokens_per_second_per_gpu:5335.756958350695 mfu:25.16967594655212 +Step 21272 | loss:0.6265676021575928 lr:3.680477905813958e-05 tokens_per_second_per_gpu:4186.984457851706 mfu:19.75072006089884 +Step 21273 | loss:0.6563529372215271 lr:3.680059661012885e-05 tokens_per_second_per_gpu:4479.523220662694 mfu:21.130675317337992 +Step 21274 | loss:0.6439270377159119 lr:3.679641426139251e-05 tokens_per_second_per_gpu:4158.216299729057 mfu:19.615015750680392 +Step 21275 | loss:0.7259176969528198 lr:3.679223201196203e-05 tokens_per_second_per_gpu:4950.783938364573 mfu:23.353692528107338 +Step 21276 | loss:0.7898859977722168 lr:3.678804986186886e-05 tokens_per_second_per_gpu:5332.599470206996 mfu:25.15478153625526 +Step 21277 | loss:0.6982170343399048 lr:3.6783867811144434e-05 tokens_per_second_per_gpu:4133.657346519566 mfu:19.499166978201732 +Step 21278 | loss:0.6447883248329163 lr:3.677968585982025e-05 tokens_per_second_per_gpu:5315.544935232188 mfu:25.074332384975378 +Step 21279 | loss:0.77333003282547 lr:3.677550400792773e-05 tokens_per_second_per_gpu:4079.2842639957084 mfu:19.242679870932196 +Step 21280 | loss:0.6811162233352661 lr:3.677132225549833e-05 tokens_per_second_per_gpu:5339.61246555673 mfu:25.187863031860342 +Step 21281 | loss:0.8451210260391235 lr:3.6767140602563496e-05 tokens_per_second_per_gpu:3904.9765827235597 mfu:18.42044079841436 +Step 21282 | loss:0.7814062237739563 lr:3.6762959049154685e-05 tokens_per_second_per_gpu:4228.441001860829 mfu:19.946277652205644 +Step 21283 | loss:0.7443264126777649 lr:3.675877759530336e-05 tokens_per_second_per_gpu:4743.670116483106 mfu:22.376701293030575 +Step 21284 | loss:0.8434737920761108 lr:3.675459624104094e-05 tokens_per_second_per_gpu:4795.592212771928 mfu:22.621626680048074 +Step 21285 | loss:0.7182053923606873 lr:3.675041498639892e-05 tokens_per_second_per_gpu:4862.394865721478 mfu:22.93674619171899 +Step 21286 | loss:0.7524468302726746 lr:3.674623383140871e-05 tokens_per_second_per_gpu:4763.4029581013065 mfu:22.46978447371358 +Step 21287 | loss:0.7811185717582703 lr:3.674205277610177e-05 tokens_per_second_per_gpu:4579.752728351729 mfu:21.603475006024574 +Step 21288 | loss:0.666174054145813 lr:3.673787182050952e-05 tokens_per_second_per_gpu:4361.6190282506905 mfu:20.574501125200996 +Step 21289 | loss:0.6467351913452148 lr:3.673369096466344e-05 tokens_per_second_per_gpu:3861.05337635096 mfu:18.21324753988304 +Step 21290 | loss:0.7796517610549927 lr:3.672951020859495e-05 tokens_per_second_per_gpu:5358.849065039669 mfu:25.27860535372368 +Step 21291 | loss:0.6913648247718811 lr:3.672532955233553e-05 tokens_per_second_per_gpu:5309.436240514456 mfu:25.04551662974104 +Step 21292 | loss:0.7359849810600281 lr:3.672114899591659e-05 tokens_per_second_per_gpu:5338.476422848494 mfu:25.18250412457672 +Step 21293 | loss:0.8754919171333313 lr:3.67169685393696e-05 tokens_per_second_per_gpu:4466.314849322578 mfu:21.068369176146135 +Step 21294 | loss:0.683887243270874 lr:3.671278818272596e-05 tokens_per_second_per_gpu:4846.897616323108 mfu:22.86364302220348 +Step 21295 | loss:0.731785237789154 lr:3.6708607926017145e-05 tokens_per_second_per_gpu:5245.986971911586 mfu:24.74621560417978 +Step 21296 | loss:0.806562066078186 lr:3.6704427769274585e-05 tokens_per_second_per_gpu:5311.420523685568 mfu:25.05487683201279 +Step 21297 | loss:0.7447496652603149 lr:3.670024771252971e-05 tokens_per_second_per_gpu:4199.913166422799 mfu:19.811706985093828 +Step 21298 | loss:0.6051830053329468 lr:3.669606775581399e-05 tokens_per_second_per_gpu:5353.146734370071 mfu:25.25170648703725 +Step 21299 | loss:0.7811520099639893 lr:3.669188789915883e-05 tokens_per_second_per_gpu:5337.233780601132 mfu:25.176642369079286 +Step 21300 | loss:0.7765501141548157 lr:3.668770814259569e-05 tokens_per_second_per_gpu:4612.856568472765 mfu:21.759631467973037 +Step 21301 | loss:0.7244971394538879 lr:3.668352848615598e-05 tokens_per_second_per_gpu:4731.264671873956 mfu:22.318182694222116 +Step 21302 | loss:0.7377764582633972 lr:3.667934892987116e-05 tokens_per_second_per_gpu:4989.548545667474 mfu:23.536551794679134 +Step 21303 | loss:0.7804619669914246 lr:3.667516947377266e-05 tokens_per_second_per_gpu:4354.11810697271 mfu:20.539117999743624 +Step 21304 | loss:0.6722108125686646 lr:3.6670990117891897e-05 tokens_per_second_per_gpu:3414.4238782085276 mfu:16.106419994295013 +Step 21305 | loss:0.6819729208946228 lr:3.666681086226032e-05 tokens_per_second_per_gpu:4168.918705159376 mfu:19.66550082311387 +Step 21306 | loss:0.5827726125717163 lr:3.6662631706909366e-05 tokens_per_second_per_gpu:3209.0660117154944 mfu:15.13771190038238 +Step 21307 | loss:0.692933976650238 lr:3.665845265187048e-05 tokens_per_second_per_gpu:3590.2290512273944 mfu:16.935722990880556 +Step 21308 | loss:0.7736794948577881 lr:3.665427369717504e-05 tokens_per_second_per_gpu:4626.997714734336 mfu:21.826337667617214 +Step 21309 | loss:0.7637882828712463 lr:3.6650094842854534e-05 tokens_per_second_per_gpu:5323.337292921492 mfu:25.111090265709137 +Step 21310 | loss:0.6465047001838684 lr:3.6645916088940355e-05 tokens_per_second_per_gpu:4184.459802221328 mfu:19.738810829539677 +Step 21311 | loss:0.7230924963951111 lr:3.6641737435463954e-05 tokens_per_second_per_gpu:5137.5407551718 mfu:24.234656296985683 +Step 21312 | loss:0.6173839569091797 lr:3.663755888245675e-05 tokens_per_second_per_gpu:5296.32738755376 mfu:24.98367993372449 +Step 21313 | loss:0.6957635879516602 lr:3.663338042995015e-05 tokens_per_second_per_gpu:5363.3547282022455 mfu:25.29985933560696 +Step 21314 | loss:0.6111065745353699 lr:3.6629202077975634e-05 tokens_per_second_per_gpu:4536.162886005922 mfu:21.3978542824854 +Step 21315 | loss:0.7501628994941711 lr:3.662502382656458e-05 tokens_per_second_per_gpu:5348.182603229634 mfu:25.2282898334797 +Step 21316 | loss:0.6840624213218689 lr:3.662084567574843e-05 tokens_per_second_per_gpu:5355.374053926381 mfu:25.262213133404266 +Step 21317 | loss:0.6632756590843201 lr:3.66166676255586e-05 tokens_per_second_per_gpu:4729.385085571619 mfu:22.3093163649435 +Step 21318 | loss:0.6460206508636475 lr:3.6612489676026515e-05 tokens_per_second_per_gpu:3848.8788412355443 mfu:18.15581818055374 +Step 21319 | loss:0.7157758474349976 lr:3.660831182718362e-05 tokens_per_second_per_gpu:5381.146142940888 mfu:25.383784474457503 +Step 21320 | loss:0.8198585510253906 lr:3.660413407906131e-05 tokens_per_second_per_gpu:5262.027997663188 mfu:24.8218838595313 +Step 21321 | loss:0.689752995967865 lr:3.659995643169103e-05 tokens_per_second_per_gpu:5305.230908412718 mfu:25.0256793607135 +Step 21322 | loss:0.8515996932983398 lr:3.6595778885104184e-05 tokens_per_second_per_gpu:4962.3361182316685 mfu:23.408186131545552 +Step 21323 | loss:0.6965605616569519 lr:3.659160143933219e-05 tokens_per_second_per_gpu:3621.5172914290238 mfu:17.083314958235892 +Step 21324 | loss:0.6459088325500488 lr:3.6587424094406464e-05 tokens_per_second_per_gpu:3906.041679862435 mfu:18.425465043342747 +Step 21325 | loss:0.8516224026679993 lr:3.6583246850358444e-05 tokens_per_second_per_gpu:5062.221076891827 mfu:23.879360523676798 +Step 21326 | loss:0.697761058807373 lr:3.6579069707219526e-05 tokens_per_second_per_gpu:3829.125147206087 mfu:18.06263663548905 +Step 21327 | loss:0.7984035611152649 lr:3.6574892665021145e-05 tokens_per_second_per_gpu:4192.874984317135 mfu:19.77850667926372 +Step 21328 | loss:0.5787899494171143 lr:3.6570715723794715e-05 tokens_per_second_per_gpu:4089.6668471870516 mfu:19.291656287297123 +Step 21329 | loss:0.6960341930389404 lr:3.6566538883571645e-05 tokens_per_second_per_gpu:4689.6658485677 mfu:22.121953947195376 +Step 21330 | loss:0.7830429673194885 lr:3.6562362144383336e-05 tokens_per_second_per_gpu:4440.087897836069 mfu:20.944652171204922 +Step 21331 | loss:0.6545770764350891 lr:3.6558185506261236e-05 tokens_per_second_per_gpu:3783.1260666917697 mfu:17.845651124450594 +Step 21332 | loss:0.6627026796340942 lr:3.655400896923672e-05 tokens_per_second_per_gpu:3821.7955351781375 mfu:18.028061604992445 +Step 21333 | loss:0.731926441192627 lr:3.654983253334122e-05 tokens_per_second_per_gpu:4649.322165308379 mfu:21.93164590991901 +Step 21334 | loss:0.7088428735733032 lr:3.654565619860615e-05 tokens_per_second_per_gpu:3681.4832590881324 mfu:17.366184658933005 +Step 21335 | loss:0.7609871625900269 lr:3.654147996506292e-05 tokens_per_second_per_gpu:5379.993768574965 mfu:25.378348527958103 +Step 21336 | loss:0.7961285710334778 lr:3.653730383274295e-05 tokens_per_second_per_gpu:5005.4636554883555 mfu:23.61162608310197 +Step 21337 | loss:0.7027596831321716 lr:3.653312780167761e-05 tokens_per_second_per_gpu:4917.896758182726 mfu:23.198558088058025 +Step 21338 | loss:0.8858675956726074 lr:3.652895187189834e-05 tokens_per_second_per_gpu:4845.074011564398 mfu:22.855040767417584 +Step 21339 | loss:0.6029646396636963 lr:3.6524776043436555e-05 tokens_per_second_per_gpu:3613.172685852726 mfu:17.04395202999597 +Step 21340 | loss:0.6857850551605225 lr:3.6520600316323625e-05 tokens_per_second_per_gpu:5191.641696878399 mfu:24.489859669588196 +Step 21341 | loss:0.7718991041183472 lr:3.6516424690591e-05 tokens_per_second_per_gpu:5334.316792941505 mfu:25.16288244059941 +Step 21342 | loss:0.6219098567962646 lr:3.6512249166270053e-05 tokens_per_second_per_gpu:5289.551444250622 mfu:24.95171665306845 +Step 21343 | loss:0.7451194524765015 lr:3.6508073743392235e-05 tokens_per_second_per_gpu:5168.5100140776585 mfu:24.380743575145047 +Step 21344 | loss:0.6727365851402283 lr:3.650389842198888e-05 tokens_per_second_per_gpu:3860.0535251822644 mfu:18.2085310713281 +Step 21345 | loss:0.7795382142066956 lr:3.649972320209144e-05 tokens_per_second_per_gpu:4716.803406227903 mfu:22.249966436823335 +Step 21346 | loss:0.7028669118881226 lr:3.649554808373129e-05 tokens_per_second_per_gpu:5041.199421395831 mfu:23.780197788037913 +Step 21347 | loss:0.6549018621444702 lr:3.6491373066939856e-05 tokens_per_second_per_gpu:3871.018367926043 mfu:18.260254105345584 +Step 21348 | loss:0.7191205024719238 lr:3.648719815174853e-05 tokens_per_second_per_gpu:5077.515940466132 mfu:23.95150900472961 +Step 21349 | loss:0.7375516891479492 lr:3.6483023338188705e-05 tokens_per_second_per_gpu:4737.334017967049 mfu:22.346812835280456 +Step 21350 | loss:0.74980628490448 lr:3.64788486262918e-05 tokens_per_second_per_gpu:3709.4877115321465 mfu:17.4982864391637 +Step 21351 | loss:0.5452383756637573 lr:3.6474674016089195e-05 tokens_per_second_per_gpu:5362.027962467967 mfu:25.29360075526879 +Step 21352 | loss:0.6990439295768738 lr:3.647049950761229e-05 tokens_per_second_per_gpu:4274.265786421192 mfu:20.162440979491343 +Step 21353 | loss:0.7776862382888794 lr:3.6466325100892475e-05 tokens_per_second_per_gpu:5001.243475116607 mfu:23.591718772251415 +Step 21354 | loss:0.5908600091934204 lr:3.646215079596117e-05 tokens_per_second_per_gpu:5374.729131552875 mfu:25.353514336885205 +Step 21355 | loss:0.6399117112159729 lr:3.6457976592849754e-05 tokens_per_second_per_gpu:5328.331575213819 mfu:25.13464914739447 +Step 21356 | loss:0.8263823986053467 lr:3.64538024915896e-05 tokens_per_second_per_gpu:5246.021992547164 mfu:24.74638080249294 +Step 21357 | loss:0.7026018500328064 lr:3.6449628492212154e-05 tokens_per_second_per_gpu:3840.6588034013416 mfu:18.11704285960673 +Step 21358 | loss:0.7571287751197815 lr:3.644545459474879e-05 tokens_per_second_per_gpu:4266.37780554793 mfu:20.12523202788385 +Step 21359 | loss:0.739201009273529 lr:3.644128079923088e-05 tokens_per_second_per_gpu:4898.5396974486275 mfu:23.107247529920162 +Step 21360 | loss:0.8077534437179565 lr:3.643710710568981e-05 tokens_per_second_per_gpu:4026.7893197407334 mfu:18.99505226232042 +Step 21361 | loss:0.8255887031555176 lr:3.643293351415701e-05 tokens_per_second_per_gpu:4064.0423045788248 mfu:19.170780947816322 +Step 21362 | loss:0.685864269733429 lr:3.642876002466382e-05 tokens_per_second_per_gpu:4527.145717718459 mfu:21.355318761185885 +Step 21363 | loss:0.7670729756355286 lr:3.642458663724169e-05 tokens_per_second_per_gpu:3323.4344584873993 mfu:15.677207377074621 +Step 21364 | loss:0.6065678596496582 lr:3.642041335192196e-05 tokens_per_second_per_gpu:4119.111234012455 mfu:19.43055048368297 +Step 21365 | loss:0.6374995112419128 lr:3.6416240168736046e-05 tokens_per_second_per_gpu:5389.075143467398 mfu:25.421186922768044 +Step 21366 | loss:0.7836629748344421 lr:3.641206708771531e-05 tokens_per_second_per_gpu:4393.59649023176 mfu:20.725344269283312 +Step 21367 | loss:0.8279834985733032 lr:3.640789410889115e-05 tokens_per_second_per_gpu:4125.310348063818 mfu:19.459792762341177 +Step 21368 | loss:0.7472944259643555 lr:3.640372123229496e-05 tokens_per_second_per_gpu:3642.0401619658583 mfu:17.180124840120925 +Step 21369 | loss:0.5786114931106567 lr:3.639954845795811e-05 tokens_per_second_per_gpu:4413.2855026751995 mfu:20.818220700270935 +Step 21370 | loss:0.7279161214828491 lr:3.6395375785912e-05 tokens_per_second_per_gpu:5314.79913104219 mfu:25.07081429936384 +Step 21371 | loss:0.7452067732810974 lr:3.6391203216188e-05 tokens_per_second_per_gpu:5316.259882842022 mfu:25.077704914833603 +Step 21372 | loss:0.7136008739471436 lr:3.6387030748817516e-05 tokens_per_second_per_gpu:5337.287928713044 mfu:25.176897794961796 +Step 21373 | loss:0.7009374499320984 lr:3.638285838383188e-05 tokens_per_second_per_gpu:4347.840348701092 mfu:20.50950474288039 +Step 21374 | loss:0.7635966539382935 lr:3.637868612126252e-05 tokens_per_second_per_gpu:4839.109121002376 mfu:22.826903360921257 +Step 21375 | loss:0.7005266547203064 lr:3.63745139611408e-05 tokens_per_second_per_gpu:5308.209753477666 mfu:25.039731081128235 +Step 21376 | loss:0.5482735633850098 lr:3.6370341903498087e-05 tokens_per_second_per_gpu:5368.29229752909 mfu:25.323150692557885 +Step 21377 | loss:0.6754598617553711 lr:3.636616994836578e-05 tokens_per_second_per_gpu:4311.720454237243 mfu:20.339121038004947 +Step 21378 | loss:0.6870815753936768 lr:3.636199809577524e-05 tokens_per_second_per_gpu:5369.578799058396 mfu:25.329219339771413 +Step 21379 | loss:0.5812488198280334 lr:3.6357826345757884e-05 tokens_per_second_per_gpu:5346.131867378375 mfu:25.218616162577696 +Step 21380 | loss:0.6219834685325623 lr:3.635365469834503e-05 tokens_per_second_per_gpu:5320.711447722249 mfu:25.098703705888923 +Step 21381 | loss:0.6745697855949402 lr:3.6349483153568085e-05 tokens_per_second_per_gpu:5188.305216072849 mfu:24.47412092036609 +Step 21382 | loss:0.6951267123222351 lr:3.634531171145841e-05 tokens_per_second_per_gpu:5298.348194085291 mfu:24.993212422918866 +Step 21383 | loss:0.7703604102134705 lr:3.63411403720474e-05 tokens_per_second_per_gpu:5272.749673387662 mfu:24.872459833230156 +Step 21384 | loss:0.7512645721435547 lr:3.633696913536642e-05 tokens_per_second_per_gpu:4586.595124512538 mfu:21.635751756148274 +Step 21385 | loss:0.6776605844497681 lr:3.6332798001446814e-05 tokens_per_second_per_gpu:5338.923188023685 mfu:25.1846115921341 +Step 21386 | loss:0.582217812538147 lr:3.632862697032e-05 tokens_per_second_per_gpu:3963.467699776591 mfu:18.696353377167146 +Step 21387 | loss:0.796384871006012 lr:3.6324456042017334e-05 tokens_per_second_per_gpu:3649.4552599867498 mfu:17.21510312263165 +Step 21388 | loss:0.7042754292488098 lr:3.6320285216570174e-05 tokens_per_second_per_gpu:4136.5600707349195 mfu:19.51285962358137 +Step 21389 | loss:0.7766050696372986 lr:3.631611449400988e-05 tokens_per_second_per_gpu:4984.606048029447 mfu:23.513237190043508 +Step 21390 | loss:0.6843386888504028 lr:3.631194387436784e-05 tokens_per_second_per_gpu:4039.732431182751 mfu:19.056107127314945 +Step 21391 | loss:0.7766876220703125 lr:3.630777335767542e-05 tokens_per_second_per_gpu:5354.528792383241 mfu:25.258225890488784 +Step 21392 | loss:0.7305784821510315 lr:3.630360294396398e-05 tokens_per_second_per_gpu:5319.980538432373 mfu:25.095255882062588 +Step 21393 | loss:0.5089524388313293 lr:3.629943263326489e-05 tokens_per_second_per_gpu:5322.120503693131 mfu:25.105350463313307 +Step 21394 | loss:0.7517539262771606 lr:3.629526242560953e-05 tokens_per_second_per_gpu:5318.847502281417 mfu:25.089911157222634 +Step 21395 | loss:0.7977705597877502 lr:3.629109232102924e-05 tokens_per_second_per_gpu:5357.217292172355 mfu:25.2709080026994 +Step 21396 | loss:0.754909873008728 lr:3.628692231955539e-05 tokens_per_second_per_gpu:4437.806079641654 mfu:20.933888445463538 +Step 21397 | loss:0.6357678771018982 lr:3.628275242121935e-05 tokens_per_second_per_gpu:5342.642691094824 mfu:25.202157122733322 +Step 21398 | loss:0.7672536969184875 lr:3.627858262605247e-05 tokens_per_second_per_gpu:3917.650935687721 mfu:18.480227883813914 +Step 21399 | loss:0.7376109957695007 lr:3.6274412934086134e-05 tokens_per_second_per_gpu:4933.128942807644 mfu:23.27041090181245 +Step 21400 | loss:0.688211977481842 lr:3.6270243345351685e-05 tokens_per_second_per_gpu:3247.615340455139 mfu:15.319555661240987 +Step 21401 | loss:0.6197413802146912 lr:3.62660738598805e-05 tokens_per_second_per_gpu:4145.284223033495 mfu:19.554012938467626 +Step 21402 | loss:0.6946650743484497 lr:3.626190447770391e-05 tokens_per_second_per_gpu:5372.969480776136 mfu:25.345213763943946 +Step 21403 | loss:0.6721519827842712 lr:3.6257735198853296e-05 tokens_per_second_per_gpu:4645.769702212193 mfu:21.914888335381594 +Step 21404 | loss:0.7592722773551941 lr:3.625356602336001e-05 tokens_per_second_per_gpu:4255.065001040792 mfu:20.0718675988602 +Step 21405 | loss:0.565025806427002 lr:3.624939695125539e-05 tokens_per_second_per_gpu:4955.144016391912 mfu:23.374259760068316 +Step 21406 | loss:0.6684471368789673 lr:3.6245227982570826e-05 tokens_per_second_per_gpu:4531.999469341625 mfu:21.378214735727777 +Step 21407 | loss:0.7101530432701111 lr:3.624105911733766e-05 tokens_per_second_per_gpu:5328.793755249974 mfu:25.13682932948106 +Step 21408 | loss:0.8095120787620544 lr:3.6236890355587255e-05 tokens_per_second_per_gpu:5088.320514324192 mfu:24.002476023068475 +Step 21409 | loss:0.5887620449066162 lr:3.6232721697350937e-05 tokens_per_second_per_gpu:4389.959184660862 mfu:20.708186478317064 +Step 21410 | loss:0.845596194267273 lr:3.622855314266007e-05 tokens_per_second_per_gpu:5348.805650694849 mfu:25.231228854675976 +Step 21411 | loss:0.742393434047699 lr:3.6224384691546024e-05 tokens_per_second_per_gpu:4399.863237022415 mfu:20.754905583112087 +Step 21412 | loss:0.7888190746307373 lr:3.6220216344040124e-05 tokens_per_second_per_gpu:5351.103079638887 mfu:25.242066218986803 +Step 21413 | loss:0.7256109118461609 lr:3.6216048100173747e-05 tokens_per_second_per_gpu:3734.017431425399 mfu:17.61399731310195 +Step 21414 | loss:0.7177401185035706 lr:3.621187995997823e-05 tokens_per_second_per_gpu:5326.911704340003 mfu:25.127951374227727 +Step 21415 | loss:0.8368609547615051 lr:3.620771192348494e-05 tokens_per_second_per_gpu:4139.2342686538805 mfu:19.525474271430326 +Step 21416 | loss:0.7120487093925476 lr:3.6203543990725186e-05 tokens_per_second_per_gpu:3972.8554754121546 mfu:18.740637116559785 +Step 21417 | loss:0.6524665951728821 lr:3.619937616173035e-05 tokens_per_second_per_gpu:4755.604374493212 mfu:22.43299722425863 +Step 21418 | loss:0.7324339747428894 lr:3.619520843653175e-05 tokens_per_second_per_gpu:3797.170020364667 mfu:17.91189885007123 +Step 21419 | loss:0.6323224902153015 lr:3.619104081516076e-05 tokens_per_second_per_gpu:5350.157024373277 mfu:25.237603514885247 +Step 21420 | loss:0.7068295478820801 lr:3.6186873297648716e-05 tokens_per_second_per_gpu:4510.493166423908 mfu:21.27676583551126 +Step 21421 | loss:0.49706849455833435 lr:3.6182705884026944e-05 tokens_per_second_per_gpu:4064.39937667497 mfu:19.172465317816624 +Step 21422 | loss:0.6905713081359863 lr:3.6178538574326824e-05 tokens_per_second_per_gpu:5345.961570605506 mfu:25.217812843644445 +Step 21423 | loss:0.6730847954750061 lr:3.617437136857968e-05 tokens_per_second_per_gpu:5141.38641879135 mfu:24.25279694841697 +Step 21424 | loss:0.7791109085083008 lr:3.6170204266816855e-05 tokens_per_second_per_gpu:4524.204909725578 mfu:21.341446468129877 +Step 21425 | loss:0.8845055103302002 lr:3.616603726906967e-05 tokens_per_second_per_gpu:4453.015613213585 mfu:21.005634410336356 +Step 21426 | loss:0.7241317629814148 lr:3.61618703753695e-05 tokens_per_second_per_gpu:3980.4068992421817 mfu:18.776258471171786 +Step 21427 | loss:0.7074669003486633 lr:3.615770358574766e-05 tokens_per_second_per_gpu:5195.094591006954 mfu:24.506147560317075 +Step 21428 | loss:0.7922089695930481 lr:3.61535369002355e-05 tokens_per_second_per_gpu:4243.185035439714 mfu:20.01582777418877 +Step 21429 | loss:0.6545483469963074 lr:3.614937031886435e-05 tokens_per_second_per_gpu:5345.112566944316 mfu:25.21380794851965 +Step 21430 | loss:0.5370757579803467 lr:3.614520384166557e-05 tokens_per_second_per_gpu:5330.533601094993 mfu:25.145036479180327 +Step 21431 | loss:0.7272171378135681 lr:3.614103746867047e-05 tokens_per_second_per_gpu:4160.198175116609 mfu:19.624364595026286 +Step 21432 | loss:0.827671468257904 lr:3.613687119991039e-05 tokens_per_second_per_gpu:2878.4038380476704 mfu:13.577922010406914 +Step 21433 | loss:0.6544405221939087 lr:3.613270503541668e-05 tokens_per_second_per_gpu:5345.349057106825 mfu:25.21492351296559 +Step 21434 | loss:0.6381016373634338 lr:3.612853897522065e-05 tokens_per_second_per_gpu:5282.111866163522 mfu:24.916622893908688 +Step 21435 | loss:0.7187566757202148 lr:3.6124373019353655e-05 tokens_per_second_per_gpu:3932.4363055068206 mfu:18.549972995894823 +Step 21436 | loss:0.7083607912063599 lr:3.612020716784703e-05 tokens_per_second_per_gpu:2892.7287769743853 mfu:13.645495191410832 +Step 21437 | loss:0.6443992853164673 lr:3.6116041420732103e-05 tokens_per_second_per_gpu:3908.877226938299 mfu:18.438840802693814 +Step 21438 | loss:0.6042605638504028 lr:3.6111875778040197e-05 tokens_per_second_per_gpu:3684.8094902726625 mfu:17.381875058943994 +Step 21439 | loss:0.5516244173049927 lr:3.610771023980263e-05 tokens_per_second_per_gpu:5343.917103711327 mfu:25.208168744482585 +Step 21440 | loss:0.7264458537101746 lr:3.6103544806050753e-05 tokens_per_second_per_gpu:5331.435204604862 mfu:25.149289496765796 +Step 21441 | loss:0.5677086114883423 lr:3.6099379476815884e-05 tokens_per_second_per_gpu:4912.1758660419555 mfu:23.171571663745155 +Step 21442 | loss:0.6403548717498779 lr:3.609521425212936e-05 tokens_per_second_per_gpu:3686.882615027702 mfu:17.39165434212507 +Step 21443 | loss:0.6499208211898804 lr:3.609104913202252e-05 tokens_per_second_per_gpu:4078.601519380479 mfu:19.239459248093922 +Step 21444 | loss:0.6764192581176758 lr:3.6086884116526666e-05 tokens_per_second_per_gpu:5067.30890709389 mfu:23.90336068681328 +Step 21445 | loss:0.5888157486915588 lr:3.6082719205673116e-05 tokens_per_second_per_gpu:5313.463505309485 mfu:25.06451392490142 +Step 21446 | loss:0.7668196558952332 lr:3.607855439949323e-05 tokens_per_second_per_gpu:4428.369563715956 mfu:20.889374789806382 +Step 21447 | loss:0.6453162431716919 lr:3.6074389698018294e-05 tokens_per_second_per_gpu:4560.8276883820845 mfu:21.514202363542985 +Step 21448 | loss:0.7798842191696167 lr:3.6070225101279656e-05 tokens_per_second_per_gpu:5265.073928188477 mfu:24.836252033508426 +Step 21449 | loss:0.6445991396903992 lr:3.606606060930864e-05 tokens_per_second_per_gpu:5309.706217885474 mfu:25.046790159063022 +Step 21450 | loss:0.6414913535118103 lr:3.606189622213654e-05 tokens_per_second_per_gpu:5312.002121048662 mfu:25.057620326005253 +Step 21451 | loss:0.7328814268112183 lr:3.605773193979472e-05 tokens_per_second_per_gpu:4407.647896656352 mfu:20.79162715080526 +Step 21452 | loss:0.697975754737854 lr:3.605356776231448e-05 tokens_per_second_per_gpu:4331.279792954573 mfu:20.431385775901898 +Step 21453 | loss:0.5799426436424255 lr:3.6049403689727134e-05 tokens_per_second_per_gpu:5369.764096419065 mfu:25.33009341903668 +Step 21454 | loss:0.7561721205711365 lr:3.6045239722063995e-05 tokens_per_second_per_gpu:5305.567000921907 mfu:25.027264766422636 +Step 21455 | loss:0.6698192954063416 lr:3.604107585935638e-05 tokens_per_second_per_gpu:5343.270026322582 mfu:25.20511637003709 +Step 21456 | loss:0.6347898840904236 lr:3.6036912101635626e-05 tokens_per_second_per_gpu:4555.730576716911 mfu:21.490158418161435 +Step 21457 | loss:0.6575420498847961 lr:3.603274844893303e-05 tokens_per_second_per_gpu:4611.480727957375 mfu:21.75314139351935 +Step 21458 | loss:0.6580559611320496 lr:3.6028584901279924e-05 tokens_per_second_per_gpu:5347.890288063784 mfu:25.226910932968273 +Step 21459 | loss:0.770318865776062 lr:3.602442145870763e-05 tokens_per_second_per_gpu:4600.458394424294 mfu:21.70114716563978 +Step 21460 | loss:0.6805554032325745 lr:3.6020258121247416e-05 tokens_per_second_per_gpu:5345.934005461099 mfu:25.217682814155182 +Step 21461 | loss:0.6953670978546143 lr:3.601609488893063e-05 tokens_per_second_per_gpu:5358.6107013715455 mfu:25.277480951632104 +Step 21462 | loss:0.6891027688980103 lr:3.6011931761788586e-05 tokens_per_second_per_gpu:3972.6039051828943 mfu:18.73945041686606 +Step 21463 | loss:0.7672543525695801 lr:3.6007768739852587e-05 tokens_per_second_per_gpu:4066.3266796071994 mfu:19.181556734579736 +Step 21464 | loss:0.6892637610435486 lr:3.600360582315393e-05 tokens_per_second_per_gpu:3130.76301202332 mfu:14.768343290965019 +Step 21465 | loss:0.6068773865699768 lr:3.599944301172396e-05 tokens_per_second_per_gpu:4717.562252209219 mfu:22.25354604278968 +Step 21466 | loss:0.7769519090652466 lr:3.599528030559397e-05 tokens_per_second_per_gpu:4453.623310862442 mfu:21.008501023829922 +Step 21467 | loss:0.5826985239982605 lr:3.599111770479525e-05 tokens_per_second_per_gpu:4405.636750883646 mfu:20.782140233058556 +Step 21468 | loss:0.7320730090141296 lr:3.598695520935912e-05 tokens_per_second_per_gpu:3987.335514331137 mfu:18.80894193069002 +Step 21469 | loss:0.6596887111663818 lr:3.598279281931689e-05 tokens_per_second_per_gpu:5183.917448964827 mfu:24.45342307428728 +Step 21470 | loss:0.6584938764572144 lr:3.597863053469987e-05 tokens_per_second_per_gpu:4333.656363006009 mfu:20.442596462319184 +Step 21471 | loss:0.7309578657150269 lr:3.597446835553935e-05 tokens_per_second_per_gpu:4225.225569622317 mfu:19.931109909727276 +Step 21472 | loss:0.7594457268714905 lr:3.5970306281866646e-05 tokens_per_second_per_gpu:4548.304759520505 mfu:21.455129571470927 +Step 21473 | loss:0.6925680041313171 lr:3.5966144313713073e-05 tokens_per_second_per_gpu:4280.418797716748 mfu:20.191465783584462 +Step 21474 | loss:0.8054245114326477 lr:3.596198245110991e-05 tokens_per_second_per_gpu:4664.067716304467 mfu:22.001203189817982 +Step 21475 | loss:0.6583626866340637 lr:3.595782069408845e-05 tokens_per_second_per_gpu:4921.302838476907 mfu:23.214625149942897 +Step 21476 | loss:0.660816490650177 lr:3.5953659042680034e-05 tokens_per_second_per_gpu:4955.13223609869 mfu:23.37420419041526 +Step 21477 | loss:0.5984272360801697 lr:3.594949749691592e-05 tokens_per_second_per_gpu:4448.251042249228 mfu:20.98315911617785 +Step 21478 | loss:0.6897880434989929 lr:3.594533605682745e-05 tokens_per_second_per_gpu:4278.9723178743125 mfu:20.184642491372784 +Step 21479 | loss:0.8004119396209717 lr:3.594117472244589e-05 tokens_per_second_per_gpu:3682.533030377018 mfu:17.371136609212126 +Step 21480 | loss:0.9337186217308044 lr:3.593701349380254e-05 tokens_per_second_per_gpu:5332.001600329163 mfu:25.151961282034403 +Step 21481 | loss:0.5856849551200867 lr:3.5932852370928736e-05 tokens_per_second_per_gpu:5348.022268720069 mfu:25.227533508242196 +Step 21482 | loss:0.6247473955154419 lr:3.592869135385573e-05 tokens_per_second_per_gpu:3882.7196462825773 mfu:18.315451031797313 +Step 21483 | loss:0.6200068593025208 lr:3.592453044261483e-05 tokens_per_second_per_gpu:5369.270145746612 mfu:25.327763369438042 +Step 21484 | loss:0.720054566860199 lr:3.592036963723732e-05 tokens_per_second_per_gpu:4300.553612623488 mfu:20.28644514085322 +Step 21485 | loss:0.7043232917785645 lr:3.5916208937754515e-05 tokens_per_second_per_gpu:4986.881770730485 mfu:23.523972162302634 +Step 21486 | loss:0.742536723613739 lr:3.5912048344197695e-05 tokens_per_second_per_gpu:5342.082626862422 mfu:25.199515204192274 +Step 21487 | loss:0.7228601574897766 lr:3.5907887856598165e-05 tokens_per_second_per_gpu:4596.8205331344325 mfu:21.683986753252363 +Step 21488 | loss:0.8511989712715149 lr:3.5903727474987224e-05 tokens_per_second_per_gpu:3809.55195189137 mfu:17.97030653365805 +Step 21489 | loss:0.7930310368537903 lr:3.589956719939613e-05 tokens_per_second_per_gpu:3797.1590158891768 mfu:17.911846940082786 +Step 21490 | loss:0.8454345464706421 lr:3.589540702985619e-05 tokens_per_second_per_gpu:5090.257973682604 mfu:24.011615349426755 +Step 21491 | loss:0.7131573557853699 lr:3.589124696639869e-05 tokens_per_second_per_gpu:4159.014272371064 mfu:19.618779923780913 +Step 21492 | loss:0.7570838332176208 lr:3.5887087009054926e-05 tokens_per_second_per_gpu:5011.335478780505 mfu:23.6393244754074 +Step 21493 | loss:0.588306188583374 lr:3.588292715785617e-05 tokens_per_second_per_gpu:4669.82612944613 mfu:22.0283666070938 +Step 21494 | loss:0.6784405708312988 lr:3.587876741283374e-05 tokens_per_second_per_gpu:5329.276894695427 mfu:25.139108380677346 +Step 21495 | loss:0.6759214997291565 lr:3.587460777401891e-05 tokens_per_second_per_gpu:4456.533026102211 mfu:21.022226647064524 +Step 21496 | loss:0.7495703101158142 lr:3.587044824144293e-05 tokens_per_second_per_gpu:5357.289601400303 mfu:25.271249097664857 +Step 21497 | loss:0.6676727533340454 lr:3.586628881513712e-05 tokens_per_second_per_gpu:3708.961960936825 mfu:17.49580638390327 +Step 21498 | loss:0.7780130505561829 lr:3.586212949513276e-05 tokens_per_second_per_gpu:3861.5723786181793 mfu:18.21569576212843 +Step 21499 | loss:0.6488595604896545 lr:3.5857970281461126e-05 tokens_per_second_per_gpu:5305.9029342006415 mfu:25.028849421014556 +Step 21500 | loss:0.6415520310401917 lr:3.585381117415349e-05 tokens_per_second_per_gpu:4462.726400012886 mfu:21.05144184849973 +Step 21501 | loss:0.7097676396369934 lr:3.584965217324117e-05 tokens_per_second_per_gpu:3696.588033823061 mfu:17.437436458497977 +Step 21502 | loss:0.9679562449455261 lr:3.584549327875542e-05 tokens_per_second_per_gpu:3370.6999792928827 mfu:15.900166903043502 +Step 21503 | loss:0.7343831062316895 lr:3.584133449072751e-05 tokens_per_second_per_gpu:3975.6698997766116 mfu:18.753913236477242 +Step 21504 | loss:0.8663228750228882 lr:3.583717580918873e-05 tokens_per_second_per_gpu:3645.2534128324373 mfu:17.19528229269571 +Step 21505 | loss:0.7102636694908142 lr:3.583301723417035e-05 tokens_per_second_per_gpu:5345.883028237911 mfu:25.217442345895794 +Step 21506 | loss:0.7329076528549194 lr:3.5828858765703666e-05 tokens_per_second_per_gpu:4236.638250340181 mfu:19.984945471877555 +Step 21507 | loss:0.7697893977165222 lr:3.582470040381993e-05 tokens_per_second_per_gpu:5032.398476569825 mfu:23.738682229697613 +Step 21508 | loss:0.7230620384216309 lr:3.582054214855045e-05 tokens_per_second_per_gpu:4551.857008768734 mfu:21.47188613725104 +Step 21509 | loss:0.5902640223503113 lr:3.581638399992646e-05 tokens_per_second_per_gpu:4402.656008373965 mfu:20.76807956207328 +Step 21510 | loss:0.7434666752815247 lr:3.581222595797929e-05 tokens_per_second_per_gpu:3699.818129641539 mfu:17.452673371584915 +Step 21511 | loss:0.7122303247451782 lr:3.580806802274016e-05 tokens_per_second_per_gpu:4103.642775308037 mfu:19.357583124783037 +Step 21512 | loss:0.8407308459281921 lr:3.580391019424036e-05 tokens_per_second_per_gpu:4816.383945573669 mfu:22.7197048311093 +Step 21513 | loss:0.5973480343818665 lr:3.579975247251115e-05 tokens_per_second_per_gpu:3282.496911037524 mfu:15.484097981087729 +Step 21514 | loss:0.8247088193893433 lr:3.579559485758383e-05 tokens_per_second_per_gpu:4463.9741732812945 mfu:21.057327807898886 +Step 21515 | loss:0.6492685079574585 lr:3.5791437349489655e-05 tokens_per_second_per_gpu:5345.182926696088 mfu:25.214139847473373 +Step 21516 | loss:0.8175056576728821 lr:3.578727994825988e-05 tokens_per_second_per_gpu:4376.685911848041 mfu:20.64557418580512 +Step 21517 | loss:0.7394179701805115 lr:3.5783122653925807e-05 tokens_per_second_per_gpu:4047.2637748421203 mfu:19.091633760335924 +Step 21518 | loss:0.7295652627944946 lr:3.5778965466518664e-05 tokens_per_second_per_gpu:4504.880501118813 mfu:21.250289935647675 +Step 21519 | loss:0.672572910785675 lr:3.577480838606975e-05 tokens_per_second_per_gpu:4004.414406816099 mfu:18.88950598050112 +Step 21520 | loss:0.7761365175247192 lr:3.57706514126103e-05 tokens_per_second_per_gpu:4627.524685244575 mfu:21.828823477424454 +Step 21521 | loss:0.6655447483062744 lr:3.57664945461716e-05 tokens_per_second_per_gpu:4253.469635065166 mfu:20.064341985355668 +Step 21522 | loss:0.7073248028755188 lr:3.576233778678492e-05 tokens_per_second_per_gpu:5260.5654898338 mfu:24.814984960570513 +Step 21523 | loss:0.7496483325958252 lr:3.5758181134481504e-05 tokens_per_second_per_gpu:4672.6905820202865 mfu:22.0418787186121 +Step 21524 | loss:0.6915216445922852 lr:3.575402458929264e-05 tokens_per_second_per_gpu:5360.447858127533 mfu:25.286147133501366 +Step 21525 | loss:0.7287708520889282 lr:3.574986815124957e-05 tokens_per_second_per_gpu:4768.60553581596 mfu:22.494325920444894 +Step 21526 | loss:0.6902046203613281 lr:3.574571182038356e-05 tokens_per_second_per_gpu:4924.401818044388 mfu:23.229243565303143 +Step 21527 | loss:0.640365481376648 lr:3.5741555596725855e-05 tokens_per_second_per_gpu:3897.7835157784084 mfu:18.386509874370308 +Step 21528 | loss:0.605483889579773 lr:3.573739948030774e-05 tokens_per_second_per_gpu:3477.7992762751132 mfu:16.405372559933003 +Step 21529 | loss:0.7554324865341187 lr:3.573324347116046e-05 tokens_per_second_per_gpu:4477.146233916678 mfu:21.119462665301278 +Step 21530 | loss:0.6870490908622742 lr:3.572908756931529e-05 tokens_per_second_per_gpu:5326.123692317249 mfu:25.124234187067685 +Step 21531 | loss:0.535827100276947 lr:3.572493177480347e-05 tokens_per_second_per_gpu:4421.338205949546 mfu:20.856206675548886 +Step 21532 | loss:0.7581472396850586 lr:3.572077608765626e-05 tokens_per_second_per_gpu:2751.018632860553 mfu:12.977024263382122 +Step 21533 | loss:0.5527356863021851 lr:3.571662050790489e-05 tokens_per_second_per_gpu:4148.059283610744 mfu:19.567103372684667 +Step 21534 | loss:0.6127553582191467 lr:3.571246503558067e-05 tokens_per_second_per_gpu:5143.066826333173 mfu:24.260723717498845 +Step 21535 | loss:0.7785991430282593 lr:3.570830967071481e-05 tokens_per_second_per_gpu:4596.5198691857895 mfu:21.682568470108055 +Step 21536 | loss:0.7334085702896118 lr:3.570415441333858e-05 tokens_per_second_per_gpu:5305.244931309326 mfu:25.025745509109353 +Step 21537 | loss:0.5763905644416809 lr:3.5699999263483216e-05 tokens_per_second_per_gpu:3469.5972968122533 mfu:16.366682423404697 +Step 21538 | loss:0.6204113960266113 lr:3.5695844221180006e-05 tokens_per_second_per_gpu:5308.76714454163 mfu:25.042360389877675 +Step 21539 | loss:0.7819872498512268 lr:3.569168928646017e-05 tokens_per_second_per_gpu:5345.5959786676 mfu:25.21608828409693 +Step 21540 | loss:0.675940752029419 lr:3.568753445935495e-05 tokens_per_second_per_gpu:4445.245996656379 mfu:20.968983803402732 +Step 21541 | loss:0.7622449994087219 lr:3.568337973989562e-05 tokens_per_second_per_gpu:5344.336294552621 mfu:25.2101461392023 +Step 21542 | loss:0.684816300868988 lr:3.5679225128113426e-05 tokens_per_second_per_gpu:3020.8166993302375 mfu:14.24970777521646 +Step 21543 | loss:0.5705010890960693 lr:3.567507062403959e-05 tokens_per_second_per_gpu:3308.30679901985 mfu:15.605847626321337 +Step 21544 | loss:0.6063193678855896 lr:3.567091622770538e-05 tokens_per_second_per_gpu:3443.2746385027453 mfu:16.24251395305013 +Step 21545 | loss:0.6055503487586975 lr:3.5666761939142036e-05 tokens_per_second_per_gpu:3791.883995560728 mfu:17.88696376918222 +Step 21546 | loss:0.6246508955955505 lr:3.566260775838083e-05 tokens_per_second_per_gpu:5260.1221287250455 mfu:24.812893550575314 +Step 21547 | loss:0.7912325263023376 lr:3.565845368545295e-05 tokens_per_second_per_gpu:4107.357405790318 mfu:19.37510567054927 +Step 21548 | loss:0.6214196085929871 lr:3.565429972038968e-05 tokens_per_second_per_gpu:5337.154992526797 mfu:25.17627071229002 +Step 21549 | loss:0.7914202809333801 lr:3.565014586322224e-05 tokens_per_second_per_gpu:3913.485880934899 mfu:18.460580609913233 +Step 21550 | loss:0.8936657309532166 lr:3.5645992113981905e-05 tokens_per_second_per_gpu:5187.895105354574 mfu:24.472186358135883 +Step 21551 | loss:0.7600584626197815 lr:3.564183847269989e-05 tokens_per_second_per_gpu:5356.5875333033055 mfu:25.267937322666153 +Step 21552 | loss:0.6815091371536255 lr:3.563768493940742e-05 tokens_per_second_per_gpu:4605.079314859644 mfu:21.722944835743654 +Step 21553 | loss:0.6579720377922058 lr:3.563353151413579e-05 tokens_per_second_per_gpu:5345.755578062303 mfu:25.216841141672205 +Step 21554 | loss:0.6383233070373535 lr:3.562937819691618e-05 tokens_per_second_per_gpu:5337.178347933536 mfu:25.17638088372845 +Step 21555 | loss:0.7325960397720337 lr:3.5625224987779856e-05 tokens_per_second_per_gpu:3652.382797019208 mfu:17.228912814303076 +Step 21556 | loss:0.7286250591278076 lr:3.562107188675804e-05 tokens_per_second_per_gpu:5347.668846827686 mfu:25.22586635687583 +Step 21557 | loss:0.7381094098091125 lr:3.5616918893881984e-05 tokens_per_second_per_gpu:4297.375301092959 mfu:20.271452503087847 +Step 21558 | loss:0.6719456315040588 lr:3.561276600918292e-05 tokens_per_second_per_gpu:5057.2661787677625 mfu:23.855987423832794 +Step 21559 | loss:0.7503620982170105 lr:3.560861323269206e-05 tokens_per_second_per_gpu:4273.114443419044 mfu:20.15700989811084 +Step 21560 | loss:0.6224749088287354 lr:3.560446056444069e-05 tokens_per_second_per_gpu:4636.3744686122745 mfu:21.870569415498633 +Step 21561 | loss:0.7290951013565063 lr:3.560030800445999e-05 tokens_per_second_per_gpu:4796.084394327879 mfu:22.623948384422327 +Step 21562 | loss:0.6934871673583984 lr:3.559615555278121e-05 tokens_per_second_per_gpu:4717.084016087441 mfu:22.251290121407756 +Step 21563 | loss:0.5659493207931519 lr:3.559200320943557e-05 tokens_per_second_per_gpu:5174.058463377847 mfu:24.406916557157103 +Step 21564 | loss:0.8474966883659363 lr:3.558785097445433e-05 tokens_per_second_per_gpu:4558.782340082024 mfu:21.504554106639194 +Step 21565 | loss:0.672110378742218 lr:3.558369884786869e-05 tokens_per_second_per_gpu:5290.863653758204 mfu:24.957906569201743 +Step 21566 | loss:0.7484913468360901 lr:3.5579546829709897e-05 tokens_per_second_per_gpu:5103.999882854661 mfu:24.076438279602822 +Step 21567 | loss:0.7140317559242249 lr:3.557539492000919e-05 tokens_per_second_per_gpu:4545.038184241426 mfu:21.4397205785443 +Step 21568 | loss:0.6485300660133362 lr:3.557124311879776e-05 tokens_per_second_per_gpu:4031.715622826301 mfu:19.018290474489024 +Step 21569 | loss:0.6340526938438416 lr:3.556709142610684e-05 tokens_per_second_per_gpu:4311.389737925354 mfu:20.337560992735746 +Step 21570 | loss:0.6186378598213196 lr:3.556293984196768e-05 tokens_per_second_per_gpu:3419.481721698621 mfu:16.13027870499487 +Step 21571 | loss:0.6052512526512146 lr:3.5558788366411494e-05 tokens_per_second_per_gpu:3898.9771053482823 mfu:18.392140240018787 +Step 21572 | loss:0.6858606934547424 lr:3.5554636999469494e-05 tokens_per_second_per_gpu:4234.995109882607 mfu:19.97719449799051 +Step 21573 | loss:0.6522461771965027 lr:3.555048574117292e-05 tokens_per_second_per_gpu:5349.882996348212 mfu:25.236310877937015 +Step 21574 | loss:0.766924262046814 lr:3.5546334591552986e-05 tokens_per_second_per_gpu:3404.7518873127633 mfu:16.06079556302788 +Step 21575 | loss:0.5466321706771851 lr:3.554218355064093e-05 tokens_per_second_per_gpu:5347.293148927699 mfu:25.22409412578092 +Step 21576 | loss:0.6816627979278564 lr:3.553803261846793e-05 tokens_per_second_per_gpu:3367.7074829300914 mfu:15.886050787127687 +Step 21577 | loss:0.7776691913604736 lr:3.5533881795065236e-05 tokens_per_second_per_gpu:4332.436931469129 mfu:20.436844195703248 +Step 21578 | loss:0.7340222597122192 lr:3.552973108046408e-05 tokens_per_second_per_gpu:3971.9516965197818 mfu:18.73637383732388 +Step 21579 | loss:0.6096580028533936 lr:3.5525580474695644e-05 tokens_per_second_per_gpu:4873.803839991281 mfu:22.990564269921382 +Step 21580 | loss:0.6120589375495911 lr:3.5521429977791166e-05 tokens_per_second_per_gpu:4632.059980228818 mfu:21.85021723766624 +Step 21581 | loss:0.6893491148948669 lr:3.551727958978186e-05 tokens_per_second_per_gpu:3979.501132840025 mfu:18.77198581651313 +Step 21582 | loss:0.8048298358917236 lr:3.5513129310698966e-05 tokens_per_second_per_gpu:4823.37808954592 mfu:22.752697401550424 +Step 21583 | loss:0.6180042028427124 lr:3.550897914057365e-05 tokens_per_second_per_gpu:5217.802100695393 mfu:24.6132627578181 +Step 21584 | loss:0.6584762930870056 lr:3.550482907943716e-05 tokens_per_second_per_gpu:3475.938823276208 mfu:16.396596485711154 +Step 21585 | loss:0.6330544948577881 lr:3.550067912732069e-05 tokens_per_second_per_gpu:3883.476864183236 mfu:18.319022957829414 +Step 21586 | loss:0.6632654070854187 lr:3.5496529284255476e-05 tokens_per_second_per_gpu:5366.512832697337 mfu:25.314756653338495 +Step 21587 | loss:0.7786648273468018 lr:3.549237955027271e-05 tokens_per_second_per_gpu:3651.4738205616886 mfu:17.224625017265584 +Step 21588 | loss:0.7591372728347778 lr:3.54882299254036e-05 tokens_per_second_per_gpu:4100.032281004619 mfu:19.340551807139963 +Step 21589 | loss:0.5332462787628174 lr:3.548408040967939e-05 tokens_per_second_per_gpu:5337.904376511194 mfu:25.179805684402506 +Step 21590 | loss:0.7899367809295654 lr:3.547993100313125e-05 tokens_per_second_per_gpu:4330.740412460134 mfu:20.428841426081824 +Step 21591 | loss:0.7717388868331909 lr:3.54757817057904e-05 tokens_per_second_per_gpu:4962.023508361479 mfu:23.406711497451127 +Step 21592 | loss:0.7363188862800598 lr:3.547163251768804e-05 tokens_per_second_per_gpu:4140.610744608842 mfu:19.53196734336089 +Step 21593 | loss:0.6667678952217102 lr:3.54674834388554e-05 tokens_per_second_per_gpu:5366.11037209367 mfu:25.312858178004138 +Step 21594 | loss:0.6731164455413818 lr:3.546333446932366e-05 tokens_per_second_per_gpu:4547.292556054918 mfu:21.450354835024037 +Step 21595 | loss:0.6155846118927002 lr:3.545918560912403e-05 tokens_per_second_per_gpu:4302.572988170248 mfu:20.295970879848284 +Step 21596 | loss:0.5884865522384644 lr:3.545503685828775e-05 tokens_per_second_per_gpu:3454.194811777617 mfu:16.294026273561382 +Step 21597 | loss:0.4958374798297882 lr:3.545088821684598e-05 tokens_per_second_per_gpu:5362.328832374036 mfu:25.29502000994942 +Step 21598 | loss:0.6257105469703674 lr:3.544673968482993e-05 tokens_per_second_per_gpu:4463.274602153205 mfu:21.054027811528545 +Step 21599 | loss:0.7747538089752197 lr:3.54425912622708e-05 tokens_per_second_per_gpu:4112.129541890311 mfu:19.397616650743313 +Step 21600 | loss:0.7238425016403198 lr:3.5438442949199816e-05 tokens_per_second_per_gpu:4916.875227811104 mfu:23.193739355004027 +Step 21601 | loss:0.7201526165008545 lr:3.543429474564814e-05 tokens_per_second_per_gpu:3913.4899105868076 mfu:18.460599618468816 +Step 21602 | loss:0.7101008892059326 lr:3.5430146651647014e-05 tokens_per_second_per_gpu:4806.806125480994 mfu:22.674524619587533 +Step 21603 | loss:0.6382227540016174 lr:3.54259986672276e-05 tokens_per_second_per_gpu:4780.484986038958 mfu:22.550363313989983 +Step 21604 | loss:0.7473739981651306 lr:3.542185079242113e-05 tokens_per_second_per_gpu:5369.327000613623 mfu:25.328031563546173 +Step 21605 | loss:0.6334576606750488 lr:3.541770302725875e-05 tokens_per_second_per_gpu:4008.6867806276287 mfu:18.909659496712248 +Step 21606 | loss:0.6785677671432495 lr:3.541355537177171e-05 tokens_per_second_per_gpu:4306.17343063016 mfu:20.312954781230516 +Step 21607 | loss:0.7377784848213196 lr:3.540940782599117e-05 tokens_per_second_per_gpu:4419.71050096994 mfu:20.848528513444975 +Step 21608 | loss:0.6601012349128723 lr:3.5405260389948336e-05 tokens_per_second_per_gpu:5275.894917467112 mfu:24.887296486186894 +Step 21609 | loss:0.6525301933288574 lr:3.54011130636744e-05 tokens_per_second_per_gpu:5308.779454480807 mfu:25.042418457961062 +Step 21610 | loss:0.7033077478408813 lr:3.539696584720056e-05 tokens_per_second_per_gpu:5380.912663069289 mfu:25.382683110067397 +Step 21611 | loss:0.756725013256073 lr:3.539281874055802e-05 tokens_per_second_per_gpu:5353.104009893225 mfu:25.251504948390423 +Step 21612 | loss:0.7581113576889038 lr:3.5388671743777926e-05 tokens_per_second_per_gpu:5248.759530183419 mfu:24.759294234595238 +Step 21613 | loss:0.7337906360626221 lr:3.538452485689151e-05 tokens_per_second_per_gpu:4069.757316028131 mfu:19.197739631904202 +Step 21614 | loss:0.818933367729187 lr:3.538037807992994e-05 tokens_per_second_per_gpu:5317.911444137488 mfu:25.085495611249726 +Step 21615 | loss:0.723447859287262 lr:3.537623141292441e-05 tokens_per_second_per_gpu:5334.397045211935 mfu:25.163261004251478 +Step 21616 | loss:0.6972050666809082 lr:3.5372084855906115e-05 tokens_per_second_per_gpu:4010.0580195782004 mfu:18.916127864799584 +Step 21617 | loss:0.8284659385681152 lr:3.5367938408906234e-05 tokens_per_second_per_gpu:4700.45819802541 mfu:22.172863300952088 +Step 21618 | loss:0.6945817470550537 lr:3.536379207195598e-05 tokens_per_second_per_gpu:4880.112103898948 mfu:23.020321427074652 +Step 21619 | loss:0.8223949074745178 lr:3.5359645845086475e-05 tokens_per_second_per_gpu:5327.807529505159 mfu:25.132177134375304 +Step 21620 | loss:0.7631234526634216 lr:3.5355499728328965e-05 tokens_per_second_per_gpu:5322.647262283051 mfu:25.107835273456242 +Step 21621 | loss:0.6149516105651855 lr:3.535135372171459e-05 tokens_per_second_per_gpu:5096.154144186228 mfu:24.03942859168277 +Step 21622 | loss:0.7908991575241089 lr:3.534720782527457e-05 tokens_per_second_per_gpu:5333.403830740395 mfu:25.158575842129363 +Step 21623 | loss:0.7237558364868164 lr:3.534306203904007e-05 tokens_per_second_per_gpu:3397.502343466318 mfu:16.026598227804474 +Step 21624 | loss:0.5524822473526001 lr:3.533891636304225e-05 tokens_per_second_per_gpu:5327.078795801573 mfu:25.128739573160697 +Step 21625 | loss:0.6566651463508606 lr:3.533477079731235e-05 tokens_per_second_per_gpu:5330.02766492591 mfu:25.14264989194915 +Step 21626 | loss:0.6188960075378418 lr:3.533062534188148e-05 tokens_per_second_per_gpu:5316.292786671247 mfu:25.07786012781003 +Step 21627 | loss:0.6869967579841614 lr:3.532647999678086e-05 tokens_per_second_per_gpu:4376.2324694434365 mfu:20.643435220617352 +Step 21628 | loss:0.8831771016120911 lr:3.532233476204164e-05 tokens_per_second_per_gpu:4556.431516165868 mfu:21.493464869134765 +Step 21629 | loss:0.7220213413238525 lr:3.531818963769503e-05 tokens_per_second_per_gpu:5331.626408727362 mfu:25.15019143923442 +Step 21630 | loss:0.552168607711792 lr:3.5314044623772194e-05 tokens_per_second_per_gpu:3948.633554251999 mfu:18.62637818176143 +Step 21631 | loss:0.7223926782608032 lr:3.530989972030428e-05 tokens_per_second_per_gpu:5178.842970562125 mfu:24.429485893866442 +Step 21632 | loss:0.6525570154190063 lr:3.5305754927322507e-05 tokens_per_second_per_gpu:4523.618390277418 mfu:21.338679755822373 +Step 21633 | loss:0.7399395108222961 lr:3.5301610244858016e-05 tokens_per_second_per_gpu:4820.174114006801 mfu:22.7375837022778 +Step 21634 | loss:0.6821370720863342 lr:3.5297465672942e-05 tokens_per_second_per_gpu:4572.077923895413 mfu:21.56727164394746 +Step 21635 | loss:0.6332048773765564 lr:3.52933212116056e-05 tokens_per_second_per_gpu:3920.4817423967597 mfu:18.493581282045877 +Step 21636 | loss:0.5979828238487244 lr:3.528917686088002e-05 tokens_per_second_per_gpu:5360.689139363628 mfu:25.287285298258848 +Step 21637 | loss:0.7400966882705688 lr:3.528503262079641e-05 tokens_per_second_per_gpu:3551.249851614527 mfu:16.751851455769426 +Step 21638 | loss:0.8177840113639832 lr:3.5280888491385954e-05 tokens_per_second_per_gpu:3534.961997086716 mfu:16.6750188669672 +Step 21639 | loss:0.7772952914237976 lr:3.527674447267981e-05 tokens_per_second_per_gpu:5038.271781420005 mfu:23.766387610765054 +Step 21640 | loss:0.7542288899421692 lr:3.5272600564709166e-05 tokens_per_second_per_gpu:5383.155181962866 mfu:25.393261454301577 +Step 21641 | loss:0.7259458303451538 lr:3.5268456767505155e-05 tokens_per_second_per_gpu:4091.730092529545 mfu:19.30138896760362 +Step 21642 | loss:1.0889877080917358 lr:3.526431308109897e-05 tokens_per_second_per_gpu:4419.602149941718 mfu:20.848017403158266 +Step 21643 | loss:0.5741110444068909 lr:3.5260169505521765e-05 tokens_per_second_per_gpu:4006.5832137478706 mfu:18.899736612834907 +Step 21644 | loss:0.6208059787750244 lr:3.525602604080469e-05 tokens_per_second_per_gpu:4633.588331040315 mfu:21.85742672920761 +Step 21645 | loss:0.7009478211402893 lr:3.525188268697894e-05 tokens_per_second_per_gpu:4467.12268046152 mfu:21.072179853458724 +Step 21646 | loss:0.7462800145149231 lr:3.524773944407567e-05 tokens_per_second_per_gpu:5355.511865913073 mfu:25.26286321605864 +Step 21647 | loss:0.7321540117263794 lr:3.524359631212603e-05 tokens_per_second_per_gpu:4868.399169919242 mfu:22.965069519059774 +Step 21648 | loss:0.6965763568878174 lr:3.523945329116117e-05 tokens_per_second_per_gpu:5344.66077577937 mfu:25.211676772511076 +Step 21649 | loss:0.7237191796302795 lr:3.523531038121227e-05 tokens_per_second_per_gpu:5322.526257465873 mfu:25.107264473088065 +Step 21650 | loss:0.7444667816162109 lr:3.52311675823105e-05 tokens_per_second_per_gpu:5217.584412326817 mfu:24.61223588464206 +Step 21651 | loss:0.6704875826835632 lr:3.5227024894486985e-05 tokens_per_second_per_gpu:4981.641014772473 mfu:23.49925062228352 +Step 21652 | loss:0.7172141671180725 lr:3.52228823177729e-05 tokens_per_second_per_gpu:4628.889711883793 mfu:21.835262540979205 +Step 21653 | loss:0.7273587584495544 lr:3.52187398521994e-05 tokens_per_second_per_gpu:4913.0936962164615 mfu:23.17590122527628 +Step 21654 | loss:0.6632649898529053 lr:3.5214597497797684e-05 tokens_per_second_per_gpu:5319.237937801421 mfu:25.091752908185658 +Step 21655 | loss:0.7419944405555725 lr:3.521045525459883e-05 tokens_per_second_per_gpu:5319.999239112676 mfu:25.095344096362243 +Step 21656 | loss:0.6789093017578125 lr:3.520631312263403e-05 tokens_per_second_per_gpu:5331.331509460736 mfu:25.14880034907874 +Step 21657 | loss:0.6013456583023071 lr:3.520217110193443e-05 tokens_per_second_per_gpu:4093.3518569272455 mfu:19.309039106969113 +Step 21658 | loss:0.818253219127655 lr:3.519802919253121e-05 tokens_per_second_per_gpu:5227.439465241165 mfu:24.658723850684453 +Step 21659 | loss:0.5835331082344055 lr:3.519388739445549e-05 tokens_per_second_per_gpu:4154.672980778559 mfu:19.59830131063834 +Step 21660 | loss:0.6817102432250977 lr:3.518974570773843e-05 tokens_per_second_per_gpu:3722.2688572587385 mfu:17.55857728424436 +Step 21661 | loss:0.664076566696167 lr:3.518560413241121e-05 tokens_per_second_per_gpu:3882.6982123110884 mfu:18.31534992409679 +Step 21662 | loss:0.6501020789146423 lr:3.518146266850493e-05 tokens_per_second_per_gpu:3819.102385934582 mfu:18.01535756051195 +Step 21663 | loss:0.7902630567550659 lr:3.517732131605075e-05 tokens_per_second_per_gpu:5315.621364768216 mfu:25.074692916137042 +Step 21664 | loss:0.7561562061309814 lr:3.517318007507984e-05 tokens_per_second_per_gpu:5275.987036283111 mfu:24.887731026359052 +Step 21665 | loss:0.6512613892555237 lr:3.5169038945623326e-05 tokens_per_second_per_gpu:4975.431138958155 mfu:23.469957578553313 +Step 21666 | loss:0.6755046844482422 lr:3.516489792771237e-05 tokens_per_second_per_gpu:5294.9300718325585 mfu:24.977088557060938 +Step 21667 | loss:0.7035442590713501 lr:3.51607570213781e-05 tokens_per_second_per_gpu:5054.951341430179 mfu:23.845067941159435 +Step 21668 | loss:0.630825400352478 lr:3.5156616226651683e-05 tokens_per_second_per_gpu:4266.138598448941 mfu:20.124103647184985 +Step 21669 | loss:0.8070024251937866 lr:3.5152475543564255e-05 tokens_per_second_per_gpu:4461.521175941457 mfu:21.04575660092237 +Step 21670 | loss:0.8200693130493164 lr:3.514833497214695e-05 tokens_per_second_per_gpu:5329.182183039712 mfu:25.138661609637868 +Step 21671 | loss:0.6209205389022827 lr:3.514419451243089e-05 tokens_per_second_per_gpu:5386.233103161297 mfu:25.407780533742084 +Step 21672 | loss:0.7416458129882812 lr:3.514005416444726e-05 tokens_per_second_per_gpu:4431.956466810444 mfu:20.906294823691855 +Step 21673 | loss:0.6613799929618835 lr:3.513591392822716e-05 tokens_per_second_per_gpu:5346.333544280409 mfu:25.219567506934215 +Step 21674 | loss:0.6601468324661255 lr:3.513177380380176e-05 tokens_per_second_per_gpu:5346.638087673567 mfu:25.221004090080203 +Step 21675 | loss:0.6802992820739746 lr:3.512763379120218e-05 tokens_per_second_per_gpu:4406.2513035312795 mfu:20.785039182750936 +Step 21676 | loss:0.6739267110824585 lr:3.512349389045959e-05 tokens_per_second_per_gpu:3711.7983162076107 mfu:17.509185955647673 +Step 21677 | loss:0.5932736396789551 lr:3.511935410160509e-05 tokens_per_second_per_gpu:5321.654493275102 mfu:25.103152212661893 +Step 21678 | loss:0.7756887078285217 lr:3.5115214424669804e-05 tokens_per_second_per_gpu:3414.271581628656 mfu:16.105701585343283 +Step 21679 | loss:0.7796783447265625 lr:3.5111074859684915e-05 tokens_per_second_per_gpu:5035.528573045809 mfu:23.75344742882436 +Step 21680 | loss:0.5736483931541443 lr:3.510693540668151e-05 tokens_per_second_per_gpu:4716.061724968636 mfu:22.246467799779502 +Step 21681 | loss:0.9020256996154785 lr:3.5102796065690754e-05 tokens_per_second_per_gpu:5063.346180976457 mfu:23.88466783160655 +Step 21682 | loss:0.7253972887992859 lr:3.509865683674377e-05 tokens_per_second_per_gpu:3892.230598762436 mfu:18.36031581738063 +Step 21683 | loss:0.6402028203010559 lr:3.50945177198717e-05 tokens_per_second_per_gpu:4990.625943970148 mfu:23.541634066295952 +Step 21684 | loss:0.6903612017631531 lr:3.5090378715105646e-05 tokens_per_second_per_gpu:4703.078695000696 mfu:22.1852246322874 +Step 21685 | loss:0.6099370121955872 lr:3.508623982247677e-05 tokens_per_second_per_gpu:4802.974956615609 mfu:22.656452342384704 +Step 21686 | loss:0.7143411636352539 lr:3.508210104201618e-05 tokens_per_second_per_gpu:3659.330351387271 mfu:17.26168561363224 +Step 21687 | loss:0.6410357356071472 lr:3.5077962373755004e-05 tokens_per_second_per_gpu:4362.773921351796 mfu:20.579948953003992 +Step 21688 | loss:0.7719777822494507 lr:3.507382381772439e-05 tokens_per_second_per_gpu:3842.448375104006 mfu:18.125484574660618 +Step 21689 | loss:0.767561137676239 lr:3.5069685373955444e-05 tokens_per_second_per_gpu:5117.61491548465 mfu:24.14066271148258 +Step 21690 | loss:0.840376079082489 lr:3.506554704247932e-05 tokens_per_second_per_gpu:4025.4273240671455 mfu:18.988627496347686 +Step 21691 | loss:0.7927296161651611 lr:3.5061408823327095e-05 tokens_per_second_per_gpu:4786.2432033900695 mfu:22.577525807688364 +Step 21692 | loss:0.7827911972999573 lr:3.5057270716529936e-05 tokens_per_second_per_gpu:4384.583396858866 mfu:20.68282797005107 +Step 21693 | loss:0.6822746992111206 lr:3.505313272211894e-05 tokens_per_second_per_gpu:3024.9185886627365 mfu:14.269057087052346 +Step 21694 | loss:0.5772310495376587 lr:3.5048994840125244e-05 tokens_per_second_per_gpu:5054.204425963025 mfu:23.84154461346395 +Step 21695 | loss:0.6574581265449524 lr:3.504485707057997e-05 tokens_per_second_per_gpu:4544.040816203418 mfu:21.435015823340482 +Step 21696 | loss:0.6966947317123413 lr:3.5040719413514224e-05 tokens_per_second_per_gpu:4272.563536458018 mfu:20.15441117598126 +Step 21697 | loss:0.5971555709838867 lr:3.503658186895914e-05 tokens_per_second_per_gpu:3656.589741489016 mfu:17.248757689145904 +Step 21698 | loss:0.8098604083061218 lr:3.5032444436945856e-05 tokens_per_second_per_gpu:5327.184650443244 mfu:25.129238907566123 +Step 21699 | loss:0.6319324970245361 lr:3.502830711750544e-05 tokens_per_second_per_gpu:5349.595300710234 mfu:25.234953768526733 +Step 21700 | loss:0.8647031784057617 lr:3.502416991066904e-05 tokens_per_second_per_gpu:4847.6690683099 mfu:22.867282092848455 +Step 21701 | loss:0.74505615234375 lr:3.502003281646777e-05 tokens_per_second_per_gpu:4053.9627023676285 mfu:19.12323374442864 +Step 21702 | loss:0.7204759120941162 lr:3.5015895834932754e-05 tokens_per_second_per_gpu:5335.652429597563 mfu:25.16918286658915 +Step 21703 | loss:0.6484425067901611 lr:3.5011758966095084e-05 tokens_per_second_per_gpu:3705.2883717812488 mfu:17.478477437077473 +Step 21704 | loss:0.6489671468734741 lr:3.5007622209985905e-05 tokens_per_second_per_gpu:4692.718585182342 mfu:22.13635422665722 +Step 21705 | loss:0.7324246168136597 lr:3.500348556663631e-05 tokens_per_second_per_gpu:3656.922501945779 mfu:17.250327376995244 +Step 21706 | loss:0.7008784413337708 lr:3.4999349036077406e-05 tokens_per_second_per_gpu:4509.6820753527145 mfu:21.272939780542032 +Step 21707 | loss:0.6602124571800232 lr:3.49952126183403e-05 tokens_per_second_per_gpu:3209.718708854142 mfu:15.140790784146997 +Step 21708 | loss:0.6989112496376038 lr:3.499107631345613e-05 tokens_per_second_per_gpu:4158.204201035256 mfu:19.614958679077485 +Step 21709 | loss:0.7859052419662476 lr:3.498694012145598e-05 tokens_per_second_per_gpu:4266.996291771069 mfu:20.12814953292303 +Step 21710 | loss:0.6910375356674194 lr:3.498280404237098e-05 tokens_per_second_per_gpu:5055.620297731043 mfu:23.848223522166325 +Step 21711 | loss:0.6916055679321289 lr:3.4978668076232223e-05 tokens_per_second_per_gpu:3706.7906234203133 mfu:17.4855638143695 +Step 21712 | loss:0.6973108053207397 lr:3.497453222307083e-05 tokens_per_second_per_gpu:3300.898516131927 mfu:15.570901491955606 +Step 21713 | loss:0.7956601977348328 lr:3.497039648291789e-05 tokens_per_second_per_gpu:4989.622511721767 mfu:23.536900705177146 +Step 21714 | loss:0.6474008560180664 lr:3.4966260855804504e-05 tokens_per_second_per_gpu:5357.078072673209 mfu:25.270251280568633 +Step 21715 | loss:0.5776861906051636 lr:3.49621253417618e-05 tokens_per_second_per_gpu:4509.749677937434 mfu:21.273258673468344 +Step 21716 | loss:0.9411330819129944 lr:3.4957989940820856e-05 tokens_per_second_per_gpu:4194.294549964149 mfu:19.78520301262356 +Step 21717 | loss:0.7206453680992126 lr:3.4953854653012805e-05 tokens_per_second_per_gpu:3833.7154954413527 mfu:18.084290091308965 +Step 21718 | loss:0.9089345335960388 lr:3.494971947836873e-05 tokens_per_second_per_gpu:4441.4403738788005 mfu:20.951032031454695 +Step 21719 | loss:0.6546677350997925 lr:3.494558441691975e-05 tokens_per_second_per_gpu:5303.3367168485665 mfu:25.016744135922988 +Step 21720 | loss:0.6417112946510315 lr:3.4941449468696924e-05 tokens_per_second_per_gpu:5308.798345638526 mfu:25.04250757077519 +Step 21721 | loss:0.7889285683631897 lr:3.493731463373139e-05 tokens_per_second_per_gpu:5337.357616877309 mfu:25.177226525922663 +Step 21722 | loss:0.6694213151931763 lr:3.4933179912054234e-05 tokens_per_second_per_gpu:4887.283272920737 mfu:23.054149054878035 +Step 21723 | loss:0.7197068333625793 lr:3.492904530369655e-05 tokens_per_second_per_gpu:5315.16105404904 mfu:25.07252155193758 +Step 21724 | loss:0.7809550762176514 lr:3.4924910808689435e-05 tokens_per_second_per_gpu:4235.348053160194 mfu:19.978859391649223 +Step 21725 | loss:0.5802344083786011 lr:3.4920776427064e-05 tokens_per_second_per_gpu:3366.586545162295 mfu:15.880763132425287 +Step 21726 | loss:0.7552692294120789 lr:3.491664215885132e-05 tokens_per_second_per_gpu:4140.220317371531 mfu:19.53012563146874 +Step 21727 | loss:0.6548769474029541 lr:3.491250800408252e-05 tokens_per_second_per_gpu:5348.9258854905465 mfu:25.231796022721458 +Step 21728 | loss:0.6539236307144165 lr:3.4908373962788665e-05 tokens_per_second_per_gpu:5087.121774403166 mfu:23.996821362334867 +Step 21729 | loss:0.6824610829353333 lr:3.4904240035000844e-05 tokens_per_second_per_gpu:4285.70901470387 mfu:20.21642063971681 +Step 21730 | loss:0.8397950530052185 lr:3.490010622075016e-05 tokens_per_second_per_gpu:4345.172252626199 mfu:20.496918878470993 +Step 21731 | loss:0.6644889712333679 lr:3.489597252006771e-05 tokens_per_second_per_gpu:3565.8645373951363 mfu:16.82079142211908 +Step 21732 | loss:0.7260220050811768 lr:3.4891838932984566e-05 tokens_per_second_per_gpu:4224.85571384467 mfu:19.929365236920216 +Step 21733 | loss:0.59916090965271 lr:3.488770545953184e-05 tokens_per_second_per_gpu:4405.143633198225 mfu:20.779814112801873 +Step 21734 | loss:0.6850164532661438 lr:3.4883572099740626e-05 tokens_per_second_per_gpu:4508.993913724004 mfu:21.269693604726925 +Step 21735 | loss:0.7626196146011353 lr:3.487943885364198e-05 tokens_per_second_per_gpu:4956.723602375664 mfu:23.38171093665883 +Step 21736 | loss:0.8151755928993225 lr:3.4875305721266994e-05 tokens_per_second_per_gpu:4252.034754430566 mfu:20.057573408586382 +Step 21737 | loss:0.8269039988517761 lr:3.487117270264677e-05 tokens_per_second_per_gpu:4295.7702019710905 mfu:20.263880976671846 +Step 21738 | loss:0.7560060620307922 lr:3.4867039797812395e-05 tokens_per_second_per_gpu:3606.644175777111 mfu:17.013155934090616 +Step 21739 | loss:0.8166841864585876 lr:3.4862907006794926e-05 tokens_per_second_per_gpu:5327.485425373832 mfu:25.1306577142312 +Step 21740 | loss:0.6986473798751831 lr:3.4858774329625475e-05 tokens_per_second_per_gpu:4236.373484785249 mfu:19.983696527581287 +Step 21741 | loss:0.725922167301178 lr:3.4854641766335124e-05 tokens_per_second_per_gpu:5325.061282103492 mfu:25.119222616823517 +Step 21742 | loss:0.5432641506195068 lr:3.485050931695495e-05 tokens_per_second_per_gpu:4621.184827898479 mfu:21.798917288631714 +Step 21743 | loss:0.735541820526123 lr:3.4846376981516e-05 tokens_per_second_per_gpu:4261.347804633001 mfu:20.10150465536148 +Step 21744 | loss:0.6066650748252869 lr:3.48422447600494e-05 tokens_per_second_per_gpu:5307.7895117544085 mfu:25.037748729218997 +Step 21745 | loss:0.7269294857978821 lr:3.4838112652586215e-05 tokens_per_second_per_gpu:5111.864723751386 mfu:24.1135380759929 +Step 21746 | loss:0.9257321953773499 lr:3.4833980659157507e-05 tokens_per_second_per_gpu:4449.1905573156855 mfu:20.98759096904413 +Step 21747 | loss:0.761397659778595 lr:3.482984877979437e-05 tokens_per_second_per_gpu:4678.198453254394 mfu:22.067860286963118 +Step 21748 | loss:0.8514411449432373 lr:3.482571701452789e-05 tokens_per_second_per_gpu:5128.212593177115 mfu:24.190653765308205 +Step 21749 | loss:0.8040376901626587 lr:3.482158536338912e-05 tokens_per_second_per_gpu:5307.9777791729275 mfu:25.038636818753886 +Step 21750 | loss:0.7680025100708008 lr:3.481745382640913e-05 tokens_per_second_per_gpu:4648.133226047479 mfu:21.926037480571388 +Step 21751 | loss:0.7652225494384766 lr:3.481332240361902e-05 tokens_per_second_per_gpu:5369.42891973768 mfu:25.32851233344355 +Step 21752 | loss:0.741619348526001 lr:3.480919109504984e-05 tokens_per_second_per_gpu:4278.028654141084 mfu:20.180191068537514 +Step 21753 | loss:0.6459816694259644 lr:3.480505990073268e-05 tokens_per_second_per_gpu:5340.709819646361 mfu:25.193039438329052 +Step 21754 | loss:0.5877166390419006 lr:3.48009288206986e-05 tokens_per_second_per_gpu:4491.728260111541 mfu:21.188248570812668 +Step 21755 | loss:0.5675625801086426 lr:3.479679785497869e-05 tokens_per_second_per_gpu:4143.316574588402 mfu:19.544731205036676 +Step 21756 | loss:0.7708545923233032 lr:3.479266700360398e-05 tokens_per_second_per_gpu:4289.943272373313 mfu:20.236394355582426 +Step 21757 | loss:0.7458986639976501 lr:3.478853626660558e-05 tokens_per_second_per_gpu:4715.166173977122 mfu:22.242243332955773 +Step 21758 | loss:0.8715012073516846 lr:3.478440564401453e-05 tokens_per_second_per_gpu:4620.718798993624 mfu:21.796718950774782 +Step 21759 | loss:0.7557584643363953 lr:3.47802751358619e-05 tokens_per_second_per_gpu:5340.479152927267 mfu:25.19195134405928 +Step 21760 | loss:0.7945107817649841 lr:3.4776144742178774e-05 tokens_per_second_per_gpu:5324.372657328814 mfu:25.115974256269862 +Step 21761 | loss:0.6290903091430664 lr:3.477201446299621e-05 tokens_per_second_per_gpu:4388.610145045787 mfu:20.701822828282864 +Step 21762 | loss:0.7783837914466858 lr:3.4767884298345244e-05 tokens_per_second_per_gpu:4620.639487716824 mfu:21.796344825950218 +Step 21763 | loss:0.8655296564102173 lr:3.4763754248256997e-05 tokens_per_second_per_gpu:4774.169779412301 mfu:22.520573406846687 +Step 21764 | loss:0.6231256127357483 lr:3.475962431276248e-05 tokens_per_second_per_gpu:4311.179758908507 mfu:20.33657048588716 +Step 21765 | loss:0.7078563570976257 lr:3.475549449189278e-05 tokens_per_second_per_gpu:5346.417146580053 mfu:25.21996187324566 +Step 21766 | loss:0.5760961771011353 lr:3.475136478567894e-05 tokens_per_second_per_gpu:4014.3811274000295 mfu:18.936520701993402 +Step 21767 | loss:0.7480689883232117 lr:3.474723519415204e-05 tokens_per_second_per_gpu:5350.590749023817 mfu:25.2396494680629 +Step 21768 | loss:0.6125760078430176 lr:3.474310571734312e-05 tokens_per_second_per_gpu:3512.7633115075246 mfu:16.570303879603838 +Step 21769 | loss:0.6402503252029419 lr:3.4738976355283255e-05 tokens_per_second_per_gpu:5366.132009228274 mfu:25.31296024405974 +Step 21770 | loss:0.6589062809944153 lr:3.473484710800351e-05 tokens_per_second_per_gpu:5326.40463235249 mfu:25.125559429147227 +Step 21771 | loss:0.8339785933494568 lr:3.473071797553492e-05 tokens_per_second_per_gpu:2990.6932208293097 mfu:14.107610187532174 +Step 21772 | loss:0.5914613008499146 lr:3.472658895790854e-05 tokens_per_second_per_gpu:4204.413680039847 mfu:19.832936675692245 +Step 21773 | loss:0.6342965364456177 lr:3.4722460055155434e-05 tokens_per_second_per_gpu:4391.266271035504 mfu:20.714352227758184 +Step 21774 | loss:0.7047330737113953 lr:3.471833126730666e-05 tokens_per_second_per_gpu:3911.8492954590974 mfu:18.452860556993603 +Step 21775 | loss:0.7437502145767212 lr:3.4714202594393254e-05 tokens_per_second_per_gpu:4948.590899391204 mfu:23.343347589098926 +Step 21776 | loss:0.6990454792976379 lr:3.47100740364463e-05 tokens_per_second_per_gpu:3292.920551610316 mfu:15.53326807212606 +Step 21777 | loss:0.828932523727417 lr:3.470594559349683e-05 tokens_per_second_per_gpu:4063.5052108365835 mfu:19.16824738499653 +Step 21778 | loss:0.8249172568321228 lr:3.4701817265575885e-05 tokens_per_second_per_gpu:5368.335836516007 mfu:25.323356073387686 +Step 21779 | loss:0.8373433351516724 lr:3.469768905271452e-05 tokens_per_second_per_gpu:3621.851808154201 mfu:17.084892930702992 +Step 21780 | loss:0.7014322280883789 lr:3.46935609549438e-05 tokens_per_second_per_gpu:4839.122643549145 mfu:22.826967149081554 +Step 21781 | loss:0.8296407461166382 lr:3.4689432972294756e-05 tokens_per_second_per_gpu:4833.557869397372 mfu:22.800717159959262 +Step 21782 | loss:0.6763718724250793 lr:3.468530510479843e-05 tokens_per_second_per_gpu:3763.0228386693198 mfu:17.750820767904308 +Step 21783 | loss:0.7769612073898315 lr:3.4681177352485896e-05 tokens_per_second_per_gpu:4003.4120159407353 mfu:18.884777531716512 +Step 21784 | loss:0.4232591986656189 lr:3.4677049715388185e-05 tokens_per_second_per_gpu:128.10782431695608 mfu:0.6043064647517791 +Step 21785 | loss:0.4982176423072815 lr:3.467292219353633e-05 tokens_per_second_per_gpu:4903.310755327886 mfu:23.129753423963038 +Step 21786 | loss:0.38433966040611267 lr:3.466879478696137e-05 tokens_per_second_per_gpu:5087.787153710373 mfu:23.99996007005238 +Step 21787 | loss:0.4492473304271698 lr:3.466466749569438e-05 tokens_per_second_per_gpu:5360.6570100745885 mfu:25.287133738920627 +Step 21788 | loss:0.41116583347320557 lr:3.466054031976636e-05 tokens_per_second_per_gpu:4468.802711944315 mfu:21.080104848606002 +Step 21789 | loss:0.4141055643558502 lr:3.4656413259208383e-05 tokens_per_second_per_gpu:5352.365410546247 mfu:25.248020849252683 +Step 21790 | loss:0.46023887395858765 lr:3.4652286314051494e-05 tokens_per_second_per_gpu:3815.718469311626 mfu:17.999395048445756 +Step 21791 | loss:0.46067261695861816 lr:3.4648159484326696e-05 tokens_per_second_per_gpu:4651.247438657487 mfu:21.940727752792256 +Step 21792 | loss:0.4393642842769623 lr:3.464403277006508e-05 tokens_per_second_per_gpu:5336.789483727515 mfu:25.174546544921572 +Step 21793 | loss:0.5394651889801025 lr:3.463990617129763e-05 tokens_per_second_per_gpu:5346.639532260375 mfu:25.22101090444265 +Step 21794 | loss:0.43694889545440674 lr:3.4635779688055414e-05 tokens_per_second_per_gpu:3943.090864215591 mfu:18.60023236717926 +Step 21795 | loss:0.39244958758354187 lr:3.463165332036945e-05 tokens_per_second_per_gpu:5344.676815278697 mfu:25.211752433566026 +Step 21796 | loss:0.5165765285491943 lr:3.462752706827079e-05 tokens_per_second_per_gpu:5102.594403613637 mfu:24.069808394223426 +Step 21797 | loss:0.5051246881484985 lr:3.4623400931790464e-05 tokens_per_second_per_gpu:4589.909090891931 mfu:21.65138430098084 +Step 21798 | loss:0.3408309519290924 lr:3.4619274910959484e-05 tokens_per_second_per_gpu:4630.09711497414 mfu:21.840958067360578 +Step 21799 | loss:0.42087775468826294 lr:3.4615149005808926e-05 tokens_per_second_per_gpu:5323.43173128712 mfu:25.11153574759265 +Step 21800 | loss:0.2589992880821228 lr:3.461102321636978e-05 tokens_per_second_per_gpu:4649.569445491961 mfu:21.932812372735317 +Step 21801 | loss:0.5116437077522278 lr:3.46068975426731e-05 tokens_per_second_per_gpu:4941.7928992817515 mfu:23.311280262724214 +Step 21802 | loss:0.45499470829963684 lr:3.460277198474989e-05 tokens_per_second_per_gpu:5130.657214096856 mfu:24.202185459281957 +Step 21803 | loss:0.28758496046066284 lr:3.459864654263121e-05 tokens_per_second_per_gpu:4769.731561156884 mfu:22.499637574097765 +Step 21804 | loss:0.44992393255233765 lr:3.4594521216348064e-05 tokens_per_second_per_gpu:4790.173657628575 mfu:22.596066430936038 +Step 21805 | loss:0.4076106548309326 lr:3.45903960059315e-05 tokens_per_second_per_gpu:3320.4511742223317 mfu:15.663134716195515 +Step 21806 | loss:0.4664522707462311 lr:3.458627091141254e-05 tokens_per_second_per_gpu:4940.316915390648 mfu:23.304317794881065 +Step 21807 | loss:0.3562757074832916 lr:3.4582145932822205e-05 tokens_per_second_per_gpu:3929.2785182601115 mfu:18.535077174678033 +Step 21808 | loss:0.321981817483902 lr:3.45780210701915e-05 tokens_per_second_per_gpu:3782.6285364731393 mfu:17.843304189521906 +Step 21809 | loss:0.3841454088687897 lr:3.457389632355147e-05 tokens_per_second_per_gpu:3749.964449428064 mfu:17.689222117862357 +Step 21810 | loss:0.45543524622917175 lr:3.456977169293314e-05 tokens_per_second_per_gpu:3916.229627607093 mfu:18.47352333109743 +Step 21811 | loss:0.44380536675453186 lr:3.456564717836752e-05 tokens_per_second_per_gpu:3661.9158541758056 mfu:17.27388187141802 +Step 21812 | loss:0.48264142870903015 lr:3.456152277988565e-05 tokens_per_second_per_gpu:4006.6617276034963 mfu:18.90010697608774 +Step 21813 | loss:0.44433948397636414 lr:3.4557398497518534e-05 tokens_per_second_per_gpu:5355.322885098478 mfu:25.261971761312743 +Step 21814 | loss:0.32644644379615784 lr:3.4553274331297194e-05 tokens_per_second_per_gpu:4386.153604254298 mfu:20.69023490624033 +Step 21815 | loss:0.37265244126319885 lr:3.4549150281252636e-05 tokens_per_second_per_gpu:5158.0563124722785 mfu:24.331431681087057 +Step 21816 | loss:0.445324182510376 lr:3.4545026347415896e-05 tokens_per_second_per_gpu:3388.314391445125 mfu:15.983257090494241 +Step 21817 | loss:0.3942756652832031 lr:3.454090252981799e-05 tokens_per_second_per_gpu:5359.213295017431 mfu:25.280323488672824 +Step 21818 | loss:0.45255792140960693 lr:3.4536778828489916e-05 tokens_per_second_per_gpu:4379.638398848996 mfu:20.65950157073485 +Step 21819 | loss:0.4575434625148773 lr:3.453265524346272e-05 tokens_per_second_per_gpu:4329.487591889123 mfu:20.422931657695134 +Step 21820 | loss:0.36605653166770935 lr:3.452853177476738e-05 tokens_per_second_per_gpu:3021.744078719319 mfu:14.254082382022 +Step 21821 | loss:0.4025823771953583 lr:3.4524408422434954e-05 tokens_per_second_per_gpu:5365.154582290589 mfu:25.30834955442845 +Step 21822 | loss:0.38361096382141113 lr:3.45202851864964e-05 tokens_per_second_per_gpu:4824.284053413195 mfu:22.75697098768573 +Step 21823 | loss:0.509637713432312 lr:3.451616206698277e-05 tokens_per_second_per_gpu:3766.5486381656638 mfu:17.76745256569199 +Step 21824 | loss:0.4546826183795929 lr:3.451203906392505e-05 tokens_per_second_per_gpu:4033.2277858643365 mfu:19.02542360554105 +Step 21825 | loss:0.4455628991127014 lr:3.4507916177354265e-05 tokens_per_second_per_gpu:5309.4307323343355 mfu:25.04549064671562 +Step 21826 | loss:0.4687769114971161 lr:3.4503793407301425e-05 tokens_per_second_per_gpu:5062.004838559376 mfu:23.878340490568565 +Step 21827 | loss:0.398898720741272 lr:3.449967075379752e-05 tokens_per_second_per_gpu:4477.3159268672725 mfu:21.120263135901837 +Step 21828 | loss:0.41044434905052185 lr:3.4495548216873584e-05 tokens_per_second_per_gpu:5364.804631626675 mfu:25.306698777439024 +Step 21829 | loss:0.5207048058509827 lr:3.449142579656061e-05 tokens_per_second_per_gpu:3205.5452286480413 mfu:15.121103765945671 +Step 21830 | loss:0.44166117906570435 lr:3.448730349288959e-05 tokens_per_second_per_gpu:5126.348162389329 mfu:24.181858927177526 +Step 21831 | loss:0.28684934973716736 lr:3.448318130589153e-05 tokens_per_second_per_gpu:5294.8530714536655 mfu:24.97672533313608 +Step 21832 | loss:0.5276876091957092 lr:3.447905923559746e-05 tokens_per_second_per_gpu:4350.637170742781 mfu:20.522697829638087 +Step 21833 | loss:0.40682461857795715 lr:3.447493728203837e-05 tokens_per_second_per_gpu:4946.2974741020425 mfu:23.332529110711242 +Step 21834 | loss:0.51658034324646 lr:3.447081544524524e-05 tokens_per_second_per_gpu:4697.033119563025 mfu:22.156706621466373 +Step 21835 | loss:0.422279417514801 lr:3.4466693725249105e-05 tokens_per_second_per_gpu:4524.241588172704 mfu:21.34161948662296 +Step 21836 | loss:0.48379746079444885 lr:3.446257212208094e-05 tokens_per_second_per_gpu:5347.428550912421 mfu:25.224732840044812 +Step 21837 | loss:0.4734213948249817 lr:3.445845063577176e-05 tokens_per_second_per_gpu:4927.369626136324 mfu:23.24324322243308 +Step 21838 | loss:0.37372636795043945 lr:3.4454329266352545e-05 tokens_per_second_per_gpu:5237.645167375128 mfu:24.706865888922692 +Step 21839 | loss:0.44316473603248596 lr:3.445020801385431e-05 tokens_per_second_per_gpu:4006.600282767412 mfu:18.89981713031236 +Step 21840 | loss:0.4215828478336334 lr:3.444608687830803e-05 tokens_per_second_per_gpu:3569.9445849997915 mfu:16.84003770279814 +Step 21841 | loss:0.5005667209625244 lr:3.444196585974473e-05 tokens_per_second_per_gpu:3664.5708338558966 mfu:17.286405863555252 +Step 21842 | loss:0.3489530682563782 lr:3.44378449581954e-05 tokens_per_second_per_gpu:5340.796163067861 mfu:25.193446734979972 +Step 21843 | loss:0.4345402419567108 lr:3.443372417369102e-05 tokens_per_second_per_gpu:4347.112438557842 mfu:20.506071066540407 +Step 21844 | loss:0.43117478489875793 lr:3.442960350626256e-05 tokens_per_second_per_gpu:5384.82212218449 mfu:25.401124695736307 +Step 21845 | loss:0.5055328607559204 lr:3.442548295594107e-05 tokens_per_second_per_gpu:4408.756399766873 mfu:20.796856149107725 +Step 21846 | loss:0.4775759279727936 lr:3.4421362522757495e-05 tokens_per_second_per_gpu:4295.044821036508 mfu:20.260459231040674 +Step 21847 | loss:0.44196268916130066 lr:3.441724220674284e-05 tokens_per_second_per_gpu:3890.3162229314103 mfu:18.351285379959485 +Step 21848 | loss:0.3370310664176941 lr:3.44131220079281e-05 tokens_per_second_per_gpu:4092.1788175045704 mfu:19.30350567987119 +Step 21849 | loss:0.5595316290855408 lr:3.4409001926344265e-05 tokens_per_second_per_gpu:5102.973939506602 mfu:24.07159872978604 +Step 21850 | loss:0.3276008069515228 lr:3.4404881962022306e-05 tokens_per_second_per_gpu:4043.0235084606193 mfu:19.071631700350334 +Step 21851 | loss:0.40234851837158203 lr:3.4400762114993214e-05 tokens_per_second_per_gpu:5352.018925419825 mfu:25.246386419794888 +Step 21852 | loss:0.3984159231185913 lr:3.439664238528799e-05 tokens_per_second_per_gpu:3220.5395331400427 mfu:15.191834458526591 +Step 21853 | loss:0.44479456543922424 lr:3.439252277293761e-05 tokens_per_second_per_gpu:4223.1180141058 mfu:19.921168210769817 +Step 21854 | loss:0.3294545114040375 lr:3.4388403277973044e-05 tokens_per_second_per_gpu:4923.973217840914 mfu:23.227221785016617 +Step 21855 | loss:0.38862019777297974 lr:3.4384283900425305e-05 tokens_per_second_per_gpu:4302.778628108553 mfu:20.296940918523056 +Step 21856 | loss:0.36187031865119934 lr:3.438016464032534e-05 tokens_per_second_per_gpu:4324.736133730942 mfu:20.400518218880638 +Step 21857 | loss:0.4825681149959564 lr:3.437604549770419e-05 tokens_per_second_per_gpu:4453.400140903163 mfu:21.007448293055283 +Step 21858 | loss:0.42289361357688904 lr:3.437192647259276e-05 tokens_per_second_per_gpu:3936.6168923576092 mfu:18.56969353735166 +Step 21859 | loss:0.5218812823295593 lr:3.436780756502208e-05 tokens_per_second_per_gpu:4755.942542343392 mfu:22.434592419705318 +Step 21860 | loss:0.3534160554409027 lr:3.43636887750231e-05 tokens_per_second_per_gpu:4682.033967774397 mfu:22.08595306336023 +Step 21861 | loss:0.34449538588523865 lr:3.435957010262682e-05 tokens_per_second_per_gpu:4407.959037802939 mfu:20.793094856679865 +Step 21862 | loss:0.42244842648506165 lr:3.4355451547864216e-05 tokens_per_second_per_gpu:3339.743594129554 mfu:15.75414035252414 +Step 21863 | loss:0.31583935022354126 lr:3.4351333110766235e-05 tokens_per_second_per_gpu:4912.152324307025 mfu:23.171460613364857 +Step 21864 | loss:0.5060797929763794 lr:3.43472147913639e-05 tokens_per_second_per_gpu:5327.103295515767 mfu:25.128855142492636 +Step 21865 | loss:0.3775385320186615 lr:3.434309658968816e-05 tokens_per_second_per_gpu:3018.966889244374 mfu:14.240981905431298 +Step 21866 | loss:0.3733743131160736 lr:3.4338978505769984e-05 tokens_per_second_per_gpu:5357.947710369165 mfu:25.274353509955947 +Step 21867 | loss:0.43015551567077637 lr:3.4334860539640326e-05 tokens_per_second_per_gpu:3968.426581673958 mfu:18.71974526914952 +Step 21868 | loss:0.42232099175453186 lr:3.43307426913302e-05 tokens_per_second_per_gpu:5176.997031646643 mfu:24.42077828505246 +Step 21869 | loss:0.3973188102245331 lr:3.432662496087056e-05 tokens_per_second_per_gpu:3577.529433691435 mfu:16.875816728185253 +Step 21870 | loss:0.5511762499809265 lr:3.432250734829236e-05 tokens_per_second_per_gpu:5344.498848477591 mfu:25.21091293380094 +Step 21871 | loss:0.41790327429771423 lr:3.431838985362661e-05 tokens_per_second_per_gpu:5337.378274628031 mfu:25.177323972057394 +Step 21872 | loss:0.3680253326892853 lr:3.431427247690423e-05 tokens_per_second_per_gpu:4904.561037405061 mfu:23.13565121784119 +Step 21873 | loss:0.39266014099121094 lr:3.4310155218156215e-05 tokens_per_second_per_gpu:5356.493372700553 mfu:25.26749315103751 +Step 21874 | loss:0.3656037151813507 lr:3.430603807741351e-05 tokens_per_second_per_gpu:3484.377662498675 mfu:16.43640393013797 +Step 21875 | loss:0.3925897777080536 lr:3.430192105470711e-05 tokens_per_second_per_gpu:3732.5951112813573 mfu:17.607287986309668 +Step 21876 | loss:0.3958609998226166 lr:3.4297804150067946e-05 tokens_per_second_per_gpu:3788.530281184486 mfu:17.87114372626681 +Step 21877 | loss:0.45831334590911865 lr:3.429368736352702e-05 tokens_per_second_per_gpu:4340.932757155248 mfu:20.476920455002 +Step 21878 | loss:0.4243479371070862 lr:3.428957069511529e-05 tokens_per_second_per_gpu:4847.376456363313 mfu:22.86590179237166 +Step 21879 | loss:0.4073111116886139 lr:3.4285454144863684e-05 tokens_per_second_per_gpu:5319.953866380195 mfu:25.095130065441747 +Step 21880 | loss:0.3615649938583374 lr:3.4281337712803176e-05 tokens_per_second_per_gpu:3587.670915502072 mfu:16.923655828201248 +Step 21881 | loss:0.3706996440887451 lr:3.427722139896474e-05 tokens_per_second_per_gpu:5045.2268940223075 mfu:23.79919606357469 +Step 21882 | loss:0.43643489480018616 lr:3.427310520337933e-05 tokens_per_second_per_gpu:5188.936376554806 mfu:24.47709821204656 +Step 21883 | loss:0.3778288662433624 lr:3.426898912607789e-05 tokens_per_second_per_gpu:4380.842761001018 mfu:20.665182752491308 +Step 21884 | loss:0.4024507403373718 lr:3.42648731670914e-05 tokens_per_second_per_gpu:5350.909986466535 mfu:25.241155365547726 +Step 21885 | loss:0.3382886052131653 lr:3.426075732645081e-05 tokens_per_second_per_gpu:4125.612001735789 mfu:19.461215714179286 +Step 21886 | loss:0.4071943163871765 lr:3.4256641604187075e-05 tokens_per_second_per_gpu:4990.56597919987 mfu:23.541351202243504 +Step 21887 | loss:0.44251447916030884 lr:3.4252526000331145e-05 tokens_per_second_per_gpu:5341.292575567919 mfu:25.19578839743945 +Step 21888 | loss:0.4421860873699188 lr:3.424841051491397e-05 tokens_per_second_per_gpu:3113.484435234255 mfu:14.686837296221812 +Step 21889 | loss:0.36999982595443726 lr:3.424429514796652e-05 tokens_per_second_per_gpu:4667.679004435361 mfu:22.0182382520807 +Step 21890 | loss:0.48530420660972595 lr:3.424017989951972e-05 tokens_per_second_per_gpu:5305.958830629092 mfu:25.02911309400447 +Step 21891 | loss:0.47236406803131104 lr:3.4236064769604545e-05 tokens_per_second_per_gpu:4855.885958430121 mfu:22.90604256549156 +Step 21892 | loss:0.36872991919517517 lr:3.4231949758251935e-05 tokens_per_second_per_gpu:4426.303283234667 mfu:20.87962779223234 +Step 21893 | loss:0.3912622332572937 lr:3.422783486549287e-05 tokens_per_second_per_gpu:5047.886287172761 mfu:23.811740874804634 +Step 21894 | loss:0.5404648184776306 lr:3.422372009135824e-05 tokens_per_second_per_gpu:5332.206613421851 mfu:25.152928363771455 +Step 21895 | loss:0.35854679346084595 lr:3.4219605435879035e-05 tokens_per_second_per_gpu:4378.74135733845 mfu:20.655270072878544 +Step 21896 | loss:0.3910672068595886 lr:3.421549089908618e-05 tokens_per_second_per_gpu:3456.5140120253027 mfu:16.304966336826084 +Step 21897 | loss:0.3573310673236847 lr:3.421137648101065e-05 tokens_per_second_per_gpu:4500.99103798496 mfu:21.23194267443445 +Step 21898 | loss:0.3475707173347473 lr:3.4207262181683366e-05 tokens_per_second_per_gpu:4914.788553876536 mfu:23.183896158031384 +Step 21899 | loss:0.4773341119289398 lr:3.4203148001135264e-05 tokens_per_second_per_gpu:3696.7184374844724 mfu:17.438051594818013 +Step 21900 | loss:0.5103985071182251 lr:3.419903393939733e-05 tokens_per_second_per_gpu:4428.646380836894 mfu:20.89068058339572 +Step 21901 | loss:0.3792276084423065 lr:3.419491999650046e-05 tokens_per_second_per_gpu:5360.92744704629 mfu:25.28840943625745 +Step 21902 | loss:0.502292811870575 lr:3.419080617247562e-05 tokens_per_second_per_gpu:5025.912430869263 mfu:23.70808644549455 +Step 21903 | loss:0.39707598090171814 lr:3.418669246735373e-05 tokens_per_second_per_gpu:4286.263954114661 mfu:20.219038383599937 +Step 21904 | loss:0.4315031170845032 lr:3.418257888116576e-05 tokens_per_second_per_gpu:5311.964362221978 mfu:25.057442211177463 +Step 21905 | loss:0.37965908646583557 lr:3.4178465413942625e-05 tokens_per_second_per_gpu:5346.046373221176 mfu:25.218212872051378 +Step 21906 | loss:0.44384539127349854 lr:3.417435206571527e-05 tokens_per_second_per_gpu:5336.490193996287 mfu:25.173134744195277 +Step 21907 | loss:0.45582881569862366 lr:3.4170238836514645e-05 tokens_per_second_per_gpu:5102.268525702558 mfu:24.06827117251636 +Step 21908 | loss:0.46215900778770447 lr:3.416612572637166e-05 tokens_per_second_per_gpu:5107.925009946182 mfu:24.094953773789268 +Step 21909 | loss:0.45800700783729553 lr:3.416201273531727e-05 tokens_per_second_per_gpu:3380.1024733188865 mfu:15.944520071595067 +Step 21910 | loss:0.35827693343162537 lr:3.4157899863382395e-05 tokens_per_second_per_gpu:5343.451454345247 mfu:25.205972196974805 +Step 21911 | loss:0.4606374502182007 lr:3.415378711059798e-05 tokens_per_second_per_gpu:3927.846273231322 mfu:18.528321030511034 +Step 21912 | loss:0.4839322865009308 lr:3.414967447699495e-05 tokens_per_second_per_gpu:5202.421642284585 mfu:24.540710511317517 +Step 21913 | loss:0.4292837679386139 lr:3.4145561962604244e-05 tokens_per_second_per_gpu:4515.370792291165 mfu:21.299774428935898 +Step 21914 | loss:0.46475258469581604 lr:3.414144956745679e-05 tokens_per_second_per_gpu:3322.8838798161155 mfu:15.674610203543768 +Step 21915 | loss:0.3041350543498993 lr:3.413733729158354e-05 tokens_per_second_per_gpu:4840.584268928325 mfu:22.833861885374148 +Step 21916 | loss:0.48298412561416626 lr:3.413322513501537e-05 tokens_per_second_per_gpu:3680.857841488861 mfu:17.363234457409995 +Step 21917 | loss:0.4306361973285675 lr:3.412911309778325e-05 tokens_per_second_per_gpu:5103.608281809005 mfu:24.074591030656432 +Step 21918 | loss:0.38088664412498474 lr:3.412500117991809e-05 tokens_per_second_per_gpu:4039.3804553472078 mfu:19.05444679724542 +Step 21919 | loss:0.35452210903167725 lr:3.4120889381450816e-05 tokens_per_second_per_gpu:5359.653033758674 mfu:25.282397811342435 +Step 21920 | loss:0.29865533113479614 lr:3.4116777702412374e-05 tokens_per_second_per_gpu:5360.621392082128 mfu:25.286965722773136 +Step 21921 | loss:0.363582044839859 lr:3.411266614283367e-05 tokens_per_second_per_gpu:4477.926886538798 mfu:21.123145136912886 +Step 21922 | loss:0.41734564304351807 lr:3.410855470274564e-05 tokens_per_second_per_gpu:4861.610767214463 mfu:22.933047465280396 +Step 21923 | loss:0.3919468820095062 lr:3.410444338217918e-05 tokens_per_second_per_gpu:4740.639541723764 mfu:22.36240555482139 +Step 21924 | loss:0.4652525782585144 lr:3.410033218116524e-05 tokens_per_second_per_gpu:5350.680722393726 mfu:25.240073887799802 +Step 21925 | loss:0.25145065784454346 lr:3.4096221099734726e-05 tokens_per_second_per_gpu:4551.323273048057 mfu:21.4693684147919 +Step 21926 | loss:0.45199379324913025 lr:3.409211013791855e-05 tokens_per_second_per_gpu:5326.972071107064 mfu:25.128236134567587 +Step 21927 | loss:0.36796772480010986 lr:3.4087999295747666e-05 tokens_per_second_per_gpu:5297.399095421653 mfu:24.988735362589562 +Step 21928 | loss:0.4286940395832062 lr:3.4083888573252954e-05 tokens_per_second_per_gpu:4044.25067716997 mfu:19.077420464528913 +Step 21929 | loss:0.4096425771713257 lr:3.4079777970465376e-05 tokens_per_second_per_gpu:5037.83457596035 mfu:23.76432523801692 +Step 21930 | loss:0.4424101412296295 lr:3.4075667487415785e-05 tokens_per_second_per_gpu:4654.568197489778 mfu:21.956392338784052 +Step 21931 | loss:0.3308427631855011 lr:3.407155712413515e-05 tokens_per_second_per_gpu:5332.447474734172 mfu:25.154064547676846 +Step 21932 | loss:0.508202850818634 lr:3.4067446880654346e-05 tokens_per_second_per_gpu:4568.344176468898 mfu:21.549658920293588 +Step 21933 | loss:0.4794025719165802 lr:3.4063336757004324e-05 tokens_per_second_per_gpu:3882.4423630050687 mfu:18.3141430392679 +Step 21934 | loss:0.35380640625953674 lr:3.4059226753215976e-05 tokens_per_second_per_gpu:4655.524012015103 mfu:21.960901078978786 +Step 21935 | loss:0.3967796266078949 lr:3.405511686932021e-05 tokens_per_second_per_gpu:5395.3315096886545 mfu:25.45069926967986 +Step 21936 | loss:0.3690375089645386 lr:3.405100710534798e-05 tokens_per_second_per_gpu:5327.9054726805225 mfu:25.132639149044323 +Step 21937 | loss:0.4712718725204468 lr:3.404689746133012e-05 tokens_per_second_per_gpu:5327.805705376815 mfu:25.132168529650677 +Step 21938 | loss:0.4353483021259308 lr:3.404278793729759e-05 tokens_per_second_per_gpu:5364.65766399782 mfu:25.306005506058455 +Step 21939 | loss:0.508402943611145 lr:3.403867853328129e-05 tokens_per_second_per_gpu:4380.93151487949 mfu:20.66560141967904 +Step 21940 | loss:0.32129934430122375 lr:3.403456924931212e-05 tokens_per_second_per_gpu:4319.7748990286545 mfu:20.377115228316022 +Step 21941 | loss:0.4396195113658905 lr:3.4030460085421e-05 tokens_per_second_per_gpu:3442.992150650693 mfu:16.2411814096546 +Step 21942 | loss:0.3360823094844818 lr:3.402635104163882e-05 tokens_per_second_per_gpu:3490.569613095404 mfu:16.46561241755845 +Step 21943 | loss:0.48342815041542053 lr:3.40222421179965e-05 tokens_per_second_per_gpu:3531.0550440886063 mfu:16.656589103080197 +Step 21944 | loss:0.42637014389038086 lr:3.4018133314524945e-05 tokens_per_second_per_gpu:5324.419497597665 mfu:25.116195209809803 +Step 21945 | loss:0.36629900336265564 lr:3.401402463125504e-05 tokens_per_second_per_gpu:3817.4577067036676 mfu:18.007599327968517 +Step 21946 | loss:0.3830568194389343 lr:3.400991606821769e-05 tokens_per_second_per_gpu:3408.4519051137727 mfu:16.078249178283375 +Step 21947 | loss:0.4306153655052185 lr:3.40058076254438e-05 tokens_per_second_per_gpu:5357.65406485762 mfu:25.272968333976902 +Step 21948 | loss:0.47221556305885315 lr:3.4001699302964265e-05 tokens_per_second_per_gpu:4671.170766093742 mfu:22.034709487579253 +Step 21949 | loss:0.5148257613182068 lr:3.399759110081001e-05 tokens_per_second_per_gpu:4247.6296835289495 mfu:20.036793937560862 +Step 21950 | loss:0.4650159180164337 lr:3.3993483019011904e-05 tokens_per_second_per_gpu:5366.550243047916 mfu:25.314933124345043 +Step 21951 | loss:0.4087374210357666 lr:3.398937505760087e-05 tokens_per_second_per_gpu:5323.370831001377 mfu:25.111248470554145 +Step 21952 | loss:0.3895433843135834 lr:3.3985267216607776e-05 tokens_per_second_per_gpu:3860.4301988794487 mfu:18.210307905424877 +Step 21953 | loss:0.4392179250717163 lr:3.398115949606352e-05 tokens_per_second_per_gpu:5332.750695179238 mfu:25.155494890250967 +Step 21954 | loss:0.42491614818573 lr:3.3977051895999024e-05 tokens_per_second_per_gpu:5246.480423703172 mfu:24.74854330047225 +Step 21955 | loss:0.44141772389411926 lr:3.397294441644515e-05 tokens_per_second_per_gpu:5321.785058183743 mfu:25.10376810961262 +Step 21956 | loss:0.28308987617492676 lr:3.396883705743282e-05 tokens_per_second_per_gpu:3700.6707361056633 mfu:17.45669526174585 +Step 21957 | loss:0.3488967716693878 lr:3.39647298189929e-05 tokens_per_second_per_gpu:3297.9212718472604 mfu:15.556857322694578 +Step 21958 | loss:0.5650444626808167 lr:3.396062270115632e-05 tokens_per_second_per_gpu:3858.6050461265118 mfu:18.201698348486104 +Step 21959 | loss:0.34604954719543457 lr:3.395651570395392e-05 tokens_per_second_per_gpu:5352.225236493095 mfu:25.247359624327803 +Step 21960 | loss:0.42849528789520264 lr:3.395240882741662e-05 tokens_per_second_per_gpu:5115.606501730096 mfu:24.13118868113163 +Step 21961 | loss:0.48588496446609497 lr:3.394830207157529e-05 tokens_per_second_per_gpu:3827.187830219337 mfu:18.053497980722184 +Step 21962 | loss:0.3759140372276306 lr:3.394419543646083e-05 tokens_per_second_per_gpu:3868.6974044437334 mfu:18.24930572460233 +Step 21963 | loss:0.5661906003952026 lr:3.3940088922104127e-05 tokens_per_second_per_gpu:3799.714234889978 mfu:17.92390034407488 +Step 21964 | loss:0.33754822611808777 lr:3.393598252853606e-05 tokens_per_second_per_gpu:5159.130168227873 mfu:24.336497241904883 +Step 21965 | loss:0.4999774694442749 lr:3.393187625578755e-05 tokens_per_second_per_gpu:5040.856655963257 mfu:23.77858090501065 +Step 21966 | loss:0.31469982862472534 lr:3.392777010388942e-05 tokens_per_second_per_gpu:4986.046004008425 mfu:23.520029708078173 +Step 21967 | loss:0.4761206805706024 lr:3.3923664072872586e-05 tokens_per_second_per_gpu:4392.780846264142 mfu:20.72149673752485 +Step 21968 | loss:0.38960516452789307 lr:3.391955816276793e-05 tokens_per_second_per_gpu:3894.2558441779915 mfu:18.369869245548305 +Step 21969 | loss:0.35886189341545105 lr:3.391545237360632e-05 tokens_per_second_per_gpu:5367.270544732708 mfu:25.318330910288342 +Step 21970 | loss:0.4751100242137909 lr:3.391134670541865e-05 tokens_per_second_per_gpu:5369.341979078522 mfu:25.32810221952069 +Step 21971 | loss:0.37628835439682007 lr:3.390724115823579e-05 tokens_per_second_per_gpu:4922.273741082646 mfu:23.219205063187236 +Step 21972 | loss:0.46620437502861023 lr:3.390313573208865e-05 tokens_per_second_per_gpu:4023.4782191775926 mfu:18.979433236032076 +Step 21973 | loss:0.5191790461540222 lr:3.389903042700805e-05 tokens_per_second_per_gpu:4135.116083646409 mfu:19.506048090115264 +Step 21974 | loss:0.47063952684402466 lr:3.389492524302491e-05 tokens_per_second_per_gpu:5341.718676296868 mfu:25.197798387278183 +Step 21975 | loss:0.45457103848457336 lr:3.389082018017007e-05 tokens_per_second_per_gpu:5343.484432696377 mfu:25.206127761483774 +Step 21976 | loss:0.4140211045742035 lr:3.388671523847445e-05 tokens_per_second_per_gpu:3939.007586917193 mfu:18.58097085148377 +Step 21977 | loss:0.5073186755180359 lr:3.3882610417968896e-05 tokens_per_second_per_gpu:3461.185410848209 mfu:16.32700212209656 +Step 21978 | loss:0.43202143907546997 lr:3.387850571868428e-05 tokens_per_second_per_gpu:5326.765957102103 mfu:25.12726385963951 +Step 21979 | loss:0.3996395170688629 lr:3.387440114065148e-05 tokens_per_second_per_gpu:4500.142912346145 mfu:21.227941921091006 +Step 21980 | loss:0.4360032379627228 lr:3.3870296683901384e-05 tokens_per_second_per_gpu:5382.13314688033 mfu:25.388440340440088 +Step 21981 | loss:0.3878431022167206 lr:3.386619234846484e-05 tokens_per_second_per_gpu:5342.715395983353 mfu:25.20250008409725 +Step 21982 | loss:0.3851250112056732 lr:3.38620881343727e-05 tokens_per_second_per_gpu:5350.12290112807 mfu:25.237442549715578 +Step 21983 | loss:0.39052310585975647 lr:3.385798404165587e-05 tokens_per_second_per_gpu:4762.383310681017 mfu:22.46497462286229 +Step 21984 | loss:0.5050036311149597 lr:3.385388007034519e-05 tokens_per_second_per_gpu:5229.835437016321 mfu:24.6700260583428 +Step 21985 | loss:0.3982856273651123 lr:3.384977622047155e-05 tokens_per_second_per_gpu:4251.395781153622 mfu:20.05455926261895 +Step 21986 | loss:0.45981475710868835 lr:3.3845672492065803e-05 tokens_per_second_per_gpu:4543.62986650332 mfu:21.433077303489867 +Step 21987 | loss:0.47262808680534363 lr:3.3841568885158826e-05 tokens_per_second_per_gpu:3821.632487694326 mfu:18.02729248219254 +Step 21988 | loss:0.40223950147628784 lr:3.3837465399781465e-05 tokens_per_second_per_gpu:4454.198595778381 mfu:21.011214740932132 +Step 21989 | loss:0.3021408021450043 lr:3.3833362035964575e-05 tokens_per_second_per_gpu:3783.737796867154 mfu:17.848536760059808 +Step 21990 | loss:0.46601176261901855 lr:3.382925879373904e-05 tokens_per_second_per_gpu:3395.3750630301265 mfu:16.016563483035487 +Step 21991 | loss:0.43680816888809204 lr:3.38251556731357e-05 tokens_per_second_per_gpu:4209.364886332472 mfu:19.856292360537463 +Step 21992 | loss:0.3905152380466461 lr:3.382105267418544e-05 tokens_per_second_per_gpu:3895.6875966168377 mfu:18.3766230660845 +Step 21993 | loss:0.43270376324653625 lr:3.3816949796919116e-05 tokens_per_second_per_gpu:4828.390640573605 mfu:22.776342459977407 +Step 21994 | loss:0.459910124540329 lr:3.3812847041367576e-05 tokens_per_second_per_gpu:5328.650481150535 mfu:25.136153481108924 +Step 21995 | loss:0.38388627767562866 lr:3.380874440756167e-05 tokens_per_second_per_gpu:5327.79035842489 mfu:25.132096135459985 +Step 21996 | loss:0.42928406596183777 lr:3.380464189553226e-05 tokens_per_second_per_gpu:3225.6053594428354 mfu:15.215730825515955 +Step 21997 | loss:0.4191983640193939 lr:3.380053950531022e-05 tokens_per_second_per_gpu:3936.7616821164024 mfu:18.57037653534757 +Step 21998 | loss:0.4529406428337097 lr:3.379643723692637e-05 tokens_per_second_per_gpu:4612.3896648650925 mfu:21.75742900399368 +Step 21999 | loss:0.43910202383995056 lr:3.37923350904116e-05 tokens_per_second_per_gpu:3868.945149316709 mfu:18.250474379437787 +Step 22000 | loss:0.4639257490634918 lr:3.3788233065796726e-05 tokens_per_second_per_gpu:4993.833704504833 mfu:23.556765620038444 +Step 22001 | loss:0.37730300426483154 lr:3.378413116311265e-05 tokens_per_second_per_gpu:4563.202592632596 mfu:21.525405192093096 +Step 22002 | loss:0.4345493018627167 lr:3.378002938239017e-05 tokens_per_second_per_gpu:4335.592260567587 mfu:20.45172842141533 +Step 22003 | loss:0.4855799376964569 lr:3.377592772366017e-05 tokens_per_second_per_gpu:3741.797034088309 mfu:17.650695026199983 +Step 22004 | loss:0.40310418605804443 lr:3.377182618695348e-05 tokens_per_second_per_gpu:4894.945244812044 mfu:23.090291883556535 +Step 22005 | loss:0.413280725479126 lr:3.376772477230095e-05 tokens_per_second_per_gpu:5363.184924957303 mfu:25.299058344729143 +Step 22006 | loss:0.41531044244766235 lr:3.376362347973344e-05 tokens_per_second_per_gpu:4048.960613647913 mfu:19.0996380384934 +Step 22007 | loss:0.5035088658332825 lr:3.375952230928179e-05 tokens_per_second_per_gpu:3745.6653905192984 mfu:17.668942723494624 +Step 22008 | loss:0.544420599937439 lr:3.3755421260976844e-05 tokens_per_second_per_gpu:5353.747149820101 mfu:25.25453874915648 +Step 22009 | loss:0.4547623097896576 lr:3.375132033484946e-05 tokens_per_second_per_gpu:4740.903470300452 mfu:22.363650550948485 +Step 22010 | loss:0.3794361650943756 lr:3.374721953093046e-05 tokens_per_second_per_gpu:4909.8689320766025 mfu:23.160689462627065 +Step 22011 | loss:0.4778716266155243 lr:3.374311884925069e-05 tokens_per_second_per_gpu:5346.702738978649 mfu:25.221309061316926 +Step 22012 | loss:0.498941570520401 lr:3.3739018289841e-05 tokens_per_second_per_gpu:4560.181425099344 mfu:21.51115382937441 +Step 22013 | loss:0.5382094979286194 lr:3.373491785273223e-05 tokens_per_second_per_gpu:5352.117837271714 mfu:25.246853003876353 +Step 22014 | loss:0.39280709624290466 lr:3.373081753795522e-05 tokens_per_second_per_gpu:5362.429150550599 mfu:25.295493227904313 +Step 22015 | loss:0.493522584438324 lr:3.3726717345540806e-05 tokens_per_second_per_gpu:4342.206607066091 mfu:20.482929422372656 +Step 22016 | loss:0.3893025517463684 lr:3.3722617275519836e-05 tokens_per_second_per_gpu:5398.131735206536 mfu:25.463908411215442 +Step 22017 | loss:0.4573167562484741 lr:3.3718517327923126e-05 tokens_per_second_per_gpu:4108.403333361743 mfu:19.380039489357323 +Step 22018 | loss:0.3778226673603058 lr:3.371441750278153e-05 tokens_per_second_per_gpu:5341.161234964277 mfu:25.195168841403053 +Step 22019 | loss:0.5402385592460632 lr:3.371031780012587e-05 tokens_per_second_per_gpu:4168.179025911532 mfu:19.662011629899496 +Step 22020 | loss:0.526893138885498 lr:3.3706218219986994e-05 tokens_per_second_per_gpu:4403.62234324184 mfu:20.772637928518776 +Step 22021 | loss:0.45841294527053833 lr:3.370211876239573e-05 tokens_per_second_per_gpu:5396.5513625973545 mfu:25.45645352397856 +Step 22022 | loss:0.436669260263443 lr:3.369801942738291e-05 tokens_per_second_per_gpu:5318.25211219743 mfu:25.087102600612436 +Step 22023 | loss:0.509900689125061 lr:3.369392021497938e-05 tokens_per_second_per_gpu:3870.280066681864 mfu:18.256771412409613 +Step 22024 | loss:0.499156653881073 lr:3.368982112521594e-05 tokens_per_second_per_gpu:4877.951918179236 mfu:23.010131462469083 +Step 22025 | loss:0.43005555868148804 lr:3.3685722158123435e-05 tokens_per_second_per_gpu:4660.098938978822 mfu:21.982481790029677 +Step 22026 | loss:0.4884510636329651 lr:3.3681623313732705e-05 tokens_per_second_per_gpu:4300.329634989898 mfu:20.28538860014105 +Step 22027 | loss:0.4059208035469055 lr:3.367752459207455e-05 tokens_per_second_per_gpu:5226.781628919033 mfu:24.655620724514506 +Step 22028 | loss:0.5047079920768738 lr:3.367342599317983e-05 tokens_per_second_per_gpu:4559.074043618517 mfu:21.505930121990584 +Step 22029 | loss:0.3209090828895569 lr:3.366932751707936e-05 tokens_per_second_per_gpu:4837.13622426686 mfu:22.817596870408934 +Step 22030 | loss:0.45725739002227783 lr:3.366522916380397e-05 tokens_per_second_per_gpu:4145.132110369299 mfu:19.553295397077513 +Step 22031 | loss:0.4353621304035187 lr:3.366113093338446e-05 tokens_per_second_per_gpu:5378.80199677755 mfu:25.37272673705246 +Step 22032 | loss:0.3631935119628906 lr:3.365703282585167e-05 tokens_per_second_per_gpu:4076.8336805468507 mfu:19.23112005069129 +Step 22033 | loss:0.37468820810317993 lr:3.365293484123643e-05 tokens_per_second_per_gpu:5341.029959195503 mfu:25.19454959120379 +Step 22034 | loss:0.4135146737098694 lr:3.364883697956953e-05 tokens_per_second_per_gpu:3980.390711840263 mfu:18.776182112435077 +Step 22035 | loss:0.4039251208305359 lr:3.364473924088183e-05 tokens_per_second_per_gpu:5327.414163677935 mfu:25.130321560653158 +Step 22036 | loss:0.37215688824653625 lr:3.3640641625204136e-05 tokens_per_second_per_gpu:5332.343151571828 mfu:25.1535724375207 +Step 22037 | loss:0.42273449897766113 lr:3.363654413256725e-05 tokens_per_second_per_gpu:3678.663279727529 mfu:17.35288233515094 +Step 22038 | loss:0.35171183943748474 lr:3.363244676300203e-05 tokens_per_second_per_gpu:4227.26746019254 mfu:19.94074185592947 +Step 22039 | loss:0.43608030676841736 lr:3.362834951653926e-05 tokens_per_second_per_gpu:5346.8994649702545 mfu:25.222237051384244 +Step 22040 | loss:0.32054394483566284 lr:3.362425239320975e-05 tokens_per_second_per_gpu:5299.867591359855 mfu:25.000379679098895 +Step 22041 | loss:0.44252103567123413 lr:3.362015539304432e-05 tokens_per_second_per_gpu:5351.057982117902 mfu:25.241853486285883 +Step 22042 | loss:0.4994407296180725 lr:3.36160585160738e-05 tokens_per_second_per_gpu:5357.27248441283 mfu:25.271168353914486 +Step 22043 | loss:0.4315589964389801 lr:3.361196176232899e-05 tokens_per_second_per_gpu:3968.773095105591 mfu:18.721379832127823 +Step 22044 | loss:0.5376952290534973 lr:3.360786513184072e-05 tokens_per_second_per_gpu:3259.9510353356263 mfu:15.377745238678822 +Step 22045 | loss:0.3862263262271881 lr:3.360376862463979e-05 tokens_per_second_per_gpu:3705.8700306644637 mfu:17.481221221270836 +Step 22046 | loss:0.4856358468532562 lr:3.3599672240757e-05 tokens_per_second_per_gpu:4660.785306723399 mfu:21.98571950378726 +Step 22047 | loss:0.3881954252719879 lr:3.359557598022316e-05 tokens_per_second_per_gpu:4513.606542168121 mfu:21.291452160093392 +Step 22048 | loss:0.39888280630111694 lr:3.35914798430691e-05 tokens_per_second_per_gpu:4268.011360626618 mfu:20.13293778590326 +Step 22049 | loss:0.5426786541938782 lr:3.3587383829325605e-05 tokens_per_second_per_gpu:4281.084236777912 mfu:20.194604773171587 +Step 22050 | loss:0.4058350920677185 lr:3.358328793902348e-05 tokens_per_second_per_gpu:5289.224556123992 mfu:24.950174666002273 +Step 22051 | loss:0.4101630747318268 lr:3.357919217219356e-05 tokens_per_second_per_gpu:5326.925267992219 mfu:25.12801535628943 +Step 22052 | loss:0.41372647881507874 lr:3.3575096528866636e-05 tokens_per_second_per_gpu:3756.0371933459787 mfu:17.717868287040115 +Step 22053 | loss:0.4286860227584839 lr:3.35710010090735e-05 tokens_per_second_per_gpu:5102.518613389015 mfu:24.069450878802733 +Step 22054 | loss:0.4465887248516083 lr:3.3566905612844945e-05 tokens_per_second_per_gpu:4072.345922602278 mfu:19.209950530776275 +Step 22055 | loss:0.3941468298435211 lr:3.3562810340211804e-05 tokens_per_second_per_gpu:5316.594431316386 mfu:25.079283037067032 +Step 22056 | loss:0.5570533275604248 lr:3.355871519120487e-05 tokens_per_second_per_gpu:4438.712931172857 mfu:20.93816621886105 +Step 22057 | loss:0.4628702700138092 lr:3.355462016585492e-05 tokens_per_second_per_gpu:4863.810017497524 mfu:22.943421704096387 +Step 22058 | loss:0.4400920569896698 lr:3.355052526419278e-05 tokens_per_second_per_gpu:5371.4280152960655 mfu:25.33794240827332 +Step 22059 | loss:0.3809788227081299 lr:3.354643048624925e-05 tokens_per_second_per_gpu:5309.434142252643 mfu:25.045506731882067 +Step 22060 | loss:0.5001879930496216 lr:3.35423358320551e-05 tokens_per_second_per_gpu:3764.1381725008173 mfu:17.756081987883057 +Step 22061 | loss:0.4741799533367157 lr:3.353824130164114e-05 tokens_per_second_per_gpu:5372.882991551252 mfu:25.34480577951359 +Step 22062 | loss:0.40892109274864197 lr:3.3534146895038174e-05 tokens_per_second_per_gpu:5278.630595479867 mfu:24.900201146127863 +Step 22063 | loss:0.3322497308254242 lr:3.353005261227699e-05 tokens_per_second_per_gpu:4118.410079650382 mfu:19.427243018927808 +Step 22064 | loss:0.44577839970588684 lr:3.3525958453388374e-05 tokens_per_second_per_gpu:4962.83691114844 mfu:23.410548457178045 +Step 22065 | loss:0.4148865342140198 lr:3.3521864418403136e-05 tokens_per_second_per_gpu:5364.293936770972 mfu:25.304289742671152 +Step 22066 | loss:0.36142873764038086 lr:3.351777050735207e-05 tokens_per_second_per_gpu:4335.235408188593 mfu:20.45004508785832 +Step 22067 | loss:0.5154767036437988 lr:3.3513676720265934e-05 tokens_per_second_per_gpu:4680.045189155306 mfu:22.076571655293353 +Step 22068 | loss:0.5263433456420898 lr:3.3509583057175546e-05 tokens_per_second_per_gpu:4484.8711421897 mfu:21.155902375630383 +Step 22069 | loss:0.3585885465145111 lr:3.350548951811169e-05 tokens_per_second_per_gpu:3266.8529773301602 mfu:15.41030287665977 +Step 22070 | loss:0.460560142993927 lr:3.350139610310514e-05 tokens_per_second_per_gpu:3963.5455290301215 mfu:18.69672051103509 +Step 22071 | loss:0.3939862549304962 lr:3.349730281218672e-05 tokens_per_second_per_gpu:4619.513702757558 mfu:21.79103430621864 +Step 22072 | loss:0.4575318396091461 lr:3.349320964538717e-05 tokens_per_second_per_gpu:3558.216720736846 mfu:16.78471536608985 +Step 22073 | loss:0.595104455947876 lr:3.3489116602737296e-05 tokens_per_second_per_gpu:5078.359956606048 mfu:23.955490372865597 +Step 22074 | loss:0.4630228877067566 lr:3.34850236842679e-05 tokens_per_second_per_gpu:5318.8048588098845 mfu:25.089710000691685 +Step 22075 | loss:0.40729793906211853 lr:3.348093089000974e-05 tokens_per_second_per_gpu:5308.918638943939 mfu:25.043075014820978 +Step 22076 | loss:0.3762252926826477 lr:3.3476838219993615e-05 tokens_per_second_per_gpu:4042.074038700933 mfu:19.06715288949765 +Step 22077 | loss:0.32004985213279724 lr:3.347274567425028e-05 tokens_per_second_per_gpu:4641.344304720531 mfu:21.894012980362753 +Step 22078 | loss:0.44082286953926086 lr:3.346865325281055e-05 tokens_per_second_per_gpu:5312.731468460733 mfu:25.06106078218749 +Step 22079 | loss:0.41641420125961304 lr:3.346456095570516e-05 tokens_per_second_per_gpu:5376.383516973875 mfu:25.36131835518332 +Step 22080 | loss:0.41574805974960327 lr:3.346046878296494e-05 tokens_per_second_per_gpu:4770.897102299864 mfu:22.505135630531026 +Step 22081 | loss:0.5488517880439758 lr:3.345637673462065e-05 tokens_per_second_per_gpu:3973.071229839606 mfu:18.74165486700558 +Step 22082 | loss:0.43375054001808167 lr:3.345228481070306e-05 tokens_per_second_per_gpu:4980.223303612553 mfu:23.49256303685571 +Step 22083 | loss:0.387642502784729 lr:3.3448193011242936e-05 tokens_per_second_per_gpu:4305.825510154547 mfu:20.311313580985697 +Step 22084 | loss:0.3931516408920288 lr:3.3444101336271074e-05 tokens_per_second_per_gpu:4476.513858390545 mfu:21.116479642050116 +Step 22085 | loss:0.5668643712997437 lr:3.344000978581823e-05 tokens_per_second_per_gpu:5063.162981494416 mfu:23.883803648392707 +Step 22086 | loss:0.45736634731292725 lr:3.343591835991519e-05 tokens_per_second_per_gpu:5326.515398433316 mfu:25.126081931649125 +Step 22087 | loss:0.4013582170009613 lr:3.343182705859271e-05 tokens_per_second_per_gpu:5307.067831618404 mfu:25.034344440886134 +Step 22088 | loss:0.2833334803581238 lr:3.3427735881881605e-05 tokens_per_second_per_gpu:5131.067726098472 mfu:24.20412191443415 +Step 22089 | loss:0.37265127897262573 lr:3.3423644829812595e-05 tokens_per_second_per_gpu:3989.375275840015 mfu:18.818563833746598 +Step 22090 | loss:0.4545336365699768 lr:3.341955390241645e-05 tokens_per_second_per_gpu:5366.4492287680605 mfu:25.31445662275215 +Step 22091 | loss:0.47586023807525635 lr:3.341546309972398e-05 tokens_per_second_per_gpu:4581.9681832720935 mfu:21.61392568487952 +Step 22092 | loss:0.35388830304145813 lr:3.341137242176592e-05 tokens_per_second_per_gpu:3618.425554986365 mfu:17.06873070993124 +Step 22093 | loss:0.4557587504386902 lr:3.3407281868573025e-05 tokens_per_second_per_gpu:4974.287147362222 mfu:23.464561175010203 +Step 22094 | loss:0.4188729226589203 lr:3.340319144017611e-05 tokens_per_second_per_gpu:4867.511535659361 mfu:22.960882396809893 +Step 22095 | loss:0.5078631043434143 lr:3.33991011366059e-05 tokens_per_second_per_gpu:5356.290590995448 mfu:25.266536595136404 +Step 22096 | loss:0.4053654670715332 lr:3.339501095789318e-05 tokens_per_second_per_gpu:5331.950754337515 mfu:25.15172143281658 +Step 22097 | loss:0.43812987208366394 lr:3.339092090406867e-05 tokens_per_second_per_gpu:4329.94434307416 mfu:20.425086230964745 +Step 22098 | loss:0.40974733233451843 lr:3.3386830975163186e-05 tokens_per_second_per_gpu:5325.214878189592 mfu:25.119947155767846 +Step 22099 | loss:0.3714201748371124 lr:3.3382741171207456e-05 tokens_per_second_per_gpu:5321.234056437253 mfu:25.101168940363834 +Step 22100 | loss:0.41226229071617126 lr:3.337865149223226e-05 tokens_per_second_per_gpu:5359.242899286017 mfu:25.280463137058753 +Step 22101 | loss:0.49751317501068115 lr:3.337456193826834e-05 tokens_per_second_per_gpu:4741.946497199531 mfu:22.368570686790186 +Step 22102 | loss:0.4057604968547821 lr:3.3370472509346456e-05 tokens_per_second_per_gpu:5303.464224094864 mfu:25.017345609358756 +Step 22103 | loss:0.366179496049881 lr:3.336638320549739e-05 tokens_per_second_per_gpu:3078.130378993375 mfu:14.52006618090958 +Step 22104 | loss:0.463235080242157 lr:3.336229402675187e-05 tokens_per_second_per_gpu:5336.273943194905 mfu:25.172114652268966 +Step 22105 | loss:0.3674354553222656 lr:3.3358204973140655e-05 tokens_per_second_per_gpu:4263.927450876401 mfu:20.113673286824636 +Step 22106 | loss:0.44329649209976196 lr:3.3354116044694497e-05 tokens_per_second_per_gpu:5105.714121994047 mfu:24.08452463810266 +Step 22107 | loss:0.4209831655025482 lr:3.335002724144417e-05 tokens_per_second_per_gpu:3202.922458055949 mfu:15.108731709571964 +Step 22108 | loss:0.47685420513153076 lr:3.334593856342041e-05 tokens_per_second_per_gpu:4528.887045480789 mfu:21.363532901341944 +Step 22109 | loss:0.33811360597610474 lr:3.334185001065395e-05 tokens_per_second_per_gpu:2958.4933417136363 mfu:13.955717863877656 +Step 22110 | loss:0.34984925389289856 lr:3.3337761583175595e-05 tokens_per_second_per_gpu:5336.730772219091 mfu:25.174269592719117 +Step 22111 | loss:0.30345043540000916 lr:3.333367328101605e-05 tokens_per_second_per_gpu:4451.552743645031 mfu:20.9987338049902 +Step 22112 | loss:0.3834679126739502 lr:3.332958510420607e-05 tokens_per_second_per_gpu:4321.127833637567 mfu:20.38349725170071 +Step 22113 | loss:0.4848896265029907 lr:3.33254970527764e-05 tokens_per_second_per_gpu:3905.787559748499 mfu:18.424266315407948 +Step 22114 | loss:0.38326510787010193 lr:3.332140912675781e-05 tokens_per_second_per_gpu:5052.534439130237 mfu:23.833667000644553 +Step 22115 | loss:0.46324223279953003 lr:3.331732132618101e-05 tokens_per_second_per_gpu:4682.090425249555 mfu:22.08621938290319 +Step 22116 | loss:0.3324209153652191 lr:3.331323365107678e-05 tokens_per_second_per_gpu:5343.52593253461 mfu:25.20632352330132 +Step 22117 | loss:0.2751140892505646 lr:3.3309146101475863e-05 tokens_per_second_per_gpu:5341.192112585198 mfu:25.1953144964092 +Step 22118 | loss:0.38413989543914795 lr:3.3305058677408974e-05 tokens_per_second_per_gpu:4112.181182513113 mfu:19.397860248371867 +Step 22119 | loss:0.4103136956691742 lr:3.330097137890685e-05 tokens_per_second_per_gpu:5376.4688717915615 mfu:25.361720988421382 +Step 22120 | loss:0.338364839553833 lr:3.3296884206000275e-05 tokens_per_second_per_gpu:5364.3567666213285 mfu:25.304586121795158 +Step 22121 | loss:0.422221302986145 lr:3.329279715871996e-05 tokens_per_second_per_gpu:4477.397878602216 mfu:21.120649716217972 +Step 22122 | loss:0.4119943678379059 lr:3.328871023709664e-05 tokens_per_second_per_gpu:3970.836761064117 mfu:18.73111449655191 +Step 22123 | loss:0.38553088903427124 lr:3.3284623441161064e-05 tokens_per_second_per_gpu:3544.067541352856 mfu:16.717971272836632 +Step 22124 | loss:0.3303995430469513 lr:3.328053677094399e-05 tokens_per_second_per_gpu:5385.784367361534 mfu:25.40566377041238 +Step 22125 | loss:0.5632629990577698 lr:3.327645022647612e-05 tokens_per_second_per_gpu:5306.134437429995 mfu:25.02994146124594 +Step 22126 | loss:0.4341755509376526 lr:3.3272363807788185e-05 tokens_per_second_per_gpu:3301.530819197158 mfu:15.573884173396328 +Step 22127 | loss:0.44506338238716125 lr:3.3268277514910954e-05 tokens_per_second_per_gpu:5343.727279767447 mfu:25.207273312551997 +Step 22128 | loss:0.37295037508010864 lr:3.326419134787514e-05 tokens_per_second_per_gpu:4665.238932021064 mfu:22.006728014183178 +Step 22129 | loss:0.34131351113319397 lr:3.3260105306711474e-05 tokens_per_second_per_gpu:5344.011175781518 mfu:25.208612498488023 +Step 22130 | loss:0.4634702205657959 lr:3.325601939145069e-05 tokens_per_second_per_gpu:4281.110055392479 mfu:20.194726563981607 +Step 22131 | loss:0.4621676206588745 lr:3.325193360212352e-05 tokens_per_second_per_gpu:3345.2261969144074 mfu:15.780002725288787 +Step 22132 | loss:0.41882753372192383 lr:3.3247847938760726e-05 tokens_per_second_per_gpu:5308.101681220799 mfu:25.039221285853564 +Step 22133 | loss:0.3842509984970093 lr:3.324376240139297e-05 tokens_per_second_per_gpu:4610.174597472333 mfu:21.74698015317256 +Step 22134 | loss:0.38517799973487854 lr:3.3239676990051025e-05 tokens_per_second_per_gpu:5324.623724974795 mfu:25.11715858519251 +Step 22135 | loss:0.48987239599227905 lr:3.32355917047656e-05 tokens_per_second_per_gpu:4712.996399809373 mfu:22.232008138004858 +Step 22136 | loss:0.3195187747478485 lr:3.3231506545567435e-05 tokens_per_second_per_gpu:3654.3724523620017 mfu:17.23829835802524 +Step 22137 | loss:0.3585014343261719 lr:3.322742151248725e-05 tokens_per_second_per_gpu:4751.376606939913 mfu:22.413054123378192 +Step 22138 | loss:0.412628173828125 lr:3.322333660555576e-05 tokens_per_second_per_gpu:4738.925579230293 mfu:22.35432050130753 +Step 22139 | loss:0.4769402742385864 lr:3.321925182480372e-05 tokens_per_second_per_gpu:5318.084283050982 mfu:25.08631092189404 +Step 22140 | loss:0.4350135326385498 lr:3.321516717026181e-05 tokens_per_second_per_gpu:5319.169312243643 mfu:25.09142918972089 +Step 22141 | loss:0.31896358728408813 lr:3.321108264196078e-05 tokens_per_second_per_gpu:4616.797864217449 mfu:21.77822323245511 +Step 22142 | loss:0.2988782823085785 lr:3.320699823993132e-05 tokens_per_second_per_gpu:5345.5355529023645 mfu:25.215803245452612 +Step 22143 | loss:0.3684658110141754 lr:3.320291396420418e-05 tokens_per_second_per_gpu:4122.703723078318 mfu:19.44749686754849 +Step 22144 | loss:0.43692120909690857 lr:3.319882981481007e-05 tokens_per_second_per_gpu:4422.144926927162 mfu:20.860012116041087 +Step 22145 | loss:0.3724208176136017 lr:3.31947457917797e-05 tokens_per_second_per_gpu:4502.331514419004 mfu:21.238265930482697 +Step 22146 | loss:0.5010953545570374 lr:3.31906618951438e-05 tokens_per_second_per_gpu:4118.938939371574 mfu:19.42973774046544 +Step 22147 | loss:0.41688135266304016 lr:3.3186578124933075e-05 tokens_per_second_per_gpu:5391.930333176654 mfu:25.4346553397722 +Step 22148 | loss:0.3240770995616913 lr:3.318249448117824e-05 tokens_per_second_per_gpu:5371.725473031564 mfu:25.339345567163196 +Step 22149 | loss:0.4223877787590027 lr:3.3178410963910004e-05 tokens_per_second_per_gpu:5324.018594968474 mfu:25.114304083706866 +Step 22150 | loss:0.4746007025241852 lr:3.3174327573159086e-05 tokens_per_second_per_gpu:5323.286308108224 mfu:25.110849761646048 +Step 22151 | loss:0.40698161721229553 lr:3.3170244308956186e-05 tokens_per_second_per_gpu:4046.8654348343825 mfu:19.089754722555607 +Step 22152 | loss:0.4822007715702057 lr:3.3166161171332045e-05 tokens_per_second_per_gpu:3488.751009300899 mfu:16.45703375317397 +Step 22153 | loss:0.4963654577732086 lr:3.316207816031737e-05 tokens_per_second_per_gpu:5356.521156121896 mfu:25.26762421017633 +Step 22154 | loss:0.48756399750709534 lr:3.3157995275942826e-05 tokens_per_second_per_gpu:4544.206624708782 mfu:21.43579797035013 +Step 22155 | loss:0.3807719647884369 lr:3.315391251823915e-05 tokens_per_second_per_gpu:5350.43110470149 mfu:25.238896398556353 +Step 22156 | loss:0.3653145134449005 lr:3.314982988723705e-05 tokens_per_second_per_gpu:5328.150368820121 mfu:25.133794365918398 +Step 22157 | loss:0.34037110209465027 lr:3.314574738296724e-05 tokens_per_second_per_gpu:4997.06308781831 mfu:23.571999172117724 +Step 22158 | loss:0.4555585980415344 lr:3.31416650054604e-05 tokens_per_second_per_gpu:4375.101477790335 mfu:20.638100140022594 +Step 22159 | loss:0.5038754940032959 lr:3.3137582754747254e-05 tokens_per_second_per_gpu:4870.343358066062 mfu:22.97424058626103 +Step 22160 | loss:0.4861050546169281 lr:3.313350063085851e-05 tokens_per_second_per_gpu:5315.362801387969 mfu:25.073473228556953 +Step 22161 | loss:0.4530613422393799 lr:3.312941863382485e-05 tokens_per_second_per_gpu:5224.07216240577 mfu:24.642839708688786 +Step 22162 | loss:0.4602833688259125 lr:3.312533676367698e-05 tokens_per_second_per_gpu:5329.3110980353995 mfu:25.13926972366769 +Step 22163 | loss:0.43859249353408813 lr:3.312125502044562e-05 tokens_per_second_per_gpu:4581.182873519436 mfu:21.610221244788573 +Step 22164 | loss:0.3158678710460663 lr:3.311717340416145e-05 tokens_per_second_per_gpu:5084.103329969129 mfu:23.982582844939795 +Step 22165 | loss:0.3593897819519043 lr:3.3113091914855155e-05 tokens_per_second_per_gpu:3952.152483417788 mfu:18.642977571027686 +Step 22166 | loss:0.449613481760025 lr:3.310901055255747e-05 tokens_per_second_per_gpu:5362.424336748471 mfu:25.29547052037836 +Step 22167 | loss:0.30252543091773987 lr:3.310492931729906e-05 tokens_per_second_per_gpu:4504.9672153510055 mfu:21.250698981475903 +Step 22168 | loss:0.3155488967895508 lr:3.310084820911066e-05 tokens_per_second_per_gpu:3849.732413115899 mfu:18.159844624747752 +Step 22169 | loss:0.3581199049949646 lr:3.3096767228022904e-05 tokens_per_second_per_gpu:4585.798255053449 mfu:21.631992786949805 +Step 22170 | loss:0.41131067276000977 lr:3.309268637406654e-05 tokens_per_second_per_gpu:4354.24233577077 mfu:20.539704008179694 +Step 22171 | loss:0.5201197862625122 lr:3.308860564727222e-05 tokens_per_second_per_gpu:5338.054036572002 mfu:25.18051165644398 +Step 22172 | loss:0.3426504135131836 lr:3.308452504767067e-05 tokens_per_second_per_gpu:5347.074901942516 mfu:25.223064617515032 +Step 22173 | loss:0.4011405110359192 lr:3.3080444575292556e-05 tokens_per_second_per_gpu:4847.063074186699 mfu:22.86442351517606 +Step 22174 | loss:0.436409056186676 lr:3.3076364230168575e-05 tokens_per_second_per_gpu:4397.917505523838 mfu:20.745727235657277 +Step 22175 | loss:0.3441565930843353 lr:3.307228401232945e-05 tokens_per_second_per_gpu:4488.673314453443 mfu:21.17383787091594 +Step 22176 | loss:0.4589908719062805 lr:3.3068203921805795e-05 tokens_per_second_per_gpu:4356.29679393136 mfu:20.549395237849904 +Step 22177 | loss:0.46570277214050293 lr:3.306412395862836e-05 tokens_per_second_per_gpu:4303.348012853923 mfu:20.299626803514027 +Step 22178 | loss:0.3308592736721039 lr:3.30600441228278e-05 tokens_per_second_per_gpu:5337.467719983462 mfu:25.17774590105991 +Step 22179 | loss:0.4871208667755127 lr:3.3055964414434815e-05 tokens_per_second_per_gpu:4906.5051775593165 mfu:23.144822058653325 +Step 22180 | loss:0.4463288187980652 lr:3.3051884833480085e-05 tokens_per_second_per_gpu:3340.8631069314893 mfu:15.759421285419448 +Step 22181 | loss:0.42846792936325073 lr:3.304780537999428e-05 tokens_per_second_per_gpu:4484.957130275118 mfu:21.15630799610038 +Step 22182 | loss:0.342411071062088 lr:3.3043726054008116e-05 tokens_per_second_per_gpu:5314.756327671784 mfu:25.070612388562626 +Step 22183 | loss:0.4608933627605438 lr:3.303964685555224e-05 tokens_per_second_per_gpu:4451.40999085337 mfu:20.998060415715795 +Step 22184 | loss:0.395782470703125 lr:3.303556778465735e-05 tokens_per_second_per_gpu:5349.67844705778 mfu:25.235345984034343 +Step 22185 | loss:0.46309033036231995 lr:3.303148884135411e-05 tokens_per_second_per_gpu:5028.231366717114 mfu:23.71902526154012 +Step 22186 | loss:0.38510486483573914 lr:3.302741002567321e-05 tokens_per_second_per_gpu:3849.589939794908 mfu:18.15917255378418 +Step 22187 | loss:0.42661890387535095 lr:3.302333133764532e-05 tokens_per_second_per_gpu:3617.96061984506 mfu:17.066537531543588 +Step 22188 | loss:0.37061741948127747 lr:3.301925277730113e-05 tokens_per_second_per_gpu:4908.033934141698 mfu:23.152033464285193 +Step 22189 | loss:0.4716242849826813 lr:3.301517434467132e-05 tokens_per_second_per_gpu:5340.139178205183 mfu:25.190347625289544 +Step 22190 | loss:0.4155172109603882 lr:3.301109603978653e-05 tokens_per_second_per_gpu:4251.319871679666 mfu:20.05420118467881 +Step 22191 | loss:0.46402886509895325 lr:3.3007017862677453e-05 tokens_per_second_per_gpu:4373.326904281441 mfu:20.629729174007682 +Step 22192 | loss:0.3863362967967987 lr:3.3002939813374774e-05 tokens_per_second_per_gpu:3888.25167186072 mfu:18.3415465403103 +Step 22193 | loss:0.4104560315608978 lr:3.299886189190915e-05 tokens_per_second_per_gpu:4969.163687181631 mfu:23.44039293918591 +Step 22194 | loss:0.4099239408969879 lr:3.2994784098311246e-05 tokens_per_second_per_gpu:5335.743875646871 mfu:25.169614233206037 +Step 22195 | loss:0.5324082970619202 lr:3.299070643261175e-05 tokens_per_second_per_gpu:3736.745320251195 mfu:17.62686523011906 +Step 22196 | loss:0.45042067766189575 lr:3.298662889484132e-05 tokens_per_second_per_gpu:4029.2090528291956 mfu:19.006466556147732 +Step 22197 | loss:0.46895021200180054 lr:3.2982551485030647e-05 tokens_per_second_per_gpu:4368.107547394099 mfu:20.605108577055976 +Step 22198 | loss:0.45105037093162537 lr:3.297847420321034e-05 tokens_per_second_per_gpu:4980.190872353384 mfu:23.49241005307287 +Step 22199 | loss:0.36833199858665466 lr:3.297439704941112e-05 tokens_per_second_per_gpu:4152.644554635689 mfu:19.588732878437238 +Step 22200 | loss:0.4707379937171936 lr:3.297032002366363e-05 tokens_per_second_per_gpu:5348.5925993453075 mfu:25.23022385510988 +Step 22201 | loss:0.38172757625579834 lr:3.2966243125998516e-05 tokens_per_second_per_gpu:5308.468410523308 mfu:25.040951210543973 +Step 22202 | loss:0.3915673494338989 lr:3.2962166356446475e-05 tokens_per_second_per_gpu:4972.948548672444 mfu:23.45824677660204 +Step 22203 | loss:0.5416062474250793 lr:3.295808971503815e-05 tokens_per_second_per_gpu:3922.0427671609605 mfu:18.500944902196903 +Step 22204 | loss:0.38626787066459656 lr:3.295401320180424e-05 tokens_per_second_per_gpu:5280.447888402481 mfu:24.908773626905976 +Step 22205 | loss:0.5426198244094849 lr:3.294993681677534e-05 tokens_per_second_per_gpu:4990.24269033274 mfu:23.539826193498484 +Step 22206 | loss:0.41868558526039124 lr:3.294586055998215e-05 tokens_per_second_per_gpu:4893.157536777562 mfu:23.081858959743442 +Step 22207 | loss:0.4475554823875427 lr:3.294178443145531e-05 tokens_per_second_per_gpu:5293.788472417337 mfu:24.971703437842393 +Step 22208 | loss:0.4375052750110626 lr:3.2937708431225486e-05 tokens_per_second_per_gpu:4819.622834649291 mfu:22.734983223490467 +Step 22209 | loss:0.3632350265979767 lr:3.2933632559323346e-05 tokens_per_second_per_gpu:3957.399945322929 mfu:18.667730744143558 +Step 22210 | loss:0.5245075225830078 lr:3.2929556815779515e-05 tokens_per_second_per_gpu:5336.796892170675 mfu:25.174581491811967 +Step 22211 | loss:0.5054966807365417 lr:3.292548120062471e-05 tokens_per_second_per_gpu:3877.44546415482 mfu:18.290571814832422 +Step 22212 | loss:0.39293670654296875 lr:3.29214057138895e-05 tokens_per_second_per_gpu:4939.048727953439 mfu:23.298335538364654 +Step 22213 | loss:0.4101392328739166 lr:3.291733035560458e-05 tokens_per_second_per_gpu:4385.001216444343 mfu:20.684798896323137 +Step 22214 | loss:0.4792547821998596 lr:3.2913255125800594e-05 tokens_per_second_per_gpu:4386.768193915365 mfu:20.693134030531393 +Step 22215 | loss:0.3774181604385376 lr:3.2909180024508206e-05 tokens_per_second_per_gpu:4505.572160251411 mfu:21.25355260978536 +Step 22216 | loss:0.4555776119232178 lr:3.2905105051758055e-05 tokens_per_second_per_gpu:3725.8980870832547 mfu:17.575696980537064 +Step 22217 | loss:0.5227242708206177 lr:3.290103020758078e-05 tokens_per_second_per_gpu:4958.818369517459 mfu:23.39159231067081 +Step 22218 | loss:0.48001590371131897 lr:3.289695549200707e-05 tokens_per_second_per_gpu:4366.9329247771575 mfu:20.59956768176058 +Step 22219 | loss:0.40247809886932373 lr:3.2892880905067515e-05 tokens_per_second_per_gpu:4382.235031574818 mfu:20.671750330333783 +Step 22220 | loss:0.49176710844039917 lr:3.288880644679279e-05 tokens_per_second_per_gpu:5349.1510042301625 mfu:25.23285794622544 +Step 22221 | loss:0.43903082609176636 lr:3.2884732117213525e-05 tokens_per_second_per_gpu:4199.994643495869 mfu:19.812091326348924 +Step 22222 | loss:0.44855692982673645 lr:3.2880657916360386e-05 tokens_per_second_per_gpu:5345.929359313175 mfu:25.21766089748271 +Step 22223 | loss:0.4221504032611847 lr:3.287658384426399e-05 tokens_per_second_per_gpu:4047.289890507452 mfu:19.091756952385083 +Step 22224 | loss:0.3410481810569763 lr:3.287250990095501e-05 tokens_per_second_per_gpu:5376.128180169962 mfu:25.360113887914316 +Step 22225 | loss:0.40300488471984863 lr:3.286843608646406e-05 tokens_per_second_per_gpu:5354.216712214168 mfu:25.2567537550849 +Step 22226 | loss:0.5367669463157654 lr:3.2864362400821804e-05 tokens_per_second_per_gpu:4026.046394737931 mfu:18.991547758326128 +Step 22227 | loss:0.44179150462150574 lr:3.286028884405885e-05 tokens_per_second_per_gpu:4692.640789924873 mfu:22.1359872531545 +Step 22228 | loss:0.5197293162345886 lr:3.285621541620584e-05 tokens_per_second_per_gpu:5353.099892064396 mfu:25.251485523889258 +Step 22229 | loss:0.4972281754016876 lr:3.2852142117293435e-05 tokens_per_second_per_gpu:4643.516154773814 mfu:21.904257967619923 +Step 22230 | loss:0.3421013653278351 lr:3.284806894735224e-05 tokens_per_second_per_gpu:3072.8159053391714 mfu:14.494996901939999 +Step 22231 | loss:0.37494096159935 lr:3.284399590641293e-05 tokens_per_second_per_gpu:5369.099700374749 mfu:25.326959349537987 +Step 22232 | loss:0.3858623802661896 lr:3.2839922994506104e-05 tokens_per_second_per_gpu:3645.4790004878396 mfu:17.19634642815546 +Step 22233 | loss:0.33211714029312134 lr:3.2835850211662426e-05 tokens_per_second_per_gpu:4679.251763117565 mfu:22.072828929300893 +Step 22234 | loss:0.4917907416820526 lr:3.283177755791248e-05 tokens_per_second_per_gpu:3943.653408366596 mfu:18.60288598391948 +Step 22235 | loss:0.4141228497028351 lr:3.2827705033286936e-05 tokens_per_second_per_gpu:3489.2041332159947 mfu:16.459171215992388 +Step 22236 | loss:0.45897701382637024 lr:3.282363263781642e-05 tokens_per_second_per_gpu:4093.7720804099176 mfu:19.3110213728344 +Step 22237 | loss:0.37883368134498596 lr:3.2819560371531545e-05 tokens_per_second_per_gpu:3736.5810347736124 mfu:17.626090267491307 +Step 22238 | loss:0.42155706882476807 lr:3.281548823446295e-05 tokens_per_second_per_gpu:4951.823731723971 mfu:23.358597410790658 +Step 22239 | loss:0.4428439736366272 lr:3.281141622664126e-05 tokens_per_second_per_gpu:4000.788552086618 mfu:18.87240220510755 +Step 22240 | loss:0.3764267563819885 lr:3.280734434809713e-05 tokens_per_second_per_gpu:5324.18384691185 mfu:25.115083605318663 +Step 22241 | loss:0.4431266486644745 lr:3.2803272598861125e-05 tokens_per_second_per_gpu:4241.629379151055 mfu:20.00848947805284 +Step 22242 | loss:0.3684021830558777 lr:3.2799200978963915e-05 tokens_per_second_per_gpu:4553.814810887468 mfu:21.481121423880598 +Step 22243 | loss:0.33510926365852356 lr:3.279512948843611e-05 tokens_per_second_per_gpu:5102.676151080284 mfu:24.070194010971154 +Step 22244 | loss:0.3287124037742615 lr:3.2791058127308303e-05 tokens_per_second_per_gpu:4302.9909081897285 mfu:20.29794227988468 +Step 22245 | loss:0.3770502507686615 lr:3.278698689561117e-05 tokens_per_second_per_gpu:5342.750562241562 mfu:25.2026659693372 +Step 22246 | loss:0.35071131587028503 lr:3.2782915793375296e-05 tokens_per_second_per_gpu:5328.719775941092 mfu:25.136480356458893 +Step 22247 | loss:0.43047791719436646 lr:3.2778844820631336e-05 tokens_per_second_per_gpu:3813.9349821430033 mfu:17.990982035177076 +Step 22248 | loss:0.49654310941696167 lr:3.2774773977409854e-05 tokens_per_second_per_gpu:5215.623437679565 mfu:24.602985632654196 +Step 22249 | loss:0.524347186088562 lr:3.27707032637415e-05 tokens_per_second_per_gpu:3647.380784763679 mfu:17.205317469062454 +Step 22250 | loss:0.5245897173881531 lr:3.2766632679656885e-05 tokens_per_second_per_gpu:5299.880116490133 mfu:25.000438762275426 +Step 22251 | loss:0.40287354588508606 lr:3.276256222518662e-05 tokens_per_second_per_gpu:4435.769001412319 mfu:20.92427919989438 +Step 22252 | loss:0.46586307883262634 lr:3.275849190036133e-05 tokens_per_second_per_gpu:3586.488715583626 mfu:16.918079189481833 +Step 22253 | loss:0.4804292619228363 lr:3.2754421705211616e-05 tokens_per_second_per_gpu:3083.615497381515 mfu:14.545940420204138 +Step 22254 | loss:0.5505555272102356 lr:3.275035163976811e-05 tokens_per_second_per_gpu:4306.206721018178 mfu:20.313111817670723 +Step 22255 | loss:0.37670597434043884 lr:3.274628170406141e-05 tokens_per_second_per_gpu:5331.584099786104 mfu:25.149991860739853 +Step 22256 | loss:0.4489386975765228 lr:3.274221189812212e-05 tokens_per_second_per_gpu:4113.677671274601 mfu:19.404919441187122 +Step 22257 | loss:0.4355998635292053 lr:3.273814222198085e-05 tokens_per_second_per_gpu:5073.627350863101 mfu:23.933165864110315 +Step 22258 | loss:0.34066125750541687 lr:3.273407267566823e-05 tokens_per_second_per_gpu:3662.9250384790375 mfu:17.278642365960962 +Step 22259 | loss:0.4733089506626129 lr:3.2730003259214845e-05 tokens_per_second_per_gpu:4026.496825181739 mfu:18.993672515580926 +Step 22260 | loss:0.44622769951820374 lr:3.2725933972651296e-05 tokens_per_second_per_gpu:5295.061876885231 mfu:24.977710303982654 +Step 22261 | loss:0.3727712631225586 lr:3.272186481600822e-05 tokens_per_second_per_gpu:2974.1764738639226 mfu:14.029697874048823 +Step 22262 | loss:0.550207793712616 lr:3.271779578931621e-05 tokens_per_second_per_gpu:5302.332240837989 mfu:25.012005851199817 +Step 22263 | loss:0.40420427918434143 lr:3.271372689260584e-05 tokens_per_second_per_gpu:5047.745510378826 mfu:23.81107680664865 +Step 22264 | loss:0.3985244929790497 lr:3.270965812590775e-05 tokens_per_second_per_gpu:4798.198377605021 mfu:22.633920404223346 +Step 22265 | loss:0.48132458329200745 lr:3.270558948925252e-05 tokens_per_second_per_gpu:3877.5113091345265 mfu:18.29088241683601 +Step 22266 | loss:0.486966997385025 lr:3.270152098267075e-05 tokens_per_second_per_gpu:3953.9671297360856 mfu:18.651537567321682 +Step 22267 | loss:0.42478886246681213 lr:3.2697452606193066e-05 tokens_per_second_per_gpu:4138.690661185899 mfu:19.52290998225486 +Step 22268 | loss:0.4032335579395294 lr:3.2693384359850046e-05 tokens_per_second_per_gpu:4293.017086738866 mfu:20.25089406239155 +Step 22269 | loss:0.43078917264938354 lr:3.26893162436723e-05 tokens_per_second_per_gpu:5368.132270217634 mfu:25.32239581642627 +Step 22270 | loss:0.3298395872116089 lr:3.2685248257690395e-05 tokens_per_second_per_gpu:4190.595819442203 mfu:19.76775546968304 +Step 22271 | loss:0.4522997736930847 lr:3.2681180401934955e-05 tokens_per_second_per_gpu:5173.29374495531 mfu:24.40330924988358 +Step 22272 | loss:0.3704128563404083 lr:3.267711267643657e-05 tokens_per_second_per_gpu:5330.462503331242 mfu:25.144701098898185 +Step 22273 | loss:0.5067898035049438 lr:3.267304508122582e-05 tokens_per_second_per_gpu:4764.653160706576 mfu:22.4756818927092 +Step 22274 | loss:0.3645695745944977 lr:3.266897761633332e-05 tokens_per_second_per_gpu:5313.218256384872 mfu:25.063357043880238 +Step 22275 | loss:0.4263473451137543 lr:3.266491028178964e-05 tokens_per_second_per_gpu:5348.913520398944 mfu:25.23173769447466 +Step 22276 | loss:0.46665307879447937 lr:3.2660843077625415e-05 tokens_per_second_per_gpu:4358.32397173225 mfu:20.558957781408903 +Step 22277 | loss:0.4181797504425049 lr:3.265677600387117e-05 tokens_per_second_per_gpu:4273.028063472312 mfu:20.156602429164167 +Step 22278 | loss:0.3844740390777588 lr:3.265270906055753e-05 tokens_per_second_per_gpu:5303.8721415407945 mfu:25.0192698255491 +Step 22279 | loss:0.39338329434394836 lr:3.2648642247715076e-05 tokens_per_second_per_gpu:5329.938038461033 mfu:25.14222710862247 +Step 22280 | loss:0.4960024654865265 lr:3.2644575565374395e-05 tokens_per_second_per_gpu:3750.619187679668 mfu:17.692310629905037 +Step 22281 | loss:0.3596547544002533 lr:3.264050901356609e-05 tokens_per_second_per_gpu:5387.576918137326 mfu:25.41411953826249 +Step 22282 | loss:0.4318167567253113 lr:3.2636442592320714e-05 tokens_per_second_per_gpu:4851.756559541278 mfu:22.886563486384784 +Step 22283 | loss:0.4654456675052643 lr:3.2632376301668884e-05 tokens_per_second_per_gpu:4509.3820370209 mfu:21.27152444853904 +Step 22284 | loss:0.38694384694099426 lr:3.262831014164117e-05 tokens_per_second_per_gpu:4560.938571660373 mfu:21.514725418885305 +Step 22285 | loss:0.3903825283050537 lr:3.262424411226815e-05 tokens_per_second_per_gpu:5377.992776632975 mfu:25.368909507563306 +Step 22286 | loss:0.3913384675979614 lr:3.26201782135804e-05 tokens_per_second_per_gpu:5341.25181753489 mfu:25.195596134843612 +Step 22287 | loss:0.4530406594276428 lr:3.26161124456085e-05 tokens_per_second_per_gpu:4019.4656002755196 mfu:18.960505028047617 +Step 22288 | loss:0.35265007615089417 lr:3.2612046808383055e-05 tokens_per_second_per_gpu:4452.945721990751 mfu:21.005304721513575 +Step 22289 | loss:0.5003184080123901 lr:3.260798130193461e-05 tokens_per_second_per_gpu:4383.104742210442 mfu:20.67585290378996 +Step 22290 | loss:0.39054355025291443 lr:3.2603915926293766e-05 tokens_per_second_per_gpu:5100.285296418325 mfu:24.058915941608824 +Step 22291 | loss:0.3878658711910248 lr:3.2599850681491094e-05 tokens_per_second_per_gpu:5027.777164773987 mfu:23.716882713479553 +Step 22292 | loss:0.41984879970550537 lr:3.259578556755717e-05 tokens_per_second_per_gpu:3503.4015709816053 mfu:16.526142952265438 +Step 22293 | loss:0.42837411165237427 lr:3.259172058452254e-05 tokens_per_second_per_gpu:5284.5504489281675 mfu:24.928126104874647 +Step 22294 | loss:0.5602695345878601 lr:3.258765573241781e-05 tokens_per_second_per_gpu:4264.637702223278 mfu:20.11702366360923 +Step 22295 | loss:0.5212392807006836 lr:3.2583591011273554e-05 tokens_per_second_per_gpu:5090.273526286422 mfu:24.01168871371251 +Step 22296 | loss:0.4529585838317871 lr:3.257952642112031e-05 tokens_per_second_per_gpu:4226.7523080774445 mfu:19.938311795508554 +Step 22297 | loss:0.375028133392334 lr:3.257546196198869e-05 tokens_per_second_per_gpu:4242.808307315613 mfu:20.014050683350945 +Step 22298 | loss:0.3661119043827057 lr:3.257139763390925e-05 tokens_per_second_per_gpu:5345.916233176661 mfu:25.21759897925722 +Step 22299 | loss:0.43550053238868713 lr:3.256733343691254e-05 tokens_per_second_per_gpu:5352.572266217528 mfu:25.248996622747327 +Step 22300 | loss:0.44529855251312256 lr:3.256326937102914e-05 tokens_per_second_per_gpu:4402.910465306544 mfu:20.769279878839082 +Step 22301 | loss:0.3736383318901062 lr:3.255920543628962e-05 tokens_per_second_per_gpu:4011.1131510783753 mfu:18.921105100109905 +Step 22302 | loss:0.43384885787963867 lr:3.2555141632724526e-05 tokens_per_second_per_gpu:5380.438581193694 mfu:25.380446784973504 +Step 22303 | loss:0.38547247648239136 lr:3.255107796036445e-05 tokens_per_second_per_gpu:5053.587641256999 mfu:23.8386351347713 +Step 22304 | loss:0.3730381429195404 lr:3.254701441923995e-05 tokens_per_second_per_gpu:5333.891569801487 mfu:25.1608765904974 +Step 22305 | loss:0.4949183464050293 lr:3.2542951009381584e-05 tokens_per_second_per_gpu:4468.758105283551 mfu:21.0798944313765 +Step 22306 | loss:0.4325534999370575 lr:3.25388877308199e-05 tokens_per_second_per_gpu:4304.428776108561 mfu:20.304724948183377 +Step 22307 | loss:0.40616002678871155 lr:3.2534824583585465e-05 tokens_per_second_per_gpu:5354.83652927222 mfu:25.259677537899424 +Step 22308 | loss:0.443931519985199 lr:3.2530761567708854e-05 tokens_per_second_per_gpu:4235.511340891789 mfu:19.979629647738918 +Step 22309 | loss:0.4557880461215973 lr:3.25266986832206e-05 tokens_per_second_per_gpu:4924.142306606612 mfu:23.22801940557393 +Step 22310 | loss:0.35699662566185 lr:3.252263593015129e-05 tokens_per_second_per_gpu:3715.109444839773 mfu:17.524805114342385 +Step 22311 | loss:0.560396134853363 lr:3.2518573308531456e-05 tokens_per_second_per_gpu:4823.034677702397 mfu:22.751077469292422 +Step 22312 | loss:0.4940631091594696 lr:3.251451081839168e-05 tokens_per_second_per_gpu:5339.995949719044 mfu:25.189671992083127 +Step 22313 | loss:0.3464412987232208 lr:3.2510448459762475e-05 tokens_per_second_per_gpu:5344.510811605764 mfu:25.21096936591765 +Step 22314 | loss:0.48168396949768066 lr:3.2506386232674434e-05 tokens_per_second_per_gpu:5337.987943775604 mfu:25.18019988544678 +Step 22315 | loss:0.47572481632232666 lr:3.250232413715809e-05 tokens_per_second_per_gpu:4055.7292398205454 mfu:19.131566802996417 +Step 22316 | loss:0.367221862077713 lr:3.249826217324399e-05 tokens_per_second_per_gpu:4831.844261295742 mfu:22.792633778172185 +Step 22317 | loss:0.3054603338241577 lr:3.249420034096271e-05 tokens_per_second_per_gpu:4411.0571709159085 mfu:20.80770927916976 +Step 22318 | loss:0.4065886437892914 lr:3.2490138640344765e-05 tokens_per_second_per_gpu:4170.897476223217 mfu:19.67483502383517 +Step 22319 | loss:0.4617297649383545 lr:3.248607707142073e-05 tokens_per_second_per_gpu:4605.002248517261 mfu:21.722581300657758 +Step 22320 | loss:0.4050874412059784 lr:3.2482015634221163e-05 tokens_per_second_per_gpu:5326.627111100864 mfu:25.126608899362093 +Step 22321 | loss:0.5351541638374329 lr:3.2477954328776574e-05 tokens_per_second_per_gpu:4364.237134631395 mfu:20.586851179693788 +Step 22322 | loss:0.45678025484085083 lr:3.247389315511752e-05 tokens_per_second_per_gpu:4781.522572774973 mfu:22.555257787654316 +Step 22323 | loss:0.39093488454818726 lr:3.2469832113274556e-05 tokens_per_second_per_gpu:3920.91277786406 mfu:18.495614549887364 +Step 22324 | loss:0.5686080455780029 lr:3.246577120327823e-05 tokens_per_second_per_gpu:5123.457970177589 mfu:24.168225397397105 +Step 22325 | loss:0.4522921144962311 lr:3.246171042515906e-05 tokens_per_second_per_gpu:5122.669409460677 mfu:24.1645056219531 +Step 22326 | loss:0.4300735592842102 lr:3.245764977894762e-05 tokens_per_second_per_gpu:4456.360967903233 mfu:21.021415019183788 +Step 22327 | loss:0.40865835547447205 lr:3.245358926467444e-05 tokens_per_second_per_gpu:5324.675722238102 mfu:25.117403865155175 +Step 22328 | loss:0.3827746510505676 lr:3.244952888237004e-05 tokens_per_second_per_gpu:5366.733337337271 mfu:25.315796811347038 +Step 22329 | loss:0.3622939884662628 lr:3.244546863206497e-05 tokens_per_second_per_gpu:4212.896253156374 mfu:19.872950420359114 +Step 22330 | loss:0.34043750166893005 lr:3.2441408513789775e-05 tokens_per_second_per_gpu:4384.340963176856 mfu:20.681684369009577 +Step 22331 | loss:0.4518427550792694 lr:3.243734852757499e-05 tokens_per_second_per_gpu:4353.290188868268 mfu:20.53521256878758 +Step 22332 | loss:0.3719997704029083 lr:3.243328867345114e-05 tokens_per_second_per_gpu:4667.3760790827555 mfu:22.016809301508065 +Step 22333 | loss:0.3318701982498169 lr:3.242922895144877e-05 tokens_per_second_per_gpu:4421.933863753541 mfu:20.85901649504003 +Step 22334 | loss:0.4817274510860443 lr:3.242516936159843e-05 tokens_per_second_per_gpu:4859.562011071304 mfu:22.923383132999277 +Step 22335 | loss:0.4122142791748047 lr:3.242110990393062e-05 tokens_per_second_per_gpu:5331.6706040614445 mfu:25.15039991616579 +Step 22336 | loss:0.33108240365982056 lr:3.2417050578475874e-05 tokens_per_second_per_gpu:5323.920208235032 mfu:25.113839976698713 +Step 22337 | loss:0.43440908193588257 lr:3.241299138526475e-05 tokens_per_second_per_gpu:5314.416333670639 mfu:25.069008578850266 +Step 22338 | loss:0.4017617404460907 lr:3.2408932324327745e-05 tokens_per_second_per_gpu:4502.90726881889 mfu:21.240981862220792 +Step 22339 | loss:0.3970381021499634 lr:3.240487339569542e-05 tokens_per_second_per_gpu:4527.420932655814 mfu:21.356616997001897 +Step 22340 | loss:0.37066832184791565 lr:3.2400814599398286e-05 tokens_per_second_per_gpu:4447.623307150351 mfu:20.980197982613795 +Step 22341 | loss:0.3681586980819702 lr:3.239675593546688e-05 tokens_per_second_per_gpu:4327.044038848997 mfu:20.411405001091723 +Step 22342 | loss:0.46439892053604126 lr:3.2392697403931694e-05 tokens_per_second_per_gpu:4546.912386899363 mfu:21.448561512253736 +Step 22343 | loss:0.4430442750453949 lr:3.2388639004823294e-05 tokens_per_second_per_gpu:4201.449769529039 mfu:19.81895540411749 +Step 22344 | loss:0.573474645614624 lr:3.2384580738172185e-05 tokens_per_second_per_gpu:3978.5064295606385 mfu:18.76729363143038 +Step 22345 | loss:0.37487950921058655 lr:3.238052260400888e-05 tokens_per_second_per_gpu:5370.417541468424 mfu:25.333175830823723 +Step 22346 | loss:0.41381198167800903 lr:3.237646460236393e-05 tokens_per_second_per_gpu:4645.3793269598555 mfu:21.913046868711433 +Step 22347 | loss:0.3758258819580078 lr:3.237240673326783e-05 tokens_per_second_per_gpu:5030.678230567637 mfu:23.73056753580106 +Step 22348 | loss:0.49001121520996094 lr:3.2368348996751094e-05 tokens_per_second_per_gpu:5343.291799371804 mfu:25.205219077225127 +Step 22349 | loss:0.4157741367816925 lr:3.236429139284429e-05 tokens_per_second_per_gpu:5324.930728958141 mfu:25.11860677536223 +Step 22350 | loss:0.40780001878738403 lr:3.2360233921577883e-05 tokens_per_second_per_gpu:4496.736936939861 mfu:21.21187535398024 +Step 22351 | loss:0.44926875829696655 lr:3.2356176582982415e-05 tokens_per_second_per_gpu:3958.9743049281674 mfu:18.675157267014917 +Step 22352 | loss:0.3357422351837158 lr:3.235211937708837e-05 tokens_per_second_per_gpu:5344.033360714724 mfu:25.2087171486031 +Step 22353 | loss:0.3781373202800751 lr:3.234806230392631e-05 tokens_per_second_per_gpu:3525.4574747813567 mfu:16.630184413614096 +Step 22354 | loss:0.43343499302864075 lr:3.23440053635267e-05 tokens_per_second_per_gpu:3254.8022696687626 mfu:15.353457632558216 +Step 22355 | loss:0.5116913318634033 lr:3.23399485559201e-05 tokens_per_second_per_gpu:4086.3269869488845 mfu:19.275901596714775 +Step 22356 | loss:0.41429731249809265 lr:3.233589188113701e-05 tokens_per_second_per_gpu:5230.530984193943 mfu:24.673307073052232 +Step 22357 | loss:0.49708637595176697 lr:3.233183533920792e-05 tokens_per_second_per_gpu:4513.256610577443 mfu:21.289801473075673 +Step 22358 | loss:0.37001803517341614 lr:3.232777893016334e-05 tokens_per_second_per_gpu:4515.984338135153 mfu:21.302668629363705 +Step 22359 | loss:0.38492751121520996 lr:3.2323722654033804e-05 tokens_per_second_per_gpu:5344.2368762181895 mfu:25.20967716595627 +Step 22360 | loss:0.3904680609703064 lr:3.23196665108498e-05 tokens_per_second_per_gpu:4620.412087658502 mfu:21.795272141076662 +Step 22361 | loss:0.411190927028656 lr:3.231561050064183e-05 tokens_per_second_per_gpu:4006.389977389316 mfu:18.898825083962084 +Step 22362 | loss:0.31461718678474426 lr:3.2311554623440425e-05 tokens_per_second_per_gpu:5323.571327486808 mfu:25.112194246684144 +Step 22363 | loss:0.3945203721523285 lr:3.230749887927608e-05 tokens_per_second_per_gpu:4814.977014038439 mfu:22.71306809501035 +Step 22364 | loss:0.5244238376617432 lr:3.230344326817929e-05 tokens_per_second_per_gpu:5312.175406959614 mfu:25.05843774521284 +Step 22365 | loss:0.43498292565345764 lr:3.2299387790180545e-05 tokens_per_second_per_gpu:5313.106074718029 mfu:25.062827863817574 +Step 22366 | loss:0.39048823714256287 lr:3.229533244531038e-05 tokens_per_second_per_gpu:4683.627944409572 mfu:22.09347212310881 +Step 22367 | loss:0.368806928396225 lr:3.229127723359927e-05 tokens_per_second_per_gpu:5246.967281974963 mfu:24.750839894006877 +Step 22368 | loss:0.3541383445262909 lr:3.228722215507772e-05 tokens_per_second_per_gpu:5269.742276363862 mfu:24.85827342835405 +Step 22369 | loss:0.4794928729534149 lr:3.228316720977623e-05 tokens_per_second_per_gpu:4551.303670192378 mfu:21.469275944777078 +Step 22370 | loss:0.5685029029846191 lr:3.2279112397725323e-05 tokens_per_second_per_gpu:3462.3965091896653 mfu:16.33271507960789 +Step 22371 | loss:0.4560500681400299 lr:3.227505771895546e-05 tokens_per_second_per_gpu:4248.708560300271 mfu:20.04188318336654 +Step 22372 | loss:0.4763239324092865 lr:3.227100317349714e-05 tokens_per_second_per_gpu:4921.977591568674 mfu:23.217808075401514 +Step 22373 | loss:0.35689637064933777 lr:3.226694876138088e-05 tokens_per_second_per_gpu:4613.9627736595985 mfu:21.764849626577398 +Step 22374 | loss:0.3998577892780304 lr:3.226289448263715e-05 tokens_per_second_per_gpu:4040.912345972171 mfu:19.06167298669086 +Step 22375 | loss:0.5017620921134949 lr:3.225884033729646e-05 tokens_per_second_per_gpu:5234.034308242915 mfu:24.689832850319952 +Step 22376 | loss:0.3816191554069519 lr:3.2254786325389296e-05 tokens_per_second_per_gpu:5356.584277016361 mfu:25.26792196220505 +Step 22377 | loss:0.4549448490142822 lr:3.225073244694614e-05 tokens_per_second_per_gpu:3949.887484713751 mfu:18.632293185692113 +Step 22378 | loss:0.3737412691116333 lr:3.2246678701997505e-05 tokens_per_second_per_gpu:4174.823175318838 mfu:19.69335321626254 +Step 22379 | loss:0.4409545958042145 lr:3.224262509057387e-05 tokens_per_second_per_gpu:5385.772040673291 mfu:25.405605623320785 +Step 22380 | loss:0.377939909696579 lr:3.22385716127057e-05 tokens_per_second_per_gpu:5347.081134676635 mfu:25.22309401838529 +Step 22381 | loss:0.5414661765098572 lr:3.2234518268423495e-05 tokens_per_second_per_gpu:3892.515976097056 mfu:18.361661990958428 +Step 22382 | loss:0.5206552147865295 lr:3.2230465057757754e-05 tokens_per_second_per_gpu:5283.8246309789165 mfu:24.924702297770715 +Step 22383 | loss:0.3714611232280731 lr:3.222641198073896e-05 tokens_per_second_per_gpu:2988.13198715083 mfu:14.095528411279258 +Step 22384 | loss:0.3249538838863373 lr:3.222235903739757e-05 tokens_per_second_per_gpu:5373.494734601121 mfu:25.347691475853946 +Step 22385 | loss:0.3317814767360687 lr:3.221830622776411e-05 tokens_per_second_per_gpu:4018.7304737961476 mfu:18.957037311018908 +Step 22386 | loss:0.3944362699985504 lr:3.2214253551869035e-05 tokens_per_second_per_gpu:4201.670464369656 mfu:19.819996459335098 +Step 22387 | loss:0.40154048800468445 lr:3.221020100974281e-05 tokens_per_second_per_gpu:5022.81169998181 mfu:23.6934597688592 +Step 22388 | loss:0.3963165283203125 lr:3.2206148601415945e-05 tokens_per_second_per_gpu:5352.029544750095 mfu:25.246436512987614 +Step 22389 | loss:0.4491857886314392 lr:3.220209632691891e-05 tokens_per_second_per_gpu:5365.094438378346 mfu:25.308065845332788 +Step 22390 | loss:0.398325651884079 lr:3.219804418628216e-05 tokens_per_second_per_gpu:4096.284435072994 mfu:19.322872578432033 +Step 22391 | loss:0.4353055953979492 lr:3.2193992179536204e-05 tokens_per_second_per_gpu:5375.788491073458 mfu:25.35851151648917 +Step 22392 | loss:0.4427053928375244 lr:3.218994030671152e-05 tokens_per_second_per_gpu:5309.496044503194 mfu:25.04579873535953 +Step 22393 | loss:0.4810708463191986 lr:3.2185888567838555e-05 tokens_per_second_per_gpu:3851.969586831964 mfu:18.170397754867626 +Step 22394 | loss:0.5367053151130676 lr:3.218183696294779e-05 tokens_per_second_per_gpu:5305.821007081169 mfu:25.0284629568138 +Step 22395 | loss:0.3845284581184387 lr:3.2177785492069704e-05 tokens_per_second_per_gpu:4570.590861651635 mfu:21.56025691762458 +Step 22396 | loss:0.4283895790576935 lr:3.217373415523477e-05 tokens_per_second_per_gpu:4856.0353639627765 mfu:22.906747336880308 +Step 22397 | loss:0.3998107612133026 lr:3.2169682952473454e-05 tokens_per_second_per_gpu:5353.354877107669 mfu:25.252688331843313 +Step 22398 | loss:0.49807390570640564 lr:3.2165631883816235e-05 tokens_per_second_per_gpu:5310.892932074458 mfu:25.05238809236678 +Step 22399 | loss:0.28666722774505615 lr:3.216158094929358e-05 tokens_per_second_per_gpu:5246.529964724767 mfu:24.74877699392368 +Step 22400 | loss:0.3368055820465088 lr:3.2157530148935945e-05 tokens_per_second_per_gpu:4839.085485630295 mfu:22.82679186883854 +Step 22401 | loss:0.4050050675868988 lr:3.2153479482773786e-05 tokens_per_second_per_gpu:4712.012792119321 mfu:22.227368292710104 +Step 22402 | loss:0.32968640327453613 lr:3.21494289508376e-05 tokens_per_second_per_gpu:3848.7268680027782 mfu:18.155101296885473 +Step 22403 | loss:0.4463241994380951 lr:3.214537855315783e-05 tokens_per_second_per_gpu:5330.931325812857 mfu:25.146912614533324 +Step 22404 | loss:0.46085116267204285 lr:3.214132828976495e-05 tokens_per_second_per_gpu:4613.977048526397 mfu:21.76491696355963 +Step 22405 | loss:0.5521950721740723 lr:3.2137278160689416e-05 tokens_per_second_per_gpu:5340.89711705433 mfu:25.193922952158673 +Step 22406 | loss:0.390903502702713 lr:3.21332281659617e-05 tokens_per_second_per_gpu:3803.7566927815583 mfu:17.9429693076639 +Step 22407 | loss:0.5055555701255798 lr:3.212917830561226e-05 tokens_per_second_per_gpu:5061.661944273674 mfu:23.876722999719107 +Step 22408 | loss:0.48256054520606995 lr:3.212512857967153e-05 tokens_per_second_per_gpu:3816.3368542321123 mfu:18.00231207562304 +Step 22409 | loss:0.3154711127281189 lr:3.212107898817e-05 tokens_per_second_per_gpu:4331.524895871164 mfu:20.43254196817832 +Step 22410 | loss:0.427223801612854 lr:3.211702953113811e-05 tokens_per_second_per_gpu:4921.63214131591 mfu:23.216178527619334 +Step 22411 | loss:0.5615839958190918 lr:3.2112980208606327e-05 tokens_per_second_per_gpu:4933.390903818234 mfu:23.271646616594555 +Step 22412 | loss:0.4042215347290039 lr:3.2108931020605105e-05 tokens_per_second_per_gpu:4377.653240422895 mfu:20.65013723973598 +Step 22413 | loss:0.41752946376800537 lr:3.2104881967164885e-05 tokens_per_second_per_gpu:5151.378587057248 mfu:24.29993171096692 +Step 22414 | loss:0.4620952308177948 lr:3.2100833048316155e-05 tokens_per_second_per_gpu:4182.40892332322 mfu:19.72913648386128 +Step 22415 | loss:0.3909865915775299 lr:3.209678426408933e-05 tokens_per_second_per_gpu:5306.605514686427 mfu:25.032163613038602 +Step 22416 | loss:0.34406447410583496 lr:3.2092735614514876e-05 tokens_per_second_per_gpu:4634.2185543673295 mfu:21.86039960016833 +Step 22417 | loss:0.34136202931404114 lr:3.2088687099623225e-05 tokens_per_second_per_gpu:3839.5245729338167 mfu:18.111692500971742 +Step 22418 | loss:0.46915683150291443 lr:3.208463871944487e-05 tokens_per_second_per_gpu:5289.825514218069 mfu:24.953009487866055 +Step 22419 | loss:0.4741206467151642 lr:3.208059047401022e-05 tokens_per_second_per_gpu:4566.740228947573 mfu:21.542092826173835 +Step 22420 | loss:0.34878990054130554 lr:3.207654236334973e-05 tokens_per_second_per_gpu:5374.628451167495 mfu:25.353039410329526 +Step 22421 | loss:0.45266827940940857 lr:3.207249438749387e-05 tokens_per_second_per_gpu:4824.178489968716 mfu:22.75647302690715 +Step 22422 | loss:0.4936707317829132 lr:3.2068446546473056e-05 tokens_per_second_per_gpu:4791.498775287829 mfu:22.602317237023122 +Step 22423 | loss:0.4194950461387634 lr:3.206439884031774e-05 tokens_per_second_per_gpu:3488.866396858483 mfu:16.457578055970238 +Step 22424 | loss:0.3368375599384308 lr:3.206035126905838e-05 tokens_per_second_per_gpu:4845.565095570791 mfu:22.857357294463313 +Step 22425 | loss:0.4078682065010071 lr:3.205630383272539e-05 tokens_per_second_per_gpu:5311.084497075642 mfu:25.053291737161718 +Step 22426 | loss:0.5019633173942566 lr:3.2052256531349234e-05 tokens_per_second_per_gpu:5345.24378687368 mfu:25.214426935314886 +Step 22427 | loss:0.34064099192619324 lr:3.2048209364960355e-05 tokens_per_second_per_gpu:3537.2589295001576 mfu:16.68585388905848 +Step 22428 | loss:0.4045208990573883 lr:3.204416233358919e-05 tokens_per_second_per_gpu:3886.097190197198 mfu:18.331383482710994 +Step 22429 | loss:0.406289279460907 lr:3.204011543726617e-05 tokens_per_second_per_gpu:5320.738713139049 mfu:25.098832321511924 +Step 22430 | loss:0.4512290060520172 lr:3.2036068676021714e-05 tokens_per_second_per_gpu:4937.16164943689 mfu:23.2894338670333 +Step 22431 | loss:0.4476148784160614 lr:3.203202204988629e-05 tokens_per_second_per_gpu:5357.490371990416 mfu:25.272196166793414 +Step 22432 | loss:0.5317384004592896 lr:3.202797555889032e-05 tokens_per_second_per_gpu:3554.458595568575 mfu:16.76698764846884 +Step 22433 | loss:0.38071873784065247 lr:3.2023929203064226e-05 tokens_per_second_per_gpu:5345.808812228419 mfu:25.217092256316462 +Step 22434 | loss:0.37836986780166626 lr:3.2019882982438474e-05 tokens_per_second_per_gpu:4684.396740347085 mfu:22.097098664716174 +Step 22435 | loss:0.4269319176673889 lr:3.2015836897043485e-05 tokens_per_second_per_gpu:5342.85197632012 mfu:25.20314435684879 +Step 22436 | loss:0.40963223576545715 lr:3.201179094690967e-05 tokens_per_second_per_gpu:5051.767050790851 mfu:23.830047098917714 +Step 22437 | loss:0.49728822708129883 lr:3.200774513206746e-05 tokens_per_second_per_gpu:4733.808704093233 mfu:22.3301833282571 +Step 22438 | loss:0.5185143351554871 lr:3.20036994525473e-05 tokens_per_second_per_gpu:4498.833667494674 mfu:21.22176598974641 +Step 22439 | loss:0.38274112343788147 lr:3.199965390837961e-05 tokens_per_second_per_gpu:4295.669656368821 mfu:20.263406685911093 +Step 22440 | loss:0.42845723032951355 lr:3.199560849959482e-05 tokens_per_second_per_gpu:4385.961902291087 mfu:20.68933061537214 +Step 22441 | loss:0.42849627137184143 lr:3.199156322622336e-05 tokens_per_second_per_gpu:5319.958811766138 mfu:25.09515339367101 +Step 22442 | loss:0.4389513432979584 lr:3.1987518088295644e-05 tokens_per_second_per_gpu:3690.6318979494868 mfu:17.409340349360757 +Step 22443 | loss:0.42362990975379944 lr:3.198347308584213e-05 tokens_per_second_per_gpu:4564.366837596227 mfu:21.530897134227214 +Step 22444 | loss:0.4403459429740906 lr:3.1979428218893186e-05 tokens_per_second_per_gpu:4513.500156281256 mfu:21.29095031971374 +Step 22445 | loss:0.44524985551834106 lr:3.197538348747927e-05 tokens_per_second_per_gpu:4196.391874668257 mfu:19.795096451093325 +Step 22446 | loss:0.4679624140262604 lr:3.1971338891630784e-05 tokens_per_second_per_gpu:3793.5783908918984 mfu:17.894956521053 +Step 22447 | loss:0.3867567777633667 lr:3.196729443137818e-05 tokens_per_second_per_gpu:4615.700589214015 mfu:21.77304719904949 +Step 22448 | loss:0.5056502223014832 lr:3.196325010675184e-05 tokens_per_second_per_gpu:3625.715789032201 mfu:17.103119987767634 +Step 22449 | loss:0.4047545790672302 lr:3.19592059177822e-05 tokens_per_second_per_gpu:5323.044316648772 mfu:25.10970824664381 +Step 22450 | loss:0.3388862609863281 lr:3.19551618644997e-05 tokens_per_second_per_gpu:3915.403624289331 mfu:18.469626932516846 +Step 22451 | loss:0.47875380516052246 lr:3.1951117946934696e-05 tokens_per_second_per_gpu:5341.312394547104 mfu:25.195881886945735 +Step 22452 | loss:0.4782552421092987 lr:3.194707416511764e-05 tokens_per_second_per_gpu:3949.728796727645 mfu:18.63154462738669 +Step 22453 | loss:0.41539841890335083 lr:3.1943030519078945e-05 tokens_per_second_per_gpu:4183.408913711393 mfu:19.733853609138247 +Step 22454 | loss:0.3462238907814026 lr:3.193898700884903e-05 tokens_per_second_per_gpu:5368.185409018747 mfu:25.322646481217532 +Step 22455 | loss:0.48133158683776855 lr:3.1934943634458295e-05 tokens_per_second_per_gpu:4542.136089676145 mfu:21.42603090333148 +Step 22456 | loss:0.4824928045272827 lr:3.1930900395937145e-05 tokens_per_second_per_gpu:5325.984694271243 mfu:25.12357850956916 +Step 22457 | loss:0.5380854606628418 lr:3.192685729331603e-05 tokens_per_second_per_gpu:4803.531561513557 mfu:22.65907794265491 +Step 22458 | loss:0.3107081651687622 lr:3.19228143266253e-05 tokens_per_second_per_gpu:3660.157942709374 mfu:17.265589503100237 +Step 22459 | loss:0.4542604386806488 lr:3.1918771495895396e-05 tokens_per_second_per_gpu:5115.967904096732 mfu:24.13289347775665 +Step 22460 | loss:0.455696702003479 lr:3.191472880115671e-05 tokens_per_second_per_gpu:3282.0644732823002 mfu:15.482058098414848 +Step 22461 | loss:0.48848986625671387 lr:3.191068624243967e-05 tokens_per_second_per_gpu:5322.228922783521 mfu:25.105861894660883 +Step 22462 | loss:0.4959755539894104 lr:3.190664381977465e-05 tokens_per_second_per_gpu:5229.42900282457 mfu:24.668108838915487 +Step 22463 | loss:0.5590645670890808 lr:3.1902601533192086e-05 tokens_per_second_per_gpu:4123.887084444919 mfu:19.453078985017015 +Step 22464 | loss:0.2732093632221222 lr:3.189855938272237e-05 tokens_per_second_per_gpu:5323.133469143514 mfu:25.110128794172493 +Step 22465 | loss:0.4227931797504425 lr:3.189451736839589e-05 tokens_per_second_per_gpu:5358.33691652314 mfu:25.276189461789663 +Step 22466 | loss:0.518775463104248 lr:3.189047549024305e-05 tokens_per_second_per_gpu:4904.198715430631 mfu:23.133942083269297 +Step 22467 | loss:0.41779693961143494 lr:3.188643374829426e-05 tokens_per_second_per_gpu:5363.021131015997 mfu:25.298285700761873 +Step 22468 | loss:0.42432165145874023 lr:3.188239214257991e-05 tokens_per_second_per_gpu:5324.398957762226 mfu:25.11609831990158 +Step 22469 | loss:0.4877682328224182 lr:3.1878350673130394e-05 tokens_per_second_per_gpu:5008.167248004866 mfu:23.624379390282797 +Step 22470 | loss:0.31558898091316223 lr:3.187430933997613e-05 tokens_per_second_per_gpu:5335.637174059882 mfu:25.169110903615046 +Step 22471 | loss:0.39425423741340637 lr:3.1870268143147495e-05 tokens_per_second_per_gpu:5233.284938998 mfu:24.686297947736108 +Step 22472 | loss:0.4319075047969818 lr:3.186622708267489e-05 tokens_per_second_per_gpu:5301.079532565236 mfu:25.00609661254317 +Step 22473 | loss:0.460243821144104 lr:3.1862186158588694e-05 tokens_per_second_per_gpu:3625.510786221624 mfu:17.10215295453306 +Step 22474 | loss:0.4022355377674103 lr:3.1858145370919315e-05 tokens_per_second_per_gpu:4685.829248906074 mfu:22.103856052000086 +Step 22475 | loss:0.4613056778907776 lr:3.1854104719697144e-05 tokens_per_second_per_gpu:3848.3787185520914 mfu:18.153459016526163 +Step 22476 | loss:0.4309505820274353 lr:3.1850064204952554e-05 tokens_per_second_per_gpu:3600.8375960559665 mfu:16.98576530684132 +Step 22477 | loss:0.4252401888370514 lr:3.184602382671596e-05 tokens_per_second_per_gpu:5325.72855001305 mfu:25.12237023340055 +Step 22478 | loss:0.47197815775871277 lr:3.184198358501772e-05 tokens_per_second_per_gpu:4059.151181095212 mfu:19.147708683832406 +Step 22479 | loss:0.3610505163669586 lr:3.183794347988828e-05 tokens_per_second_per_gpu:5357.184910839459 mfu:25.270755254427304 +Step 22480 | loss:0.4890543520450592 lr:3.1833903511357946e-05 tokens_per_second_per_gpu:4457.064430870021 mfu:21.024733374021285 +Step 22481 | loss:0.37493836879730225 lr:3.182986367945715e-05 tokens_per_second_per_gpu:4585.578459880791 mfu:21.630955975619454 +Step 22482 | loss:0.4497021436691284 lr:3.182582398421626e-05 tokens_per_second_per_gpu:4015.6831543067156 mfu:18.942662585060937 +Step 22483 | loss:0.3894376754760742 lr:3.182178442566568e-05 tokens_per_second_per_gpu:5110.625163965899 mfu:24.10769086099405 +Step 22484 | loss:0.46727219223976135 lr:3.1817745003835774e-05 tokens_per_second_per_gpu:4837.4698384583235 mfu:22.81917058547052 +Step 22485 | loss:0.4500609338283539 lr:3.1813705718756926e-05 tokens_per_second_per_gpu:4355.214416074857 mfu:20.544289476828073 +Step 22486 | loss:0.5070180296897888 lr:3.180966657045954e-05 tokens_per_second_per_gpu:4297.158809412099 mfu:20.270431274891944 +Step 22487 | loss:0.39368686079978943 lr:3.180562755897395e-05 tokens_per_second_per_gpu:3390.3807518407048 mfu:15.99300446503763 +Step 22488 | loss:0.2697373628616333 lr:3.180158868433056e-05 tokens_per_second_per_gpu:5216.388350169044 mfu:24.6065938553746 +Step 22489 | loss:0.5655459761619568 lr:3.1797549946559734e-05 tokens_per_second_per_gpu:3850.9201412836883 mfu:18.165447341162537 +Step 22490 | loss:0.5076060891151428 lr:3.1793511345691876e-05 tokens_per_second_per_gpu:5349.5572267893895 mfu:25.234774167346032 +Step 22491 | loss:0.39750248193740845 lr:3.178947288175733e-05 tokens_per_second_per_gpu:4304.901873217369 mfu:20.306956627964315 +Step 22492 | loss:0.3407321870326996 lr:3.178543455478648e-05 tokens_per_second_per_gpu:5326.188771748932 mfu:25.124541177850624 +Step 22493 | loss:0.34093669056892395 lr:3.178139636480972e-05 tokens_per_second_per_gpu:5144.32545371608 mfu:24.266660877607993 +Step 22494 | loss:0.41309091448783875 lr:3.177735831185739e-05 tokens_per_second_per_gpu:4077.552359702682 mfu:19.234510182888652 +Step 22495 | loss:0.5693356990814209 lr:3.1773320395959874e-05 tokens_per_second_per_gpu:4796.585784036004 mfu:22.626313525221455 +Step 22496 | loss:0.3589343726634979 lr:3.176928261714753e-05 tokens_per_second_per_gpu:5341.327825267296 mfu:25.195954676285634 +Step 22497 | loss:0.3951404392719269 lr:3.176524497545075e-05 tokens_per_second_per_gpu:4336.305106293109 mfu:20.455091036326635 +Step 22498 | loss:0.48392409086227417 lr:3.176120747089988e-05 tokens_per_second_per_gpu:4837.86205412159 mfu:22.821020733673034 +Step 22499 | loss:0.4206666052341461 lr:3.17571701035253e-05 tokens_per_second_per_gpu:4276.125397570173 mfu:20.171213082563465 +Step 22500 | loss:0.3089796006679535 lr:3.175313287335738e-05 tokens_per_second_per_gpu:5302.3933058719695 mfu:25.01229390538387 +Step 22501 | loss:0.3247118294239044 lr:3.174909578042649e-05 tokens_per_second_per_gpu:5337.179981718598 mfu:25.176388590571342 +Step 22502 | loss:0.4720969796180725 lr:3.174505882476296e-05 tokens_per_second_per_gpu:5353.375080561745 mfu:25.252783634983253 +Step 22503 | loss:0.49768730998039246 lr:3.174102200639716e-05 tokens_per_second_per_gpu:5345.758261605852 mfu:25.216853800404984 +Step 22504 | loss:0.3766016364097595 lr:3.173698532535948e-05 tokens_per_second_per_gpu:5304.092976898682 mfu:25.020311543610653 +Step 22505 | loss:0.41484957933425903 lr:3.173294878168025e-05 tokens_per_second_per_gpu:4077.2088956432835 mfu:19.232890004319458 +Step 22506 | loss:0.37330693006515503 lr:3.172891237538985e-05 tokens_per_second_per_gpu:3920.5729686497607 mfu:18.494011611846087 +Step 22507 | loss:0.39102673530578613 lr:3.1724876106518635e-05 tokens_per_second_per_gpu:5347.34297301796 mfu:25.224329154515544 +Step 22508 | loss:0.5069999694824219 lr:3.172083997509697e-05 tokens_per_second_per_gpu:4069.44976840228 mfu:19.196288877280026 +Step 22509 | loss:0.3903306722640991 lr:3.171680398115518e-05 tokens_per_second_per_gpu:4129.660865608125 mfu:19.480314895872507 +Step 22510 | loss:0.36507293581962585 lr:3.171276812472365e-05 tokens_per_second_per_gpu:4849.384197769575 mfu:22.875372651141003 +Step 22511 | loss:0.4199869632720947 lr:3.170873240583272e-05 tokens_per_second_per_gpu:4516.546457269249 mfu:21.30532024122665 +Step 22512 | loss:0.3111204206943512 lr:3.1704696824512744e-05 tokens_per_second_per_gpu:5124.864614364051 mfu:24.174860778022893 +Step 22513 | loss:0.48102909326553345 lr:3.170066138079408e-05 tokens_per_second_per_gpu:5043.283120371797 mfu:23.790026951623247 +Step 22514 | loss:0.22189070284366608 lr:3.1696626074707075e-05 tokens_per_second_per_gpu:5256.166982527903 mfu:24.794236451145863 +Step 22515 | loss:0.4050773084163666 lr:3.169259090628211e-05 tokens_per_second_per_gpu:5309.705462040893 mfu:25.04678659361517 +Step 22516 | loss:0.3492056131362915 lr:3.1688555875549474e-05 tokens_per_second_per_gpu:5316.295901800251 mfu:25.077874822405036 +Step 22517 | loss:0.3516542911529541 lr:3.168452098253955e-05 tokens_per_second_per_gpu:3801.712497939928 mfu:17.933326491820328 +Step 22518 | loss:0.38961344957351685 lr:3.168048622728269e-05 tokens_per_second_per_gpu:5334.645256189234 mfu:25.16443185778056 +Step 22519 | loss:0.5229941010475159 lr:3.167645160980921e-05 tokens_per_second_per_gpu:3449.6882253712456 mfu:16.272767936579605 +Step 22520 | loss:0.4034099578857422 lr:3.167241713014949e-05 tokens_per_second_per_gpu:3529.172774127506 mfu:16.647710114525314 +Step 22521 | loss:0.4424823820590973 lr:3.166838278833385e-05 tokens_per_second_per_gpu:3492.373870947008 mfu:16.4741234096826 +Step 22522 | loss:0.5497177243232727 lr:3.1664348584392675e-05 tokens_per_second_per_gpu:5088.622965499743 mfu:24.003902736867037 +Step 22523 | loss:0.37389829754829407 lr:3.1660314518356236e-05 tokens_per_second_per_gpu:4556.651332366902 mfu:21.494501779659554 +Step 22524 | loss:0.4563426971435547 lr:3.165628059025493e-05 tokens_per_second_per_gpu:5365.252795593778 mfu:25.308812843336504 +Step 22525 | loss:0.43147316575050354 lr:3.1652246800119066e-05 tokens_per_second_per_gpu:5308.773821763396 mfu:25.042391887471997 +Step 22526 | loss:0.4050164222717285 lr:3.164821314797901e-05 tokens_per_second_per_gpu:5083.480975061764 mfu:23.979647090657057 +Step 22527 | loss:0.42199909687042236 lr:3.164417963386508e-05 tokens_per_second_per_gpu:4767.911096544778 mfu:22.49105013191917 +Step 22528 | loss:0.45081081986427307 lr:3.1640146257807604e-05 tokens_per_second_per_gpu:4816.531409929082 mfu:22.720400445633796 +Step 22529 | loss:0.4642164707183838 lr:3.163611301983695e-05 tokens_per_second_per_gpu:5366.549899260594 mfu:25.31493150264159 +Step 22530 | loss:0.46346336603164673 lr:3.1632079919983423e-05 tokens_per_second_per_gpu:3354.27803518514 mfu:15.822701790814696 +Step 22531 | loss:0.3117513358592987 lr:3.162804695827737e-05 tokens_per_second_per_gpu:4122.418649997289 mfu:19.446152129186796 +Step 22532 | loss:0.46068498492240906 lr:3.162401413474911e-05 tokens_per_second_per_gpu:4172.31181453812 mfu:19.68150669897756 +Step 22533 | loss:0.41368749737739563 lr:3.1619981449429e-05 tokens_per_second_per_gpu:4421.372187030081 mfu:20.856366970103537 +Step 22534 | loss:0.4725217819213867 lr:3.161594890234735e-05 tokens_per_second_per_gpu:4984.520288486435 mfu:23.51283264764703 +Step 22535 | loss:0.4109460711479187 lr:3.1611916493534486e-05 tokens_per_second_per_gpu:5314.202004583642 mfu:25.067997551978586 +Step 22536 | loss:0.41403263807296753 lr:3.160788422302075e-05 tokens_per_second_per_gpu:4917.2040728680195 mfu:23.19529057324425 +Step 22537 | loss:0.4033641815185547 lr:3.160385209083647e-05 tokens_per_second_per_gpu:5248.623973532521 mfu:24.758654790744604 +Step 22538 | loss:0.46145159006118774 lr:3.159982009701197e-05 tokens_per_second_per_gpu:4533.586843203604 mfu:21.385702649068644 +Step 22539 | loss:0.4514451026916504 lr:3.159578824157755e-05 tokens_per_second_per_gpu:4840.261918257027 mfu:22.83234130225894 +Step 22540 | loss:0.3999563753604889 lr:3.1591756524563565e-05 tokens_per_second_per_gpu:3233.3812627041216 mfu:15.252411087905346 +Step 22541 | loss:0.43666356801986694 lr:3.158772494600032e-05 tokens_per_second_per_gpu:5349.167485447542 mfu:25.232935690939804 +Step 22542 | loss:0.4931267499923706 lr:3.158369350591816e-05 tokens_per_second_per_gpu:5340.645741484664 mfu:25.192737170707463 +Step 22543 | loss:0.34273281693458557 lr:3.1579662204347384e-05 tokens_per_second_per_gpu:3559.126986410965 mfu:16.789009244582004 +Step 22544 | loss:0.46387162804603577 lr:3.157563104131833e-05 tokens_per_second_per_gpu:4158.446699169641 mfu:19.616102584151825 +Step 22545 | loss:0.3845582604408264 lr:3.157160001686129e-05 tokens_per_second_per_gpu:5315.350284819921 mfu:25.07341418576993 +Step 22546 | loss:0.2791423201560974 lr:3.15675691310066e-05 tokens_per_second_per_gpu:4172.085381812051 mfu:19.680438577175288 +Step 22547 | loss:0.474320650100708 lr:3.1563538383784585e-05 tokens_per_second_per_gpu:4735.273291494681 mfu:22.337092036914107 +Step 22548 | loss:0.4666315019130707 lr:3.1559507775225533e-05 tokens_per_second_per_gpu:4971.393843943768 mfu:23.45091296913689 +Step 22549 | loss:0.366237610578537 lr:3.1555477305359786e-05 tokens_per_second_per_gpu:4463.307276440693 mfu:21.054181941717435 +Step 22550 | loss:0.38066723942756653 lr:3.155144697421764e-05 tokens_per_second_per_gpu:4865.029011260953 mfu:22.94917190566026 +Step 22551 | loss:0.6320831775665283 lr:3.154741678182945e-05 tokens_per_second_per_gpu:3912.885487497808 mfu:18.45774845163279 +Step 22552 | loss:0.4402678608894348 lr:3.154338672822545e-05 tokens_per_second_per_gpu:5349.88965375239 mfu:25.236342282048387 +Step 22553 | loss:0.3384729027748108 lr:3.1539356813436013e-05 tokens_per_second_per_gpu:5320.518584620153 mfu:25.097793937730497 +Step 22554 | loss:0.472817063331604 lr:3.153532703749144e-05 tokens_per_second_per_gpu:5329.889473148245 mfu:25.141998017755956 +Step 22555 | loss:0.5143977403640747 lr:3.1531297400422e-05 tokens_per_second_per_gpu:4017.2076138304724 mfu:18.949853720734115 +Step 22556 | loss:0.35639461874961853 lr:3.152726790225804e-05 tokens_per_second_per_gpu:5297.609665928419 mfu:24.98972865959701 +Step 22557 | loss:0.3876335620880127 lr:3.1523238543029855e-05 tokens_per_second_per_gpu:3450.189448913648 mfu:16.27513229354667 +Step 22558 | loss:0.4713190197944641 lr:3.151920932276777e-05 tokens_per_second_per_gpu:4109.159652925813 mfu:19.383607177782466 +Step 22559 | loss:0.40336307883262634 lr:3.1515180241502054e-05 tokens_per_second_per_gpu:3544.0959546838412 mfu:16.718105303366702 +Step 22560 | loss:0.4194124639034271 lr:3.151115129926302e-05 tokens_per_second_per_gpu:4755.520910504781 mfu:22.432603510384777 +Step 22561 | loss:0.43576958775520325 lr:3.1507122496080975e-05 tokens_per_second_per_gpu:5019.851638351052 mfu:23.67949665310832 +Step 22562 | loss:0.4327423870563507 lr:3.150309383198623e-05 tokens_per_second_per_gpu:4361.336711145019 mfu:20.57316938724528 +Step 22563 | loss:0.3455519378185272 lr:3.149906530700908e-05 tokens_per_second_per_gpu:4456.907973830797 mfu:21.023995339472947 +Step 22564 | loss:0.439537912607193 lr:3.149503692117981e-05 tokens_per_second_per_gpu:5345.841290344983 mfu:25.21724546113363 +Step 22565 | loss:0.504359781742096 lr:3.1491008674528734e-05 tokens_per_second_per_gpu:5328.723560969219 mfu:25.13649821108236 +Step 22566 | loss:0.2795700430870056 lr:3.1486980567086157e-05 tokens_per_second_per_gpu:5333.23359366822 mfu:25.157772804814474 +Step 22567 | loss:0.4454170763492584 lr:3.1482952598882345e-05 tokens_per_second_per_gpu:4422.485564576849 mfu:20.861618961953443 +Step 22568 | loss:0.48205018043518066 lr:3.14789247699476e-05 tokens_per_second_per_gpu:5314.921878644203 mfu:25.071393320745436 +Step 22569 | loss:0.5126267075538635 lr:3.1474897080312246e-05 tokens_per_second_per_gpu:4842.242750621568 mfu:22.84168522648733 +Step 22570 | loss:0.4820531904697418 lr:3.147086953000655e-05 tokens_per_second_per_gpu:4350.431761311092 mfu:20.521728878302344 +Step 22571 | loss:0.3208814263343811 lr:3.14668421190608e-05 tokens_per_second_per_gpu:5067.989417214216 mfu:23.906570769157852 +Step 22572 | loss:0.4234042465686798 lr:3.14628148475053e-05 tokens_per_second_per_gpu:4858.752127216954 mfu:22.91956277267787 +Step 22573 | loss:0.4946650266647339 lr:3.1458787715370345e-05 tokens_per_second_per_gpu:5324.562465430662 mfu:25.11686961347088 +Step 22574 | loss:0.5396488308906555 lr:3.1454760722686206e-05 tokens_per_second_per_gpu:3818.407355657415 mfu:18.01207898411016 +Step 22575 | loss:0.37657400965690613 lr:3.1450733869483166e-05 tokens_per_second_per_gpu:5367.290784581952 mfu:25.3184263851105 +Step 22576 | loss:0.47150006890296936 lr:3.1446707155791535e-05 tokens_per_second_per_gpu:4529.258528104039 mfu:21.365285248257287 +Step 22577 | loss:0.4578847885131836 lr:3.144268058164158e-05 tokens_per_second_per_gpu:5345.587839196314 mfu:25.216049888822134 +Step 22578 | loss:0.4009271264076233 lr:3.14386541470636e-05 tokens_per_second_per_gpu:3876.1916765056512 mfu:18.28465748457306 +Step 22579 | loss:0.44595852494239807 lr:3.143462785208787e-05 tokens_per_second_per_gpu:4483.437476584529 mfu:21.149139530362213 +Step 22580 | loss:0.3501315414905548 lr:3.143060169674468e-05 tokens_per_second_per_gpu:4903.140150237788 mfu:23.128948650644812 +Step 22581 | loss:0.4494470953941345 lr:3.142657568106429e-05 tokens_per_second_per_gpu:4693.793215901881 mfu:22.14142344311219 +Step 22582 | loss:0.4615429937839508 lr:3.142254980507699e-05 tokens_per_second_per_gpu:3704.063437962107 mfu:17.47269921525732 +Step 22583 | loss:0.35543373227119446 lr:3.141852406881308e-05 tokens_per_second_per_gpu:4601.566424115476 mfu:21.706373930742448 +Step 22584 | loss:0.37180811166763306 lr:3.141449847230281e-05 tokens_per_second_per_gpu:4603.698330037329 mfu:21.71643049471648 +Step 22585 | loss:0.44440263509750366 lr:3.141047301557647e-05 tokens_per_second_per_gpu:5348.261435937782 mfu:25.228661700814282 +Step 22586 | loss:0.48777350783348083 lr:3.1406447698664345e-05 tokens_per_second_per_gpu:3776.017630884035 mfu:17.812119419921295 +Step 22587 | loss:0.4311216473579407 lr:3.14024225215967e-05 tokens_per_second_per_gpu:3946.45794074042 mfu:18.616115441629372 +Step 22588 | loss:0.46905526518821716 lr:3.139839748440379e-05 tokens_per_second_per_gpu:4666.069289264898 mfu:22.010644950975895 +Step 22589 | loss:0.4754761755466461 lr:3.1394372587115916e-05 tokens_per_second_per_gpu:5312.053463071358 mfu:25.057862515086164 +Step 22590 | loss:0.44370102882385254 lr:3.139034782976335e-05 tokens_per_second_per_gpu:5226.032167909791 mfu:24.652085389062673 +Step 22591 | loss:0.5402511954307556 lr:3.138632321237633e-05 tokens_per_second_per_gpu:4565.416560062198 mfu:21.535848854200324 +Step 22592 | loss:0.40757402777671814 lr:3.138229873498516e-05 tokens_per_second_per_gpu:4408.914050599171 mfu:20.797599814981677 +Step 22593 | loss:0.33012887835502625 lr:3.137827439762009e-05 tokens_per_second_per_gpu:5331.634082384773 mfu:25.150227637185687 +Step 22594 | loss:0.4967966675758362 lr:3.1374250200311405e-05 tokens_per_second_per_gpu:4887.5349973979555 mfu:23.055336482185698 +Step 22595 | loss:0.44408172369003296 lr:3.137022614308937e-05 tokens_per_second_per_gpu:5355.876912590873 mfu:25.26458520352121 +Step 22596 | loss:0.47397804260253906 lr:3.136620222598423e-05 tokens_per_second_per_gpu:5309.203845865471 mfu:25.044420384531158 +Step 22597 | loss:0.3466448485851288 lr:3.136217844902625e-05 tokens_per_second_per_gpu:3396.976503683548 mfu:16.02411775183163 +Step 22598 | loss:0.4406963884830475 lr:3.135815481224572e-05 tokens_per_second_per_gpu:3769.190369178034 mfu:17.779914061603126 +Step 22599 | loss:0.3719392418861389 lr:3.135413131567288e-05 tokens_per_second_per_gpu:5226.463978675721 mfu:24.65412231412008 +Step 22600 | loss:0.38743653893470764 lr:3.1350107959337985e-05 tokens_per_second_per_gpu:4848.340165919751 mfu:22.870447774775183 +Step 22601 | loss:0.4005054235458374 lr:3.134608474327133e-05 tokens_per_second_per_gpu:5371.383348636062 mfu:25.33773170801717 +Step 22602 | loss:0.37055689096450806 lr:3.1342061667503155e-05 tokens_per_second_per_gpu:5349.948321622866 mfu:25.2366190284032 +Step 22603 | loss:0.38463062047958374 lr:3.13380387320637e-05 tokens_per_second_per_gpu:5176.577991481338 mfu:24.41880160109709 +Step 22604 | loss:0.49812740087509155 lr:3.133401593698323e-05 tokens_per_second_per_gpu:4142.6311118143085 mfu:19.54149776017927 +Step 22605 | loss:0.4320167303085327 lr:3.132999328229202e-05 tokens_per_second_per_gpu:4881.247931880359 mfu:23.025679321455208 +Step 22606 | loss:0.44893693923950195 lr:3.1325970768020314e-05 tokens_per_second_per_gpu:4855.665059500018 mfu:22.90500054754902 +Step 22607 | loss:0.5210211277008057 lr:3.132194839419835e-05 tokens_per_second_per_gpu:4353.726956354684 mfu:20.53727287554122 +Step 22608 | loss:0.33311134576797485 lr:3.131792616085641e-05 tokens_per_second_per_gpu:5383.0776463615575 mfu:25.392895705641266 +Step 22609 | loss:0.5383377075195312 lr:3.131390406802474e-05 tokens_per_second_per_gpu:4172.649158505518 mfu:19.683098008028594 +Step 22610 | loss:0.5715117454528809 lr:3.130988211573357e-05 tokens_per_second_per_gpu:4948.422764472278 mfu:23.34255446799957 +Step 22611 | loss:0.4085075259208679 lr:3.1305860304013154e-05 tokens_per_second_per_gpu:4068.297207305354 mfu:19.190852049938623 +Step 22612 | loss:0.43418624997138977 lr:3.130183863289375e-05 tokens_per_second_per_gpu:4326.948527052368 mfu:20.41095445565103 +Step 22613 | loss:0.42029881477355957 lr:3.129781710240561e-05 tokens_per_second_per_gpu:5143.7160753035205 mfu:24.26378633566527 +Step 22614 | loss:0.4937760829925537 lr:3.129379571257897e-05 tokens_per_second_per_gpu:5126.500066076821 mfu:24.182575482788867 +Step 22615 | loss:0.42214664816856384 lr:3.128977446344409e-05 tokens_per_second_per_gpu:4559.125357103857 mfu:21.50617217645592 +Step 22616 | loss:0.474719762802124 lr:3.12857533550312e-05 tokens_per_second_per_gpu:4160.141881015505 mfu:19.62409904614661 +Step 22617 | loss:0.43625393509864807 lr:3.128173238737054e-05 tokens_per_second_per_gpu:5350.014579142204 mfu:25.236931576426613 +Step 22618 | loss:0.4160878658294678 lr:3.1277711560492364e-05 tokens_per_second_per_gpu:4121.388020286656 mfu:19.441290472998176 +Step 22619 | loss:0.5894079208374023 lr:3.1273690874426905e-05 tokens_per_second_per_gpu:5127.261954449131 mfu:24.18616944023261 +Step 22620 | loss:0.44901421666145325 lr:3.12696703292044e-05 tokens_per_second_per_gpu:5336.922080175983 mfu:25.175172024992282 +Step 22621 | loss:0.43921172618865967 lr:3.12656499248551e-05 tokens_per_second_per_gpu:3798.0892024768555 mfu:17.916234788923063 +Step 22622 | loss:0.4801546335220337 lr:3.126162966140924e-05 tokens_per_second_per_gpu:4620.581578914668 mfu:21.79607166025019 +Step 22623 | loss:0.5791383981704712 lr:3.125760953889707e-05 tokens_per_second_per_gpu:3275.2335646883944 mfu:15.449835537104208 +Step 22624 | loss:0.397535115480423 lr:3.125358955734878e-05 tokens_per_second_per_gpu:4961.935293150875 mfu:23.40629537125164 +Step 22625 | loss:0.3801344931125641 lr:3.124956971679466e-05 tokens_per_second_per_gpu:4789.203203218622 mfu:22.591488631907694 +Step 22626 | loss:0.4085211753845215 lr:3.1245550017264904e-05 tokens_per_second_per_gpu:3111.713430785705 mfu:14.678483166073482 +Step 22627 | loss:0.46120327711105347 lr:3.1241530458789756e-05 tokens_per_second_per_gpu:5339.991292814957 mfu:25.189650024671995 +Step 22628 | loss:0.4973335266113281 lr:3.123751104139946e-05 tokens_per_second_per_gpu:4913.886296642753 mfu:23.179640056718675 +Step 22629 | loss:0.4356326162815094 lr:3.123349176512423e-05 tokens_per_second_per_gpu:5157.016977292777 mfu:24.32652895971648 +Step 22630 | loss:0.4376240372657776 lr:3.122947262999431e-05 tokens_per_second_per_gpu:5015.621313139927 mfu:23.659541487319537 +Step 22631 | loss:0.46765026450157166 lr:3.122545363603993e-05 tokens_per_second_per_gpu:5217.474345145174 mfu:24.61171667896685 +Step 22632 | loss:0.3314715325832367 lr:3.1221434783291304e-05 tokens_per_second_per_gpu:3923.0114785365504 mfu:18.505514479034762 +Step 22633 | loss:0.4498026967048645 lr:3.121741607177865e-05 tokens_per_second_per_gpu:5300.070992819228 mfu:25.001339158486626 +Step 22634 | loss:0.3961949646472931 lr:3.121339750153221e-05 tokens_per_second_per_gpu:4590.560515246005 mfu:21.654457180803465 +Step 22635 | loss:0.4857841432094574 lr:3.1209379072582215e-05 tokens_per_second_per_gpu:4875.937619644558 mfu:23.00062967260624 +Step 22636 | loss:0.4490906000137329 lr:3.1205360784958864e-05 tokens_per_second_per_gpu:5343.420190289913 mfu:25.2058247190916 +Step 22637 | loss:0.4749087989330292 lr:3.12013426386924e-05 tokens_per_second_per_gpu:4674.996132026121 mfu:22.052754391357016 +Step 22638 | loss:0.4220500886440277 lr:3.119732463381305e-05 tokens_per_second_per_gpu:4916.380110136432 mfu:23.19140380045682 +Step 22639 | loss:0.5319597125053406 lr:3.119330677035101e-05 tokens_per_second_per_gpu:4212.233241361594 mfu:19.869822880601664 +Step 22640 | loss:0.4427241086959839 lr:3.1189289048336504e-05 tokens_per_second_per_gpu:3441.3514108929594 mfu:16.233441760278303 +Step 22641 | loss:0.4171174168586731 lr:3.118527146779976e-05 tokens_per_second_per_gpu:4193.524682878943 mfu:19.78157141822977 +Step 22642 | loss:0.6010527610778809 lr:3.118125402877099e-05 tokens_per_second_per_gpu:4839.863719832682 mfu:22.830462932351647 +Step 22643 | loss:0.43411484360694885 lr:3.11772367312804e-05 tokens_per_second_per_gpu:4857.248412410315 mfu:22.91246949337475 +Step 22644 | loss:0.3847373425960541 lr:3.117321957535822e-05 tokens_per_second_per_gpu:4358.547101050554 mfu:20.56001032047313 +Step 22645 | loss:0.4446379542350769 lr:3.116920256103468e-05 tokens_per_second_per_gpu:5357.749410944265 mfu:25.273418097735323 +Step 22646 | loss:0.4365047812461853 lr:3.1165185688339946e-05 tokens_per_second_per_gpu:4168.355668987106 mfu:19.66284488542539 +Step 22647 | loss:0.49932050704956055 lr:3.116116895730425e-05 tokens_per_second_per_gpu:4915.178132019761 mfu:23.18573386460186 +Step 22648 | loss:0.48647186160087585 lr:3.115715236795782e-05 tokens_per_second_per_gpu:5023.761108832515 mfu:23.697938292393825 +Step 22649 | loss:0.5562907457351685 lr:3.115313592033084e-05 tokens_per_second_per_gpu:3408.226310436888 mfu:16.077185009702088 +Step 22650 | loss:0.3725842237472534 lr:3.114911961445354e-05 tokens_per_second_per_gpu:5369.343398267823 mfu:25.328108914078765 +Step 22651 | loss:0.4468700587749481 lr:3.114510345035612e-05 tokens_per_second_per_gpu:4341.126666653569 mfu:20.477835159189954 +Step 22652 | loss:0.43768230080604553 lr:3.1141087428068785e-05 tokens_per_second_per_gpu:5382.856350024255 mfu:25.391851813061166 +Step 22653 | loss:0.5063401460647583 lr:3.113707154762171e-05 tokens_per_second_per_gpu:5358.159652883205 mfu:25.275353278955798 +Step 22654 | loss:0.4035152792930603 lr:3.113305580904515e-05 tokens_per_second_per_gpu:4143.702775123893 mfu:19.546552978855196 +Step 22655 | loss:0.5202662944793701 lr:3.112904021236929e-05 tokens_per_second_per_gpu:5135.484171744446 mfu:24.224955042069542 +Step 22656 | loss:0.34242409467697144 lr:3.1125024757624307e-05 tokens_per_second_per_gpu:5057.589749937358 mfu:23.857513764246704 +Step 22657 | loss:0.4797116816043854 lr:3.1121009444840434e-05 tokens_per_second_per_gpu:5327.232577664592 mfu:25.12946498844645 +Step 22658 | loss:0.49801838397979736 lr:3.111699427404785e-05 tokens_per_second_per_gpu:3712.431333244144 mfu:17.512172005012847 +Step 22659 | loss:0.32084986567497253 lr:3.111297924527676e-05 tokens_per_second_per_gpu:4787.823890149919 mfu:22.584982176827545 +Step 22660 | loss:0.3916829824447632 lr:3.110896435855738e-05 tokens_per_second_per_gpu:3292.566093299569 mfu:15.53159603179744 +Step 22661 | loss:0.37336987257003784 lr:3.110494961391989e-05 tokens_per_second_per_gpu:3570.3939510305527 mfu:16.84215743903499 +Step 22662 | loss:0.5066022276878357 lr:3.110093501139447e-05 tokens_per_second_per_gpu:3878.460251939563 mfu:18.295358741953702 +Step 22663 | loss:0.40550127625465393 lr:3.109692055101132e-05 tokens_per_second_per_gpu:3726.7174821718377 mfu:17.57956220697313 +Step 22664 | loss:0.4502231180667877 lr:3.109290623280067e-05 tokens_per_second_per_gpu:4819.380172067289 mfu:22.733838542688662 +Step 22665 | loss:0.38377296924591064 lr:3.1088892056792656e-05 tokens_per_second_per_gpu:5004.799569612782 mfu:23.608493476722153 +Step 22666 | loss:0.3303023874759674 lr:3.1084878023017515e-05 tokens_per_second_per_gpu:4108.645685784768 mfu:19.38118270708632 +Step 22667 | loss:0.2699272036552429 lr:3.108086413150543e-05 tokens_per_second_per_gpu:4265.753695951276 mfu:20.12228799643236 +Step 22668 | loss:0.4629516303539276 lr:3.107685038228657e-05 tokens_per_second_per_gpu:5351.141442509288 mfu:25.242247183191868 +Step 22669 | loss:0.28070637583732605 lr:3.107283677539112e-05 tokens_per_second_per_gpu:5331.974624990122 mfu:25.151834034757677 +Step 22670 | loss:0.4645671844482422 lr:3.1068823310849293e-05 tokens_per_second_per_gpu:4342.735087789142 mfu:20.485422356111396 +Step 22671 | loss:0.4664362072944641 lr:3.1064809988691265e-05 tokens_per_second_per_gpu:5332.6932347574275 mfu:25.15522383963753 +Step 22672 | loss:0.4022359848022461 lr:3.10607968089472e-05 tokens_per_second_per_gpu:4560.078195742879 mfu:21.510666878887218 +Step 22673 | loss:0.37078753113746643 lr:3.105678377164731e-05 tokens_per_second_per_gpu:4600.548984147147 mfu:21.701574492818683 +Step 22674 | loss:0.4199434816837311 lr:3.105277087682178e-05 tokens_per_second_per_gpu:4037.298168791761 mfu:19.044624296288656 +Step 22675 | loss:0.37921053171157837 lr:3.1048758124500763e-05 tokens_per_second_per_gpu:4238.740461064259 mfu:19.994861958537168 +Step 22676 | loss:0.4782707095146179 lr:3.1044745514714445e-05 tokens_per_second_per_gpu:4623.755375486859 mfu:21.811043000186743 +Step 22677 | loss:0.43363088369369507 lr:3.104073304749303e-05 tokens_per_second_per_gpu:3996.163817327535 mfu:18.85058654218784 +Step 22678 | loss:0.4050842821598053 lr:3.103672072286668e-05 tokens_per_second_per_gpu:3460.518292174181 mfu:16.323855209488986 +Step 22679 | loss:0.4217662215232849 lr:3.1032708540865564e-05 tokens_per_second_per_gpu:5391.8556216487195 mfu:25.43430291274783 +Step 22680 | loss:0.40888532996177673 lr:3.102869650151987e-05 tokens_per_second_per_gpu:5333.238685036097 mfu:25.15779682166543 +Step 22681 | loss:0.2556735575199127 lr:3.102468460485979e-05 tokens_per_second_per_gpu:5335.974392441968 mfu:25.170701620258733 +Step 22682 | loss:0.28152379393577576 lr:3.102067285091546e-05 tokens_per_second_per_gpu:4196.365204558956 mfu:19.794970643637367 +Step 22683 | loss:0.4427981972694397 lr:3.1016661239717064e-05 tokens_per_second_per_gpu:5365.177430720378 mfu:25.308457334370107 +Step 22684 | loss:0.38566258549690247 lr:3.101264977129479e-05 tokens_per_second_per_gpu:4910.289365847454 mfu:23.162672720457575 +Step 22685 | loss:0.5208349227905273 lr:3.1008638445678796e-05 tokens_per_second_per_gpu:5358.613286844456 mfu:25.27749314774895 +Step 22686 | loss:0.4587208032608032 lr:3.100462726289925e-05 tokens_per_second_per_gpu:4765.177492117241 mfu:22.478155253433467 +Step 22687 | loss:0.36595675349235535 lr:3.100061622298634e-05 tokens_per_second_per_gpu:4650.780163429326 mfu:21.938523535815648 +Step 22688 | loss:0.45000141859054565 lr:3.0996605325970216e-05 tokens_per_second_per_gpu:5346.420227244921 mfu:25.21997640526746 +Step 22689 | loss:0.36707887053489685 lr:3.099259457188106e-05 tokens_per_second_per_gpu:4606.662107297002 mfu:21.730411137722903 +Step 22690 | loss:0.4740746319293976 lr:3.098858396074901e-05 tokens_per_second_per_gpu:5315.980916313821 mfu:25.07638898212348 +Step 22691 | loss:0.44068437814712524 lr:3.098457349260425e-05 tokens_per_second_per_gpu:4745.407645890352 mfu:22.384897515697645 +Step 22692 | loss:0.45569533109664917 lr:3.098056316747693e-05 tokens_per_second_per_gpu:4844.803296158219 mfu:22.85376375665783 +Step 22693 | loss:0.44933202862739563 lr:3.0976552985397236e-05 tokens_per_second_per_gpu:4259.28718469391 mfu:20.091784359530774 +Step 22694 | loss:0.42171549797058105 lr:3.09725429463953e-05 tokens_per_second_per_gpu:4879.880208804282 mfu:23.01922753834773 +Step 22695 | loss:0.36169323325157166 lr:3.09685330505013e-05 tokens_per_second_per_gpu:4999.624880267677 mfu:23.58408358418825 +Step 22696 | loss:0.3198247253894806 lr:3.0964523297745404e-05 tokens_per_second_per_gpu:4528.581214376798 mfu:21.362090243844253 +Step 22697 | loss:0.43209147453308105 lr:3.0960513688157754e-05 tokens_per_second_per_gpu:3794.4880682957737 mfu:17.899247624574052 +Step 22698 | loss:0.3185917139053345 lr:3.095650422176851e-05 tokens_per_second_per_gpu:4834.449467503019 mfu:22.804922980345925 +Step 22699 | loss:0.549789547920227 lr:3.095249489860781e-05 tokens_per_second_per_gpu:4183.37502916019 mfu:19.733693769928923 +Step 22700 | loss:0.3891347646713257 lr:3.0948485718705846e-05 tokens_per_second_per_gpu:5329.52865776388 mfu:25.140295990026452 +Step 22701 | loss:0.31520578265190125 lr:3.094447668209273e-05 tokens_per_second_per_gpu:4227.412910291672 mfu:19.941427968863426 +Step 22702 | loss:0.42923396825790405 lr:3.0940467788798656e-05 tokens_per_second_per_gpu:4927.857473348086 mfu:23.24554448096628 +Step 22703 | loss:0.3399386405944824 lr:3.093645903885376e-05 tokens_per_second_per_gpu:5347.836813861881 mfu:25.226658686034256 +Step 22704 | loss:0.48986315727233887 lr:3.093245043228818e-05 tokens_per_second_per_gpu:3934.4119560721756 mfu:18.55929247669235 +Step 22705 | loss:0.455160915851593 lr:3.0928441969132075e-05 tokens_per_second_per_gpu:4952.656734056359 mfu:23.362526824917385 +Step 22706 | loss:0.43688759207725525 lr:3.092443364941558e-05 tokens_per_second_per_gpu:4681.170987285217 mfu:22.081882237153117 +Step 22707 | loss:0.4495159089565277 lr:3.0920425473168865e-05 tokens_per_second_per_gpu:4322.8964637377485 mfu:20.391840181642678 +Step 22708 | loss:0.5024126768112183 lr:3.091641744042206e-05 tokens_per_second_per_gpu:4072.7513065610156 mfu:19.211862796075277 +Step 22709 | loss:0.5031439661979675 lr:3.091240955120531e-05 tokens_per_second_per_gpu:5326.2666381932795 mfu:25.124908487154002 +Step 22710 | loss:0.45756202936172485 lr:3.0908401805548783e-05 tokens_per_second_per_gpu:5355.001774812128 mfu:25.26045702930497 +Step 22711 | loss:0.5255154967308044 lr:3.0904394203482595e-05 tokens_per_second_per_gpu:3724.3058357243262 mfu:17.568186059210905 +Step 22712 | loss:0.4854884147644043 lr:3.090038674503688e-05 tokens_per_second_per_gpu:5147.3667046552355 mfu:24.281006977178905 +Step 22713 | loss:0.37383201718330383 lr:3.089637943024181e-05 tokens_per_second_per_gpu:3421.962951690598 mfu:16.141983090208512 +Step 22714 | loss:0.39667320251464844 lr:3.0892372259127497e-05 tokens_per_second_per_gpu:5321.170494392729 mfu:25.100869107355013 +Step 22715 | loss:0.3974367678165436 lr:3.088836523172408e-05 tokens_per_second_per_gpu:5312.484549688554 mfu:25.059896024210435 +Step 22716 | loss:0.4052933156490326 lr:3.0884358348061725e-05 tokens_per_second_per_gpu:4987.848710620044 mfu:23.528533382738637 +Step 22717 | loss:0.40912747383117676 lr:3.088035160817054e-05 tokens_per_second_per_gpu:4727.523873859546 mfu:22.30053671174219 +Step 22718 | loss:0.4523380398750305 lr:3.08763450120807e-05 tokens_per_second_per_gpu:4698.215663684471 mfu:22.162284883850067 +Step 22719 | loss:0.35596683621406555 lr:3.087233855982227e-05 tokens_per_second_per_gpu:5148.389644764143 mfu:24.2858323602042 +Step 22720 | loss:0.4773598611354828 lr:3.086833225142544e-05 tokens_per_second_per_gpu:5299.285036505487 mfu:24.997631668456737 +Step 22721 | loss:0.3303056061267853 lr:3.086432608692031e-05 tokens_per_second_per_gpu:4005.51317607256 mfu:18.894689062553212 +Step 22722 | loss:0.34568068385124207 lr:3.086032006633705e-05 tokens_per_second_per_gpu:4517.622780724522 mfu:21.31039744260489 +Step 22723 | loss:0.39425143599510193 lr:3.085631418970575e-05 tokens_per_second_per_gpu:5337.218143595161 mfu:25.176568606654172 +Step 22724 | loss:0.3897838294506073 lr:3.0852308457056544e-05 tokens_per_second_per_gpu:4865.57311659712 mfu:22.951738543364907 +Step 22725 | loss:0.445610374212265 lr:3.084830286841961e-05 tokens_per_second_per_gpu:5304.967961304473 mfu:25.024438994340503 +Step 22726 | loss:0.4592410922050476 lr:3.084429742382499e-05 tokens_per_second_per_gpu:4057.3515371270873 mfu:19.139219456183834 +Step 22727 | loss:0.43667107820510864 lr:3.084029212330287e-05 tokens_per_second_per_gpu:3689.808417868885 mfu:17.40545585332037 +Step 22728 | loss:0.4240398406982422 lr:3.083628696688334e-05 tokens_per_second_per_gpu:4483.282860036736 mfu:21.14841017772597 +Step 22729 | loss:0.3982354998588562 lr:3.0832281954596564e-05 tokens_per_second_per_gpu:4298.079174435822 mfu:20.274772793739313 +Step 22730 | loss:0.3392334580421448 lr:3.082827708647263e-05 tokens_per_second_per_gpu:4252.967202671381 mfu:20.061971926030377 +Step 22731 | loss:0.44343605637550354 lr:3.0824272362541674e-05 tokens_per_second_per_gpu:5207.865308185747 mfu:24.566389212158477 +Step 22732 | loss:0.39347219467163086 lr:3.082026778283382e-05 tokens_per_second_per_gpu:3662.806963485695 mfu:17.278085386071695 +Step 22733 | loss:0.5315667986869812 lr:3.081626334737918e-05 tokens_per_second_per_gpu:4932.172537355978 mfu:23.265899374117343 +Step 22734 | loss:0.498897522687912 lr:3.0812259056207864e-05 tokens_per_second_per_gpu:5361.7387810435475 mfu:25.292236637150367 +Step 22735 | loss:0.38967791199684143 lr:3.080825490934999e-05 tokens_per_second_per_gpu:5342.222013825671 mfu:25.200172716279784 +Step 22736 | loss:0.582159161567688 lr:3.0804250906835685e-05 tokens_per_second_per_gpu:4658.719336264429 mfu:21.97597396864139 +Step 22737 | loss:0.3538966476917267 lr:3.0800247048695056e-05 tokens_per_second_per_gpu:3761.2860639825444 mfu:17.742628105382472 +Step 22738 | loss:0.42139819264411926 lr:3.0796243334958216e-05 tokens_per_second_per_gpu:5340.8204534851675 mfu:25.193561317022777 +Step 22739 | loss:0.40547794103622437 lr:3.07922397656553e-05 tokens_per_second_per_gpu:4534.544316076357 mfu:21.390219211971136 +Step 22740 | loss:0.4605601131916046 lr:3.078823634081639e-05 tokens_per_second_per_gpu:3810.862641530573 mfu:17.97648928031288 +Step 22741 | loss:0.46907395124435425 lr:3.07842330604716e-05 tokens_per_second_per_gpu:4472.8243794219015 mfu:21.099075740265004 +Step 22742 | loss:0.32845786213874817 lr:3.078022992465104e-05 tokens_per_second_per_gpu:3586.7612055056406 mfu:16.919364570935553 +Step 22743 | loss:0.45953404903411865 lr:3.0776226933384835e-05 tokens_per_second_per_gpu:4469.796061860146 mfu:21.08479064964195 +Step 22744 | loss:0.3797840476036072 lr:3.077222408670306e-05 tokens_per_second_per_gpu:5314.230665771208 mfu:25.068132751690438 +Step 22745 | loss:0.4369634687900543 lr:3.076822138463586e-05 tokens_per_second_per_gpu:5354.732808178784 mfu:25.259188267805044 +Step 22746 | loss:0.3577265441417694 lr:3.0764218827213334e-05 tokens_per_second_per_gpu:5357.909539436759 mfu:25.274173451155153 +Step 22747 | loss:0.43328481912612915 lr:3.076021641446555e-05 tokens_per_second_per_gpu:4350.14555194348 mfu:20.520378779882968 +Step 22748 | loss:0.3439177870750427 lr:3.0756214146422635e-05 tokens_per_second_per_gpu:5362.852668198799 mfu:25.29749103291043 +Step 22749 | loss:0.39493125677108765 lr:3.075221202311469e-05 tokens_per_second_per_gpu:5303.722057489777 mfu:25.018561853473475 +Step 22750 | loss:0.5194951295852661 lr:3.074821004457182e-05 tokens_per_second_per_gpu:4572.01961362347 mfu:21.56699658444595 +Step 22751 | loss:0.37755629420280457 lr:3.074420821082411e-05 tokens_per_second_per_gpu:3519.0923890470212 mfu:16.600159218209622 +Step 22752 | loss:0.5047882795333862 lr:3.074020652190167e-05 tokens_per_second_per_gpu:4228.208546591302 mfu:19.94518112103833 +Step 22753 | loss:0.3920620381832123 lr:3.073620497783459e-05 tokens_per_second_per_gpu:3906.1402296062915 mfu:18.425929919298838 +Step 22754 | loss:0.3737127184867859 lr:3.0732203578653e-05 tokens_per_second_per_gpu:5361.458124210056 mfu:25.29091273098179 +Step 22755 | loss:0.3490366041660309 lr:3.0728202324386934e-05 tokens_per_second_per_gpu:4173.891536228914 mfu:19.68895851572135 +Step 22756 | loss:0.36155930161476135 lr:3.0724201215066526e-05 tokens_per_second_per_gpu:4894.245048085218 mfu:23.086988936130215 +Step 22757 | loss:0.36760982871055603 lr:3.072020025072186e-05 tokens_per_second_per_gpu:5360.2206459257295 mfu:25.28507533477906 +Step 22758 | loss:0.400789737701416 lr:3.071619943138303e-05 tokens_per_second_per_gpu:5274.490081013991 mfu:24.880669632947708 +Step 22759 | loss:0.4727417230606079 lr:3.0712198757080123e-05 tokens_per_second_per_gpu:4641.3258101338315 mfu:21.89392573824179 +Step 22760 | loss:0.4838540554046631 lr:3.070819822784323e-05 tokens_per_second_per_gpu:4577.730425603186 mfu:21.59393545891927 +Step 22761 | loss:0.4004044532775879 lr:3.0704197843702456e-05 tokens_per_second_per_gpu:3885.42811964911 mfu:18.328227362780456 +Step 22762 | loss:0.3371294438838959 lr:3.0700197604687856e-05 tokens_per_second_per_gpu:4168.780293803974 mfu:19.664847913135024 +Step 22763 | loss:0.3036361336708069 lr:3.069619751082953e-05 tokens_per_second_per_gpu:4736.7189732366805 mfu:22.343911564349856 +Step 22764 | loss:0.42963436245918274 lr:3.069219756215757e-05 tokens_per_second_per_gpu:4115.910288920156 mfu:19.41545107954505 +Step 22765 | loss:0.5011053085327148 lr:3.068819775870205e-05 tokens_per_second_per_gpu:4792.250207355839 mfu:22.60586187029546 +Step 22766 | loss:0.4322005808353424 lr:3.068419810049307e-05 tokens_per_second_per_gpu:5326.806339583679 mfu:25.127454350695064 +Step 22767 | loss:0.44303104281425476 lr:3.068019858756068e-05 tokens_per_second_per_gpu:3694.8861640724585 mfu:17.429408448515463 +Step 22768 | loss:0.44154492020606995 lr:3.067619921993501e-05 tokens_per_second_per_gpu:5361.662243527297 mfu:25.291875596627563 +Step 22769 | loss:0.33635640144348145 lr:3.06721999976461e-05 tokens_per_second_per_gpu:5329.897926597924 mfu:25.1420378941204 +Step 22770 | loss:0.3400513231754303 lr:3.066820092072404e-05 tokens_per_second_per_gpu:3968.8106101840626 mfu:18.721556797153703 +Step 22771 | loss:0.5083785057067871 lr:3.0664201989198896e-05 tokens_per_second_per_gpu:4217.569969589596 mfu:19.894997138192878 +Step 22772 | loss:0.44754302501678467 lr:3.066020320310077e-05 tokens_per_second_per_gpu:4322.544578374213 mfu:20.39018027834505 +Step 22773 | loss:0.43685752153396606 lr:3.065620456245971e-05 tokens_per_second_per_gpu:4750.769340846869 mfu:22.41018954560703 +Step 22774 | loss:0.3322029709815979 lr:3.065220606730581e-05 tokens_per_second_per_gpu:5238.078669271668 mfu:24.708910791331725 +Step 22775 | loss:0.4267677962779999 lr:3.064820771766915e-05 tokens_per_second_per_gpu:5336.465648333719 mfu:25.173018958117023 +Step 22776 | loss:0.4280198812484741 lr:3.0644209513579784e-05 tokens_per_second_per_gpu:4741.92385282769 mfu:22.36846386942468 +Step 22777 | loss:0.32164594531059265 lr:3.064021145506779e-05 tokens_per_second_per_gpu:5325.584732807599 mfu:25.121691823104687 +Step 22778 | loss:0.4023609459400177 lr:3.0636213542163217e-05 tokens_per_second_per_gpu:5289.606473237196 mfu:24.951976234187033 +Step 22779 | loss:0.3877542316913605 lr:3.063221577489617e-05 tokens_per_second_per_gpu:3671.9149852429664 mfu:17.321049478689496 +Step 22780 | loss:0.442788690328598 lr:3.062821815329668e-05 tokens_per_second_per_gpu:5079.302465560213 mfu:23.959936348411006 +Step 22781 | loss:0.34427616000175476 lr:3.062422067739485e-05 tokens_per_second_per_gpu:3874.9012166419056 mfu:18.278570165220614 +Step 22782 | loss:0.3983912467956543 lr:3.0620223347220725e-05 tokens_per_second_per_gpu:4248.653189731372 mfu:20.041621990945853 +Step 22783 | loss:0.4480685591697693 lr:3.061622616280439e-05 tokens_per_second_per_gpu:4161.193193651131 mfu:19.629058267221293 +Step 22784 | loss:0.4031856060028076 lr:3.061222912417587e-05 tokens_per_second_per_gpu:3934.1202883544925 mfu:18.557916630304394 +Step 22785 | loss:0.32301995158195496 lr:3.060823223136525e-05 tokens_per_second_per_gpu:5310.931522226138 mfu:25.052570128696413 +Step 22786 | loss:0.34229761362075806 lr:3.06042354844026e-05 tokens_per_second_per_gpu:5362.941401561903 mfu:25.297909603323667 +Step 22787 | loss:0.48525458574295044 lr:3.060023888331796e-05 tokens_per_second_per_gpu:5308.819362850296 mfu:25.042606712549016 +Step 22788 | loss:0.46008139848709106 lr:3.0596242428141394e-05 tokens_per_second_per_gpu:4654.818034712296 mfu:21.957570863589297 +Step 22789 | loss:0.3739222586154938 lr:3.059224611890297e-05 tokens_per_second_per_gpu:5377.823506746686 mfu:25.368111032629262 +Step 22790 | loss:0.36685505509376526 lr:3.058824995563276e-05 tokens_per_second_per_gpu:5317.879896308218 mfu:25.08534679475645 +Step 22791 | loss:0.4656527042388916 lr:3.0584253938360774e-05 tokens_per_second_per_gpu:4416.24350255975 mfu:20.832174090413186 +Step 22792 | loss:0.43433672189712524 lr:3.058025806711711e-05 tokens_per_second_per_gpu:3777.637556993029 mfu:17.819760887765316 +Step 22793 | loss:0.37922656536102295 lr:3.057626234193179e-05 tokens_per_second_per_gpu:5297.723482863874 mfu:24.99026555350072 +Step 22794 | loss:0.463640958070755 lr:3.057226676283487e-05 tokens_per_second_per_gpu:3993.4282900025487 mfu:18.83768259306679 +Step 22795 | loss:0.42379671335220337 lr:3.0568271329856426e-05 tokens_per_second_per_gpu:3090.828450985781 mfu:14.579965153012246 +Step 22796 | loss:0.3547828197479248 lr:3.056427604302648e-05 tokens_per_second_per_gpu:5356.156093373786 mfu:25.26590214690736 +Step 22797 | loss:0.3689691722393036 lr:3.0560280902375124e-05 tokens_per_second_per_gpu:5150.157761443819 mfu:24.294172868254048 +Step 22798 | loss:0.3428085148334503 lr:3.055628590793235e-05 tokens_per_second_per_gpu:4696.614512098385 mfu:22.15473197863387 +Step 22799 | loss:0.3255513310432434 lr:3.0552291059728234e-05 tokens_per_second_per_gpu:3786.127754513281 mfu:17.859810598045417 +Step 22800 | loss:0.40819138288497925 lr:3.054829635779282e-05 tokens_per_second_per_gpu:4412.532448572497 mfu:20.81466841558421 +Step 22801 | loss:0.4186006188392639 lr:3.054430180215615e-05 tokens_per_second_per_gpu:4330.535947316726 mfu:20.427876929115023 +Step 22802 | loss:0.4662346839904785 lr:3.0540307392848266e-05 tokens_per_second_per_gpu:4219.987620238125 mfu:19.906401608795722 +Step 22803 | loss:0.3782500922679901 lr:3.053631312989921e-05 tokens_per_second_per_gpu:5068.235875684536 mfu:23.9077333558125 +Step 22804 | loss:0.46070754528045654 lr:3.053231901333905e-05 tokens_per_second_per_gpu:5321.913322421877 mfu:25.104373153908032 +Step 22805 | loss:0.3391273319721222 lr:3.0528325043197785e-05 tokens_per_second_per_gpu:5328.241059234955 mfu:25.13422216807855 +Step 22806 | loss:0.3924426734447479 lr:3.052433121950548e-05 tokens_per_second_per_gpu:4148.846561782333 mfu:19.570817098143554 +Step 22807 | loss:0.4413210153579712 lr:3.0520337542292144e-05 tokens_per_second_per_gpu:5335.988868177263 mfu:25.17076990477193 +Step 22808 | loss:0.4098384380340576 lr:3.051634401158785e-05 tokens_per_second_per_gpu:5324.768199423928 mfu:25.117840095818952 +Step 22809 | loss:0.5296558141708374 lr:3.0512350627422626e-05 tokens_per_second_per_gpu:3862.6044944471355 mfu:18.220564428590734 +Step 22810 | loss:0.47316861152648926 lr:3.050835738982648e-05 tokens_per_second_per_gpu:4512.684219627708 mfu:21.28710140730581 +Step 22811 | loss:0.4063516855239868 lr:3.050436429882948e-05 tokens_per_second_per_gpu:5344.810931863655 mfu:25.212385084379925 +Step 22812 | loss:0.48054975271224976 lr:3.0500371354461653e-05 tokens_per_second_per_gpu:4756.110198763597 mfu:22.435383283644555 +Step 22813 | loss:0.36347565054893494 lr:3.0496378556753013e-05 tokens_per_second_per_gpu:3535.5908225829858 mfu:16.677985144121795 +Step 22814 | loss:0.38739460706710815 lr:3.0492385905733583e-05 tokens_per_second_per_gpu:4522.150081945201 mfu:21.331753494900074 +Step 22815 | loss:0.42136719822883606 lr:3.0488393401433425e-05 tokens_per_second_per_gpu:3227.2190908968296 mfu:15.223343072115423 +Step 22816 | loss:0.49080008268356323 lr:3.0484401043882532e-05 tokens_per_second_per_gpu:5024.550532712971 mfu:23.701662139527542 +Step 22817 | loss:0.46888861060142517 lr:3.0480408833110964e-05 tokens_per_second_per_gpu:4828.956246388991 mfu:22.779010519110884 +Step 22818 | loss:0.35386115312576294 lr:3.047641676914873e-05 tokens_per_second_per_gpu:2798.86100062223 mfu:13.202704874863608 +Step 22819 | loss:0.4446503520011902 lr:3.047242485202587e-05 tokens_per_second_per_gpu:5353.562977381972 mfu:25.253669976342767 +Step 22820 | loss:0.44671913981437683 lr:3.0468433081772372e-05 tokens_per_second_per_gpu:4112.9243516801625 mfu:19.401365904130575 +Step 22821 | loss:0.3870581090450287 lr:3.046444145841829e-05 tokens_per_second_per_gpu:4109.628340869228 mfu:19.385818058777964 +Step 22822 | loss:0.6032959818840027 lr:3.0460449981993632e-05 tokens_per_second_per_gpu:4695.696674465261 mfu:22.150402381919175 +Step 22823 | loss:0.4144311249256134 lr:3.045645865252842e-05 tokens_per_second_per_gpu:3884.9447424826158 mfu:18.32594719021345 +Step 22824 | loss:0.5539618134498596 lr:3.0452467470052687e-05 tokens_per_second_per_gpu:4052.7542956359284 mfu:19.117533483699763 +Step 22825 | loss:0.3901960849761963 lr:3.044847643459644e-05 tokens_per_second_per_gpu:5057.8745059267785 mfu:23.858857006833183 +Step 22826 | loss:0.5312908291816711 lr:3.0444485546189706e-05 tokens_per_second_per_gpu:4031.9756125596746 mfu:19.019516890420167 +Step 22827 | loss:0.3223772644996643 lr:3.0440494804862468e-05 tokens_per_second_per_gpu:4753.330510701543 mfu:22.422271020792163 +Step 22828 | loss:0.4826088845729828 lr:3.0436504210644778e-05 tokens_per_second_per_gpu:5102.988157826812 mfu:24.071665800028367 +Step 22829 | loss:0.40313178300857544 lr:3.0432513763566635e-05 tokens_per_second_per_gpu:4013.704777709359 mfu:18.933330245105356 +Step 22830 | loss:0.3510594666004181 lr:3.0428523463658044e-05 tokens_per_second_per_gpu:5258.963237885796 mfu:24.807426864759407 +Step 22831 | loss:0.44202330708503723 lr:3.0424533310949034e-05 tokens_per_second_per_gpu:5342.998958926509 mfu:25.203837698880942 +Step 22832 | loss:0.30563846230506897 lr:3.0420543305469597e-05 tokens_per_second_per_gpu:3595.658388509499 mfu:16.961334101180707 +Step 22833 | loss:0.41121599078178406 lr:3.041655344724978e-05 tokens_per_second_per_gpu:5349.401899592421 mfu:25.234041462456375 +Step 22834 | loss:0.36562034487724304 lr:3.0412563736319534e-05 tokens_per_second_per_gpu:5315.137556918998 mfu:25.072410711966146 +Step 22835 | loss:0.44596076011657715 lr:3.0408574172708904e-05 tokens_per_second_per_gpu:4523.071640014925 mfu:21.336100641548978 +Step 22836 | loss:0.311899870634079 lr:3.0404584756447886e-05 tokens_per_second_per_gpu:4252.610501307717 mfu:20.060289304838445 +Step 22837 | loss:0.40777531266212463 lr:3.0400595487566484e-05 tokens_per_second_per_gpu:4803.650079955371 mfu:22.659637014366286 +Step 22838 | loss:0.4508996903896332 lr:3.0396606366094717e-05 tokens_per_second_per_gpu:5135.290080273872 mfu:24.224039479487427 +Step 22839 | loss:0.4934489130973816 lr:3.039261739206256e-05 tokens_per_second_per_gpu:3818.1339928865823 mfu:18.010789485279645 +Step 22840 | loss:0.5992917418479919 lr:3.0388628565500045e-05 tokens_per_second_per_gpu:5341.838539159204 mfu:25.198363800850537 +Step 22841 | loss:0.4794362783432007 lr:3.038463988643715e-05 tokens_per_second_per_gpu:4561.657013807668 mfu:21.518114433072967 +Step 22842 | loss:0.35561442375183105 lr:3.0380651354903882e-05 tokens_per_second_per_gpu:5365.7364237490865 mfu:25.31109419986057 +Step 22843 | loss:0.3840023875236511 lr:3.037666297093023e-05 tokens_per_second_per_gpu:4051.70484007811 mfu:19.11258302277795 +Step 22844 | loss:0.399395227432251 lr:3.037267473454621e-05 tokens_per_second_per_gpu:4773.174397638722 mfu:22.515878021191075 +Step 22845 | loss:0.4041990339756012 lr:3.0368686645781806e-05 tokens_per_second_per_gpu:4061.1677681410806 mfu:19.157221268992995 +Step 22846 | loss:0.38908833265304565 lr:3.0364698704667006e-05 tokens_per_second_per_gpu:5374.203007597587 mfu:25.351032520422113 +Step 22847 | loss:0.38433408737182617 lr:3.0360710911231817e-05 tokens_per_second_per_gpu:5360.298096119432 mfu:25.285440680557116 +Step 22848 | loss:0.4804915487766266 lr:3.035672326550624e-05 tokens_per_second_per_gpu:5176.9541048566325 mfu:24.420575792059914 +Step 22849 | loss:0.46601739525794983 lr:3.0352735767520247e-05 tokens_per_second_per_gpu:3016.61835400075 mfu:14.229903463985693 +Step 22850 | loss:0.34738653898239136 lr:3.0348748417303823e-05 tokens_per_second_per_gpu:4255.306616756371 mfu:20.073007341414478 +Step 22851 | loss:0.4006619453430176 lr:3.0344761214886985e-05 tokens_per_second_per_gpu:4958.605155629704 mfu:23.390586544384227 +Step 22852 | loss:0.3774506449699402 lr:3.034077416029969e-05 tokens_per_second_per_gpu:3250.778246325202 mfu:15.334475627877461 +Step 22853 | loss:0.44703370332717896 lr:3.0336787253571952e-05 tokens_per_second_per_gpu:3294.85156118218 mfu:15.542376973740746 +Step 22854 | loss:0.3791941702365875 lr:3.0332800494733742e-05 tokens_per_second_per_gpu:4845.503849136628 mfu:22.85706838458365 +Step 22855 | loss:0.4091414213180542 lr:3.0328813883815067e-05 tokens_per_second_per_gpu:4548.9322180215795 mfu:21.458089400275956 +Step 22856 | loss:0.4443403482437134 lr:3.032482742084587e-05 tokens_per_second_per_gpu:3868.9426720142683 mfu:18.250462693579504 +Step 22857 | loss:0.42916008830070496 lr:3.0320841105856162e-05 tokens_per_second_per_gpu:4548.614252513419 mfu:21.456589502723453 +Step 22858 | loss:0.3451153337955475 lr:3.031685493887593e-05 tokens_per_second_per_gpu:5326.665058505467 mfu:25.126787903744102 +Step 22859 | loss:0.49464449286460876 lr:3.0312868919935123e-05 tokens_per_second_per_gpu:3859.4956128801887 mfu:18.205899303809556 +Step 22860 | loss:0.5020871758460999 lr:3.030888304906375e-05 tokens_per_second_per_gpu:4828.764239547586 mfu:22.77810479008023 +Step 22861 | loss:0.38440924882888794 lr:3.0304897326291786e-05 tokens_per_second_per_gpu:4663.915982529833 mfu:22.00048743571458 +Step 22862 | loss:0.32174256443977356 lr:3.0300911751649207e-05 tokens_per_second_per_gpu:3683.517084757632 mfu:17.37577856162247 +Step 22863 | loss:0.3833705186843872 lr:3.0296926325165976e-05 tokens_per_second_per_gpu:5333.537986917184 mfu:25.15920867970468 +Step 22864 | loss:0.42232099175453186 lr:3.0292941046872074e-05 tokens_per_second_per_gpu:5273.247842899996 mfu:24.874809783816314 +Step 22865 | loss:0.37747809290885925 lr:3.028895591679748e-05 tokens_per_second_per_gpu:4779.768268531281 mfu:22.546982435221654 +Step 22866 | loss:0.48742878437042236 lr:3.0284970934972155e-05 tokens_per_second_per_gpu:5321.515129580715 mfu:25.102494810337568 +Step 22867 | loss:0.3442681133747101 lr:3.028098610142609e-05 tokens_per_second_per_gpu:4831.244983423857 mfu:22.789806882203155 +Step 22868 | loss:0.497890830039978 lr:3.0277001416189233e-05 tokens_per_second_per_gpu:4979.201641349411 mfu:23.487743681646915 +Step 22869 | loss:0.585996687412262 lr:3.0273016879291597e-05 tokens_per_second_per_gpu:4266.805871829517 mfu:20.127251289569738 +Step 22870 | loss:0.4549729526042938 lr:3.0269032490763095e-05 tokens_per_second_per_gpu:3516.4426467845524 mfu:16.58765993186491 +Step 22871 | loss:0.4378480911254883 lr:3.0265048250633716e-05 tokens_per_second_per_gpu:4712.617545378962 mfu:22.230221017017378 +Step 22872 | loss:0.3890188932418823 lr:3.0261064158933416e-05 tokens_per_second_per_gpu:5333.538905375364 mfu:25.15921301222862 +Step 22873 | loss:0.4036524295806885 lr:3.025708021569219e-05 tokens_per_second_per_gpu:4333.9442488626055 mfu:20.44395446902314 +Step 22874 | loss:0.36960965394973755 lr:3.025309642093998e-05 tokens_per_second_per_gpu:5331.845532734217 mfu:25.151225084561165 +Step 22875 | loss:0.44138747453689575 lr:3.024911277470674e-05 tokens_per_second_per_gpu:5309.019830497572 mfu:25.04355235264452 +Step 22876 | loss:0.4843023419380188 lr:3.0245129277022456e-05 tokens_per_second_per_gpu:4778.906034743677 mfu:22.542915131333274 +Step 22877 | loss:0.5097467303276062 lr:3.024114592791708e-05 tokens_per_second_per_gpu:5233.562689534636 mfu:24.687608144406546 +Step 22878 | loss:0.3546220362186432 lr:3.0237162727420564e-05 tokens_per_second_per_gpu:3457.546952252942 mfu:16.309838892117632 +Step 22879 | loss:0.5297129154205322 lr:3.023317967556285e-05 tokens_per_second_per_gpu:3933.4736544794787 mfu:18.55486634798884 +Step 22880 | loss:0.5053430199623108 lr:3.022919677237393e-05 tokens_per_second_per_gpu:5378.194720548173 mfu:25.369862111466478 +Step 22881 | loss:0.39246752858161926 lr:3.022521401788374e-05 tokens_per_second_per_gpu:3603.2191115904643 mfu:16.996999321945996 +Step 22882 | loss:0.42820796370506287 lr:3.022123141212223e-05 tokens_per_second_per_gpu:5346.301598214223 mfu:25.219416811889456 +Step 22883 | loss:0.4379061162471771 lr:3.0217248955119377e-05 tokens_per_second_per_gpu:5331.3145705606785 mfu:25.148720445397096 +Step 22884 | loss:0.49261993169784546 lr:3.0213266646905125e-05 tokens_per_second_per_gpu:4117.8921454416895 mfu:19.424799834896717 +Step 22885 | loss:0.3387722373008728 lr:3.0209284487509408e-05 tokens_per_second_per_gpu:5356.976540315154 mfu:25.26977233511248 +Step 22886 | loss:0.3606197237968445 lr:3.020530247696218e-05 tokens_per_second_per_gpu:5050.114807198402 mfu:23.82225318399031 +Step 22887 | loss:0.41715487837791443 lr:3.020132061529341e-05 tokens_per_second_per_gpu:4219.386887151551 mfu:19.903567848330717 +Step 22888 | loss:0.44589126110076904 lr:3.0197338902533023e-05 tokens_per_second_per_gpu:5246.1049728155 mfu:24.74677223457658 +Step 22889 | loss:0.38064274191856384 lr:3.0193357338710983e-05 tokens_per_second_per_gpu:4774.307341258762 mfu:22.52122230954691 +Step 22890 | loss:0.4017459452152252 lr:3.0189375923857234e-05 tokens_per_second_per_gpu:5318.683252700576 mfu:25.089136363925938 +Step 22891 | loss:0.3326951563358307 lr:3.018539465800172e-05 tokens_per_second_per_gpu:4654.201165214348 mfu:21.95466098491869 +Step 22892 | loss:0.3455806076526642 lr:3.018141354117437e-05 tokens_per_second_per_gpu:4175.6782785022915 mfu:19.697386883874536 +Step 22893 | loss:0.3427245318889618 lr:3.017743257340514e-05 tokens_per_second_per_gpu:4511.76791553871 mfu:21.28277904458041 +Step 22894 | loss:0.4110446870326996 lr:3.0173451754723975e-05 tokens_per_second_per_gpu:4250.162891177448 mfu:20.04874351024862 +Step 22895 | loss:0.45241010189056396 lr:3.0169471085160795e-05 tokens_per_second_per_gpu:4956.802945562713 mfu:23.382085212009486 +Step 22896 | loss:0.5157610774040222 lr:3.0165490564745567e-05 tokens_per_second_per_gpu:4799.018506424304 mfu:22.637789091792307 +Step 22897 | loss:0.33766689896583557 lr:3.0161510193508214e-05 tokens_per_second_per_gpu:5256.63263113409 mfu:24.796432995069093 +Step 22898 | loss:0.3456263840198517 lr:3.0157529971478682e-05 tokens_per_second_per_gpu:3322.197413775079 mfu:15.671372026105013 +Step 22899 | loss:0.3565574586391449 lr:3.0153549898686882e-05 tokens_per_second_per_gpu:5214.867574644104 mfu:24.599420097752397 +Step 22900 | loss:0.46631282567977905 lr:3.0149569975162773e-05 tokens_per_second_per_gpu:4584.801471208502 mfu:21.62729078748433 +Step 22901 | loss:0.4275485873222351 lr:3.014559020093628e-05 tokens_per_second_per_gpu:4270.505707177304 mfu:20.144704044162253 +Step 22902 | loss:0.3961377739906311 lr:3.0141610576037326e-05 tokens_per_second_per_gpu:4890.840921105175 mfu:23.070931088361398 +Step 22903 | loss:0.4238448739051819 lr:3.0137631100495867e-05 tokens_per_second_per_gpu:5354.489784446905 mfu:25.25804188339765 +Step 22904 | loss:0.4616644084453583 lr:3.0133651774341808e-05 tokens_per_second_per_gpu:5225.58344173471 mfu:24.649968671134236 +Step 22905 | loss:0.4898792803287506 lr:3.0129672597605103e-05 tokens_per_second_per_gpu:3994.8659743134285 mfu:18.84446439525545 +Step 22906 | loss:0.3708188235759735 lr:3.0125693570315672e-05 tokens_per_second_per_gpu:5328.3981973134 mfu:25.134963415205096 +Step 22907 | loss:0.47201451659202576 lr:3.0121714692503422e-05 tokens_per_second_per_gpu:3919.114273478552 mfu:18.487130697844435 +Step 22908 | loss:0.40029677748680115 lr:3.0117735964198285e-05 tokens_per_second_per_gpu:4427.868618455042 mfu:20.88701174554058 +Step 22909 | loss:0.3045494556427002 lr:3.011375738543021e-05 tokens_per_second_per_gpu:5335.626647190986 mfu:25.16906124657839 +Step 22910 | loss:0.49342527985572815 lr:3.010977895622909e-05 tokens_per_second_per_gpu:3937.1527580790707 mfu:18.572221307387917 +Step 22911 | loss:0.40723398327827454 lr:3.0105800676624856e-05 tokens_per_second_per_gpu:4973.065570150901 mfu:23.45879878688184 +Step 22912 | loss:0.4680415689945221 lr:3.0101822546647445e-05 tokens_per_second_per_gpu:3688.5384821019097 mfu:17.399465349634163 +Step 22913 | loss:0.43163084983825684 lr:3.0097844566326776e-05 tokens_per_second_per_gpu:4858.014676301834 mfu:22.91608409088915 +Step 22914 | loss:0.49577224254608154 lr:3.0093866735692743e-05 tokens_per_second_per_gpu:4392.301588047429 mfu:20.719235994746725 +Step 22915 | loss:0.4458409249782562 lr:3.0089889054775277e-05 tokens_per_second_per_gpu:5155.183011409449 mfu:24.317877829737977 +Step 22916 | loss:0.38972774147987366 lr:3.00859115236043e-05 tokens_per_second_per_gpu:5038.699187842 mfu:23.76840375978066 +Step 22917 | loss:0.3349088728427887 lr:3.008193414220972e-05 tokens_per_second_per_gpu:5337.050719865131 mfu:25.175778840354187 +Step 22918 | loss:0.4424402713775635 lr:3.007795691062145e-05 tokens_per_second_per_gpu:4967.012643752725 mfu:23.43024610032591 +Step 22919 | loss:0.41230508685112 lr:3.007397982886942e-05 tokens_per_second_per_gpu:4008.7242671043846 mfu:18.909836326818958 +Step 22920 | loss:0.5273115038871765 lr:3.007000289698354e-05 tokens_per_second_per_gpu:4340.616717576228 mfu:20.47542964238585 +Step 22921 | loss:0.4874895215034485 lr:3.0066026114993695e-05 tokens_per_second_per_gpu:4592.334046462891 mfu:21.662823230149144 +Step 22922 | loss:0.5059206485748291 lr:3.006204948292981e-05 tokens_per_second_per_gpu:4402.257088497469 mfu:20.766197788953097 +Step 22923 | loss:0.4280873239040375 lr:3.0058073000821797e-05 tokens_per_second_per_gpu:5331.041343262366 mfu:25.147431585613575 +Step 22924 | loss:0.4439176321029663 lr:3.0054096668699555e-05 tokens_per_second_per_gpu:4065.036334801146 mfu:19.17546995797406 +Step 22925 | loss:0.35331910848617554 lr:3.0050120486593008e-05 tokens_per_second_per_gpu:4636.423691728312 mfu:21.870801609335313 +Step 22926 | loss:0.37973514199256897 lr:3.004614445453205e-05 tokens_per_second_per_gpu:4783.787753189654 mfu:22.56594303015 +Step 22927 | loss:0.42564356327056885 lr:3.0042168572546592e-05 tokens_per_second_per_gpu:5012.880147757863 mfu:23.6466109425213 +Step 22928 | loss:0.37584149837493896 lr:3.0038192840666513e-05 tokens_per_second_per_gpu:3286.659147320226 mfu:15.503731959783972 +Step 22929 | loss:0.36180636286735535 lr:3.0034217258921747e-05 tokens_per_second_per_gpu:5342.48588963286 mfu:25.201417463484148 +Step 22930 | loss:0.46161532402038574 lr:3.0030241827342177e-05 tokens_per_second_per_gpu:4224.384575716868 mfu:19.927142797986917 +Step 22931 | loss:0.36440643668174744 lr:3.0026266545957693e-05 tokens_per_second_per_gpu:4736.874815905466 mfu:22.34464670080802 +Step 22932 | loss:0.39946281909942627 lr:3.0022291414798222e-05 tokens_per_second_per_gpu:5344.042477597108 mfu:25.2087601544928 +Step 22933 | loss:0.4642087519168854 lr:3.0018316433893644e-05 tokens_per_second_per_gpu:4499.509338581137 mfu:21.224953245542633 +Step 22934 | loss:0.4555092751979828 lr:3.0014341603273854e-05 tokens_per_second_per_gpu:5307.233791419182 mfu:25.035127301582072 +Step 22935 | loss:0.3998037278652191 lr:3.0010366922968763e-05 tokens_per_second_per_gpu:5186.246982848638 mfu:24.46441188307669 +Step 22936 | loss:0.4941772520542145 lr:3.0006392393008244e-05 tokens_per_second_per_gpu:4483.610365699979 mfu:21.14995507781772 +Step 22937 | loss:0.4160148799419403 lr:3.0002418013422208e-05 tokens_per_second_per_gpu:4207.913184962625 mfu:19.849444437490096 +Step 22938 | loss:0.4509299397468567 lr:2.9998443784240516e-05 tokens_per_second_per_gpu:4789.81825538456 mfu:22.594389937913146 +Step 22939 | loss:0.5825589299201965 lr:2.9994469705493095e-05 tokens_per_second_per_gpu:3841.4106831053577 mfu:18.12058960445448 +Step 22940 | loss:0.4934728145599365 lr:2.999049577720981e-05 tokens_per_second_per_gpu:4151.7383025399895 mfu:19.58445793267927 +Step 22941 | loss:0.41243410110473633 lr:2.998652199942057e-05 tokens_per_second_per_gpu:4134.621212474818 mfu:19.503713698365143 +Step 22942 | loss:0.47902101278305054 lr:2.9982548372155263e-05 tokens_per_second_per_gpu:3287.2370296426625 mfu:15.506457929295795 +Step 22943 | loss:0.5005319118499756 lr:2.997857489544375e-05 tokens_per_second_per_gpu:3980.97661420375 mfu:18.778945913848897 +Step 22944 | loss:0.4494413733482361 lr:2.9974601569315918e-05 tokens_per_second_per_gpu:5336.435111856808 mfu:25.17287491234537 +Step 22945 | loss:0.36887601017951965 lr:2.997062839380168e-05 tokens_per_second_per_gpu:4497.947378110389 mfu:21.21758521150418 +Step 22946 | loss:0.4732467830181122 lr:2.99666553689309e-05 tokens_per_second_per_gpu:5307.709750599771 mfu:25.037372482243914 +Step 22947 | loss:0.3972330391407013 lr:2.9962682494733453e-05 tokens_per_second_per_gpu:5371.198482321023 mfu:25.336859662067685 +Step 22948 | loss:0.43322673439979553 lr:2.9958709771239236e-05 tokens_per_second_per_gpu:3289.8303896078564 mfu:15.518691250724451 +Step 22949 | loss:0.4398425221443176 lr:2.9954737198478123e-05 tokens_per_second_per_gpu:3531.21198525927 mfu:16.657329421359144 +Step 22950 | loss:0.36962011456489563 lr:2.9950764776479988e-05 tokens_per_second_per_gpu:4507.170023313556 mfu:21.261090002473047 +Step 22951 | loss:0.3763534724712372 lr:2.9946792505274697e-05 tokens_per_second_per_gpu:5362.695869942846 mfu:25.29675138878456 +Step 22952 | loss:0.39185822010040283 lr:2.9942820384892146e-05 tokens_per_second_per_gpu:4275.382320326646 mfu:20.167707860423697 +Step 22953 | loss:0.3103068768978119 lr:2.993884841536221e-05 tokens_per_second_per_gpu:3707.8621597263223 mfu:17.49061843394789 +Step 22954 | loss:0.3484061062335968 lr:2.9934876596714733e-05 tokens_per_second_per_gpu:5370.399065010467 mfu:25.333088674219134 +Step 22955 | loss:0.5646620988845825 lr:2.993090492897963e-05 tokens_per_second_per_gpu:4238.270658139334 mfu:19.992645817983657 +Step 22956 | loss:0.4647091329097748 lr:2.9926933412186754e-05 tokens_per_second_per_gpu:5357.1240743411845 mfu:25.270468278285165 +Step 22957 | loss:0.3929491639137268 lr:2.9922962046365964e-05 tokens_per_second_per_gpu:3937.861834553645 mfu:18.575566142099404 +Step 22958 | loss:0.4434216320514679 lr:2.9918990831547134e-05 tokens_per_second_per_gpu:5114.98084325056 mfu:24.12823734333531 +Step 22959 | loss:0.5139546394348145 lr:2.9915019767760143e-05 tokens_per_second_per_gpu:5199.1414073713395 mfu:24.525237083565948 +Step 22960 | loss:0.429130494594574 lr:2.991104885503484e-05 tokens_per_second_per_gpu:4140.0779411390495 mfu:19.529454018488234 +Step 22961 | loss:0.4593290388584137 lr:2.990707809340112e-05 tokens_per_second_per_gpu:5313.82761391161 mfu:25.066231487300975 +Step 22962 | loss:0.49127745628356934 lr:2.9903107482888815e-05 tokens_per_second_per_gpu:5323.364376388291 mfu:25.111218023042944 +Step 22963 | loss:0.42038676142692566 lr:2.9899137023527825e-05 tokens_per_second_per_gpu:3936.8448966385336 mfu:18.57076907244632 +Step 22964 | loss:0.4631514251232147 lr:2.989516671534798e-05 tokens_per_second_per_gpu:5313.052830779063 mfu:25.062576703073116 +Step 22965 | loss:0.47660011053085327 lr:2.989119655837913e-05 tokens_per_second_per_gpu:5352.256680202857 mfu:25.2475079496716 +Step 22966 | loss:0.431067556142807 lr:2.9887226552651175e-05 tokens_per_second_per_gpu:5346.452740174461 mfu:25.220129774303377 +Step 22967 | loss:0.4426788091659546 lr:2.988325669819395e-05 tokens_per_second_per_gpu:4880.034134280894 mfu:23.019953631083307 +Step 22968 | loss:0.4146565794944763 lr:2.987928699503732e-05 tokens_per_second_per_gpu:3881.672578832764 mfu:18.310511835988418 +Step 22969 | loss:0.4846900701522827 lr:2.9875317443211143e-05 tokens_per_second_per_gpu:4879.809880336459 mfu:23.01889578696574 +Step 22970 | loss:0.3633020222187042 lr:2.987134804274526e-05 tokens_per_second_per_gpu:4909.758700129649 mfu:23.160169479725777 +Step 22971 | loss:0.5209866762161255 lr:2.9867378793669558e-05 tokens_per_second_per_gpu:4555.485519635337 mfu:21.489002442096602 +Step 22972 | loss:0.4725906550884247 lr:2.986340969601386e-05 tokens_per_second_per_gpu:4776.400514845507 mfu:22.53109616648783 +Step 22973 | loss:0.4325128495693207 lr:2.9859440749808032e-05 tokens_per_second_per_gpu:4072.9581498247467 mfu:19.212838511041394 +Step 22974 | loss:0.4227008521556854 lr:2.9855471955081903e-05 tokens_per_second_per_gpu:4634.357896796403 mfu:21.86105690218054 +Step 22975 | loss:0.3210710287094116 lr:2.9851503311865346e-05 tokens_per_second_per_gpu:5403.783970586807 mfu:25.49057096987498 +Step 22976 | loss:0.4728558659553528 lr:2.9847534820188205e-05 tokens_per_second_per_gpu:5211.436954870126 mfu:24.58323727895445 +Step 22977 | loss:0.41795384883880615 lr:2.9843566480080325e-05 tokens_per_second_per_gpu:4492.136534062328 mfu:21.190174468697247 +Step 22978 | loss:0.4110874533653259 lr:2.983959829157157e-05 tokens_per_second_per_gpu:4829.255516422647 mfu:22.780422226920372 +Step 22979 | loss:0.40089288353919983 lr:2.983563025469176e-05 tokens_per_second_per_gpu:5051.575703509972 mfu:23.8291444811466 +Step 22980 | loss:0.3263704478740692 lr:2.9831662369470737e-05 tokens_per_second_per_gpu:5360.636959071283 mfu:25.28703915491698 +Step 22981 | loss:0.46474170684814453 lr:2.9827694635938357e-05 tokens_per_second_per_gpu:4976.3686997436735 mfu:23.474380212742968 +Step 22982 | loss:0.3363538086414337 lr:2.9823727054124468e-05 tokens_per_second_per_gpu:5296.705350407227 mfu:24.98546284899141 +Step 22983 | loss:0.42835667729377747 lr:2.9819759624058886e-05 tokens_per_second_per_gpu:5353.188840531311 mfu:25.25190510898387 +Step 22984 | loss:0.4770027697086334 lr:2.981579234577149e-05 tokens_per_second_per_gpu:2989.7216116678155 mfu:14.10302694134379 +Step 22985 | loss:0.493368536233902 lr:2.981182521929209e-05 tokens_per_second_per_gpu:4710.3762164071 mfu:22.21964828584695 +Step 22986 | loss:0.47002506256103516 lr:2.9807858244650523e-05 tokens_per_second_per_gpu:5309.956727194682 mfu:25.04797185421585 +Step 22987 | loss:0.3902443051338196 lr:2.9803891421876624e-05 tokens_per_second_per_gpu:4361.240683148591 mfu:20.572716406802062 +Step 22988 | loss:0.39635327458381653 lr:2.9799924751000243e-05 tokens_per_second_per_gpu:3845.30776660238 mfu:18.138972812220942 +Step 22989 | loss:0.3816445767879486 lr:2.97959582320512e-05 tokens_per_second_per_gpu:5169.598228450923 mfu:24.38587686801225 +Step 22990 | loss:0.43195703625679016 lr:2.9791991865059334e-05 tokens_per_second_per_gpu:4137.567031596102 mfu:19.51760962976881 +Step 22991 | loss:0.4256965219974518 lr:2.978802565005448e-05 tokens_per_second_per_gpu:4560.670529582924 mfu:21.51346101867329 +Step 22992 | loss:0.3470820188522339 lr:2.978405958706647e-05 tokens_per_second_per_gpu:4283.9020935591925 mfu:20.20789708438483 +Step 22993 | loss:0.3187634348869324 lr:2.9780093676125125e-05 tokens_per_second_per_gpu:2995.8756313314625 mfu:14.132056502081237 +Step 22994 | loss:0.4268400967121124 lr:2.9776127917260264e-05 tokens_per_second_per_gpu:4698.021263052235 mfu:22.161367862899642 +Step 22995 | loss:0.4084470272064209 lr:2.9772162310501735e-05 tokens_per_second_per_gpu:4868.0385417889665 mfu:22.96336837463976 +Step 22996 | loss:0.42531687021255493 lr:2.976819685587934e-05 tokens_per_second_per_gpu:5338.45181031136 mfu:25.182388023039728 +Step 22997 | loss:0.32846543192863464 lr:2.9764231553422932e-05 tokens_per_second_per_gpu:5311.714438662214 mfu:25.056263279104698 +Step 22998 | loss:0.4297955632209778 lr:2.9760266403162317e-05 tokens_per_second_per_gpu:5316.009018391844 mfu:25.076521544420224 +Step 22999 | loss:0.47171494364738464 lr:2.9756301405127318e-05 tokens_per_second_per_gpu:3547.853071271414 mfu:16.735828263340476 +Step 23000 | loss:0.32765498757362366 lr:2.9752336559347783e-05 tokens_per_second_per_gpu:3589.0777558073128 mfu:16.93029213395278 +Step 23001 | loss:0.38122445344924927 lr:2.9748371865853486e-05 tokens_per_second_per_gpu:3060.5720808017004 mfu:14.43724069258487 +Step 23002 | loss:0.48793303966522217 lr:2.9744407324674273e-05 tokens_per_second_per_gpu:5365.581885373464 mfu:25.310365215975793 +Step 23003 | loss:0.4679444134235382 lr:2.974044293583994e-05 tokens_per_second_per_gpu:4523.718530912915 mfu:21.33915213628577 +Step 23004 | loss:0.3898192346096039 lr:2.9736478699380345e-05 tokens_per_second_per_gpu:3900.4770857630892 mfu:18.399215903558478 +Step 23005 | loss:0.3702026307582855 lr:2.9732514615325267e-05 tokens_per_second_per_gpu:4430.293378820242 mfu:20.898449753890002 +Step 23006 | loss:0.41797977685928345 lr:2.9728550683704526e-05 tokens_per_second_per_gpu:5228.298041395683 mfu:24.662773900893512 +Step 23007 | loss:0.4852125644683838 lr:2.972458690454797e-05 tokens_per_second_per_gpu:5326.028894482612 mfu:25.12378700950752 +Step 23008 | loss:0.4892180860042572 lr:2.972062327788536e-05 tokens_per_second_per_gpu:5051.837376281456 mfu:23.83037883625567 +Step 23009 | loss:0.47814175486564636 lr:2.971665980374654e-05 tokens_per_second_per_gpu:5267.741084233824 mfu:24.848833463638076 +Step 23010 | loss:0.41067495942115784 lr:2.9712696482161294e-05 tokens_per_second_per_gpu:4659.363866013614 mfu:21.979014325436413 +Step 23011 | loss:0.42249685525894165 lr:2.9708733313159464e-05 tokens_per_second_per_gpu:4320.143578892521 mfu:20.378854354137154 +Step 23012 | loss:0.48549631237983704 lr:2.9704770296770822e-05 tokens_per_second_per_gpu:4836.220338327306 mfu:22.813276480165804 +Step 23013 | loss:0.32377737760543823 lr:2.9700807433025206e-05 tokens_per_second_per_gpu:5228.965092598787 mfu:24.665920495229226 +Step 23014 | loss:0.4625922739505768 lr:2.9696844721952422e-05 tokens_per_second_per_gpu:5345.987317651965 mfu:25.21793429685549 +Step 23015 | loss:0.326566606760025 lr:2.969288216358224e-05 tokens_per_second_per_gpu:5349.257435355602 mfu:25.233360000003163 +Step 23016 | loss:0.3866674602031708 lr:2.968891975794449e-05 tokens_per_second_per_gpu:5261.7606351453205 mfu:24.82062266491777 +Step 23017 | loss:0.4781663417816162 lr:2.968495750506895e-05 tokens_per_second_per_gpu:5324.296502672771 mfu:25.11561502176398 +Step 23018 | loss:0.4666287899017334 lr:2.968099540498545e-05 tokens_per_second_per_gpu:4875.582860559192 mfu:22.99895621347243 +Step 23019 | loss:0.4124959111213684 lr:2.9677033457723768e-05 tokens_per_second_per_gpu:5345.199244426036 mfu:25.214216820989627 +Step 23020 | loss:0.5265257954597473 lr:2.9673071663313716e-05 tokens_per_second_per_gpu:5186.678158778325 mfu:24.46644581350322 +Step 23021 | loss:0.3647262454032898 lr:2.9669110021785095e-05 tokens_per_second_per_gpu:3873.0672962391104 mfu:18.269919249781545 +Step 23022 | loss:0.440828412771225 lr:2.966514853316769e-05 tokens_per_second_per_gpu:5203.788430809975 mfu:24.54715788598995 +Step 23023 | loss:0.42833974957466125 lr:2.966118719749128e-05 tokens_per_second_per_gpu:4145.342120277333 mfu:19.554286049645096 +Step 23024 | loss:0.4080302119255066 lr:2.9657226014785688e-05 tokens_per_second_per_gpu:4984.232844786625 mfu:23.511476726672072 +Step 23025 | loss:0.38078659772872925 lr:2.9653264985080698e-05 tokens_per_second_per_gpu:4633.753966861647 mfu:21.85820806163722 +Step 23026 | loss:0.3229473829269409 lr:2.964930410840609e-05 tokens_per_second_per_gpu:5317.76153218958 mfu:25.084788451013832 +Step 23027 | loss:0.4186006486415863 lr:2.964534338479167e-05 tokens_per_second_per_gpu:4800.400079465717 mfu:22.64430620754932 +Step 23028 | loss:0.48263418674468994 lr:2.9641382814267204e-05 tokens_per_second_per_gpu:3194.1179057519894 mfu:15.067199134142662 +Step 23029 | loss:0.376266747713089 lr:2.9637422396862534e-05 tokens_per_second_per_gpu:5325.544317256687 mfu:25.121501176055435 +Step 23030 | loss:0.5030519962310791 lr:2.963346213260737e-05 tokens_per_second_per_gpu:4221.008440808597 mfu:19.911216993596586 +Step 23031 | loss:0.4449932873249054 lr:2.962950202153155e-05 tokens_per_second_per_gpu:3880.8381756576246 mfu:18.30657581384742 +Step 23032 | loss:0.32755643129348755 lr:2.9625542063664847e-05 tokens_per_second_per_gpu:5272.130494611026 mfu:24.86953906129797 +Step 23033 | loss:0.46605777740478516 lr:2.9621582259037033e-05 tokens_per_second_per_gpu:3971.283615513116 mfu:18.733222385229066 +Step 23034 | loss:0.4774321913719177 lr:2.961762260767791e-05 tokens_per_second_per_gpu:4830.717512224099 mfu:22.787318710558026 +Step 23035 | loss:0.37485918402671814 lr:2.961366310961723e-05 tokens_per_second_per_gpu:4213.215599163643 mfu:19.874456829961453 +Step 23036 | loss:0.3787819743156433 lr:2.9609703764884826e-05 tokens_per_second_per_gpu:5334.347623826411 mfu:25.163027875143804 +Step 23037 | loss:0.34501054883003235 lr:2.9605744573510414e-05 tokens_per_second_per_gpu:4205.566916518943 mfu:19.838376688926576 +Step 23038 | loss:0.34571775794029236 lr:2.960178553552381e-05 tokens_per_second_per_gpu:4673.316693304099 mfu:22.044832192363447 +Step 23039 | loss:0.37265631556510925 lr:2.959782665095476e-05 tokens_per_second_per_gpu:3911.7046861711674 mfu:18.452178410309546 +Step 23040 | loss:0.3764382004737854 lr:2.9593867919833072e-05 tokens_per_second_per_gpu:4501.564833240107 mfu:21.234649364552546 +Step 23041 | loss:0.46408310532569885 lr:2.9589909342188514e-05 tokens_per_second_per_gpu:4356.818413140244 mfu:20.551855804655624 +Step 23042 | loss:0.3846590220928192 lr:2.9585950918050843e-05 tokens_per_second_per_gpu:5344.337565798662 mfu:25.21015213588677 +Step 23043 | loss:0.3476365804672241 lr:2.958199264744985e-05 tokens_per_second_per_gpu:5241.149001608465 mfu:24.72339407281711 +Step 23044 | loss:0.46607622504234314 lr:2.9578034530415298e-05 tokens_per_second_per_gpu:4295.806474496092 mfu:20.26405208036101 +Step 23045 | loss:0.4298664629459381 lr:2.957407656697695e-05 tokens_per_second_per_gpu:5308.5074027568935 mfu:25.041135143562556 +Step 23046 | loss:0.4274700880050659 lr:2.9570118757164567e-05 tokens_per_second_per_gpu:5225.904715224235 mfu:24.651484172999236 +Step 23047 | loss:0.4045593738555908 lr:2.9566161101007932e-05 tokens_per_second_per_gpu:5334.359761532928 mfu:25.163085130776352 +Step 23048 | loss:0.3605066239833832 lr:2.9562203598536802e-05 tokens_per_second_per_gpu:4083.573204374902 mfu:19.26291153446954 +Step 23049 | loss:0.4361313581466675 lr:2.9558246249780956e-05 tokens_per_second_per_gpu:3662.1201022663563 mfu:17.274845344509465 +Step 23050 | loss:0.45412662625312805 lr:2.9554289054770163e-05 tokens_per_second_per_gpu:5330.602649419853 mfu:25.14536219190955 +Step 23051 | loss:0.46109920740127563 lr:2.9550332013534155e-05 tokens_per_second_per_gpu:5327.857112211473 mfu:25.132411024460666 +Step 23052 | loss:0.3379189372062683 lr:2.954637512610271e-05 tokens_per_second_per_gpu:4368.2357982357225 mfu:20.605713558157763 +Step 23053 | loss:0.35172146558761597 lr:2.9542418392505577e-05 tokens_per_second_per_gpu:3953.6957960391283 mfu:18.650257639978836 +Step 23054 | loss:0.4906921684741974 lr:2.9538461812772543e-05 tokens_per_second_per_gpu:5332.330366834412 mfu:25.153512129733016 +Step 23055 | loss:0.5352635979652405 lr:2.953450538693333e-05 tokens_per_second_per_gpu:4372.5913428486265 mfu:20.626259405229373 +Step 23056 | loss:0.48589399456977844 lr:2.9530549115017725e-05 tokens_per_second_per_gpu:3280.614375171023 mfu:15.475217738211702 +Step 23057 | loss:0.41541072726249695 lr:2.9526592997055487e-05 tokens_per_second_per_gpu:3762.6428270689394 mfu:17.74902818834865 +Step 23058 | loss:0.42271241545677185 lr:2.952263703307634e-05 tokens_per_second_per_gpu:5299.582870495561 mfu:24.99903660220363 +Step 23059 | loss:0.5339603424072266 lr:2.9518681223110044e-05 tokens_per_second_per_gpu:5317.413197740981 mfu:25.0831452979878 +Step 23060 | loss:0.4477122724056244 lr:2.9514725567186376e-05 tokens_per_second_per_gpu:5239.468143018495 mfu:24.71546517606417 +Step 23061 | loss:0.3895121216773987 lr:2.951077006533506e-05 tokens_per_second_per_gpu:5339.439859661361 mfu:25.187048820402286 +Step 23062 | loss:0.33181077241897583 lr:2.950681471758585e-05 tokens_per_second_per_gpu:5322.241287181309 mfu:25.105920219634836 +Step 23063 | loss:0.3363622725009918 lr:2.950285952396852e-05 tokens_per_second_per_gpu:5307.386579675571 mfu:25.035848029855817 +Step 23064 | loss:0.3833636939525604 lr:2.949890448451278e-05 tokens_per_second_per_gpu:5338.541895926488 mfu:25.18281297225648 +Step 23065 | loss:0.3856211304664612 lr:2.949494959924843e-05 tokens_per_second_per_gpu:5247.360925374708 mfu:24.752696777085962 +Step 23066 | loss:0.4642556607723236 lr:2.9490994868205145e-05 tokens_per_second_per_gpu:4400.430056797225 mfu:20.757579368699293 +Step 23067 | loss:0.5099674463272095 lr:2.948704029141272e-05 tokens_per_second_per_gpu:4226.508689907333 mfu:19.937162607034594 +Step 23068 | loss:0.4378044009208679 lr:2.9483085868900883e-05 tokens_per_second_per_gpu:4743.7498572750255 mfu:22.377077443951247 +Step 23069 | loss:0.3779418170452118 lr:2.9479131600699363e-05 tokens_per_second_per_gpu:3914.9542735546015 mfu:18.467507268433987 +Step 23070 | loss:0.4037254750728607 lr:2.9475177486837925e-05 tokens_per_second_per_gpu:3312.785107239214 mfu:15.626972570270249 +Step 23071 | loss:0.4091869294643402 lr:2.947122352734628e-05 tokens_per_second_per_gpu:5326.765270006823 mfu:25.127260618493843 +Step 23072 | loss:0.49945998191833496 lr:2.9467269722254213e-05 tokens_per_second_per_gpu:3503.708856931131 mfu:16.52759247251774 +Step 23073 | loss:0.40513384342193604 lr:2.9463316071591408e-05 tokens_per_second_per_gpu:4324.408506140643 mfu:20.39897274363817 +Step 23074 | loss:0.42376139760017395 lr:2.945936257538763e-05 tokens_per_second_per_gpu:4041.3994401443165 mfu:19.063970693007708 +Step 23075 | loss:0.5167953968048096 lr:2.9455409233672592e-05 tokens_per_second_per_gpu:5334.595759028061 mfu:25.16419837122622 +Step 23076 | loss:0.38742610812187195 lr:2.9451456046476055e-05 tokens_per_second_per_gpu:4933.838406338293 mfu:23.273757562333476 +Step 23077 | loss:0.3671382665634155 lr:2.9447503013827742e-05 tokens_per_second_per_gpu:4406.683586028016 mfu:20.78707833304304 +Step 23078 | loss:0.43289732933044434 lr:2.944355013575737e-05 tokens_per_second_per_gpu:4894.817109060816 mfu:23.08968744535578 +Step 23079 | loss:0.39335867762565613 lr:2.94395974122947e-05 tokens_per_second_per_gpu:4210.702701647453 mfu:19.862603063633056 +Step 23080 | loss:0.41710546612739563 lr:2.9435644843469436e-05 tokens_per_second_per_gpu:4539.76484988014 mfu:21.414845360638495 +Step 23081 | loss:0.4095154404640198 lr:2.9431692429311303e-05 tokens_per_second_per_gpu:4819.903912872763 mfu:22.736309117427474 +Step 23082 | loss:0.4633558988571167 lr:2.9427740169850027e-05 tokens_per_second_per_gpu:4669.552433205756 mfu:22.02707553523056 +Step 23083 | loss:0.39310139417648315 lr:2.9423788065115354e-05 tokens_per_second_per_gpu:4511.204511161148 mfu:21.280121370004743 +Step 23084 | loss:0.45418837666511536 lr:2.9419836115137e-05 tokens_per_second_per_gpu:3240.5157930988585 mfu:15.286065885053727 +Step 23085 | loss:0.4915734827518463 lr:2.9415884319944677e-05 tokens_per_second_per_gpu:5318.861859510242 mfu:25.08997888272059 +Step 23086 | loss:0.39737042784690857 lr:2.9411932679568126e-05 tokens_per_second_per_gpu:5340.45935310071 mfu:25.191857944899215 +Step 23087 | loss:0.4069116711616516 lr:2.940798119403705e-05 tokens_per_second_per_gpu:5372.501120745264 mfu:25.343004429767866 +Step 23088 | loss:0.4071498513221741 lr:2.940402986338118e-05 tokens_per_second_per_gpu:4303.887553118619 mfu:20.302171906997653 +Step 23089 | loss:0.413473904132843 lr:2.9400078687630206e-05 tokens_per_second_per_gpu:4181.346939538928 mfu:19.72412692515766 +Step 23090 | loss:0.5155087113380432 lr:2.939612766681389e-05 tokens_per_second_per_gpu:5188.267356753459 mfu:24.47394233149706 +Step 23091 | loss:0.37572452425956726 lr:2.939217680096191e-05 tokens_per_second_per_gpu:3399.5630930979055 mfu:16.036319135416814 +Step 23092 | loss:0.4375576078891754 lr:2.9388226090104016e-05 tokens_per_second_per_gpu:4911.691128354828 mfu:23.1692850733702 +Step 23093 | loss:0.4053248167037964 lr:2.93842755342699e-05 tokens_per_second_per_gpu:4266.741819096334 mfu:20.12694914189879 +Step 23094 | loss:0.2923109233379364 lr:2.9380325133489288e-05 tokens_per_second_per_gpu:4012.494325293858 mfu:18.927620334536957 +Step 23095 | loss:0.4977836608886719 lr:2.937637488779186e-05 tokens_per_second_per_gpu:4455.199968822971 mfu:21.015938388435323 +Step 23096 | loss:0.3936496675014496 lr:2.9372424797207365e-05 tokens_per_second_per_gpu:4092.6625755973596 mfu:19.30578764933252 +Step 23097 | loss:0.35398608446121216 lr:2.9368474861765483e-05 tokens_per_second_per_gpu:3994.5364368337855 mfu:18.84290991073908 +Step 23098 | loss:0.43462279438972473 lr:2.936452508149593e-05 tokens_per_second_per_gpu:3966.3199841738965 mfu:18.709808089319136 +Step 23099 | loss:0.4440361559391022 lr:2.936057545642843e-05 tokens_per_second_per_gpu:4635.356575836466 mfu:21.865767841604725 +Step 23100 | loss:0.47860226035118103 lr:2.935662598659268e-05 tokens_per_second_per_gpu:5273.811235428289 mfu:24.877467402496954 +Step 23101 | loss:0.4162936806678772 lr:2.9352676672018382e-05 tokens_per_second_per_gpu:4704.723254434551 mfu:22.192982299727664 +Step 23102 | loss:0.36443498730659485 lr:2.934872751273522e-05 tokens_per_second_per_gpu:5360.322427058124 mfu:25.285555453746213 +Step 23103 | loss:0.4212634265422821 lr:2.934477850877292e-05 tokens_per_second_per_gpu:4311.9252750893265 mfu:20.340087212910266 +Step 23104 | loss:0.485352098941803 lr:2.934082966016118e-05 tokens_per_second_per_gpu:5336.627859800811 mfu:25.173784137283427 +Step 23105 | loss:0.4030091464519501 lr:2.933688096692968e-05 tokens_per_second_per_gpu:5311.66553290065 mfu:25.05603258228322 +Step 23106 | loss:0.44592562317848206 lr:2.933293242910815e-05 tokens_per_second_per_gpu:5317.507687512592 mfu:25.08359102236211 +Step 23107 | loss:0.4263308346271515 lr:2.9328984046726267e-05 tokens_per_second_per_gpu:4858.222621596609 mfu:22.917065004323728 +Step 23108 | loss:0.33095353841781616 lr:2.9325035819813757e-05 tokens_per_second_per_gpu:5332.826765138588 mfu:25.15585372522809 +Step 23109 | loss:0.5451978445053101 lr:2.932108774840026e-05 tokens_per_second_per_gpu:5358.6841728653035 mfu:25.27782752920369 +Step 23110 | loss:0.3951219916343689 lr:2.9317139832515515e-05 tokens_per_second_per_gpu:5259.864318408475 mfu:24.81167741532507 +Step 23111 | loss:0.3456394374370575 lr:2.931319207218919e-05 tokens_per_second_per_gpu:4660.930891446541 mfu:21.986406251765636 +Step 23112 | loss:0.6143704056739807 lr:2.9309244467450993e-05 tokens_per_second_per_gpu:4005.052622961353 mfu:18.89251655494919 +Step 23113 | loss:0.5459306240081787 lr:2.9305297018330613e-05 tokens_per_second_per_gpu:3467.854590741016 mfu:16.358461781530355 +Step 23114 | loss:0.35995081067085266 lr:2.930134972485772e-05 tokens_per_second_per_gpu:5328.629981155671 mfu:25.13605677913549 +Step 23115 | loss:0.516648530960083 lr:2.929740258706204e-05 tokens_per_second_per_gpu:4712.241084100854 mfu:22.228445184937627 +Step 23116 | loss:0.4637109339237213 lr:2.9293455604973225e-05 tokens_per_second_per_gpu:4802.782571467086 mfu:22.655544828814833 +Step 23117 | loss:0.488895058631897 lr:2.928950877862098e-05 tokens_per_second_per_gpu:4574.8270431255405 mfu:21.580239708404086 +Step 23118 | loss:0.483699768781662 lr:2.928556210803496e-05 tokens_per_second_per_gpu:5103.247773882491 mfu:24.072890453258072 +Step 23119 | loss:0.37094706296920776 lr:2.9281615593244887e-05 tokens_per_second_per_gpu:5354.8501458572 mfu:25.259741769654006 +Step 23120 | loss:0.5007907152175903 lr:2.9277669234280424e-05 tokens_per_second_per_gpu:5376.781439760945 mfu:25.36319542486256 +Step 23121 | loss:0.41673675179481506 lr:2.9273723031171253e-05 tokens_per_second_per_gpu:5231.843166863817 mfu:24.679496862587474 +Step 23122 | loss:0.35600385069847107 lr:2.9269776983947055e-05 tokens_per_second_per_gpu:4807.808166652287 mfu:22.67925141875838 +Step 23123 | loss:0.46560972929000854 lr:2.9265831092637525e-05 tokens_per_second_per_gpu:5347.088300307297 mfu:25.223127819887704 +Step 23124 | loss:0.556723415851593 lr:2.926188535727231e-05 tokens_per_second_per_gpu:4232.61529840466 mfu:19.96596852121133 +Step 23125 | loss:0.36487239599227905 lr:2.92579397778811e-05 tokens_per_second_per_gpu:5324.897455291986 mfu:25.11844981780191 +Step 23126 | loss:0.49650537967681885 lr:2.9253994354493573e-05 tokens_per_second_per_gpu:4896.983855279199 mfu:23.09990835695292 +Step 23127 | loss:0.42868828773498535 lr:2.92500490871394e-05 tokens_per_second_per_gpu:4008.416657883164 mfu:18.908385281638918 +Step 23128 | loss:0.5169916152954102 lr:2.924610397584826e-05 tokens_per_second_per_gpu:5355.6029553926 mfu:25.263292900675037 +Step 23129 | loss:0.41226065158843994 lr:2.9242159020649824e-05 tokens_per_second_per_gpu:5177.114570402872 mfu:24.421332735419774 +Step 23130 | loss:0.4309884011745453 lr:2.923821422157377e-05 tokens_per_second_per_gpu:4350.568793329753 mfu:20.522375281714446 +Step 23131 | loss:0.4475775361061096 lr:2.9234269578649732e-05 tokens_per_second_per_gpu:5345.686578455291 mfu:25.216515658753384 +Step 23132 | loss:0.4308166205883026 lr:2.9230325091907407e-05 tokens_per_second_per_gpu:5318.320554919307 mfu:25.08742545660907 +Step 23133 | loss:0.46052277088165283 lr:2.9226380761376467e-05 tokens_per_second_per_gpu:4845.8287554504495 mfu:22.85860102310068 +Step 23134 | loss:0.4477570652961731 lr:2.9222436587086554e-05 tokens_per_second_per_gpu:5336.116808023851 mfu:25.17137341885708 +Step 23135 | loss:0.40390744805336 lr:2.9218492569067356e-05 tokens_per_second_per_gpu:4647.892480168813 mfu:21.924901841186237 +Step 23136 | loss:0.41168317198753357 lr:2.9214548707348528e-05 tokens_per_second_per_gpu:5358.8522024901395 mfu:25.278620153612852 +Step 23137 | loss:0.3774746358394623 lr:2.9210605001959744e-05 tokens_per_second_per_gpu:3704.923824585507 mfu:17.476757805756996 +Step 23138 | loss:0.4365094304084778 lr:2.9206661452930627e-05 tokens_per_second_per_gpu:3325.1465924670406 mfu:15.685283805176768 +Step 23139 | loss:0.41468095779418945 lr:2.920271806029088e-05 tokens_per_second_per_gpu:4861.674307244652 mfu:22.933347194443847 +Step 23140 | loss:0.4166499972343445 lr:2.9198774824070136e-05 tokens_per_second_per_gpu:4094.4132591112243 mfu:19.314045922164826 +Step 23141 | loss:0.5169550180435181 lr:2.9194831744298058e-05 tokens_per_second_per_gpu:4241.039231840973 mfu:20.005705652501668 +Step 23142 | loss:0.3207691013813019 lr:2.9190888821004298e-05 tokens_per_second_per_gpu:4916.096973762512 mfu:23.190068197872986 +Step 23143 | loss:0.3418121635913849 lr:2.9186946054218544e-05 tokens_per_second_per_gpu:4579.253514826331 mfu:21.601120130650667 +Step 23144 | loss:0.4926903247833252 lr:2.918300344397042e-05 tokens_per_second_per_gpu:4290.036562406263 mfu:20.236834420584774 +Step 23145 | loss:0.4605028033256531 lr:2.9179060990289563e-05 tokens_per_second_per_gpu:4038.113437056355 mfu:19.048470055791853 +Step 23146 | loss:0.5575392842292786 lr:2.917511869320565e-05 tokens_per_second_per_gpu:4190.839809487066 mfu:19.76890641235365 +Step 23147 | loss:0.34601110219955444 lr:2.917117655274835e-05 tokens_per_second_per_gpu:5347.059121082245 mfu:25.22299017650464 +Step 23148 | loss:0.42509329319000244 lr:2.9167234568947254e-05 tokens_per_second_per_gpu:3874.3642976609517 mfu:18.27603742677963 +Step 23149 | loss:0.48541364073753357 lr:2.916329274183206e-05 tokens_per_second_per_gpu:5320.090342457519 mfu:25.095773846383704 +Step 23150 | loss:0.5474311113357544 lr:2.9159351071432413e-05 tokens_per_second_per_gpu:4399.955848609208 mfu:20.755342447768157 +Step 23151 | loss:0.4600733518600464 lr:2.9155409557777925e-05 tokens_per_second_per_gpu:4747.003346883771 mfu:22.392424709538265 +Step 23152 | loss:0.619402289390564 lr:2.9151468200898285e-05 tokens_per_second_per_gpu:4064.7991707682972 mfu:19.174351214766755 +Step 23153 | loss:0.46862608194351196 lr:2.9147527000823095e-05 tokens_per_second_per_gpu:3247.826737188243 mfu:15.32055285569904 +Step 23154 | loss:0.3253630995750427 lr:2.914358595758203e-05 tokens_per_second_per_gpu:3335.0605341209803 mfu:15.732049559451267 +Step 23155 | loss:0.4812193512916565 lr:2.91396450712047e-05 tokens_per_second_per_gpu:5376.348396209525 mfu:25.36115268454566 +Step 23156 | loss:0.514025866985321 lr:2.913570434172076e-05 tokens_per_second_per_gpu:3779.7123367497743 mfu:17.829547977871812 +Step 23157 | loss:0.4334128499031067 lr:2.913176376915987e-05 tokens_per_second_per_gpu:5324.688344588354 mfu:25.117463406934913 +Step 23158 | loss:0.39399489760398865 lr:2.9127823353551625e-05 tokens_per_second_per_gpu:5362.266291920192 mfu:25.294724995958134 +Step 23159 | loss:0.4861409366130829 lr:2.912388309492571e-05 tokens_per_second_per_gpu:4352.313821527809 mfu:20.530606877457185 +Step 23160 | loss:0.40361350774765015 lr:2.9119942993311706e-05 tokens_per_second_per_gpu:4298.974187356173 mfu:20.27899472238956 +Step 23161 | loss:0.38939687609672546 lr:2.9116003048739298e-05 tokens_per_second_per_gpu:4329.939075266116 mfu:20.425061381815418 +Step 23162 | loss:0.3669070303440094 lr:2.9112063261238077e-05 tokens_per_second_per_gpu:5322.979484149008 mfu:25.109402420680862 +Step 23163 | loss:0.2981216013431549 lr:2.910812363083769e-05 tokens_per_second_per_gpu:3778.003688398009 mfu:17.821487992071066 +Step 23164 | loss:0.33049336075782776 lr:2.9104184157567797e-05 tokens_per_second_per_gpu:4746.421234483724 mfu:22.389678786028625 +Step 23165 | loss:0.4134582281112671 lr:2.910024484145798e-05 tokens_per_second_per_gpu:4684.551572396409 mfu:22.097829033909903 +Step 23166 | loss:0.4637182950973511 lr:2.9096305682537907e-05 tokens_per_second_per_gpu:5265.392163451081 mfu:24.83775320353866 +Step 23167 | loss:0.4466143548488617 lr:2.9092366680837168e-05 tokens_per_second_per_gpu:4784.51495160165 mfu:22.569373349132228 +Step 23168 | loss:0.4203841984272003 lr:2.9088427836385423e-05 tokens_per_second_per_gpu:5349.868329618143 mfu:25.236241692468877 +Step 23169 | loss:0.3887821137905121 lr:2.9084489149212263e-05 tokens_per_second_per_gpu:3821.4971751540006 mfu:18.026654189853122 +Step 23170 | loss:0.4388269782066345 lr:2.908055061934733e-05 tokens_per_second_per_gpu:4292.078384039068 mfu:20.246466041597404 +Step 23171 | loss:0.3708059787750244 lr:2.9076612246820246e-05 tokens_per_second_per_gpu:4292.6290044233765 mfu:20.249063411895712 +Step 23172 | loss:0.433302640914917 lr:2.9072674031660647e-05 tokens_per_second_per_gpu:5332.757238885399 mfu:25.1555257580294 +Step 23173 | loss:0.40942591428756714 lr:2.906873597389814e-05 tokens_per_second_per_gpu:5098.281247508445 mfu:24.049462500974915 +Step 23174 | loss:0.40505582094192505 lr:2.9064798073562316e-05 tokens_per_second_per_gpu:4732.421922064301 mfu:22.323641640817215 +Step 23175 | loss:0.37218067049980164 lr:2.9060860330682827e-05 tokens_per_second_per_gpu:5360.990793510201 mfu:25.288708252335738 +Step 23176 | loss:0.4956274628639221 lr:2.9056922745289287e-05 tokens_per_second_per_gpu:5325.226851861003 mfu:25.120003637618808 +Step 23177 | loss:0.39982643723487854 lr:2.9052985317411292e-05 tokens_per_second_per_gpu:5333.755794872012 mfu:25.160236116989473 +Step 23178 | loss:0.37051674723625183 lr:2.9049048047078464e-05 tokens_per_second_per_gpu:3809.7472266335667 mfu:17.971227677934294 +Step 23179 | loss:0.4361087679862976 lr:2.904511093432044e-05 tokens_per_second_per_gpu:3415.6730564049 mfu:16.11231258097908 +Step 23180 | loss:0.3567502200603485 lr:2.9041173979166813e-05 tokens_per_second_per_gpu:5345.421597014971 mfu:25.2152656960889 +Step 23181 | loss:0.30398181080818176 lr:2.9037237181647166e-05 tokens_per_second_per_gpu:4990.713731599419 mfu:23.542048175521334 +Step 23182 | loss:0.47630760073661804 lr:2.9033300541791132e-05 tokens_per_second_per_gpu:4087.7029113774397 mfu:19.282392067001187 +Step 23183 | loss:0.46301499009132385 lr:2.902936405962835e-05 tokens_per_second_per_gpu:4655.431353664936 mfu:21.960463993731903 +Step 23184 | loss:0.3711138367652893 lr:2.9025427735188364e-05 tokens_per_second_per_gpu:3424.2698489574577 mfu:16.15286511821355 +Step 23185 | loss:0.4079710841178894 lr:2.9021491568500818e-05 tokens_per_second_per_gpu:4808.750434398341 mfu:22.683696256483973 +Step 23186 | loss:0.5185885429382324 lr:2.9017555559595334e-05 tokens_per_second_per_gpu:4843.417318423126 mfu:22.84722586320921 +Step 23187 | loss:0.3744620382785797 lr:2.9013619708501473e-05 tokens_per_second_per_gpu:3193.017731905376 mfu:15.062009426399172 +Step 23188 | loss:0.3785941004753113 lr:2.9009684015248867e-05 tokens_per_second_per_gpu:4993.179139399191 mfu:23.55367792475481 +Step 23189 | loss:0.4908222258090973 lr:2.900574847986709e-05 tokens_per_second_per_gpu:5373.889481893331 mfu:25.349553566182134 +Step 23190 | loss:0.47245046496391296 lr:2.9001813102385778e-05 tokens_per_second_per_gpu:5325.659190985646 mfu:25.122043055034414 +Step 23191 | loss:0.552388072013855 lr:2.899787788283449e-05 tokens_per_second_per_gpu:3928.442720705287 mfu:18.53113457501016 +Step 23192 | loss:0.5130769610404968 lr:2.899394282124285e-05 tokens_per_second_per_gpu:4538.763558342077 mfu:21.4101220976155 +Step 23193 | loss:0.5592927932739258 lr:2.899000791764045e-05 tokens_per_second_per_gpu:4034.6810507975206 mfu:19.03227890418345 +Step 23194 | loss:0.5361030101776123 lr:2.898607317205687e-05 tokens_per_second_per_gpu:4942.246646300496 mfu:23.313420664828858 +Step 23195 | loss:0.3826156556606293 lr:2.8982138584521735e-05 tokens_per_second_per_gpu:3085.270189779128 mfu:14.553745886563313 +Step 23196 | loss:0.3281163275241852 lr:2.8978204155064605e-05 tokens_per_second_per_gpu:5349.699870541065 mfu:25.23544704226022 +Step 23197 | loss:0.4987015426158905 lr:2.8974269883715088e-05 tokens_per_second_per_gpu:4666.663150459493 mfu:22.01344629555397 +Step 23198 | loss:0.44017335772514343 lr:2.897033577050276e-05 tokens_per_second_per_gpu:2963.509972715173 mfu:13.979382168237414 +Step 23199 | loss:0.3490774929523468 lr:2.8966401815457223e-05 tokens_per_second_per_gpu:5036.224688956404 mfu:23.75673112634426 +Step 23200 | loss:0.4132480323314667 lr:2.8962468018608076e-05 tokens_per_second_per_gpu:3546.5464192763766 mfu:16.729664562942062 +Step 23201 | loss:0.5369042158126831 lr:2.895853437998487e-05 tokens_per_second_per_gpu:5334.697235829629 mfu:25.16467705461294 +Step 23202 | loss:0.4192669689655304 lr:2.8954600899617236e-05 tokens_per_second_per_gpu:4979.838754957048 mfu:23.490749055236925 +Step 23203 | loss:0.3181341290473938 lr:2.895066757753471e-05 tokens_per_second_per_gpu:3740.7191368654217 mfu:17.645610401091393 +Step 23204 | loss:0.44773203134536743 lr:2.8946734413766923e-05 tokens_per_second_per_gpu:4611.747947096972 mfu:21.754401911793114 +Step 23205 | loss:0.4914979636669159 lr:2.8942801408343413e-05 tokens_per_second_per_gpu:5327.029558316407 mfu:25.12850731154239 +Step 23206 | loss:0.42184728384017944 lr:2.893886856129378e-05 tokens_per_second_per_gpu:3951.576489009232 mfu:18.640260507127746 +Step 23207 | loss:0.41171446442604065 lr:2.8934935872647605e-05 tokens_per_second_per_gpu:4756.086069466658 mfu:22.435269461634014 +Step 23208 | loss:0.43624094128608704 lr:2.893100334243448e-05 tokens_per_second_per_gpu:5243.417248698132 mfu:24.73409378134193 +Step 23209 | loss:0.3553575873374939 lr:2.8927070970683973e-05 tokens_per_second_per_gpu:5315.691470225652 mfu:25.075023615540992 +Step 23210 | loss:0.5227465629577637 lr:2.8923138757425628e-05 tokens_per_second_per_gpu:5299.096029351492 mfu:24.996740089463394 +Step 23211 | loss:0.38092857599258423 lr:2.891920670268905e-05 tokens_per_second_per_gpu:5321.398219673303 mfu:25.101943326357418 +Step 23212 | loss:0.42589884996414185 lr:2.8915274806503833e-05 tokens_per_second_per_gpu:4784.984954930453 mfu:22.571590435025026 +Step 23213 | loss:0.42730990052223206 lr:2.89113430688995e-05 tokens_per_second_per_gpu:4461.712370803649 mfu:21.046658499708535 +Step 23214 | loss:0.48555371165275574 lr:2.8907411489905646e-05 tokens_per_second_per_gpu:4731.055763213598 mfu:22.317197236427713 +Step 23215 | loss:0.46389156579971313 lr:2.8903480069551858e-05 tokens_per_second_per_gpu:3655.8609783754787 mfu:17.24531998919954 +Step 23216 | loss:0.37248894572257996 lr:2.889954880786767e-05 tokens_per_second_per_gpu:3252.356515985531 mfu:15.341920595145806 +Step 23217 | loss:0.4197755455970764 lr:2.8895617704882694e-05 tokens_per_second_per_gpu:4738.32696182391 mfu:22.351496720866965 +Step 23218 | loss:0.4792175590991974 lr:2.8891686760626447e-05 tokens_per_second_per_gpu:4573.91630167597 mfu:21.575943585597944 +Step 23219 | loss:0.4828343391418457 lr:2.888775597512854e-05 tokens_per_second_per_gpu:4261.235011976695 mfu:20.100972593157273 +Step 23220 | loss:0.3786388337612152 lr:2.8883825348418486e-05 tokens_per_second_per_gpu:4496.598484273241 mfu:21.21122224912933 +Step 23221 | loss:0.4662793278694153 lr:2.887989488052588e-05 tokens_per_second_per_gpu:4104.78246752587 mfu:19.362959247426094 +Step 23222 | loss:0.4111291468143463 lr:2.8875964571480303e-05 tokens_per_second_per_gpu:4130.384146085152 mfu:19.483726733287018 +Step 23223 | loss:0.38370686769485474 lr:2.8872034421311268e-05 tokens_per_second_per_gpu:4414.792878536109 mfu:20.82533124939169 +Step 23224 | loss:0.4467923045158386 lr:2.886810443004838e-05 tokens_per_second_per_gpu:3805.161294146029 mfu:17.949595051949814 +Step 23225 | loss:0.4135318100452423 lr:2.886417459772115e-05 tokens_per_second_per_gpu:5394.597981415527 mfu:25.447239091662684 +Step 23226 | loss:0.5445794463157654 lr:2.8860244924359187e-05 tokens_per_second_per_gpu:5230.449266226386 mfu:24.67292159545674 +Step 23227 | loss:0.4129777252674103 lr:2.8856315409991995e-05 tokens_per_second_per_gpu:4835.831575192707 mfu:22.811442618130066 +Step 23228 | loss:0.4516453742980957 lr:2.8852386054649156e-05 tokens_per_second_per_gpu:4027.9320976092727 mfu:19.000442940504332 +Step 23229 | loss:0.47720134258270264 lr:2.884845685836024e-05 tokens_per_second_per_gpu:5363.297074012397 mfu:25.299587370956626 +Step 23230 | loss:0.3836124539375305 lr:2.8844527821154754e-05 tokens_per_second_per_gpu:3497.7477240154235 mfu:16.499472791480194 +Step 23231 | loss:0.45468807220458984 lr:2.8840598943062303e-05 tokens_per_second_per_gpu:3673.8403303566024 mfu:17.33013166008827 +Step 23232 | loss:0.43697482347488403 lr:2.8836670224112378e-05 tokens_per_second_per_gpu:5361.261634199386 mfu:25.289985854076807 +Step 23233 | loss:0.4533896744251251 lr:2.883274166433458e-05 tokens_per_second_per_gpu:5339.998997053646 mfu:25.189686366880373 +Step 23234 | loss:0.4841301441192627 lr:2.8828813263758412e-05 tokens_per_second_per_gpu:4106.423053131462 mfu:19.370698169640445 +Step 23235 | loss:0.3782505691051483 lr:2.882488502241344e-05 tokens_per_second_per_gpu:5369.88207083969 mfu:25.330649924507277 +Step 23236 | loss:0.4135744869709015 lr:2.8820956940329236e-05 tokens_per_second_per_gpu:5052.866232945467 mfu:23.835232128680857 +Step 23237 | loss:0.4145332872867584 lr:2.881702901753529e-05 tokens_per_second_per_gpu:5336.82332053451 mfu:25.174706158913317 +Step 23238 | loss:0.3806540071964264 lr:2.8813101254061193e-05 tokens_per_second_per_gpu:4703.222028259448 mfu:22.18590075972411 +Step 23239 | loss:0.3663480281829834 lr:2.880917364993645e-05 tokens_per_second_per_gpu:4521.775708741796 mfu:21.329987512624953 +Step 23240 | loss:0.4024220108985901 lr:2.8805246205190627e-05 tokens_per_second_per_gpu:3811.4708420960396 mfu:17.9793582661499 +Step 23241 | loss:0.4237542152404785 lr:2.8801318919853238e-05 tokens_per_second_per_gpu:5359.4659401606805 mfu:25.28151525891924 +Step 23242 | loss:0.4588065445423126 lr:2.8797391793953833e-05 tokens_per_second_per_gpu:4565.5993313923855 mfu:21.53671101774882 +Step 23243 | loss:0.4223799705505371 lr:2.8793464827521953e-05 tokens_per_second_per_gpu:4216.626253292159 mfu:19.890545467403264 +Step 23244 | loss:0.4307883381843567 lr:2.878953802058715e-05 tokens_per_second_per_gpu:5068.098554144274 mfu:23.907085586677617 +Step 23245 | loss:0.47935235500335693 lr:2.8785611373178913e-05 tokens_per_second_per_gpu:3674.122930671925 mfu:17.33146473399223 +Step 23246 | loss:0.4718085527420044 lr:2.878168488532682e-05 tokens_per_second_per_gpu:4973.929852188532 mfu:23.462875752715064 +Step 23247 | loss:0.3764622211456299 lr:2.8777758557060385e-05 tokens_per_second_per_gpu:4453.153544813006 mfu:21.00628505722438 +Step 23248 | loss:0.42103755474090576 lr:2.8773832388409116e-05 tokens_per_second_per_gpu:5303.194125590552 mfu:25.016071508630358 +Step 23249 | loss:0.398759126663208 lr:2.876990637940256e-05 tokens_per_second_per_gpu:5312.45436244436 mfu:25.0597536258291 +Step 23250 | loss:0.42771169543266296 lr:2.8765980530070248e-05 tokens_per_second_per_gpu:4161.547164747721 mfu:19.630728009277604 +Step 23251 | loss:0.38711273670196533 lr:2.876205484044173e-05 tokens_per_second_per_gpu:5342.692259813744 mfu:25.202390946837767 +Step 23252 | loss:0.3483777642250061 lr:2.8758129310546484e-05 tokens_per_second_per_gpu:3095.0397877353143 mfu:14.599830746987937 +Step 23253 | loss:0.3699512183666229 lr:2.8754203940414076e-05 tokens_per_second_per_gpu:5328.157921126589 mfu:25.133829991436563 +Step 23254 | loss:0.4381725788116455 lr:2.8750278730073992e-05 tokens_per_second_per_gpu:4877.026343949804 mfu:23.005765370909607 +Step 23255 | loss:0.4524346888065338 lr:2.874635367955579e-05 tokens_per_second_per_gpu:4061.859703037718 mfu:19.1604852439567 +Step 23256 | loss:0.40388545393943787 lr:2.8742428788888965e-05 tokens_per_second_per_gpu:4643.072782674702 mfu:21.902166505781327 +Step 23257 | loss:0.3578821122646332 lr:2.873850405810304e-05 tokens_per_second_per_gpu:5330.557753026166 mfu:25.14515040796041 +Step 23258 | loss:0.5606568455696106 lr:2.8734579487227555e-05 tokens_per_second_per_gpu:5369.946023350811 mfu:25.33095159941366 +Step 23259 | loss:0.3773534297943115 lr:2.8730655076291996e-05 tokens_per_second_per_gpu:4867.684238574001 mfu:22.961697065924408 +Step 23260 | loss:0.3882863223552704 lr:2.8726730825325908e-05 tokens_per_second_per_gpu:4545.459928724216 mfu:21.441710019226647 +Step 23261 | loss:0.38408979773521423 lr:2.8722806734358777e-05 tokens_per_second_per_gpu:4880.288974886273 mfu:23.021155757699166 +Step 23262 | loss:0.46455711126327515 lr:2.8718882803420154e-05 tokens_per_second_per_gpu:5248.162918703859 mfu:24.75647991645374 +Step 23263 | loss:0.3786298334598541 lr:2.8714959032539502e-05 tokens_per_second_per_gpu:5307.851880123264 mfu:25.03804293145603 +Step 23264 | loss:0.4717339277267456 lr:2.8711035421746367e-05 tokens_per_second_per_gpu:5070.22486989424 mfu:23.917115780857085 +Step 23265 | loss:0.35749953985214233 lr:2.8707111971070265e-05 tokens_per_second_per_gpu:3634.17198207163 mfu:17.14300929310993 +Step 23266 | loss:0.5734938979148865 lr:2.870318868054067e-05 tokens_per_second_per_gpu:4519.0536636952365 mfu:21.31714716171156 +Step 23267 | loss:0.4267451763153076 lr:2.8699265550187136e-05 tokens_per_second_per_gpu:5336.358605465876 mfu:25.17251401864601 +Step 23268 | loss:0.36091986298561096 lr:2.869534258003912e-05 tokens_per_second_per_gpu:4730.20773822964 mfu:22.3131969578904 +Step 23269 | loss:0.4298570454120636 lr:2.869141977012617e-05 tokens_per_second_per_gpu:4847.4601454601225 mfu:22.86629656812014 +Step 23270 | loss:0.368935227394104 lr:2.8687497120477756e-05 tokens_per_second_per_gpu:5378.46682222411 mfu:25.37114566149721 +Step 23271 | loss:0.43989160656929016 lr:2.8683574631123388e-05 tokens_per_second_per_gpu:4804.7934520812905 mfu:22.665030495763645 +Step 23272 | loss:0.4975239932537079 lr:2.86796523020926e-05 tokens_per_second_per_gpu:3693.9614115053605 mfu:17.425046232877406 +Step 23273 | loss:0.33213818073272705 lr:2.8675730133414848e-05 tokens_per_second_per_gpu:4826.54393634462 mfu:22.767631241049056 +Step 23274 | loss:0.33970901370048523 lr:2.867180812511967e-05 tokens_per_second_per_gpu:5332.618398722616 mfu:25.154870825292974 +Step 23275 | loss:0.36866694688796997 lr:2.8667886277236522e-05 tokens_per_second_per_gpu:4060.1298727351837 mfu:19.15232533928016 +Step 23276 | loss:0.413865864276886 lr:2.8663964589794943e-05 tokens_per_second_per_gpu:4037.6537630415337 mfu:19.046301695035467 +Step 23277 | loss:0.35262107849121094 lr:2.8660043062824394e-05 tokens_per_second_per_gpu:5363.317305161391 mfu:25.29968280473822 +Step 23278 | loss:0.5297623872756958 lr:2.8656121696354387e-05 tokens_per_second_per_gpu:4780.7697060695555 mfu:22.551706386952556 +Step 23279 | loss:0.4700163006782532 lr:2.8652200490414404e-05 tokens_per_second_per_gpu:4632.128127544802 mfu:21.850538700182867 +Step 23280 | loss:0.4736509621143341 lr:2.8648279445033964e-05 tokens_per_second_per_gpu:5179.99039010506 mfu:24.43489846762043 +Step 23281 | loss:0.4652478098869324 lr:2.8644358560242524e-05 tokens_per_second_per_gpu:4163.483397332706 mfu:19.639861548736324 +Step 23282 | loss:0.443105548620224 lr:2.8640437836069612e-05 tokens_per_second_per_gpu:3943.948033153291 mfu:18.60427577930649 +Step 23283 | loss:0.4208253026008606 lr:2.863651727254468e-05 tokens_per_second_per_gpu:3506.220849329839 mfu:16.539441969252543 +Step 23284 | loss:0.5650292634963989 lr:2.863259686969722e-05 tokens_per_second_per_gpu:5326.45331802785 mfu:25.125789087784533 +Step 23285 | loss:0.3956562876701355 lr:2.8628676627556716e-05 tokens_per_second_per_gpu:4354.742423888631 mfu:20.54206300915547 +Step 23286 | loss:0.43379679322242737 lr:2.862475654615267e-05 tokens_per_second_per_gpu:3622.691276929168 mfu:17.088852848142654 +Step 23287 | loss:0.38295602798461914 lr:2.8620836625514574e-05 tokens_per_second_per_gpu:3553.7148255755806 mfu:16.76347915851164 +Step 23288 | loss:0.3198370039463043 lr:2.861691686567187e-05 tokens_per_second_per_gpu:5182.8365588852175 mfu:24.4483243313629 +Step 23289 | loss:0.37116050720214844 lr:2.861299726665409e-05 tokens_per_second_per_gpu:4950.750664521843 mfu:23.353535569714076 +Step 23290 | loss:0.3787272572517395 lr:2.860907782849066e-05 tokens_per_second_per_gpu:4855.32469990643 mfu:22.90339501327473 +Step 23291 | loss:0.3978716731071472 lr:2.860515855121111e-05 tokens_per_second_per_gpu:3846.1883356437643 mfu:18.14312660662972 +Step 23292 | loss:0.349613219499588 lr:2.8601239434844872e-05 tokens_per_second_per_gpu:5078.871959993602 mfu:23.957905580201476 +Step 23293 | loss:0.3318050503730774 lr:2.859732047942144e-05 tokens_per_second_per_gpu:4757.064063956211 mfu:22.4398828285043 +Step 23294 | loss:0.4307798147201538 lr:2.8593401684970312e-05 tokens_per_second_per_gpu:4541.996405372471 mfu:21.425371988638446 +Step 23295 | loss:0.44872474670410156 lr:2.8589483051520926e-05 tokens_per_second_per_gpu:5331.917937589229 mfu:25.151566630615797 +Step 23296 | loss:0.39802056550979614 lr:2.8585564579102786e-05 tokens_per_second_per_gpu:4574.305139459854 mfu:21.577777799767095 +Step 23297 | loss:0.4075655937194824 lr:2.8581646267745326e-05 tokens_per_second_per_gpu:4775.670336435547 mfu:22.527651790346766 +Step 23298 | loss:0.5572316646575928 lr:2.8577728117478065e-05 tokens_per_second_per_gpu:4949.8380526062865 mfu:23.349230623600747 +Step 23299 | loss:0.4000520408153534 lr:2.8573810128330413e-05 tokens_per_second_per_gpu:5349.455732058935 mfu:25.234295399385687 +Step 23300 | loss:0.45140716433525085 lr:2.856989230033188e-05 tokens_per_second_per_gpu:4824.135276190889 mfu:22.756269180144105 +Step 23301 | loss:0.30906108021736145 lr:2.8565974633511933e-05 tokens_per_second_per_gpu:5309.428868689892 mfu:25.045481855586807 +Step 23302 | loss:0.41031137108802795 lr:2.856205712790001e-05 tokens_per_second_per_gpu:4274.111270972613 mfu:20.161712103757342 +Step 23303 | loss:0.36985328793525696 lr:2.8558139783525605e-05 tokens_per_second_per_gpu:3330.296122317169 mfu:15.709575016080214 +Step 23304 | loss:0.39355769753456116 lr:2.8554222600418158e-05 tokens_per_second_per_gpu:4507.88315581826 mfu:21.26445397017069 +Step 23305 | loss:0.40783393383026123 lr:2.855030557860715e-05 tokens_per_second_per_gpu:4186.099726866506 mfu:19.746546633890976 +Step 23306 | loss:0.48020780086517334 lr:2.8546388718122012e-05 tokens_per_second_per_gpu:4731.858871427346 mfu:22.320985634896353 +Step 23307 | loss:0.42692986130714417 lr:2.8542472018992227e-05 tokens_per_second_per_gpu:4698.881253490265 mfu:22.16542458452533 +Step 23308 | loss:0.33177709579467773 lr:2.8538555481247263e-05 tokens_per_second_per_gpu:5144.2421060349325 mfu:24.266267712375438 +Step 23309 | loss:0.38207772374153137 lr:2.853463910491654e-05 tokens_per_second_per_gpu:4436.003517621528 mfu:20.925385452865807 +Step 23310 | loss:0.41296127438545227 lr:2.8530722890029537e-05 tokens_per_second_per_gpu:5374.667252258863 mfu:25.353222441697657 +Step 23311 | loss:0.35776975750923157 lr:2.852680683661573e-05 tokens_per_second_per_gpu:5366.19162314568 mfu:25.31324145307933 +Step 23312 | loss:0.4038603603839874 lr:2.852289094470455e-05 tokens_per_second_per_gpu:5334.472096169319 mfu:25.163615032422474 +Step 23313 | loss:0.391803115606308 lr:2.8518975214325422e-05 tokens_per_second_per_gpu:5355.287862884248 mfu:25.261806555552816 +Step 23314 | loss:0.38540002703666687 lr:2.8515059645507826e-05 tokens_per_second_per_gpu:4576.583483181179 mfu:21.588525135826107 +Step 23315 | loss:0.4144074618816376 lr:2.851114423828123e-05 tokens_per_second_per_gpu:4456.045902995682 mfu:21.019928804259255 +Step 23316 | loss:0.4982537031173706 lr:2.8507228992675035e-05 tokens_per_second_per_gpu:4152.818764582473 mfu:19.5895546564795 +Step 23317 | loss:0.4418350160121918 lr:2.850331390871871e-05 tokens_per_second_per_gpu:5374.4689883412075 mfu:25.352287196970764 +Step 23318 | loss:0.37739747762680054 lr:2.8499398986441734e-05 tokens_per_second_per_gpu:5341.4129334131185 mfu:25.196356145930423 +Step 23319 | loss:0.38810622692108154 lr:2.8495484225873514e-05 tokens_per_second_per_gpu:4132.796291258364 mfu:19.495105233623462 +Step 23320 | loss:0.525305986404419 lr:2.8491569627043483e-05 tokens_per_second_per_gpu:4835.537614385058 mfu:22.810055954845506 +Step 23321 | loss:0.38732481002807617 lr:2.8487655189981106e-05 tokens_per_second_per_gpu:5133.818680901316 mfu:24.217098637600458 +Step 23322 | loss:0.4365748167037964 lr:2.848374091471582e-05 tokens_per_second_per_gpu:3667.559998625813 mfu:17.30050626377905 +Step 23323 | loss:0.4549691379070282 lr:2.8479826801277077e-05 tokens_per_second_per_gpu:5389.622822064525 mfu:25.423770416154294 +Step 23324 | loss:0.38312456011772156 lr:2.8475912849694287e-05 tokens_per_second_per_gpu:3711.8457938058054 mfu:17.509409915578875 +Step 23325 | loss:0.4179155230522156 lr:2.8471999059996918e-05 tokens_per_second_per_gpu:5005.221153844406 mfu:23.61048216147241 +Step 23326 | loss:0.3950042724609375 lr:2.8468085432214375e-05 tokens_per_second_per_gpu:5039.631458246913 mfu:23.772801438342366 +Step 23327 | loss:0.4166029393672943 lr:2.846417196637613e-05 tokens_per_second_per_gpu:4204.5738594361155 mfu:19.833692269233868 +Step 23328 | loss:0.5111679434776306 lr:2.8460258662511574e-05 tokens_per_second_per_gpu:4540.422165686491 mfu:21.417946031446693 +Step 23329 | loss:0.440904438495636 lr:2.8456345520650164e-05 tokens_per_second_per_gpu:5347.433609277504 mfu:25.224756701215956 +Step 23330 | loss:0.2857646942138672 lr:2.8452432540821338e-05 tokens_per_second_per_gpu:4930.4603773671315 mfu:23.257822823325334 +Step 23331 | loss:0.45143648982048035 lr:2.8448519723054506e-05 tokens_per_second_per_gpu:4325.517011248334 mfu:20.40420175136162 +Step 23332 | loss:0.4717985987663269 lr:2.8444607067379124e-05 tokens_per_second_per_gpu:3739.7954559830378 mfu:17.64125324077298 +Step 23333 | loss:0.3731918931007385 lr:2.8440694573824588e-05 tokens_per_second_per_gpu:3998.4168735548815 mfu:18.861214592823266 +Step 23334 | loss:0.42769181728363037 lr:2.843678224242035e-05 tokens_per_second_per_gpu:4358.099806028186 mfu:20.55790035353622 +Step 23335 | loss:0.43027135729789734 lr:2.843287007319581e-05 tokens_per_second_per_gpu:3144.125384785681 mfu:14.831375883140963 +Step 23336 | loss:0.42537227272987366 lr:2.8428958066180405e-05 tokens_per_second_per_gpu:5362.220791931265 mfu:25.294510364747268 +Step 23337 | loss:0.4918980598449707 lr:2.8425046221403583e-05 tokens_per_second_per_gpu:4415.4356686414185 mfu:20.828363399989737 +Step 23338 | loss:0.40199732780456543 lr:2.8421134538894712e-05 tokens_per_second_per_gpu:4954.04921696107 mfu:23.369095404360987 +Step 23339 | loss:0.34366944432258606 lr:2.841722301868325e-05 tokens_per_second_per_gpu:4205.030856796309 mfu:19.83584800375369 +Step 23340 | loss:0.4749281406402588 lr:2.8413311660798627e-05 tokens_per_second_per_gpu:5123.0026176525425 mfu:24.166077421845312 +Step 23341 | loss:0.4575507342815399 lr:2.8409400465270237e-05 tokens_per_second_per_gpu:3196.4315444477947 mfu:15.07811296261824 +Step 23342 | loss:0.41208067536354065 lr:2.8405489432127486e-05 tokens_per_second_per_gpu:5361.574094991689 mfu:25.291459784945413 +Step 23343 | loss:0.44248008728027344 lr:2.84015785613998e-05 tokens_per_second_per_gpu:3020.465656246415 mfu:14.248051845095945 +Step 23344 | loss:0.39238014817237854 lr:2.8397667853116616e-05 tokens_per_second_per_gpu:5193.047073772085 mfu:24.496489072177734 +Step 23345 | loss:0.43336185812950134 lr:2.8393757307307312e-05 tokens_per_second_per_gpu:5334.731793500816 mfu:25.164840069044075 +Step 23346 | loss:0.48527106642723083 lr:2.8389846924001316e-05 tokens_per_second_per_gpu:4674.885864402604 mfu:22.052234240163287 +Step 23347 | loss:0.3890696167945862 lr:2.838593670322806e-05 tokens_per_second_per_gpu:5355.416242412437 mfu:25.26241214369109 +Step 23348 | loss:0.4830070734024048 lr:2.8382026645016936e-05 tokens_per_second_per_gpu:4555.274351683024 mfu:21.488006326836576 +Step 23349 | loss:0.4280893802642822 lr:2.8378116749397322e-05 tokens_per_second_per_gpu:4980.570925743698 mfu:23.494202829758773 +Step 23350 | loss:0.31150269508361816 lr:2.8374207016398645e-05 tokens_per_second_per_gpu:4136.9072129555725 mfu:19.51449715266476 +Step 23351 | loss:0.4758259952068329 lr:2.837029744605035e-05 tokens_per_second_per_gpu:3969.7618017640775 mfu:18.726043730126655 +Step 23352 | loss:0.34335601329803467 lr:2.836638803838178e-05 tokens_per_second_per_gpu:4495.24837655387 mfu:21.204853560664926 +Step 23353 | loss:0.38814327120780945 lr:2.8362478793422363e-05 tokens_per_second_per_gpu:4397.243541867322 mfu:20.742548034099435 +Step 23354 | loss:0.473926305770874 lr:2.8358569711201533e-05 tokens_per_second_per_gpu:4736.31352249031 mfu:22.341998984002196 +Step 23355 | loss:0.32820162177085876 lr:2.8354660791748654e-05 tokens_per_second_per_gpu:5349.772792034725 mfu:25.235791025387503 +Step 23356 | loss:0.5304900407791138 lr:2.835075203509312e-05 tokens_per_second_per_gpu:4848.610764788094 mfu:22.871724235806084 +Step 23357 | loss:0.4250186085700989 lr:2.8346843441264337e-05 tokens_per_second_per_gpu:5158.070585031946 mfu:24.33149900718616 +Step 23358 | loss:0.43705040216445923 lr:2.834293501029171e-05 tokens_per_second_per_gpu:5334.66562926631 mfu:25.164527961061133 +Step 23359 | loss:0.4516677260398865 lr:2.833902674220465e-05 tokens_per_second_per_gpu:5335.86062373039 mfu:25.170164953835275 +Step 23360 | loss:0.40816208720207214 lr:2.8335118637032516e-05 tokens_per_second_per_gpu:4721.497038300613 mfu:22.27210710012652 +Step 23361 | loss:0.3537508547306061 lr:2.833121069480474e-05 tokens_per_second_per_gpu:4820.967424013277 mfu:22.741325880930994 +Step 23362 | loss:0.3297981321811676 lr:2.832730291555067e-05 tokens_per_second_per_gpu:5354.425990950163 mfu:25.257740958589224 +Step 23363 | loss:0.5395148396492004 lr:2.8323395299299736e-05 tokens_per_second_per_gpu:5347.171087568587 mfu:25.22351834152417 +Step 23364 | loss:0.5178683996200562 lr:2.8319487846081306e-05 tokens_per_second_per_gpu:3832.7419330390203 mfu:18.079697631350292 +Step 23365 | loss:0.4372161030769348 lr:2.8315580555924758e-05 tokens_per_second_per_gpu:5358.295085124499 mfu:25.2759921359451 +Step 23366 | loss:0.4071049392223358 lr:2.8311673428859532e-05 tokens_per_second_per_gpu:4200.654497718642 mfu:19.81520397130048 +Step 23367 | loss:0.3872242569923401 lr:2.8307766464914942e-05 tokens_per_second_per_gpu:5299.545188132525 mfu:24.99885884806792 +Step 23368 | loss:0.4111793339252472 lr:2.830385966412042e-05 tokens_per_second_per_gpu:4126.899902672741 mfu:19.467290962637446 +Step 23369 | loss:0.5321913957595825 lr:2.8299953026505353e-05 tokens_per_second_per_gpu:4778.265257481494 mfu:22.53989247565968 +Step 23370 | loss:0.4403662383556366 lr:2.8296046552099108e-05 tokens_per_second_per_gpu:4133.162992738657 mfu:19.496835027071725 +Step 23371 | loss:0.42859259247779846 lr:2.8292140240931047e-05 tokens_per_second_per_gpu:4234.823532571565 mfu:19.97638513854022 +Step 23372 | loss:0.3906593322753906 lr:2.8288234093030568e-05 tokens_per_second_per_gpu:4473.413418157979 mfu:21.101854336483456 +Step 23373 | loss:0.3296826183795929 lr:2.828432810842707e-05 tokens_per_second_per_gpu:3441.0337006341474 mfu:16.231943066780527 +Step 23374 | loss:0.4348313808441162 lr:2.82804222871499e-05 tokens_per_second_per_gpu:3471.3010728838135 mfu:16.37471942582842 +Step 23375 | loss:0.36581090092658997 lr:2.8276516629228433e-05 tokens_per_second_per_gpu:5125.660524115737 mfu:24.1786152201178 +Step 23376 | loss:0.4863378405570984 lr:2.8272611134692084e-05 tokens_per_second_per_gpu:5241.253973888176 mfu:24.723889244970625 +Step 23377 | loss:0.44543173909187317 lr:2.8268705803570196e-05 tokens_per_second_per_gpu:4896.38645451121 mfu:23.097090315603236 +Step 23378 | loss:0.4967385530471802 lr:2.826480063589213e-05 tokens_per_second_per_gpu:4865.313091014462 mfu:22.95051195832657 +Step 23379 | loss:0.3413057029247284 lr:2.8260895631687267e-05 tokens_per_second_per_gpu:3317.3577030433444 mfu:15.648542284843623 +Step 23380 | loss:0.5297884345054626 lr:2.8256990790985004e-05 tokens_per_second_per_gpu:4458.8551499076075 mfu:21.033180501249756 +Step 23381 | loss:0.36222922801971436 lr:2.8253086113814662e-05 tokens_per_second_per_gpu:5356.346026711874 mfu:25.266798094869095 +Step 23382 | loss:0.34433189034461975 lr:2.8249181600205644e-05 tokens_per_second_per_gpu:2961.873268529803 mfu:13.971661555344495 +Step 23383 | loss:0.3933756351470947 lr:2.8245277250187318e-05 tokens_per_second_per_gpu:5355.192276725477 mfu:25.26135565933322 +Step 23384 | loss:0.47818228602409363 lr:2.8241373063789038e-05 tokens_per_second_per_gpu:4129.416633358742 mfu:19.47916281068185 +Step 23385 | loss:0.3231457769870758 lr:2.8237469041040153e-05 tokens_per_second_per_gpu:4347.913858716688 mfu:20.50985150216606 +Step 23386 | loss:0.45721957087516785 lr:2.8233565181970034e-05 tokens_per_second_per_gpu:4586.825766545115 mfu:21.63683973396753 +Step 23387 | loss:0.34104907512664795 lr:2.8229661486608073e-05 tokens_per_second_per_gpu:4588.393333603988 mfu:21.644234215238754 +Step 23388 | loss:0.5156452655792236 lr:2.822575795498359e-05 tokens_per_second_per_gpu:3368.3493491116396 mfu:15.889078579419715 +Step 23389 | loss:0.39789482951164246 lr:2.8221854587125952e-05 tokens_per_second_per_gpu:3943.1932250404407 mfu:18.60071522065463 +Step 23390 | loss:0.4576089680194855 lr:2.8217951383064544e-05 tokens_per_second_per_gpu:4006.6371212847926 mfu:18.89999090388414 +Step 23391 | loss:0.45986098051071167 lr:2.8214048342828714e-05 tokens_per_second_per_gpu:5283.282928142364 mfu:24.92214699306666 +Step 23392 | loss:0.3823859989643097 lr:2.821014546644778e-05 tokens_per_second_per_gpu:4171.103224864179 mfu:19.675805575278915 +Step 23393 | loss:0.3927590250968933 lr:2.8206242753951117e-05 tokens_per_second_per_gpu:5295.413136128872 mfu:24.97936725376592 +Step 23394 | loss:0.41535139083862305 lr:2.8202340205368087e-05 tokens_per_second_per_gpu:5085.732397303397 mfu:23.990267433503085 +Step 23395 | loss:0.4787835478782654 lr:2.8198437820728057e-05 tokens_per_second_per_gpu:5299.1528563176735 mfu:24.997008151958553 +Step 23396 | loss:0.32822996377944946 lr:2.819453560006034e-05 tokens_per_second_per_gpu:4772.805358431139 mfu:22.51413720028432 +Step 23397 | loss:0.38934171199798584 lr:2.8190633543394324e-05 tokens_per_second_per_gpu:5314.806475346472 mfu:25.070848943700206 +Step 23398 | loss:0.36959290504455566 lr:2.8186731650759313e-05 tokens_per_second_per_gpu:4695.91043641586 mfu:22.1514107335117 +Step 23399 | loss:0.4883384704589844 lr:2.8182829922184695e-05 tokens_per_second_per_gpu:3565.3473439611234 mfu:16.81835173244854 +Step 23400 | loss:0.4336252212524414 lr:2.8178928357699786e-05 tokens_per_second_per_gpu:5320.058995617511 mfu:25.095625977991062 +Step 23401 | loss:0.364203542470932 lr:2.817502695733393e-05 tokens_per_second_per_gpu:4467.043241801413 mfu:21.07180512774537 +Step 23402 | loss:0.4956335127353668 lr:2.817112572111651e-05 tokens_per_second_per_gpu:4639.38843162591 mfu:21.88478679326907 +Step 23403 | loss:0.43227148056030273 lr:2.8167224649076817e-05 tokens_per_second_per_gpu:4340.4025923781255 mfu:20.474419577292878 +Step 23404 | loss:0.44402337074279785 lr:2.816332374124421e-05 tokens_per_second_per_gpu:4986.209122712299 mfu:23.520799166835264 +Step 23405 | loss:0.35661181807518005 lr:2.8159422997648053e-05 tokens_per_second_per_gpu:3636.5046447595473 mfu:17.15401286100213 +Step 23406 | loss:0.38815608620643616 lr:2.815552241831766e-05 tokens_per_second_per_gpu:5078.076034450832 mfu:23.954151063617328 +Step 23407 | loss:0.4827395975589752 lr:2.815162200328235e-05 tokens_per_second_per_gpu:3801.87453545949 mfu:17.93409085044655 +Step 23408 | loss:0.41315171122550964 lr:2.8147721752571486e-05 tokens_per_second_per_gpu:4573.86111762951 mfu:21.575683273035423 +Step 23409 | loss:0.3521992564201355 lr:2.814382166621441e-05 tokens_per_second_per_gpu:5322.907220383048 mfu:25.109061540167357 +Step 23410 | loss:0.433987021446228 lr:2.8139921744240422e-05 tokens_per_second_per_gpu:4349.66595876208 mfu:20.518116457019268 +Step 23411 | loss:0.35147881507873535 lr:2.8136021986678872e-05 tokens_per_second_per_gpu:4925.738971521208 mfu:23.23555114639521 +Step 23412 | loss:0.452396422624588 lr:2.813212239355911e-05 tokens_per_second_per_gpu:4796.014161497873 mfu:22.623617084180232 +Step 23413 | loss:0.38857024908065796 lr:2.8128222964910443e-05 tokens_per_second_per_gpu:4417.572847062602 mfu:20.838444835242697 +Step 23414 | loss:0.49841994047164917 lr:2.8124323700762183e-05 tokens_per_second_per_gpu:5334.742036971073 mfu:25.164888389240996 +Step 23415 | loss:0.5186214447021484 lr:2.812042460114368e-05 tokens_per_second_per_gpu:4808.487628950301 mfu:22.682456558346335 +Step 23416 | loss:0.5597056746482849 lr:2.811652566608427e-05 tokens_per_second_per_gpu:4565.1087557229985 mfu:21.534396888615483 +Step 23417 | loss:0.3849157691001892 lr:2.8112626895613248e-05 tokens_per_second_per_gpu:4301.1936247828025 mfu:20.289464187406043 +Step 23418 | loss:0.4051155149936676 lr:2.8108728289759943e-05 tokens_per_second_per_gpu:4453.93523298951 mfu:21.009972413722707 +Step 23419 | loss:0.5002543926239014 lr:2.8104829848553715e-05 tokens_per_second_per_gpu:4060.0559803583783 mfu:19.15197677633143 +Step 23420 | loss:0.5537160038948059 lr:2.8100931572023852e-05 tokens_per_second_per_gpu:3731.892228086492 mfu:17.603972366355247 +Step 23421 | loss:0.3695521354675293 lr:2.809703346019966e-05 tokens_per_second_per_gpu:4110.044371640735 mfu:19.387780546909333 +Step 23422 | loss:0.3839053511619568 lr:2.809313551311047e-05 tokens_per_second_per_gpu:3942.9747760238824 mfu:18.599684759372284 +Step 23423 | loss:0.4274250566959381 lr:2.8089237730785624e-05 tokens_per_second_per_gpu:5033.783036521947 mfu:23.745213435222063 +Step 23424 | loss:0.46099039912223816 lr:2.80853401132544e-05 tokens_per_second_per_gpu:4489.269665501521 mfu:21.1766509605577 +Step 23425 | loss:0.2700786888599396 lr:2.8081442660546125e-05 tokens_per_second_per_gpu:5315.6645338275775 mfu:25.07489655195545 +Step 23426 | loss:0.356117308139801 lr:2.807754537269014e-05 tokens_per_second_per_gpu:5354.709505876681 mfu:25.25907834687024 +Step 23427 | loss:0.4641636312007904 lr:2.8073648249715734e-05 tokens_per_second_per_gpu:5348.286109744085 mfu:25.228778091368397 +Step 23428 | loss:0.39011722803115845 lr:2.806975129165219e-05 tokens_per_second_per_gpu:4763.692560747473 mfu:22.471150578806856 +Step 23429 | loss:0.4773326516151428 lr:2.8065854498528853e-05 tokens_per_second_per_gpu:3953.363822239622 mfu:18.64869166292599 +Step 23430 | loss:0.46648716926574707 lr:2.8061957870375023e-05 tokens_per_second_per_gpu:4930.139567785227 mfu:23.256309509791734 +Step 23431 | loss:0.48279017210006714 lr:2.8058061407220022e-05 tokens_per_second_per_gpu:4756.975981144834 mfu:22.43946732685456 +Step 23432 | loss:0.30884313583374023 lr:2.805416510909312e-05 tokens_per_second_per_gpu:5314.108089749931 mfu:25.06755453968445 +Step 23433 | loss:0.41734579205513 lr:2.805026897602365e-05 tokens_per_second_per_gpu:5226.899879774607 mfu:24.656178533975922 +Step 23434 | loss:0.4260800778865814 lr:2.8046373008040927e-05 tokens_per_second_per_gpu:5336.939027847769 mfu:25.175251970051665 +Step 23435 | loss:0.3670329451560974 lr:2.8042477205174228e-05 tokens_per_second_per_gpu:3531.998951746733 mfu:16.661041676550834 +Step 23436 | loss:0.46713775396347046 lr:2.8038581567452844e-05 tokens_per_second_per_gpu:4975.268815753371 mfu:23.469191872301156 +Step 23437 | loss:0.42692825198173523 lr:2.8034686094906087e-05 tokens_per_second_per_gpu:4915.471896049364 mfu:23.18711959965081 +Step 23438 | loss:0.47874215245246887 lr:2.8030790787563278e-05 tokens_per_second_per_gpu:5308.985531083323 mfu:25.043390556455424 +Step 23439 | loss:0.41659635305404663 lr:2.8026895645453675e-05 tokens_per_second_per_gpu:3930.6310186210867 mfu:18.54145714964111 +Step 23440 | loss:0.5029523968696594 lr:2.80230006686066e-05 tokens_per_second_per_gpu:3584.9140146258983 mfu:16.910651056392545 +Step 23441 | loss:0.47485101222991943 lr:2.801910585705136e-05 tokens_per_second_per_gpu:3637.745335400498 mfu:17.159865410439146 +Step 23442 | loss:0.5138148665428162 lr:2.8015211210817226e-05 tokens_per_second_per_gpu:4321.859851267517 mfu:20.386950303756272 +Step 23443 | loss:0.50186687707901 lr:2.8011316729933472e-05 tokens_per_second_per_gpu:3354.5066657553734 mfu:15.823780280222884 +Step 23444 | loss:0.4576467275619507 lr:2.800742241442942e-05 tokens_per_second_per_gpu:3337.9425269952253 mfu:15.745644411557771 +Step 23445 | loss:0.3569352924823761 lr:2.8003528264334355e-05 tokens_per_second_per_gpu:4473.908304660774 mfu:21.10418880055348 +Step 23446 | loss:0.5475246906280518 lr:2.7999634279677555e-05 tokens_per_second_per_gpu:4422.217361039452 mfu:20.860353800107298 +Step 23447 | loss:0.5733948945999146 lr:2.7995740460488306e-05 tokens_per_second_per_gpu:5000.520127014804 mfu:23.588306615839294 +Step 23448 | loss:0.5854122042655945 lr:2.799184680679592e-05 tokens_per_second_per_gpu:4340.124552536733 mfu:20.47310801592254 +Step 23449 | loss:0.4448010325431824 lr:2.7987953318629666e-05 tokens_per_second_per_gpu:4763.871927880842 mfu:22.471996684178148 +Step 23450 | loss:0.33639776706695557 lr:2.7984059996018802e-05 tokens_per_second_per_gpu:5206.22341738259 mfu:24.558644133104526 +Step 23451 | loss:0.4123772382736206 lr:2.7980166838992626e-05 tokens_per_second_per_gpu:5334.380588568498 mfu:25.163183375456594 +Step 23452 | loss:0.3325289487838745 lr:2.7976273847580457e-05 tokens_per_second_per_gpu:5338.294723422328 mfu:25.18164701738238 +Step 23453 | loss:0.4443906545639038 lr:2.7972381021811517e-05 tokens_per_second_per_gpu:4704.246191463038 mfu:22.190731912295703 +Step 23454 | loss:0.4291746914386749 lr:2.7968488361715105e-05 tokens_per_second_per_gpu:3765.054341937653 mfu:17.760403715431234 +Step 23455 | loss:0.5071345567703247 lr:2.796459586732053e-05 tokens_per_second_per_gpu:4678.532703357893 mfu:22.06943700173028 +Step 23456 | loss:0.45714646577835083 lr:2.796070353865704e-05 tokens_per_second_per_gpu:4457.998988566568 mfu:21.02914184212789 +Step 23457 | loss:0.4293731451034546 lr:2.7956811375753888e-05 tokens_per_second_per_gpu:5274.310770924812 mfu:24.879823796663594 +Step 23458 | loss:0.3880723714828491 lr:2.7952919378640373e-05 tokens_per_second_per_gpu:4702.3007912079 mfu:22.181555127372707 +Step 23459 | loss:0.4701979458332062 lr:2.7949027547345786e-05 tokens_per_second_per_gpu:5336.967790718247 mfu:25.175387649419164 +Step 23460 | loss:0.38628458976745605 lr:2.7945135881899355e-05 tokens_per_second_per_gpu:4705.330447902607 mfu:22.195846534914377 +Step 23461 | loss:0.3785332143306732 lr:2.7941244382330374e-05 tokens_per_second_per_gpu:4332.28840946803 mfu:20.436143592083596 +Step 23462 | loss:0.38760703802108765 lr:2.7937353048668103e-05 tokens_per_second_per_gpu:5335.22116213377 mfu:25.167148504380624 +Step 23463 | loss:0.38335445523262024 lr:2.7933461880941858e-05 tokens_per_second_per_gpu:4479.010979517358 mfu:21.12825898845809 +Step 23464 | loss:0.34313100576400757 lr:2.792957087918082e-05 tokens_per_second_per_gpu:5349.860260103782 mfu:25.236203627192832 +Step 23465 | loss:0.40677934885025024 lr:2.79256800434143e-05 tokens_per_second_per_gpu:4085.2499787810175 mfu:19.2708211654304 +Step 23466 | loss:0.4298468828201294 lr:2.7921789373671553e-05 tokens_per_second_per_gpu:5346.567360748157 mfu:25.220670459105786 +Step 23467 | loss:0.4600229263305664 lr:2.791789886998187e-05 tokens_per_second_per_gpu:5241.875104546165 mfu:24.72681922426018 +Step 23468 | loss:0.34044453501701355 lr:2.7914008532374457e-05 tokens_per_second_per_gpu:4466.643986203328 mfu:21.069921770969174 +Step 23469 | loss:0.440938800573349 lr:2.7910118360878612e-05 tokens_per_second_per_gpu:3755.074850960604 mfu:17.713328753815762 +Step 23470 | loss:0.4779941737651825 lr:2.7906228355523605e-05 tokens_per_second_per_gpu:5116.080721188608 mfu:24.13342565522761 +Step 23471 | loss:0.48363786935806274 lr:2.7902338516338677e-05 tokens_per_second_per_gpu:4263.22485959602 mfu:20.110359043880642 +Step 23472 | loss:0.3203146457672119 lr:2.789844884335305e-05 tokens_per_second_per_gpu:5317.344106561573 mfu:25.082819383106365 +Step 23473 | loss:0.46382394433021545 lr:2.7894559336596023e-05 tokens_per_second_per_gpu:4245.968708335621 mfu:20.02895883418234 +Step 23474 | loss:0.3652277886867523 lr:2.7890669996096847e-05 tokens_per_second_per_gpu:3984.8852094757535 mfu:18.79738342462206 +Step 23475 | loss:0.5029817223548889 lr:2.7886780821884746e-05 tokens_per_second_per_gpu:5354.722469865845 mfu:25.259139500219014 +Step 23476 | loss:0.3667774200439453 lr:2.7882891813988988e-05 tokens_per_second_per_gpu:5368.466438055256 mfu:25.323972143131243 +Step 23477 | loss:0.5038127899169922 lr:2.787900297243884e-05 tokens_per_second_per_gpu:4182.401341292242 mfu:19.72910071812753 +Step 23478 | loss:0.4240524470806122 lr:2.7875114297263538e-05 tokens_per_second_per_gpu:5337.057478139914 mfu:25.17581072028942 +Step 23479 | loss:0.43415170907974243 lr:2.78712257884923e-05 tokens_per_second_per_gpu:4317.602704591939 mfu:20.366868616543446 +Step 23480 | loss:0.29977351427078247 lr:2.7867337446154396e-05 tokens_per_second_per_gpu:4266.551874052758 mfu:20.126053138720277 +Step 23481 | loss:0.23886626958847046 lr:2.7863449270279095e-05 tokens_per_second_per_gpu:5328.2287153582465 mfu:25.134163939906035 +Step 23482 | loss:0.5220602750778198 lr:2.7859561260895594e-05 tokens_per_second_per_gpu:3827.231662861348 mfu:18.053704746773175 +Step 23483 | loss:0.42837509512901306 lr:2.7855673418033163e-05 tokens_per_second_per_gpu:4640.1077614677515 mfu:21.88817999486336 +Step 23484 | loss:0.3315844237804413 lr:2.7851785741721055e-05 tokens_per_second_per_gpu:3896.043331090989 mfu:18.37830112629363 +Step 23485 | loss:0.47280165553092957 lr:2.784789823198849e-05 tokens_per_second_per_gpu:3697.5641654692204 mfu:17.442041038018143 +Step 23486 | loss:0.41359183192253113 lr:2.784401088886469e-05 tokens_per_second_per_gpu:4456.932785328748 mfu:21.024112379542053 +Step 23487 | loss:0.4035343527793884 lr:2.7840123712378914e-05 tokens_per_second_per_gpu:4172.833322420827 mfu:19.68396674063871 +Step 23488 | loss:0.460445374250412 lr:2.7836236702560414e-05 tokens_per_second_per_gpu:4977.0960033896545 mfu:23.477811028131857 +Step 23489 | loss:0.40739649534225464 lr:2.7832349859438384e-05 tokens_per_second_per_gpu:4641.733128595363 mfu:21.895847128920533 +Step 23490 | loss:0.4569510817527771 lr:2.782846318304208e-05 tokens_per_second_per_gpu:3686.7666826514856 mfu:17.391107469326354 +Step 23491 | loss:0.3885591924190521 lr:2.7824576673400753e-05 tokens_per_second_per_gpu:3597.1364537818376 mfu:16.968306387254266 +Step 23492 | loss:0.504682719707489 lr:2.782069033054362e-05 tokens_per_second_per_gpu:4886.978506833501 mfu:23.052711421246087 +Step 23493 | loss:0.36603328585624695 lr:2.7816804154499875e-05 tokens_per_second_per_gpu:5377.688369421588 mfu:25.367473566809966 +Step 23494 | loss:0.359358549118042 lr:2.7812918145298783e-05 tokens_per_second_per_gpu:5312.224423169399 mfu:25.058668963037423 +Step 23495 | loss:0.44685500860214233 lr:2.7809032302969586e-05 tokens_per_second_per_gpu:5299.318787491912 mfu:24.99779087761821 +Step 23496 | loss:0.4402383267879486 lr:2.7805146627541467e-05 tokens_per_second_per_gpu:4695.873070829262 mfu:22.151234473664495 +Step 23497 | loss:0.44338715076446533 lr:2.780126111904367e-05 tokens_per_second_per_gpu:3799.790161985889 mfu:17.92425850514079 +Step 23498 | loss:0.40256980061531067 lr:2.7797375777505418e-05 tokens_per_second_per_gpu:5316.394329868174 mfu:25.07833912439499 +Step 23499 | loss:0.4604377746582031 lr:2.7793490602955985e-05 tokens_per_second_per_gpu:4603.66697982781 mfu:21.716282610429285 +Step 23500 | loss:0.48013725876808167 lr:2.7789605595424496e-05 tokens_per_second_per_gpu:3729.7945856369984 mfu:17.594077429026278 +Step 23501 | loss:0.4577735364437103 lr:2.778572075494022e-05 tokens_per_second_per_gpu:5376.643074880365 mfu:25.362542734113383 +Step 23502 | loss:0.47516223788261414 lr:2.7781836081532377e-05 tokens_per_second_per_gpu:5092.2175546761055 mfu:24.020859027312092 +Step 23503 | loss:0.4409923553466797 lr:2.7777951575230188e-05 tokens_per_second_per_gpu:4160.047190045206 mfu:19.623652372683765 +Step 23504 | loss:0.3705407977104187 lr:2.777406723606285e-05 tokens_per_second_per_gpu:4706.592716814145 mfu:22.201800872735387 +Step 23505 | loss:0.4081767201423645 lr:2.7770183064059585e-05 tokens_per_second_per_gpu:5318.342589766883 mfu:25.087529398744614 +Step 23506 | loss:0.416871041059494 lr:2.7766299059249652e-05 tokens_per_second_per_gpu:5213.912172747722 mfu:24.59491330399879 +Step 23507 | loss:0.35002636909484863 lr:2.7762415221662174e-05 tokens_per_second_per_gpu:5330.666873122981 mfu:25.145665146074897 +Step 23508 | loss:0.407542884349823 lr:2.7758531551326417e-05 tokens_per_second_per_gpu:3489.3715090745236 mfu:16.459960756474334 +Step 23509 | loss:0.34753963351249695 lr:2.7754648048271582e-05 tokens_per_second_per_gpu:4150.635794443854 mfu:19.579257213882524 +Step 23510 | loss:0.42434173822402954 lr:2.7750764712526894e-05 tokens_per_second_per_gpu:4757.746726624273 mfu:22.44310306478381 +Step 23511 | loss:0.3252342641353607 lr:2.774688154412153e-05 tokens_per_second_per_gpu:4032.1077626509423 mfu:19.020140264947806 +Step 23512 | loss:0.37976154685020447 lr:2.7742998543084698e-05 tokens_per_second_per_gpu:3126.085062335471 mfu:14.746276604146917 +Step 23513 | loss:0.32844972610473633 lr:2.773911570944564e-05 tokens_per_second_per_gpu:5317.6236357071175 mfu:25.084137969778492 +Step 23514 | loss:0.4598689675331116 lr:2.7735233043233533e-05 tokens_per_second_per_gpu:4038.334106395274 mfu:19.049510990713607 +Step 23515 | loss:0.2672433853149414 lr:2.7731350544477563e-05 tokens_per_second_per_gpu:2754.7362299452216 mfu:12.99456080311797 +Step 23516 | loss:0.3574081063270569 lr:2.7727468213206942e-05 tokens_per_second_per_gpu:4849.55411105624 mfu:22.8761741611044 +Step 23517 | loss:0.48235467076301575 lr:2.77235860494509e-05 tokens_per_second_per_gpu:5134.356333321461 mfu:24.21963483579924 +Step 23518 | loss:0.3558112680912018 lr:2.771970405323858e-05 tokens_per_second_per_gpu:4130.9743989620065 mfu:19.486511056814667 +Step 23519 | loss:0.38280537724494934 lr:2.771582222459922e-05 tokens_per_second_per_gpu:5034.104773990066 mfu:23.74673112575325 +Step 23520 | loss:0.4980846643447876 lr:2.7711940563562015e-05 tokens_per_second_per_gpu:4091.6846906290684 mfu:19.301174799092706 +Step 23521 | loss:0.4020625352859497 lr:2.7708059070156155e-05 tokens_per_second_per_gpu:5321.9698359288295 mfu:25.10463973776253 +Step 23522 | loss:0.4242439568042755 lr:2.7704177744410808e-05 tokens_per_second_per_gpu:4018.41866119898 mfu:18.955566437797355 +Step 23523 | loss:0.4519699215888977 lr:2.770029658635518e-05 tokens_per_second_per_gpu:4389.675502256163 mfu:20.706848300012886 +Step 23524 | loss:0.31842905282974243 lr:2.7696415596018486e-05 tokens_per_second_per_gpu:3275.8603545272445 mfu:15.452792211715439 +Step 23525 | loss:0.5401764512062073 lr:2.7692534773429883e-05 tokens_per_second_per_gpu:4747.1839950964395 mfu:22.39327685797919 +Step 23526 | loss:0.5323793292045593 lr:2.7688654118618574e-05 tokens_per_second_per_gpu:3686.669419312907 mfu:17.390648661563443 +Step 23527 | loss:0.3615923821926117 lr:2.768477363161376e-05 tokens_per_second_per_gpu:4625.148321404696 mfu:21.817613763742262 +Step 23528 | loss:0.4018213152885437 lr:2.7680893312444605e-05 tokens_per_second_per_gpu:4315.444412198561 mfu:20.35668758308123 +Step 23529 | loss:0.3621206283569336 lr:2.7677013161140285e-05 tokens_per_second_per_gpu:4676.731901771771 mfu:22.060942313400105 +Step 23530 | loss:0.3501591384410858 lr:2.7673133177729997e-05 tokens_per_second_per_gpu:5344.814861574271 mfu:25.212403621495376 +Step 23531 | loss:0.3817724585533142 lr:2.766925336224294e-05 tokens_per_second_per_gpu:4979.868340529505 mfu:23.490888615430027 +Step 23532 | loss:0.47227662801742554 lr:2.7665373714708265e-05 tokens_per_second_per_gpu:4039.23663220395 mfu:19.053768358939916 +Step 23533 | loss:0.36333584785461426 lr:2.766149423515516e-05 tokens_per_second_per_gpu:4885.656001805002 mfu:23.04647293938403 +Step 23534 | loss:0.4852609932422638 lr:2.7657614923612806e-05 tokens_per_second_per_gpu:5330.471263861033 mfu:25.144742423811905 +Step 23535 | loss:0.4748961627483368 lr:2.7653735780110424e-05 tokens_per_second_per_gpu:4062.1110437897196 mfu:19.161670861167046 +Step 23536 | loss:0.5023590326309204 lr:2.764985680467711e-05 tokens_per_second_per_gpu:4614.126117412637 mfu:21.7656201469298 +Step 23537 | loss:0.4931882619857788 lr:2.7645977997342064e-05 tokens_per_second_per_gpu:4465.776558524258 mfu:21.06582996660858 +Step 23538 | loss:0.42372220754623413 lr:2.76420993581345e-05 tokens_per_second_per_gpu:4279.739116708067 mfu:20.188259612300975 +Step 23539 | loss:0.5006740689277649 lr:2.763822088708354e-05 tokens_per_second_per_gpu:4753.314890610034 mfu:22.42219733815545 +Step 23540 | loss:0.5342109799385071 lr:2.7634342584218365e-05 tokens_per_second_per_gpu:4322.361675192417 mfu:20.389317492830013 +Step 23541 | loss:0.4819270968437195 lr:2.7630464449568162e-05 tokens_per_second_per_gpu:4690.000435800914 mfu:22.123532252260915 +Step 23542 | loss:0.38214677572250366 lr:2.7626586483162127e-05 tokens_per_second_per_gpu:3872.1099230377363 mfu:18.26540315704595 +Step 23543 | loss:0.4421961307525635 lr:2.762270868502935e-05 tokens_per_second_per_gpu:5059.1510633161015 mfu:23.86487874584177 +Step 23544 | loss:0.5191441178321838 lr:2.7618831055199035e-05 tokens_per_second_per_gpu:3991.078020832658 mfu:18.826595972395147 +Step 23545 | loss:0.5106181502342224 lr:2.7614953593700345e-05 tokens_per_second_per_gpu:5019.495503165452 mfu:23.677816702674917 +Step 23546 | loss:0.4002418518066406 lr:2.7611076300562467e-05 tokens_per_second_per_gpu:5340.004735129307 mfu:25.189713434362282 +Step 23547 | loss:0.46751055121421814 lr:2.7607199175814525e-05 tokens_per_second_per_gpu:4210.365415675876 mfu:19.861012028158203 +Step 23548 | loss:0.3984576463699341 lr:2.7603322219485695e-05 tokens_per_second_per_gpu:3845.6228525672886 mfu:18.140459126476625 +Step 23549 | loss:0.41365450620651245 lr:2.7599445431605152e-05 tokens_per_second_per_gpu:3796.2201093696826 mfu:17.90741795783636 +Step 23550 | loss:0.36692970991134644 lr:2.7595568812202044e-05 tokens_per_second_per_gpu:4857.872267100893 mfu:22.915412322390793 +Step 23551 | loss:0.44032472372055054 lr:2.75916923613055e-05 tokens_per_second_per_gpu:5302.846606444899 mfu:25.014432201527434 +Step 23552 | loss:0.31158846616744995 lr:2.7587816078944695e-05 tokens_per_second_per_gpu:4314.944108326899 mfu:20.35432756435796 +Step 23553 | loss:0.31401026248931885 lr:2.7583939965148804e-05 tokens_per_second_per_gpu:5349.062650176569 mfu:25.232441165079884 +Step 23554 | loss:0.3391711711883545 lr:2.758006401994694e-05 tokens_per_second_per_gpu:4094.9283864760737 mfu:19.31647586583461 +Step 23555 | loss:0.37694698572158813 lr:2.757618824336828e-05 tokens_per_second_per_gpu:4739.129709519654 mfu:22.35528341871067 +Step 23556 | loss:0.3884223401546478 lr:2.757231263544199e-05 tokens_per_second_per_gpu:3947.708925755654 mfu:18.622016551386285 +Step 23557 | loss:0.40407615900039673 lr:2.7568437196197194e-05 tokens_per_second_per_gpu:5359.863145792374 mfu:25.28338894565421 +Step 23558 | loss:0.41084933280944824 lr:2.7564561925663023e-05 tokens_per_second_per_gpu:5104.635468116125 mfu:24.07943644372317 +Step 23559 | loss:0.4255655109882355 lr:2.7560686823868646e-05 tokens_per_second_per_gpu:3871.1973870737534 mfu:18.261098569209153 +Step 23560 | loss:0.4572695791721344 lr:2.7556811890843233e-05 tokens_per_second_per_gpu:5262.525142943105 mfu:24.8242289786384 +Step 23561 | loss:0.45817434787750244 lr:2.7552937126615874e-05 tokens_per_second_per_gpu:4305.926007814024 mfu:20.311787645592116 +Step 23562 | loss:0.5041452646255493 lr:2.754906253121574e-05 tokens_per_second_per_gpu:5327.524921736763 mfu:25.13084402531392 +Step 23563 | loss:0.3638594150543213 lr:2.7545188104671994e-05 tokens_per_second_per_gpu:4298.081745830156 mfu:20.27478492344511 +Step 23564 | loss:0.4380389451980591 lr:2.754131384701375e-05 tokens_per_second_per_gpu:3572.320985691872 mfu:16.85124759031845 +Step 23565 | loss:0.4381144940853119 lr:2.7537439758270134e-05 tokens_per_second_per_gpu:5336.683653383667 mfu:25.17404732513312 +Step 23566 | loss:0.4737463593482971 lr:2.75335658384703e-05 tokens_per_second_per_gpu:4623.359404990463 mfu:21.809175139795713 +Step 23567 | loss:0.43367788195610046 lr:2.7529692087643403e-05 tokens_per_second_per_gpu:4900.122288897649 mfu:23.11471288380314 +Step 23568 | loss:0.5272216200828552 lr:2.752581850581854e-05 tokens_per_second_per_gpu:4585.438561614285 mfu:21.63029605162724 +Step 23569 | loss:0.4397483468055725 lr:2.7521945093024866e-05 tokens_per_second_per_gpu:3147.6858308505984 mfu:14.848171114703916 +Step 23570 | loss:0.45847493410110474 lr:2.7518071849291504e-05 tokens_per_second_per_gpu:5138.401096503761 mfu:24.23871467383791 +Step 23571 | loss:0.4101177155971527 lr:2.7514198774647635e-05 tokens_per_second_per_gpu:5347.116574587828 mfu:25.22326119449306 +Step 23572 | loss:0.46438702940940857 lr:2.751032586912231e-05 tokens_per_second_per_gpu:4090.1264151909886 mfu:19.293824147982082 +Step 23573 | loss:0.3949703276157379 lr:2.7506453132744693e-05 tokens_per_second_per_gpu:5366.246416084064 mfu:25.31349992071832 +Step 23574 | loss:0.41309303045272827 lr:2.7502580565543928e-05 tokens_per_second_per_gpu:4432.435052515278 mfu:20.9085523941167 +Step 23575 | loss:0.4225190579891205 lr:2.7498708167549107e-05 tokens_per_second_per_gpu:5328.761837670919 mfu:25.136678768814956 +Step 23576 | loss:0.4304875433444977 lr:2.7494835938789376e-05 tokens_per_second_per_gpu:3730.3714907770723 mfu:17.596798789002175 +Step 23577 | loss:0.3622717261314392 lr:2.749096387929385e-05 tokens_per_second_per_gpu:5384.960562206858 mfu:25.401777740942123 +Step 23578 | loss:0.3946188986301422 lr:2.7487091989091708e-05 tokens_per_second_per_gpu:5327.151789096226 mfu:25.129083894985516 +Step 23579 | loss:0.3659687042236328 lr:2.748322026821197e-05 tokens_per_second_per_gpu:5331.917601456801 mfu:25.151565045021783 +Step 23580 | loss:0.6063692569732666 lr:2.7479348716683807e-05 tokens_per_second_per_gpu:4083.6306505514644 mfu:19.263182517885713 +Step 23581 | loss:0.4815008044242859 lr:2.747547733453634e-05 tokens_per_second_per_gpu:4999.27213347923 mfu:23.58241961740234 +Step 23582 | loss:0.42308464646339417 lr:2.7471606121798692e-05 tokens_per_second_per_gpu:4590.160928244478 mfu:21.652572260740687 +Step 23583 | loss:0.41655847430229187 lr:2.7467735078499958e-05 tokens_per_second_per_gpu:5356.620394299703 mfu:25.268092333592822 +Step 23584 | loss:0.4011206030845642 lr:2.746386420466926e-05 tokens_per_second_per_gpu:3429.5312528170293 mfu:16.177684057907214 +Step 23585 | loss:0.481407105922699 lr:2.7459993500335734e-05 tokens_per_second_per_gpu:5312.833658639985 mfu:25.061542830698464 +Step 23586 | loss:0.5064240097999573 lr:2.7456122965528475e-05 tokens_per_second_per_gpu:5083.725849922488 mfu:23.98080220715505 +Step 23587 | loss:0.42377719283103943 lr:2.7452252600276573e-05 tokens_per_second_per_gpu:4664.162201970097 mfu:22.001648894823678 +Step 23588 | loss:0.42717915773391724 lr:2.7448382404609157e-05 tokens_per_second_per_gpu:3767.871718496929 mfu:17.77369376135535 +Step 23589 | loss:0.4668293595314026 lr:2.7444512378555352e-05 tokens_per_second_per_gpu:4559.106757662662 mfu:21.50608443971841 +Step 23590 | loss:0.4204833209514618 lr:2.744064252214423e-05 tokens_per_second_per_gpu:5341.38642389476 mfu:25.196231096009335 +Step 23591 | loss:0.274370402097702 lr:2.743677283540491e-05 tokens_per_second_per_gpu:5314.715302401053 mfu:25.070418865360953 +Step 23592 | loss:0.46616703271865845 lr:2.743290331836652e-05 tokens_per_second_per_gpu:5318.378337897053 mfu:25.087698028773886 +Step 23593 | loss:0.37998297810554504 lr:2.742903397105815e-05 tokens_per_second_per_gpu:5112.184683334591 mfu:24.11504737993761 +Step 23594 | loss:0.5565013289451599 lr:2.7425164793508872e-05 tokens_per_second_per_gpu:5358.827587804377 mfu:25.278504041940412 +Step 23595 | loss:0.46088093519210815 lr:2.7421295785747814e-05 tokens_per_second_per_gpu:3707.0628488590587 mfu:17.48684794821051 +Step 23596 | loss:0.3275533616542816 lr:2.7417426947804088e-05 tokens_per_second_per_gpu:5340.3839876757675 mfu:25.191502433331095 +Step 23597 | loss:0.477629691362381 lr:2.7413558279706752e-05 tokens_per_second_per_gpu:5012.879034324205 mfu:23.646605690264764 +Step 23598 | loss:0.3122013807296753 lr:2.7409689781484933e-05 tokens_per_second_per_gpu:4132.925764908279 mfu:19.495715982920593 +Step 23599 | loss:0.44792231917381287 lr:2.7405821453167735e-05 tokens_per_second_per_gpu:3918.526352761906 mfu:18.4843573755144 +Step 23600 | loss:0.4360780715942383 lr:2.7401953294784238e-05 tokens_per_second_per_gpu:5313.763718281169 mfu:25.065930080710466 +Step 23601 | loss:0.4747258126735687 lr:2.7398085306363513e-05 tokens_per_second_per_gpu:4841.083786045942 mfu:22.836218192844356 +Step 23602 | loss:0.44411811232566833 lr:2.7394217487934677e-05 tokens_per_second_per_gpu:4488.2063689830675 mfu:21.171635209463066 +Step 23603 | loss:0.4082745313644409 lr:2.739034983952683e-05 tokens_per_second_per_gpu:4205.453521653719 mfu:19.837841786000155 +Step 23604 | loss:0.3898921012878418 lr:2.738648236116903e-05 tokens_per_second_per_gpu:3690.0480099880474 mfu:17.40658605022509 +Step 23605 | loss:0.4394725263118744 lr:2.7382615052890383e-05 tokens_per_second_per_gpu:5319.478222427678 mfu:25.092886371764497 +Step 23606 | loss:0.4041825830936432 lr:2.737874791471999e-05 tokens_per_second_per_gpu:3457.2450116420564 mfu:16.30841458673971 +Step 23607 | loss:0.46643513441085815 lr:2.7374880946686922e-05 tokens_per_second_per_gpu:3001.371397701416 mfu:14.15798096972274 +Step 23608 | loss:0.4296606779098511 lr:2.737101414882024e-05 tokens_per_second_per_gpu:4902.934931056956 mfu:23.127980596754828 +Step 23609 | loss:0.477479487657547 lr:2.736714752114905e-05 tokens_per_second_per_gpu:5320.062602641164 mfu:25.095642992937055 +Step 23610 | loss:0.535346508026123 lr:2.7363281063702452e-05 tokens_per_second_per_gpu:5320.79923779498 mfu:25.099117826640526 +Step 23611 | loss:0.3723011314868927 lr:2.7359414776509485e-05 tokens_per_second_per_gpu:3875.5361963654577 mfu:18.28156547291532 +Step 23612 | loss:0.4378359019756317 lr:2.735554865959925e-05 tokens_per_second_per_gpu:4211.691536050267 mfu:19.867267564223678 +Step 23613 | loss:0.3825865387916565 lr:2.735168271300082e-05 tokens_per_second_per_gpu:4205.478285923716 mfu:19.83795860328695 +Step 23614 | loss:0.43604743480682373 lr:2.734781693674332e-05 tokens_per_second_per_gpu:4341.495578882432 mfu:20.479575381116415 +Step 23615 | loss:0.40503424406051636 lr:2.734395133085573e-05 tokens_per_second_per_gpu:2707.2423434836464 mfu:12.770524037386105 +Step 23616 | loss:0.34565380215644836 lr:2.734008589536718e-05 tokens_per_second_per_gpu:5391.1455730033595 mfu:25.430953492140876 +Step 23617 | loss:0.37578344345092773 lr:2.733622063030673e-05 tokens_per_second_per_gpu:4468.091289762001 mfu:21.076748948790883 +Step 23618 | loss:0.42254483699798584 lr:2.733235553570348e-05 tokens_per_second_per_gpu:4754.874531260677 mfu:22.429554429206668 +Step 23619 | loss:0.6469424366950989 lr:2.7328490611586455e-05 tokens_per_second_per_gpu:4405.446193774347 mfu:20.7812413426616 +Step 23620 | loss:0.45177316665649414 lr:2.7324625857984752e-05 tokens_per_second_per_gpu:5296.971784003268 mfu:24.98671966172219 +Step 23621 | loss:0.3426401615142822 lr:2.732076127492744e-05 tokens_per_second_per_gpu:5371.710397613759 mfu:25.339274453845274 +Step 23622 | loss:0.43408119678497314 lr:2.7316896862443585e-05 tokens_per_second_per_gpu:5359.460463878278 mfu:25.281489426360796 +Step 23623 | loss:0.48112133145332336 lr:2.7313032620562217e-05 tokens_per_second_per_gpu:5291.708692298502 mfu:24.961892760174766 +Step 23624 | loss:0.401742160320282 lr:2.7309168549312432e-05 tokens_per_second_per_gpu:4809.859657903657 mfu:22.68892865301159 +Step 23625 | loss:0.39742207527160645 lr:2.7305304648723302e-05 tokens_per_second_per_gpu:4096.514874451612 mfu:19.323959600297986 +Step 23626 | loss:0.4418230652809143 lr:2.730144091882385e-05 tokens_per_second_per_gpu:5218.496222227149 mfu:24.616537047512736 +Step 23627 | loss:0.5395281910896301 lr:2.729757735964316e-05 tokens_per_second_per_gpu:3507.2703058958973 mfu:16.544392434930394 +Step 23628 | loss:0.5181957483291626 lr:2.729371397121031e-05 tokens_per_second_per_gpu:5077.014823088336 mfu:23.949145148559108 +Step 23629 | loss:0.3789493441581726 lr:2.7289850753554324e-05 tokens_per_second_per_gpu:5186.418531153146 mfu:24.465221105698213 +Step 23630 | loss:0.3636764585971832 lr:2.7285987706704264e-05 tokens_per_second_per_gpu:4936.838833814967 mfu:23.287911090666665 +Step 23631 | loss:0.4485197365283966 lr:2.7282124830689175e-05 tokens_per_second_per_gpu:4854.159701754321 mfu:22.89789951822205 +Step 23632 | loss:0.40662136673927307 lr:2.7278262125538152e-05 tokens_per_second_per_gpu:5345.37929298726 mfu:25.21506614077237 +Step 23633 | loss:0.4037702977657318 lr:2.7274399591280197e-05 tokens_per_second_per_gpu:3889.5227433897653 mfu:18.347542401579975 +Step 23634 | loss:0.35545504093170166 lr:2.7270537227944383e-05 tokens_per_second_per_gpu:5352.094915943768 mfu:25.24674488006165 +Step 23635 | loss:0.3701557219028473 lr:2.7266675035559774e-05 tokens_per_second_per_gpu:5340.341610959757 mfu:25.19130253513146 +Step 23636 | loss:0.392572283744812 lr:2.726281301415541e-05 tokens_per_second_per_gpu:4899.119315575777 mfu:23.109981687518978 +Step 23637 | loss:0.46412235498428345 lr:2.7258951163760305e-05 tokens_per_second_per_gpu:4584.739000594881 mfu:21.626996102941295 +Step 23638 | loss:0.4161994159221649 lr:2.725508948440353e-05 tokens_per_second_per_gpu:4616.131905202594 mfu:21.77508179015771 +Step 23639 | loss:0.45616304874420166 lr:2.7251227976114146e-05 tokens_per_second_per_gpu:5035.7505100882445 mfu:23.75449434371989 +Step 23640 | loss:0.46288010478019714 lr:2.7247366638921167e-05 tokens_per_second_per_gpu:5325.675606759623 mfu:25.122120491041084 +Step 23641 | loss:0.3157021105289459 lr:2.7243505472853647e-05 tokens_per_second_per_gpu:3570.7263167271785 mfu:16.84372526473341 +Step 23642 | loss:0.40576407313346863 lr:2.723964447794064e-05 tokens_per_second_per_gpu:5375.796517433704 mfu:25.358549378199886 +Step 23643 | loss:0.38765424489974976 lr:2.7235783654211177e-05 tokens_per_second_per_gpu:3327.2532738723035 mfu:15.695221380802305 +Step 23644 | loss:0.4971059560775757 lr:2.723192300169427e-05 tokens_per_second_per_gpu:4174.483026852179 mfu:19.69174867791 +Step 23645 | loss:0.3776658773422241 lr:2.7228062520418974e-05 tokens_per_second_per_gpu:5366.320820642363 mfu:25.31385089971455 +Step 23646 | loss:0.3715025782585144 lr:2.722420221041434e-05 tokens_per_second_per_gpu:5350.072885294513 mfu:25.23720661649511 +Step 23647 | loss:0.3726654648780823 lr:2.722034207170937e-05 tokens_per_second_per_gpu:3497.8260278510625 mfu:16.499842164032916 +Step 23648 | loss:0.4357713460922241 lr:2.7216482104333117e-05 tokens_per_second_per_gpu:5032.9589770696675 mfu:23.741326206186617 +Step 23649 | loss:0.35886913537979126 lr:2.7212622308314607e-05 tokens_per_second_per_gpu:3112.0078848087583 mfu:14.679872155939261 +Step 23650 | loss:0.3262273967266083 lr:2.7208762683682893e-05 tokens_per_second_per_gpu:5233.775435720427 mfu:24.688611704463167 +Step 23651 | loss:0.450229287147522 lr:2.720490323046698e-05 tokens_per_second_per_gpu:4241.1046377251405 mfu:20.006014183216678 +Step 23652 | loss:0.3946341574192047 lr:2.720104394869588e-05 tokens_per_second_per_gpu:3994.7685961113552 mfu:18.844005045661834 +Step 23653 | loss:0.5220838785171509 lr:2.7197184838398643e-05 tokens_per_second_per_gpu:3899.1232486041768 mfu:18.39282962269145 +Step 23654 | loss:0.6239100098609924 lr:2.7193325899604306e-05 tokens_per_second_per_gpu:4642.049121318393 mfu:21.897337720508983 +Step 23655 | loss:0.4491311013698578 lr:2.718946713234185e-05 tokens_per_second_per_gpu:4636.7307301079945 mfu:21.872249961758424 +Step 23656 | loss:0.40806758403778076 lr:2.7185608536640327e-05 tokens_per_second_per_gpu:3883.328832913495 mfu:18.318324669073306 +Step 23657 | loss:0.3830145001411438 lr:2.718175011252877e-05 tokens_per_second_per_gpu:3727.230894828881 mfu:17.581984062073634 +Step 23658 | loss:0.42331641912460327 lr:2.7177891860036185e-05 tokens_per_second_per_gpu:3992.3486594754986 mfu:18.832589791666734 +Step 23659 | loss:0.404755562543869 lr:2.7174033779191566e-05 tokens_per_second_per_gpu:4068.1480879184073 mfu:19.19014862834801 +Step 23660 | loss:0.4299713969230652 lr:2.7170175870023957e-05 tokens_per_second_per_gpu:3720.4748511522107 mfu:17.550114651350867 +Step 23661 | loss:0.442188024520874 lr:2.7166318132562385e-05 tokens_per_second_per_gpu:5331.831730706136 mfu:25.15115997803984 +Step 23662 | loss:0.49489089846611023 lr:2.7162460566835824e-05 tokens_per_second_per_gpu:4285.114749807065 mfu:20.213617390806533 +Step 23663 | loss:0.4761314392089844 lr:2.7158603172873315e-05 tokens_per_second_per_gpu:4717.992916257639 mfu:22.255577558584942 +Step 23664 | loss:0.3763703405857086 lr:2.7154745950703887e-05 tokens_per_second_per_gpu:5342.257736525533 mfu:25.200341226349952 +Step 23665 | loss:0.3812958300113678 lr:2.7150888900356526e-05 tokens_per_second_per_gpu:3914.688299704757 mfu:18.46625262440445 +Step 23666 | loss:0.38697364926338196 lr:2.7147032021860223e-05 tokens_per_second_per_gpu:3782.5002686915145 mfu:17.84269912851124 +Step 23667 | loss:0.5142823457717896 lr:2.714317531524402e-05 tokens_per_second_per_gpu:5175.986544718519 mfu:24.416011645805654 +Step 23668 | loss:0.40150973200798035 lr:2.7139318780536916e-05 tokens_per_second_per_gpu:5341.4543167471475 mfu:25.196551358177764 +Step 23669 | loss:0.4529496431350708 lr:2.7135462417767906e-05 tokens_per_second_per_gpu:5355.863607113956 mfu:25.264522439316444 +Step 23670 | loss:0.4324873685836792 lr:2.7131606226965988e-05 tokens_per_second_per_gpu:4844.241000176775 mfu:22.85111131057577 +Step 23671 | loss:0.43004167079925537 lr:2.7127750208160208e-05 tokens_per_second_per_gpu:5355.555029907881 mfu:25.26306682798669 +Step 23672 | loss:0.519984781742096 lr:2.7123894361379532e-05 tokens_per_second_per_gpu:4014.536228214306 mfu:18.937252338997276 +Step 23673 | loss:0.45964479446411133 lr:2.712003868665295e-05 tokens_per_second_per_gpu:4477.394480868269 mfu:21.12063368852723 +Step 23674 | loss:0.32190945744514465 lr:2.7116183184009475e-05 tokens_per_second_per_gpu:3449.8460195825214 mfu:16.27351227879666 +Step 23675 | loss:0.4615160822868347 lr:2.7112327853478124e-05 tokens_per_second_per_gpu:5354.372335446982 mfu:25.25748785642615 +Step 23676 | loss:0.35611799359321594 lr:2.710847269508786e-05 tokens_per_second_per_gpu:4920.155617404987 mfu:23.209213512410557 +Step 23677 | loss:0.4735599458217621 lr:2.7104617708867692e-05 tokens_per_second_per_gpu:3833.252884263921 mfu:18.082107875455417 +Step 23678 | loss:0.40405821800231934 lr:2.710076289484663e-05 tokens_per_second_per_gpu:4633.261605204497 mfu:21.85588550769488 +Step 23679 | loss:0.43393340706825256 lr:2.7096908253053644e-05 tokens_per_second_per_gpu:3132.0067536976835 mfu:14.774210232647205 +Step 23680 | loss:0.3690209984779358 lr:2.7093053783517743e-05 tokens_per_second_per_gpu:3592.1362690216456 mfu:16.94471966262095 +Step 23681 | loss:0.4693949818611145 lr:2.7089199486267892e-05 tokens_per_second_per_gpu:5344.074117446255 mfu:25.208909405059543 +Step 23682 | loss:0.415512353181839 lr:2.7085345361333115e-05 tokens_per_second_per_gpu:5340.682486981189 mfu:25.192910505484 +Step 23683 | loss:0.35495588183403015 lr:2.7081491408742356e-05 tokens_per_second_per_gpu:4985.841748335497 mfu:23.519066199219356 +Step 23684 | loss:0.4503525197505951 lr:2.707763762852462e-05 tokens_per_second_per_gpu:3985.048752702337 mfu:18.798154885925204 +Step 23685 | loss:0.3355066478252411 lr:2.707378402070891e-05 tokens_per_second_per_gpu:4276.43545496407 mfu:20.172675676191727 +Step 23686 | loss:0.3799496591091156 lr:2.706993058532421e-05 tokens_per_second_per_gpu:4264.665404913179 mfu:20.11715434192406 +Step 23687 | loss:0.3359799087047577 lr:2.7066077322399485e-05 tokens_per_second_per_gpu:3348.117376081183 mfu:15.7936409107045 +Step 23688 | loss:0.32010141015052795 lr:2.7062224231963702e-05 tokens_per_second_per_gpu:5339.329686520695 mfu:25.18652911490029 +Step 23689 | loss:0.4401151239871979 lr:2.7058371314045857e-05 tokens_per_second_per_gpu:5342.862377565694 mfu:25.203193421298796 +Step 23690 | loss:0.4259817600250244 lr:2.705451856867495e-05 tokens_per_second_per_gpu:5369.301227689157 mfu:25.327909988264153 +Step 23691 | loss:0.3883236348628998 lr:2.705066599587991e-05 tokens_per_second_per_gpu:3280.498139660861 mfu:15.474669435478445 +Step 23692 | loss:0.33520227670669556 lr:2.7046813595689746e-05 tokens_per_second_per_gpu:4926.885711954692 mfu:23.240960516674047 +Step 23693 | loss:0.3878401815891266 lr:2.7042961368133446e-05 tokens_per_second_per_gpu:5236.10573349357 mfu:24.69960411664888 +Step 23694 | loss:0.36623015999794006 lr:2.7039109313239958e-05 tokens_per_second_per_gpu:5359.229996662967 mfu:25.280402273184414 +Step 23695 | loss:0.49265533685684204 lr:2.7035257431038242e-05 tokens_per_second_per_gpu:4452.388238472765 mfu:21.002674976642737 +Step 23696 | loss:0.4653068482875824 lr:2.7031405721557278e-05 tokens_per_second_per_gpu:4649.7127278504595 mfu:21.933488260066845 +Step 23697 | loss:0.389013409614563 lr:2.7027554184826075e-05 tokens_per_second_per_gpu:5366.0141815190655 mfu:25.312404430651924 +Step 23698 | loss:0.3490538001060486 lr:2.7023702820873543e-05 tokens_per_second_per_gpu:4700.309526282423 mfu:22.172161990974402 +Step 23699 | loss:0.4202521741390228 lr:2.7019851629728672e-05 tokens_per_second_per_gpu:4995.483169446414 mfu:23.56454642759562 +Step 23700 | loss:0.38427868485450745 lr:2.701600061142045e-05 tokens_per_second_per_gpu:4879.284418504418 mfu:23.016417093850965 +Step 23701 | loss:0.47376179695129395 lr:2.7012149765977823e-05 tokens_per_second_per_gpu:4590.22015604632 mfu:21.652851648387283 +Step 23702 | loss:0.3955976665019989 lr:2.7008299093429724e-05 tokens_per_second_per_gpu:4529.915827644324 mfu:21.368385842335677 +Step 23703 | loss:0.43004918098449707 lr:2.7004448593805143e-05 tokens_per_second_per_gpu:5270.516134940588 mfu:24.861923851294353 +Step 23704 | loss:0.48524200916290283 lr:2.7000598267133065e-05 tokens_per_second_per_gpu:3830.9169110791518 mfu:18.071088691384738 +Step 23705 | loss:0.4533618092536926 lr:2.6996748113442394e-05 tokens_per_second_per_gpu:3881.261288334714 mfu:18.30857170853572 +Step 23706 | loss:0.37699493765830994 lr:2.6992898132762123e-05 tokens_per_second_per_gpu:4387.928587305349 mfu:20.69860780413547 +Step 23707 | loss:0.5743898153305054 lr:2.698904832512122e-05 tokens_per_second_per_gpu:4150.023407314017 mfu:19.57636847930706 +Step 23708 | loss:0.4088650941848755 lr:2.698519869054862e-05 tokens_per_second_per_gpu:3540.0219849390464 mfu:16.69888769298959 +Step 23709 | loss:0.38065305352211 lr:2.698134922907325e-05 tokens_per_second_per_gpu:5157.106132115862 mfu:24.326949518228357 +Step 23710 | loss:0.37758585810661316 lr:2.69774999407241e-05 tokens_per_second_per_gpu:3803.051173976421 mfu:17.93964125508615 +Step 23711 | loss:0.45614469051361084 lr:2.6973650825530127e-05 tokens_per_second_per_gpu:4766.72919557744 mfu:22.485474903403137 +Step 23712 | loss:0.3529698848724365 lr:2.696980188352024e-05 tokens_per_second_per_gpu:4866.375017221446 mfu:22.955521245428205 +Step 23713 | loss:0.4497183859348297 lr:2.696595311472342e-05 tokens_per_second_per_gpu:4534.128949852398 mfu:21.388259858623936 +Step 23714 | loss:0.26590976119041443 lr:2.696210451916862e-05 tokens_per_second_per_gpu:3556.1258827442307 mfu:16.774852526544937 +Step 23715 | loss:0.38809195160865784 lr:2.695825609688475e-05 tokens_per_second_per_gpu:5147.571505319768 mfu:24.281973056856103 +Step 23716 | loss:0.3751301169395447 lr:2.6954407847900797e-05 tokens_per_second_per_gpu:5340.971609994644 mfu:25.194274348067935 +Step 23717 | loss:0.5238369703292847 lr:2.6950559772245664e-05 tokens_per_second_per_gpu:4162.307219587023 mfu:19.634313317633318 +Step 23718 | loss:0.3622702360153198 lr:2.6946711869948337e-05 tokens_per_second_per_gpu:4918.746743310391 mfu:23.202567612927638 +Step 23719 | loss:0.46123743057250977 lr:2.6942864141037705e-05 tokens_per_second_per_gpu:5360.832857673948 mfu:25.287963242049496 +Step 23720 | loss:0.3793020248413086 lr:2.6939016585542737e-05 tokens_per_second_per_gpu:4157.953471655387 mfu:19.6137759458138 +Step 23721 | loss:0.3510412871837616 lr:2.6935169203492365e-05 tokens_per_second_per_gpu:3972.72631610117 mfu:18.740027850053025 +Step 23722 | loss:0.4403102993965149 lr:2.6931321994915547e-05 tokens_per_second_per_gpu:3270.3598204710956 mfu:15.4268452540847 +Step 23723 | loss:0.471274733543396 lr:2.692747495984121e-05 tokens_per_second_per_gpu:3470.069900990642 mfu:16.368911777948743 +Step 23724 | loss:0.3701976537704468 lr:2.692362809829825e-05 tokens_per_second_per_gpu:4414.35765046878 mfu:20.823278204340532 +Step 23725 | loss:0.3528888523578644 lr:2.6919781410315632e-05 tokens_per_second_per_gpu:4749.632559307104 mfu:22.404827153129027 +Step 23726 | loss:0.4000730812549591 lr:2.69159348959223e-05 tokens_per_second_per_gpu:4644.643536608911 mfu:21.909576020087247 +Step 23727 | loss:0.538394033908844 lr:2.6912088555147142e-05 tokens_per_second_per_gpu:3991.2214421971 mfu:18.82727251544143 +Step 23728 | loss:0.5109087824821472 lr:2.690824238801912e-05 tokens_per_second_per_gpu:4044.552238677215 mfu:19.07884298161029 +Step 23729 | loss:0.4487531781196594 lr:2.6904396394567173e-05 tokens_per_second_per_gpu:5344.199347716503 mfu:25.209500137610792 +Step 23730 | loss:0.5484023094177246 lr:2.6900550574820205e-05 tokens_per_second_per_gpu:4401.552288518894 mfu:20.76287312720288 +Step 23731 | loss:0.43933916091918945 lr:2.6896704928807125e-05 tokens_per_second_per_gpu:5322.873252194791 mfu:25.10890130642778 +Step 23732 | loss:0.45255881547927856 lr:2.6892859456556873e-05 tokens_per_second_per_gpu:5233.308890707055 mfu:24.686410932034306 +Step 23733 | loss:0.4052789509296417 lr:2.6889014158098392e-05 tokens_per_second_per_gpu:4501.710930674795 mfu:21.235338531078764 +Step 23734 | loss:0.4778117537498474 lr:2.6885169033460573e-05 tokens_per_second_per_gpu:5388.949683134228 mfu:25.420595104970733 +Step 23735 | loss:0.4373980760574341 lr:2.6881324082672342e-05 tokens_per_second_per_gpu:5336.358163556908 mfu:25.17251193408601 +Step 23736 | loss:0.4274464249610901 lr:2.6877479305762642e-05 tokens_per_second_per_gpu:3410.4890997182074 mfu:16.087858972813734 +Step 23737 | loss:0.4660240411758423 lr:2.687363470276037e-05 tokens_per_second_per_gpu:4631.187979332607 mfu:21.846103860660122 +Step 23738 | loss:0.4473402798175812 lr:2.6869790273694422e-05 tokens_per_second_per_gpu:4367.6577061184535 mfu:20.60298659900792 +Step 23739 | loss:0.36477816104888916 lr:2.6865946018593734e-05 tokens_per_second_per_gpu:4134.952650999577 mfu:19.505277150435674 +Step 23740 | loss:0.43550342321395874 lr:2.6862101937487237e-05 tokens_per_second_per_gpu:4948.636126577222 mfu:23.343560933451922 +Step 23741 | loss:0.34209945797920227 lr:2.6858258030403806e-05 tokens_per_second_per_gpu:4460.988692816334 mfu:21.043244787170178 +Step 23742 | loss:0.45286989212036133 lr:2.6854414297372366e-05 tokens_per_second_per_gpu:5345.495985576033 mfu:25.21561659962342 +Step 23743 | loss:0.49433714151382446 lr:2.6850570738421854e-05 tokens_per_second_per_gpu:4652.234407628013 mfu:21.9453834538205 +Step 23744 | loss:0.3812147378921509 lr:2.684672735358113e-05 tokens_per_second_per_gpu:5379.733196108449 mfu:25.3771193631753 +Step 23745 | loss:0.3763832747936249 lr:2.6842884142879144e-05 tokens_per_second_per_gpu:5344.8958522482435 mfu:25.212785668322937 +Step 23746 | loss:0.5465419888496399 lr:2.6839041106344763e-05 tokens_per_second_per_gpu:4159.861561044421 mfu:19.622776729015495 +Step 23747 | loss:0.5293612480163574 lr:2.6835198244006927e-05 tokens_per_second_per_gpu:5212.161582134754 mfu:24.58665546939586 +Step 23748 | loss:0.3330945074558258 lr:2.6831355555894498e-05 tokens_per_second_per_gpu:3596.3760142137244 mfu:16.964719264067107 +Step 23749 | loss:0.4380079209804535 lr:2.682751304203641e-05 tokens_per_second_per_gpu:4968.512919092736 mfu:23.43732315507811 +Step 23750 | loss:0.3128030598163605 lr:2.6823670702461567e-05 tokens_per_second_per_gpu:4508.672031603877 mfu:21.268175231847533 +Step 23751 | loss:0.40710070729255676 lr:2.6819828537198832e-05 tokens_per_second_per_gpu:5360.695317300456 mfu:25.28731444064093 +Step 23752 | loss:0.415945440530777 lr:2.681598654627715e-05 tokens_per_second_per_gpu:4606.86731557373 mfu:21.731379140176372 +Step 23753 | loss:0.3524097502231598 lr:2.6812144729725363e-05 tokens_per_second_per_gpu:5242.670644319801 mfu:24.7305719211055 +Step 23754 | loss:0.4553253948688507 lr:2.6808303087572423e-05 tokens_per_second_per_gpu:5339.237119316886 mfu:25.186092459606325 +Step 23755 | loss:0.36897674202919006 lr:2.680446161984717e-05 tokens_per_second_per_gpu:3898.4084153244407 mfu:18.389457632147803 +Step 23756 | loss:0.4514991044998169 lr:2.6800620326578523e-05 tokens_per_second_per_gpu:4536.314613479565 mfu:21.39857000686595 +Step 23757 | loss:0.5306324362754822 lr:2.6796779207795376e-05 tokens_per_second_per_gpu:4127.305993511649 mfu:19.469206562410818 +Step 23758 | loss:0.47836941480636597 lr:2.6792938263526623e-05 tokens_per_second_per_gpu:5028.559599403579 mfu:23.72057359112445 +Step 23759 | loss:0.34585630893707275 lr:2.6789097493801146e-05 tokens_per_second_per_gpu:3509.2929710536105 mfu:16.55393369158159 +Step 23760 | loss:0.49848148226737976 lr:2.678525689864781e-05 tokens_per_second_per_gpu:5048.0963440250025 mfu:23.812731748816056 +Step 23761 | loss:0.49003565311431885 lr:2.6781416478095517e-05 tokens_per_second_per_gpu:5350.306526155474 mfu:25.2383087402995 +Step 23762 | loss:0.2881254255771637 lr:2.6777576232173174e-05 tokens_per_second_per_gpu:5316.320446077501 mfu:25.077990601948514 +Step 23763 | loss:0.2803872525691986 lr:2.6773736160909625e-05 tokens_per_second_per_gpu:5331.714231332103 mfu:25.150605713445064 +Step 23764 | loss:0.307000994682312 lr:2.676989626433377e-05 tokens_per_second_per_gpu:4166.609365718909 mfu:19.654607274959247 +Step 23765 | loss:0.3373417854309082 lr:2.6766056542474503e-05 tokens_per_second_per_gpu:5291.831995161632 mfu:24.962474400817793 +Step 23766 | loss:0.3637501001358032 lr:2.67622169953607e-05 tokens_per_second_per_gpu:5366.779193904238 mfu:25.316013124597355 +Step 23767 | loss:0.5024823546409607 lr:2.67583776230212e-05 tokens_per_second_per_gpu:4676.273343529898 mfu:22.058779215935086 +Step 23768 | loss:0.299917608499527 lr:2.6754538425484914e-05 tokens_per_second_per_gpu:4466.866454406042 mfu:21.070971191438367 +Step 23769 | loss:0.45009684562683105 lr:2.6750699402780726e-05 tokens_per_second_per_gpu:4619.131530631024 mfu:21.78923153509249 +Step 23770 | loss:0.48627960681915283 lr:2.674686055493748e-05 tokens_per_second_per_gpu:2842.9948844966225 mfu:13.410891935116176 +Step 23771 | loss:0.3993290662765503 lr:2.6743021881984053e-05 tokens_per_second_per_gpu:4367.992721390176 mfu:20.604566923204132 +Step 23772 | loss:0.4543876349925995 lr:2.6739183383949352e-05 tokens_per_second_per_gpu:3866.626819373494 mfu:18.239538421547767 +Step 23773 | loss:0.4277346134185791 lr:2.6735345060862205e-05 tokens_per_second_per_gpu:4653.57909161417 mfu:21.951726557609945 +Step 23774 | loss:0.5056893825531006 lr:2.6731506912751512e-05 tokens_per_second_per_gpu:4903.90081376521 mfu:23.132536830286245 +Step 23775 | loss:0.42215004563331604 lr:2.67276689396461e-05 tokens_per_second_per_gpu:4529.133985175643 mfu:21.364697758014827 +Step 23776 | loss:0.3927382230758667 lr:2.6723831141574884e-05 tokens_per_second_per_gpu:5127.967645341268 mfu:24.189498304574098 +Step 23777 | loss:0.4039749205112457 lr:2.6719993518566688e-05 tokens_per_second_per_gpu:5027.455162701017 mfu:23.71536377476213 +Step 23778 | loss:0.42995402216911316 lr:2.6716156070650388e-05 tokens_per_second_per_gpu:4062.1857726814214 mfu:19.162023370099266 +Step 23779 | loss:0.5432997941970825 lr:2.671231879785487e-05 tokens_per_second_per_gpu:5358.820841209716 mfu:25.27847221710231 +Step 23780 | loss:0.42275309562683105 lr:2.6708481700208952e-05 tokens_per_second_per_gpu:5083.079676457617 mfu:23.97775409667269 +Step 23781 | loss:0.37228021025657654 lr:2.6704644777741534e-05 tokens_per_second_per_gpu:5328.969883681387 mfu:25.13766015734264 +Step 23782 | loss:0.3172493278980255 lr:2.6700808030481438e-05 tokens_per_second_per_gpu:5325.727084458432 mfu:25.122363320129363 +Step 23783 | loss:0.4334275722503662 lr:2.6696971458457554e-05 tokens_per_second_per_gpu:5338.215526598913 mfu:25.181273432453963 +Step 23784 | loss:0.4579131007194519 lr:2.6693135061698703e-05 tokens_per_second_per_gpu:3280.1607588134634 mfu:15.473077952458164 +Step 23785 | loss:0.4484982192516327 lr:2.6689298840233756e-05 tokens_per_second_per_gpu:3968.5086950099308 mfu:18.720132611765287 +Step 23786 | loss:0.4793923497200012 lr:2.6685462794091592e-05 tokens_per_second_per_gpu:3921.380722773669 mfu:18.497821925865626 +Step 23787 | loss:0.44080278277397156 lr:2.6681626923301017e-05 tokens_per_second_per_gpu:4699.995672462289 mfu:22.170681488955847 +Step 23788 | loss:0.35340315103530884 lr:2.6677791227890914e-05 tokens_per_second_per_gpu:4734.298920358216 mfu:22.33249576201854 +Step 23789 | loss:0.41341927647590637 lr:2.667395570789011e-05 tokens_per_second_per_gpu:5248.142505792415 mfu:24.756383625267656 +Step 23790 | loss:0.4373548626899719 lr:2.6670120363327466e-05 tokens_per_second_per_gpu:5043.5139871508245 mfu:23.791115989609857 +Step 23791 | loss:0.43187227845191956 lr:2.6666285194231815e-05 tokens_per_second_per_gpu:3995.1691090133154 mfu:18.845894333354927 +Step 23792 | loss:0.3096020817756653 lr:2.6662450200632006e-05 tokens_per_second_per_gpu:4266.37107229191 mfu:20.1252002659664 +Step 23793 | loss:0.4506772458553314 lr:2.6658615382556895e-05 tokens_per_second_per_gpu:4540.950886012261 mfu:21.42044009543234 +Step 23794 | loss:0.3432619571685791 lr:2.6654780740035328e-05 tokens_per_second_per_gpu:4532.942139511314 mfu:21.382661471754318 +Step 23795 | loss:0.3821980059146881 lr:2.6650946273096133e-05 tokens_per_second_per_gpu:3666.1547673944906 mfu:17.29387754830373 +Step 23796 | loss:0.535794734954834 lr:2.6647111981768135e-05 tokens_per_second_per_gpu:5332.454740203123 mfu:25.154098820133505 +Step 23797 | loss:0.5402044057846069 lr:2.6643277866080208e-05 tokens_per_second_per_gpu:4277.0123940387475 mfu:20.175397196242937 +Step 23798 | loss:0.2654164135456085 lr:2.6639443926061147e-05 tokens_per_second_per_gpu:4872.177347011135 mfu:22.982891825025714 +Step 23799 | loss:0.39437830448150635 lr:2.663561016173981e-05 tokens_per_second_per_gpu:4920.165018565739 mfu:23.209257859289824 +Step 23800 | loss:0.371627539396286 lr:2.663177657314503e-05 tokens_per_second_per_gpu:4921.104838756716 mfu:23.21369115148052 +Step 23801 | loss:0.38018685579299927 lr:2.6627943160305658e-05 tokens_per_second_per_gpu:5320.107772901904 mfu:25.09585606876381 +Step 23802 | loss:0.3686179220676422 lr:2.6624109923250486e-05 tokens_per_second_per_gpu:5346.99092965923 mfu:25.222668505927622 +Step 23803 | loss:0.5244125723838806 lr:2.6620276862008386e-05 tokens_per_second_per_gpu:4064.421381229644 mfu:19.17256911705538 +Step 23804 | loss:0.4427906274795532 lr:2.661644397660815e-05 tokens_per_second_per_gpu:5082.769561899627 mfu:23.976291233391652 +Step 23805 | loss:0.4882088601589203 lr:2.6612611267078646e-05 tokens_per_second_per_gpu:5082.215604276581 mfu:23.973678120768827 +Step 23806 | loss:0.42446377873420715 lr:2.6608778733448653e-05 tokens_per_second_per_gpu:5345.222951258014 mfu:25.214328650160873 +Step 23807 | loss:0.4222923219203949 lr:2.660494637574702e-05 tokens_per_second_per_gpu:5350.972644460018 mfu:25.24145093399354 +Step 23808 | loss:0.4163733720779419 lr:2.660111419400259e-05 tokens_per_second_per_gpu:4286.317558382376 mfu:20.21929124407658 +Step 23809 | loss:0.43568935990333557 lr:2.659728218824415e-05 tokens_per_second_per_gpu:4129.537901543431 mfu:19.47973485339951 +Step 23810 | loss:0.4584660530090332 lr:2.6593450358500547e-05 tokens_per_second_per_gpu:5044.420067144814 mfu:23.795390123534695 +Step 23811 | loss:0.5425443649291992 lr:2.6589618704800584e-05 tokens_per_second_per_gpu:4773.13360450688 mfu:22.515685593028145 +Step 23812 | loss:0.4516262710094452 lr:2.6585787227173098e-05 tokens_per_second_per_gpu:5303.7774945363835 mfu:25.018823359480848 +Step 23813 | loss:0.4228392541408539 lr:2.6581955925646872e-05 tokens_per_second_per_gpu:4452.143199261251 mfu:21.001519084874023 +Step 23814 | loss:0.40918025374412537 lr:2.6578124800250746e-05 tokens_per_second_per_gpu:5210.409451552299 mfu:24.57839037049412 +Step 23815 | loss:0.43610039353370667 lr:2.6574293851013553e-05 tokens_per_second_per_gpu:3586.1709336108556 mfu:16.916580157697084 +Step 23816 | loss:0.40691104531288147 lr:2.6570463077964068e-05 tokens_per_second_per_gpu:5370.58407796382 mfu:25.333961411886563 +Step 23817 | loss:0.3852880895137787 lr:2.6566632481131138e-05 tokens_per_second_per_gpu:4692.687096259241 mfu:22.136205688034387 +Step 23818 | loss:0.4869282841682434 lr:2.6562802060543535e-05 tokens_per_second_per_gpu:4149.890308962449 mfu:19.575740631673796 +Step 23819 | loss:0.6048335433006287 lr:2.6558971816230105e-05 tokens_per_second_per_gpu:4159.835755190103 mfu:19.62265499839774 +Step 23820 | loss:0.3593308627605438 lr:2.6555141748219627e-05 tokens_per_second_per_gpu:3804.4734475457403 mfu:17.946350362177522 +Step 23821 | loss:0.438468337059021 lr:2.6551311856540907e-05 tokens_per_second_per_gpu:4933.1914612185565 mfu:23.27070581182346 +Step 23822 | loss:0.35749974846839905 lr:2.654748214122279e-05 tokens_per_second_per_gpu:4842.714072156839 mfu:22.84390853058495 +Step 23823 | loss:0.442699670791626 lr:2.6543652602294033e-05 tokens_per_second_per_gpu:4806.002398271701 mfu:22.6707333012113 +Step 23824 | loss:0.6037154793739319 lr:2.653982323978348e-05 tokens_per_second_per_gpu:4617.60957685178 mfu:21.782052219443706 +Step 23825 | loss:0.4830223321914673 lr:2.653599405371988e-05 tokens_per_second_per_gpu:5044.5339815216175 mfu:23.795927477085872 +Step 23826 | loss:0.47188669443130493 lr:2.6532165044132094e-05 tokens_per_second_per_gpu:3635.640543046259 mfu:17.149936745789596 +Step 23827 | loss:0.4025241732597351 lr:2.6528336211048867e-05 tokens_per_second_per_gpu:4129.160986683166 mfu:19.477956881695334 +Step 23828 | loss:0.5182642340660095 lr:2.652450755449901e-05 tokens_per_second_per_gpu:4192.599325608978 mfu:19.77720635010511 +Step 23829 | loss:0.3858124017715454 lr:2.6520679074511334e-05 tokens_per_second_per_gpu:5400.187592655044 mfu:25.473606241565506 +Step 23830 | loss:0.3715061843395233 lr:2.6516850771114645e-05 tokens_per_second_per_gpu:5344.4511166488965 mfu:25.2106877746211 +Step 23831 | loss:0.4282521903514862 lr:2.651302264433772e-05 tokens_per_second_per_gpu:4569.1952923987865 mfu:21.553673779350227 +Step 23832 | loss:0.39926818013191223 lr:2.650919469420933e-05 tokens_per_second_per_gpu:4417.592933535834 mfu:20.83853958656404 +Step 23833 | loss:0.41733184456825256 lr:2.65053669207583e-05 tokens_per_second_per_gpu:3571.3487972068538 mfu:16.84666161136216 +Step 23834 | loss:0.38269779086112976 lr:2.650153932401338e-05 tokens_per_second_per_gpu:5279.647339617231 mfu:24.90499730169825 +Step 23835 | loss:0.31940415501594543 lr:2.6497711904003386e-05 tokens_per_second_per_gpu:4022.001187331228 mfu:18.972465824805152 +Step 23836 | loss:0.43389031291007996 lr:2.64938846607571e-05 tokens_per_second_per_gpu:5320.756579114665 mfu:25.098916598367158 +Step 23837 | loss:0.4392695128917694 lr:2.6490057594303332e-05 tokens_per_second_per_gpu:5320.809834390571 mfu:25.099167812589894 +Step 23838 | loss:0.4800046980381012 lr:2.648623070467081e-05 tokens_per_second_per_gpu:5345.507717649511 mfu:25.215671941815717 +Step 23839 | loss:0.4171464145183563 lr:2.6482403991888372e-05 tokens_per_second_per_gpu:5336.498469284738 mfu:25.17317378014281 +Step 23840 | loss:0.3798779547214508 lr:2.6478577455984756e-05 tokens_per_second_per_gpu:5330.601636674473 mfu:25.1453574146168 +Step 23841 | loss:0.424138605594635 lr:2.647475109698878e-05 tokens_per_second_per_gpu:5358.46245190149 mfu:25.276781633587884 +Step 23842 | loss:0.4799521863460541 lr:2.6470924914929184e-05 tokens_per_second_per_gpu:3825.8825454301937 mfu:18.047340729667248 +Step 23843 | loss:0.6091315150260925 lr:2.646709890983476e-05 tokens_per_second_per_gpu:3866.2807031399066 mfu:18.237905732220415 +Step 23844 | loss:0.3778330981731415 lr:2.646327308173431e-05 tokens_per_second_per_gpu:4768.445516185997 mfu:22.493571080548193 +Step 23845 | loss:0.5325872898101807 lr:2.645944743065657e-05 tokens_per_second_per_gpu:5313.061511712661 mfu:25.062617652518018 +Step 23846 | loss:0.4211377501487732 lr:2.645562195663034e-05 tokens_per_second_per_gpu:4747.821361879373 mfu:22.39628342584027 +Step 23847 | loss:0.4426235258579254 lr:2.645179665968437e-05 tokens_per_second_per_gpu:4539.099362605331 mfu:21.411706143619384 +Step 23848 | loss:0.44838204979896545 lr:2.644797153984746e-05 tokens_per_second_per_gpu:5045.695053415969 mfu:23.80140445131082 +Step 23849 | loss:0.4902518093585968 lr:2.644414659714834e-05 tokens_per_second_per_gpu:5139.166836384972 mfu:24.24232679950628 +Step 23850 | loss:0.38598769903182983 lr:2.6440321831615798e-05 tokens_per_second_per_gpu:5363.097916610617 mfu:25.29864791151267 +Step 23851 | loss:0.43825700879096985 lr:2.6436497243278625e-05 tokens_per_second_per_gpu:3917.6411320577904 mfu:18.48018163841886 +Step 23852 | loss:0.3875284492969513 lr:2.643267283216554e-05 tokens_per_second_per_gpu:5356.838703505006 mfu:25.26912213536163 +Step 23853 | loss:0.42113661766052246 lr:2.6428848598305356e-05 tokens_per_second_per_gpu:5356.213013578868 mfu:25.266170649226325 +Step 23854 | loss:0.45364636182785034 lr:2.6425024541726783e-05 tokens_per_second_per_gpu:4746.770289665478 mfu:22.39132533887592 +Step 23855 | loss:0.2600820064544678 lr:2.6421200662458634e-05 tokens_per_second_per_gpu:5267.35471979041 mfu:24.847010916637917 +Step 23856 | loss:0.4327932298183441 lr:2.6417376960529616e-05 tokens_per_second_per_gpu:4783.143546835481 mfu:22.56290419886443 +Step 23857 | loss:0.5430595278739929 lr:2.6413553435968517e-05 tokens_per_second_per_gpu:5056.788608133926 mfu:23.85373464167103 +Step 23858 | loss:0.440320760011673 lr:2.6409730088804118e-05 tokens_per_second_per_gpu:4528.509401434675 mfu:21.36175148994371 +Step 23859 | loss:0.3450465500354767 lr:2.6405906919065128e-05 tokens_per_second_per_gpu:2911.8517368847833 mfu:13.735701456021584 +Step 23860 | loss:0.5227882862091064 lr:2.640208392678034e-05 tokens_per_second_per_gpu:4837.066114409978 mfu:22.817266150252046 +Step 23861 | loss:0.5094716548919678 lr:2.6398261111978468e-05 tokens_per_second_per_gpu:4281.036064697261 mfu:20.19437753724815 +Step 23862 | loss:0.6264607906341553 lr:2.639443847468831e-05 tokens_per_second_per_gpu:5067.647824867236 mfu:23.904959419775484 +Step 23863 | loss:0.37477821111679077 lr:2.639061601493857e-05 tokens_per_second_per_gpu:5375.791401981526 mfu:25.35852524773918 +Step 23864 | loss:0.3533572554588318 lr:2.638679373275802e-05 tokens_per_second_per_gpu:5313.988129390146 mfu:25.066988666199997 +Step 23865 | loss:0.4142109751701355 lr:2.6382971628175424e-05 tokens_per_second_per_gpu:4139.125830383066 mfu:19.524962749605415 +Step 23866 | loss:0.35697442293167114 lr:2.6379149701219495e-05 tokens_per_second_per_gpu:3597.034897685734 mfu:16.967827329821688 +Step 23867 | loss:0.40397244691848755 lr:2.6375327951918993e-05 tokens_per_second_per_gpu:5180.149978037934 mfu:24.435651271128304 +Step 23868 | loss:0.4313533902168274 lr:2.6371506380302692e-05 tokens_per_second_per_gpu:4762.362410780949 mfu:22.46487603446778 +Step 23869 | loss:0.3645508885383606 lr:2.6367684986399298e-05 tokens_per_second_per_gpu:5359.78089584392 mfu:25.283000958614064 +Step 23870 | loss:0.35764428973197937 lr:2.636386377023754e-05 tokens_per_second_per_gpu:4393.702682538893 mfu:20.725845196514335 +Step 23871 | loss:0.35330286622047424 lr:2.6360042731846184e-05 tokens_per_second_per_gpu:5347.332628461003 mfu:25.22428035747542 +Step 23872 | loss:0.5162334442138672 lr:2.635622187125396e-05 tokens_per_second_per_gpu:3466.156701305496 mfu:16.350452547373195 +Step 23873 | loss:0.28038373589515686 lr:2.6352401188489633e-05 tokens_per_second_per_gpu:4727.910529878643 mfu:22.302360634154585 +Step 23874 | loss:0.4719897210597992 lr:2.6348580683581893e-05 tokens_per_second_per_gpu:5121.915603575757 mfu:24.160949790981384 +Step 23875 | loss:0.39028188586235046 lr:2.634476035655951e-05 tokens_per_second_per_gpu:4849.438299079794 mfu:22.875627856251942 +Step 23876 | loss:0.4080554246902466 lr:2.6340940207451186e-05 tokens_per_second_per_gpu:5161.033595647204 mfu:24.345476033800068 +Step 23877 | loss:0.4543970227241516 lr:2.633712023628569e-05 tokens_per_second_per_gpu:4680.27324189983 mfu:22.07764741899909 +Step 23878 | loss:0.3486403226852417 lr:2.6333300443091714e-05 tokens_per_second_per_gpu:4921.477065064828 mfu:23.215447006484244 +Step 23879 | loss:0.3985091745853424 lr:2.6329480827898e-05 tokens_per_second_per_gpu:5333.426787900817 mfu:25.158684134971967 +Step 23880 | loss:0.3704342246055603 lr:2.63256613907333e-05 tokens_per_second_per_gpu:5346.646989200647 mfu:25.221046080102195 +Step 23881 | loss:0.2646563649177551 lr:2.6321842131626312e-05 tokens_per_second_per_gpu:4413.465654807531 mfu:20.819070508616267 +Step 23882 | loss:0.348525732755661 lr:2.6318023050605784e-05 tokens_per_second_per_gpu:5361.061405897222 mfu:25.289041343013032 +Step 23883 | loss:0.4212487041950226 lr:2.6314204147700405e-05 tokens_per_second_per_gpu:3432.4161931469916 mfu:16.191292813665225 +Step 23884 | loss:0.5030226111412048 lr:2.6310385422938943e-05 tokens_per_second_per_gpu:3945.0591604503484 mfu:18.609517156344577 +Step 23885 | loss:0.4599364995956421 lr:2.630656687635007e-05 tokens_per_second_per_gpu:4399.954648193935 mfu:20.755336785204502 +Step 23886 | loss:0.4199647009372711 lr:2.6302748507962537e-05 tokens_per_second_per_gpu:5327.52892125815 mfu:25.130862891738694 +Step 23887 | loss:0.4405633807182312 lr:2.629893031780507e-05 tokens_per_second_per_gpu:3397.92460240846 mfu:16.028590095279363 +Step 23888 | loss:0.49206387996673584 lr:2.6295112305906357e-05 tokens_per_second_per_gpu:5312.492965665265 mfu:25.059935723808493 +Step 23889 | loss:0.5267477631568909 lr:2.6291294472295147e-05 tokens_per_second_per_gpu:3769.756665561836 mfu:17.782585378265647 +Step 23890 | loss:0.38989290595054626 lr:2.6287476817000116e-05 tokens_per_second_per_gpu:3851.436108654026 mfu:18.16788124728153 +Step 23891 | loss:0.36935660243034363 lr:2.6283659340050015e-05 tokens_per_second_per_gpu:3446.9730002101232 mfu:16.259959756229406 +Step 23892 | loss:0.484596312046051 lr:2.6279842041473525e-05 tokens_per_second_per_gpu:4153.983908370452 mfu:19.595050838521423 +Step 23893 | loss:0.3452601432800293 lr:2.627602492129937e-05 tokens_per_second_per_gpu:4757.503397828634 mfu:22.441955241338608 +Step 23894 | loss:0.41442835330963135 lr:2.6272207979556275e-05 tokens_per_second_per_gpu:4978.64671097368 mfu:23.485125980383703 +Step 23895 | loss:0.3590793013572693 lr:2.626839121627291e-05 tokens_per_second_per_gpu:5339.295307467934 mfu:25.186366943042746 +Step 23896 | loss:0.4498903751373291 lr:2.6264574631478002e-05 tokens_per_second_per_gpu:5338.221042163367 mfu:25.18129945031254 +Step 23897 | loss:0.44023144245147705 lr:2.6260758225200276e-05 tokens_per_second_per_gpu:5362.305455741163 mfu:25.294909738383694 +Step 23898 | loss:0.41032639145851135 lr:2.625694199746842e-05 tokens_per_second_per_gpu:4059.5369553203464 mfu:19.14952844667223 +Step 23899 | loss:0.4632546305656433 lr:2.6253125948311118e-05 tokens_per_second_per_gpu:4358.188608033146 mfu:20.558319247744798 +Step 23900 | loss:0.3906281292438507 lr:2.6249310077757083e-05 tokens_per_second_per_gpu:5016.239816660319 mfu:23.662459073952803 +Step 23901 | loss:0.43598371744155884 lr:2.6245494385835036e-05 tokens_per_second_per_gpu:5325.095007426399 mfu:25.119381704925804 +Step 23902 | loss:0.5102490782737732 lr:2.6241678872573633e-05 tokens_per_second_per_gpu:3626.6522347771074 mfu:17.10753736212048 +Step 23903 | loss:0.48458144068717957 lr:2.6237863538001596e-05 tokens_per_second_per_gpu:4663.7946172617585 mfu:21.999914935038028 +Step 23904 | loss:0.37550732493400574 lr:2.623404838214764e-05 tokens_per_second_per_gpu:5336.920567315974 mfu:25.1751648885735 +Step 23905 | loss:0.37451493740081787 lr:2.6230233405040437e-05 tokens_per_second_per_gpu:5196.682136441266 mfu:24.513636282993904 +Step 23906 | loss:0.42654719948768616 lr:2.622641860670867e-05 tokens_per_second_per_gpu:4125.574365412915 mfu:19.461038177222868 +Step 23907 | loss:0.3673108220100403 lr:2.6222603987181037e-05 tokens_per_second_per_gpu:5168.081658403167 mfu:24.37872294834346 +Step 23908 | loss:0.4255880117416382 lr:2.6218789546486234e-05 tokens_per_second_per_gpu:4838.332044369968 mfu:22.82323775786324 +Step 23909 | loss:0.44829922914505005 lr:2.6214975284652964e-05 tokens_per_second_per_gpu:5349.686968276917 mfu:25.235386180078876 +Step 23910 | loss:0.36931493878364563 lr:2.621116120170989e-05 tokens_per_second_per_gpu:4656.72086780954 mfu:21.966546851965806 +Step 23911 | loss:0.44313305616378784 lr:2.620734729768572e-05 tokens_per_second_per_gpu:5330.815578393974 mfu:25.146366614209906 +Step 23912 | loss:0.4838887155056 lr:2.6203533572609107e-05 tokens_per_second_per_gpu:4577.6939581786155 mfu:21.593763435855582 +Step 23913 | loss:0.39586344361305237 lr:2.6199720026508778e-05 tokens_per_second_per_gpu:4081.6508804857676 mfu:19.25384360470222 +Step 23914 | loss:0.45351797342300415 lr:2.619590665941337e-05 tokens_per_second_per_gpu:5025.368602905525 mfu:23.705521116202984 +Step 23915 | loss:0.3738172948360443 lr:2.6192093471351586e-05 tokens_per_second_per_gpu:4184.6032532661 mfu:19.73948751259314 +Step 23916 | loss:0.3758879005908966 lr:2.618828046235212e-05 tokens_per_second_per_gpu:5350.666438612027 mfu:25.240006508764466 +Step 23917 | loss:0.4490900933742523 lr:2.618446763244362e-05 tokens_per_second_per_gpu:5327.557292379953 mfu:25.130996723160848 +Step 23918 | loss:0.3741697669029236 lr:2.61806549816548e-05 tokens_per_second_per_gpu:4221.47405854516 mfu:19.91341339190253 +Step 23919 | loss:0.45266082882881165 lr:2.6176842510014294e-05 tokens_per_second_per_gpu:4639.886863987888 mfu:21.88713798376185 +Step 23920 | loss:0.4527244567871094 lr:2.6173030217550808e-05 tokens_per_second_per_gpu:5297.263909004155 mfu:24.988097665193038 +Step 23921 | loss:0.4207862615585327 lr:2.616921810429299e-05 tokens_per_second_per_gpu:4272.484926763491 mfu:20.15404036063998 +Step 23922 | loss:0.5041458010673523 lr:2.6165406170269518e-05 tokens_per_second_per_gpu:5322.46735259848 mfu:25.106986608778367 +Step 23923 | loss:0.4015454947948456 lr:2.616159441550909e-05 tokens_per_second_per_gpu:3809.7515816747195 mfu:17.97124822140646 +Step 23924 | loss:0.44212400913238525 lr:2.6157782840040334e-05 tokens_per_second_per_gpu:4965.2961072014405 mfu:23.422148904542162 +Step 23925 | loss:0.36095672845840454 lr:2.615397144389195e-05 tokens_per_second_per_gpu:4266.260720319454 mfu:20.12467971688433 +Step 23926 | loss:0.26483389735221863 lr:2.6150160227092575e-05 tokens_per_second_per_gpu:5325.930270584226 mfu:25.123321783751862 +Step 23927 | loss:0.5033054351806641 lr:2.6146349189670905e-05 tokens_per_second_per_gpu:5150.77209223988 mfu:24.29707077143471 +Step 23928 | loss:0.5369732975959778 lr:2.6142538331655564e-05 tokens_per_second_per_gpu:4625.539921689732 mfu:21.819461009100674 +Step 23929 | loss:0.4718141555786133 lr:2.613872765307524e-05 tokens_per_second_per_gpu:3896.3924048944273 mfu:18.379947766982582 +Step 23930 | loss:0.3630197048187256 lr:2.613491715395861e-05 tokens_per_second_per_gpu:3571.1902602842147 mfu:16.845913765648884 +Step 23931 | loss:0.4144322872161865 lr:2.6131106834334295e-05 tokens_per_second_per_gpu:3335.6983520728504 mfu:15.735058255553204 +Step 23932 | loss:0.5315162539482117 lr:2.6127296694230967e-05 tokens_per_second_per_gpu:3989.013906599166 mfu:18.816859193381564 +Step 23933 | loss:0.4148350954055786 lr:2.612348673367731e-05 tokens_per_second_per_gpu:4700.550163696899 mfu:22.173297118715467 +Step 23934 | loss:0.37826210260391235 lr:2.6119676952701955e-05 tokens_per_second_per_gpu:4947.59429953461 mfu:23.3386464575379 +Step 23935 | loss:0.44157281517982483 lr:2.6115867351333534e-05 tokens_per_second_per_gpu:4098.48394114386 mfu:19.333248023842742 +Step 23936 | loss:0.5305095911026001 lr:2.6112057929600732e-05 tokens_per_second_per_gpu:5225.452810799983 mfu:24.6493524627272 +Step 23937 | loss:0.37486398220062256 lr:2.61082486875322e-05 tokens_per_second_per_gpu:4471.301128153261 mfu:21.091890304137358 +Step 23938 | loss:0.3041192591190338 lr:2.6104439625156563e-05 tokens_per_second_per_gpu:4131.3400200894075 mfu:19.488235754054013 +Step 23939 | loss:0.36467012763023376 lr:2.610063074250248e-05 tokens_per_second_per_gpu:3954.796745633083 mfu:18.65545100705489 +Step 23940 | loss:0.3583640158176422 lr:2.609682203959862e-05 tokens_per_second_per_gpu:5102.9227167743975 mfu:24.071357103418734 +Step 23941 | loss:0.4160243272781372 lr:2.6093013516473618e-05 tokens_per_second_per_gpu:3781.4978179883087 mfu:17.837970397508492 +Step 23942 | loss:0.3642146587371826 lr:2.6089205173156094e-05 tokens_per_second_per_gpu:4159.848868585798 mfu:19.62271685652261 +Step 23943 | loss:0.41659706830978394 lr:2.6085397009674706e-05 tokens_per_second_per_gpu:5371.360362037145 mfu:25.33762327630816 +Step 23944 | loss:0.36825570464134216 lr:2.6081589026058102e-05 tokens_per_second_per_gpu:5322.1410356137985 mfu:25.105447315886206 +Step 23945 | loss:0.47493407130241394 lr:2.6077781222334935e-05 tokens_per_second_per_gpu:5032.837182753453 mfu:23.740751681616782 +Step 23946 | loss:0.4314294755458832 lr:2.6073973598533807e-05 tokens_per_second_per_gpu:4976.846630300047 mfu:23.476634692720744 +Step 23947 | loss:0.5256165266036987 lr:2.6070166154683402e-05 tokens_per_second_per_gpu:4670.553780379483 mfu:22.03179906069648 +Step 23948 | loss:0.38398948311805725 lr:2.6066358890812303e-05 tokens_per_second_per_gpu:2989.960899041575 mfu:14.104155700712449 +Step 23949 | loss:0.35780730843544006 lr:2.60625518069492e-05 tokens_per_second_per_gpu:5373.106507883348 mfu:25.345860144187903 +Step 23950 | loss:0.42955559492111206 lr:2.605874490312268e-05 tokens_per_second_per_gpu:5251.9593897581435 mfu:24.774388518161555 +Step 23951 | loss:0.3860801160335541 lr:2.605493817936139e-05 tokens_per_second_per_gpu:5339.151406403406 mfu:25.185688137169315 +Step 23952 | loss:0.4373715817928314 lr:2.6051131635693982e-05 tokens_per_second_per_gpu:3972.9589447743615 mfu:18.741125199195007 +Step 23953 | loss:0.5583235621452332 lr:2.6047325272149052e-05 tokens_per_second_per_gpu:3434.199018592275 mfu:16.199702705472014 +Step 23954 | loss:0.4292422831058502 lr:2.6043519088755264e-05 tokens_per_second_per_gpu:4460.916810891839 mfu:21.042905707868012 +Step 23955 | loss:0.45534834265708923 lr:2.6039713085541213e-05 tokens_per_second_per_gpu:5325.15657420744 mfu:25.119672125936347 +Step 23956 | loss:0.4039539694786072 lr:2.6035907262535553e-05 tokens_per_second_per_gpu:5326.027264292517 mfu:25.123779319622706 +Step 23957 | loss:0.31686291098594666 lr:2.603210161976687e-05 tokens_per_second_per_gpu:5354.366670750626 mfu:25.257461135086942 +Step 23958 | loss:0.3871062397956848 lr:2.6028296157263808e-05 tokens_per_second_per_gpu:4426.408900788204 mfu:20.880126008252585 +Step 23959 | loss:0.5427175164222717 lr:2.6024490875055014e-05 tokens_per_second_per_gpu:5340.1790691462065 mfu:25.190535797664445 +Step 23960 | loss:0.500786304473877 lr:2.602068577316906e-05 tokens_per_second_per_gpu:3986.293490889347 mfu:18.80402652832757 +Step 23961 | loss:0.4280926585197449 lr:2.601688085163459e-05 tokens_per_second_per_gpu:4959.039131449303 mfu:23.392633682369183 +Step 23962 | loss:0.3922919034957886 lr:2.601307611048024e-05 tokens_per_second_per_gpu:3990.529816077742 mfu:18.82400999703283 +Step 23963 | loss:0.4318576455116272 lr:2.6009271549734604e-05 tokens_per_second_per_gpu:4872.376258651431 mfu:22.98383012517082 +Step 23964 | loss:0.4058879315853119 lr:2.6005467169426278e-05 tokens_per_second_per_gpu:3579.204848389416 mfu:16.883719945170853 +Step 23965 | loss:0.4941137731075287 lr:2.6001662969583895e-05 tokens_per_second_per_gpu:5388.3368357900745 mfu:25.41770419948583 +Step 23966 | loss:0.4147096276283264 lr:2.5997858950236088e-05 tokens_per_second_per_gpu:4201.517620799757 mfu:19.819275470138088 +Step 23967 | loss:0.3943355679512024 lr:2.599405511141142e-05 tokens_per_second_per_gpu:4728.290315425774 mfu:22.3041521473785 +Step 23968 | loss:0.4894155263900757 lr:2.5990251453138537e-05 tokens_per_second_per_gpu:4133.905967940481 mfu:19.500339767863373 +Step 23969 | loss:0.49722012877464294 lr:2.598644797544605e-05 tokens_per_second_per_gpu:4591.842929924234 mfu:21.66050654964318 +Step 23970 | loss:0.47268128395080566 lr:2.5982644678362556e-05 tokens_per_second_per_gpu:4342.935707785382 mfu:20.486368714862962 +Step 23971 | loss:0.34835347533226013 lr:2.5978841561916634e-05 tokens_per_second_per_gpu:4395.2510549285535 mfu:20.733149133255935 +Step 23972 | loss:0.421142041683197 lr:2.5975038626136904e-05 tokens_per_second_per_gpu:5351.379220261524 mfu:25.24336882141823 +Step 23973 | loss:0.46134352684020996 lr:2.5971235871052002e-05 tokens_per_second_per_gpu:3675.0843594121643 mfu:17.335999957396595 +Step 23974 | loss:0.4842773377895355 lr:2.596743329669048e-05 tokens_per_second_per_gpu:4947.517873086849 mfu:23.338285940944115 +Step 23975 | loss:0.4416787028312683 lr:2.5963630903080953e-05 tokens_per_second_per_gpu:4707.847076846845 mfu:22.20771790302552 +Step 23976 | loss:0.4367839992046356 lr:2.5959828690252053e-05 tokens_per_second_per_gpu:4634.314067808464 mfu:21.86085015336643 +Step 23977 | loss:0.41646286845207214 lr:2.5956026658232347e-05 tokens_per_second_per_gpu:4418.411320351616 mfu:20.842400056805275 +Step 23978 | loss:0.3982785940170288 lr:2.5952224807050408e-05 tokens_per_second_per_gpu:5308.858082396595 mfu:25.042789359255142 +Step 23979 | loss:0.4419100284576416 lr:2.5948423136734855e-05 tokens_per_second_per_gpu:4448.174695179568 mfu:20.98279897402411 +Step 23980 | loss:0.3772564232349396 lr:2.594462164731428e-05 tokens_per_second_per_gpu:4060.52355312463 mfu:19.154182396845982 +Step 23981 | loss:0.43309804797172546 lr:2.5940820338817308e-05 tokens_per_second_per_gpu:4106.803864920827 mfu:19.37249452382407 +Step 23982 | loss:0.4932289719581604 lr:2.593701921127246e-05 tokens_per_second_per_gpu:3889.8431243559644 mfu:18.349053693260167 +Step 23983 | loss:0.3296290636062622 lr:2.593321826470839e-05 tokens_per_second_per_gpu:5030.598843770313 mfu:23.7301930547333 +Step 23984 | loss:0.4983375668525696 lr:2.592941749915363e-05 tokens_per_second_per_gpu:5334.645070156612 mfu:25.16443098023294 +Step 23985 | loss:0.4444582760334015 lr:2.592561691463682e-05 tokens_per_second_per_gpu:5046.290410570363 mfu:23.8042128525865 +Step 23986 | loss:0.40745869278907776 lr:2.592181651118649e-05 tokens_per_second_per_gpu:4852.077758198739 mfu:22.888078635254168 +Step 23987 | loss:0.5199843645095825 lr:2.591801628883126e-05 tokens_per_second_per_gpu:3415.437835813557 mfu:16.11120300531694 +Step 23988 | loss:0.4064925014972687 lr:2.591421624759971e-05 tokens_per_second_per_gpu:3756.038171744631 mfu:17.71787290231349 +Step 23989 | loss:0.45934411883354187 lr:2.5910416387520398e-05 tokens_per_second_per_gpu:4423.858282741555 mfu:20.86809430774623 +Step 23990 | loss:0.4141465723514557 lr:2.5906616708621923e-05 tokens_per_second_per_gpu:3654.9242791073852 mfu:17.24090141893457 +Step 23991 | loss:0.43854960799217224 lr:2.5902817210932873e-05 tokens_per_second_per_gpu:4324.157852416031 mfu:20.397790367253236 +Step 23992 | loss:0.3637392520904541 lr:2.5899017894481816e-05 tokens_per_second_per_gpu:4261.567008977441 mfu:20.102538679654327 +Step 23993 | loss:0.429531067609787 lr:2.589521875929729e-05 tokens_per_second_per_gpu:3671.0909470131924 mfu:17.31716234976423 +Step 23994 | loss:0.3892095386981964 lr:2.5891419805407908e-05 tokens_per_second_per_gpu:3702.785350873198 mfu:17.466670260394793 +Step 23995 | loss:0.40609148144721985 lr:2.5887621032842246e-05 tokens_per_second_per_gpu:3402.2421404048664 mfu:16.048956658656355 +Step 23996 | loss:0.3559016287326813 lr:2.588382244162885e-05 tokens_per_second_per_gpu:3819.5852955771743 mfu:18.01763552768894 +Step 23997 | loss:0.44907721877098083 lr:2.58800240317963e-05 tokens_per_second_per_gpu:5340.923981685685 mfu:25.19404967720835 +Step 23998 | loss:0.4200417697429657 lr:2.587622580337319e-05 tokens_per_second_per_gpu:4444.126034824486 mfu:20.9637007523565 +Step 23999 | loss:0.43059009313583374 lr:2.5872427756388062e-05 tokens_per_second_per_gpu:3797.210556548819 mfu:17.912090066168062 +Step 24000 | loss:0.4445280432701111 lr:2.5868629890869468e-05 tokens_per_second_per_gpu:5334.53063828798 mfu:25.16389118558451 +Step 24001 | loss:0.42863860726356506 lr:2.586483220684598e-05 tokens_per_second_per_gpu:5069.347564969567 mfu:23.912977383843746 +Step 24002 | loss:0.3421899080276489 lr:2.5861034704346187e-05 tokens_per_second_per_gpu:4861.860107695037 mfu:22.93422364686914 +Step 24003 | loss:0.43883875012397766 lr:2.585723738339862e-05 tokens_per_second_per_gpu:4711.883634651608 mfu:22.226759034898365 +Step 24004 | loss:0.3194815218448639 lr:2.5853440244031847e-05 tokens_per_second_per_gpu:5093.89257399713 mfu:24.02876037923713 +Step 24005 | loss:0.5494819283485413 lr:2.584964328627445e-05 tokens_per_second_per_gpu:4564.950297245366 mfu:21.533649412940683 +Step 24006 | loss:0.4630007743835449 lr:2.584584651015497e-05 tokens_per_second_per_gpu:4097.276691973074 mfu:19.327553223526195 +Step 24007 | loss:0.3789623975753784 lr:2.5842049915701937e-05 tokens_per_second_per_gpu:4067.4557935968514 mfu:19.186882957915635 +Step 24008 | loss:0.3811793029308319 lr:2.5838253502943937e-05 tokens_per_second_per_gpu:5324.086725368637 mfu:25.114625466428684 +Step 24009 | loss:0.41882041096687317 lr:2.583445727190953e-05 tokens_per_second_per_gpu:4845.780634201387 mfu:22.85837402695852 +Step 24010 | loss:0.41722530126571655 lr:2.5830661222627235e-05 tokens_per_second_per_gpu:4781.426468543081 mfu:22.55480444759541 +Step 24011 | loss:0.4170682728290558 lr:2.5826865355125616e-05 tokens_per_second_per_gpu:4330.730652793051 mfu:20.428795388067027 +Step 24012 | loss:0.4282991290092468 lr:2.5823069669433254e-05 tokens_per_second_per_gpu:4703.4059247571195 mfu:22.186768230879604 +Step 24013 | loss:0.4314599931240082 lr:2.5819274165578665e-05 tokens_per_second_per_gpu:3891.4000940682695 mfu:18.356398185039534 +Step 24014 | loss:0.43679279088974 lr:2.5815478843590385e-05 tokens_per_second_per_gpu:5363.7746666504945 mfu:25.30184025691624 +Step 24015 | loss:0.3530940115451813 lr:2.581168370349697e-05 tokens_per_second_per_gpu:5108.878687548377 mfu:24.09945243375299 +Step 24016 | loss:0.4936787188053131 lr:2.5807888745326975e-05 tokens_per_second_per_gpu:5352.421357807965 mfu:25.248284762031794 +Step 24017 | loss:0.4681691825389862 lr:2.580409396910895e-05 tokens_per_second_per_gpu:4607.877294160474 mfu:21.73614338149016 +Step 24018 | loss:0.34091320633888245 lr:2.5800299374871407e-05 tokens_per_second_per_gpu:4764.3593986905025 mfu:22.47429616715841 +Step 24019 | loss:0.4739876687526703 lr:2.5796504962642892e-05 tokens_per_second_per_gpu:4928.233567008615 mfu:23.247318578931186 +Step 24020 | loss:0.4471127688884735 lr:2.5792710732451997e-05 tokens_per_second_per_gpu:5163.249384051375 mfu:24.355928285755034 +Step 24021 | loss:0.4012817442417145 lr:2.5788916684327168e-05 tokens_per_second_per_gpu:4553.2917155656 mfu:21.47865389399808 +Step 24022 | loss:0.37770724296569824 lr:2.578512281829699e-05 tokens_per_second_per_gpu:3522.0549789679276 mfu:16.614134260336503 +Step 24023 | loss:0.44367513060569763 lr:2.5781329134389996e-05 tokens_per_second_per_gpu:5369.139093530911 mfu:25.327145173776767 +Step 24024 | loss:0.5214215517044067 lr:2.577753563263473e-05 tokens_per_second_per_gpu:5373.319975578702 mfu:25.34686710772823 +Step 24025 | loss:0.48685088753700256 lr:2.5773742313059696e-05 tokens_per_second_per_gpu:4639.614178684486 mfu:21.885851680660792 +Step 24026 | loss:0.49448883533477783 lr:2.576994917569343e-05 tokens_per_second_per_gpu:4613.791401889867 mfu:21.76404123670054 +Step 24027 | loss:0.43683210015296936 lr:2.5766156220564497e-05 tokens_per_second_per_gpu:4957.559610623999 mfu:23.385654530203944 +Step 24028 | loss:0.5027326941490173 lr:2.5762363447701387e-05 tokens_per_second_per_gpu:4952.144091448012 mfu:23.360108602267964 +Step 24029 | loss:0.26639220118522644 lr:2.5758570857132618e-05 tokens_per_second_per_gpu:5312.814859087066 mfu:25.06145414999981 +Step 24030 | loss:0.3280786871910095 lr:2.575477844888673e-05 tokens_per_second_per_gpu:5361.950508037136 mfu:25.293235389503476 +Step 24031 | loss:0.3972759544849396 lr:2.575098622299227e-05 tokens_per_second_per_gpu:5346.652787071393 mfu:25.221073429647724 +Step 24032 | loss:0.5183061361312866 lr:2.5747194179477713e-05 tokens_per_second_per_gpu:5102.516221511429 mfu:24.069439595908065 +Step 24033 | loss:0.457552433013916 lr:2.57434023183716e-05 tokens_per_second_per_gpu:4194.782531389788 mfu:19.787504904266516 +Step 24034 | loss:0.4588240087032318 lr:2.5739610639702484e-05 tokens_per_second_per_gpu:4835.793201491937 mfu:22.811261602836307 +Step 24035 | loss:0.340724378824234 lr:2.573581914349884e-05 tokens_per_second_per_gpu:4709.729476837624 mfu:22.21659750495258 +Step 24036 | loss:0.5223605632781982 lr:2.5732027829789183e-05 tokens_per_second_per_gpu:5311.828981167521 mfu:25.056803595645203 +Step 24037 | loss:0.46692603826522827 lr:2.5728236698602042e-05 tokens_per_second_per_gpu:4700.051713261126 mfu:22.17094584296551 +Step 24038 | loss:0.4119068682193756 lr:2.5724445749965953e-05 tokens_per_second_per_gpu:5317.449637359196 mfu:25.083317189884163 +Step 24039 | loss:0.4808904528617859 lr:2.5720654983909375e-05 tokens_per_second_per_gpu:4977.183338513701 mfu:23.478223002812893 +Step 24040 | loss:0.4746907353401184 lr:2.571686440046086e-05 tokens_per_second_per_gpu:3494.7030656838647 mfu:16.485110618657952 +Step 24041 | loss:0.5051085352897644 lr:2.5713073999648917e-05 tokens_per_second_per_gpu:5022.404066847054 mfu:23.69153689381317 +Step 24042 | loss:0.37435808777809143 lr:2.5709283781502048e-05 tokens_per_second_per_gpu:5364.780803133896 mfu:25.30658637437303 +Step 24043 | loss:0.46204546093940735 lr:2.5705493746048738e-05 tokens_per_second_per_gpu:5141.026751439646 mfu:24.25110033615341 +Step 24044 | loss:0.4406989514827728 lr:2.5701703893317504e-05 tokens_per_second_per_gpu:4456.517045941961 mfu:21.02215126592213 +Step 24045 | loss:0.5416535139083862 lr:2.569791422333687e-05 tokens_per_second_per_gpu:5372.372139010876 mfu:25.342396000920203 +Step 24046 | loss:0.3131214380264282 lr:2.5694124736135315e-05 tokens_per_second_per_gpu:3663.38366606705 mfu:17.28080579054355 +Step 24047 | loss:0.42215755581855774 lr:2.5690335431741346e-05 tokens_per_second_per_gpu:3863.7291066253492 mfu:18.225869416113994 +Step 24048 | loss:0.44234293699264526 lr:2.568654631018349e-05 tokens_per_second_per_gpu:5274.294253115734 mfu:24.879745879339946 +Step 24049 | loss:0.3424903452396393 lr:2.568275737149022e-05 tokens_per_second_per_gpu:5260.684600072052 mfu:24.815546823886645 +Step 24050 | loss:0.44025346636772156 lr:2.567896861569001e-05 tokens_per_second_per_gpu:4437.206018574426 mfu:20.93105785502845 +Step 24051 | loss:0.4917434751987457 lr:2.567518004281138e-05 tokens_per_second_per_gpu:5245.384291200051 mfu:24.74337265643554 +Step 24052 | loss:0.3520858883857727 lr:2.5671391652882826e-05 tokens_per_second_per_gpu:4385.64627857781 mfu:20.68784176446565 +Step 24053 | loss:0.4357035160064697 lr:2.5667603445932865e-05 tokens_per_second_per_gpu:4096.390570842789 mfu:19.323373238966788 +Step 24054 | loss:0.3838574290275574 lr:2.5663815421989935e-05 tokens_per_second_per_gpu:5303.880101701344 mfu:25.019307374984557 +Step 24055 | loss:0.3467620611190796 lr:2.5660027581082558e-05 tokens_per_second_per_gpu:5305.638587403407 mfu:25.027602452069786 +Step 24056 | loss:0.44032022356987 lr:2.565623992323927e-05 tokens_per_second_per_gpu:4044.2033208686066 mfu:19.07719707677557 +Step 24057 | loss:0.3807118535041809 lr:2.5652452448488452e-05 tokens_per_second_per_gpu:4400.347176887705 mfu:20.757188410025318 +Step 24058 | loss:0.4682278633117676 lr:2.564866515685865e-05 tokens_per_second_per_gpu:5119.054913900037 mfu:24.147455429697022 +Step 24059 | loss:0.3268921971321106 lr:2.564487804837835e-05 tokens_per_second_per_gpu:3851.762551843427 mfu:18.169421135502887 +Step 24060 | loss:0.507422685623169 lr:2.5641091123076045e-05 tokens_per_second_per_gpu:5323.301414171568 mfu:25.1109210195242 +Step 24061 | loss:0.3697543442249298 lr:2.5637304380980187e-05 tokens_per_second_per_gpu:5268.636402353308 mfu:24.8530568319647 +Step 24062 | loss:0.43249964714050293 lr:2.563351782211927e-05 tokens_per_second_per_gpu:4088.896782284465 mfu:19.288023759765004 +Step 24063 | loss:0.5176740288734436 lr:2.5629731446521788e-05 tokens_per_second_per_gpu:5329.8917887610205 mfu:25.142008940896503 +Step 24064 | loss:0.4744633734226227 lr:2.5625945254216212e-05 tokens_per_second_per_gpu:4961.2349517717075 mfu:23.402991741475546 +Step 24065 | loss:0.4844140410423279 lr:2.5622159245230988e-05 tokens_per_second_per_gpu:5180.186957921537 mfu:24.43582571154868 +Step 24066 | loss:0.35311296582221985 lr:2.5618373419594614e-05 tokens_per_second_per_gpu:3440.552463583563 mfu:16.229672989505413 +Step 24067 | loss:0.5193790793418884 lr:2.5614587777335585e-05 tokens_per_second_per_gpu:3498.0098720371575 mfu:16.50070938842578 +Step 24068 | loss:0.5299972295761108 lr:2.561080231848234e-05 tokens_per_second_per_gpu:5328.733837321476 mfu:25.13654668638928 +Step 24069 | loss:0.46193063259124756 lr:2.560701704306336e-05 tokens_per_second_per_gpu:4971.39200075469 mfu:23.450904274499532 +Step 24070 | loss:0.4733788073062897 lr:2.560323195110713e-05 tokens_per_second_per_gpu:3166.4545664450284 mfu:14.936706442779345 +Step 24071 | loss:0.40225303173065186 lr:2.5599447042642115e-05 tokens_per_second_per_gpu:5339.415803818493 mfu:25.18693534488713 +Step 24072 | loss:0.3635671138763428 lr:2.559566231769675e-05 tokens_per_second_per_gpu:3708.642525220112 mfu:17.49429955112618 +Step 24073 | loss:0.4899146556854248 lr:2.559187777629952e-05 tokens_per_second_per_gpu:5369.3326249026795 mfu:25.328058094277257 +Step 24074 | loss:0.47740504145622253 lr:2.5588093418478915e-05 tokens_per_second_per_gpu:3427.85900894964 mfu:16.169795798270517 +Step 24075 | loss:0.4317217469215393 lr:2.558430924426335e-05 tokens_per_second_per_gpu:4609.373382268889 mfu:21.74320068435651 +Step 24076 | loss:0.4098987281322479 lr:2.5580525253681314e-05 tokens_per_second_per_gpu:4377.476880717985 mfu:20.649305320917854 +Step 24077 | loss:0.42849305272102356 lr:2.557674144676128e-05 tokens_per_second_per_gpu:4574.75336512999 mfu:21.579892156728306 +Step 24078 | loss:0.49692365527153015 lr:2.5572957823531696e-05 tokens_per_second_per_gpu:5330.563256007994 mfu:25.1451763664646 +Step 24079 | loss:0.3470524549484253 lr:2.556917438402099e-05 tokens_per_second_per_gpu:3763.670404673539 mfu:17.753875447232364 +Step 24080 | loss:0.3974836766719818 lr:2.5565391128257644e-05 tokens_per_second_per_gpu:4704.794689519534 mfu:22.19331927121161 +Step 24081 | loss:0.520199716091156 lr:2.556160805627012e-05 tokens_per_second_per_gpu:4052.3091662295365 mfu:19.1154337323429 +Step 24082 | loss:0.46191149950027466 lr:2.5557825168086847e-05 tokens_per_second_per_gpu:5377.498791770079 mfu:25.36657929668252 +Step 24083 | loss:0.3807431757450104 lr:2.555404246373629e-05 tokens_per_second_per_gpu:5365.494786145493 mfu:25.309954354056792 +Step 24084 | loss:0.5341723561286926 lr:2.5550259943246897e-05 tokens_per_second_per_gpu:3708.477821160237 mfu:17.49352261397434 +Step 24085 | loss:0.5234780311584473 lr:2.5546477606647156e-05 tokens_per_second_per_gpu:3892.997422056357 mfu:18.363933053691603 +Step 24086 | loss:0.516869068145752 lr:2.5542695453965443e-05 tokens_per_second_per_gpu:3785.0518828856993 mfu:17.854735528015485 +Step 24087 | loss:0.4511927366256714 lr:2.5538913485230236e-05 tokens_per_second_per_gpu:5328.7054235506685 mfu:25.136412653784493 +Step 24088 | loss:0.3402056396007538 lr:2.5535131700470007e-05 tokens_per_second_per_gpu:4610.5562802639315 mfu:21.748780616022334 +Step 24089 | loss:0.4368298351764679 lr:2.5531350099713146e-05 tokens_per_second_per_gpu:5370.9166768999585 mfu:25.33553033781556 +Step 24090 | loss:0.5017246007919312 lr:2.5527568682988122e-05 tokens_per_second_per_gpu:5317.855354365318 mfu:25.085231026224502 +Step 24091 | loss:0.4876694977283478 lr:2.5523787450323385e-05 tokens_per_second_per_gpu:4411.403043968189 mfu:20.809340821369396 +Step 24092 | loss:0.44097900390625 lr:2.5520006401747398e-05 tokens_per_second_per_gpu:5147.946312510125 mfu:24.283741086321715 +Step 24093 | loss:0.4294949471950531 lr:2.551622553728853e-05 tokens_per_second_per_gpu:4133.489921613807 mfu:19.49837720635563 +Step 24094 | loss:0.4586372673511505 lr:2.551244485697526e-05 tokens_per_second_per_gpu:4887.442460039027 mfu:23.05489996767513 +Step 24095 | loss:0.4674016833305359 lr:2.550866436083601e-05 tokens_per_second_per_gpu:4907.992414609508 mfu:23.151837609567888 +Step 24096 | loss:0.45094454288482666 lr:2.5504884048899236e-05 tokens_per_second_per_gpu:4794.723917630617 mfu:22.61753078372008 +Step 24097 | loss:0.5123192071914673 lr:2.5501103921193342e-05 tokens_per_second_per_gpu:4562.129992944249 mfu:21.520345556358787 +Step 24098 | loss:0.4188607633113861 lr:2.5497323977746767e-05 tokens_per_second_per_gpu:4858.1406772352575 mfu:22.916678458790084 +Step 24099 | loss:0.4027984142303467 lr:2.549354421858797e-05 tokens_per_second_per_gpu:4272.5370277302 mfu:20.154286129789288 +Step 24100 | loss:0.41031327843666077 lr:2.548976464374535e-05 tokens_per_second_per_gpu:5003.191503584173 mfu:23.600907954900848 +Step 24101 | loss:0.5727521777153015 lr:2.5485985253247325e-05 tokens_per_second_per_gpu:4907.137222858818 mfu:23.147803524169124 +Step 24102 | loss:0.380315363407135 lr:2.5482206047122324e-05 tokens_per_second_per_gpu:5333.339591568663 mfu:25.15827281499597 +Step 24103 | loss:0.3251801133155823 lr:2.54784270253988e-05 tokens_per_second_per_gpu:5358.263475767685 mfu:25.2758430292159 +Step 24104 | loss:0.36791637539863586 lr:2.547464818810514e-05 tokens_per_second_per_gpu:5326.2159521199565 mfu:25.124669392298202 +Step 24105 | loss:0.36815086007118225 lr:2.547086953526978e-05 tokens_per_second_per_gpu:4102.027027676852 mfu:19.349961367531765 +Step 24106 | loss:0.4864028990268707 lr:2.5467091066921157e-05 tokens_per_second_per_gpu:4579.486552436465 mfu:21.602219408817977 +Step 24107 | loss:0.3693002164363861 lr:2.5463312783087672e-05 tokens_per_second_per_gpu:3579.092196864658 mfu:16.88318854870824 +Step 24108 | loss:0.38815709948539734 lr:2.5459534683797727e-05 tokens_per_second_per_gpu:5258.575927369644 mfu:24.805599854972726 +Step 24109 | loss:0.4576992690563202 lr:2.5455756769079754e-05 tokens_per_second_per_gpu:4248.032188567308 mfu:20.03869262250124 +Step 24110 | loss:0.42804571986198425 lr:2.5451979038962183e-05 tokens_per_second_per_gpu:5358.870281263047 mfu:25.278705434269217 +Step 24111 | loss:0.469175785779953 lr:2.5448201493473383e-05 tokens_per_second_per_gpu:4520.61791085504 mfu:21.324525982452435 +Step 24112 | loss:0.4543065130710602 lr:2.5444424132641797e-05 tokens_per_second_per_gpu:4063.525627231398 mfu:19.168343692614265 +Step 24113 | loss:0.5201919674873352 lr:2.5440646956495856e-05 tokens_per_second_per_gpu:4090.055829822758 mfu:19.29349118475703 +Step 24114 | loss:0.3662043511867523 lr:2.5436869965063938e-05 tokens_per_second_per_gpu:4682.743661206213 mfu:22.089300808364296 +Step 24115 | loss:0.4858158826828003 lr:2.543309315837444e-05 tokens_per_second_per_gpu:5042.06982170502 mfu:23.784303614801967 +Step 24116 | loss:0.4177335202693939 lr:2.5429316536455782e-05 tokens_per_second_per_gpu:4697.433062253451 mfu:22.15859321937437 +Step 24117 | loss:0.37960219383239746 lr:2.5425540099336387e-05 tokens_per_second_per_gpu:3973.4571160538603 mfu:18.74347515811719 +Step 24118 | loss:0.3862976133823395 lr:2.5421763847044623e-05 tokens_per_second_per_gpu:5222.352856618396 mfu:24.63472944994593 +Step 24119 | loss:0.45683103799819946 lr:2.5417987779608905e-05 tokens_per_second_per_gpu:5071.703571577199 mfu:23.924091068988123 +Step 24120 | loss:0.4363366663455963 lr:2.5414211897057637e-05 tokens_per_second_per_gpu:5317.595008632865 mfu:25.08400293098496 +Step 24121 | loss:0.5041514039039612 lr:2.5410436199419262e-05 tokens_per_second_per_gpu:4663.586730319097 mfu:21.99893429686032 +Step 24122 | loss:0.4359508454799652 lr:2.5406660686722095e-05 tokens_per_second_per_gpu:4323.020294314121 mfu:20.3924243115991 +Step 24123 | loss:0.4540475010871887 lr:2.540288535899457e-05 tokens_per_second_per_gpu:3740.2489727444104 mfu:17.64339255671431 +Step 24124 | loss:0.45177412033081055 lr:2.53991102162651e-05 tokens_per_second_per_gpu:5344.443669079752 mfu:25.210652643166757 +Step 24125 | loss:0.4593346416950226 lr:2.5395335258562048e-05 tokens_per_second_per_gpu:4038.142532592602 mfu:19.04860730440054 +Step 24126 | loss:0.5014083981513977 lr:2.5391560485913813e-05 tokens_per_second_per_gpu:5307.259278740875 mfu:25.035247529627078 +Step 24127 | loss:0.4546787738800049 lr:2.5387785898348792e-05 tokens_per_second_per_gpu:5313.630442677952 mfu:25.06530139695093 +Step 24128 | loss:0.4432309567928314 lr:2.5384011495895415e-05 tokens_per_second_per_gpu:3066.0934290361542 mfu:14.463285834245935 +Step 24129 | loss:0.4708828330039978 lr:2.5380237278581987e-05 tokens_per_second_per_gpu:4276.441074486158 mfu:20.1727021844362 +Step 24130 | loss:0.40504929423332214 lr:2.5376463246436944e-05 tokens_per_second_per_gpu:4903.108847272055 mfu:23.128800989214614 +Step 24131 | loss:0.44242292642593384 lr:2.5372689399488658e-05 tokens_per_second_per_gpu:5393.163510434232 mfu:25.44047244729793 +Step 24132 | loss:0.4646146595478058 lr:2.536891573776553e-05 tokens_per_second_per_gpu:4094.010142125246 mfu:19.31214435056312 +Step 24133 | loss:0.4109855592250824 lr:2.5365142261295916e-05 tokens_per_second_per_gpu:5332.311815335842 mfu:25.153424619149042 +Step 24134 | loss:0.5586763024330139 lr:2.5361368970108213e-05 tokens_per_second_per_gpu:4293.3307160570885 mfu:20.25237350539628 +Step 24135 | loss:0.36214473843574524 lr:2.5357595864230815e-05 tokens_per_second_per_gpu:5357.36039040843 mfu:25.271583021494042 +Step 24136 | loss:0.4231470823287964 lr:2.5353822943692086e-05 tokens_per_second_per_gpu:3782.527446528281 mfu:17.842827331004287 +Step 24137 | loss:0.42511996626853943 lr:2.5350050208520372e-05 tokens_per_second_per_gpu:4009.13999534376 mfu:18.91179738785464 +Step 24138 | loss:0.41971465945243835 lr:2.5346277658744085e-05 tokens_per_second_per_gpu:4294.541822446508 mfu:20.258086500871237 +Step 24139 | loss:0.4466719925403595 lr:2.5342505294391604e-05 tokens_per_second_per_gpu:5356.834831127123 mfu:25.269103868694458 +Step 24140 | loss:0.4001157283782959 lr:2.5338733115491264e-05 tokens_per_second_per_gpu:4353.580708186017 mfu:20.536582997977103 +Step 24141 | loss:0.42039424180984497 lr:2.5334961122071465e-05 tokens_per_second_per_gpu:3307.0332577516333 mfu:15.599840114870611 +Step 24142 | loss:0.3943282961845398 lr:2.5331189314160585e-05 tokens_per_second_per_gpu:4041.535872851612 mfu:19.06461426936584 +Step 24143 | loss:0.37061968445777893 lr:2.532741769178698e-05 tokens_per_second_per_gpu:4219.979126783337 mfu:19.906361543720354 +Step 24144 | loss:0.45421990752220154 lr:2.5323646254979e-05 tokens_per_second_per_gpu:5332.624824103258 mfu:25.154901134909746 +Step 24145 | loss:0.288386732339859 lr:2.5319875003765027e-05 tokens_per_second_per_gpu:5356.575609854192 mfu:25.267881077722333 +Step 24146 | loss:0.43225398659706116 lr:2.531610393817343e-05 tokens_per_second_per_gpu:5147.689730266763 mfu:24.282530744102335 +Step 24147 | loss:0.4549078643321991 lr:2.5312333058232558e-05 tokens_per_second_per_gpu:5350.307627761435 mfu:25.238313936762776 +Step 24148 | loss:0.4172990024089813 lr:2.5308562363970772e-05 tokens_per_second_per_gpu:3738.5313686333507 mfu:17.63529032506896 +Step 24149 | loss:0.39516404271125793 lr:2.5304791855416467e-05 tokens_per_second_per_gpu:3579.0419430954616 mfu:16.882951493104763 +Step 24150 | loss:0.49993839859962463 lr:2.5301021532597964e-05 tokens_per_second_per_gpu:3949.7563725276677 mfu:18.631674707140316 +Step 24151 | loss:0.3244824707508087 lr:2.529725139554361e-05 tokens_per_second_per_gpu:5175.910097454432 mfu:24.415651031017703 +Step 24152 | loss:0.39683210849761963 lr:2.5293481444281787e-05 tokens_per_second_per_gpu:5310.565856203374 mfu:25.050845219677978 +Step 24153 | loss:0.4795263707637787 lr:2.5289711678840854e-05 tokens_per_second_per_gpu:5059.920041390111 mfu:23.868506146618063 +Step 24154 | loss:0.4787094295024872 lr:2.5285942099249137e-05 tokens_per_second_per_gpu:4120.817476779232 mfu:19.438599121928895 +Step 24155 | loss:0.4343740940093994 lr:2.5282172705535013e-05 tokens_per_second_per_gpu:5383.585146375979 mfu:25.395289669797727 +Step 24156 | loss:0.41418805718421936 lr:2.5278403497726833e-05 tokens_per_second_per_gpu:5315.641435394466 mfu:25.074787592705466 +Step 24157 | loss:0.3984803557395935 lr:2.5274634475852932e-05 tokens_per_second_per_gpu:5323.10355279577 mfu:25.10998767365593 +Step 24158 | loss:0.4112201929092407 lr:2.527086563994164e-05 tokens_per_second_per_gpu:4507.129219962879 mfu:21.26089752610613 +Step 24159 | loss:0.33524906635284424 lr:2.5267096990021328e-05 tokens_per_second_per_gpu:3653.375949564251 mfu:17.23359768430734 +Step 24160 | loss:0.5481148958206177 lr:2.5263328526120355e-05 tokens_per_second_per_gpu:5086.4492100500665 mfu:23.99364876152999 +Step 24161 | loss:0.42263075709342957 lr:2.5259560248267022e-05 tokens_per_second_per_gpu:5345.244207517994 mfu:25.21442891956589 +Step 24162 | loss:0.43913161754608154 lr:2.5255792156489694e-05 tokens_per_second_per_gpu:5008.313212904505 mfu:23.625067931618577 +Step 24163 | loss:0.49873390793800354 lr:2.525202425081671e-05 tokens_per_second_per_gpu:5357.01401847319 mfu:25.26994912597837 +Step 24164 | loss:0.4416326880455017 lr:2.524825653127645e-05 tokens_per_second_per_gpu:5019.890198656796 mfu:23.67967854864958 +Step 24165 | loss:0.3438236117362976 lr:2.524448899789717e-05 tokens_per_second_per_gpu:3300.6747365363944 mfu:15.569845885422742 +Step 24166 | loss:0.3325873017311096 lr:2.524072165070725e-05 tokens_per_second_per_gpu:4992.4869237223165 mfu:23.550412625302815 +Step 24167 | loss:0.3914770185947418 lr:2.523695448973502e-05 tokens_per_second_per_gpu:4602.817493499288 mfu:21.712275438480333 +Step 24168 | loss:0.4143834412097931 lr:2.5233187515008838e-05 tokens_per_second_per_gpu:3724.167771253548 mfu:17.567534785546012 +Step 24169 | loss:0.45186731219291687 lr:2.5229420726556985e-05 tokens_per_second_per_gpu:4219.556342501561 mfu:19.90436719812878 +Step 24170 | loss:0.35454675555229187 lr:2.5225654124407827e-05 tokens_per_second_per_gpu:5357.804313907351 mfu:25.27367708437961 +Step 24171 | loss:0.4021921455860138 lr:2.5221887708589702e-05 tokens_per_second_per_gpu:5010.23092443376 mfu:23.6341141040974 +Step 24172 | loss:0.4535538852214813 lr:2.5218121479130925e-05 tokens_per_second_per_gpu:4154.390664189427 mfu:19.596969575119232 +Step 24173 | loss:0.4451060891151428 lr:2.5214355436059796e-05 tokens_per_second_per_gpu:5326.579473689888 mfu:25.126384185566742 +Step 24174 | loss:0.36707815527915955 lr:2.5210589579404663e-05 tokens_per_second_per_gpu:3153.002981282276 mfu:14.873253020489551 +Step 24175 | loss:0.347797155380249 lr:2.5206823909193866e-05 tokens_per_second_per_gpu:4742.832470622183 mfu:22.372749974587485 +Step 24176 | loss:0.44254592061042786 lr:2.5203058425455696e-05 tokens_per_second_per_gpu:3331.5047836038016 mfu:15.715276477588139 +Step 24177 | loss:0.39932557940483093 lr:2.5199293128218483e-05 tokens_per_second_per_gpu:4308.872266330674 mfu:20.325685649698695 +Step 24178 | loss:0.3381248414516449 lr:2.5195528017510573e-05 tokens_per_second_per_gpu:5349.156617233776 mfu:25.232884423721163 +Step 24179 | loss:0.47708430886268616 lr:2.5191763093360253e-05 tokens_per_second_per_gpu:5368.647795388494 mfu:25.32482763661488 +Step 24180 | loss:0.5793077349662781 lr:2.518799835579584e-05 tokens_per_second_per_gpu:4328.61745694846 mfu:20.418827082719698 +Step 24181 | loss:0.4891372323036194 lr:2.5184233804845654e-05 tokens_per_second_per_gpu:4905.8272177138315 mfu:23.1416240063903 +Step 24182 | loss:0.4426901638507843 lr:2.5180469440538033e-05 tokens_per_second_per_gpu:5339.996940102408 mfu:25.18967666389043 +Step 24183 | loss:0.38855794072151184 lr:2.5176705262901245e-05 tokens_per_second_per_gpu:5147.064464868781 mfu:24.279581260538773 +Step 24184 | loss:0.4672107398509979 lr:2.5172941271963624e-05 tokens_per_second_per_gpu:4086.1875920839097 mfu:19.27524404735347 +Step 24185 | loss:0.32082584500312805 lr:2.51691774677535e-05 tokens_per_second_per_gpu:3789.394555000655 mfu:17.875220653317996 +Step 24186 | loss:0.3141871988773346 lr:2.5165413850299168e-05 tokens_per_second_per_gpu:3959.9236016868094 mfu:18.679635261792157 +Step 24187 | loss:0.4209214746952057 lr:2.51616504196289e-05 tokens_per_second_per_gpu:4583.918318450759 mfu:21.62312480524466 +Step 24188 | loss:0.37871214747428894 lr:2.5157887175771032e-05 tokens_per_second_per_gpu:4144.148398475023 mfu:19.54865506023366 +Step 24189 | loss:0.3705979883670807 lr:2.515412411875388e-05 tokens_per_second_per_gpu:4779.202568859881 mfu:22.544313933353344 +Step 24190 | loss:0.41891980171203613 lr:2.5150361248605714e-05 tokens_per_second_per_gpu:4990.944750879472 mfu:23.543137932881297 +Step 24191 | loss:0.32369738817214966 lr:2.514659856535485e-05 tokens_per_second_per_gpu:5360.1948019273295 mfu:25.284953424229165 +Step 24192 | loss:0.41490188241004944 lr:2.5142836069029612e-05 tokens_per_second_per_gpu:5322.579302378579 mfu:25.107514694991696 +Step 24193 | loss:0.43508636951446533 lr:2.5139073759658277e-05 tokens_per_second_per_gpu:5322.43245397457 mfu:25.106821986015067 +Step 24194 | loss:0.44918984174728394 lr:2.513531163726912e-05 tokens_per_second_per_gpu:5294.246021485106 mfu:24.973861774860982 +Step 24195 | loss:0.4340982139110565 lr:2.5131549701890456e-05 tokens_per_second_per_gpu:3964.506312395682 mfu:18.701252690096982 +Step 24196 | loss:0.49464109539985657 lr:2.5127787953550598e-05 tokens_per_second_per_gpu:5336.409174579899 mfu:25.172752561784833 +Step 24197 | loss:0.42125043272972107 lr:2.51240263922778e-05 tokens_per_second_per_gpu:5356.464354949413 mfu:25.267356269354448 +Step 24198 | loss:0.5182522535324097 lr:2.5120265018100374e-05 tokens_per_second_per_gpu:4889.522813591451 mfu:23.064713350326834 +Step 24199 | loss:0.4136519730091095 lr:2.511650383104661e-05 tokens_per_second_per_gpu:4526.625728848697 mfu:21.352865884967965 +Step 24200 | loss:0.46081623435020447 lr:2.511274283114484e-05 tokens_per_second_per_gpu:4449.765767186504 mfu:20.9903043321457 +Step 24201 | loss:0.5618675947189331 lr:2.5108982018423265e-05 tokens_per_second_per_gpu:4096.799888247165 mfu:19.32530405899998 +Step 24202 | loss:0.48043447732925415 lr:2.5105221392910204e-05 tokens_per_second_per_gpu:4070.9633398433816 mfu:19.203428652009617 +Step 24203 | loss:0.528592586517334 lr:2.510146095463396e-05 tokens_per_second_per_gpu:4045.027978198202 mfu:19.081087126100314 +Step 24204 | loss:0.3739960491657257 lr:2.5097700703622824e-05 tokens_per_second_per_gpu:3600.2051528667093 mfu:16.98278196441166 +Step 24205 | loss:0.34474778175354004 lr:2.5093940639905033e-05 tokens_per_second_per_gpu:3803.7388846069503 mfu:17.94288530346589 +Step 24206 | loss:0.3859734833240509 lr:2.5090180763508896e-05 tokens_per_second_per_gpu:5130.789367319238 mfu:24.202808848579604 +Step 24207 | loss:0.32525965571403503 lr:2.508642107446271e-05 tokens_per_second_per_gpu:5306.161004776666 mfu:25.030066783953046 +Step 24208 | loss:0.5765222311019897 lr:2.508266157279473e-05 tokens_per_second_per_gpu:3533.59750512198 mfu:16.6685823255632 +Step 24209 | loss:0.3503585159778595 lr:2.5078902258533206e-05 tokens_per_second_per_gpu:4084.2762878438803 mfu:19.26622809915112 +Step 24210 | loss:0.38502126932144165 lr:2.507514313170644e-05 tokens_per_second_per_gpu:3886.8828616281858 mfu:18.33508962890033 +Step 24211 | loss:0.33582034707069397 lr:2.507138419234273e-05 tokens_per_second_per_gpu:5337.627529140109 mfu:25.178499748118067 +Step 24212 | loss:0.4144074022769928 lr:2.5067625440470287e-05 tokens_per_second_per_gpu:5209.7774127806415 mfu:24.575408935771275 +Step 24213 | loss:0.48591935634613037 lr:2.5063866876117425e-05 tokens_per_second_per_gpu:5191.0762784332 mfu:24.487192494311856 +Step 24214 | loss:0.5082643628120422 lr:2.5060108499312417e-05 tokens_per_second_per_gpu:4191.482936794959 mfu:19.77194015359381 +Step 24215 | loss:0.4539927542209625 lr:2.505635031008352e-05 tokens_per_second_per_gpu:5030.830309248288 mfu:23.73128491688499 +Step 24216 | loss:0.5327274799346924 lr:2.5052592308458977e-05 tokens_per_second_per_gpu:4655.874555983283 mfu:21.96255465468566 +Step 24217 | loss:0.43993598222732544 lr:2.5048834494467066e-05 tokens_per_second_per_gpu:5308.310337820741 mfu:25.040205554635975 +Step 24218 | loss:0.4278246760368347 lr:2.5045076868136075e-05 tokens_per_second_per_gpu:4258.02687419848 mfu:20.085839259892687 +Step 24219 | loss:0.3786880671977997 lr:2.5041319429494225e-05 tokens_per_second_per_gpu:4651.753289734346 mfu:21.943113938628894 +Step 24220 | loss:0.3097337484359741 lr:2.5037562178569797e-05 tokens_per_second_per_gpu:5326.463173320575 mfu:25.125835576881805 +Step 24221 | loss:0.44830241799354553 lr:2.5033805115391062e-05 tokens_per_second_per_gpu:5353.832748706884 mfu:25.25494253371011 +Step 24222 | loss:0.3672883212566376 lr:2.5030048239986274e-05 tokens_per_second_per_gpu:4783.7887670303 mfu:22.565947812609306 +Step 24223 | loss:0.44594016671180725 lr:2.5026291552383656e-05 tokens_per_second_per_gpu:4699.5023964534375 mfu:22.16835462186053 +Step 24224 | loss:0.47860386967658997 lr:2.5022535052611485e-05 tokens_per_second_per_gpu:5346.441605142125 mfu:25.220077248456015 +Step 24225 | loss:0.49369174242019653 lr:2.501877874069804e-05 tokens_per_second_per_gpu:5024.096558345431 mfu:23.69952066497984 +Step 24226 | loss:0.46718305349349976 lr:2.5015022616671523e-05 tokens_per_second_per_gpu:3513.691308898261 mfu:16.574681401628624 +Step 24227 | loss:0.38223034143447876 lr:2.5011266680560208e-05 tokens_per_second_per_gpu:4363.872842724959 mfu:20.58513275261652 +Step 24228 | loss:0.3905044496059418 lr:2.500751093239237e-05 tokens_per_second_per_gpu:4687.329093543168 mfu:22.11093107505314 +Step 24229 | loss:0.34580063819885254 lr:2.5003755372196224e-05 tokens_per_second_per_gpu:5368.451218738418 mfu:25.323900351017034 +Step 24230 | loss:0.3362308740615845 lr:2.500000000000001e-05 tokens_per_second_per_gpu:4755.509868832383 mfu:22.432551424932175 +Step 24231 | loss:0.4473828077316284 lr:2.4996244815831987e-05 tokens_per_second_per_gpu:4677.361161320579 mfu:22.063910638054583 +Step 24232 | loss:0.41666465997695923 lr:2.4992489819720412e-05 tokens_per_second_per_gpu:3526.043782380865 mfu:16.632950126595368 +Step 24233 | loss:0.3942422568798065 lr:2.4988735011693492e-05 tokens_per_second_per_gpu:5318.883418012628 mfu:25.090080577854607 +Step 24234 | loss:0.4428505003452301 lr:2.4984980391779488e-05 tokens_per_second_per_gpu:5334.744326065271 mfu:25.164899187288885 +Step 24235 | loss:0.3649596571922302 lr:2.4981225960006643e-05 tokens_per_second_per_gpu:4107.851294295558 mfu:19.377435426894554 +Step 24236 | loss:0.35551801323890686 lr:2.4977471716403204e-05 tokens_per_second_per_gpu:5362.002589697335 mfu:25.29348106758068 +Step 24237 | loss:0.41100478172302246 lr:2.4973717660997398e-05 tokens_per_second_per_gpu:4519.365013466627 mfu:21.318615851704955 +Step 24238 | loss:0.48515066504478455 lr:2.4969963793817437e-05 tokens_per_second_per_gpu:4540.17351747163 mfu:21.416773115393433 +Step 24239 | loss:0.3885447680950165 lr:2.4966210114891565e-05 tokens_per_second_per_gpu:5192.40771634945 mfu:24.49347311412954 +Step 24240 | loss:0.460694283246994 lr:2.4962456624248042e-05 tokens_per_second_per_gpu:4238.690564208816 mfu:19.994626586556766 +Step 24241 | loss:0.4401492178440094 lr:2.495870332191506e-05 tokens_per_second_per_gpu:4274.270143551828 mfu:20.16246153281996 +Step 24242 | loss:0.43234890699386597 lr:2.495495020792087e-05 tokens_per_second_per_gpu:4305.4826608273515 mfu:20.30969630221319 +Step 24243 | loss:0.5919060111045837 lr:2.495119728229371e-05 tokens_per_second_per_gpu:5230.691711396974 mfu:24.674065250691815 +Step 24244 | loss:0.324320524930954 lr:2.4947444545061786e-05 tokens_per_second_per_gpu:4513.693136984413 mfu:21.291860642616438 +Step 24245 | loss:0.43936893343925476 lr:2.4943691996253317e-05 tokens_per_second_per_gpu:4337.272021556863 mfu:20.459652140598525 +Step 24246 | loss:0.5168201327323914 lr:2.4939939635896536e-05 tokens_per_second_per_gpu:5326.92164843075 mfu:25.12799828220042 +Step 24247 | loss:0.42873409390449524 lr:2.4936187464019683e-05 tokens_per_second_per_gpu:4509.889449664936 mfu:21.273918000554488 +Step 24248 | loss:0.38846108317375183 lr:2.4932435480650944e-05 tokens_per_second_per_gpu:5371.368699904595 mfu:25.337662607451506 +Step 24249 | loss:0.38477903604507446 lr:2.492868368581856e-05 tokens_per_second_per_gpu:3910.5371649337035 mfu:18.446671013432763 +Step 24250 | loss:0.4624699354171753 lr:2.4924932079550763e-05 tokens_per_second_per_gpu:4145.03705052141 mfu:19.552846983556137 +Step 24251 | loss:0.433580219745636 lr:2.492118066187575e-05 tokens_per_second_per_gpu:5346.356588513807 mfu:25.21967621051491 +Step 24252 | loss:0.43587517738342285 lr:2.4917429432821716e-05 tokens_per_second_per_gpu:3444.3825408796515 mfu:16.247740117589633 +Step 24253 | loss:0.3817504942417145 lr:2.49136783924169e-05 tokens_per_second_per_gpu:5350.387766337092 mfu:25.2386919640972 +Step 24254 | loss:0.45199054479599 lr:2.4909927540689527e-05 tokens_per_second_per_gpu:4583.355304121428 mfu:21.620468970593052 +Step 24255 | loss:0.4064512848854065 lr:2.490617687766777e-05 tokens_per_second_per_gpu:5040.849037201522 mfu:23.778544966011648 +Step 24256 | loss:0.6318190097808838 lr:2.490242640337986e-05 tokens_per_second_per_gpu:4691.6652491127 mfu:22.13138546069905 +Step 24257 | loss:0.45211413502693176 lr:2.4898676117854027e-05 tokens_per_second_per_gpu:4948.724655754449 mfu:23.343978540685537 +Step 24258 | loss:0.4395516812801361 lr:2.4894926021118448e-05 tokens_per_second_per_gpu:5324.172482178913 mfu:25.115029995934375 +Step 24259 | loss:0.38318300247192383 lr:2.4891176113201316e-05 tokens_per_second_per_gpu:4308.982201924046 mfu:20.326204234649577 +Step 24260 | loss:0.40093010663986206 lr:2.4887426394130854e-05 tokens_per_second_per_gpu:5233.932492759369 mfu:24.68935256931255 +Step 24261 | loss:0.35782116651535034 lr:2.4883676863935278e-05 tokens_per_second_per_gpu:5341.739054777422 mfu:25.197894516047885 +Step 24262 | loss:0.37029504776000977 lr:2.4879927522642753e-05 tokens_per_second_per_gpu:3896.195785299743 mfu:18.379020278807694 +Step 24263 | loss:0.47884002327919006 lr:2.4876178370281504e-05 tokens_per_second_per_gpu:3827.336816035813 mfu:18.054200772238108 +Step 24264 | loss:0.4990224540233612 lr:2.4872429406879742e-05 tokens_per_second_per_gpu:5361.091899744346 mfu:25.2891851876927 +Step 24265 | loss:0.3443748950958252 lr:2.486868063246564e-05 tokens_per_second_per_gpu:4929.333650258931 mfu:23.252507859316484 +Step 24266 | loss:0.44036176800727844 lr:2.4864932047067385e-05 tokens_per_second_per_gpu:5279.334753909183 mfu:24.90352278158076 +Step 24267 | loss:0.4649125337600708 lr:2.4861183650713175e-05 tokens_per_second_per_gpu:5166.158893296219 mfu:24.369652937276424 +Step 24268 | loss:0.41272884607315063 lr:2.4857435443431237e-05 tokens_per_second_per_gpu:4291.201201692545 mfu:20.242328222806115 +Step 24269 | loss:0.45407700538635254 lr:2.4853687425249717e-05 tokens_per_second_per_gpu:4654.181569876897 mfu:21.954568550368624 +Step 24270 | loss:0.43141186237335205 lr:2.484993959619682e-05 tokens_per_second_per_gpu:4806.135221386577 mfu:22.67135985050613 +Step 24271 | loss:0.462465763092041 lr:2.4846191956300734e-05 tokens_per_second_per_gpu:3788.5381586520502 mfu:17.871180885625343 +Step 24272 | loss:0.25986313819885254 lr:2.4842444505589668e-05 tokens_per_second_per_gpu:4179.041622969152 mfu:19.713252353570454 +Step 24273 | loss:0.44636270403862 lr:2.4838697244091787e-05 tokens_per_second_per_gpu:4547.495157807983 mfu:21.451310542060664 +Step 24274 | loss:0.51188063621521 lr:2.483495017183526e-05 tokens_per_second_per_gpu:5039.337898660841 mfu:23.771416667688438 +Step 24275 | loss:0.47470030188560486 lr:2.4831203288848276e-05 tokens_per_second_per_gpu:4470.200603489304 mfu:21.086698941528603 +Step 24276 | loss:0.39233919978141785 lr:2.482745659515905e-05 tokens_per_second_per_gpu:3714.036426280697 mfu:17.51974350272338 +Step 24277 | loss:0.446390300989151 lr:2.482371009079571e-05 tokens_per_second_per_gpu:4457.873149751561 mfu:21.02854823896718 +Step 24278 | loss:0.448923259973526 lr:2.4819963775786463e-05 tokens_per_second_per_gpu:4598.389135654773 mfu:21.691386118972 +Step 24279 | loss:0.42073649168014526 lr:2.48162176501595e-05 tokens_per_second_per_gpu:5318.003281396417 mfu:25.08592882326915 +Step 24280 | loss:0.40534746646881104 lr:2.4812471713942974e-05 tokens_per_second_per_gpu:4966.986078217227 mfu:23.43012078616242 +Step 24281 | loss:0.4592689573764801 lr:2.480872596716505e-05 tokens_per_second_per_gpu:5328.573447163629 mfu:25.135790098649334 +Step 24282 | loss:0.4822497069835663 lr:2.4804980409853906e-05 tokens_per_second_per_gpu:4863.651955294309 mfu:22.942676097715633 +Step 24283 | loss:0.3597438931465149 lr:2.4801235042037742e-05 tokens_per_second_per_gpu:3779.7316675833895 mfu:17.829639164712155 +Step 24284 | loss:0.5451978445053101 lr:2.479748986374468e-05 tokens_per_second_per_gpu:4334.36003041062 mfu:20.44591578152477 +Step 24285 | loss:0.5055687427520752 lr:2.4793744875002917e-05 tokens_per_second_per_gpu:4368.277737302344 mfu:20.60591139189056 +Step 24286 | loss:0.37015271186828613 lr:2.4790000075840635e-05 tokens_per_second_per_gpu:5191.3420725122805 mfu:24.488446290332025 +Step 24287 | loss:0.48155418038368225 lr:2.478625546628598e-05 tokens_per_second_per_gpu:4412.318665415745 mfu:20.813659963958646 +Step 24288 | loss:0.368437260389328 lr:2.4782511046367097e-05 tokens_per_second_per_gpu:5324.259819148188 mfu:25.11544197931967 +Step 24289 | loss:0.30479755997657776 lr:2.477876681611216e-05 tokens_per_second_per_gpu:4935.99854816602 mfu:23.283947319893063 +Step 24290 | loss:0.4652342200279236 lr:2.4775022775549356e-05 tokens_per_second_per_gpu:3356.252838870719 mfu:15.832017276735824 +Step 24291 | loss:0.33258941769599915 lr:2.4771278924706808e-05 tokens_per_second_per_gpu:4891.951418530841 mfu:23.07616949418867 +Step 24292 | loss:0.3666108250617981 lr:2.476753526361269e-05 tokens_per_second_per_gpu:3612.2993856599087 mfu:17.039832523986078 +Step 24293 | loss:0.4150446057319641 lr:2.476379179229518e-05 tokens_per_second_per_gpu:3924.106349171284 mfu:18.510679170622993 +Step 24294 | loss:0.4056876003742218 lr:2.4760048510782414e-05 tokens_per_second_per_gpu:5148.5170864872225 mfu:24.286433524555765 +Step 24295 | loss:0.41307714581489563 lr:2.4756305419102516e-05 tokens_per_second_per_gpu:5346.980931625437 mfu:25.222621343496385 +Step 24296 | loss:0.3901202976703644 lr:2.475256251728367e-05 tokens_per_second_per_gpu:4284.683022376499 mfu:20.21158085885626 +Step 24297 | loss:0.4686720073223114 lr:2.4748819805354045e-05 tokens_per_second_per_gpu:5259.619621205293 mfu:24.810523136868014 +Step 24298 | loss:0.5009089112281799 lr:2.474507728334175e-05 tokens_per_second_per_gpu:5341.827647179176 mfu:25.19831242152238 +Step 24299 | loss:0.35822606086730957 lr:2.4741334951274947e-05 tokens_per_second_per_gpu:5075.685932187571 mfu:23.94287654344803 +Step 24300 | loss:0.48362404108047485 lr:2.473759280918181e-05 tokens_per_second_per_gpu:3325.2788070359456 mfu:15.685907483856425 +Step 24301 | loss:0.34377601742744446 lr:2.4733850857090434e-05 tokens_per_second_per_gpu:3962.709842562544 mfu:18.69277843538471 +Step 24302 | loss:0.5033837556838989 lr:2.4730109095029013e-05 tokens_per_second_per_gpu:4218.335614718134 mfu:19.898608816896715 +Step 24303 | loss:0.5309258699417114 lr:2.472636752302565e-05 tokens_per_second_per_gpu:4666.95063267076 mfu:22.01480239819404 +Step 24304 | loss:0.3194076120853424 lr:2.4722626141108517e-05 tokens_per_second_per_gpu:3424.9537141213664 mfu:16.156091026871138 +Step 24305 | loss:0.436698853969574 lr:2.4718884949305726e-05 tokens_per_second_per_gpu:5350.396621773401 mfu:25.238733736701164 +Step 24306 | loss:0.44734230637550354 lr:2.471514394764542e-05 tokens_per_second_per_gpu:4908.5263139172 mfu:23.154356103694887 +Step 24307 | loss:0.4364187717437744 lr:2.471140313615575e-05 tokens_per_second_per_gpu:4574.466257235643 mfu:21.578537819805014 +Step 24308 | loss:0.3190475404262543 lr:2.4707662514864854e-05 tokens_per_second_per_gpu:3934.2872660631524 mfu:18.558704292645476 +Step 24309 | loss:0.42156627774238586 lr:2.4703922083800858e-05 tokens_per_second_per_gpu:4157.80704877577 mfu:19.613085244108326 +Step 24310 | loss:0.5448899865150452 lr:2.4700181842991878e-05 tokens_per_second_per_gpu:5362.78952863621 mfu:25.297193192820977 +Step 24311 | loss:0.2513193190097809 lr:2.469644179246607e-05 tokens_per_second_per_gpu:5368.21250628097 mfu:25.32277430362671 +Step 24312 | loss:0.4898759722709656 lr:2.4692701932251537e-05 tokens_per_second_per_gpu:4171.691588919869 mfu:19.678580988914867 +Step 24313 | loss:0.4624917209148407 lr:2.4688962262376424e-05 tokens_per_second_per_gpu:4244.988268543083 mfu:20.02433393240066 +Step 24314 | loss:0.42826536297798157 lr:2.468522278286885e-05 tokens_per_second_per_gpu:3319.1687901853147 mfu:15.657085491895629 +Step 24315 | loss:0.37637144327163696 lr:2.468148349375697e-05 tokens_per_second_per_gpu:5210.524353432581 mfu:24.578932382267688 +Step 24316 | loss:0.4478161334991455 lr:2.4677744395068882e-05 tokens_per_second_per_gpu:3652.932849059656 mfu:17.231507503626425 +Step 24317 | loss:0.4246649444103241 lr:2.467400548683269e-05 tokens_per_second_per_gpu:4251.392053093713 mfu:20.05454167672429 +Step 24318 | loss:0.4680319130420685 lr:2.4670266769076534e-05 tokens_per_second_per_gpu:5355.879563910716 mfu:25.26459771024927 +Step 24319 | loss:0.44756925106048584 lr:2.4666528241828552e-05 tokens_per_second_per_gpu:5204.583843954883 mfu:24.550909985505957 +Step 24320 | loss:0.3092985451221466 lr:2.466278990511683e-05 tokens_per_second_per_gpu:5251.844721313757 mfu:24.77384760754493 +Step 24321 | loss:0.3728506863117218 lr:2.465905175896949e-05 tokens_per_second_per_gpu:3931.1429043508406 mfu:18.54387180196507 +Step 24322 | loss:0.4103013277053833 lr:2.4655313803414677e-05 tokens_per_second_per_gpu:3786.248547799474 mfu:17.860380400585846 +Step 24323 | loss:0.44026902318000793 lr:2.4651576038480484e-05 tokens_per_second_per_gpu:5353.457728644965 mfu:25.25317350009302 +Step 24324 | loss:0.420404851436615 lr:2.4647838464195006e-05 tokens_per_second_per_gpu:5333.409696940642 mfu:25.158603513996805 +Step 24325 | loss:0.4932264983654022 lr:2.4644101080586362e-05 tokens_per_second_per_gpu:5322.731075607927 mfu:25.10823063520971 +Step 24326 | loss:0.35846108198165894 lr:2.4640363887682693e-05 tokens_per_second_per_gpu:5260.537580943616 mfu:24.814853309573767 +Step 24327 | loss:0.3118525743484497 lr:2.4636626885512067e-05 tokens_per_second_per_gpu:3637.4688424041674 mfu:17.158561145800885 +Step 24328 | loss:0.41261574625968933 lr:2.4632890074102605e-05 tokens_per_second_per_gpu:5004.942128674819 mfu:23.60916595214087 +Step 24329 | loss:0.4177658259868622 lr:2.462915345348243e-05 tokens_per_second_per_gpu:3730.1635497521174 mfu:17.595817895709065 +Step 24330 | loss:0.48741647601127625 lr:2.4625417023679615e-05 tokens_per_second_per_gpu:4438.406279404476 mfu:20.936719690149406 +Step 24331 | loss:0.3962368667125702 lr:2.462168078472229e-05 tokens_per_second_per_gpu:5328.932880588284 mfu:25.1374856074391 +Step 24332 | loss:0.5172480344772339 lr:2.4617944736638527e-05 tokens_per_second_per_gpu:5228.869989474417 mfu:24.665471877565302 +Step 24333 | loss:0.4628896713256836 lr:2.4614208879456458e-05 tokens_per_second_per_gpu:5331.36310853008 mfu:25.148949407280195 +Step 24334 | loss:0.3840100169181824 lr:2.461047321320414e-05 tokens_per_second_per_gpu:4210.564068626269 mfu:19.861949108018894 +Step 24335 | loss:0.4089663028717041 lr:2.4606737737909697e-05 tokens_per_second_per_gpu:5364.440897557598 mfu:25.30498298177576 +Step 24336 | loss:0.42368119955062866 lr:2.4603002453601238e-05 tokens_per_second_per_gpu:5366.242834555316 mfu:25.31348302603615 +Step 24337 | loss:0.36390623450279236 lr:2.4599267360306823e-05 tokens_per_second_per_gpu:4564.964920793831 mfu:21.533718394713826 +Step 24338 | loss:0.4880019426345825 lr:2.459553245805457e-05 tokens_per_second_per_gpu:5318.666083309818 mfu:25.089055372980326 +Step 24339 | loss:0.3714744448661804 lr:2.4591797746872547e-05 tokens_per_second_per_gpu:5242.237666007091 mfu:24.72852948853073 +Step 24340 | loss:0.3078228235244751 lr:2.4588063226788872e-05 tokens_per_second_per_gpu:3772.6072784885437 mfu:17.796032205805645 +Step 24341 | loss:0.37985000014305115 lr:2.4584328897831598e-05 tokens_per_second_per_gpu:5011.415811791137 mfu:23.63970341992477 +Step 24342 | loss:0.39955848455429077 lr:2.4580594760028823e-05 tokens_per_second_per_gpu:4519.430859962473 mfu:21.318926460860432 +Step 24343 | loss:0.4404605031013489 lr:2.457686081340865e-05 tokens_per_second_per_gpu:5351.514677862845 mfu:25.24400779803511 +Step 24344 | loss:0.47380441427230835 lr:2.4573127057999162e-05 tokens_per_second_per_gpu:5330.88892936319 mfu:25.146712623246664 +Step 24345 | loss:0.3769066631793976 lr:2.456939349382843e-05 tokens_per_second_per_gpu:5095.956710548726 mfu:24.03849726352902 +Step 24346 | loss:0.44924482703208923 lr:2.4565660120924516e-05 tokens_per_second_per_gpu:4114.72698414804 mfu:19.40986922904237 +Step 24347 | loss:0.4984261393547058 lr:2.4561926939315537e-05 tokens_per_second_per_gpu:4751.381839473657 mfu:22.413078806132628 +Step 24348 | loss:0.4682317078113556 lr:2.4558193949029533e-05 tokens_per_second_per_gpu:5353.0649038060565 mfu:25.25132047830506 +Step 24349 | loss:0.4888746440410614 lr:2.4554461150094604e-05 tokens_per_second_per_gpu:4813.275105297012 mfu:22.70503990110146 +Step 24350 | loss:0.31424421072006226 lr:2.4550728542538814e-05 tokens_per_second_per_gpu:5339.927019291087 mfu:25.189346835493712 +Step 24351 | loss:0.48568302392959595 lr:2.4546996126390258e-05 tokens_per_second_per_gpu:4206.124036855443 mfu:19.841004720608375 +Step 24352 | loss:0.3953072726726532 lr:2.4543263901677e-05 tokens_per_second_per_gpu:3403.1981220081525 mfu:16.053466186986334 +Step 24353 | loss:0.5140792727470398 lr:2.4539531868427074e-05 tokens_per_second_per_gpu:3572.4178858437413 mfu:16.851704684867688 +Step 24354 | loss:0.4705512225627899 lr:2.453580002666858e-05 tokens_per_second_per_gpu:4947.59314088046 mfu:23.338640991968585 +Step 24355 | loss:0.3905889093875885 lr:2.4532068376429608e-05 tokens_per_second_per_gpu:3382.8886333644555 mfu:15.957662863897937 +Step 24356 | loss:0.43900713324546814 lr:2.4528336917738178e-05 tokens_per_second_per_gpu:3391.4836995945375 mfu:15.99820725777475 +Step 24357 | loss:0.426357239484787 lr:2.4524605650622374e-05 tokens_per_second_per_gpu:5371.676246553192 mfu:25.33911335746581 +Step 24358 | loss:0.3188798427581787 lr:2.4520874575110282e-05 tokens_per_second_per_gpu:4336.279916877923 mfu:20.454972213557447 +Step 24359 | loss:0.49467360973358154 lr:2.4517143691229945e-05 tokens_per_second_per_gpu:5190.386091424294 mfu:24.4839367644328 +Step 24360 | loss:0.4669210910797119 lr:2.45134129990094e-05 tokens_per_second_per_gpu:3645.4012169852667 mfu:17.195979510102593 +Step 24361 | loss:0.48578736186027527 lr:2.4509682498476732e-05 tokens_per_second_per_gpu:5291.024683487946 mfu:24.958666173911443 +Step 24362 | loss:0.5325605869293213 lr:2.450595218966001e-05 tokens_per_second_per_gpu:5353.733284677566 mfu:25.254473344913503 +Step 24363 | loss:0.41162797808647156 lr:2.4502222072587255e-05 tokens_per_second_per_gpu:4529.523850603456 mfu:21.36653681975572 +Step 24364 | loss:0.319792777299881 lr:2.4498492147286545e-05 tokens_per_second_per_gpu:4872.370418114971 mfu:22.983802574363835 +Step 24365 | loss:0.4054126441478729 lr:2.4494762413785943e-05 tokens_per_second_per_gpu:3640.5234376198723 mfu:17.17297019260105 +Step 24366 | loss:0.3912583589553833 lr:2.449103287211347e-05 tokens_per_second_per_gpu:4559.102091315717 mfu:21.50606242776371 +Step 24367 | loss:0.4580991268157959 lr:2.4487303522297215e-05 tokens_per_second_per_gpu:5038.605533836441 mfu:23.76796197785742 +Step 24368 | loss:0.3934750556945801 lr:2.448357436436519e-05 tokens_per_second_per_gpu:5359.618335187892 mfu:25.282234132263866 +Step 24369 | loss:0.41096195578575134 lr:2.4479845398345475e-05 tokens_per_second_per_gpu:3879.6043079783763 mfu:18.300755449484875 +Step 24370 | loss:0.40228432416915894 lr:2.4476116624266075e-05 tokens_per_second_per_gpu:5365.654181647741 mfu:25.310706249836574 +Step 24371 | loss:0.5112428069114685 lr:2.4472388042155065e-05 tokens_per_second_per_gpu:5344.194858813088 mfu:25.209478962687495 +Step 24372 | loss:0.3695223033428192 lr:2.4468659652040497e-05 tokens_per_second_per_gpu:5033.346087899482 mfu:23.74315227401877 +Step 24373 | loss:0.4412939250469208 lr:2.446493145395038e-05 tokens_per_second_per_gpu:4631.571879995941 mfu:21.847914785589275 +Step 24374 | loss:0.4563002288341522 lr:2.4461203447912788e-05 tokens_per_second_per_gpu:5344.880488595612 mfu:25.212713195352165 +Step 24375 | loss:0.4492243528366089 lr:2.445747563395573e-05 tokens_per_second_per_gpu:4658.647385846112 mfu:21.975634566242185 +Step 24376 | loss:0.3772067725658417 lr:2.4453748012107276e-05 tokens_per_second_per_gpu:4203.965833757936 mfu:19.830824108370017 +Step 24377 | loss:0.5913859009742737 lr:2.4450020582395415e-05 tokens_per_second_per_gpu:4770.274166650985 mfu:22.502197136791484 +Step 24378 | loss:0.37379947304725647 lr:2.444629334484822e-05 tokens_per_second_per_gpu:4892.518531146143 mfu:23.078844661154434 +Step 24379 | loss:0.41237184405326843 lr:2.4442566299493726e-05 tokens_per_second_per_gpu:4880.496486519449 mfu:23.022134625477992 +Step 24380 | loss:0.3680429458618164 lr:2.4438839446359933e-05 tokens_per_second_per_gpu:4005.119162839371 mfu:18.892830434906674 +Step 24381 | loss:0.3686813712120056 lr:2.443511278547491e-05 tokens_per_second_per_gpu:4777.196016107132 mfu:22.534848681665423 +Step 24382 | loss:0.39313650131225586 lr:2.443138631686665e-05 tokens_per_second_per_gpu:5008.368970177607 mfu:23.625330948188967 +Step 24383 | loss:0.3532631993293762 lr:2.442766004056321e-05 tokens_per_second_per_gpu:5224.0737960729875 mfu:24.642847414975787 +Step 24384 | loss:0.4695638120174408 lr:2.442393395659259e-05 tokens_per_second_per_gpu:4971.138705771869 mfu:23.449709438848984 +Step 24385 | loss:0.40935689210891724 lr:2.4420208064982818e-05 tokens_per_second_per_gpu:5340.4248687127865 mfu:25.191695276157734 +Step 24386 | loss:0.454402357339859 lr:2.4416482365761935e-05 tokens_per_second_per_gpu:5330.812292767508 mfu:25.14635111534928 +Step 24387 | loss:0.4769890308380127 lr:2.4412756858957962e-05 tokens_per_second_per_gpu:5115.185062861075 mfu:24.12920068208157 +Step 24388 | loss:0.4174441993236542 lr:2.4409031544598916e-05 tokens_per_second_per_gpu:3692.8617047923412 mfu:17.41985872868278 +Step 24389 | loss:0.4292861819267273 lr:2.4405306422712786e-05 tokens_per_second_per_gpu:4485.695086363754 mfu:21.159789060878765 +Step 24390 | loss:0.3972436189651489 lr:2.4401581493327613e-05 tokens_per_second_per_gpu:4960.5160084693225 mfu:23.39960036325382 +Step 24391 | loss:0.3850468695163727 lr:2.4397856756471432e-05 tokens_per_second_per_gpu:5363.777410835702 mfu:25.301853201706066 +Step 24392 | loss:0.3115430176258087 lr:2.439413221217222e-05 tokens_per_second_per_gpu:4849.925155327149 mfu:22.87792444023696 +Step 24393 | loss:0.47940686345100403 lr:2.4390407860458005e-05 tokens_per_second_per_gpu:4946.613282451792 mfu:23.33401883257947 +Step 24394 | loss:0.5218608379364014 lr:2.4386683701356818e-05 tokens_per_second_per_gpu:5264.308689622662 mfu:24.832642272630313 +Step 24395 | loss:0.40341585874557495 lr:2.4382959734896632e-05 tokens_per_second_per_gpu:3222.2939168767753 mfu:15.200110185941488 +Step 24396 | loss:0.3829803764820099 lr:2.4379235961105494e-05 tokens_per_second_per_gpu:4293.385157485717 mfu:20.25263031490379 +Step 24397 | loss:0.40271279215812683 lr:2.4375512380011378e-05 tokens_per_second_per_gpu:4107.365811485355 mfu:19.37514532164692 +Step 24398 | loss:0.4207216501235962 lr:2.437178899164232e-05 tokens_per_second_per_gpu:5266.166338826666 mfu:24.841405120873237 +Step 24399 | loss:0.4127140939235687 lr:2.4368065796026286e-05 tokens_per_second_per_gpu:4594.253703138309 mfu:21.671878578214294 +Step 24400 | loss:0.38812389969825745 lr:2.4364342793191296e-05 tokens_per_second_per_gpu:5354.030262081511 mfu:25.255874237997485 +Step 24401 | loss:0.4281138777732849 lr:2.436061998316538e-05 tokens_per_second_per_gpu:4516.357162868361 mfu:21.304427307240704 +Step 24402 | loss:0.45184701681137085 lr:2.4356897365976488e-05 tokens_per_second_per_gpu:3588.3362014173663 mfu:16.926794095373175 +Step 24403 | loss:0.3385758697986603 lr:2.4353174941652668e-05 tokens_per_second_per_gpu:5357.916869560166 mfu:25.27420802859791 +Step 24404 | loss:0.37794438004493713 lr:2.434945271022187e-05 tokens_per_second_per_gpu:5369.275289498803 mfu:25.327787633394742 +Step 24405 | loss:0.35114040970802307 lr:2.434573067171213e-05 tokens_per_second_per_gpu:4295.129792889344 mfu:20.26086005776819 +Step 24406 | loss:0.4607301950454712 lr:2.4342008826151403e-05 tokens_per_second_per_gpu:5344.662323218266 mfu:25.21168407204437 +Step 24407 | loss:0.5275799036026001 lr:2.4338287173567705e-05 tokens_per_second_per_gpu:3551.0168662851806 mfu:16.750752424219503 +Step 24408 | loss:0.374324768781662 lr:2.433456571398904e-05 tokens_per_second_per_gpu:4918.494037677481 mfu:23.201375557341173 +Step 24409 | loss:0.3978591561317444 lr:2.4330844447443363e-05 tokens_per_second_per_gpu:4205.804417159955 mfu:19.839497019972015 +Step 24410 | loss:0.38585150241851807 lr:2.43271233739587e-05 tokens_per_second_per_gpu:4498.683275137886 mfu:21.22105656333986 +Step 24411 | loss:0.4361107349395752 lr:2.4323402493562995e-05 tokens_per_second_per_gpu:5148.532395359929 mfu:24.286505739120287 +Step 24412 | loss:0.4892967939376831 lr:2.4319681806284284e-05 tokens_per_second_per_gpu:4448.796087050933 mfu:20.985730185501726 +Step 24413 | loss:0.5244255065917969 lr:2.43159613121505e-05 tokens_per_second_per_gpu:5361.852438474325 mfu:25.2927727786433 +Step 24414 | loss:0.5104116201400757 lr:2.431224101118964e-05 tokens_per_second_per_gpu:4653.8947281110895 mfu:21.953215468819057 +Step 24415 | loss:0.3786335289478302 lr:2.430852090342972e-05 tokens_per_second_per_gpu:4309.787929215244 mfu:20.33000498775341 +Step 24416 | loss:0.4943256378173828 lr:2.4304800988898675e-05 tokens_per_second_per_gpu:5352.667561804438 mfu:25.249446148289888 +Step 24417 | loss:0.45253124833106995 lr:2.4301081267624513e-05 tokens_per_second_per_gpu:3564.8691786011955 mfu:16.81609614486234 +Step 24418 | loss:0.5312596559524536 lr:2.4297361739635182e-05 tokens_per_second_per_gpu:4157.932392498004 mfu:19.61367651183195 +Step 24419 | loss:0.4322625696659088 lr:2.4293642404958695e-05 tokens_per_second_per_gpu:5334.855135133215 mfu:25.165421892568357 +Step 24420 | loss:0.5624969005584717 lr:2.428992326362298e-05 tokens_per_second_per_gpu:3854.158247742601 mfu:18.180722041807158 +Step 24421 | loss:0.444265216588974 lr:2.4286204315656042e-05 tokens_per_second_per_gpu:5326.845696842281 mfu:25.12764000559893 +Step 24422 | loss:0.5353496670722961 lr:2.428248556108583e-05 tokens_per_second_per_gpu:3790.7070113028662 mfu:17.88141173362366 +Step 24423 | loss:0.38136187195777893 lr:2.4278766999940357e-05 tokens_per_second_per_gpu:4308.433538551286 mfu:20.323616095908644 +Step 24424 | loss:0.5067219138145447 lr:2.4275048632247532e-05 tokens_per_second_per_gpu:3688.144360456214 mfu:17.39760621058733 +Step 24425 | loss:0.42947056889533997 lr:2.4271330458035368e-05 tokens_per_second_per_gpu:4539.255361200482 mfu:21.412442015608818 +Step 24426 | loss:0.37278738617897034 lr:2.426761247733179e-05 tokens_per_second_per_gpu:3125.536315626613 mfu:14.743688072295393 +Step 24427 | loss:0.3788672983646393 lr:2.4263894690164807e-05 tokens_per_second_per_gpu:4178.59645913046 mfu:19.711152439790563 +Step 24428 | loss:0.38834348320961 lr:2.4260177096562338e-05 tokens_per_second_per_gpu:5094.682431402491 mfu:24.03248627138179 +Step 24429 | loss:0.37989088892936707 lr:2.4256459696552352e-05 tokens_per_second_per_gpu:4425.840017005186 mfu:20.877442486386506 +Step 24430 | loss:0.43826839327812195 lr:2.4252742490162843e-05 tokens_per_second_per_gpu:4697.159801681848 mfu:22.15730420263506 +Step 24431 | loss:0.3356498181819916 lr:2.4249025477421728e-05 tokens_per_second_per_gpu:5337.171514869091 mfu:25.176348650997618 +Step 24432 | loss:0.3665660321712494 lr:2.424530865835699e-05 tokens_per_second_per_gpu:3221.3783163015105 mfu:15.195791141810398 +Step 24433 | loss:0.3441857397556305 lr:2.4241592032996558e-05 tokens_per_second_per_gpu:3684.0644516936723 mfu:17.378360584850203 +Step 24434 | loss:0.4254801869392395 lr:2.4237875601368414e-05 tokens_per_second_per_gpu:4822.418862072332 mfu:22.748172561896425 +Step 24435 | loss:0.49206602573394775 lr:2.4234159363500476e-05 tokens_per_second_per_gpu:5333.502744557471 mfu:25.159042435480945 +Step 24436 | loss:0.5609738230705261 lr:2.4230443319420714e-05 tokens_per_second_per_gpu:5349.573882944688 mfu:25.23485273727241 +Step 24437 | loss:0.3563592731952667 lr:2.4226727469157096e-05 tokens_per_second_per_gpu:3442.2119742199984 mfu:16.23750118431919 +Step 24438 | loss:0.3934984803199768 lr:2.4223011812737527e-05 tokens_per_second_per_gpu:4279.289176523748 mfu:20.186137167683835 +Step 24439 | loss:0.4313778281211853 lr:2.4219296350190003e-05 tokens_per_second_per_gpu:4770.375203787423 mfu:22.502673746202774 +Step 24440 | loss:0.45765140652656555 lr:2.4215581081542415e-05 tokens_per_second_per_gpu:3833.8182510004945 mfu:18.084774806813346 +Step 24441 | loss:0.4113011062145233 lr:2.4211866006822754e-05 tokens_per_second_per_gpu:3576.2401343557244 mfu:16.869734883241556 +Step 24442 | loss:0.3330734074115753 lr:2.4208151126058915e-05 tokens_per_second_per_gpu:3679.738078458506 mfu:17.357952344144536 +Step 24443 | loss:0.3891533315181732 lr:2.4204436439278866e-05 tokens_per_second_per_gpu:5338.616413905973 mfu:25.18316448627979 +Step 24444 | loss:0.37385404109954834 lr:2.4200721946510564e-05 tokens_per_second_per_gpu:3662.1551857180903 mfu:17.27501083913715 +Step 24445 | loss:0.3317834138870239 lr:2.41970076477819e-05 tokens_per_second_per_gpu:4282.286114869761 mfu:20.200274237192385 +Step 24446 | loss:0.3726899027824402 lr:2.4193293543120855e-05 tokens_per_second_per_gpu:4434.783462931642 mfu:20.919630246730822 +Step 24447 | loss:0.43335840106010437 lr:2.418957963255532e-05 tokens_per_second_per_gpu:4598.238708631532 mfu:21.690676529037873 +Step 24448 | loss:0.4192121624946594 lr:2.4185865916113278e-05 tokens_per_second_per_gpu:3844.2947089423246 mfu:18.134194046393002 +Step 24449 | loss:0.40670499205589294 lr:2.4182152393822604e-05 tokens_per_second_per_gpu:5338.581086580507 mfu:25.182997841258103 +Step 24450 | loss:0.3505796194076538 lr:2.4178439065711263e-05 tokens_per_second_per_gpu:4833.903441257274 mfu:22.802347281383057 +Step 24451 | loss:0.47644904255867004 lr:2.4174725931807192e-05 tokens_per_second_per_gpu:4953.80888708735 mfu:23.36796172734215 +Step 24452 | loss:0.4165000021457672 lr:2.4171012992138282e-05 tokens_per_second_per_gpu:4015.8243837141763 mfu:18.943328788272154 +Step 24453 | loss:0.4168119728565216 lr:2.4167300246732483e-05 tokens_per_second_per_gpu:3696.5363215311027 mfu:17.437192522793783 +Step 24454 | loss:0.4591580331325531 lr:2.416358769561773e-05 tokens_per_second_per_gpu:5323.035245582881 mfu:25.109665456878318 +Step 24455 | loss:0.4162956178188324 lr:2.4159875338821934e-05 tokens_per_second_per_gpu:5324.557240739452 mfu:25.116844967711017 +Step 24456 | loss:0.38812562823295593 lr:2.415616317637299e-05 tokens_per_second_per_gpu:3606.5474000047593 mfu:17.012699426260788 +Step 24457 | loss:0.31229084730148315 lr:2.4152451208298845e-05 tokens_per_second_per_gpu:5063.147529487049 mfu:23.883730758637572 +Step 24458 | loss:0.39255276322364807 lr:2.414873943462741e-05 tokens_per_second_per_gpu:5356.973776984456 mfu:25.269759300010104 +Step 24459 | loss:0.4740675091743469 lr:2.4145027855386627e-05 tokens_per_second_per_gpu:4877.931045767377 mfu:23.01003300374115 +Step 24460 | loss:0.483163446187973 lr:2.4141316470604362e-05 tokens_per_second_per_gpu:5349.063718741743 mfu:25.232446205684127 +Step 24461 | loss:0.4879288971424103 lr:2.4137605280308584e-05 tokens_per_second_per_gpu:3421.590847927454 mfu:16.140227813270315 +Step 24462 | loss:0.3414095342159271 lr:2.4133894284527152e-05 tokens_per_second_per_gpu:4985.537596424982 mfu:23.51763146276379 +Step 24463 | loss:0.3334468901157379 lr:2.4130183483288023e-05 tokens_per_second_per_gpu:4590.017673310368 mfu:21.651896502774683 +Step 24464 | loss:0.4500499367713928 lr:2.412647287661906e-05 tokens_per_second_per_gpu:3980.7275527257734 mfu:18.777771048363014 +Step 24465 | loss:0.40556222200393677 lr:2.4122762464548205e-05 tokens_per_second_per_gpu:5351.922168035631 mfu:25.245929998705137 +Step 24466 | loss:0.3997369110584259 lr:2.411905224710337e-05 tokens_per_second_per_gpu:3074.8613507863947 mfu:14.504645617103531 +Step 24467 | loss:0.38017523288726807 lr:2.4115342224312425e-05 tokens_per_second_per_gpu:5338.927974307156 mfu:25.184634169849854 +Step 24468 | loss:0.47431138157844543 lr:2.4111632396203316e-05 tokens_per_second_per_gpu:3509.4068344779976 mfu:16.554470804781527 +Step 24469 | loss:0.5362000465393066 lr:2.4107922762803903e-05 tokens_per_second_per_gpu:4673.482441096957 mfu:22.045614052981843 +Step 24470 | loss:0.5843170285224915 lr:2.4104213324142122e-05 tokens_per_second_per_gpu:4050.235704463952 mfu:19.105652859425582 +Step 24471 | loss:0.3582412004470825 lr:2.4100504080245843e-05 tokens_per_second_per_gpu:3639.5030740832235 mfu:17.168156963706707 +Step 24472 | loss:0.2941794693470001 lr:2.4096795031142978e-05 tokens_per_second_per_gpu:4271.702949738956 mfu:20.15035164159621 +Step 24473 | loss:0.4188528060913086 lr:2.409308617686144e-05 tokens_per_second_per_gpu:4642.806555962764 mfu:21.900910668958154 +Step 24474 | loss:0.3311830759048462 lr:2.4089377517429084e-05 tokens_per_second_per_gpu:3658.506884197987 mfu:17.257801178402396 +Step 24475 | loss:0.44884446263313293 lr:2.408566905287385e-05 tokens_per_second_per_gpu:4031.9222319833257 mfu:19.019265085133863 +Step 24476 | loss:0.47011834383010864 lr:2.4081960783223585e-05 tokens_per_second_per_gpu:4757.911790883376 mfu:22.443881701056867 +Step 24477 | loss:0.42426079511642456 lr:2.4078252708506223e-05 tokens_per_second_per_gpu:5350.2791342264145 mfu:25.238179527896587 +Step 24478 | loss:0.36683088541030884 lr:2.40745448287496e-05 tokens_per_second_per_gpu:5346.18763403404 mfu:25.218879223407235 +Step 24479 | loss:0.39782679080963135 lr:2.4070837143981647e-05 tokens_per_second_per_gpu:5336.240783471957 mfu:25.17195823219819 +Step 24480 | loss:0.3742343783378601 lr:2.406712965423025e-05 tokens_per_second_per_gpu:4374.725295525612 mfu:20.636325624096582 +Step 24481 | loss:0.3648555278778076 lr:2.4063422359523257e-05 tokens_per_second_per_gpu:5108.585192812112 mfu:24.098067969006664 +Step 24482 | loss:0.38352176547050476 lr:2.4059715259888594e-05 tokens_per_second_per_gpu:5039.243040445049 mfu:23.770969205300066 +Step 24483 | loss:0.4558052122592926 lr:2.405600835535411e-05 tokens_per_second_per_gpu:5083.44044049604 mfu:23.979455882194625 +Step 24484 | loss:0.4304391145706177 lr:2.405230164594771e-05 tokens_per_second_per_gpu:5347.887899885358 mfu:25.226899667523174 +Step 24485 | loss:0.421019047498703 lr:2.4048595131697243e-05 tokens_per_second_per_gpu:4577.7414924591485 mfu:21.593987663167045 +Step 24486 | loss:0.428937703371048 lr:2.4044888812630607e-05 tokens_per_second_per_gpu:5333.613060598158 mfu:25.159562815066725 +Step 24487 | loss:0.42709580063819885 lr:2.4041182688775687e-05 tokens_per_second_per_gpu:5338.982540813707 mfu:25.18489156937126 +Step 24488 | loss:0.5946515798568726 lr:2.4037476760160326e-05 tokens_per_second_per_gpu:4374.4494549688225 mfu:20.635024437126958 +Step 24489 | loss:0.4593188464641571 lr:2.403377102681242e-05 tokens_per_second_per_gpu:4985.639996122005 mfu:23.518114499605836 +Step 24490 | loss:0.4398285448551178 lr:2.403006548875985e-05 tokens_per_second_per_gpu:3708.003263408189 mfu:17.491284044090026 +Step 24491 | loss:0.4452269673347473 lr:2.4026360146030474e-05 tokens_per_second_per_gpu:4667.541616935347 mfu:22.017590171802407 +Step 24492 | loss:0.4444042146205902 lr:2.402265499865214e-05 tokens_per_second_per_gpu:5393.399078196111 mfu:25.44158366062272 +Step 24493 | loss:0.35741356015205383 lr:2.4018950046652726e-05 tokens_per_second_per_gpu:5172.82366297487 mfu:24.401091792977585 +Step 24494 | loss:0.46340230107307434 lr:2.4015245290060114e-05 tokens_per_second_per_gpu:4381.467648150632 mfu:20.668130451792784 +Step 24495 | loss:0.4806954562664032 lr:2.4011540728902166e-05 tokens_per_second_per_gpu:5357.6708889513175 mfu:25.273047696097354 +Step 24496 | loss:0.41508182883262634 lr:2.4007836363206728e-05 tokens_per_second_per_gpu:5358.237880752682 mfu:25.27572229316317 +Step 24497 | loss:0.5404707193374634 lr:2.400413219300168e-05 tokens_per_second_per_gpu:5126.638272833501 mfu:24.183227427640627 +Step 24498 | loss:0.4743252694606781 lr:2.4000428218314853e-05 tokens_per_second_per_gpu:4252.278393113538 mfu:20.058722693823007 +Step 24499 | loss:0.4511944651603699 lr:2.3996724439174144e-05 tokens_per_second_per_gpu:4737.590003142824 mfu:22.34802036103016 +Step 24500 | loss:0.45408812165260315 lr:2.3993020855607367e-05 tokens_per_second_per_gpu:4134.417070662206 mfu:19.50275072660487 +Step 24501 | loss:0.3834141194820404 lr:2.39893174676424e-05 tokens_per_second_per_gpu:5370.117858193024 mfu:25.331762173682428 +Step 24502 | loss:0.4139357805252075 lr:2.3985614275307115e-05 tokens_per_second_per_gpu:3734.8124949507446 mfu:17.617747763402832 +Step 24503 | loss:0.37130171060562134 lr:2.3981911278629328e-05 tokens_per_second_per_gpu:5098.266029614205 mfu:24.049390715571345 +Step 24504 | loss:0.2918755114078522 lr:2.397820847763692e-05 tokens_per_second_per_gpu:3636.544841057033 mfu:17.154202473795564 +Step 24505 | loss:0.48798295855522156 lr:2.397450587235771e-05 tokens_per_second_per_gpu:5360.285769856521 mfu:25.28538253547189 +Step 24506 | loss:0.3720850646495819 lr:2.3970803462819584e-05 tokens_per_second_per_gpu:5174.295625301665 mfu:24.40803529041575 +Step 24507 | loss:0.4514721930027008 lr:2.3967101249050345e-05 tokens_per_second_per_gpu:4924.508113717864 mfu:23.22974498013084 +Step 24508 | loss:0.3419814705848694 lr:2.396339923107786e-05 tokens_per_second_per_gpu:5306.428420778514 mfu:25.031328230859636 +Step 24509 | loss:0.5295530557632446 lr:2.395969740892999e-05 tokens_per_second_per_gpu:4515.028037855963 mfu:21.298157597785114 +Step 24510 | loss:0.32903242111206055 lr:2.395599578263454e-05 tokens_per_second_per_gpu:5184.900520049193 mfu:24.45806038832146 +Step 24511 | loss:0.40685975551605225 lr:2.3952294352219384e-05 tokens_per_second_per_gpu:4773.014291961436 mfu:22.515122775394467 +Step 24512 | loss:0.5330129861831665 lr:2.3948593117712332e-05 tokens_per_second_per_gpu:5339.8629709904735 mfu:25.18904470873195 +Step 24513 | loss:0.39155712723731995 lr:2.3944892079141253e-05 tokens_per_second_per_gpu:4743.721335167103 mfu:22.3769429003018 +Step 24514 | loss:0.49290138483047485 lr:2.3941191236533943e-05 tokens_per_second_per_gpu:3927.48345470142 mfu:18.526609553602263 +Step 24515 | loss:0.3801480233669281 lr:2.3937490589918266e-05 tokens_per_second_per_gpu:4183.2510655601955 mfu:19.733109012477374 +Step 24516 | loss:0.36892271041870117 lr:2.393379013932206e-05 tokens_per_second_per_gpu:4550.301247315825 mfu:21.464547345037396 +Step 24517 | loss:0.40675145387649536 lr:2.3930089884773127e-05 tokens_per_second_per_gpu:4388.170984626812 mfu:20.699751233658063 +Step 24518 | loss:0.47355443239212036 lr:2.3926389826299317e-05 tokens_per_second_per_gpu:4006.902281998529 mfu:18.90124171221192 +Step 24519 | loss:0.42200225591659546 lr:2.3922689963928475e-05 tokens_per_second_per_gpu:3823.5894236028116 mfu:18.03652368276056 +Step 24520 | loss:0.4338149130344391 lr:2.391899029768841e-05 tokens_per_second_per_gpu:4097.86570868956 mfu:19.330331715874475 +Step 24521 | loss:0.3936258554458618 lr:2.3915290827606925e-05 tokens_per_second_per_gpu:3898.965608348479 mfu:18.392086006709125 +Step 24522 | loss:0.4535960853099823 lr:2.3911591553711863e-05 tokens_per_second_per_gpu:5086.865130265293 mfu:23.99561072814852 +Step 24523 | loss:0.42568105459213257 lr:2.3907892476031073e-05 tokens_per_second_per_gpu:3502.5111420802596 mfu:16.521942647214953 +Step 24524 | loss:0.3814719319343567 lr:2.3904193594592333e-05 tokens_per_second_per_gpu:5193.558155147939 mfu:24.49889993022711 +Step 24525 | loss:0.39941471815109253 lr:2.3900494909423486e-05 tokens_per_second_per_gpu:4490.689891064577 mfu:21.183350406854107 +Step 24526 | loss:0.39768722653388977 lr:2.3896796420552364e-05 tokens_per_second_per_gpu:5379.774849615927 mfu:25.377315849876897 +Step 24527 | loss:0.5526028275489807 lr:2.3893098128006768e-05 tokens_per_second_per_gpu:4418.872288895253 mfu:20.844574524074954 +Step 24528 | loss:0.39497798681259155 lr:2.3889400031814492e-05 tokens_per_second_per_gpu:4527.030567910568 mfu:21.354775579895474 +Step 24529 | loss:0.42086338996887207 lr:2.388570213200337e-05 tokens_per_second_per_gpu:5344.076661312892 mfu:25.2089214049125 +Step 24530 | loss:0.33050572872161865 lr:2.3882004428601213e-05 tokens_per_second_per_gpu:3852.485929935247 mfu:18.17283343338303 +Step 24531 | loss:0.4769217073917389 lr:2.3878306921635856e-05 tokens_per_second_per_gpu:4492.882060486117 mfu:21.19369124403818 +Step 24532 | loss:0.4275742173194885 lr:2.3874609611135062e-05 tokens_per_second_per_gpu:4341.467911393044 mfu:20.47944486884841 +Step 24533 | loss:0.3380485773086548 lr:2.3870912497126684e-05 tokens_per_second_per_gpu:5382.152665790419 mfu:25.388532414469246 +Step 24534 | loss:0.412539541721344 lr:2.386721557963851e-05 tokens_per_second_per_gpu:4936.654682628012 mfu:23.287042418098366 +Step 24535 | loss:0.3672790825366974 lr:2.386351885869832e-05 tokens_per_second_per_gpu:5315.180691263196 mfu:25.072614184027213 +Step 24536 | loss:0.36828142404556274 lr:2.385982233433394e-05 tokens_per_second_per_gpu:3771.5342264945084 mfu:17.790970436468232 +Step 24537 | loss:0.6387996673583984 lr:2.385612600657317e-05 tokens_per_second_per_gpu:3827.3157286888318 mfu:18.054101299624524 +Step 24538 | loss:0.5533222556114197 lr:2.3852429875443837e-05 tokens_per_second_per_gpu:4461.772344842555 mfu:21.04694140748268 +Step 24539 | loss:0.44174841046333313 lr:2.3848733940973693e-05 tokens_per_second_per_gpu:3921.5145036865115 mfu:18.498452993256883 +Step 24540 | loss:0.5115336179733276 lr:2.384503820319058e-05 tokens_per_second_per_gpu:4306.6320624261 mfu:20.315118225662616 +Step 24541 | loss:0.4927385747432709 lr:2.3841342662122246e-05 tokens_per_second_per_gpu:4576.997093316507 mfu:21.59047620540361 +Step 24542 | loss:0.4033365845680237 lr:2.3837647317796535e-05 tokens_per_second_per_gpu:5265.641595474442 mfu:24.838929816950344 +Step 24543 | loss:0.4823039770126343 lr:2.3833952170241196e-05 tokens_per_second_per_gpu:3772.3026781652784 mfu:17.79459535411063 +Step 24544 | loss:0.4391332268714905 lr:2.383025721948404e-05 tokens_per_second_per_gpu:5035.117636754696 mfu:23.751508972226215 +Step 24545 | loss:0.42366835474967957 lr:2.382656246555289e-05 tokens_per_second_per_gpu:5010.006320137594 mfu:23.633054607310935 +Step 24546 | loss:0.4986216723918915 lr:2.3822867908475472e-05 tokens_per_second_per_gpu:2452.515788678658 mfu:11.568935417539185 +Step 24547 | loss:0.4787750840187073 lr:2.3819173548279605e-05 tokens_per_second_per_gpu:4915.953436329919 mfu:23.189391107313437 +Step 24548 | loss:0.4073933959007263 lr:2.3815479384993095e-05 tokens_per_second_per_gpu:5230.17678193625 mfu:24.67163624056956 +Step 24549 | loss:0.4069395959377289 lr:2.381178541864371e-05 tokens_per_second_per_gpu:5248.499343278201 mfu:24.758066888570863 +Step 24550 | loss:0.36587828397750854 lr:2.3808091649259206e-05 tokens_per_second_per_gpu:5369.8339320737105 mfu:25.33042284573483 +Step 24551 | loss:0.5370378494262695 lr:2.3804398076867386e-05 tokens_per_second_per_gpu:4039.580982010775 mfu:19.055392715730843 +Step 24552 | loss:0.38432419300079346 lr:2.3800704701496053e-05 tokens_per_second_per_gpu:4985.397938619868 mfu:23.516972673068988 +Step 24553 | loss:0.5577444434165955 lr:2.3797011523172942e-05 tokens_per_second_per_gpu:5347.763338772005 mfu:25.22631209149917 +Step 24554 | loss:0.41438814997673035 lr:2.3793318541925853e-05 tokens_per_second_per_gpu:5341.423637196315 mfu:25.196406637502015 +Step 24555 | loss:0.42788195610046387 lr:2.3789625757782575e-05 tokens_per_second_per_gpu:3694.1723962798023 mfu:17.42604148405616 +Step 24556 | loss:0.38826632499694824 lr:2.3785933170770864e-05 tokens_per_second_per_gpu:5367.549535683241 mfu:25.31964695820277 +Step 24557 | loss:0.492872416973114 lr:2.378224078091848e-05 tokens_per_second_per_gpu:4493.317110545001 mfu:21.19574344939313 +Step 24558 | loss:0.48448556661605835 lr:2.377854858825321e-05 tokens_per_second_per_gpu:5217.504489179078 mfu:24.61185887351788 +Step 24559 | loss:0.42462798953056335 lr:2.3774856592802837e-05 tokens_per_second_per_gpu:5079.168551938915 mfu:23.95930465501132 +Step 24560 | loss:0.5439170002937317 lr:2.3771164794595098e-05 tokens_per_second_per_gpu:5343.858845621759 mfu:25.207893931134237 +Step 24561 | loss:0.3768305778503418 lr:2.3767473193657768e-05 tokens_per_second_per_gpu:5341.441013095715 mfu:25.196488602584118 +Step 24562 | loss:0.319247305393219 lr:2.3763781790018647e-05 tokens_per_second_per_gpu:4817.70922765017 mfu:22.725956412780594 +Step 24563 | loss:0.3818921148777008 lr:2.376009058370547e-05 tokens_per_second_per_gpu:5317.874228187579 mfu:25.085320057264315 +Step 24564 | loss:0.4462929964065552 lr:2.3756399574745976e-05 tokens_per_second_per_gpu:5030.036797573302 mfu:23.727541786927024 +Step 24565 | loss:0.38704174757003784 lr:2.3752708763167953e-05 tokens_per_second_per_gpu:4588.955227509816 mfu:21.64688476466153 +Step 24566 | loss:0.46796610951423645 lr:2.374901814899916e-05 tokens_per_second_per_gpu:4328.07190351114 mfu:20.416253614074844 +Step 24567 | loss:0.5033493041992188 lr:2.3745327732267364e-05 tokens_per_second_per_gpu:4245.1624245123585 mfu:20.025155455821803 +Step 24568 | loss:0.39384719729423523 lr:2.3741637513000288e-05 tokens_per_second_per_gpu:3831.4257774028865 mfu:18.07348910065534 +Step 24569 | loss:0.4227506220340729 lr:2.3737947491225733e-05 tokens_per_second_per_gpu:4999.375854615649 mfu:23.582908887699475 +Step 24570 | loss:0.3827143609523773 lr:2.3734257666971422e-05 tokens_per_second_per_gpu:5110.813489161405 mfu:24.108579223072844 +Step 24571 | loss:0.4707336127758026 lr:2.373056804026509e-05 tokens_per_second_per_gpu:4570.886148270607 mfu:21.561649834987364 +Step 24572 | loss:0.46945950388908386 lr:2.3726878611134507e-05 tokens_per_second_per_gpu:4697.410449370643 mfu:22.158486550548012 +Step 24573 | loss:0.5190526247024536 lr:2.3723189379607424e-05 tokens_per_second_per_gpu:4752.21976775767 mfu:22.41703145681362 +Step 24574 | loss:0.47745004296302795 lr:2.371950034571161e-05 tokens_per_second_per_gpu:5348.6536943053015 mfu:25.23051205046005 +Step 24575 | loss:0.4430590271949768 lr:2.371581150947476e-05 tokens_per_second_per_gpu:3665.5562112425537 mfu:17.291054056810832 +Step 24576 | loss:0.4081067144870758 lr:2.3712122870924673e-05 tokens_per_second_per_gpu:5317.198479010907 mfu:25.08213243310325 +Step 24577 | loss:0.39308100938796997 lr:2.3708434430089044e-05 tokens_per_second_per_gpu:4119.268057650769 mfu:19.431290247541774 +Step 24578 | loss:0.4619709253311157 lr:2.3704746186995648e-05 tokens_per_second_per_gpu:4738.454207549379 mfu:22.35209696066436 +Step 24579 | loss:0.35681477189064026 lr:2.3701058141672194e-05 tokens_per_second_per_gpu:4188.848761975912 mfu:19.7595143015358 +Step 24580 | loss:0.4041789174079895 lr:2.3697370294146444e-05 tokens_per_second_per_gpu:4980.658316984488 mfu:23.494615069152065 +Step 24581 | loss:0.3616914451122284 lr:2.369368264444614e-05 tokens_per_second_per_gpu:4500.329786961497 mfu:21.22882344053572 +Step 24582 | loss:0.31922322511672974 lr:2.368999519259899e-05 tokens_per_second_per_gpu:4751.5045341601635 mfu:22.41365757790275 +Step 24583 | loss:0.3144446015357971 lr:2.368630793863274e-05 tokens_per_second_per_gpu:3782.6991606398724 mfu:17.843637335766108 +Step 24584 | loss:0.502572238445282 lr:2.3682620882575145e-05 tokens_per_second_per_gpu:5332.910847624685 mfu:25.156250356660962 +Step 24585 | loss:0.4253145456314087 lr:2.367893402445392e-05 tokens_per_second_per_gpu:5312.240717726475 mfu:25.05874582724329 +Step 24586 | loss:0.39101099967956543 lr:2.367524736429677e-05 tokens_per_second_per_gpu:4016.8942622709988 mfu:18.94837558796479 +Step 24587 | loss:0.5030223727226257 lr:2.3671560902131447e-05 tokens_per_second_per_gpu:4447.407223684609 mfu:20.97917868003825 +Step 24588 | loss:0.4184325039386749 lr:2.3667874637985687e-05 tokens_per_second_per_gpu:3867.1978424025347 mfu:18.242232034602324 +Step 24589 | loss:0.5042176246643066 lr:2.3664188571887186e-05 tokens_per_second_per_gpu:4007.3445333677714 mfu:18.903327887376502 +Step 24590 | loss:0.39714038372039795 lr:2.3660502703863678e-05 tokens_per_second_per_gpu:4957.870287497737 mfu:23.387120046024307 +Step 24591 | loss:0.30124229192733765 lr:2.3656817033942918e-05 tokens_per_second_per_gpu:3941.554541724009 mfu:18.592985271862823 +Step 24592 | loss:0.47118934988975525 lr:2.3653131562152593e-05 tokens_per_second_per_gpu:5355.807641675523 mfu:25.26425844079467 +Step 24593 | loss:0.36753615736961365 lr:2.364944628852041e-05 tokens_per_second_per_gpu:4219.98448219122 mfu:19.906386806093067 +Step 24594 | loss:0.47331205010414124 lr:2.3645761213074102e-05 tokens_per_second_per_gpu:5333.197913945335 mfu:25.157604497474033 +Step 24595 | loss:0.38992545008659363 lr:2.3642076335841407e-05 tokens_per_second_per_gpu:4745.346958137686 mfu:22.38461124121392 +Step 24596 | loss:0.3202694356441498 lr:2.363839165685e-05 tokens_per_second_per_gpu:4172.915139918726 mfu:19.684352687735803 +Step 24597 | loss:0.3637270927429199 lr:2.363470717612761e-05 tokens_per_second_per_gpu:5355.291450997689 mfu:25.261823481296112 +Step 24598 | loss:0.5048422813415527 lr:2.3631022893701977e-05 tokens_per_second_per_gpu:4413.871442837576 mfu:20.82098467998857 +Step 24599 | loss:0.561140239238739 lr:2.3627338809600786e-05 tokens_per_second_per_gpu:3955.1272162704836 mfu:18.65700989343561 +Step 24600 | loss:0.4654713273048401 lr:2.3623654923851717e-05 tokens_per_second_per_gpu:5080.488829695829 mfu:23.9655326304533 +Step 24601 | loss:0.4096750319004059 lr:2.3619971236482513e-05 tokens_per_second_per_gpu:4548.995288632841 mfu:21.458386915110232 +Step 24602 | loss:0.454517126083374 lr:2.3616287747520888e-05 tokens_per_second_per_gpu:5366.592575063472 mfu:25.315132811685007 +Step 24603 | loss:0.5894802212715149 lr:2.3612604456994508e-05 tokens_per_second_per_gpu:5326.2827874430905 mfu:25.124984665920703 +Step 24604 | loss:0.3842123746871948 lr:2.3608921364931102e-05 tokens_per_second_per_gpu:5343.508937470316 mfu:25.20624335468339 +Step 24605 | loss:0.4576508402824402 lr:2.360523847135838e-05 tokens_per_second_per_gpu:3934.2404084554637 mfu:18.558483257315277 +Step 24606 | loss:0.3596128225326538 lr:2.360155577630403e-05 tokens_per_second_per_gpu:5066.419610673063 mfu:23.89916572386662 +Step 24607 | loss:0.3180122971534729 lr:2.359787327979573e-05 tokens_per_second_per_gpu:2962.50137621086 mfu:13.974624446442263 +Step 24608 | loss:0.35659077763557434 lr:2.3594190981861193e-05 tokens_per_second_per_gpu:5376.160565828094 mfu:25.360266656589285 +Step 24609 | loss:0.37858879566192627 lr:2.3590508882528113e-05 tokens_per_second_per_gpu:5349.954993832387 mfu:25.236650502353903 +Step 24610 | loss:0.4191495478153229 lr:2.3586826981824208e-05 tokens_per_second_per_gpu:4304.556546456452 mfu:20.30532766271425 +Step 24611 | loss:0.4868824779987335 lr:2.3583145279777125e-05 tokens_per_second_per_gpu:5096.905773975381 mfu:24.04297415763947 +Step 24612 | loss:0.47787848114967346 lr:2.357946377641458e-05 tokens_per_second_per_gpu:4983.345763366669 mfu:23.507292212262445 +Step 24613 | loss:0.39861661195755005 lr:2.3575782471764273e-05 tokens_per_second_per_gpu:5075.357788906834 mfu:23.941328635605775 +Step 24614 | loss:0.44956710934638977 lr:2.3572101365853883e-05 tokens_per_second_per_gpu:4810.525541058121 mfu:22.69206973746258 +Step 24615 | loss:0.4865991473197937 lr:2.3568420458711073e-05 tokens_per_second_per_gpu:5072.278137024673 mfu:23.926801392234832 +Step 24616 | loss:0.5346174240112305 lr:2.3564739750363542e-05 tokens_per_second_per_gpu:4096.628050759308 mfu:19.32449347225126 +Step 24617 | loss:0.523370623588562 lr:2.3561059240838994e-05 tokens_per_second_per_gpu:5286.112780240428 mfu:24.935495888235785 +Step 24618 | loss:0.3902322053909302 lr:2.3557378930165076e-05 tokens_per_second_per_gpu:5348.571807538039 mfu:25.230125776607547 +Step 24619 | loss:0.4499395489692688 lr:2.3553698818369483e-05 tokens_per_second_per_gpu:4076.271711902905 mfu:19.228469148716055 +Step 24620 | loss:0.36765867471694946 lr:2.355001890547992e-05 tokens_per_second_per_gpu:5321.851587512534 mfu:25.104081939807603 +Step 24621 | loss:0.5699514150619507 lr:2.3546339191524034e-05 tokens_per_second_per_gpu:5338.853131685291 mfu:25.184281124433053 +Step 24622 | loss:0.34632766246795654 lr:2.3542659676529493e-05 tokens_per_second_per_gpu:5302.280600733129 mfu:25.01176225601448 +Step 24623 | loss:0.5805578231811523 lr:2.3538980360523983e-05 tokens_per_second_per_gpu:4353.141815212512 mfu:20.534512664938216 +Step 24624 | loss:0.3883156180381775 lr:2.3535301243535197e-05 tokens_per_second_per_gpu:3720.3212960986534 mfu:17.549390305964042 +Step 24625 | loss:0.48365318775177 lr:2.3531622325590768e-05 tokens_per_second_per_gpu:4995.015390365891 mfu:23.56233983386145 +Step 24626 | loss:0.4427981972694397 lr:2.3527943606718384e-05 tokens_per_second_per_gpu:4348.3913654233065 mfu:20.512103982772228 +Step 24627 | loss:0.4339797794818878 lr:2.3524265086945735e-05 tokens_per_second_per_gpu:5386.117779842952 mfu:25.40723653397326 +Step 24628 | loss:0.4273415505886078 lr:2.3520586766300463e-05 tokens_per_second_per_gpu:4699.051763544977 mfu:22.16622890954542 +Step 24629 | loss:0.4322604537010193 lr:2.351690864481022e-05 tokens_per_second_per_gpu:5202.931784513117 mfu:24.543116939248744 +Step 24630 | loss:0.49201634526252747 lr:2.3513230722502684e-05 tokens_per_second_per_gpu:4878.250696593086 mfu:23.01154085122407 +Step 24631 | loss:0.46954482793807983 lr:2.3509552999405543e-05 tokens_per_second_per_gpu:4671.52796962293 mfu:22.03639447757164 +Step 24632 | loss:0.5350980758666992 lr:2.3505875475546414e-05 tokens_per_second_per_gpu:5305.898158840754 mfu:25.02882689482721 +Step 24633 | loss:0.4733107089996338 lr:2.3502198150952968e-05 tokens_per_second_per_gpu:5341.493876161665 mfu:25.196737966685575 +Step 24634 | loss:0.38489899039268494 lr:2.3498521025652904e-05 tokens_per_second_per_gpu:4071.495920047911 mfu:19.205940923701938 +Step 24635 | loss:0.331406831741333 lr:2.3494844099673834e-05 tokens_per_second_per_gpu:4577.488323644374 mfu:21.5927934226727 +Step 24636 | loss:0.4582907259464264 lr:2.349116737304341e-05 tokens_per_second_per_gpu:4811.877756252394 mfu:22.698448367245668 +Step 24637 | loss:0.5090488195419312 lr:2.3487490845789294e-05 tokens_per_second_per_gpu:3666.985078202226 mfu:17.297794266049447 +Step 24638 | loss:0.5855199694633484 lr:2.348381451793917e-05 tokens_per_second_per_gpu:3512.369499708079 mfu:16.568446202154632 +Step 24639 | loss:0.44336557388305664 lr:2.3480138389520633e-05 tokens_per_second_per_gpu:3922.0082143870163 mfu:18.500781910866895 +Step 24640 | loss:0.49736133217811584 lr:2.3476462460561362e-05 tokens_per_second_per_gpu:4602.054983214824 mfu:21.708678547370848 +Step 24641 | loss:0.3920721113681793 lr:2.3472786731089002e-05 tokens_per_second_per_gpu:5370.858090994775 mfu:25.335253978105023 +Step 24642 | loss:0.46097663044929504 lr:2.346911120113123e-05 tokens_per_second_per_gpu:4141.936026549764 mfu:19.538218924392638 +Step 24643 | loss:0.3269138038158417 lr:2.346543587071563e-05 tokens_per_second_per_gpu:4949.778782474485 mfu:23.34895103627651 +Step 24644 | loss:0.38939014077186584 lr:2.3461760739869866e-05 tokens_per_second_per_gpu:3815.743386394012 mfu:17.999512586574664 +Step 24645 | loss:0.4466763734817505 lr:2.345808580862159e-05 tokens_per_second_per_gpu:4622.22971315897 mfu:21.80384619068126 +Step 24646 | loss:0.4463363587856293 lr:2.345441107699845e-05 tokens_per_second_per_gpu:4721.443812968907 mfu:22.271856027155685 +Step 24647 | loss:0.38072577118873596 lr:2.3450736545028057e-05 tokens_per_second_per_gpu:3149.574229085255 mfu:14.857079011370754 +Step 24648 | loss:0.48784980177879333 lr:2.3447062212738065e-05 tokens_per_second_per_gpu:4365.148231684468 mfu:20.59114897994232 +Step 24649 | loss:0.5500738620758057 lr:2.344338808015612e-05 tokens_per_second_per_gpu:3479.5858445760327 mfu:16.413800107428425 +Step 24650 | loss:0.42584195733070374 lr:2.3439714147309845e-05 tokens_per_second_per_gpu:3728.0064798525523 mfu:17.58564262895874 +Step 24651 | loss:0.3891257345676422 lr:2.343604041422685e-05 tokens_per_second_per_gpu:5335.129051217592 mfu:25.166714001473267 +Step 24652 | loss:0.4188883602619171 lr:2.343236688093478e-05 tokens_per_second_per_gpu:3841.8922289758902 mfu:18.12286113848598 +Step 24653 | loss:0.38983356952667236 lr:2.3428693547461296e-05 tokens_per_second_per_gpu:5366.254425434163 mfu:25.31353770218928 +Step 24654 | loss:0.4674941599369049 lr:2.342502041383397e-05 tokens_per_second_per_gpu:4477.397448649601 mfu:21.120647688058128 +Step 24655 | loss:0.43421056866645813 lr:2.3421347480080464e-05 tokens_per_second_per_gpu:3785.19064851318 mfu:17.855390109156165 +Step 24656 | loss:0.37501394748687744 lr:2.3417674746228407e-05 tokens_per_second_per_gpu:3503.3219013928197 mfu:16.525767137222104 +Step 24657 | loss:0.36564210057258606 lr:2.341400221230542e-05 tokens_per_second_per_gpu:4004.7917199286967 mfu:18.8912858308295 +Step 24658 | loss:0.4242817163467407 lr:2.3410329878339086e-05 tokens_per_second_per_gpu:3313.325614142982 mfu:15.629522233555308 +Step 24659 | loss:0.38729315996170044 lr:2.3406657744357056e-05 tokens_per_second_per_gpu:4753.059921760666 mfu:22.42099460659081 +Step 24660 | loss:0.6804580092430115 lr:2.340298581038697e-05 tokens_per_second_per_gpu:5334.0014652606615 mfu:25.16139498612842 +Step 24661 | loss:0.44638973474502563 lr:2.3399314076456402e-05 tokens_per_second_per_gpu:4745.383607191206 mfu:22.38478412105235 +Step 24662 | loss:0.5184869766235352 lr:2.3395642542592984e-05 tokens_per_second_per_gpu:4929.920114447369 mfu:23.255274310954455 +Step 24663 | loss:0.5115225911140442 lr:2.3391971208824355e-05 tokens_per_second_per_gpu:4181.812569662485 mfu:19.726323381895167 +Step 24664 | loss:0.325945109128952 lr:2.33883000751781e-05 tokens_per_second_per_gpu:5327.444224326568 mfu:25.130463361861633 +Step 24665 | loss:0.4597654938697815 lr:2.3384629141681825e-05 tokens_per_second_per_gpu:5089.450025861641 mfu:24.007804121705156 +Step 24666 | loss:0.5414262413978577 lr:2.338095840836314e-05 tokens_per_second_per_gpu:5336.249153487129 mfu:25.171997714987825 +Step 24667 | loss:0.3873370289802551 lr:2.3377287875249693e-05 tokens_per_second_per_gpu:5380.594132903344 mfu:25.381180548927787 +Step 24668 | loss:0.48117855191230774 lr:2.3373617542369036e-05 tokens_per_second_per_gpu:5298.347988931293 mfu:24.993211455172457 +Step 24669 | loss:0.4427051544189453 lr:2.33699474097488e-05 tokens_per_second_per_gpu:4470.748667945669 mfu:21.089284255078717 +Step 24670 | loss:0.4303167760372162 lr:2.336627747741661e-05 tokens_per_second_per_gpu:3218.6974314427325 mfu:15.183144950525419 +Step 24671 | loss:0.42756184935569763 lr:2.336260774540004e-05 tokens_per_second_per_gpu:3919.9124290903305 mfu:18.49089573404458 +Step 24672 | loss:0.4992713928222656 lr:2.335893821372668e-05 tokens_per_second_per_gpu:3895.682394709666 mfu:18.376598527800834 +Step 24673 | loss:0.4087355136871338 lr:2.335526888242414e-05 tokens_per_second_per_gpu:5335.476757499506 mfu:25.16835419132986 +Step 24674 | loss:0.46916747093200684 lr:2.3351599751520048e-05 tokens_per_second_per_gpu:5307.372487679852 mfu:25.035781555507672 +Step 24675 | loss:0.4039337635040283 lr:2.3347930821041957e-05 tokens_per_second_per_gpu:4844.051721272792 mfu:22.850218449691372 +Step 24676 | loss:0.5502400994300842 lr:2.3344262091017477e-05 tokens_per_second_per_gpu:4657.093329986139 mfu:21.96830381960142 +Step 24677 | loss:0.3909033238887787 lr:2.33405935614742e-05 tokens_per_second_per_gpu:5310.169383597529 mfu:25.048974990751006 +Step 24678 | loss:0.3948952555656433 lr:2.333692523243976e-05 tokens_per_second_per_gpu:4944.818605630333 mfu:23.32555303580921 +Step 24679 | loss:0.46897977590560913 lr:2.3333257103941674e-05 tokens_per_second_per_gpu:4612.477071239123 mfu:21.757841314773042 +Step 24680 | loss:0.40828782320022583 lr:2.332958917600756e-05 tokens_per_second_per_gpu:5358.527126977076 mfu:25.277086716954127 +Step 24681 | loss:0.4028087556362152 lr:2.332592144866501e-05 tokens_per_second_per_gpu:4553.612616536355 mfu:21.48016763862848 +Step 24682 | loss:0.3607083261013031 lr:2.3322253921941627e-05 tokens_per_second_per_gpu:5331.8013248043535 mfu:25.151016548213345 +Step 24683 | loss:0.42379143834114075 lr:2.3318586595864957e-05 tokens_per_second_per_gpu:4229.13535432049 mfu:19.949553031226614 +Step 24684 | loss:0.39864736795425415 lr:2.33149194704626e-05 tokens_per_second_per_gpu:5314.611365237599 mfu:25.06992857602743 +Step 24685 | loss:0.3839125633239746 lr:2.331125254576216e-05 tokens_per_second_per_gpu:5319.54962745715 mfu:25.09322320147147 +Step 24686 | loss:0.4980214536190033 lr:2.3307585821791195e-05 tokens_per_second_per_gpu:5334.948605299609 mfu:25.165862807290893 +Step 24687 | loss:0.4196520447731018 lr:2.3303919298577265e-05 tokens_per_second_per_gpu:3742.75783784804 mfu:17.65522730146466 +Step 24688 | loss:0.47704610228538513 lr:2.330025297614796e-05 tokens_per_second_per_gpu:4933.913087954563 mfu:23.27410984825942 +Step 24689 | loss:0.4343603253364563 lr:2.3296586854530884e-05 tokens_per_second_per_gpu:4477.0076030146965 mfu:21.118808719683763 +Step 24690 | loss:0.4059034585952759 lr:2.3292920933753566e-05 tokens_per_second_per_gpu:4217.804326814232 mfu:19.896102641206948 +Step 24691 | loss:0.3771190643310547 lr:2.3289255213843592e-05 tokens_per_second_per_gpu:4173.118255824045 mfu:19.68531082011634 +Step 24692 | loss:0.39186522364616394 lr:2.3285589694828563e-05 tokens_per_second_per_gpu:5311.666753368625 mfu:25.056038339438892 +Step 24693 | loss:0.45435360074043274 lr:2.328192437673602e-05 tokens_per_second_per_gpu:5310.602028895981 mfu:25.0510158524407 +Step 24694 | loss:0.3851878345012665 lr:2.3278259259593516e-05 tokens_per_second_per_gpu:3328.2809230507264 mfu:15.700068977311956 +Step 24695 | loss:0.41785818338394165 lr:2.327459434342863e-05 tokens_per_second_per_gpu:3515.6407720078164 mfu:16.583877351729104 +Step 24696 | loss:0.4511367976665497 lr:2.327092962826895e-05 tokens_per_second_per_gpu:5358.07076435518 mfu:25.274933976603155 +Step 24697 | loss:0.4347456395626068 lr:2.3267265114142e-05 tokens_per_second_per_gpu:5336.752001464997 mfu:25.174369734694142 +Step 24698 | loss:0.3898608982563019 lr:2.3263600801075364e-05 tokens_per_second_per_gpu:4676.346757081484 mfu:22.05912552018356 +Step 24699 | loss:0.4436383545398712 lr:2.3259936689096613e-05 tokens_per_second_per_gpu:5343.531674946349 mfu:25.206350611237255 +Step 24700 | loss:0.5365185141563416 lr:2.325627277823329e-05 tokens_per_second_per_gpu:4493.914645269875 mfu:21.1985621226403 +Step 24701 | loss:0.4101971983909607 lr:2.3252609068512937e-05 tokens_per_second_per_gpu:3885.8902659977734 mfu:18.3304073859574 +Step 24702 | loss:0.3830348253250122 lr:2.324894555996312e-05 tokens_per_second_per_gpu:5345.2859961808945 mfu:25.21462604381867 +Step 24703 | loss:0.402751088142395 lr:2.3245282252611415e-05 tokens_per_second_per_gpu:4884.613820029537 mfu:23.041556790134617 +Step 24704 | loss:0.42051705718040466 lr:2.3241619146485342e-05 tokens_per_second_per_gpu:3650.8911529771635 mfu:17.221876474856035 +Step 24705 | loss:0.4058617949485779 lr:2.323795624161246e-05 tokens_per_second_per_gpu:5326.515701629994 mfu:25.126083361879584 +Step 24706 | loss:0.5850329399108887 lr:2.3234293538020352e-05 tokens_per_second_per_gpu:4408.463157852375 mfu:20.795472876964805 +Step 24707 | loss:0.2606733739376068 lr:2.323063103573653e-05 tokens_per_second_per_gpu:4123.869318183381 mfu:19.452995178530106 +Step 24708 | loss:0.2767230272293091 lr:2.3226968734788528e-05 tokens_per_second_per_gpu:5397.235999386383 mfu:25.45968307252349 +Step 24709 | loss:0.3417823612689972 lr:2.3223306635203917e-05 tokens_per_second_per_gpu:3517.5956296236673 mfu:16.593098748636173 +Step 24710 | loss:0.3862471878528595 lr:2.321964473701025e-05 tokens_per_second_per_gpu:5319.81835930435 mfu:25.094490855445095 +Step 24711 | loss:0.450629860162735 lr:2.321598304023503e-05 tokens_per_second_per_gpu:5120.625214968796 mfu:24.154862807759354 +Step 24712 | loss:0.563705325126648 lr:2.3212321544905818e-05 tokens_per_second_per_gpu:5242.35516338961 mfu:24.729083743731195 +Step 24713 | loss:0.4470542371273041 lr:2.3208660251050158e-05 tokens_per_second_per_gpu:3442.132604349338 mfu:16.237126783097388 +Step 24714 | loss:0.436640202999115 lr:2.3204999158695617e-05 tokens_per_second_per_gpu:5105.88468103513 mfu:24.085329194199822 +Step 24715 | loss:0.4168121814727783 lr:2.3201338267869665e-05 tokens_per_second_per_gpu:4964.2690688052335 mfu:23.417304189196436 +Step 24716 | loss:0.4194677770137787 lr:2.3197677578599865e-05 tokens_per_second_per_gpu:3867.3469211467136 mfu:18.242935264474074 +Step 24717 | loss:0.29422473907470703 lr:2.319401709091375e-05 tokens_per_second_per_gpu:3545.129470885696 mfu:16.7229805756269 +Step 24718 | loss:0.4251766800880432 lr:2.319035680483887e-05 tokens_per_second_per_gpu:4678.241152061352 mfu:22.068061704520712 +Step 24719 | loss:0.37474140524864197 lr:2.3186696720402723e-05 tokens_per_second_per_gpu:4486.076922645822 mfu:21.16159024776927 +Step 24720 | loss:0.35496440529823303 lr:2.318303683763285e-05 tokens_per_second_per_gpu:3985.369012016902 mfu:18.79966560375388 +Step 24721 | loss:0.5379796028137207 lr:2.3179377156556792e-05 tokens_per_second_per_gpu:5317.898681890287 mfu:25.08543540955222 +Step 24722 | loss:0.29799339175224304 lr:2.3175717677202068e-05 tokens_per_second_per_gpu:4572.640234655315 mfu:21.56992415974203 +Step 24723 | loss:0.41824740171432495 lr:2.3172058399596175e-05 tokens_per_second_per_gpu:3948.8208480527255 mfu:18.627261678575078 +Step 24724 | loss:0.4455678164958954 lr:2.316839932376665e-05 tokens_per_second_per_gpu:4565.051085289275 mfu:21.534124847340017 +Step 24725 | loss:0.4027886986732483 lr:2.3164740449741035e-05 tokens_per_second_per_gpu:4284.188051989167 mfu:20.20924599908856 +Step 24726 | loss:0.49283578991889954 lr:2.316108177754682e-05 tokens_per_second_per_gpu:5346.811396537576 mfu:25.221821617561275 +Step 24727 | loss:0.5839208960533142 lr:2.3157423307211527e-05 tokens_per_second_per_gpu:4266.064090199803 mfu:20.12375217906138 +Step 24728 | loss:0.349955290555954 lr:2.3153765038762697e-05 tokens_per_second_per_gpu:3971.179426180964 mfu:18.732730906372772 +Step 24729 | loss:0.4798480272293091 lr:2.3150106972227832e-05 tokens_per_second_per_gpu:5338.693978191009 mfu:25.183530370246135 +Step 24730 | loss:0.3600030243396759 lr:2.3146449107634414e-05 tokens_per_second_per_gpu:5353.239631389397 mfu:25.252144698127275 +Step 24731 | loss:0.4849497079849243 lr:2.3142791445009982e-05 tokens_per_second_per_gpu:4046.2898619005186 mfu:19.087039646823705 +Step 24732 | loss:0.4819307327270508 lr:2.313913398438206e-05 tokens_per_second_per_gpu:4654.533630774533 mfu:21.95622928169055 +Step 24733 | loss:0.40746334195137024 lr:2.313547672577812e-05 tokens_per_second_per_gpu:5312.868654048414 mfu:25.06170791001086 +Step 24734 | loss:0.4934169352054596 lr:2.3131819669225685e-05 tokens_per_second_per_gpu:5157.967382573147 mfu:24.33101218357985 +Step 24735 | loss:0.42460399866104126 lr:2.3128162814752287e-05 tokens_per_second_per_gpu:5042.382181411104 mfu:23.78577706882962 +Step 24736 | loss:0.4153711497783661 lr:2.31245061623854e-05 tokens_per_second_per_gpu:4333.51843185171 mfu:20.441945817530872 +Step 24737 | loss:0.381545752286911 lr:2.3120849712152513e-05 tokens_per_second_per_gpu:5372.963873480377 mfu:25.34518731337315 +Step 24738 | loss:0.49881690740585327 lr:2.311719346408115e-05 tokens_per_second_per_gpu:5285.399976401634 mfu:24.93213347091132 +Step 24739 | loss:0.5145347118377686 lr:2.311353741819882e-05 tokens_per_second_per_gpu:4492.370882953548 mfu:21.191279932401173 +Step 24740 | loss:0.3438187837600708 lr:2.3109881574532987e-05 tokens_per_second_per_gpu:3178.674946360016 mfu:14.994352059850154 +Step 24741 | loss:0.40553849935531616 lr:2.310622593311117e-05 tokens_per_second_per_gpu:3886.8748392708153 mfu:18.33505178607186 +Step 24742 | loss:0.4167574346065521 lr:2.3102570493960873e-05 tokens_per_second_per_gpu:3959.5610681856283 mfu:18.677925129412493 +Step 24743 | loss:0.47949889302253723 lr:2.309891525710958e-05 tokens_per_second_per_gpu:5204.712602268966 mfu:24.55151736044192 +Step 24744 | loss:0.3873361349105835 lr:2.3095260222584757e-05 tokens_per_second_per_gpu:4397.038515549093 mfu:20.741580889975246 +Step 24745 | loss:0.40233883261680603 lr:2.3091605390413916e-05 tokens_per_second_per_gpu:4740.303928021259 mfu:22.36082240772507 +Step 24746 | loss:0.42237377166748047 lr:2.308795076062456e-05 tokens_per_second_per_gpu:3660.6199163715746 mfu:17.267768711685672 +Step 24747 | loss:0.3893243968486786 lr:2.308429633324414e-05 tokens_per_second_per_gpu:4624.76165843654 mfu:21.815789808549965 +Step 24748 | loss:0.32054591178894043 lr:2.308064210830017e-05 tokens_per_second_per_gpu:5173.078275350586 mfu:24.40229284299519 +Step 24749 | loss:0.45324161648750305 lr:2.307698808582012e-05 tokens_per_second_per_gpu:5225.885546840173 mfu:24.651393752461157 +Step 24750 | loss:0.4588755965232849 lr:2.3073334265831514e-05 tokens_per_second_per_gpu:4941.575078182466 mfu:23.310252763434775 +Step 24751 | loss:0.41592204570770264 lr:2.3069680648361765e-05 tokens_per_second_per_gpu:3869.454879292856 mfu:18.252878862704176 +Step 24752 | loss:0.4992586374282837 lr:2.306602723343838e-05 tokens_per_second_per_gpu:4498.968403526579 mfu:21.222401562597245 +Step 24753 | loss:0.273293673992157 lr:2.3062374021088844e-05 tokens_per_second_per_gpu:4465.63979949219 mfu:21.06518485092082 +Step 24754 | loss:0.43068307638168335 lr:2.3058721011340644e-05 tokens_per_second_per_gpu:5261.751503035978 mfu:24.8205795871999 +Step 24755 | loss:0.37523025274276733 lr:2.3055068204221224e-05 tokens_per_second_per_gpu:5378.503082273878 mfu:25.37131670633857 +Step 24756 | loss:0.4889283776283264 lr:2.305141559975808e-05 tokens_per_second_per_gpu:4987.079875162989 mfu:23.524906654710854 +Step 24757 | loss:0.39866799116134644 lr:2.304776319797869e-05 tokens_per_second_per_gpu:5327.190720746741 mfu:25.12926754222342 +Step 24758 | loss:0.3922012150287628 lr:2.304411099891051e-05 tokens_per_second_per_gpu:5328.171859061216 mfu:25.13389573905226 +Step 24759 | loss:0.43305444717407227 lr:2.3040459002581e-05 tokens_per_second_per_gpu:5322.118677854029 mfu:25.10534185051874 +Step 24760 | loss:0.38439232110977173 lr:2.3036807209017637e-05 tokens_per_second_per_gpu:4097.182085837512 mfu:19.32710695024328 +Step 24761 | loss:0.4069353938102722 lr:2.3033155618247903e-05 tokens_per_second_per_gpu:5124.280137867383 mfu:24.172103702666135 +Step 24762 | loss:0.38436633348464966 lr:2.3029504230299227e-05 tokens_per_second_per_gpu:5101.73549504256 mfu:24.06575677594874 +Step 24763 | loss:0.36048901081085205 lr:2.302585304519909e-05 tokens_per_second_per_gpu:3246.1075302470776 mfu:15.312443063230612 +Step 24764 | loss:0.4593711793422699 lr:2.302220206297498e-05 tokens_per_second_per_gpu:5293.669782768275 mfu:24.971143558517227 +Step 24765 | loss:0.42260411381721497 lr:2.301855128365432e-05 tokens_per_second_per_gpu:4064.6674420296936 mfu:19.17372982783152 +Step 24766 | loss:0.37792518734931946 lr:2.3014900707264568e-05 tokens_per_second_per_gpu:5343.490275647634 mfu:25.206155323681763 +Step 24767 | loss:0.3348928689956665 lr:2.301125033383319e-05 tokens_per_second_per_gpu:3763.82124259485 mfu:17.75458697544279 +Step 24768 | loss:0.415476530790329 lr:2.300760016338766e-05 tokens_per_second_per_gpu:3867.6583457837232 mfu:18.24440430762137 +Step 24769 | loss:0.5368222594261169 lr:2.3003950195955392e-05 tokens_per_second_per_gpu:4124.189950041595 mfu:19.454507653710735 +Step 24770 | loss:0.35362106561660767 lr:2.3000300431563863e-05 tokens_per_second_per_gpu:4314.0001865087615 mfu:20.349874924092134 +Step 24771 | loss:0.43943339586257935 lr:2.2996650870240538e-05 tokens_per_second_per_gpu:5322.58186930251 mfu:25.10752680360984 +Step 24772 | loss:0.367568701505661 lr:2.2993001512012847e-05 tokens_per_second_per_gpu:4961.570827523048 mfu:23.404576124700935 +Step 24773 | loss:0.40554702281951904 lr:2.2989352356908218e-05 tokens_per_second_per_gpu:4142.894103184332 mfu:19.542738335342626 +Step 24774 | loss:0.38065141439437866 lr:2.2985703404954117e-05 tokens_per_second_per_gpu:5371.388969033753 mfu:25.337758220392015 +Step 24775 | loss:0.3983604907989502 lr:2.298205465617801e-05 tokens_per_second_per_gpu:5094.620959335753 mfu:24.032196297154766 +Step 24776 | loss:0.4416384696960449 lr:2.2978406110607293e-05 tokens_per_second_per_gpu:4483.061633107521 mfu:21.147366612555658 +Step 24777 | loss:0.4296434223651886 lr:2.2974757768269435e-05 tokens_per_second_per_gpu:3551.245751175735 mfu:16.75183211330004 +Step 24778 | loss:0.41439351439476013 lr:2.2971109629191888e-05 tokens_per_second_per_gpu:5395.543510193532 mfu:25.451699312232382 +Step 24779 | loss:0.457129567861557 lr:2.2967461693402075e-05 tokens_per_second_per_gpu:4128.397105131052 mfu:19.47435352208238 +Step 24780 | loss:0.326223760843277 lr:2.2963813960927417e-05 tokens_per_second_per_gpu:5058.289118553028 mfu:23.860812805331413 +Step 24781 | loss:0.384689599275589 lr:2.296016643179536e-05 tokens_per_second_per_gpu:5055.6606161628 mfu:23.848413711087957 +Step 24782 | loss:0.5050977468490601 lr:2.2956519106033365e-05 tokens_per_second_per_gpu:5348.109422671899 mfu:25.22794462830297 +Step 24783 | loss:0.4724959433078766 lr:2.2952871983668817e-05 tokens_per_second_per_gpu:4199.209895819294 mfu:19.808389537666756 +Step 24784 | loss:0.4355800151824951 lr:2.294922506472917e-05 tokens_per_second_per_gpu:3722.236877280635 mfu:17.5584264292313 +Step 24785 | loss:0.4016868770122528 lr:2.2945578349241858e-05 tokens_per_second_per_gpu:4805.77831139472 mfu:22.669676245179417 +Step 24786 | loss:0.4630829095840454 lr:2.2941931837234337e-05 tokens_per_second_per_gpu:4450.491223081514 mfu:20.993726431378008 +Step 24787 | loss:0.39148983359336853 lr:2.293828552873396e-05 tokens_per_second_per_gpu:4434.451899301062 mfu:20.91806620451479 +Step 24788 | loss:0.42663273215293884 lr:2.2934639423768194e-05 tokens_per_second_per_gpu:5372.889755549564 mfu:25.34483768644768 +Step 24789 | loss:0.38926568627357483 lr:2.2930993522364458e-05 tokens_per_second_per_gpu:5359.6367979556735 mfu:25.282321224289568 +Step 24790 | loss:0.4424196183681488 lr:2.2927347824550194e-05 tokens_per_second_per_gpu:4482.678221358498 mfu:21.14555799391873 +Step 24791 | loss:0.5162680149078369 lr:2.292370233035278e-05 tokens_per_second_per_gpu:4593.175205123957 mfu:21.66679111906986 +Step 24792 | loss:0.5021633505821228 lr:2.2920057039799652e-05 tokens_per_second_per_gpu:5346.771809827604 mfu:25.221634880296147 +Step 24793 | loss:0.433982253074646 lr:2.2916411952918256e-05 tokens_per_second_per_gpu:5320.294625676964 mfu:25.096737485184132 +Step 24794 | loss:0.4789261519908905 lr:2.291276706973598e-05 tokens_per_second_per_gpu:4596.760641470574 mfu:21.683704234055373 +Step 24795 | loss:0.30553382635116577 lr:2.290912239028022e-05 tokens_per_second_per_gpu:5137.6217020897275 mfu:24.235038137408516 +Step 24796 | loss:0.32541516423225403 lr:2.2905477914578404e-05 tokens_per_second_per_gpu:3563.9893552579383 mfu:16.81194586803885 +Step 24797 | loss:0.5899728536605835 lr:2.2901833642657973e-05 tokens_per_second_per_gpu:4445.424364492913 mfu:20.969825194920357 +Step 24798 | loss:0.46945032477378845 lr:2.289818957454628e-05 tokens_per_second_per_gpu:3590.562766749091 mfu:16.937297183934138 +Step 24799 | loss:0.3363126218318939 lr:2.289454571027077e-05 tokens_per_second_per_gpu:5315.924713049104 mfu:25.076123861734544 +Step 24800 | loss:0.3944300711154938 lr:2.2890902049858853e-05 tokens_per_second_per_gpu:3956.6892474019155 mfu:18.66437826079257 +Step 24801 | loss:0.43005508184432983 lr:2.2887258593337925e-05 tokens_per_second_per_gpu:5337.708912701824 mfu:25.17888364826415 +Step 24802 | loss:0.348360151052475 lr:2.2883615340735364e-05 tokens_per_second_per_gpu:3696.8340297060145 mfu:17.438596863049096 +Step 24803 | loss:0.5247630476951599 lr:2.2879972292078595e-05 tokens_per_second_per_gpu:5104.451748087031 mfu:24.07856980500007 +Step 24804 | loss:0.5068656802177429 lr:2.287632944739503e-05 tokens_per_second_per_gpu:4193.422466683988 mfu:19.781089246998288 +Step 24805 | loss:0.5015008449554443 lr:2.2872686806712035e-05 tokens_per_second_per_gpu:5352.986668913518 mfu:25.250951430968705 +Step 24806 | loss:0.46645891666412354 lr:2.286904437005702e-05 tokens_per_second_per_gpu:5274.533476786325 mfu:24.880874338209892 +Step 24807 | loss:0.3983131945133209 lr:2.286540213745741e-05 tokens_per_second_per_gpu:3993.086213533433 mfu:18.836068959997707 +Step 24808 | loss:0.3620617091655731 lr:2.2861760108940566e-05 tokens_per_second_per_gpu:4886.936837331034 mfu:23.052514859093403 +Step 24809 | loss:0.39678728580474854 lr:2.2858118284533874e-05 tokens_per_second_per_gpu:5322.410375562371 mfu:25.106717838377755 +Step 24810 | loss:0.4026591181755066 lr:2.2854476664264735e-05 tokens_per_second_per_gpu:5351.018269519092 mfu:25.241666155181633 +Step 24811 | loss:0.43815839290618896 lr:2.2850835248160556e-05 tokens_per_second_per_gpu:4888.935112369616 mfu:23.061941063391266 +Step 24812 | loss:0.4871273338794708 lr:2.2847194036248698e-05 tokens_per_second_per_gpu:5007.578640688534 mfu:23.621602829144894 +Step 24813 | loss:0.37952497601509094 lr:2.284355302855655e-05 tokens_per_second_per_gpu:5324.170090966649 mfu:25.115018716178138 +Step 24814 | loss:0.4020509719848633 lr:2.2839912225111532e-05 tokens_per_second_per_gpu:5265.73626775661 mfu:24.83937640225809 +Step 24815 | loss:0.3668566346168518 lr:2.2836271625940998e-05 tokens_per_second_per_gpu:3940.4976080192055 mfu:18.58799953524578 +Step 24816 | loss:0.5263113975524902 lr:2.283263123107231e-05 tokens_per_second_per_gpu:4741.4853934924295 mfu:22.366395581931382 +Step 24817 | loss:0.6559931635856628 lr:2.282899104053287e-05 tokens_per_second_per_gpu:4401.50330903111 mfu:20.76264208260224 +Step 24818 | loss:0.3904847502708435 lr:2.2825351054350075e-05 tokens_per_second_per_gpu:4457.600021424846 mfu:21.02725984604957 +Step 24819 | loss:0.35158771276474 lr:2.282171127255126e-05 tokens_per_second_per_gpu:5344.638848682762 mfu:25.21157333865523 +Step 24820 | loss:0.3077716529369354 lr:2.2818071695163828e-05 tokens_per_second_per_gpu:5344.835716742651 mfu:25.212501998882885 +Step 24821 | loss:0.549770176410675 lr:2.2814432322215145e-05 tokens_per_second_per_gpu:4690.224128106858 mfu:22.124587447033896 +Step 24822 | loss:0.3391565680503845 lr:2.2810793153732617e-05 tokens_per_second_per_gpu:3324.4873638982795 mfu:15.68217411154193 +Step 24823 | loss:0.3752592206001282 lr:2.280715418974355e-05 tokens_per_second_per_gpu:5387.832127512023 mfu:25.415323404426147 +Step 24824 | loss:0.528989851474762 lr:2.2803515430275353e-05 tokens_per_second_per_gpu:4945.998227926346 mfu:23.331117515443818 +Step 24825 | loss:0.5630337595939636 lr:2.279987687535538e-05 tokens_per_second_per_gpu:3940.104720717036 mfu:18.586146218807958 +Step 24826 | loss:0.4211653470993042 lr:2.2796238525011022e-05 tokens_per_second_per_gpu:4191.033976379278 mfu:19.7698223307125 +Step 24827 | loss:0.47313380241394043 lr:2.2792600379269606e-05 tokens_per_second_per_gpu:5344.696888333001 mfu:25.211847121587994 +Step 24828 | loss:0.409891813993454 lr:2.2788962438158516e-05 tokens_per_second_per_gpu:3428.657561053312 mfu:16.173562704790573 +Step 24829 | loss:0.43397173285484314 lr:2.278532470170513e-05 tokens_per_second_per_gpu:5297.5616092211685 mfu:24.98950196790959 +Step 24830 | loss:0.4516840875148773 lr:2.2781687169936795e-05 tokens_per_second_per_gpu:5339.569812606028 mfu:25.18766183061454 +Step 24831 | loss:0.5063908100128174 lr:2.2778049842880838e-05 tokens_per_second_per_gpu:4360.9079778223895 mfu:20.571146979012987 +Step 24832 | loss:0.3726254999637604 lr:2.2774412720564643e-05 tokens_per_second_per_gpu:5349.034965216702 mfu:25.23231057040065 +Step 24833 | loss:0.4341539144515991 lr:2.2770775803015583e-05 tokens_per_second_per_gpu:4087.580737839699 mfu:19.281815753578705 +Step 24834 | loss:0.5092834234237671 lr:2.2767139090260974e-05 tokens_per_second_per_gpu:3667.7202846918835 mfu:17.301262360500324 +Step 24835 | loss:0.39002013206481934 lr:2.2763502582328187e-05 tokens_per_second_per_gpu:3805.0642458768807 mfu:17.949137258706102 +Step 24836 | loss:0.5538493990898132 lr:2.275986627924459e-05 tokens_per_second_per_gpu:4926.773556637384 mfu:23.24043146090664 +Step 24837 | loss:0.4520048201084137 lr:2.275623018103752e-05 tokens_per_second_per_gpu:4974.846447865509 mfu:23.467199490910716 +Step 24838 | loss:0.35973820090293884 lr:2.275259428773429e-05 tokens_per_second_per_gpu:5348.560076646001 mfu:25.230070439988307 +Step 24839 | loss:0.2751494348049164 lr:2.2748958599362286e-05 tokens_per_second_per_gpu:5346.097217090931 mfu:25.21845271125987 +Step 24840 | loss:0.3239010274410248 lr:2.2745323115948856e-05 tokens_per_second_per_gpu:4516.14043820907 mfu:21.30340498004579 +Step 24841 | loss:0.3519028127193451 lr:2.274168783752131e-05 tokens_per_second_per_gpu:4395.037788896539 mfu:20.732143120995985 +Step 24842 | loss:0.40585440397262573 lr:2.2738052764107002e-05 tokens_per_second_per_gpu:4765.9353749903785 mfu:22.481730316254207 +Step 24843 | loss:0.369148850440979 lr:2.2734417895733307e-05 tokens_per_second_per_gpu:4314.585128013955 mfu:20.352634192973444 +Step 24844 | loss:0.42309054732322693 lr:2.2730783232427527e-05 tokens_per_second_per_gpu:5250.738382829137 mfu:24.768628820152166 +Step 24845 | loss:0.48856016993522644 lr:2.272714877421699e-05 tokens_per_second_per_gpu:3603.508848427739 mfu:16.998366060041622 +Step 24846 | loss:0.3967628479003906 lr:2.2723514521129047e-05 tokens_per_second_per_gpu:5369.353470272576 mfu:25.328156425443645 +Step 24847 | loss:0.3373051583766937 lr:2.271988047319105e-05 tokens_per_second_per_gpu:4782.175172164577 mfu:22.558336210320014 +Step 24848 | loss:0.3576212227344513 lr:2.2716246630430292e-05 tokens_per_second_per_gpu:2861.843755125338 mfu:13.499805273820773 +Step 24849 | loss:0.5388937592506409 lr:2.2712612992874134e-05 tokens_per_second_per_gpu:4232.798859648695 mfu:19.966834410918217 +Step 24850 | loss:0.42321401834487915 lr:2.2708979560549902e-05 tokens_per_second_per_gpu:5021.305501170399 mfu:23.686354772081756 +Step 24851 | loss:0.4659073054790497 lr:2.270534633348492e-05 tokens_per_second_per_gpu:4395.25589321722 mfu:20.733171956289066 +Step 24852 | loss:0.3805820345878601 lr:2.2701713311706497e-05 tokens_per_second_per_gpu:3610.17787608141 mfu:17.02982500133744 +Step 24853 | loss:0.35205286741256714 lr:2.2698080495241967e-05 tokens_per_second_per_gpu:5358.281284545731 mfu:25.275927036260434 +Step 24854 | loss:0.40559545159339905 lr:2.269444788411868e-05 tokens_per_second_per_gpu:3574.225500298235 mfu:16.860231510660366 +Step 24855 | loss:0.47956350445747375 lr:2.2690815478363914e-05 tokens_per_second_per_gpu:5347.007152166173 mfu:25.222745030260725 +Step 24856 | loss:0.28387531638145447 lr:2.2687183278005008e-05 tokens_per_second_per_gpu:4792.289546302405 mfu:22.60604743881833 +Step 24857 | loss:0.3838319778442383 lr:2.2683551283069287e-05 tokens_per_second_per_gpu:3374.854098271608 mfu:15.919762591031933 +Step 24858 | loss:0.652496874332428 lr:2.2679919493584074e-05 tokens_per_second_per_gpu:5306.098205617483 mfu:25.029770549604535 +Step 24859 | loss:0.4019513726234436 lr:2.2676287909576677e-05 tokens_per_second_per_gpu:5176.347451739888 mfu:24.417714105802535 +Step 24860 | loss:0.362624853849411 lr:2.2672656531074388e-05 tokens_per_second_per_gpu:5353.205512792995 mfu:25.251983754886815 +Step 24861 | loss:0.41707149147987366 lr:2.2669025358104552e-05 tokens_per_second_per_gpu:4213.280483530582 mfu:19.874762900590717 +Step 24862 | loss:0.416045606136322 lr:2.2665394390694443e-05 tokens_per_second_per_gpu:3962.3640638049496 mfu:18.691147337989197 +Step 24863 | loss:0.35498476028442383 lr:2.266176362887139e-05 tokens_per_second_per_gpu:3765.3401306349556 mfu:17.761751829476992 +Step 24864 | loss:0.4271574020385742 lr:2.2658133072662707e-05 tokens_per_second_per_gpu:5338.765797962972 mfu:25.183869156364203 +Step 24865 | loss:0.39902064204216003 lr:2.265450272209571e-05 tokens_per_second_per_gpu:4312.7671564617285 mfu:20.34405851098366 +Step 24866 | loss:0.39569398760795593 lr:2.2650872577197683e-05 tokens_per_second_per_gpu:5308.4533725082265 mfu:25.04088027366108 +Step 24867 | loss:0.38271328806877136 lr:2.2647242637995926e-05 tokens_per_second_per_gpu:5316.377243464542 mfu:25.078258524913817 +Step 24868 | loss:0.47934508323669434 lr:2.2643612904517737e-05 tokens_per_second_per_gpu:4671.921377572926 mfu:22.038250249994086 +Step 24869 | loss:0.3876400291919708 lr:2.2639983376790447e-05 tokens_per_second_per_gpu:3827.4062248365326 mfu:18.054528185393465 +Step 24870 | loss:0.4184482991695404 lr:2.2636354054841314e-05 tokens_per_second_per_gpu:4587.6712775054675 mfu:21.64082815342656 +Step 24871 | loss:0.5077939629554749 lr:2.2632724938697657e-05 tokens_per_second_per_gpu:3700.942058984417 mfu:17.45797513805739 +Step 24872 | loss:0.4260462820529938 lr:2.262909602838678e-05 tokens_per_second_per_gpu:4193.203652121614 mfu:19.780057061373952 +Step 24873 | loss:0.3172147274017334 lr:2.2625467323935972e-05 tokens_per_second_per_gpu:3710.0797839582838 mfu:17.501079345815764 +Step 24874 | loss:0.37953752279281616 lr:2.2621838825372493e-05 tokens_per_second_per_gpu:4127.536542751026 mfu:19.470294102508717 +Step 24875 | loss:0.41878628730773926 lr:2.261821053272366e-05 tokens_per_second_per_gpu:5346.427106759347 mfu:25.220008857110773 +Step 24876 | loss:0.3752918839454651 lr:2.261458244601678e-05 tokens_per_second_per_gpu:4941.620067425995 mfu:23.310464985372455 +Step 24877 | loss:0.3985992670059204 lr:2.26109545652791e-05 tokens_per_second_per_gpu:4101.909333224219 mfu:19.34940618271795 +Step 24878 | loss:0.37542635202407837 lr:2.260732689053793e-05 tokens_per_second_per_gpu:4029.061219717452 mfu:19.005769202136708 +Step 24879 | loss:0.465084433555603 lr:2.2603699421820557e-05 tokens_per_second_per_gpu:5003.915453731575 mfu:23.604322951264777 +Step 24880 | loss:0.41055482625961304 lr:2.2600072159154263e-05 tokens_per_second_per_gpu:3610.3833681769306 mfu:17.030794342612577 +Step 24881 | loss:0.3842892050743103 lr:2.2596445102566304e-05 tokens_per_second_per_gpu:4908.089461659266 mfu:23.152295397059532 +Step 24882 | loss:0.41015544533729553 lr:2.259281825208398e-05 tokens_per_second_per_gpu:4454.544192139987 mfu:21.01284497793465 +Step 24883 | loss:0.40841448307037354 lr:2.2589191607734582e-05 tokens_per_second_per_gpu:3542.6920324658236 mfu:16.711482762730483 +Step 24884 | loss:0.35979461669921875 lr:2.2585565169545354e-05 tokens_per_second_per_gpu:4479.821192369242 mfu:21.132080900717 +Step 24885 | loss:0.44158855080604553 lr:2.258193893754359e-05 tokens_per_second_per_gpu:5363.4749442815555 mfu:25.300426415364043 +Step 24886 | loss:0.34688401222229004 lr:2.2578312911756578e-05 tokens_per_second_per_gpu:5340.404057678784 mfu:25.191597106959616 +Step 24887 | loss:0.3806522786617279 lr:2.2574687092211554e-05 tokens_per_second_per_gpu:5328.271763444095 mfu:25.13436700507174 +Step 24888 | loss:0.4644550681114197 lr:2.2571061478935835e-05 tokens_per_second_per_gpu:4194.9700487971195 mfu:19.788389455870664 +Step 24889 | loss:0.3737916648387909 lr:2.2567436071956633e-05 tokens_per_second_per_gpu:4689.328150738295 mfu:22.120360968916884 +Step 24890 | loss:0.3999996781349182 lr:2.256381087130127e-05 tokens_per_second_per_gpu:4716.880045321453 mfu:22.250327956503586 +Step 24891 | loss:0.43033868074417114 lr:2.2560185876996965e-05 tokens_per_second_per_gpu:3683.187912374035 mfu:17.374225799326446 +Step 24892 | loss:0.39185404777526855 lr:2.2556561089071003e-05 tokens_per_second_per_gpu:5330.964534278333 mfu:25.147069264530924 +Step 24893 | loss:0.4068385660648346 lr:2.255293650755067e-05 tokens_per_second_per_gpu:5119.52533134474 mfu:24.149674469045184 +Step 24894 | loss:0.36446189880371094 lr:2.254931213246318e-05 tokens_per_second_per_gpu:4435.391732239861 mfu:20.92249955683914 +Step 24895 | loss:0.486356645822525 lr:2.2545687963835837e-05 tokens_per_second_per_gpu:4942.274287570989 mfu:23.31355105341786 +Step 24896 | loss:0.5040189623832703 lr:2.2542064001695862e-05 tokens_per_second_per_gpu:3785.27080057209 mfu:17.855768200093397 +Step 24897 | loss:0.45148274302482605 lr:2.2538440246070542e-05 tokens_per_second_per_gpu:4975.506522824118 mfu:23.470313177110874 +Step 24898 | loss:0.4623899459838867 lr:2.25348166969871e-05 tokens_per_second_per_gpu:3652.2150517550303 mfu:17.228121531271658 +Step 24899 | loss:0.4927330017089844 lr:2.2531193354472802e-05 tokens_per_second_per_gpu:3743.0846039130984 mfu:17.656768712745578 +Step 24900 | loss:0.4352889358997345 lr:2.252757021855491e-05 tokens_per_second_per_gpu:4468.536874419741 mfu:21.078850847646027 +Step 24901 | loss:0.32842862606048584 lr:2.2523947289260682e-05 tokens_per_second_per_gpu:5362.119061273473 mfu:25.29403048387742 +Step 24902 | loss:0.4099010229110718 lr:2.2520324566617358e-05 tokens_per_second_per_gpu:4785.608644451164 mfu:22.574532484906626 +Step 24903 | loss:0.40105482935905457 lr:2.2516702050652156e-05 tokens_per_second_per_gpu:3002.867719708959 mfu:14.165039375931183 +Step 24904 | loss:0.5226405262947083 lr:2.2513079741392345e-05 tokens_per_second_per_gpu:5350.595902878168 mfu:25.239673779673215 +Step 24905 | loss:0.440622478723526 lr:2.250945763886519e-05 tokens_per_second_per_gpu:5347.101749121614 mfu:25.223191260239442 +Step 24906 | loss:0.4986649751663208 lr:2.2505835743097888e-05 tokens_per_second_per_gpu:5341.924344860081 mfu:25.198768560981538 +Step 24907 | loss:0.48185357451438904 lr:2.250221405411771e-05 tokens_per_second_per_gpu:4605.129667272009 mfu:21.723182356663784 +Step 24908 | loss:0.4697553813457489 lr:2.2498592571951903e-05 tokens_per_second_per_gpu:5330.958500603632 mfu:25.14704080265791 +Step 24909 | loss:0.5122370719909668 lr:2.2494971296627694e-05 tokens_per_second_per_gpu:5028.725981005103 mfu:23.721358441526462 +Step 24910 | loss:0.42207860946655273 lr:2.249135022817229e-05 tokens_per_second_per_gpu:5230.6251065012875 mfu:24.673751064034896 +Step 24911 | loss:0.3350023627281189 lr:2.2487729366612957e-05 tokens_per_second_per_gpu:4001.5108417839656 mfu:18.87580936884527 +Step 24912 | loss:0.33979493379592896 lr:2.248410871197694e-05 tokens_per_second_per_gpu:4691.063187778656 mfu:22.12854543466411 +Step 24913 | loss:0.3632274866104126 lr:2.2480488264291433e-05 tokens_per_second_per_gpu:5125.537556953813 mfu:24.178035163034647 +Step 24914 | loss:0.500318706035614 lr:2.2476868023583685e-05 tokens_per_second_per_gpu:4160.031058719165 mfu:19.623576278466548 +Step 24915 | loss:0.3365638554096222 lr:2.2473247989880946e-05 tokens_per_second_per_gpu:5318.126408552039 mfu:25.086509635069877 +Step 24916 | loss:0.5027853846549988 lr:2.246962816321042e-05 tokens_per_second_per_gpu:4663.836745110755 mfu:22.000113659289493 +Step 24917 | loss:0.4458393454551697 lr:2.2466008543599315e-05 tokens_per_second_per_gpu:3589.5656728482727 mfu:16.932593721882217 +Step 24918 | loss:0.399611234664917 lr:2.246238913107487e-05 tokens_per_second_per_gpu:4681.686173402796 mfu:22.084312457969553 +Step 24919 | loss:0.46753212809562683 lr:2.245876992566433e-05 tokens_per_second_per_gpu:5337.148735955046 mfu:25.176241198973585 +Step 24920 | loss:0.40218237042427063 lr:2.245515092739488e-05 tokens_per_second_per_gpu:4232.180858757917 mfu:19.963919195274595 +Step 24921 | loss:0.3701326251029968 lr:2.245153213629375e-05 tokens_per_second_per_gpu:5333.5161105657535 mfu:25.15910548522249 +Step 24922 | loss:0.49097153544425964 lr:2.2447913552388184e-05 tokens_per_second_per_gpu:5330.182858253461 mfu:25.143381965353917 +Step 24923 | loss:0.46600767970085144 lr:2.2444295175705354e-05 tokens_per_second_per_gpu:4045.030146502509 mfu:19.08109735436168 +Step 24924 | loss:0.3489365875720978 lr:2.244067700627252e-05 tokens_per_second_per_gpu:5354.818107710969 mfu:25.259590640251965 +Step 24925 | loss:0.45896902680397034 lr:2.2437059044116847e-05 tokens_per_second_per_gpu:3973.31505016431 mfu:18.74280500907724 +Step 24926 | loss:0.4124618470668793 lr:2.2433441289265594e-05 tokens_per_second_per_gpu:4141.744972859347 mfu:19.53731769153782 +Step 24927 | loss:0.319780170917511 lr:2.2429823741745925e-05 tokens_per_second_per_gpu:5342.505289143043 mfu:25.201508974283584 +Step 24928 | loss:0.3771376609802246 lr:2.242620640158507e-05 tokens_per_second_per_gpu:3950.5259690966495 mfu:18.635305025462916 +Step 24929 | loss:0.3500858545303345 lr:2.242258926881025e-05 tokens_per_second_per_gpu:4882.695492999102 mfu:23.032507714231713 +Step 24930 | loss:0.41665956377983093 lr:2.241897234344864e-05 tokens_per_second_per_gpu:5354.343559044081 mfu:25.257352113223902 +Step 24931 | loss:0.5478463768959045 lr:2.2415355625527478e-05 tokens_per_second_per_gpu:5262.341006343627 mfu:24.82336037488172 +Step 24932 | loss:0.5034176707267761 lr:2.2411739115073926e-05 tokens_per_second_per_gpu:5344.197508320449 mfu:25.20949146086577 +Step 24933 | loss:0.47505202889442444 lr:2.2408122812115224e-05 tokens_per_second_per_gpu:5342.706046733525 mfu:25.20245598209066 +Step 24934 | loss:0.3051040768623352 lr:2.240450671667853e-05 tokens_per_second_per_gpu:3626.081595599821 mfu:17.104845559760136 +Step 24935 | loss:0.4139261841773987 lr:2.2400890828791066e-05 tokens_per_second_per_gpu:5347.958952763696 mfu:25.227234836073166 +Step 24936 | loss:0.47902312874794006 lr:2.2397275148480022e-05 tokens_per_second_per_gpu:5330.310754011161 mfu:25.14398527146426 +Step 24937 | loss:0.4168008863925934 lr:2.239365967577261e-05 tokens_per_second_per_gpu:5351.6376262520735 mfu:25.24458776656426 +Step 24938 | loss:0.4214051067829132 lr:2.239004441069601e-05 tokens_per_second_per_gpu:4726.769770963882 mfu:22.29697947972012 +Step 24939 | loss:0.3189588189125061 lr:2.2386429353277387e-05 tokens_per_second_per_gpu:4449.570312055631 mfu:20.98938233694529 +Step 24940 | loss:0.3547220826148987 lr:2.2382814503543954e-05 tokens_per_second_per_gpu:4984.824916586462 mfu:23.51426963036977 +Step 24941 | loss:0.4326288402080536 lr:2.2379199861522914e-05 tokens_per_second_per_gpu:4187.895399408448 mfu:19.755017127644646 +Step 24942 | loss:0.5436481237411499 lr:2.2375585427241418e-05 tokens_per_second_per_gpu:4156.254817959506 mfu:19.605763106510125 +Step 24943 | loss:0.3108009994029999 lr:2.2371971200726672e-05 tokens_per_second_per_gpu:5281.022058163559 mfu:24.911482083632446 +Step 24944 | loss:0.44947612285614014 lr:2.2368357182005872e-05 tokens_per_second_per_gpu:4766.77850762388 mfu:22.485707516739698 +Step 24945 | loss:0.40859347581863403 lr:2.236474337110619e-05 tokens_per_second_per_gpu:4997.420630034531 mfu:23.573685759754632 +Step 24946 | loss:0.4543296694755554 lr:2.2361129768054773e-05 tokens_per_second_per_gpu:4279.80014524673 mfu:20.18854749433039 +Step 24947 | loss:0.3937647044658661 lr:2.235751637287883e-05 tokens_per_second_per_gpu:5314.570355048978 mfu:25.069735123970645 +Step 24948 | loss:0.39146825671195984 lr:2.2353903185605547e-05 tokens_per_second_per_gpu:3549.9400936120314 mfu:16.745673103803895 +Step 24949 | loss:0.4408186078071594 lr:2.235029020626207e-05 tokens_per_second_per_gpu:4735.313963275354 mfu:22.33728389264286 +Step 24950 | loss:0.44686979055404663 lr:2.234667743487559e-05 tokens_per_second_per_gpu:4842.681554306229 mfu:22.843755138335506 +Step 24951 | loss:0.44675806164741516 lr:2.234306487147329e-05 tokens_per_second_per_gpu:5355.670267731811 mfu:25.263610424463767 +Step 24952 | loss:0.4015611410140991 lr:2.2339452516082315e-05 tokens_per_second_per_gpu:5347.111368925285 mfu:25.223236638494665 +Step 24953 | loss:0.4594877362251282 lr:2.2335840368729864e-05 tokens_per_second_per_gpu:4254.007574919296 mfu:20.066879539429603 +Step 24954 | loss:0.3857123851776123 lr:2.2332228429443074e-05 tokens_per_second_per_gpu:3748.0269276911044 mfu:17.680082497253824 +Step 24955 | loss:0.4581720530986786 lr:2.232861669824914e-05 tokens_per_second_per_gpu:3707.5681411340524 mfu:17.48923149808341 +Step 24956 | loss:0.33131512999534607 lr:2.232500517517519e-05 tokens_per_second_per_gpu:5367.570673461287 mfu:25.31974666870829 +Step 24957 | loss:0.3081182539463043 lr:2.2321393860248406e-05 tokens_per_second_per_gpu:5314.95420091471 mfu:25.071545790410163 +Step 24958 | loss:0.3674428164958954 lr:2.2317782753495976e-05 tokens_per_second_per_gpu:5349.339273174575 mfu:25.233746042958227 +Step 24959 | loss:0.4356701970100403 lr:2.231417185494501e-05 tokens_per_second_per_gpu:3433.8767120247676 mfu:16.19818233040191 +Step 24960 | loss:0.44733670353889465 lr:2.2310561164622713e-05 tokens_per_second_per_gpu:4088.313095790937 mfu:19.285270410987845 +Step 24961 | loss:0.4230878949165344 lr:2.2306950682556193e-05 tokens_per_second_per_gpu:4620.806537275437 mfu:21.79713282721979 +Step 24962 | loss:0.4909655451774597 lr:2.2303340408772655e-05 tokens_per_second_per_gpu:4793.031567339119 mfu:22.60954767865035 +Step 24963 | loss:0.4533826410770416 lr:2.229973034329921e-05 tokens_per_second_per_gpu:5352.080114963266 mfu:25.246675061311315 +Step 24964 | loss:0.40700066089630127 lr:2.2296120486163023e-05 tokens_per_second_per_gpu:4539.971415564936 mfu:21.41581976621742 +Step 24965 | loss:0.40439313650131226 lr:2.2292510837391267e-05 tokens_per_second_per_gpu:5224.631820592896 mfu:24.64547971184499 +Step 24966 | loss:0.4343855381011963 lr:2.228890139701106e-05 tokens_per_second_per_gpu:5344.32535828983 mfu:25.210094550984792 +Step 24967 | loss:0.4345993399620056 lr:2.2285292165049577e-05 tokens_per_second_per_gpu:4396.317011773046 mfu:20.7381774335624 +Step 24968 | loss:0.550973117351532 lr:2.2281683141533927e-05 tokens_per_second_per_gpu:4859.623511752022 mfu:22.923673242203318 +Step 24969 | loss:0.4758533239364624 lr:2.22780743264913e-05 tokens_per_second_per_gpu:5358.239189629108 mfu:25.27572846735659 +Step 24970 | loss:0.3316513001918793 lr:2.227446571994879e-05 tokens_per_second_per_gpu:5244.048811849469 mfu:24.73707297248257 +Step 24971 | loss:0.4176054894924164 lr:2.2270857321933557e-05 tokens_per_second_per_gpu:5315.283800703861 mfu:25.07310056885111 +Step 24972 | loss:0.45685410499572754 lr:2.226724913247275e-05 tokens_per_second_per_gpu:4878.910858018747 mfu:23.014654945304066 +Step 24973 | loss:0.46962809562683105 lr:2.2263641151593524e-05 tokens_per_second_per_gpu:5173.656567291491 mfu:24.40502074474722 +Step 24974 | loss:0.42963355779647827 lr:2.226003337932298e-05 tokens_per_second_per_gpu:4568.26316253413 mfu:21.549276763740888 +Step 24975 | loss:0.46234598755836487 lr:2.225642581568826e-05 tokens_per_second_per_gpu:4609.755215643946 mfu:21.745001857534145 +Step 24976 | loss:0.33842355012893677 lr:2.225281846071649e-05 tokens_per_second_per_gpu:4755.829878318156 mfu:22.434060964275815 +Step 24977 | loss:0.4857167601585388 lr:2.2249211314434836e-05 tokens_per_second_per_gpu:4111.104762005345 mfu:19.39278258918045 +Step 24978 | loss:0.28735587000846863 lr:2.224560437687039e-05 tokens_per_second_per_gpu:5368.114967573038 mfu:25.322314196899573 +Step 24979 | loss:0.46587178111076355 lr:2.2241997648050288e-05 tokens_per_second_per_gpu:3249.4789259856107 mfu:15.3283465121487 +Step 24980 | loss:0.4341568648815155 lr:2.2238391128001683e-05 tokens_per_second_per_gpu:2614.462720433852 mfu:12.332866725624212 +Step 24981 | loss:0.4056583046913147 lr:2.2234784816751664e-05 tokens_per_second_per_gpu:4484.839402456349 mfu:21.155752653892804 +Step 24982 | loss:0.49180394411087036 lr:2.2231178714327396e-05 tokens_per_second_per_gpu:5052.8133927841045 mfu:23.834982872624245 +Step 24983 | loss:0.556495726108551 lr:2.222757282075595e-05 tokens_per_second_per_gpu:5095.323208002652 mfu:24.03550892393251 +Step 24984 | loss:0.31660816073417664 lr:2.222396713606449e-05 tokens_per_second_per_gpu:4805.306316507117 mfu:22.667449764763962 +Step 24985 | loss:0.458954781293869 lr:2.2220361660280104e-05 tokens_per_second_per_gpu:4269.477690599603 mfu:20.139854714567523 +Step 24986 | loss:0.37824565172195435 lr:2.221675639342992e-05 tokens_per_second_per_gpu:4405.422458568154 mfu:20.78112937964434 +Step 24987 | loss:0.4238513708114624 lr:2.2213151335541076e-05 tokens_per_second_per_gpu:5343.054916640016 mfu:25.204101660962923 +Step 24988 | loss:0.3635731041431427 lr:2.220954648664065e-05 tokens_per_second_per_gpu:5365.181225175534 mfu:25.30847523346248 +Step 24989 | loss:0.4447893500328064 lr:2.2205941846755784e-05 tokens_per_second_per_gpu:5318.499078612884 mfu:25.088267583330953 +Step 24990 | loss:0.45210686326026917 lr:2.220233741591356e-05 tokens_per_second_per_gpu:5306.963183058096 mfu:25.033850795772274 +Step 24991 | loss:0.4140801727771759 lr:2.219873319414112e-05 tokens_per_second_per_gpu:4783.375462912156 mfu:22.563998186567066 +Step 24992 | loss:0.42598602175712585 lr:2.219512918146554e-05 tokens_per_second_per_gpu:4888.00787282367 mfu:23.05756711624966 +Step 24993 | loss:0.39549022912979126 lr:2.2191525377913942e-05 tokens_per_second_per_gpu:5309.081826225335 mfu:25.043844797069944 +Step 24994 | loss:0.4923584461212158 lr:2.2187921783513444e-05 tokens_per_second_per_gpu:4409.234121422955 mfu:20.79910964366718 +Step 24995 | loss:0.408170223236084 lr:2.218431839829112e-05 tokens_per_second_per_gpu:5318.679568019491 mfu:25.089118982656586 +Step 24996 | loss:0.4505886733531952 lr:2.21807152222741e-05 tokens_per_second_per_gpu:4913.689705353741 mfu:23.17871270206647 +Step 24997 | loss:0.4988443851470947 lr:2.2177112255489458e-05 tokens_per_second_per_gpu:3909.7981639114923 mfu:18.44318501952442 +Step 24998 | loss:0.4892120957374573 lr:2.2173509497964322e-05 tokens_per_second_per_gpu:4750.992918647927 mfu:22.411244200240922 +Step 24999 | loss:0.34324169158935547 lr:2.2169906949725754e-05 tokens_per_second_per_gpu:3760.4811651088135 mfu:17.738831260065176 +Step 25000 | loss:0.4412592053413391 lr:2.2166304610800865e-05 tokens_per_second_per_gpu:5336.381594545441 mfu:25.172622462056655 +Step 25001 | loss:0.4221212565898895 lr:2.2162702481216767e-05 tokens_per_second_per_gpu:4971.148171485234 mfu:23.449754090233945 +Step 25002 | loss:0.4869745969772339 lr:2.2159100561000522e-05 tokens_per_second_per_gpu:5405.805217716515 mfu:25.50010553744613 +Step 25003 | loss:0.49831289052963257 lr:2.215549885017925e-05 tokens_per_second_per_gpu:4361.788830113291 mfu:20.575302109558066 +Step 25004 | loss:0.45085611939430237 lr:2.215189734878001e-05 tokens_per_second_per_gpu:5050.75328803281 mfu:23.825265007022278 +Step 25005 | loss:0.4650091230869293 lr:2.214829605682992e-05 tokens_per_second_per_gpu:3544.3544401743966 mfu:16.719324623527825 +Step 25006 | loss:0.3651634752750397 lr:2.2144694974356028e-05 tokens_per_second_per_gpu:3335.1011023721016 mfu:15.732240926813466 +Step 25007 | loss:0.3507280945777893 lr:2.214109410138544e-05 tokens_per_second_per_gpu:4884.541074387894 mfu:23.0412136365313 +Step 25008 | loss:0.40840989351272583 lr:2.213749343794524e-05 tokens_per_second_per_gpu:5054.529068688674 mfu:23.843076008590597 +Step 25009 | loss:0.4764057397842407 lr:2.2133892984062522e-05 tokens_per_second_per_gpu:3636.2273131011884 mfu:17.15270464025199 +Step 25010 | loss:0.3491814434528351 lr:2.2130292739764334e-05 tokens_per_second_per_gpu:5346.595325657662 mfu:25.220802374355227 +Step 25011 | loss:0.3000237047672272 lr:2.212669270507779e-05 tokens_per_second_per_gpu:5332.659435753338 mfu:25.15506440396853 +Step 25012 | loss:0.46911951899528503 lr:2.212309288002993e-05 tokens_per_second_per_gpu:5334.988225343035 mfu:25.166049701795608 +Step 25013 | loss:0.39579614996910095 lr:2.211949326464786e-05 tokens_per_second_per_gpu:4237.398951361423 mfu:19.988533828383737 +Step 25014 | loss:0.4492773115634918 lr:2.2115893858958626e-05 tokens_per_second_per_gpu:5351.270546259592 mfu:25.24285618760942 +Step 25015 | loss:0.31850382685661316 lr:2.211229466298931e-05 tokens_per_second_per_gpu:3397.4072892440145 mfu:16.026149840819944 +Step 25016 | loss:0.4388800263404846 lr:2.2108695676767e-05 tokens_per_second_per_gpu:4349.339401230098 mfu:20.51657602942445 +Step 25017 | loss:0.3732937276363373 lr:2.2105096900318728e-05 tokens_per_second_per_gpu:5351.566368146257 mfu:25.244251629921237 +Step 25018 | loss:0.3951016664505005 lr:2.2101498333671604e-05 tokens_per_second_per_gpu:3933.803321546981 mfu:18.556421443793244 +Step 25019 | loss:0.42513880133628845 lr:2.2097899976852654e-05 tokens_per_second_per_gpu:4206.656429777994 mfu:19.84351610885965 +Step 25020 | loss:0.3391539454460144 lr:2.2094301829888976e-05 tokens_per_second_per_gpu:5357.761528106212 mfu:25.27347525645563 +Step 25021 | loss:0.3439303934574127 lr:2.2090703892807597e-05 tokens_per_second_per_gpu:5306.292374240983 mfu:25.03068647613017 +Step 25022 | loss:0.5054872632026672 lr:2.208710616563559e-05 tokens_per_second_per_gpu:4312.492209855089 mfu:20.342761540929413 +Step 25023 | loss:0.4647041857242584 lr:2.2083508648400043e-05 tokens_per_second_per_gpu:5334.17921076142 mfu:25.16223344198201 +Step 25024 | loss:0.47311481833457947 lr:2.207991134112797e-05 tokens_per_second_per_gpu:5346.941911698471 mfu:25.2224372798434 +Step 25025 | loss:0.3616029918193817 lr:2.2076314243846463e-05 tokens_per_second_per_gpu:5333.1603527735615 mfu:25.157427315018182 +Step 25026 | loss:0.3982037603855133 lr:2.2072717356582544e-05 tokens_per_second_per_gpu:4592.792591309006 mfu:21.664986264424083 +Step 25027 | loss:0.3756842315196991 lr:2.2069120679363297e-05 tokens_per_second_per_gpu:4703.023399366868 mfu:22.184963793348228 +Step 25028 | loss:0.4270181953907013 lr:2.2065524212215743e-05 tokens_per_second_per_gpu:4348.998323579364 mfu:20.514967107952092 +Step 25029 | loss:0.5664120316505432 lr:2.2061927955166945e-05 tokens_per_second_per_gpu:4804.604013200679 mfu:22.664136880242197 +Step 25030 | loss:0.45402562618255615 lr:2.2058331908243966e-05 tokens_per_second_per_gpu:3503.870042151448 mfu:16.528352810703012 +Step 25031 | loss:0.40391620993614197 lr:2.205473607147382e-05 tokens_per_second_per_gpu:4743.551389586562 mfu:22.376141238002685 +Step 25032 | loss:0.3795228600502014 lr:2.2051140444883594e-05 tokens_per_second_per_gpu:4530.565456270046 mfu:21.371450251401278 +Step 25033 | loss:0.49899667501449585 lr:2.2047545028500282e-05 tokens_per_second_per_gpu:3267.109102995615 mfu:15.411511065123266 +Step 25034 | loss:0.32470831274986267 lr:2.2043949822350975e-05 tokens_per_second_per_gpu:3711.147949476269 mfu:17.50611806481193 +Step 25035 | loss:0.3762783110141754 lr:2.2040354826462668e-05 tokens_per_second_per_gpu:4387.734443464395 mfu:20.697691994513338 +Step 25036 | loss:0.46030405163764954 lr:2.2036760040862414e-05 tokens_per_second_per_gpu:4154.32599953394 mfu:19.59666454090637 +Step 25037 | loss:0.29348018765449524 lr:2.2033165465577282e-05 tokens_per_second_per_gpu:3919.7317166302546 mfu:18.490043282537684 +Step 25038 | loss:0.5130884647369385 lr:2.202957110063426e-05 tokens_per_second_per_gpu:4094.701313065036 mfu:19.31540472181203 +Step 25039 | loss:0.3368930220603943 lr:2.2025976946060423e-05 tokens_per_second_per_gpu:3596.846252730222 mfu:16.966937459380393 +Step 25040 | loss:0.3666837215423584 lr:2.2022383001882762e-05 tokens_per_second_per_gpu:3796.713625989626 mfu:17.90974595993525 +Step 25041 | loss:0.3746187686920166 lr:2.2018789268128347e-05 tokens_per_second_per_gpu:5000.431860385393 mfu:23.58789024708852 +Step 25042 | loss:0.3529643714427948 lr:2.201519574482417e-05 tokens_per_second_per_gpu:3952.323406801708 mfu:18.643783845792175 +Step 25043 | loss:0.3641350269317627 lr:2.201160243199728e-05 tokens_per_second_per_gpu:5356.979943098196 mfu:25.269788386620665 +Step 25044 | loss:0.34933286905288696 lr:2.2008009329674694e-05 tokens_per_second_per_gpu:4587.691431117622 mfu:21.640923221453654 +Step 25045 | loss:0.42474743723869324 lr:2.2004416437883463e-05 tokens_per_second_per_gpu:4884.945428478089 mfu:23.04312104376467 +Step 25046 | loss:0.39257189631462097 lr:2.200082375665057e-05 tokens_per_second_per_gpu:5007.40446321295 mfu:23.620781204274824 +Step 25047 | loss:0.4281596839427948 lr:2.199723128600307e-05 tokens_per_second_per_gpu:4703.422903674729 mfu:22.18684832333087 +Step 25048 | loss:0.4998934864997864 lr:2.1993639025967956e-05 tokens_per_second_per_gpu:4399.243527503069 mfu:20.75198230757599 +Step 25049 | loss:0.39321523904800415 lr:2.1990046976572264e-05 tokens_per_second_per_gpu:4745.513909562466 mfu:22.38539877956946 +Step 25050 | loss:0.3074607253074646 lr:2.198645513784299e-05 tokens_per_second_per_gpu:4008.587235717322 mfu:18.909189926386215 +Step 25051 | loss:0.3522905111312866 lr:2.1982863509807167e-05 tokens_per_second_per_gpu:3894.447898905165 mfu:18.370775200464298 +Step 25052 | loss:0.48950016498565674 lr:2.1979272092491815e-05 tokens_per_second_per_gpu:4426.730302220333 mfu:20.88164211364468 +Step 25053 | loss:0.44919005036354065 lr:2.1975680885923912e-05 tokens_per_second_per_gpu:4273.890827355746 mfu:20.160672233605023 +Step 25054 | loss:0.3933410346508026 lr:2.1972089890130508e-05 tokens_per_second_per_gpu:5357.568896850141 mfu:25.272566581954457 +Step 25055 | loss:0.37594926357269287 lr:2.196849910513858e-05 tokens_per_second_per_gpu:4229.638363203929 mfu:19.95192580995196 +Step 25056 | loss:0.4386107623577118 lr:2.196490853097516e-05 tokens_per_second_per_gpu:5330.172966581749 mfu:25.143335304650762 +Step 25057 | loss:0.34484928846359253 lr:2.1961318167667226e-05 tokens_per_second_per_gpu:4872.832385429681 mfu:22.985981753007078 +Step 25058 | loss:0.49637866020202637 lr:2.195772801524179e-05 tokens_per_second_per_gpu:5329.900243832843 mfu:25.142048824912873 +Step 25059 | loss:0.47938084602355957 lr:2.1954138073725884e-05 tokens_per_second_per_gpu:4294.638707986829 mfu:20.258543526495306 +Step 25060 | loss:0.3866285979747772 lr:2.1950548343146465e-05 tokens_per_second_per_gpu:4819.168218182476 mfu:22.732838720050935 +Step 25061 | loss:0.4864030182361603 lr:2.1946958823530568e-05 tokens_per_second_per_gpu:3699.325313720679 mfu:17.45034867480315 +Step 25062 | loss:0.4566080570220947 lr:2.1943369514905155e-05 tokens_per_second_per_gpu:3981.089571537553 mfu:18.779478752864957 +Step 25063 | loss:0.5356726050376892 lr:2.1939780417297258e-05 tokens_per_second_per_gpu:4901.226671462041 mfu:23.11992244478647 +Step 25064 | loss:0.3926908075809479 lr:2.1936191530733833e-05 tokens_per_second_per_gpu:4465.034954095263 mfu:21.062331691986245 +Step 25065 | loss:0.47889789938926697 lr:2.1932602855241895e-05 tokens_per_second_per_gpu:5317.534905638709 mfu:25.08371941490689 +Step 25066 | loss:0.511810302734375 lr:2.192901439084845e-05 tokens_per_second_per_gpu:5016.319809263408 mfu:23.662836412709723 +Step 25067 | loss:0.4813631772994995 lr:2.192542613758045e-05 tokens_per_second_per_gpu:3995.262652400775 mfu:18.84633559347372 +Step 25068 | loss:0.5118376612663269 lr:2.1921838095464924e-05 tokens_per_second_per_gpu:3442.112619029755 mfu:16.23703250893507 +Step 25069 | loss:0.4105086326599121 lr:2.1918250264528816e-05 tokens_per_second_per_gpu:4507.0910955476875 mfu:21.26071768673498 +Step 25070 | loss:0.47183042764663696 lr:2.191466264479915e-05 tokens_per_second_per_gpu:4322.935402646708 mfu:20.392023863119903 +Step 25071 | loss:0.38201063871383667 lr:2.191107523630287e-05 tokens_per_second_per_gpu:4744.390864201318 mfu:22.38010118298963 +Step 25072 | loss:0.42856600880622864 lr:2.1907488039066976e-05 tokens_per_second_per_gpu:4767.300807413008 mfu:22.488171293958548 +Step 25073 | loss:0.38398584723472595 lr:2.190390105311847e-05 tokens_per_second_per_gpu:3845.384032661638 mfu:18.139332572234792 +Step 25074 | loss:0.3275856673717499 lr:2.190031427848429e-05 tokens_per_second_per_gpu:4491.899381777423 mfu:21.189055780906994 +Step 25075 | loss:0.4430456757545471 lr:2.1896727715191433e-05 tokens_per_second_per_gpu:4620.724983314544 mfu:21.796748123271716 +Step 25076 | loss:0.30303698778152466 lr:2.1893141363266896e-05 tokens_per_second_per_gpu:5095.441551809932 mfu:24.036067171862985 +Step 25077 | loss:0.3185056746006012 lr:2.188955522273762e-05 tokens_per_second_per_gpu:4853.634292161386 mfu:22.895421071528055 +Step 25078 | loss:0.42889997363090515 lr:2.188596929363057e-05 tokens_per_second_per_gpu:4422.986293725879 mfu:20.863980986782618 +Step 25079 | loss:0.4358617663383484 lr:2.1882383575972737e-05 tokens_per_second_per_gpu:5335.142949602787 mfu:25.166779562527537 +Step 25080 | loss:0.4018072187900543 lr:2.187879806979108e-05 tokens_per_second_per_gpu:3632.5108333507947 mfu:17.135173371172552 +Step 25081 | loss:0.42869964241981506 lr:2.1875212775112592e-05 tokens_per_second_per_gpu:5310.219838549891 mfu:25.0492129953698 +Step 25082 | loss:0.402310848236084 lr:2.1871627691964196e-05 tokens_per_second_per_gpu:4113.788714788458 mfu:19.40544325238796 +Step 25083 | loss:0.37057721614837646 lr:2.1868042820372903e-05 tokens_per_second_per_gpu:4334.225718976449 mfu:20.445282211573893 +Step 25084 | loss:0.48801013827323914 lr:2.1864458160365643e-05 tokens_per_second_per_gpu:4754.247264085593 mfu:22.426595502918982 +Step 25085 | loss:0.3720431327819824 lr:2.1860873711969365e-05 tokens_per_second_per_gpu:5349.601132631896 mfu:25.234981278696242 +Step 25086 | loss:0.36010923981666565 lr:2.1857289475211052e-05 tokens_per_second_per_gpu:5330.873090435855 mfu:25.146637908324028 +Step 25087 | loss:0.4244439899921417 lr:2.185370545011765e-05 tokens_per_second_per_gpu:3840.9253493012893 mfu:18.118300202094204 +Step 25088 | loss:0.4375222325325012 lr:2.185012163671614e-05 tokens_per_second_per_gpu:3994.220496618761 mfu:18.84141956683985 +Step 25089 | loss:0.4820760190486908 lr:2.1846538035033444e-05 tokens_per_second_per_gpu:4380.7244977944065 mfu:20.664624884767942 +Step 25090 | loss:0.40610283613204956 lr:2.1842954645096537e-05 tokens_per_second_per_gpu:4328.103679644279 mfu:20.416403507516424 +Step 25091 | loss:0.33263683319091797 lr:2.183937146693235e-05 tokens_per_second_per_gpu:4904.117267439905 mfu:23.133557879200588 +Step 25092 | loss:0.46064335107803345 lr:2.183578850056786e-05 tokens_per_second_per_gpu:4516.6919697093435 mfu:21.306006648233556 +Step 25093 | loss:0.40988633036613464 lr:2.1832205746029975e-05 tokens_per_second_per_gpu:4976.446312162859 mfu:23.474746323766308 +Step 25094 | loss:0.4281134009361267 lr:2.182862320334567e-05 tokens_per_second_per_gpu:3941.7281220179875 mfu:18.59380407972537 +Step 25095 | loss:0.4142071604728699 lr:2.1825040872541908e-05 tokens_per_second_per_gpu:5351.082782636263 mfu:25.241970474562397 +Step 25096 | loss:0.4174162447452545 lr:2.1821458753645586e-05 tokens_per_second_per_gpu:5295.453749178073 mfu:24.979558832448298 +Step 25097 | loss:0.46828025579452515 lr:2.1817876846683683e-05 tokens_per_second_per_gpu:4427.755603193794 mfu:20.886478633270876 +Step 25098 | loss:0.3147198557853699 lr:2.1814295151683108e-05 tokens_per_second_per_gpu:5350.543087326231 mfu:25.239424639703465 +Step 25099 | loss:0.39165961742401123 lr:2.1810713668670834e-05 tokens_per_second_per_gpu:4200.881501237197 mfu:19.8162747856283 +Step 25100 | loss:0.45629486441612244 lr:2.1807132397673764e-05 tokens_per_second_per_gpu:4194.835112877769 mfu:19.787752940116643 +Step 25101 | loss:0.40737900137901306 lr:2.180355133871884e-05 tokens_per_second_per_gpu:5115.948886248671 mfu:24.13280376732257 +Step 25102 | loss:0.3045979142189026 lr:2.1799970491833034e-05 tokens_per_second_per_gpu:5309.1893363559 mfu:25.04435194069894 +Step 25103 | loss:0.468597412109375 lr:2.1796389857043227e-05 tokens_per_second_per_gpu:3900.494432388144 mfu:18.3992977305485 +Step 25104 | loss:0.42874929308891296 lr:2.1792809434376365e-05 tokens_per_second_per_gpu:3290.691715817197 mfu:15.522754273411314 +Step 25105 | loss:0.45479339361190796 lr:2.1789229223859404e-05 tokens_per_second_per_gpu:4780.17660055608 mfu:22.54890860705126 +Step 25106 | loss:0.36444318294525146 lr:2.178564922551925e-05 tokens_per_second_per_gpu:4198.79265236882 mfu:19.8064213291218 +Step 25107 | loss:0.36182212829589844 lr:2.178206943938281e-05 tokens_per_second_per_gpu:4751.087505902379 mfu:22.411690384458428 +Step 25108 | loss:0.4050884246826172 lr:2.1778489865477027e-05 tokens_per_second_per_gpu:5234.985113472263 mfu:24.694317960810885 +Step 25109 | loss:0.37856540083885193 lr:2.1774910503828842e-05 tokens_per_second_per_gpu:3246.597501637495 mfu:15.31475434187695 +Step 25110 | loss:0.4371301829814911 lr:2.1771331354465136e-05 tokens_per_second_per_gpu:4210.516939483552 mfu:19.86172679181164 +Step 25111 | loss:0.3640972077846527 lr:2.176775241741285e-05 tokens_per_second_per_gpu:4186.27547706354 mfu:19.747375677556466 +Step 25112 | loss:0.4543733298778534 lr:2.176417369269892e-05 tokens_per_second_per_gpu:5377.876936227834 mfu:25.36836306860785 +Step 25113 | loss:0.3836113214492798 lr:2.176059518035024e-05 tokens_per_second_per_gpu:4302.824195540443 mfu:20.29715586787389 +Step 25114 | loss:0.34442272782325745 lr:2.175701688039371e-05 tokens_per_second_per_gpu:4951.298583154438 mfu:23.356120195388588 +Step 25115 | loss:0.39673566818237305 lr:2.175343879285626e-05 tokens_per_second_per_gpu:4533.075583307489 mfu:21.383290948908737 +Step 25116 | loss:0.46716606616973877 lr:2.1749860917764803e-05 tokens_per_second_per_gpu:5368.142308194167 mfu:25.32244316727423 +Step 25117 | loss:0.45562276244163513 lr:2.1746283255146262e-05 tokens_per_second_per_gpu:3870.1928229117475 mfu:18.256359868660653 +Step 25118 | loss:0.3986746072769165 lr:2.1742705805027514e-05 tokens_per_second_per_gpu:3156.4799740453254 mfu:14.889654588588993 +Step 25119 | loss:0.40835174918174744 lr:2.1739128567435496e-05 tokens_per_second_per_gpu:5353.112061504845 mfu:25.25154292921618 +Step 25120 | loss:0.4574647843837738 lr:2.17355515423971e-05 tokens_per_second_per_gpu:3488.3230188967964 mfu:16.455014849415125 +Step 25121 | loss:0.4159586727619171 lr:2.1731974729939208e-05 tokens_per_second_per_gpu:4003.2842913709 mfu:18.884175033128543 +Step 25122 | loss:0.38250216841697693 lr:2.1728398130088744e-05 tokens_per_second_per_gpu:4178.3616055205075 mfu:19.710044595242262 +Step 25123 | loss:0.3510587811470032 lr:2.1724821742872604e-05 tokens_per_second_per_gpu:4643.14555691136 mfu:21.902509794272206 +Step 25124 | loss:0.4247165024280548 lr:2.1721245568317706e-05 tokens_per_second_per_gpu:4314.880335378744 mfu:20.354026736480723 +Step 25125 | loss:0.4841470420360565 lr:2.1717669606450907e-05 tokens_per_second_per_gpu:3999.0222493519937 mfu:18.864070253746007 +Step 25126 | loss:0.3618558347225189 lr:2.1714093857299146e-05 tokens_per_second_per_gpu:4216.633682853586 mfu:19.890580513912127 +Step 25127 | loss:0.4922470152378082 lr:2.1710518320889278e-05 tokens_per_second_per_gpu:4461.751034768025 mfu:21.046840884225254 +Step 25128 | loss:0.3660221993923187 lr:2.1706942997248225e-05 tokens_per_second_per_gpu:4692.685209773933 mfu:22.13619678916132 +Step 25129 | loss:0.4251609742641449 lr:2.1703367886402848e-05 tokens_per_second_per_gpu:4847.933604853382 mfu:22.86852995685865 +Step 25130 | loss:0.4755478501319885 lr:2.1699792988380052e-05 tokens_per_second_per_gpu:5379.043401941759 mfu:25.373865486400003 +Step 25131 | loss:0.4339912533760071 lr:2.169621830320674e-05 tokens_per_second_per_gpu:4416.956249403092 mfu:20.835536238880383 +Step 25132 | loss:0.3356945812702179 lr:2.1692643830909764e-05 tokens_per_second_per_gpu:3921.824068328802 mfu:18.49991326249174 +Step 25133 | loss:0.48636898398399353 lr:2.1689069571516047e-05 tokens_per_second_per_gpu:5384.723278485121 mfu:25.40065843314191 +Step 25134 | loss:0.4563850164413452 lr:2.1685495525052428e-05 tokens_per_second_per_gpu:4060.8596224747166 mfu:19.15576769330968 +Step 25135 | loss:0.33803364634513855 lr:2.168192169154583e-05 tokens_per_second_per_gpu:4820.839559833053 mfu:22.740722723776994 +Step 25136 | loss:0.38484734296798706 lr:2.1678348071023096e-05 tokens_per_second_per_gpu:4299.748683147805 mfu:20.282648151181295 +Step 25137 | loss:0.46958115696907043 lr:2.1674774663511116e-05 tokens_per_second_per_gpu:5350.067682009384 mfu:25.237182071711384 +Step 25138 | loss:0.4056648910045624 lr:2.1671201469036796e-05 tokens_per_second_per_gpu:3769.8827127187474 mfu:17.78317996421066 +Step 25139 | loss:0.47021448612213135 lr:2.1667628487626956e-05 tokens_per_second_per_gpu:4359.39769190906 mfu:20.564022702678464 +Step 25140 | loss:0.37101083993911743 lr:2.1664055719308505e-05 tokens_per_second_per_gpu:5368.696953269173 mfu:25.32505952272524 +Step 25141 | loss:0.42418888211250305 lr:2.166048316410832e-05 tokens_per_second_per_gpu:4537.209805739811 mfu:21.402792781493076 +Step 25142 | loss:0.37943020462989807 lr:2.165691082205326e-05 tokens_per_second_per_gpu:4822.623324151949 mfu:22.74913704441081 +Step 25143 | loss:0.4969596266746521 lr:2.1653338693170173e-05 tokens_per_second_per_gpu:5329.970830835729 mfu:25.142381795848873 +Step 25144 | loss:0.3263370096683502 lr:2.1649766777485943e-05 tokens_per_second_per_gpu:5346.478978283185 mfu:25.220253543938913 +Step 25145 | loss:0.45426705479621887 lr:2.1646195075027447e-05 tokens_per_second_per_gpu:4643.722338406622 mfu:21.90523057099443 +Step 25146 | loss:0.3040885031223297 lr:2.1642623585821527e-05 tokens_per_second_per_gpu:3389.7277380416435 mfu:15.989924087531817 +Step 25147 | loss:0.4126008152961731 lr:2.1639052309895047e-05 tokens_per_second_per_gpu:4070.8847244695717 mfu:19.203057809878192 +Step 25148 | loss:0.4020976126194 lr:2.1635481247274892e-05 tokens_per_second_per_gpu:5310.008470127355 mfu:25.048215934457524 +Step 25149 | loss:0.419247567653656 lr:2.1631910397987903e-05 tokens_per_second_per_gpu:4441.298591643573 mfu:20.95036322046058 +Step 25150 | loss:0.43427467346191406 lr:2.1628339762060917e-05 tokens_per_second_per_gpu:4706.110279287879 mfu:22.19952513261166 +Step 25151 | loss:0.45646724104881287 lr:2.1624769339520806e-05 tokens_per_second_per_gpu:3737.2846163097256 mfu:17.629409181640543 +Step 25152 | loss:0.34600308537483215 lr:2.1621199130394447e-05 tokens_per_second_per_gpu:4041.7420900673897 mfu:19.065587031156948 +Step 25153 | loss:0.5584568381309509 lr:2.161762913470865e-05 tokens_per_second_per_gpu:4644.90932363603 mfu:21.910829782842114 +Step 25154 | loss:0.364753782749176 lr:2.161405935249029e-05 tokens_per_second_per_gpu:4437.927513488545 mfu:20.93446126963807 +Step 25155 | loss:0.41392937302589417 lr:2.1610489783766226e-05 tokens_per_second_per_gpu:4768.968996173644 mfu:22.49604042496482 +Step 25156 | loss:0.505433201789856 lr:2.1606920428563292e-05 tokens_per_second_per_gpu:4381.392456833243 mfu:20.667775761519696 +Step 25157 | loss:0.4529571235179901 lr:2.1603351286908314e-05 tokens_per_second_per_gpu:4737.600636517762 mfu:22.348070520473986 +Step 25158 | loss:0.4873638153076172 lr:2.159978235882815e-05 tokens_per_second_per_gpu:5314.389659587219 mfu:25.068882752647706 +Step 25159 | loss:0.4496152102947235 lr:2.1596213644349656e-05 tokens_per_second_per_gpu:5025.99616351084 mfu:23.70848142665114 +Step 25160 | loss:0.3848106563091278 lr:2.159264514349968e-05 tokens_per_second_per_gpu:5208.374545369649 mfu:24.56879137083976 +Step 25161 | loss:0.4667592942714691 lr:2.1589076856305018e-05 tokens_per_second_per_gpu:4158.485912524293 mfu:19.616287560236206 +Step 25162 | loss:0.42888423800468445 lr:2.1585508782792556e-05 tokens_per_second_per_gpu:3642.9044040106533 mfu:17.184201617301117 +Step 25163 | loss:0.5711369514465332 lr:2.1581940922989096e-05 tokens_per_second_per_gpu:4725.434783494243 mfu:22.290682116053482 +Step 25164 | loss:0.45575687289237976 lr:2.1578373276921497e-05 tokens_per_second_per_gpu:4797.784124910504 mfu:22.63196630358451 +Step 25165 | loss:0.34249940514564514 lr:2.1574805844616565e-05 tokens_per_second_per_gpu:3412.497267016019 mfu:16.097331840586182 +Step 25166 | loss:0.2635393738746643 lr:2.1571238626101142e-05 tokens_per_second_per_gpu:4167.824036847994 mfu:19.660337085919405 +Step 25167 | loss:0.44756487011909485 lr:2.156767162140208e-05 tokens_per_second_per_gpu:4829.485674582201 mfu:22.78150792222802 +Step 25168 | loss:0.29645928740501404 lr:2.1564104830546166e-05 tokens_per_second_per_gpu:4724.284672470499 mfu:22.285256846125304 +Step 25169 | loss:0.4301758110523224 lr:2.1560538253560253e-05 tokens_per_second_per_gpu:4937.227174019112 mfu:23.28974295766729 +Step 25170 | loss:0.41543498635292053 lr:2.155697189047118e-05 tokens_per_second_per_gpu:5343.31657178671 mfu:25.20533593293285 +Step 25171 | loss:0.4174148440361023 lr:2.155340574130575e-05 tokens_per_second_per_gpu:4763.941124871597 mfu:22.472323098189758 +Step 25172 | loss:0.38713547587394714 lr:2.154983980609076e-05 tokens_per_second_per_gpu:4292.557761697133 mfu:20.24872734780076 +Step 25173 | loss:0.4551664888858795 lr:2.1546274084853063e-05 tokens_per_second_per_gpu:4856.558414293891 mfu:22.90921465453376 +Step 25174 | loss:0.37777578830718994 lr:2.1542708577619487e-05 tokens_per_second_per_gpu:3819.8432648704966 mfu:18.01885241285964 +Step 25175 | loss:0.35383474826812744 lr:2.1539143284416807e-05 tokens_per_second_per_gpu:5335.322263063108 mfu:25.167625414713903 +Step 25176 | loss:0.2947438955307007 lr:2.1535578205271866e-05 tokens_per_second_per_gpu:4939.078136827809 mfu:23.298474265042735 +Step 25177 | loss:0.46053892374038696 lr:2.1532013340211493e-05 tokens_per_second_per_gpu:3311.7757181410025 mfu:15.622211109674446 +Step 25178 | loss:0.4166302978992462 lr:2.1528448689262476e-05 tokens_per_second_per_gpu:4038.98616413452 mfu:19.05258685832215 +Step 25179 | loss:0.40293073654174805 lr:2.1524884252451615e-05 tokens_per_second_per_gpu:3384.3360265896276 mfu:15.964490464691824 +Step 25180 | loss:0.4404720962047577 lr:2.152132002980573e-05 tokens_per_second_per_gpu:5297.6575522318735 mfu:24.98995454746065 +Step 25181 | loss:0.4521574378013611 lr:2.151775602135165e-05 tokens_per_second_per_gpu:5337.0050072403055 mfu:25.175563206103508 +Step 25182 | loss:0.4371853768825531 lr:2.1514192227116143e-05 tokens_per_second_per_gpu:5326.970579059591 mfu:25.12822909632509 +Step 25183 | loss:0.3511907160282135 lr:2.1510628647126037e-05 tokens_per_second_per_gpu:3965.4236873251816 mfu:18.705580104159736 +Step 25184 | loss:0.4532375633716583 lr:2.150706528140814e-05 tokens_per_second_per_gpu:3697.177382277117 mfu:17.440216515708975 +Step 25185 | loss:0.361344575881958 lr:2.1503502129989238e-05 tokens_per_second_per_gpu:5003.788881139458 mfu:23.603725886752247 +Step 25186 | loss:0.25975561141967773 lr:2.1499939192896118e-05 tokens_per_second_per_gpu:4353.639955677596 mfu:20.536862478503547 +Step 25187 | loss:0.36094582080841064 lr:2.1496376470155592e-05 tokens_per_second_per_gpu:4391.701171543678 mfu:20.716403727656918 +Step 25188 | loss:0.4953942596912384 lr:2.1492813961794478e-05 tokens_per_second_per_gpu:5242.37554934106 mfu:24.72917990774238 +Step 25189 | loss:0.40416356921195984 lr:2.1489251667839523e-05 tokens_per_second_per_gpu:4851.041181207141 mfu:22.88318892472653 +Step 25190 | loss:0.413975328207016 lr:2.1485689588317548e-05 tokens_per_second_per_gpu:4435.978443619793 mfu:20.92526717452156 +Step 25191 | loss:0.4079788625240326 lr:2.1482127723255358e-05 tokens_per_second_per_gpu:5323.98152024736 mfu:25.114129195921777 +Step 25192 | loss:0.3479773700237274 lr:2.1478566072679725e-05 tokens_per_second_per_gpu:4184.1475277585 mfu:19.737337777618997 +Step 25193 | loss:0.3844713270664215 lr:2.147500463661741e-05 tokens_per_second_per_gpu:3709.3664063055003 mfu:17.497714221712805 +Step 25194 | loss:0.40514346957206726 lr:2.147144341509523e-05 tokens_per_second_per_gpu:5336.038172513884 mfu:25.171002481739965 +Step 25195 | loss:0.4165516197681427 lr:2.1467882408139972e-05 tokens_per_second_per_gpu:5292.155783547123 mfu:24.964001765876148 +Step 25196 | loss:0.3536589741706848 lr:2.1464321615778422e-05 tokens_per_second_per_gpu:3856.487178844652 mfu:18.191708007172153 +Step 25197 | loss:0.4588996171951294 lr:2.146076103803733e-05 tokens_per_second_per_gpu:5303.517519424014 mfu:25.017597012519474 +Step 25198 | loss:0.5133633017539978 lr:2.14572006749435e-05 tokens_per_second_per_gpu:3853.2810806711364 mfu:18.176584295070924 +Step 25199 | loss:0.39048635959625244 lr:2.145364052652372e-05 tokens_per_second_per_gpu:5065.849789826773 mfu:23.89647778171321 +Step 25200 | loss:0.41942185163497925 lr:2.145008059280476e-05 tokens_per_second_per_gpu:4159.523957262982 mfu:19.6211841943773 +Step 25201 | loss:0.35156089067459106 lr:2.1446520873813364e-05 tokens_per_second_per_gpu:4669.927977557001 mfu:22.028847042009858 +Step 25202 | loss:0.49036091566085815 lr:2.1442961369576326e-05 tokens_per_second_per_gpu:4328.060562613012 mfu:20.416200117123417 +Step 25203 | loss:0.33037397265434265 lr:2.1439402080120442e-05 tokens_per_second_per_gpu:3721.7439645911345 mfu:17.55610127597535 +Step 25204 | loss:0.30094003677368164 lr:2.1435843005472432e-05 tokens_per_second_per_gpu:3674.8879828052977 mfu:17.335073615436702 +Step 25205 | loss:0.39102309942245483 lr:2.1432284145659104e-05 tokens_per_second_per_gpu:4364.384780134004 mfu:20.587547648720516 +Step 25206 | loss:0.4317309856414795 lr:2.142872550070722e-05 tokens_per_second_per_gpu:5369.847644707246 mfu:25.330487530566835 +Step 25207 | loss:0.42980673909187317 lr:2.1425167070643543e-05 tokens_per_second_per_gpu:5143.8253106307875 mfu:24.264301617341456 +Step 25208 | loss:0.4411388039588928 lr:2.1421608855494813e-05 tokens_per_second_per_gpu:5336.374036713947 mfu:25.172586810476005 +Step 25209 | loss:0.5285815000534058 lr:2.141805085528781e-05 tokens_per_second_per_gpu:5342.220432109542 mfu:25.200165255054937 +Step 25210 | loss:0.3355005979537964 lr:2.1414493070049307e-05 tokens_per_second_per_gpu:5050.486541263025 mfu:23.82400671699751 +Step 25211 | loss:0.5682898163795471 lr:2.141093549980603e-05 tokens_per_second_per_gpu:3371.0235879201327 mfu:15.901693420151565 +Step 25212 | loss:0.37824153900146484 lr:2.140737814458476e-05 tokens_per_second_per_gpu:4763.790345284994 mfu:22.471611845154108 +Step 25213 | loss:0.3902103006839752 lr:2.140382100441226e-05 tokens_per_second_per_gpu:4374.257626866841 mfu:20.63411955124066 +Step 25214 | loss:0.4577792286872864 lr:2.1400264079315278e-05 tokens_per_second_per_gpu:3573.858482853097 mfu:16.858500226752017 +Step 25215 | loss:0.3643917143344879 lr:2.139670736932054e-05 tokens_per_second_per_gpu:4700.829933773283 mfu:22.174616841899468 +Step 25216 | loss:0.4372575283050537 lr:2.139315087445481e-05 tokens_per_second_per_gpu:4093.559722500417 mfu:19.31001964434329 +Step 25217 | loss:0.4420672655105591 lr:2.1389594594744864e-05 tokens_per_second_per_gpu:5342.894628175065 mfu:25.203345552925718 +Step 25218 | loss:0.5263558030128479 lr:2.1386038530217413e-05 tokens_per_second_per_gpu:4235.083520312543 mfu:19.977611545072897 +Step 25219 | loss:0.37955936789512634 lr:2.1382482680899213e-05 tokens_per_second_per_gpu:3851.849072282301 mfu:18.169829267174965 +Step 25220 | loss:0.38727498054504395 lr:2.1378927046817032e-05 tokens_per_second_per_gpu:4141.323679665835 mfu:19.535330379663993 +Step 25221 | loss:0.3087339997291565 lr:2.1375371627997592e-05 tokens_per_second_per_gpu:4631.235811926664 mfu:21.846329495167367 +Step 25222 | loss:0.405593603849411 lr:2.1371816424467615e-05 tokens_per_second_per_gpu:5317.459066713642 mfu:25.083361669757895 +Step 25223 | loss:0.39752498269081116 lr:2.1368261436253867e-05 tokens_per_second_per_gpu:3876.081502963583 mfu:18.284137777177577 +Step 25224 | loss:0.36687374114990234 lr:2.1364706663383087e-05 tokens_per_second_per_gpu:4834.508361222573 mfu:22.805200792069364 +Step 25225 | loss:0.39259010553359985 lr:2.1361152105881988e-05 tokens_per_second_per_gpu:4201.179577299706 mfu:19.817680861272173 +Step 25226 | loss:0.463766485452652 lr:2.1357597763777322e-05 tokens_per_second_per_gpu:4787.5454400933795 mfu:22.58366868040237 +Step 25227 | loss:0.47778865694999695 lr:2.135404363709581e-05 tokens_per_second_per_gpu:4493.364827040177 mfu:21.19596853624215 +Step 25228 | loss:0.4317840337753296 lr:2.1350489725864232e-05 tokens_per_second_per_gpu:4723.21586134014 mfu:22.28021508166542 +Step 25229 | loss:0.3975341320037842 lr:2.1346936030109242e-05 tokens_per_second_per_gpu:5320.067963311579 mfu:25.095668280134028 +Step 25230 | loss:0.32930150628089905 lr:2.1343382549857598e-05 tokens_per_second_per_gpu:3999.8160102610573 mfu:18.867814559383635 +Step 25231 | loss:0.3911086916923523 lr:2.133982928513603e-05 tokens_per_second_per_gpu:4131.417872537968 mfu:19.48860299733687 +Step 25232 | loss:0.36745449900627136 lr:2.1336276235971286e-05 tokens_per_second_per_gpu:5349.814501862378 mfu:25.235987777760972 +Step 25233 | loss:0.4667437970638275 lr:2.133272340239005e-05 tokens_per_second_per_gpu:4420.10791353504 mfu:20.85040317632013 +Step 25234 | loss:0.376726359128952 lr:2.132917078441905e-05 tokens_per_second_per_gpu:4729.889421115084 mfu:22.311695401750537 +Step 25235 | loss:0.3402191400527954 lr:2.132561838208504e-05 tokens_per_second_per_gpu:5308.380406937227 mfu:25.040536082613468 +Step 25236 | loss:0.5627798438072205 lr:2.132206619541471e-05 tokens_per_second_per_gpu:4861.822373487071 mfu:22.93404564817204 +Step 25237 | loss:0.5083541870117188 lr:2.131851422443476e-05 tokens_per_second_per_gpu:5371.80369236946 mfu:25.33971454112564 +Step 25238 | loss:0.505702793598175 lr:2.131496246917193e-05 tokens_per_second_per_gpu:4407.044604629627 mfu:20.78878131938297 +Step 25239 | loss:0.46222779154777527 lr:2.1311410929652942e-05 tokens_per_second_per_gpu:4295.777520652714 mfu:20.26391550014176 +Step 25240 | loss:0.40389537811279297 lr:2.1307859605904474e-05 tokens_per_second_per_gpu:3362.3191383316394 mfu:15.8606330463104 +Step 25241 | loss:0.48718029260635376 lr:2.130430849795326e-05 tokens_per_second_per_gpu:5369.9964606249 mfu:25.331189520641022 +Step 25242 | loss:0.35202261805534363 lr:2.1300757605826018e-05 tokens_per_second_per_gpu:4430.444102306211 mfu:20.899160742289396 +Step 25243 | loss:0.3330177068710327 lr:2.1297206929549436e-05 tokens_per_second_per_gpu:4303.55286053962 mfu:20.30059310499808 +Step 25244 | loss:0.5011687874794006 lr:2.129365646915021e-05 tokens_per_second_per_gpu:3308.3488612747938 mfu:15.606046041154467 +Step 25245 | loss:0.43372318148612976 lr:2.1290106224655053e-05 tokens_per_second_per_gpu:5329.816028574456 mfu:25.141651567170456 +Step 25246 | loss:0.3217042088508606 lr:2.128655619609069e-05 tokens_per_second_per_gpu:4870.599478039507 mfu:22.97544874787434 +Step 25247 | loss:0.34171614050865173 lr:2.1283006383483788e-05 tokens_per_second_per_gpu:5066.495231199776 mfu:23.899522438793316 +Step 25248 | loss:0.41150587797164917 lr:2.1279456786861056e-05 tokens_per_second_per_gpu:5344.77838856075 mfu:25.212231572067655 +Step 25249 | loss:0.42986199259757996 lr:2.1275907406249212e-05 tokens_per_second_per_gpu:5303.149401619593 mfu:25.015860538028647 +Step 25250 | loss:0.33812853693962097 lr:2.1272358241674934e-05 tokens_per_second_per_gpu:4743.778989485638 mfu:22.37721486555917 +Step 25251 | loss:0.4162740707397461 lr:2.12688092931649e-05 tokens_per_second_per_gpu:5357.130663318132 mfu:25.270499359613623 +Step 25252 | loss:0.42374706268310547 lr:2.1265260560745808e-05 tokens_per_second_per_gpu:5303.385329913943 mfu:25.016973452046614 +Step 25253 | loss:0.417688250541687 lr:2.1261712044444388e-05 tokens_per_second_per_gpu:4410.113142484156 mfu:20.803256135989322 +Step 25254 | loss:0.36598142981529236 lr:2.1258163744287273e-05 tokens_per_second_per_gpu:5227.141881152206 mfu:24.657320095763755 +Step 25255 | loss:0.5183243155479431 lr:2.125461566030118e-05 tokens_per_second_per_gpu:4874.195181445491 mfu:22.99241029433063 +Step 25256 | loss:0.4421999454498291 lr:2.12510677925128e-05 tokens_per_second_per_gpu:5373.921812678363 mfu:25.34970607601133 +Step 25257 | loss:0.3037355840206146 lr:2.1247520140948818e-05 tokens_per_second_per_gpu:3377.7281077435264 mfu:15.933319784067551 +Step 25258 | loss:0.3272647261619568 lr:2.1243972705635883e-05 tokens_per_second_per_gpu:5296.71840158438 mfu:24.9855244136208 +Step 25259 | loss:0.47530636191368103 lr:2.12404254866007e-05 tokens_per_second_per_gpu:5363.882701030174 mfu:25.302349873517752 +Step 25260 | loss:0.46387752890586853 lr:2.1236878483869966e-05 tokens_per_second_per_gpu:3868.8369145266042 mfu:18.249963817466103 +Step 25261 | loss:0.5045118927955627 lr:2.123333169747032e-05 tokens_per_second_per_gpu:4609.6569546932 mfu:21.744538343864424 +Step 25262 | loss:0.4416445791721344 lr:2.122978512742846e-05 tokens_per_second_per_gpu:3975.376977775487 mfu:18.75253147342029 +Step 25263 | loss:0.3639339506626129 lr:2.1226238773771052e-05 tokens_per_second_per_gpu:5333.75212074239 mfu:25.16021878549317 +Step 25264 | loss:0.38705769181251526 lr:2.122269263652481e-05 tokens_per_second_per_gpu:4267.755496537595 mfu:20.131730831340423 +Step 25265 | loss:0.4916686713695526 lr:2.1219146715716332e-05 tokens_per_second_per_gpu:3935.995352322303 mfu:18.566761626959458 +Step 25266 | loss:0.3337846100330353 lr:2.1215601011372325e-05 tokens_per_second_per_gpu:4597.634760699631 mfu:21.687827603598958 +Step 25267 | loss:0.4128642976284027 lr:2.1212055523519457e-05 tokens_per_second_per_gpu:3357.1746636880557 mfu:15.836365681678734 +Step 25268 | loss:0.4627258777618408 lr:2.120851025218441e-05 tokens_per_second_per_gpu:4346.7786867702125 mfu:20.504496702414233 +Step 25269 | loss:0.3437941372394562 lr:2.1204965197393817e-05 tokens_per_second_per_gpu:3747.5811896578584 mfu:17.677979874900252 +Step 25270 | loss:0.4545527994632721 lr:2.120142035917435e-05 tokens_per_second_per_gpu:3860.6432577110913 mfu:18.21131294028532 +Step 25271 | loss:0.5660484433174133 lr:2.1197875737552696e-05 tokens_per_second_per_gpu:5364.555197961221 mfu:25.305522156281306 +Step 25272 | loss:0.4756942689418793 lr:2.1194331332555495e-05 tokens_per_second_per_gpu:4247.9661475699395 mfu:20.038381095848894 +Step 25273 | loss:0.4379006028175354 lr:2.1190787144209384e-05 tokens_per_second_per_gpu:4198.298772932748 mfu:19.804091615557322 +Step 25274 | loss:0.38865208625793457 lr:2.118724317254104e-05 tokens_per_second_per_gpu:5357.042004681533 mfu:25.270081141698064 +Step 25275 | loss:0.40243610739707947 lr:2.118369941757713e-05 tokens_per_second_per_gpu:5206.367325417884 mfu:24.559322971860247 +Step 25276 | loss:0.4135446548461914 lr:2.1180155879344277e-05 tokens_per_second_per_gpu:3353.633005881397 mfu:15.819659077535592 +Step 25277 | loss:0.4691295027732849 lr:2.1176612557869153e-05 tokens_per_second_per_gpu:4451.434697799835 mfu:20.99817696259771 +Step 25278 | loss:0.2863360345363617 lr:2.1173069453178416e-05 tokens_per_second_per_gpu:4822.271164694523 mfu:22.74747584816553 +Step 25279 | loss:0.4109630584716797 lr:2.116952656529871e-05 tokens_per_second_per_gpu:5038.311724182033 mfu:23.766576027588478 +Step 25280 | loss:0.4233401119709015 lr:2.1165983894256647e-05 tokens_per_second_per_gpu:4238.908236275459 mfu:19.995653382833805 +Step 25281 | loss:0.42207369208335876 lr:2.1162441440078907e-05 tokens_per_second_per_gpu:4551.789246707733 mfu:21.4715664920479 +Step 25282 | loss:0.45805492997169495 lr:2.1158899202792138e-05 tokens_per_second_per_gpu:4209.77463756831 mfu:19.85822522702768 +Step 25283 | loss:0.32269516587257385 lr:2.1155357182422948e-05 tokens_per_second_per_gpu:5076.27451914926 mfu:23.94565300856961 +Step 25284 | loss:0.4751726984977722 lr:2.1151815378997998e-05 tokens_per_second_per_gpu:4885.5043102041745 mfu:23.045757384221673 +Step 25285 | loss:0.44671401381492615 lr:2.114827379254395e-05 tokens_per_second_per_gpu:4476.447084180584 mfu:21.11616465670888 +Step 25286 | loss:0.35552170872688293 lr:2.114473242308741e-05 tokens_per_second_per_gpu:4300.502452729023 mfu:20.286203810902254 +Step 25287 | loss:0.46768465638160706 lr:2.1141191270655005e-05 tokens_per_second_per_gpu:4478.067733932831 mfu:21.12380953810159 +Step 25288 | loss:0.4127923548221588 lr:2.113765033527338e-05 tokens_per_second_per_gpu:4174.180763019647 mfu:19.690322847840417 +Step 25289 | loss:0.463270902633667 lr:2.1134109616969183e-05 tokens_per_second_per_gpu:4248.451675097634 mfu:20.040671412036566 +Step 25290 | loss:0.280473917722702 lr:2.1130569115769023e-05 tokens_per_second_per_gpu:4095.1803754277894 mfu:19.317664540713604 +Step 25291 | loss:0.3807247579097748 lr:2.112702883169953e-05 tokens_per_second_per_gpu:5326.350640335871 mfu:25.125304739592845 +Step 25292 | loss:0.34407126903533936 lr:2.112348876478736e-05 tokens_per_second_per_gpu:4139.264193239763 mfu:19.52561543080759 +Step 25293 | loss:0.49901261925697327 lr:2.1119948915059107e-05 tokens_per_second_per_gpu:4779.4434122606735 mfu:22.545450032767004 +Step 25294 | loss:0.3653256595134735 lr:2.1116409282541393e-05 tokens_per_second_per_gpu:4225.38991821991 mfu:19.931885170102884 +Step 25295 | loss:0.4542454183101654 lr:2.1112869867260847e-05 tokens_per_second_per_gpu:5349.301643700074 mfu:25.23356853830676 +Step 25296 | loss:0.5180383324623108 lr:2.1109330669244116e-05 tokens_per_second_per_gpu:3809.544636880807 mfu:17.970272027505153 +Step 25297 | loss:0.41932380199432373 lr:2.1105791688517774e-05 tokens_per_second_per_gpu:5315.43193321293 mfu:25.073799335170403 +Step 25298 | loss:0.4613328278064728 lr:2.1102252925108458e-05 tokens_per_second_per_gpu:3863.0607026347284 mfu:18.222716440448753 +Step 25299 | loss:0.4321180284023285 lr:2.1098714379042782e-05 tokens_per_second_per_gpu:5344.479803054235 mfu:25.210823093289488 +Step 25300 | loss:0.37234628200531006 lr:2.109517605034741e-05 tokens_per_second_per_gpu:5355.656690997585 mfu:25.263546380691995 +Step 25301 | loss:0.39905107021331787 lr:2.1091637939048865e-05 tokens_per_second_per_gpu:5348.577996604394 mfu:25.230154971489508 +Step 25302 | loss:0.3536987900733948 lr:2.1088100045173797e-05 tokens_per_second_per_gpu:4663.999543625447 mfu:22.000881607659583 +Step 25303 | loss:0.40890762209892273 lr:2.1084562368748818e-05 tokens_per_second_per_gpu:5364.8266574241525 mfu:25.306802676883716 +Step 25304 | loss:0.44636017084121704 lr:2.1081024909800558e-05 tokens_per_second_per_gpu:5323.927574546688 mfu:25.11387472484762 +Step 25305 | loss:0.4218401610851288 lr:2.107748766835558e-05 tokens_per_second_per_gpu:3854.4873532398324 mfu:18.182274488588785 +Step 25306 | loss:0.3479040861129761 lr:2.1073950644440503e-05 tokens_per_second_per_gpu:4561.643300674063 mfu:21.51804974588205 +Step 25307 | loss:0.4230145215988159 lr:2.1070413838081977e-05 tokens_per_second_per_gpu:4455.281083973316 mfu:21.016321022439172 +Step 25308 | loss:0.32765477895736694 lr:2.106687724930652e-05 tokens_per_second_per_gpu:5299.712596246545 mfu:24.999648540704523 +Step 25309 | loss:0.3943372964859009 lr:2.106334087814077e-05 tokens_per_second_per_gpu:4033.7249888808806 mfu:19.027768997001676 +Step 25310 | loss:0.41236603260040283 lr:2.1059804724611325e-05 tokens_per_second_per_gpu:4055.275073429446 mfu:19.12942442264082 +Step 25311 | loss:0.4061238169670105 lr:2.1056268788744805e-05 tokens_per_second_per_gpu:5040.96950913788 mfu:23.779113252690077 +Step 25312 | loss:0.4510749578475952 lr:2.105273307056776e-05 tokens_per_second_per_gpu:4416.32229068053 mfu:20.832545747421545 +Step 25313 | loss:0.5439252853393555 lr:2.1049197570106804e-05 tokens_per_second_per_gpu:4907.473877288724 mfu:23.14939158055416 +Step 25314 | loss:0.40232422947883606 lr:2.1045662287388545e-05 tokens_per_second_per_gpu:4595.982595600936 mfu:21.68003405893998 +Step 25315 | loss:0.45871803164482117 lr:2.104212722243955e-05 tokens_per_second_per_gpu:4322.991436913222 mfu:20.39228818631548 +Step 25316 | loss:0.45308971405029297 lr:2.1038592375286398e-05 tokens_per_second_per_gpu:4539.880948571711 mfu:21.415393017975113 +Step 25317 | loss:0.6036180257797241 lr:2.103505774595569e-05 tokens_per_second_per_gpu:5356.195384885003 mfu:25.266087491669598 +Step 25318 | loss:0.39462754130363464 lr:2.1031523334474025e-05 tokens_per_second_per_gpu:5315.590417495892 mfu:25.07454693257334 +Step 25319 | loss:0.4992981553077698 lr:2.1027989140867948e-05 tokens_per_second_per_gpu:4681.516479648 mfu:22.08351198357544 +Step 25320 | loss:0.39257460832595825 lr:2.1024455165164064e-05 tokens_per_second_per_gpu:4169.542204850649 mfu:19.66844197753762 +Step 25321 | loss:0.41461148858070374 lr:2.1020921407388964e-05 tokens_per_second_per_gpu:4607.77044890946 mfu:21.73563937421145 +Step 25322 | loss:0.42484718561172485 lr:2.101738786756922e-05 tokens_per_second_per_gpu:3912.7817541031086 mfu:18.457259123511253 +Step 25323 | loss:0.4777641296386719 lr:2.1013854545731378e-05 tokens_per_second_per_gpu:5139.339661056645 mfu:24.24314204296949 +Step 25324 | loss:0.4512811601161957 lr:2.101032144190203e-05 tokens_per_second_per_gpu:5342.935688835044 mfu:25.203539243064508 +Step 25325 | loss:0.3280113637447357 lr:2.1006788556107783e-05 tokens_per_second_per_gpu:4537.199959817078 mfu:21.40274633659566 +Step 25326 | loss:0.4294041395187378 lr:2.100325588837515e-05 tokens_per_second_per_gpu:4595.073390197329 mfu:21.675745181924725 +Step 25327 | loss:0.34744706749916077 lr:2.0999723438730735e-05 tokens_per_second_per_gpu:5203.140982764286 mfu:24.544103763092405 +Step 25328 | loss:0.3926919102668762 lr:2.0996191207201114e-05 tokens_per_second_per_gpu:5325.7253174101525 mfu:25.12235498466114 +Step 25329 | loss:0.49411216378211975 lr:2.0992659193812842e-05 tokens_per_second_per_gpu:3766.209355427305 mfu:17.765852111127547 +Step 25330 | loss:0.3555780351161957 lr:2.0989127398592467e-05 tokens_per_second_per_gpu:5344.694884593092 mfu:25.21183766960497 +Step 25331 | loss:0.46883347630500793 lr:2.098559582156656e-05 tokens_per_second_per_gpu:3432.686344765954 mfu:16.19256716494449 +Step 25332 | loss:0.401679664850235 lr:2.0982064462761714e-05 tokens_per_second_per_gpu:5358.840096718158 mfu:25.27856304862096 +Step 25333 | loss:0.3777625262737274 lr:2.097853332220444e-05 tokens_per_second_per_gpu:5228.456030196749 mfu:24.66351916102381 +Step 25334 | loss:0.39774036407470703 lr:2.0975002399921324e-05 tokens_per_second_per_gpu:5347.26322793862 mfu:25.223952983370385 +Step 25335 | loss:0.3702343702316284 lr:2.0971471695938917e-05 tokens_per_second_per_gpu:4047.972456343559 mfu:19.09497673189174 +Step 25336 | loss:0.534959614276886 lr:2.0967941210283815e-05 tokens_per_second_per_gpu:5276.616679274447 mfu:24.890701159777358 +Step 25337 | loss:0.47517624497413635 lr:2.0964410942982494e-05 tokens_per_second_per_gpu:3724.663016879985 mfu:17.56987094366354 +Step 25338 | loss:0.41956573724746704 lr:2.0960880894061534e-05 tokens_per_second_per_gpu:4391.1577615402975 mfu:20.713840369955665 +Step 25339 | loss:0.3997947871685028 lr:2.0957351063547505e-05 tokens_per_second_per_gpu:3977.2813709730144 mfu:18.76151482105601 +Step 25340 | loss:0.3920755088329315 lr:2.0953821451466958e-05 tokens_per_second_per_gpu:3920.086725793967 mfu:18.491717921333723 +Step 25341 | loss:0.4555061161518097 lr:2.095029205784641e-05 tokens_per_second_per_gpu:4056.530110051385 mfu:19.135344644517836 +Step 25342 | loss:0.32499969005584717 lr:2.094676288271242e-05 tokens_per_second_per_gpu:5329.423194301533 mfu:25.139798500880637 +Step 25343 | loss:0.4898165762424469 lr:2.0943233926091577e-05 tokens_per_second_per_gpu:5341.606955872925 mfu:25.197271382976975 +Step 25344 | loss:0.34033122658729553 lr:2.0939705188010332e-05 tokens_per_second_per_gpu:3909.170447517841 mfu:18.440223974196105 +Step 25345 | loss:0.4351092278957367 lr:2.0936176668495284e-05 tokens_per_second_per_gpu:4127.529320125215 mfu:19.47026003215046 +Step 25346 | loss:0.3272811472415924 lr:2.0932648367572954e-05 tokens_per_second_per_gpu:4090.8038678518956 mfu:19.297019807768923 +Step 25347 | loss:0.3346187472343445 lr:2.09291202852699e-05 tokens_per_second_per_gpu:4233.31095506761 mfu:19.969250052381728 +Step 25348 | loss:0.47700586915016174 lr:2.0925592421612628e-05 tokens_per_second_per_gpu:5389.11785635177 mfu:25.42138840673119 +Step 25349 | loss:0.46861183643341064 lr:2.0922064776627683e-05 tokens_per_second_per_gpu:4434.672868309038 mfu:20.919108553026607 +Step 25350 | loss:0.38902613520622253 lr:2.0918537350341617e-05 tokens_per_second_per_gpu:4684.381315978401 mfu:22.097025905337425 +Step 25351 | loss:0.3743225634098053 lr:2.0915010142780945e-05 tokens_per_second_per_gpu:5187.913851307913 mfu:24.472274785996174 +Step 25352 | loss:0.445339560508728 lr:2.0911483153972173e-05 tokens_per_second_per_gpu:3676.5960837184853 mfu:17.343131018876278 +Step 25353 | loss:0.4306669235229492 lr:2.0907956383941844e-05 tokens_per_second_per_gpu:5353.065014568908 mfu:25.251321000792327 +Step 25354 | loss:0.3236287534236908 lr:2.090442983271651e-05 tokens_per_second_per_gpu:3403.4825136731033 mfu:16.05480771099213 +Step 25355 | loss:0.34076356887817383 lr:2.0900903500322657e-05 tokens_per_second_per_gpu:3776.040685980259 mfu:17.81222817474379 +Step 25356 | loss:0.33510392904281616 lr:2.0897377386786816e-05 tokens_per_second_per_gpu:4229.923737846078 mfu:19.953271970828915 +Step 25357 | loss:0.3982697129249573 lr:2.0893851492135537e-05 tokens_per_second_per_gpu:4413.514069208546 mfu:20.819298887606205 +Step 25358 | loss:0.5372314453125 lr:2.0890325816395305e-05 tokens_per_second_per_gpu:4127.396502871681 mfu:19.46963351050458 +Step 25359 | loss:0.4795110523700714 lr:2.088680035959264e-05 tokens_per_second_per_gpu:4353.4888384480755 mfu:20.536149632748643 +Step 25360 | loss:0.48401308059692383 lr:2.088327512175406e-05 tokens_per_second_per_gpu:5310.346772178639 mfu:25.049811762953716 +Step 25361 | loss:0.3685222864151001 lr:2.0879750102906097e-05 tokens_per_second_per_gpu:3928.743984202176 mfu:18.53255568632584 +Step 25362 | loss:0.5314741134643555 lr:2.0876225303075236e-05 tokens_per_second_per_gpu:4064.5593269389765 mfu:19.17321983050227 +Step 25363 | loss:0.30621734261512756 lr:2.0872700722288004e-05 tokens_per_second_per_gpu:4411.928614827375 mfu:20.81182002878381 +Step 25364 | loss:0.3440262973308563 lr:2.086917636057092e-05 tokens_per_second_per_gpu:5371.998705244744 mfu:25.340634450130985 +Step 25365 | loss:0.5195802450180054 lr:2.0865652217950476e-05 tokens_per_second_per_gpu:3223.6161166362836 mfu:15.20634722779763 +Step 25366 | loss:0.3557719886302948 lr:2.086212829445316e-05 tokens_per_second_per_gpu:5341.621851449612 mfu:25.197341647953653 +Step 25367 | loss:0.47180938720703125 lr:2.08586045901055e-05 tokens_per_second_per_gpu:5324.779225020401 mfu:25.117892105438678 +Step 25368 | loss:0.4195416569709778 lr:2.085508110493401e-05 tokens_per_second_per_gpu:5247.103377891757 mfu:24.75148188166675 +Step 25369 | loss:0.5306566953659058 lr:2.0851557838965153e-05 tokens_per_second_per_gpu:5088.620761396603 mfu:24.003892339736463 +Step 25370 | loss:0.3550575375556946 lr:2.084803479222545e-05 tokens_per_second_per_gpu:5303.96598756467 mfu:25.019712513255506 +Step 25371 | loss:0.4037172794342041 lr:2.0844511964741403e-05 tokens_per_second_per_gpu:4526.8083345371915 mfu:21.353727267156337 +Step 25372 | loss:0.3330460488796234 lr:2.084098935653953e-05 tokens_per_second_per_gpu:5348.870203183847 mfu:25.231533359780368 +Step 25373 | loss:0.4888187050819397 lr:2.0837466967646263e-05 tokens_per_second_per_gpu:4024.461840353451 mfu:18.98407314494168 +Step 25374 | loss:0.42893216013908386 lr:2.083394479808813e-05 tokens_per_second_per_gpu:5352.424175069154 mfu:25.248298051533496 +Step 25375 | loss:0.4024409353733063 lr:2.083042284789164e-05 tokens_per_second_per_gpu:5301.766897755409 mfu:25.009339031421813 +Step 25376 | loss:0.4639444649219513 lr:2.0826901117083235e-05 tokens_per_second_per_gpu:4873.991946331387 mfu:22.991451599621918 +Step 25377 | loss:0.34187397360801697 lr:2.0823379605689442e-05 tokens_per_second_per_gpu:5347.42897677698 mfu:25.224734848920598 +Step 25378 | loss:0.47379666566848755 lr:2.0819858313736723e-05 tokens_per_second_per_gpu:5104.9272307445435 mfu:24.08081273782126 +Step 25379 | loss:0.3538990020751953 lr:2.0816337241251616e-05 tokens_per_second_per_gpu:4386.4786500515775 mfu:20.691768202724628 +Step 25380 | loss:0.40574389696121216 lr:2.0812816388260518e-05 tokens_per_second_per_gpu:4325.171205454743 mfu:20.4025705264327 +Step 25381 | loss:0.44970086216926575 lr:2.080929575478996e-05 tokens_per_second_per_gpu:4871.116978359863 mfu:22.977889885180122 +Step 25382 | loss:0.39098334312438965 lr:2.0805775340866405e-05 tokens_per_second_per_gpu:3741.5204111777557 mfu:17.649390148734167 +Step 25383 | loss:0.4123455584049225 lr:2.0802255146516364e-05 tokens_per_second_per_gpu:4629.405603901442 mfu:21.83769609164671 +Step 25384 | loss:0.43550926446914673 lr:2.0798735171766264e-05 tokens_per_second_per_gpu:5346.314295243494 mfu:25.219476705942864 +Step 25385 | loss:0.40449127554893494 lr:2.0795215416642606e-05 tokens_per_second_per_gpu:5342.441167622313 mfu:25.201206502130038 +Step 25386 | loss:0.3849641978740692 lr:2.0791695881171874e-05 tokens_per_second_per_gpu:4690.251334310028 mfu:22.12471578333611 +Step 25387 | loss:0.4588519036769867 lr:2.078817656538053e-05 tokens_per_second_per_gpu:5333.141308853689 mfu:25.157337481598915 +Step 25388 | loss:0.4013126492500305 lr:2.0784657469295017e-05 tokens_per_second_per_gpu:4024.7996639860894 mfu:18.98566671665549 +Step 25389 | loss:0.4610467553138733 lr:2.0781138592941824e-05 tokens_per_second_per_gpu:4130.203311655203 mfu:19.482873706427412 +Step 25390 | loss:0.3921928405761719 lr:2.077761993634743e-05 tokens_per_second_per_gpu:5322.916769625865 mfu:25.10910658557499 +Step 25391 | loss:0.38742193579673767 lr:2.0774101499538275e-05 tokens_per_second_per_gpu:5280.1132897461575 mfu:24.90719526795568 +Step 25392 | loss:0.456149160861969 lr:2.0770583282540824e-05 tokens_per_second_per_gpu:3941.2513866796503 mfu:18.591555237794996 +Step 25393 | loss:0.5350326895713806 lr:2.0767065285381565e-05 tokens_per_second_per_gpu:4051.7106069005376 mfu:19.11261022586326 +Step 25394 | loss:0.4385868012905121 lr:2.076354750808694e-05 tokens_per_second_per_gpu:3707.172309130639 mfu:17.487364290987404 +Step 25395 | loss:0.3945244550704956 lr:2.0760029950683384e-05 tokens_per_second_per_gpu:4837.009714313367 mfu:22.817000101373484 +Step 25396 | loss:0.3816778361797333 lr:2.0756512613197376e-05 tokens_per_second_per_gpu:3994.912678068759 mfu:18.844684704837825 +Step 25397 | loss:0.30407920479774475 lr:2.0752995495655392e-05 tokens_per_second_per_gpu:4041.5955977369435 mfu:19.064896001840065 +Step 25398 | loss:0.4615672826766968 lr:2.0749478598083833e-05 tokens_per_second_per_gpu:3862.908474937116 mfu:18.22199835642639 +Step 25399 | loss:0.42977601289749146 lr:2.0745961920509184e-05 tokens_per_second_per_gpu:3522.8630201077726 mfu:16.617945928259186 +Step 25400 | loss:0.256626695394516 lr:2.0742445462957905e-05 tokens_per_second_per_gpu:4381.3773547605515 mfu:20.66770452246613 +Step 25401 | loss:0.43053364753723145 lr:2.0738929225456428e-05 tokens_per_second_per_gpu:5138.9677745438685 mfu:24.241387790838342 +Step 25402 | loss:0.4217948913574219 lr:2.0735413208031174e-05 tokens_per_second_per_gpu:4360.58603405329 mfu:20.569628315324874 +Step 25403 | loss:0.4093063473701477 lr:2.0731897410708616e-05 tokens_per_second_per_gpu:3831.6602330872047 mfu:18.074595068120534 +Step 25404 | loss:0.4647260308265686 lr:2.0728381833515214e-05 tokens_per_second_per_gpu:4320.857524191379 mfu:20.38222215592349 +Step 25405 | loss:0.48000138998031616 lr:2.072486647647736e-05 tokens_per_second_per_gpu:5350.417561205411 mfu:25.238832511574593 +Step 25406 | loss:0.3356102406978607 lr:2.072135133962153e-05 tokens_per_second_per_gpu:5326.8124412836 mfu:25.12748313345465 +Step 25407 | loss:0.5103591680526733 lr:2.0717836422974147e-05 tokens_per_second_per_gpu:5003.117640562192 mfu:23.600559530423833 +Step 25408 | loss:0.49625495076179504 lr:2.071432172656169e-05 tokens_per_second_per_gpu:4858.742389708551 mfu:22.919516839189342 +Step 25409 | loss:0.3716641068458557 lr:2.0710807250410523e-05 tokens_per_second_per_gpu:5330.215842848398 mfu:25.143537559315988 +Step 25410 | loss:0.3137773275375366 lr:2.0707292994547105e-05 tokens_per_second_per_gpu:5073.5055331068725 mfu:23.93259122896993 +Step 25411 | loss:0.3837401568889618 lr:2.07037789589979e-05 tokens_per_second_per_gpu:4437.24989885888 mfu:20.931264845816965 +Step 25412 | loss:0.4322623908519745 lr:2.0700265143789287e-05 tokens_per_second_per_gpu:5338.160525462962 mfu:25.18101398271152 +Step 25413 | loss:0.5158437490463257 lr:2.0696751548947722e-05 tokens_per_second_per_gpu:5087.156559484766 mfu:23.996985449499725 +Step 25414 | loss:0.38769015669822693 lr:2.069323817449963e-05 tokens_per_second_per_gpu:4426.150105024568 mfu:20.878905224480405 +Step 25415 | loss:0.30124399065971375 lr:2.0689725020471453e-05 tokens_per_second_per_gpu:4912.927836894849 mfu:23.175118838557715 +Step 25416 | loss:0.40160486102104187 lr:2.068621208688959e-05 tokens_per_second_per_gpu:4983.211789553243 mfu:23.506660234926226 +Step 25417 | loss:0.49182456731796265 lr:2.0682699373780446e-05 tokens_per_second_per_gpu:4996.891782319024 mfu:23.571191094849862 +Step 25418 | loss:0.4974464476108551 lr:2.0679186881170464e-05 tokens_per_second_per_gpu:4535.98126846404 mfu:21.396997561553192 +Step 25419 | loss:0.3388040065765381 lr:2.0675674609086076e-05 tokens_per_second_per_gpu:3999.221052358374 mfu:18.865008041446394 +Step 25420 | loss:0.32203030586242676 lr:2.0672162557553665e-05 tokens_per_second_per_gpu:3640.992322162317 mfu:17.175182000987714 +Step 25421 | loss:0.4512174427509308 lr:2.066865072659966e-05 tokens_per_second_per_gpu:3384.2888202834256 mfu:15.964267784491241 +Step 25422 | loss:0.394661009311676 lr:2.066513911625049e-05 tokens_per_second_per_gpu:5193.988154082324 mfu:24.50092830856599 +Step 25423 | loss:0.45744699239730835 lr:2.0661627726532556e-05 tokens_per_second_per_gpu:5322.75793012047 mfu:25.10835731252723 +Step 25424 | loss:0.5077276825904846 lr:2.0658116557472246e-05 tokens_per_second_per_gpu:4479.515357043659 mfu:21.13063822330533 +Step 25425 | loss:0.39464008808135986 lr:2.0654605609095982e-05 tokens_per_second_per_gpu:5336.923509140982 mfu:25.175178765663986 +Step 25426 | loss:0.47703245282173157 lr:2.0651094881430193e-05 tokens_per_second_per_gpu:5262.678835519074 mfu:24.824953972741902 +Step 25427 | loss:0.3904874324798584 lr:2.064758437450125e-05 tokens_per_second_per_gpu:4185.827347408971 mfu:19.74526177351706 +Step 25428 | loss:0.39392849802970886 lr:2.064407408833557e-05 tokens_per_second_per_gpu:5310.068477150401 mfu:25.048498997823486 +Step 25429 | loss:0.3077830374240875 lr:2.064056402295957e-05 tokens_per_second_per_gpu:5373.51814488178 mfu:25.34780190614202 +Step 25430 | loss:0.36562487483024597 lr:2.063705417839963e-05 tokens_per_second_per_gpu:5280.948907237981 mfu:24.91113701823566 +Step 25431 | loss:0.539604663848877 lr:2.063354455468214e-05 tokens_per_second_per_gpu:5104.0166463338355 mfu:24.076517355794234 +Step 25432 | loss:0.44639337062835693 lr:2.063003515183351e-05 tokens_per_second_per_gpu:4841.7297011517985 mfu:22.839265084603163 +Step 25433 | loss:0.3285958170890808 lr:2.0626525969880146e-05 tokens_per_second_per_gpu:5019.65975120553 mfu:23.678591488703567 +Step 25434 | loss:0.47073912620544434 lr:2.062301700884841e-05 tokens_per_second_per_gpu:5328.701956520581 mfu:25.136396299212038 +Step 25435 | loss:0.40241843461990356 lr:2.0619508268764716e-05 tokens_per_second_per_gpu:4288.313244130829 mfu:20.22870523425113 +Step 25436 | loss:0.38012751936912537 lr:2.061599974965547e-05 tokens_per_second_per_gpu:5342.407317946809 mfu:25.201046827435338 +Step 25437 | loss:0.42648622393608093 lr:2.0612491451547035e-05 tokens_per_second_per_gpu:4660.960682873676 mfu:21.986546783010375 +Step 25438 | loss:0.4028964936733246 lr:2.060898337446578e-05 tokens_per_second_per_gpu:5336.277952626022 mfu:25.172133565439623 +Step 25439 | loss:0.3867638409137726 lr:2.0605475518438116e-05 tokens_per_second_per_gpu:5361.5325790551115 mfu:25.291263947189226 +Step 25440 | loss:0.32588595151901245 lr:2.0601967883490435e-05 tokens_per_second_per_gpu:5315.2705419805525 mfu:25.073038025191092 +Step 25441 | loss:0.4728199541568756 lr:2.0598460469649093e-05 tokens_per_second_per_gpu:5240.277080208687 mfu:24.719281070809224 +Step 25442 | loss:0.33579695224761963 lr:2.0594953276940475e-05 tokens_per_second_per_gpu:4964.020004224521 mfu:23.416129309074414 +Step 25443 | loss:0.363799124956131 lr:2.0591446305390993e-05 tokens_per_second_per_gpu:5343.125234216472 mfu:25.20443336096848 +Step 25444 | loss:0.48820140957832336 lr:2.0587939555026968e-05 tokens_per_second_per_gpu:5372.177850379614 mfu:25.341479508297482 +Step 25445 | loss:0.4257183074951172 lr:2.0584433025874826e-05 tokens_per_second_per_gpu:5309.447267716692 mfu:25.045568646935422 +Step 25446 | loss:0.40144455432891846 lr:2.0580926717960897e-05 tokens_per_second_per_gpu:5267.581947317599 mfu:24.84808278765268 +Step 25447 | loss:0.41902342438697815 lr:2.0577420631311588e-05 tokens_per_second_per_gpu:4524.911761930271 mfu:21.344780810580836 +Step 25448 | loss:0.4832347333431244 lr:2.057391476595324e-05 tokens_per_second_per_gpu:5367.396111225896 mfu:25.318923228860562 +Step 25449 | loss:0.3751373589038849 lr:2.0570409121912232e-05 tokens_per_second_per_gpu:5341.08469353927 mfu:25.19480778244198 +Step 25450 | loss:0.43889400362968445 lr:2.056690369921493e-05 tokens_per_second_per_gpu:5117.039625833023 mfu:24.13794897204132 +Step 25451 | loss:0.4717182517051697 lr:2.056339849788772e-05 tokens_per_second_per_gpu:4122.028455511305 mfu:19.444311515222363 +Step 25452 | loss:0.46034669876098633 lr:2.0559893517956948e-05 tokens_per_second_per_gpu:5329.472926501897 mfu:25.140033096154944 +Step 25453 | loss:0.3991157114505768 lr:2.0556388759448954e-05 tokens_per_second_per_gpu:3527.596283202076 mfu:16.640273537852973 +Step 25454 | loss:0.43638718128204346 lr:2.0552884222390117e-05 tokens_per_second_per_gpu:5357.6103057395185 mfu:25.27276191475074 +Step 25455 | loss:0.39344102144241333 lr:2.0549379906806815e-05 tokens_per_second_per_gpu:5224.619402678635 mfu:24.645421134424712 +Step 25456 | loss:0.5757313966751099 lr:2.0545875812725363e-05 tokens_per_second_per_gpu:4988.914691632675 mfu:23.53356179705034 +Step 25457 | loss:0.5625211596488953 lr:2.054237194017214e-05 tokens_per_second_per_gpu:5348.542228356711 mfu:25.229986246562497 +Step 25458 | loss:0.42397430539131165 lr:2.0538868289173514e-05 tokens_per_second_per_gpu:5325.664569389544 mfu:25.122068425883253 +Step 25459 | loss:0.41592472791671753 lr:2.0535364859755817e-05 tokens_per_second_per_gpu:4571.394234076703 mfu:21.564046562423094 +Step 25460 | loss:0.426521897315979 lr:2.0531861651945382e-05 tokens_per_second_per_gpu:5217.296229217921 mfu:24.610876475748288 +Step 25461 | loss:0.4074831008911133 lr:2.0528358665768577e-05 tokens_per_second_per_gpu:5313.335208069535 mfu:25.063908724930812 +Step 25462 | loss:0.4260264039039612 lr:2.0524855901251768e-05 tokens_per_second_per_gpu:5349.837532187112 mfu:25.236096415732117 +Step 25463 | loss:0.3712662160396576 lr:2.0521353358421258e-05 tokens_per_second_per_gpu:5072.985445620826 mfu:23.930137887562175 +Step 25464 | loss:0.441243439912796 lr:2.051785103730341e-05 tokens_per_second_per_gpu:4120.513310638965 mfu:19.43716431834916 +Step 25465 | loss:0.44380152225494385 lr:2.051434893792459e-05 tokens_per_second_per_gpu:4359.754932833601 mfu:20.56570786907111 +Step 25466 | loss:0.48378732800483704 lr:2.0510847060311106e-05 tokens_per_second_per_gpu:5338.182767134678 mfu:25.181118900471823 +Step 25467 | loss:0.4228810667991638 lr:2.0507345404489294e-05 tokens_per_second_per_gpu:3936.462112441606 mfu:18.5689634140797 +Step 25468 | loss:0.41393619775772095 lr:2.050384397048549e-05 tokens_per_second_per_gpu:5071.288966427372 mfu:23.92213530575745 +Step 25469 | loss:0.4549851417541504 lr:2.050034275832607e-05 tokens_per_second_per_gpu:5353.1777360644655 mfu:25.251852727319143 +Step 25470 | loss:0.3411673605442047 lr:2.049684176803731e-05 tokens_per_second_per_gpu:3276.560016586491 mfu:15.456092637023561 +Step 25471 | loss:0.4413422644138336 lr:2.0493340999645572e-05 tokens_per_second_per_gpu:4839.764206314243 mfu:22.829993510106412 +Step 25472 | loss:0.5311010479927063 lr:2.0489840453177193e-05 tokens_per_second_per_gpu:5332.45357805003 mfu:25.15409333805908 +Step 25473 | loss:0.4835772216320038 lr:2.0486340128658494e-05 tokens_per_second_per_gpu:5330.153294780145 mfu:25.143242509406242 +Step 25474 | loss:0.4412963092327118 lr:2.048284002611578e-05 tokens_per_second_per_gpu:3987.1388142945725 mfu:18.80801406305706 +Step 25475 | loss:0.45752498507499695 lr:2.0479340145575387e-05 tokens_per_second_per_gpu:4241.081665229692 mfu:20.005905818036137 +Step 25476 | loss:0.340579628944397 lr:2.0475840487063668e-05 tokens_per_second_per_gpu:4865.090071922272 mfu:22.94945993921773 +Step 25477 | loss:0.5115754008293152 lr:2.04723410506069e-05 tokens_per_second_per_gpu:5302.504150040769 mfu:25.012816776240065 +Step 25478 | loss:0.40994584560394287 lr:2.0468841836231422e-05 tokens_per_second_per_gpu:4979.251322834069 mfu:23.48797803768658 +Step 25479 | loss:0.5232138633728027 lr:2.046534284396357e-05 tokens_per_second_per_gpu:5367.635766482382 mfu:25.32005372359481 +Step 25480 | loss:0.5647779107093811 lr:2.046184407382963e-05 tokens_per_second_per_gpu:4432.75943415956 mfu:20.910082557677995 +Step 25481 | loss:0.42951473593711853 lr:2.0458345525855943e-05 tokens_per_second_per_gpu:3858.528435762612 mfu:18.201336964328515 +Step 25482 | loss:0.32607486844062805 lr:2.0454847200068787e-05 tokens_per_second_per_gpu:3599.539456757605 mfu:16.979641762285546 +Step 25483 | loss:0.4797765016555786 lr:2.045134909649452e-05 tokens_per_second_per_gpu:3445.0607785204375 mfu:16.250939480260495 +Step 25484 | loss:0.3522278666496277 lr:2.0447851215159407e-05 tokens_per_second_per_gpu:5290.689520073637 mfu:24.95708515090137 +Step 25485 | loss:0.43880337476730347 lr:2.044435355608978e-05 tokens_per_second_per_gpu:5355.235595327107 mfu:25.26156000056802 +Step 25486 | loss:0.3570675551891327 lr:2.0440856119311935e-05 tokens_per_second_per_gpu:4682.516463293945 mfu:22.088229077048165 +Step 25487 | loss:0.4551358222961426 lr:2.0437358904852206e-05 tokens_per_second_per_gpu:5333.902600868026 mfu:25.160928625920356 +Step 25488 | loss:0.39059877395629883 lr:2.0433861912736867e-05 tokens_per_second_per_gpu:4262.179873339686 mfu:20.105429665416477 +Step 25489 | loss:0.285739541053772 lr:2.0430365142992207e-05 tokens_per_second_per_gpu:4219.422061471882 mfu:19.90373377160108 +Step 25490 | loss:0.39647120237350464 lr:2.0426868595644548e-05 tokens_per_second_per_gpu:5315.659180421378 mfu:25.07487129902502 +Step 25491 | loss:0.3791847825050354 lr:2.0423372270720202e-05 tokens_per_second_per_gpu:5324.830416233236 mfu:25.11813358312374 +Step 25492 | loss:0.48240336775779724 lr:2.0419876168245428e-05 tokens_per_second_per_gpu:4906.174759225268 mfu:23.143263418996437 +Step 25493 | loss:0.28813791275024414 lr:2.041638028824654e-05 tokens_per_second_per_gpu:3912.2675771928766 mfu:18.454833663298057 +Step 25494 | loss:0.45728424191474915 lr:2.0412884630749845e-05 tokens_per_second_per_gpu:3757.598952213358 mfu:17.725235370081084 +Step 25495 | loss:0.39892110228538513 lr:2.0409389195781625e-05 tokens_per_second_per_gpu:5338.734339210126 mfu:25.183720760059607 +Step 25496 | loss:0.38355112075805664 lr:2.0405893983368146e-05 tokens_per_second_per_gpu:5356.6899080665535 mfu:25.268420241891327 +Step 25497 | loss:0.38515329360961914 lr:2.0402398993535716e-05 tokens_per_second_per_gpu:4449.2345422969975 mfu:20.98779845370559 +Step 25498 | loss:0.5294004678726196 lr:2.0398904226310632e-05 tokens_per_second_per_gpu:5001.161853689454 mfu:23.59133375005348 +Step 25499 | loss:0.3893112242221832 lr:2.0395409681719152e-05 tokens_per_second_per_gpu:5360.980983204729 mfu:25.288661975451014 +Step 25500 | loss:0.5098270177841187 lr:2.0391915359787576e-05 tokens_per_second_per_gpu:5339.293998107153 mfu:25.18636076656454 +Step 25501 | loss:0.456241250038147 lr:2.0388421260542195e-05 tokens_per_second_per_gpu:4685.45403134991 mfu:22.102086086768946 +Step 25502 | loss:0.312096506357193 lr:2.0384927384009284e-05 tokens_per_second_per_gpu:4139.329470937027 mfu:19.52592335684311 +Step 25503 | loss:0.3286570608615875 lr:2.0381433730215087e-05 tokens_per_second_per_gpu:4853.186167503561 mfu:22.893307191059144 +Step 25504 | loss:0.3688097596168518 lr:2.037794029918591e-05 tokens_per_second_per_gpu:5314.03610838958 mfu:25.067214991326388 +Step 25505 | loss:0.41362401843070984 lr:2.037444709094804e-05 tokens_per_second_per_gpu:4278.6599749790985 mfu:20.183169116644866 +Step 25506 | loss:0.532034695148468 lr:2.0370954105527718e-05 tokens_per_second_per_gpu:4908.912358126118 mfu:23.15617714009429 +Step 25507 | loss:0.47594353556632996 lr:2.0367461342951234e-05 tokens_per_second_per_gpu:3584.7494942635612 mfu:16.90987498577333 +Step 25508 | loss:0.4175771176815033 lr:2.0363968803244863e-05 tokens_per_second_per_gpu:5031.488358175967 mfu:23.734389045951577 +Step 25509 | loss:0.4596908986568451 lr:2.0360476486434854e-05 tokens_per_second_per_gpu:3454.477657663121 mfu:16.29536050586179 +Step 25510 | loss:0.419013649225235 lr:2.0356984392547497e-05 tokens_per_second_per_gpu:5386.115205688637 mfu:25.40722439124816 +Step 25511 | loss:0.3585074543952942 lr:2.0353492521609024e-05 tokens_per_second_per_gpu:5067.409706141298 mfu:23.903836173117988 +Step 25512 | loss:0.4815279245376587 lr:2.035000087364574e-05 tokens_per_second_per_gpu:3280.4744177354155 mfu:15.474557535108742 +Step 25513 | loss:0.47077909111976624 lr:2.0346509448683866e-05 tokens_per_second_per_gpu:3104.070196209615 mfu:14.642428724507775 +Step 25514 | loss:0.4368780255317688 lr:2.0343018246749673e-05 tokens_per_second_per_gpu:5358.037799909404 mfu:25.274778477688116 +Step 25515 | loss:0.40390104055404663 lr:2.033952726786944e-05 tokens_per_second_per_gpu:5029.344895347845 mfu:23.724277966078855 +Step 25516 | loss:0.5513443946838379 lr:2.0336036512069396e-05 tokens_per_second_per_gpu:5098.11076857399 mfu:24.048658322754385 +Step 25517 | loss:0.3484792113304138 lr:2.0332545979375816e-05 tokens_per_second_per_gpu:4959.919347672993 mfu:23.396785812476708 +Step 25518 | loss:0.4188219904899597 lr:2.0329055669814934e-05 tokens_per_second_per_gpu:3983.2988340844727 mfu:18.789900221237836 +Step 25519 | loss:0.400023490190506 lr:2.0325565583413026e-05 tokens_per_second_per_gpu:3916.1729167186904 mfu:18.473255816160954 +Step 25520 | loss:0.40317049622535706 lr:2.0322075720196304e-05 tokens_per_second_per_gpu:5059.455628373767 mfu:23.866315431182937 +Step 25521 | loss:0.6231471300125122 lr:2.031858608019104e-05 tokens_per_second_per_gpu:5336.887948409255 mfu:25.175011019625146 +Step 25522 | loss:0.38297542929649353 lr:2.0315096663423476e-05 tokens_per_second_per_gpu:5354.772158676246 mfu:25.25937389081546 +Step 25523 | loss:0.44241300225257874 lr:2.031160746991988e-05 tokens_per_second_per_gpu:4098.61979754699 mfu:19.3338888816758 +Step 25524 | loss:0.33385521173477173 lr:2.030811849970647e-05 tokens_per_second_per_gpu:3071.081860193576 mfu:14.486817115129616 +Step 25525 | loss:0.35060951113700867 lr:2.0304629752809468e-05 tokens_per_second_per_gpu:5284.882748256958 mfu:24.929693617504658 +Step 25526 | loss:0.4364292025566101 lr:2.030114122925514e-05 tokens_per_second_per_gpu:5005.693156207531 mfu:23.61270867715118 +Step 25527 | loss:0.3770824372768402 lr:2.0297652929069734e-05 tokens_per_second_per_gpu:4923.762528871121 mfu:23.22622792919886 +Step 25528 | loss:0.4858914315700531 lr:2.029416485227945e-05 tokens_per_second_per_gpu:4392.841572524222 mfu:20.721783193654613 +Step 25529 | loss:0.44268274307250977 lr:2.0290676998910545e-05 tokens_per_second_per_gpu:4788.583167033135 mfu:22.588563815431673 +Step 25530 | loss:0.5329942107200623 lr:2.0287189368989267e-05 tokens_per_second_per_gpu:4912.294219197845 mfu:23.172129955774622 +Step 25531 | loss:0.37749096751213074 lr:2.028370196254183e-05 tokens_per_second_per_gpu:4270.860205783021 mfu:20.14637627456912 +Step 25532 | loss:0.3326749801635742 lr:2.028021477959445e-05 tokens_per_second_per_gpu:5267.685918847707 mfu:24.848573239099576 +Step 25533 | loss:0.38417747616767883 lr:2.027672782017336e-05 tokens_per_second_per_gpu:4971.605167405085 mfu:23.451909817959425 +Step 25534 | loss:0.4061603248119354 lr:2.0273241084304806e-05 tokens_per_second_per_gpu:4603.1372223896315 mfu:21.713783654207457 +Step 25535 | loss:0.36485305428504944 lr:2.0269754572014988e-05 tokens_per_second_per_gpu:3494.691904920033 mfu:16.485057971430734 +Step 25536 | loss:0.4509885907173157 lr:2.0266268283330138e-05 tokens_per_second_per_gpu:5356.790961424101 mfu:25.26889692782036 +Step 25537 | loss:0.36215659976005554 lr:2.02627822182765e-05 tokens_per_second_per_gpu:4013.809609121899 mfu:18.933824752764387 +Step 25538 | loss:0.34020963311195374 lr:2.025929637688025e-05 tokens_per_second_per_gpu:5331.124364589773 mfu:25.14782321137985 +Step 25539 | loss:0.342503160238266 lr:2.0255810759167642e-05 tokens_per_second_per_gpu:4510.065497165272 mfu:21.27474844665037 +Step 25540 | loss:0.3384553790092468 lr:2.025232536516486e-05 tokens_per_second_per_gpu:5360.048957209055 mfu:25.284265449809386 +Step 25541 | loss:0.42752280831336975 lr:2.0248840194898156e-05 tokens_per_second_per_gpu:3687.898301023623 mfu:17.396445506261728 +Step 25542 | loss:0.3924314081668854 lr:2.02453552483937e-05 tokens_per_second_per_gpu:4785.13530696488 mfu:22.572299671224126 +Step 25543 | loss:0.40877336263656616 lr:2.0241870525677725e-05 tokens_per_second_per_gpu:4052.853608600331 mfu:19.118001959897384 +Step 25544 | loss:0.3511248826980591 lr:2.0238386026776457e-05 tokens_per_second_per_gpu:5373.183840206145 mfu:25.34622493394879 +Step 25545 | loss:0.29335451126098633 lr:2.0234901751716067e-05 tokens_per_second_per_gpu:5077.515572599195 mfu:23.95150726943851 +Step 25546 | loss:0.4444330334663391 lr:2.0231417700522797e-05 tokens_per_second_per_gpu:5245.076270155185 mfu:24.741919668613097 +Step 25547 | loss:0.42995738983154297 lr:2.0227933873222817e-05 tokens_per_second_per_gpu:4118.830876998082 mfu:19.42922799181234 +Step 25548 | loss:0.42944788932800293 lr:2.022445026984236e-05 tokens_per_second_per_gpu:4079.9485000156346 mfu:19.245813185568544 +Step 25549 | loss:0.435533344745636 lr:2.0220966890407596e-05 tokens_per_second_per_gpu:3819.718482107554 mfu:18.018263791276706 +Step 25550 | loss:0.47827160358428955 lr:2.0217483734944743e-05 tokens_per_second_per_gpu:4536.970836648166 mfu:21.40166552351546 +Step 25551 | loss:0.44633790850639343 lr:2.0214000803480004e-05 tokens_per_second_per_gpu:5365.263356237363 mfu:25.30886265969413 +Step 25552 | loss:0.38754045963287354 lr:2.0210518096039555e-05 tokens_per_second_per_gpu:4616.117717821446 mfu:21.775014865860218 +Step 25553 | loss:0.4827367961406708 lr:2.0207035612649617e-05 tokens_per_second_per_gpu:4808.765308593986 mfu:22.683766420602634 +Step 25554 | loss:0.626258373260498 lr:2.0203553353336347e-05 tokens_per_second_per_gpu:5099.896568271587 mfu:24.0570822446168 +Step 25555 | loss:0.4519825279712677 lr:2.0200071318125975e-05 tokens_per_second_per_gpu:4451.768830934219 mfu:20.999753125601593 +Step 25556 | loss:0.4206496775150299 lr:2.019658950704465e-05 tokens_per_second_per_gpu:4170.079520006571 mfu:19.670976584803523 +Step 25557 | loss:0.3590753972530365 lr:2.0193107920118576e-05 tokens_per_second_per_gpu:4352.321660824349 mfu:20.53064385675649 +Step 25558 | loss:0.44310152530670166 lr:2.0189626557373942e-05 tokens_per_second_per_gpu:4398.247988916757 mfu:20.747286182207954 +Step 25559 | loss:0.28830012679100037 lr:2.0186145418836948e-05 tokens_per_second_per_gpu:4006.913046013217 mfu:18.901292487905735 +Step 25560 | loss:0.42005765438079834 lr:2.0182664504533756e-05 tokens_per_second_per_gpu:4716.418756124969 mfu:22.248151976659724 +Step 25561 | loss:0.39184272289276123 lr:2.0179183814490536e-05 tokens_per_second_per_gpu:5342.587915079508 mfu:25.20189873492333 +Step 25562 | loss:0.3171424865722656 lr:2.0175703348733475e-05 tokens_per_second_per_gpu:5149.2996453793185 mfu:24.290124988368095 +Step 25563 | loss:0.41225647926330566 lr:2.017222310728878e-05 tokens_per_second_per_gpu:5111.422902504299 mfu:24.11145392978791 +Step 25564 | loss:0.37949079275131226 lr:2.0168743090182575e-05 tokens_per_second_per_gpu:3580.6970561038843 mfu:16.89075894355689 +Step 25565 | loss:0.4358011484146118 lr:2.016526329744106e-05 tokens_per_second_per_gpu:3767.9020415396776 mfu:17.77383680032164 +Step 25566 | loss:0.5643817186355591 lr:2.0161783729090434e-05 tokens_per_second_per_gpu:4604.421104438633 mfu:21.719839944885464 +Step 25567 | loss:0.4598182737827301 lr:2.015830438515683e-05 tokens_per_second_per_gpu:5188.6868717602865 mfu:24.475921255360905 +Step 25568 | loss:0.34645333886146545 lr:2.0154825265666416e-05 tokens_per_second_per_gpu:4317.032807863975 mfu:20.364180316443232 +Step 25569 | loss:0.42280468344688416 lr:2.0151346370645362e-05 tokens_per_second_per_gpu:4337.582877399624 mfu:20.461118500646297 +Step 25570 | loss:0.28654131293296814 lr:2.0147867700119867e-05 tokens_per_second_per_gpu:4868.626933890224 mfu:22.966143920571437 +Step 25571 | loss:0.3839940130710602 lr:2.014438925411604e-05 tokens_per_second_per_gpu:4340.28438700782 mfu:20.47386198239323 +Step 25572 | loss:0.43861833214759827 lr:2.0140911032660082e-05 tokens_per_second_per_gpu:5066.836214713531 mfu:23.901130916205485 +Step 25573 | loss:0.5080772638320923 lr:2.013743303577815e-05 tokens_per_second_per_gpu:5296.059429183072 mfu:24.982415928371545 +Step 25574 | loss:0.41412657499313354 lr:2.0133955263496386e-05 tokens_per_second_per_gpu:5358.223443986045 mfu:25.275654192471777 +Step 25575 | loss:0.40265077352523804 lr:2.0130477715840968e-05 tokens_per_second_per_gpu:4626.965162352454 mfu:21.82618411247787 +Step 25576 | loss:0.3515353500843048 lr:2.012700039283802e-05 tokens_per_second_per_gpu:4327.707190939529 mfu:20.414533202648173 +Step 25577 | loss:0.4350016117095947 lr:2.012352329451373e-05 tokens_per_second_per_gpu:5332.022675608737 mfu:25.152060697723964 +Step 25578 | loss:0.4579307436943054 lr:2.0120046420894222e-05 tokens_per_second_per_gpu:5112.067835674394 mfu:24.114496189588195 +Step 25579 | loss:0.44702577590942383 lr:2.011656977200565e-05 tokens_per_second_per_gpu:5001.356494741242 mfu:23.59225190510395 +Step 25580 | loss:0.40053656697273254 lr:2.0113093347874195e-05 tokens_per_second_per_gpu:5337.456900285823 mfu:25.177694862700122 +Step 25581 | loss:0.45329275727272034 lr:2.0109617148525955e-05 tokens_per_second_per_gpu:4465.5267963014 mfu:21.06465179558953 +Step 25582 | loss:0.47921594977378845 lr:2.010614117398712e-05 tokens_per_second_per_gpu:4148.142409637637 mfu:19.567495492336278 +Step 25583 | loss:0.4545370936393738 lr:2.0102665424283795e-05 tokens_per_second_per_gpu:4580.323510422437 mfu:21.606167482437566 +Step 25584 | loss:0.4277718663215637 lr:2.0099189899442154e-05 tokens_per_second_per_gpu:4639.10168265174 mfu:21.883434149433484 +Step 25585 | loss:0.41748788952827454 lr:2.0095714599488303e-05 tokens_per_second_per_gpu:3431.9026489958414 mfu:16.18887033828458 +Step 25586 | loss:0.35131222009658813 lr:2.00922395244484e-05 tokens_per_second_per_gpu:4428.171529069556 mfu:20.88844062659112 +Step 25587 | loss:0.4280630052089691 lr:2.0088764674348592e-05 tokens_per_second_per_gpu:4528.878735029428 mfu:21.363493699524966 +Step 25588 | loss:0.38867929577827454 lr:2.0085290049214988e-05 tokens_per_second_per_gpu:4764.49759928031 mfu:22.474948082919987 +Step 25589 | loss:0.29321253299713135 lr:2.0081815649073754e-05 tokens_per_second_per_gpu:5351.028697701728 mfu:25.241715346698353 +Step 25590 | loss:0.37545713782310486 lr:2.0078341473950983e-05 tokens_per_second_per_gpu:5364.867565059945 mfu:25.306995645181313 +Step 25591 | loss:0.4112091064453125 lr:2.0074867523872842e-05 tokens_per_second_per_gpu:4391.462033404785 mfu:20.715275672254585 +Step 25592 | loss:0.5140029191970825 lr:2.0071393798865422e-05 tokens_per_second_per_gpu:5076.466574405575 mfu:23.94655896598165 +Step 25593 | loss:0.43729981780052185 lr:2.0067920298954867e-05 tokens_per_second_per_gpu:4084.50255024626 mfu:19.267295417507494 +Step 25594 | loss:0.49902620911598206 lr:2.0064447024167306e-05 tokens_per_second_per_gpu:4531.7268863395075 mfu:21.376928915199333 +Step 25595 | loss:0.2758472263813019 lr:2.0060973974528874e-05 tokens_per_second_per_gpu:4275.463371279173 mfu:20.168090191595486 +Step 25596 | loss:0.3270316421985626 lr:2.0057501150065678e-05 tokens_per_second_per_gpu:3561.407207377764 mfu:16.79976543592821 +Step 25597 | loss:0.4884929060935974 lr:2.0054028550803823e-05 tokens_per_second_per_gpu:3300.310547946462 mfu:15.56812794570705 +Step 25598 | loss:0.41974589228630066 lr:2.0050556176769454e-05 tokens_per_second_per_gpu:4494.255781551917 mfu:21.200171320686543 +Step 25599 | loss:0.4286474585533142 lr:2.0047084027988648e-05 tokens_per_second_per_gpu:3971.5261548630215 mfu:18.734366484724053 +Step 25600 | loss:0.46481990814208984 lr:2.0043612104487557e-05 tokens_per_second_per_gpu:5369.682861711743 mfu:25.329710221062168 +Step 25601 | loss:0.5269212126731873 lr:2.004014040629228e-05 tokens_per_second_per_gpu:4155.099328579979 mfu:19.600312465960073 +Step 25602 | loss:0.39822888374328613 lr:2.0036668933428942e-05 tokens_per_second_per_gpu:5306.571850216538 mfu:25.032004811990383 +Step 25603 | loss:0.5131927132606506 lr:2.0033197685923627e-05 tokens_per_second_per_gpu:4505.413272712889 mfu:21.252803110157142 +Step 25604 | loss:0.5141746401786804 lr:2.0029726663802472e-05 tokens_per_second_per_gpu:4052.743883826787 mfu:19.117484369419603 +Step 25605 | loss:0.5052942037582397 lr:2.0026255867091547e-05 tokens_per_second_per_gpu:4201.433995557495 mfu:19.8188809956024 +Step 25606 | loss:0.5110449194908142 lr:2.0022785295816997e-05 tokens_per_second_per_gpu:3700.4240011253382 mfu:17.45553137074633 +Step 25607 | loss:0.451678603887558 lr:2.0019314950004888e-05 tokens_per_second_per_gpu:5341.720705379294 mfu:25.197807958806184 +Step 25608 | loss:0.5069958567619324 lr:2.0015844829681334e-05 tokens_per_second_per_gpu:4124.79307902087 mfu:19.45735271601039 +Step 25609 | loss:0.36124512553215027 lr:2.0012374934872457e-05 tokens_per_second_per_gpu:3893.1242742761715 mfu:18.364531437255685 +Step 25610 | loss:0.37350866198539734 lr:2.0008905265604316e-05 tokens_per_second_per_gpu:5340.450177347311 mfu:25.191814661304885 +Step 25611 | loss:0.40207648277282715 lr:2.000543582190304e-05 tokens_per_second_per_gpu:4425.320645970139 mfu:20.87499252460035 +Step 25612 | loss:0.3448173999786377 lr:2.0001966603794693e-05 tokens_per_second_per_gpu:5312.068564798685 mfu:25.057933752510582 +Step 25613 | loss:0.4015996754169464 lr:1.99984976113054e-05 tokens_per_second_per_gpu:3921.7419921312417 mfu:18.499526095064205 +Step 25614 | loss:0.4054836928844452 lr:1.9995028844461217e-05 tokens_per_second_per_gpu:4688.722185139354 mfu:22.11750252579829 +Step 25615 | loss:0.4467855393886566 lr:1.9991560303288255e-05 tokens_per_second_per_gpu:5341.977464056343 mfu:25.199019133293373 +Step 25616 | loss:0.4676889181137085 lr:1.9988091987812613e-05 tokens_per_second_per_gpu:4428.892577349971 mfu:20.89184193435336 +Step 25617 | loss:0.33731672167778015 lr:1.998462389806034e-05 tokens_per_second_per_gpu:5148.985205766554 mfu:24.28864172306576 +Step 25618 | loss:0.44304364919662476 lr:1.998115603405756e-05 tokens_per_second_per_gpu:4692.11932844932 mfu:22.133527430403813 +Step 25619 | loss:0.4482695758342743 lr:1.9977688395830318e-05 tokens_per_second_per_gpu:3498.1474979907257 mfu:16.50135859353017 +Step 25620 | loss:0.3673885762691498 lr:1.9974220983404727e-05 tokens_per_second_per_gpu:3791.254833098359 mfu:17.883995902501038 +Step 25621 | loss:0.4361409544944763 lr:1.997075379680684e-05 tokens_per_second_per_gpu:5359.590904978269 mfu:25.282104739285 +Step 25622 | loss:0.4223177134990692 lr:1.996728683606274e-05 tokens_per_second_per_gpu:4288.347470969861 mfu:20.228866688090548 +Step 25623 | loss:0.5040444135665894 lr:1.9963820101198528e-05 tokens_per_second_per_gpu:3438.506500234706 mfu:16.22002183131151 +Step 25624 | loss:0.320844829082489 lr:1.9960353592240237e-05 tokens_per_second_per_gpu:4270.527460817857 mfu:20.144806659796295 +Step 25625 | loss:0.49797558784484863 lr:1.9956887309213985e-05 tokens_per_second_per_gpu:4654.883013104036 mfu:21.95787737774959 +Step 25626 | loss:0.42691880464553833 lr:1.9953421252145793e-05 tokens_per_second_per_gpu:5359.115740222519 mfu:25.27986330606066 +Step 25627 | loss:0.3001982867717743 lr:1.9949955421061773e-05 tokens_per_second_per_gpu:5335.736125421242 mfu:25.169577674069423 +Step 25628 | loss:0.3848218321800232 lr:1.9946489815987952e-05 tokens_per_second_per_gpu:4804.316144449935 mfu:22.66277895422946 +Step 25629 | loss:0.4315837025642395 lr:1.994302443695042e-05 tokens_per_second_per_gpu:5359.746595839961 mfu:25.2828391596432 +Step 25630 | loss:0.4225408136844635 lr:1.9939559283975235e-05 tokens_per_second_per_gpu:4454.0777797401615 mfu:21.01064483106651 +Step 25631 | loss:0.41241806745529175 lr:1.993609435708848e-05 tokens_per_second_per_gpu:5383.643303478361 mfu:25.395564006772286 +Step 25632 | loss:0.3902701437473297 lr:1.9932629656316177e-05 tokens_per_second_per_gpu:5334.931667987451 mfu:25.16578291109964 +Step 25633 | loss:0.34416696429252625 lr:1.9929165181684417e-05 tokens_per_second_per_gpu:4172.820322351157 mfu:19.683905417092035 +Step 25634 | loss:0.46106770634651184 lr:1.9925700933219242e-05 tokens_per_second_per_gpu:5341.7855573524475 mfu:25.198113876628437 +Step 25635 | loss:0.43732789158821106 lr:1.9922236910946695e-05 tokens_per_second_per_gpu:5107.94085259584 mfu:24.095028506270733 +Step 25636 | loss:0.3844771683216095 lr:1.9918773114892842e-05 tokens_per_second_per_gpu:3577.763343310601 mfu:16.87692011976796 +Step 25637 | loss:0.5109910368919373 lr:1.9915309545083733e-05 tokens_per_second_per_gpu:4120.435216091847 mfu:19.436795933046103 +Step 25638 | loss:0.3584177792072296 lr:1.9911846201545442e-05 tokens_per_second_per_gpu:4656.26107203198 mfu:21.9643779168338 +Step 25639 | loss:0.46259188652038574 lr:1.990838308430397e-05 tokens_per_second_per_gpu:4185.142093634747 mfu:19.742029314547693 +Step 25640 | loss:0.4714648127555847 lr:1.9904920193385412e-05 tokens_per_second_per_gpu:5341.9254949240185 mfu:25.19877398602935 +Step 25641 | loss:0.44868502020835876 lr:1.990145752881577e-05 tokens_per_second_per_gpu:4245.090119264667 mfu:20.024814379631895 +Step 25642 | loss:0.33496394753456116 lr:1.9897995090621126e-05 tokens_per_second_per_gpu:4182.63755112901 mfu:19.730214960229116 +Step 25643 | loss:0.37654682993888855 lr:1.9894532878827476e-05 tokens_per_second_per_gpu:5339.876513151433 mfu:25.189108589415724 +Step 25644 | loss:0.31242644786834717 lr:1.9891070893460895e-05 tokens_per_second_per_gpu:5339.550111898753 mfu:25.187568899017034 +Step 25645 | loss:0.4397096335887909 lr:1.9887609134547423e-05 tokens_per_second_per_gpu:5354.20394622084 mfu:25.25669353571627 +Step 25646 | loss:0.5687538385391235 lr:1.9884147602113073e-05 tokens_per_second_per_gpu:5073.332721019455 mfu:23.9317760448688 +Step 25647 | loss:0.3109533190727234 lr:1.9880686296183904e-05 tokens_per_second_per_gpu:3263.3952502544007 mfu:15.39399218809434 +Step 25648 | loss:0.3978944718837738 lr:1.987722521678592e-05 tokens_per_second_per_gpu:4747.694933861589 mfu:22.395687043309536 +Step 25649 | loss:0.45182153582572937 lr:1.987376436394519e-05 tokens_per_second_per_gpu:5237.190866491186 mfu:24.704722874141417 +Step 25650 | loss:0.4178628921508789 lr:1.9870303737687702e-05 tokens_per_second_per_gpu:4196.526498620961 mfu:19.79573149524745 +Step 25651 | loss:0.38686349987983704 lr:1.9866843338039504e-05 tokens_per_second_per_gpu:5187.161838581289 mfu:24.468727413658083 +Step 25652 | loss:0.4268273711204529 lr:1.9863383165026644e-05 tokens_per_second_per_gpu:5359.008181831306 mfu:25.279355934777954 +Step 25653 | loss:0.47578999400138855 lr:1.9859923218675103e-05 tokens_per_second_per_gpu:5332.3138669306045 mfu:25.15343429687158 +Step 25654 | loss:0.3847789764404297 lr:1.9856463499010942e-05 tokens_per_second_per_gpu:4137.827682529235 mfu:19.518839164692032 +Step 25655 | loss:0.4183632433414459 lr:1.985300400606015e-05 tokens_per_second_per_gpu:3474.307249025456 mfu:16.388900071595085 +Step 25656 | loss:0.446090430021286 lr:1.984954473984878e-05 tokens_per_second_per_gpu:3837.4437893416607 mfu:18.101877089749255 +Step 25657 | loss:0.3932599723339081 lr:1.984608570040281e-05 tokens_per_second_per_gpu:3327.440051153174 mfu:15.69610244110366 +Step 25658 | loss:0.5005481243133545 lr:1.9842626887748283e-05 tokens_per_second_per_gpu:3925.476717757242 mfu:18.51714343305195 +Step 25659 | loss:0.4075133502483368 lr:1.983916830191122e-05 tokens_per_second_per_gpu:4121.983070873649 mfu:19.444097428143124 +Step 25660 | loss:0.5114777088165283 lr:1.9835709942917607e-05 tokens_per_second_per_gpu:3812.329837933879 mfu:17.983410296076794 +Step 25661 | loss:0.4788508415222168 lr:1.983225181079346e-05 tokens_per_second_per_gpu:5313.888058483313 mfu:25.06651661465861 +Step 25662 | loss:0.392367422580719 lr:1.9828793905564824e-05 tokens_per_second_per_gpu:5083.017589628835 mfu:23.977461222508218 +Step 25663 | loss:0.39608222246170044 lr:1.9825336227257672e-05 tokens_per_second_per_gpu:4492.274627405763 mfu:21.19082587855938 +Step 25664 | loss:0.3355839252471924 lr:1.9821878775898006e-05 tokens_per_second_per_gpu:4804.67259305455 mfu:22.66446038311383 +Step 25665 | loss:0.4246975779533386 lr:1.9818421551511833e-05 tokens_per_second_per_gpu:5294.1440846630185 mfu:24.973380921478977 +Step 25666 | loss:0.3893073499202728 lr:1.9814964554125187e-05 tokens_per_second_per_gpu:3723.9291458132006 mfu:17.5664091486302 +Step 25667 | loss:0.40653494000434875 lr:1.9811507783764022e-05 tokens_per_second_per_gpu:5363.20839918624 mfu:25.299169076672158 +Step 25668 | loss:0.4919929802417755 lr:1.980805124045436e-05 tokens_per_second_per_gpu:4708.539518232105 mfu:22.210984267182475 +Step 25669 | loss:0.47847193479537964 lr:1.980459492422222e-05 tokens_per_second_per_gpu:5044.695950645617 mfu:23.79669151307856 +Step 25670 | loss:0.4719173014163971 lr:1.980113883509357e-05 tokens_per_second_per_gpu:4119.743675199998 mfu:19.43353381667021 +Step 25671 | loss:0.3035222291946411 lr:1.979768297309439e-05 tokens_per_second_per_gpu:5292.07755531253 mfu:24.963632749946463 +Step 25672 | loss:0.415389746427536 lr:1.9794227338250686e-05 tokens_per_second_per_gpu:4807.880618171751 mfu:22.679593184937197 +Step 25673 | loss:0.437729150056839 lr:1.979077193058846e-05 tokens_per_second_per_gpu:4605.812963576258 mfu:21.726405581913298 +Step 25674 | loss:0.37916022539138794 lr:1.978731675013371e-05 tokens_per_second_per_gpu:5342.446216556225 mfu:25.20123031881274 +Step 25675 | loss:0.3395865261554718 lr:1.9783861796912384e-05 tokens_per_second_per_gpu:5365.802567192165 mfu:25.311406209766808 +Step 25676 | loss:0.5215902924537659 lr:1.978040707095051e-05 tokens_per_second_per_gpu:4625.253352280257 mfu:21.8181092123024 +Step 25677 | loss:0.4872426986694336 lr:1.9776952572274032e-05 tokens_per_second_per_gpu:4010.007510370778 mfu:18.915889604250413 +Step 25678 | loss:0.40350785851478577 lr:1.9773498300908967e-05 tokens_per_second_per_gpu:3543.8987729191317 mfu:16.71717516333989 +Step 25679 | loss:0.5182750225067139 lr:1.977004425688126e-05 tokens_per_second_per_gpu:3359.6473360589803 mfu:15.848029699133793 +Step 25680 | loss:0.4528348743915558 lr:1.9766590440216908e-05 tokens_per_second_per_gpu:3887.0805134566963 mfu:18.33602198629829 +Step 25681 | loss:0.4526160955429077 lr:1.9763136850941905e-05 tokens_per_second_per_gpu:3419.63524194681 mfu:16.13100288619883 +Step 25682 | loss:0.4574059844017029 lr:1.9759683489082194e-05 tokens_per_second_per_gpu:3777.4505806982384 mfu:17.818878888681343 +Step 25683 | loss:0.3854869306087494 lr:1.9756230354663775e-05 tokens_per_second_per_gpu:5369.815131816199 mfu:25.330334161712496 +Step 25684 | loss:0.41071784496307373 lr:1.975277744771259e-05 tokens_per_second_per_gpu:4673.82028144848 mfu:22.047207703561487 +Step 25685 | loss:0.3386269807815552 lr:1.9749324768254644e-05 tokens_per_second_per_gpu:3489.225825549034 mfu:16.459273542428427 +Step 25686 | loss:0.4729781150817871 lr:1.9745872316315865e-05 tokens_per_second_per_gpu:3764.733619201956 mfu:17.758890811566012 +Step 25687 | loss:0.4535450339317322 lr:1.974242009192225e-05 tokens_per_second_per_gpu:5328.757323012364 mfu:25.136657472400277 +Step 25688 | loss:0.4809454381465912 lr:1.9738968095099765e-05 tokens_per_second_per_gpu:5343.873313845933 mfu:25.2079621802162 +Step 25689 | loss:0.3497786819934845 lr:1.9735516325874342e-05 tokens_per_second_per_gpu:4080.9189867578407 mfu:19.250391137113642 +Step 25690 | loss:0.326549768447876 lr:1.9732064784271985e-05 tokens_per_second_per_gpu:5348.486306873219 mfu:25.229722455383676 +Step 25691 | loss:0.37042883038520813 lr:1.9728613470318613e-05 tokens_per_second_per_gpu:5354.04719703994 mfu:25.255954123085797 +Step 25692 | loss:0.4745841324329376 lr:1.972516238404022e-05 tokens_per_second_per_gpu:5223.83180828285 mfu:24.641705917282334 +Step 25693 | loss:0.46233370900154114 lr:1.972171152546272e-05 tokens_per_second_per_gpu:4426.9195561946435 mfu:20.882534856931557 +Step 25694 | loss:0.4652356505393982 lr:1.9718260894612094e-05 tokens_per_second_per_gpu:5364.680184278865 mfu:25.306111738066505 +Step 25695 | loss:0.4155615270137787 lr:1.9714810491514314e-05 tokens_per_second_per_gpu:2959.2013460771377 mfu:13.959057641258294 +Step 25696 | loss:0.2904750406742096 lr:1.9711360316195288e-05 tokens_per_second_per_gpu:4635.765227596799 mfu:21.867695521681398 +Step 25697 | loss:0.44335711002349854 lr:1.9707910368680986e-05 tokens_per_second_per_gpu:5356.760492017888 mfu:25.268753198432638 +Step 25698 | loss:0.40418845415115356 lr:1.9704460648997374e-05 tokens_per_second_per_gpu:5335.605972250088 mfu:25.168963719354664 +Step 25699 | loss:0.3150934875011444 lr:1.970101115717038e-05 tokens_per_second_per_gpu:4591.117080852333 mfu:21.6570825957284 +Step 25700 | loss:0.48777058720588684 lr:1.9697561893225926e-05 tokens_per_second_per_gpu:4915.225231499714 mfu:23.1859560408848 +Step 25701 | loss:0.4654701352119446 lr:1.9694112857189978e-05 tokens_per_second_per_gpu:4822.398784535517 mfu:22.748077852729686 +Step 25702 | loss:0.47565957903862 lr:1.9690664049088493e-05 tokens_per_second_per_gpu:5210.980287705537 mfu:24.581083102023634 +Step 25703 | loss:0.4396969676017761 lr:1.968721546894737e-05 tokens_per_second_per_gpu:3844.57196315678 mfu:18.135501901827034 +Step 25704 | loss:0.457549124956131 lr:1.9683767116792575e-05 tokens_per_second_per_gpu:5125.62994579326 mfu:24.178470976953477 +Step 25705 | loss:0.4114565849304199 lr:1.968031899265004e-05 tokens_per_second_per_gpu:5247.764137298396 mfu:24.75459879652505 +Step 25706 | loss:0.4918835163116455 lr:1.967687109654571e-05 tokens_per_second_per_gpu:4745.817345271274 mfu:22.386830137579373 +Step 25707 | loss:0.32869067788124084 lr:1.967342342850547e-05 tokens_per_second_per_gpu:5364.290828930814 mfu:25.304275082458872 +Step 25708 | loss:0.43345150351524353 lr:1.9669975988555295e-05 tokens_per_second_per_gpu:5341.262160588881 mfu:25.195644924794 +Step 25709 | loss:0.42976564168930054 lr:1.9666528776721093e-05 tokens_per_second_per_gpu:5292.612388866253 mfu:24.966155651071386 +Step 25710 | loss:0.4056945741176605 lr:1.9663081793028816e-05 tokens_per_second_per_gpu:5319.222779199545 mfu:25.091681402474283 +Step 25711 | loss:0.2952970564365387 lr:1.965963503750436e-05 tokens_per_second_per_gpu:4750.824805687839 mfu:22.41045118272516 +Step 25712 | loss:0.3922021687030792 lr:1.9656188510173674e-05 tokens_per_second_per_gpu:4255.676665886377 mfu:20.07475292629761 +Step 25713 | loss:0.5279576778411865 lr:1.9652742211062647e-05 tokens_per_second_per_gpu:3700.481412705849 mfu:17.455802190967024 +Step 25714 | loss:0.4591875374317169 lr:1.964929614019724e-05 tokens_per_second_per_gpu:5308.542033633117 mfu:25.041298503314337 +Step 25715 | loss:0.38755714893341064 lr:1.9645850297603323e-05 tokens_per_second_per_gpu:5306.985693937222 mfu:25.033956983429867 +Step 25716 | loss:0.26550012826919556 lr:1.9642404683306846e-05 tokens_per_second_per_gpu:5345.470643953315 mfu:25.215497058865335 +Step 25717 | loss:0.4341607391834259 lr:1.9638959297333736e-05 tokens_per_second_per_gpu:4912.785372509115 mfu:23.17444680974328 +Step 25718 | loss:0.40185463428497314 lr:1.9635514139709865e-05 tokens_per_second_per_gpu:5144.230999250093 mfu:24.26621531977634 +Step 25719 | loss:0.37579160928726196 lr:1.963206921046119e-05 tokens_per_second_per_gpu:5321.5023725669835 mfu:25.102434633327228 +Step 25720 | loss:0.5370933413505554 lr:1.9628624509613574e-05 tokens_per_second_per_gpu:4599.1854388947295 mfu:21.695142417218218 +Step 25721 | loss:0.3092103898525238 lr:1.962518003719297e-05 tokens_per_second_per_gpu:5338.214109238376 mfu:25.18126674652248 +Step 25722 | loss:0.388069212436676 lr:1.962173579322524e-05 tokens_per_second_per_gpu:3478.6339098502513 mfu:16.4093096689101 +Step 25723 | loss:0.4243090748786926 lr:1.9618291777736318e-05 tokens_per_second_per_gpu:5335.8195115219305 mfu:25.169971020533506 +Step 25724 | loss:0.5430811047554016 lr:1.9614847990752117e-05 tokens_per_second_per_gpu:4097.940817106939 mfu:19.33068601509407 +Step 25725 | loss:0.43404126167297363 lr:1.9611404432298502e-05 tokens_per_second_per_gpu:5357.805394126726 mfu:25.27368217995871 +Step 25726 | loss:0.43591973185539246 lr:1.9607961102401397e-05 tokens_per_second_per_gpu:5339.2803616713545 mfu:25.186296441170263 +Step 25727 | loss:0.4285789728164673 lr:1.9604518001086704e-05 tokens_per_second_per_gpu:4245.323205444198 mfu:20.025913886909347 +Step 25728 | loss:0.3485988974571228 lr:1.9601075128380314e-05 tokens_per_second_per_gpu:4967.573084554717 mfu:23.432889795209967 +Step 25729 | loss:0.36730799078941345 lr:1.9597632484308097e-05 tokens_per_second_per_gpu:5181.434198506775 mfu:24.441709158190434 +Step 25730 | loss:0.4795398712158203 lr:1.9594190068895968e-05 tokens_per_second_per_gpu:3626.080406053429 mfu:17.10483994846685 +Step 25731 | loss:0.4522043466567993 lr:1.959074788216983e-05 tokens_per_second_per_gpu:5367.083690318319 mfu:25.317449486135022 +Step 25732 | loss:0.4268604815006256 lr:1.9587305924155543e-05 tokens_per_second_per_gpu:4980.314467342996 mfu:23.492993071726346 +Step 25733 | loss:0.2933602035045624 lr:1.9583864194879004e-05 tokens_per_second_per_gpu:5149.658565021506 mfu:24.2918180735583 +Step 25734 | loss:0.32900094985961914 lr:1.9580422694366124e-05 tokens_per_second_per_gpu:4042.8696221396904 mfu:19.070905792318797 +Step 25735 | loss:0.38978782296180725 lr:1.957698142264277e-05 tokens_per_second_per_gpu:4711.194510815806 mfu:22.223508320188238 +Step 25736 | loss:0.37321263551712036 lr:1.9573540379734802e-05 tokens_per_second_per_gpu:5379.833546308565 mfu:25.37759273219076 +Step 25737 | loss:0.43674570322036743 lr:1.9570099565668116e-05 tokens_per_second_per_gpu:5289.748682396834 mfu:24.952647059056428 +Step 25738 | loss:0.44774001836776733 lr:1.956665898046861e-05 tokens_per_second_per_gpu:3720.5072003893283 mfu:17.550267248221697 +Step 25739 | loss:0.45756879448890686 lr:1.956321862416213e-05 tokens_per_second_per_gpu:4259.874586820451 mfu:20.094555235582806 +Step 25740 | loss:0.4144229292869568 lr:1.9559778496774566e-05 tokens_per_second_per_gpu:4465.115388188935 mfu:21.06271111332957 +Step 25741 | loss:0.35604286193847656 lr:1.9556338598331814e-05 tokens_per_second_per_gpu:5353.074895285315 mfu:25.251367609817443 +Step 25742 | loss:0.4086127281188965 lr:1.955289892885972e-05 tokens_per_second_per_gpu:4206.4037015346485 mfu:19.842323946615902 +Step 25743 | loss:0.35590898990631104 lr:1.9549459488384138e-05 tokens_per_second_per_gpu:3947.291849473554 mfu:18.620049131403064 +Step 25744 | loss:0.4436599016189575 lr:1.954602027693096e-05 tokens_per_second_per_gpu:5203.460002050875 mfu:24.545608631497554 +Step 25745 | loss:0.4064648449420929 lr:1.954258129452605e-05 tokens_per_second_per_gpu:5359.208576780609 mfu:25.280301231944723 +Step 25746 | loss:0.3822585344314575 lr:1.953914254119529e-05 tokens_per_second_per_gpu:4314.034060643761 mfu:20.35003471416645 +Step 25747 | loss:0.4244292080402374 lr:1.9535704016964506e-05 tokens_per_second_per_gpu:4740.226065455731 mfu:22.360455116718754 +Step 25748 | loss:0.42659991979599 lr:1.95322657218596e-05 tokens_per_second_per_gpu:5319.425379243298 mfu:25.092637101447803 +Step 25749 | loss:0.4522678852081299 lr:1.952882765590639e-05 tokens_per_second_per_gpu:4575.622059027994 mfu:21.583989934059655 +Step 25750 | loss:0.4535766541957855 lr:1.952538981913078e-05 tokens_per_second_per_gpu:5149.825363706315 mfu:24.29260489141334 +Step 25751 | loss:0.4800507724285126 lr:1.9521952211558577e-05 tokens_per_second_per_gpu:4254.914159808671 mfu:20.071156055032116 +Step 25752 | loss:0.4529305398464203 lr:1.9518514833215663e-05 tokens_per_second_per_gpu:4218.515526080446 mfu:19.89945748948879 +Step 25753 | loss:0.3666992485523224 lr:1.9515077684127904e-05 tokens_per_second_per_gpu:4972.543423197531 mfu:23.45633573061534 +Step 25754 | loss:0.46623772382736206 lr:1.951164076432112e-05 tokens_per_second_per_gpu:5274.9778028573555 mfu:24.88297030009679 +Step 25755 | loss:0.4656517803668976 lr:1.9508204073821173e-05 tokens_per_second_per_gpu:4467.301609180574 mfu:21.073023890754893 +Step 25756 | loss:0.4071536064147949 lr:1.950476761265394e-05 tokens_per_second_per_gpu:4942.154391879875 mfu:23.31298548498656 +Step 25757 | loss:0.42137083411216736 lr:1.9501331380845235e-05 tokens_per_second_per_gpu:3746.6607333563147 mfu:17.67363792547983 +Step 25758 | loss:0.3939254879951477 lr:1.9497895378420888e-05 tokens_per_second_per_gpu:4534.45477224795 mfu:21.389796818454787 +Step 25759 | loss:0.41188037395477295 lr:1.9494459605406766e-05 tokens_per_second_per_gpu:4334.149535131119 mfu:20.444922839377163 +Step 25760 | loss:0.29552993178367615 lr:1.9491024061828722e-05 tokens_per_second_per_gpu:4161.588889676953 mfu:19.630924832887803 +Step 25761 | loss:0.43291234970092773 lr:1.9487588747712564e-05 tokens_per_second_per_gpu:5297.303907637639 mfu:24.988286345571655 +Step 25762 | loss:0.36615097522735596 lr:1.9484153663084144e-05 tokens_per_second_per_gpu:4713.438537424837 mfu:22.234093776573456 +Step 25763 | loss:0.5511543154716492 lr:1.9480718807969313e-05 tokens_per_second_per_gpu:4144.689755503113 mfu:19.55120873370019 +Step 25764 | loss:0.46941298246383667 lr:1.947728418239389e-05 tokens_per_second_per_gpu:4408.999350864862 mfu:20.79800219088867 +Step 25765 | loss:0.44391852617263794 lr:1.947384978638369e-05 tokens_per_second_per_gpu:4657.078137607834 mfu:21.968232154560866 +Step 25766 | loss:0.42445674538612366 lr:1.947041561996456e-05 tokens_per_second_per_gpu:4924.766132314982 mfu:23.23096209787596 +Step 25767 | loss:0.5380200743675232 lr:1.9466981683162348e-05 tokens_per_second_per_gpu:4741.06301322791 mfu:22.36440314215814 +Step 25768 | loss:0.4931543171405792 lr:1.946354797600285e-05 tokens_per_second_per_gpu:4458.280154594321 mfu:21.030468150252666 +Step 25769 | loss:0.3780762553215027 lr:1.94601144985119e-05 tokens_per_second_per_gpu:5333.02527306136 mfu:25.15679012096875 +Step 25770 | loss:0.38075771927833557 lr:1.9456681250715346e-05 tokens_per_second_per_gpu:5324.998303950886 mfu:25.118925538132494 +Step 25771 | loss:0.4224768579006195 lr:1.9453248232638988e-05 tokens_per_second_per_gpu:5336.087937789521 mfu:25.17123723303597 +Step 25772 | loss:0.4201526343822479 lr:1.9449815444308628e-05 tokens_per_second_per_gpu:4900.2953190146145 mfu:23.115529096386847 +Step 25773 | loss:0.4772796630859375 lr:1.944638288575011e-05 tokens_per_second_per_gpu:4654.8396037560615 mfu:21.957672608448796 +Step 25774 | loss:0.29817384481430054 lr:1.9442950556989252e-05 tokens_per_second_per_gpu:4025.6096143709005 mfu:18.989487390812428 +Step 25775 | loss:0.45483145117759705 lr:1.9439518458051852e-05 tokens_per_second_per_gpu:5368.891983137073 mfu:25.325979511887653 +Step 25776 | loss:0.32060402631759644 lr:1.9436086588963736e-05 tokens_per_second_per_gpu:4262.10187587158 mfu:20.10506173805168 +Step 25777 | loss:0.47964176535606384 lr:1.9432654949750727e-05 tokens_per_second_per_gpu:4898.825238026816 mfu:23.108594473545722 +Step 25778 | loss:0.45414265990257263 lr:1.9429223540438618e-05 tokens_per_second_per_gpu:4334.046890559734 mfu:20.44443864742097 +Step 25779 | loss:0.4223027527332306 lr:1.942579236105321e-05 tokens_per_second_per_gpu:4763.7369819765 mfu:22.47136012132322 +Step 25780 | loss:0.4308250844478607 lr:1.942236141162031e-05 tokens_per_second_per_gpu:4630.823976727873 mfu:21.84438679826834 +Step 25781 | loss:0.34828487038612366 lr:1.9418930692165733e-05 tokens_per_second_per_gpu:5345.249018134521 mfu:25.21445161206482 +Step 25782 | loss:0.304385244846344 lr:1.9415500202715297e-05 tokens_per_second_per_gpu:5071.957432572198 mfu:23.925288574614683 +Step 25783 | loss:0.41103318333625793 lr:1.9412069943294774e-05 tokens_per_second_per_gpu:4258.6946011125 mfu:20.088989041672082 +Step 25784 | loss:0.40926051139831543 lr:1.9408639913929992e-05 tokens_per_second_per_gpu:5369.562247546732 mfu:25.329141263466912 +Step 25785 | loss:0.4329453706741333 lr:1.9405210114646715e-05 tokens_per_second_per_gpu:3775.6087850937042 mfu:17.81019082457202 +Step 25786 | loss:0.42308640480041504 lr:1.9401780545470773e-05 tokens_per_second_per_gpu:3764.0692079348782 mfu:17.755756670258954 +Step 25787 | loss:0.4614681005477905 lr:1.939835120642793e-05 tokens_per_second_per_gpu:3956.0185055214592 mfu:18.661214256902092 +Step 25788 | loss:0.33413147926330566 lr:1.939492209754399e-05 tokens_per_second_per_gpu:5316.528770841118 mfu:25.07897330540236 +Step 25789 | loss:0.39297163486480713 lr:1.9391493218844758e-05 tokens_per_second_per_gpu:5325.900825209846 mfu:25.123182884897012 +Step 25790 | loss:0.48768651485443115 lr:1.9388064570356e-05 tokens_per_second_per_gpu:4574.154768240768 mfu:21.57706847307069 +Step 25791 | loss:0.40596500039100647 lr:1.9384636152103515e-05 tokens_per_second_per_gpu:4676.064457873825 mfu:22.057793866655818 +Step 25792 | loss:0.43827083706855774 lr:1.9381207964113103e-05 tokens_per_second_per_gpu:4827.306998137306 mfu:22.771230733716923 +Step 25793 | loss:0.4288369119167328 lr:1.9377780006410533e-05 tokens_per_second_per_gpu:5338.2864539012435 mfu:25.1816080086405 +Step 25794 | loss:0.25948697328567505 lr:1.9374352279021568e-05 tokens_per_second_per_gpu:5341.144934396874 mfu:25.19509194884545 +Step 25795 | loss:0.403128057718277 lr:1.937092478197201e-05 tokens_per_second_per_gpu:4875.283418741782 mfu:22.99754369532966 +Step 25796 | loss:0.5924339890480042 lr:1.9367497515287654e-05 tokens_per_second_per_gpu:5329.3643089469915 mfu:25.13952072861639 +Step 25797 | loss:0.3902736306190491 lr:1.936407047899424e-05 tokens_per_second_per_gpu:5376.201958658258 mfu:25.360461913631514 +Step 25798 | loss:0.32085344195365906 lr:1.9360643673117557e-05 tokens_per_second_per_gpu:4074.1212839649393 mfu:19.218325213231836 +Step 25799 | loss:0.4096575975418091 lr:1.9357217097683405e-05 tokens_per_second_per_gpu:3723.571888591277 mfu:17.564723905360015 +Step 25800 | loss:0.4270976185798645 lr:1.9353790752717526e-05 tokens_per_second_per_gpu:5363.842052873921 mfu:25.302158129229415 +Step 25801 | loss:0.25201255083084106 lr:1.9350364638245687e-05 tokens_per_second_per_gpu:5365.3838997267485 mfu:25.309431283900402 +Step 25802 | loss:0.49449825286865234 lr:1.9346938754293665e-05 tokens_per_second_per_gpu:4798.230654897989 mfu:22.634072661721348 +Step 25803 | loss:0.401581346988678 lr:1.9343513100887246e-05 tokens_per_second_per_gpu:4217.6609919908315 mfu:19.895426506389526 +Step 25804 | loss:0.3842051029205322 lr:1.9340087678052154e-05 tokens_per_second_per_gpu:5334.254656713982 mfu:25.16258933341265 +Step 25805 | loss:0.3956557810306549 lr:1.9336662485814178e-05 tokens_per_second_per_gpu:5323.098612193545 mfu:25.10996436799228 +Step 25806 | loss:0.41479718685150146 lr:1.9333237524199093e-05 tokens_per_second_per_gpu:5335.4995625986 mfu:25.168461766873243 +Step 25807 | loss:0.5377100110054016 lr:1.932981279323265e-05 tokens_per_second_per_gpu:4720.753749872478 mfu:22.26860088179292 +Step 25808 | loss:0.47512608766555786 lr:1.9326388292940574e-05 tokens_per_second_per_gpu:5301.443601763368 mfu:25.00781398906737 +Step 25809 | loss:0.3389086127281189 lr:1.9322964023348645e-05 tokens_per_second_per_gpu:5282.307589766015 mfu:24.917546155535536 +Step 25810 | loss:0.5046665072441101 lr:1.931953998448263e-05 tokens_per_second_per_gpu:4573.987801838385 mfu:21.576280864063243 +Step 25811 | loss:0.39461666345596313 lr:1.931611617636826e-05 tokens_per_second_per_gpu:4064.938068065564 mfu:19.175006417016284 +Step 25812 | loss:0.445241779088974 lr:1.9312692599031296e-05 tokens_per_second_per_gpu:4340.508172185293 mfu:20.474917615257056 +Step 25813 | loss:0.35616710782051086 lr:1.9309269252497503e-05 tokens_per_second_per_gpu:4941.10821519296 mfu:23.308050491058236 +Step 25814 | loss:0.4531504511833191 lr:1.930584613679261e-05 tokens_per_second_per_gpu:5335.617436012492 mfu:25.16901779587785 +Step 25815 | loss:0.3745373487472534 lr:1.9302423251942347e-05 tokens_per_second_per_gpu:3510.6727457746706 mfu:16.560442324354675 +Step 25816 | loss:0.44883930683135986 lr:1.9299000597972477e-05 tokens_per_second_per_gpu:5320.383578147914 mfu:25.097157089166462 +Step 25817 | loss:0.4986003041267395 lr:1.929557817490874e-05 tokens_per_second_per_gpu:4500.982762617757 mfu:21.23190363811543 +Step 25818 | loss:0.502812922000885 lr:1.9292155982776894e-05 tokens_per_second_per_gpu:4420.586139109969 mfu:20.852659047950766 +Step 25819 | loss:0.26264286041259766 lr:1.928873402160265e-05 tokens_per_second_per_gpu:5330.100521904701 mfu:25.14299357074879 +Step 25820 | loss:0.40377190709114075 lr:1.928531229141175e-05 tokens_per_second_per_gpu:3522.3308531879084 mfu:16.615435606101148 +Step 25821 | loss:0.40681588649749756 lr:1.9281890792229955e-05 tokens_per_second_per_gpu:4673.708242221213 mfu:22.046679195410608 +Step 25822 | loss:0.39568158984184265 lr:1.9278469524082975e-05 tokens_per_second_per_gpu:4229.567680336046 mfu:19.951592386804414 +Step 25823 | loss:0.5019886493682861 lr:1.9275048486996533e-05 tokens_per_second_per_gpu:5338.057216129483 mfu:25.180526654959106 +Step 25824 | loss:0.4038563370704651 lr:1.9271627680996374e-05 tokens_per_second_per_gpu:4235.147624696342 mfu:19.97791393638861 +Step 25825 | loss:0.4105673134326935 lr:1.926820710610824e-05 tokens_per_second_per_gpu:4517.510894757635 mfu:21.30986965740937 +Step 25826 | loss:0.38215240836143494 lr:1.9264786762357822e-05 tokens_per_second_per_gpu:5341.0572710939105 mfu:25.19467842608847 +Step 25827 | loss:0.40839824080467224 lr:1.926136664977086e-05 tokens_per_second_per_gpu:5372.8882590011635 mfu:25.344830626973536 +Step 25828 | loss:0.3574429154396057 lr:1.9257946768373107e-05 tokens_per_second_per_gpu:4437.979508329951 mfu:20.934706538176222 +Step 25829 | loss:0.44025298953056335 lr:1.925452711819026e-05 tokens_per_second_per_gpu:5337.119009343669 mfu:25.176100973475837 +Step 25830 | loss:0.4072224497795105 lr:1.9251107699248017e-05 tokens_per_second_per_gpu:5219.08156671416 mfu:24.619298217328033 +Step 25831 | loss:0.35755670070648193 lr:1.9247688511572114e-05 tokens_per_second_per_gpu:5338.939340549581 mfu:25.184687786354658 +Step 25832 | loss:0.3407613933086395 lr:1.9244269555188294e-05 tokens_per_second_per_gpu:5337.882673259726 mfu:25.17970330646238 +Step 25833 | loss:0.4145732522010803 lr:1.9240850830122226e-05 tokens_per_second_per_gpu:4530.868666364621 mfu:21.37288054515037 +Step 25834 | loss:0.5468019843101501 lr:1.923743233639964e-05 tokens_per_second_per_gpu:5262.4082792875 mfu:24.823677712835917 +Step 25835 | loss:0.3473530411720276 lr:1.923401407404627e-05 tokens_per_second_per_gpu:3502.292760953318 mfu:16.52091250617951 +Step 25836 | loss:0.5036470293998718 lr:1.923059604308781e-05 tokens_per_second_per_gpu:5333.987105429325 mfu:25.16132724835397 +Step 25837 | loss:0.4455358684062958 lr:1.9227178243549942e-05 tokens_per_second_per_gpu:5323.933042902339 mfu:25.113900520014223 +Step 25838 | loss:0.42778781056404114 lr:1.9223760675458395e-05 tokens_per_second_per_gpu:5054.716352804969 mfu:23.843959459721123 +Step 25839 | loss:0.49138233065605164 lr:1.9220343338838885e-05 tokens_per_second_per_gpu:4413.876905731317 mfu:20.82101044939041 +Step 25840 | loss:0.3622717559337616 lr:1.9216926233717085e-05 tokens_per_second_per_gpu:5320.751427869429 mfu:25.098892299064484 +Step 25841 | loss:0.35423845052719116 lr:1.9213509360118713e-05 tokens_per_second_per_gpu:5344.775624809656 mfu:25.212218534982203 +Step 25842 | loss:0.36213546991348267 lr:1.9210092718069483e-05 tokens_per_second_per_gpu:5356.828711019829 mfu:25.269074999104152 +Step 25843 | loss:0.5034571886062622 lr:1.9206676307595073e-05 tokens_per_second_per_gpu:3381.598772714101 mfu:15.951578371137295 +Step 25844 | loss:0.3246038854122162 lr:1.920326012872116e-05 tokens_per_second_per_gpu:5135.236548928209 mfu:24.22378696299654 +Step 25845 | loss:0.39161697030067444 lr:1.9199844181473457e-05 tokens_per_second_per_gpu:4937.567811248792 mfu:23.291349801598393 +Step 25846 | loss:0.42941567301750183 lr:1.919642846587768e-05 tokens_per_second_per_gpu:5260.239765642373 mfu:24.81344846398527 +Step 25847 | loss:0.47763293981552124 lr:1.9193012981959475e-05 tokens_per_second_per_gpu:4360.960262824617 mfu:20.571393616289235 +Step 25848 | loss:0.4751688838005066 lr:1.9189597729744548e-05 tokens_per_second_per_gpu:4515.359138967419 mfu:21.299719458219528 +Step 25849 | loss:0.41759297251701355 lr:1.9186182709258587e-05 tokens_per_second_per_gpu:4211.224932533226 mfu:19.865066515823024 +Step 25850 | loss:0.4911591708660126 lr:1.9182767920527318e-05 tokens_per_second_per_gpu:4431.1005798613105 mfu:20.902257458923422 +Step 25851 | loss:0.36258721351623535 lr:1.9179353363576342e-05 tokens_per_second_per_gpu:5350.74785246963 mfu:25.240390551821413 +Step 25852 | loss:0.39886465668678284 lr:1.9175939038431383e-05 tokens_per_second_per_gpu:4739.507738630008 mfu:22.357066646522625 +Step 25853 | loss:0.41529399156570435 lr:1.9172524945118125e-05 tokens_per_second_per_gpu:5312.601121200107 mfu:25.06044591191958 +Step 25854 | loss:0.5046575665473938 lr:1.9169111083662257e-05 tokens_per_second_per_gpu:4828.424720682345 mfu:22.776503221665 +Step 25855 | loss:0.5172165036201477 lr:1.9165697454089417e-05 tokens_per_second_per_gpu:5370.207143609293 mfu:25.33218334822463 +Step 25856 | loss:0.3922847807407379 lr:1.91622840564253e-05 tokens_per_second_per_gpu:4477.978751430814 mfu:21.12338979245759 +Step 25857 | loss:0.41459938883781433 lr:1.9158870890695617e-05 tokens_per_second_per_gpu:5342.792410903004 mfu:25.20286337661334 +Step 25858 | loss:0.5911403298377991 lr:1.915545795692597e-05 tokens_per_second_per_gpu:5090.427627038158 mfu:24.01241563325075 +Step 25859 | loss:0.355249285697937 lr:1.915204525514206e-05 tokens_per_second_per_gpu:4452.218477068151 mfu:21.00187418313289 +Step 25860 | loss:0.5070734620094299 lr:1.9148632785369548e-05 tokens_per_second_per_gpu:4310.97542310408 mfu:20.335606599034396 +Step 25861 | loss:0.31250089406967163 lr:1.9145220547634125e-05 tokens_per_second_per_gpu:3117.704710412802 mfu:14.70674505429221 +Step 25862 | loss:0.4980961084365845 lr:1.9141808541961416e-05 tokens_per_second_per_gpu:3596.2819798503674 mfu:16.964275687931234 +Step 25863 | loss:0.32101738452911377 lr:1.9138396768377104e-05 tokens_per_second_per_gpu:3827.637547529594 mfu:18.055619374004355 +Step 25864 | loss:0.40599381923675537 lr:1.9134985226906867e-05 tokens_per_second_per_gpu:3980.311334063112 mfu:18.77580767391701 +Step 25865 | loss:0.2905586063861847 lr:1.9131573917576346e-05 tokens_per_second_per_gpu:4348.655689366459 mfu:20.513350843910395 +Step 25866 | loss:0.517469584941864 lr:1.9128162840411173e-05 tokens_per_second_per_gpu:5348.988235826803 mfu:25.23209013989564 +Step 25867 | loss:0.31560367345809937 lr:1.912475199543703e-05 tokens_per_second_per_gpu:5361.513792761787 mfu:25.291175329038342 +Step 25868 | loss:0.6131957173347473 lr:1.9121341382679587e-05 tokens_per_second_per_gpu:5335.612739812507 mfu:25.168995643101265 +Step 25869 | loss:0.39416205883026123 lr:1.911793100216445e-05 tokens_per_second_per_gpu:3444.340201294532 mfu:16.24754039454275 +Step 25870 | loss:0.39414384961128235 lr:1.9114520853917305e-05 tokens_per_second_per_gpu:4074.8300409457343 mfu:19.221668540836657 +Step 25871 | loss:0.42164668440818787 lr:1.91111109379638e-05 tokens_per_second_per_gpu:5104.758285403236 mfu:24.080015793821264 +Step 25872 | loss:0.4682065546512604 lr:1.9107701254329573e-05 tokens_per_second_per_gpu:4135.790205588062 mfu:19.509228038331106 +Step 25873 | loss:0.5018191933631897 lr:1.9104291803040253e-05 tokens_per_second_per_gpu:5326.128942658255 mfu:25.12425895382201 +Step 25874 | loss:0.3713201582431793 lr:1.9100882584121492e-05 tokens_per_second_per_gpu:4733.921028538555 mfu:22.33071318183021 +Step 25875 | loss:0.4353334605693817 lr:1.9097473597598953e-05 tokens_per_second_per_gpu:5343.685820627079 mfu:25.20707774271322 +Step 25876 | loss:0.4069109261035919 lr:1.9094064843498244e-05 tokens_per_second_per_gpu:4832.257514967793 mfu:22.794583166251595 +Step 25877 | loss:0.396748423576355 lr:1.909065632184501e-05 tokens_per_second_per_gpu:4040.1910774397297 mfu:19.05827063996223 +Step 25878 | loss:0.34321218729019165 lr:1.908724803266492e-05 tokens_per_second_per_gpu:3345.136043462824 mfu:15.779577456075902 +Step 25879 | loss:0.4053933322429657 lr:1.908383997598358e-05 tokens_per_second_per_gpu:5175.466020776961 mfu:24.413556245562784 +Step 25880 | loss:0.3961697220802307 lr:1.9080432151826606e-05 tokens_per_second_per_gpu:4562.206635066607 mfu:21.520707090326443 +Step 25881 | loss:0.5041995644569397 lr:1.907702456021965e-05 tokens_per_second_per_gpu:5310.361705508418 mfu:25.049882206018168 +Step 25882 | loss:0.4641300439834595 lr:1.907361720118835e-05 tokens_per_second_per_gpu:4100.2127525961405 mfu:19.3414031224288 +Step 25883 | loss:0.32216474413871765 lr:1.9070210074758304e-05 tokens_per_second_per_gpu:3879.587021811289 mfu:18.3006739076854 +Step 25884 | loss:0.34633585810661316 lr:1.9066803180955155e-05 tokens_per_second_per_gpu:3959.2741928876817 mfu:18.67657188968611 +Step 25885 | loss:0.43168923258781433 lr:1.9063396519804526e-05 tokens_per_second_per_gpu:4844.8972431590455 mfu:22.854206920689737 +Step 25886 | loss:0.4899868965148926 lr:1.905999009133208e-05 tokens_per_second_per_gpu:5098.690257204871 mfu:24.05139186949721 +Step 25887 | loss:0.4629816710948944 lr:1.9056583895563358e-05 tokens_per_second_per_gpu:3843.4825319037204 mfu:18.130362868730213 +Step 25888 | loss:0.4917518198490143 lr:1.905317793252402e-05 tokens_per_second_per_gpu:3837.0679868821862 mfu:18.100104365429402 +Step 25889 | loss:0.3209899663925171 lr:1.9049772202239692e-05 tokens_per_second_per_gpu:3865.9068440168226 mfu:18.236142174949883 +Step 25890 | loss:0.5397843718528748 lr:1.9046366704735957e-05 tokens_per_second_per_gpu:4344.922348027668 mfu:20.495740035841596 +Step 25891 | loss:0.3829648196697235 lr:1.9042961440038455e-05 tokens_per_second_per_gpu:4890.140433322144 mfu:23.067626767973305 +Step 25892 | loss:0.4007357954978943 lr:1.9039556408172786e-05 tokens_per_second_per_gpu:4968.991391490706 mfu:23.43958019101514 +Step 25893 | loss:0.48179203271865845 lr:1.90361516091646e-05 tokens_per_second_per_gpu:4829.971469487907 mfu:22.783799499683393 +Step 25894 | loss:0.3724110424518585 lr:1.9032747043039435e-05 tokens_per_second_per_gpu:5028.262298253364 mfu:23.71917117087408 +Step 25895 | loss:0.3715318441390991 lr:1.902934270982293e-05 tokens_per_second_per_gpu:4845.747935816947 mfu:22.858219783100193 +Step 25896 | loss:0.38478443026542664 lr:1.9025938609540688e-05 tokens_per_second_per_gpu:5312.9433363147655 mfu:25.062060199003344 +Step 25897 | loss:0.3352225124835968 lr:1.9022534742218335e-05 tokens_per_second_per_gpu:4395.52858840248 mfu:20.734458306004537 +Step 25898 | loss:0.46670523285865784 lr:1.9019131107881427e-05 tokens_per_second_per_gpu:5357.464759887527 mfu:25.272075350134212 +Step 25899 | loss:0.3895602822303772 lr:1.901572770655559e-05 tokens_per_second_per_gpu:4497.5932609322745 mfu:21.215914780356393 +Step 25900 | loss:0.3959234356880188 lr:1.901232453826644e-05 tokens_per_second_per_gpu:5333.4035493959445 mfu:25.158574514979588 +Step 25901 | loss:0.3520335257053375 lr:1.9008921603039555e-05 tokens_per_second_per_gpu:5328.196538276123 mfu:25.13401215511968 +Step 25902 | loss:0.3944106698036194 lr:1.9005518900900498e-05 tokens_per_second_per_gpu:4713.094674734469 mfu:22.232471717593967 +Step 25903 | loss:0.3916161060333252 lr:1.900211643187489e-05 tokens_per_second_per_gpu:4564.014346144793 mfu:21.52923437190956 +Step 25904 | loss:0.47608107328414917 lr:1.8998714195988343e-05 tokens_per_second_per_gpu:4426.106351576783 mfu:20.878698832002094 +Step 25905 | loss:0.3463878631591797 lr:1.89953121932664e-05 tokens_per_second_per_gpu:4956.177603201238 mfu:23.379135365395996 +Step 25906 | loss:0.4614906311035156 lr:1.899191042373467e-05 tokens_per_second_per_gpu:5106.804242796658 mfu:24.089666923922366 +Step 25907 | loss:0.31750887632369995 lr:1.8988508887418755e-05 tokens_per_second_per_gpu:4891.979687687324 mfu:23.07630284462301 +Step 25908 | loss:0.5118666887283325 lr:1.8985107584344223e-05 tokens_per_second_per_gpu:4228.950054429413 mfu:19.948678940024983 +Step 25909 | loss:0.4574972093105316 lr:1.898170651453664e-05 tokens_per_second_per_gpu:4952.002331808673 mfu:23.359439897862515 +Step 25910 | loss:0.4919315576553345 lr:1.8978305678021595e-05 tokens_per_second_per_gpu:4517.524316037106 mfu:21.309932967874534 +Step 25911 | loss:0.4995107054710388 lr:1.8974905074824696e-05 tokens_per_second_per_gpu:4835.953346596018 mfu:22.81201703461583 +Step 25912 | loss:0.3283676207065582 lr:1.8971504704971472e-05 tokens_per_second_per_gpu:3386.282992142729 mfu:15.97367463339238 +Step 25913 | loss:0.4268830120563507 lr:1.896810456848752e-05 tokens_per_second_per_gpu:5291.767114810292 mfu:24.962168349130792 +Step 25914 | loss:0.5320203304290771 lr:1.8964704665398437e-05 tokens_per_second_per_gpu:5329.545450994775 mfu:25.140375206561803 +Step 25915 | loss:0.45349326729774475 lr:1.8961304995729763e-05 tokens_per_second_per_gpu:3550.3455060077586 mfu:16.747585503245027 +Step 25916 | loss:0.47526833415031433 lr:1.895790555950706e-05 tokens_per_second_per_gpu:5369.960203718446 mfu:25.331018490627212 +Step 25917 | loss:0.38334572315216064 lr:1.8954506356755908e-05 tokens_per_second_per_gpu:4421.155305355244 mfu:20.855343902239504 +Step 25918 | loss:0.4869695007801056 lr:1.8951107387501898e-05 tokens_per_second_per_gpu:5269.993411040027 mfu:24.85945807346951 +Step 25919 | loss:0.4578903019428253 lr:1.8947708651770545e-05 tokens_per_second_per_gpu:3540.087938099285 mfu:16.699198805299208 +Step 25920 | loss:0.38071882724761963 lr:1.8944310149587446e-05 tokens_per_second_per_gpu:5326.731891260635 mfu:25.127103165253075 +Step 25921 | loss:0.38078320026397705 lr:1.8940911880978152e-05 tokens_per_second_per_gpu:5349.119313995522 mfu:25.23270845798173 +Step 25922 | loss:0.36363738775253296 lr:1.8937513845968252e-05 tokens_per_second_per_gpu:2211.022325138188 mfu:10.429769546985186 +Step 25923 | loss:0.4906841218471527 lr:1.8934116044583243e-05 tokens_per_second_per_gpu:3894.7402207383875 mfu:18.372154132426857 +Step 25924 | loss:0.4125497341156006 lr:1.8930718476848715e-05 tokens_per_second_per_gpu:5341.996580866833 mfu:25.19910931055012 +Step 25925 | loss:0.46441224217414856 lr:1.8927321142790227e-05 tokens_per_second_per_gpu:3880.40108233341 mfu:18.30451397006129 +Step 25926 | loss:0.4724554121494293 lr:1.892392404243331e-05 tokens_per_second_per_gpu:4326.774760913493 mfu:20.410134771126394 +Step 25927 | loss:0.36896970868110657 lr:1.8920527175803525e-05 tokens_per_second_per_gpu:5334.3342245354415 mfu:25.162964668402147 +Step 25928 | loss:0.4920152425765991 lr:1.891713054292642e-05 tokens_per_second_per_gpu:4710.153560195341 mfu:22.218597978507017 +Step 25929 | loss:0.33628225326538086 lr:1.8913734143827578e-05 tokens_per_second_per_gpu:4710.517294599382 mfu:22.220313775750366 +Step 25930 | loss:0.4218015968799591 lr:1.8910337978532473e-05 tokens_per_second_per_gpu:5312.396688869014 mfu:25.059481569734064 +Step 25931 | loss:0.4117332100868225 lr:1.8906942047066685e-05 tokens_per_second_per_gpu:4100.994919786999 mfu:19.345092738519565 +Step 25932 | loss:0.49639028310775757 lr:1.8903546349455747e-05 tokens_per_second_per_gpu:5366.592123221587 mfu:25.31513068026974 +Step 25933 | loss:0.38811349868774414 lr:1.8900150885725225e-05 tokens_per_second_per_gpu:4425.308493804092 mfu:20.874935200759737 +Step 25934 | loss:0.46089768409729004 lr:1.8896755655900622e-05 tokens_per_second_per_gpu:5308.514276354787 mfu:25.04116756749658 +Step 25935 | loss:0.27831342816352844 lr:1.8893360660007488e-05 tokens_per_second_per_gpu:4595.330572055591 mfu:21.676958352629896 +Step 25936 | loss:0.4330070912837982 lr:1.8889965898071373e-05 tokens_per_second_per_gpu:3494.93318965755 mfu:16.486196152705027 +Step 25937 | loss:0.37270626425743103 lr:1.8886571370117796e-05 tokens_per_second_per_gpu:3876.0734948172903 mfu:18.284100001385184 +Step 25938 | loss:0.36589303612709045 lr:1.8883177076172266e-05 tokens_per_second_per_gpu:3874.0897209029104 mfu:18.274742201364322 +Step 25939 | loss:0.38990074396133423 lr:1.8879783016260334e-05 tokens_per_second_per_gpu:4538.706250491078 mfu:21.409851766704612 +Step 25940 | loss:0.3907521665096283 lr:1.887638919040754e-05 tokens_per_second_per_gpu:3903.7629196321877 mfu:18.414715742540086 +Step 25941 | loss:0.4072299599647522 lr:1.887299559863938e-05 tokens_per_second_per_gpu:4136.698644651145 mfu:19.513513300387352 +Step 25942 | loss:0.333634614944458 lr:1.8869602240981394e-05 tokens_per_second_per_gpu:4709.471103229987 mfu:22.215378712562277 +Step 25943 | loss:0.3992230296134949 lr:1.886620911745911e-05 tokens_per_second_per_gpu:5332.3350184675555 mfu:25.1535340722802 +Step 25944 | loss:0.4807753562927246 lr:1.886281622809805e-05 tokens_per_second_per_gpu:5317.804060876844 mfu:25.084989066087797 +Step 25945 | loss:0.36430323123931885 lr:1.88594235729237e-05 tokens_per_second_per_gpu:3600.1026646285964 mfu:16.982298509906183 +Step 25946 | loss:0.4196002185344696 lr:1.8856031151961602e-05 tokens_per_second_per_gpu:4805.208982343254 mfu:22.6669906229061 +Step 25947 | loss:0.4269201457500458 lr:1.885263896523728e-05 tokens_per_second_per_gpu:3739.0068650039643 mfu:17.637533322577195 +Step 25948 | loss:0.5142464637756348 lr:1.8849247012776216e-05 tokens_per_second_per_gpu:5317.216425682531 mfu:25.08221709061532 +Step 25949 | loss:0.5247007608413696 lr:1.8845855294603946e-05 tokens_per_second_per_gpu:5314.529507648581 mfu:25.06954243981369 +Step 25950 | loss:0.4531291425228119 lr:1.884246381074599e-05 tokens_per_second_per_gpu:5358.062877329829 mfu:25.274896772158908 +Step 25951 | loss:0.4399329721927643 lr:1.883907256122784e-05 tokens_per_second_per_gpu:5331.141817329673 mfu:25.147905538931706 +Step 25952 | loss:0.42860522866249084 lr:1.883568154607498e-05 tokens_per_second_per_gpu:5334.493550086635 mfu:25.163716234210867 +Step 25953 | loss:0.4093174636363983 lr:1.883229076531294e-05 tokens_per_second_per_gpu:4827.467895724232 mfu:22.77198971508641 +Step 25954 | loss:0.4663037955760956 lr:1.882890021896723e-05 tokens_per_second_per_gpu:3983.1802743834364 mfu:18.789340954899675 +Step 25955 | loss:0.47605249285697937 lr:1.8825509907063327e-05 tokens_per_second_per_gpu:5346.043250747658 mfu:25.218198142811044 +Step 25956 | loss:0.3937789499759674 lr:1.8822119829626738e-05 tokens_per_second_per_gpu:4641.449743443469 mfu:21.894510352808553 +Step 25957 | loss:0.4542698264122009 lr:1.8818729986682987e-05 tokens_per_second_per_gpu:4985.171560165694 mfu:23.515904807276502 +Step 25958 | loss:0.47990190982818604 lr:1.881534037825754e-05 tokens_per_second_per_gpu:4957.411153606105 mfu:23.384954233121157 +Step 25959 | loss:0.439270943403244 lr:1.8811951004375882e-05 tokens_per_second_per_gpu:5346.4108926463105 mfu:25.219932372373165 +Step 25960 | loss:0.3617269992828369 lr:1.8808561865063522e-05 tokens_per_second_per_gpu:5142.692958253307 mfu:24.258960117977644 +Step 25961 | loss:0.4224323630332947 lr:1.8805172960345967e-05 tokens_per_second_per_gpu:5348.502105915318 mfu:25.22979698216085 +Step 25962 | loss:0.36038580536842346 lr:1.880178429024867e-05 tokens_per_second_per_gpu:4998.85560808032 mfu:23.58045479602912 +Step 25963 | loss:0.4138562083244324 lr:1.8798395854797128e-05 tokens_per_second_per_gpu:5338.540442751794 mfu:25.18280611738351 +Step 25964 | loss:0.33417874574661255 lr:1.8795007654016834e-05 tokens_per_second_per_gpu:4637.740167053139 mfu:21.87701164805637 +Step 25965 | loss:0.2924008369445801 lr:1.8791619687933306e-05 tokens_per_second_per_gpu:4617.125559717828 mfu:21.779769028041017 +Step 25966 | loss:0.5048659443855286 lr:1.8788231956571955e-05 tokens_per_second_per_gpu:4275.0139079748415 mfu:20.165969996502607 +Step 25967 | loss:0.3999931812286377 lr:1.8784844459958288e-05 tokens_per_second_per_gpu:4826.551942165212 mfu:22.76766900587072 +Step 25968 | loss:0.4041294753551483 lr:1.8781457198117796e-05 tokens_per_second_per_gpu:4800.61001143748 mfu:22.645296492478217 +Step 25969 | loss:0.3611212968826294 lr:1.8778070171075963e-05 tokens_per_second_per_gpu:5315.844720060873 mfu:25.07574652116085 +Step 25970 | loss:0.34252139925956726 lr:1.8774683378858232e-05 tokens_per_second_per_gpu:5315.057635793295 mfu:25.072033710380257 +Step 25971 | loss:0.42817673087120056 lr:1.8771296821490086e-05 tokens_per_second_per_gpu:4838.477031653123 mfu:22.823921687615268 +Step 25972 | loss:0.41011881828308105 lr:1.8767910498997027e-05 tokens_per_second_per_gpu:5198.930200410301 mfu:24.52424078429511 +Step 25973 | loss:0.46937477588653564 lr:1.8764524411404494e-05 tokens_per_second_per_gpu:5349.142814893659 mfu:25.232819315727912 +Step 25974 | loss:0.33114153146743774 lr:1.876113855873794e-05 tokens_per_second_per_gpu:4431.325960222968 mfu:20.903320616543215 +Step 25975 | loss:0.43260905146598816 lr:1.8757752941022848e-05 tokens_per_second_per_gpu:4767.9366685992145 mfu:22.49117075966299 +Step 25976 | loss:0.33779430389404297 lr:1.8754367558284702e-05 tokens_per_second_per_gpu:4835.894907639838 mfu:22.811741368088814 +Step 25977 | loss:0.36740633845329285 lr:1.875098241054893e-05 tokens_per_second_per_gpu:4031.5612582514095 mfu:19.017562310451947 +Step 25978 | loss:0.31843337416648865 lr:1.8747597497841e-05 tokens_per_second_per_gpu:3833.8631111918726 mfu:18.084986419990013 +Step 25979 | loss:0.5038460493087769 lr:1.8744212820186403e-05 tokens_per_second_per_gpu:5355.831457375984 mfu:25.26437078351712 +Step 25980 | loss:0.5052056312561035 lr:1.8740828377610563e-05 tokens_per_second_per_gpu:4025.6134203714687 mfu:18.989505344366478 +Step 25981 | loss:0.35512271523475647 lr:1.873744417013893e-05 tokens_per_second_per_gpu:5373.308104693568 mfu:25.346811110737672 +Step 25982 | loss:0.3851684033870697 lr:1.873406019779696e-05 tokens_per_second_per_gpu:5089.827802808933 mfu:24.009586160020927 +Step 25983 | loss:0.37100517749786377 lr:1.8730676460610135e-05 tokens_per_second_per_gpu:3734.4668904377327 mfu:17.616117487948745 +Step 25984 | loss:0.3518485426902771 lr:1.8727292958603864e-05 tokens_per_second_per_gpu:5306.492931507747 mfu:25.031632538976094 +Step 25985 | loss:0.3525521457195282 lr:1.8723909691803616e-05 tokens_per_second_per_gpu:5327.106823377029 mfu:25.12887178401612 +Step 25986 | loss:0.4374750852584839 lr:1.872052666023485e-05 tokens_per_second_per_gpu:3357.8201867159023 mfu:15.839410723938672 +Step 25987 | loss:0.44376814365386963 lr:1.8717143863922993e-05 tokens_per_second_per_gpu:4445.878504781827 mfu:20.971967452147485 +Step 25988 | loss:0.44190797209739685 lr:1.8713761302893473e-05 tokens_per_second_per_gpu:5376.633521064633 mfu:25.362497667134523 +Step 25989 | loss:0.32944047451019287 lr:1.8710378977171743e-05 tokens_per_second_per_gpu:3427.455939070724 mfu:16.167894448880855 +Step 25990 | loss:0.4167954921722412 lr:1.8706996886783266e-05 tokens_per_second_per_gpu:4389.610808375928 mfu:20.706543127941906 +Step 25991 | loss:0.4626423120498657 lr:1.870361503175344e-05 tokens_per_second_per_gpu:4723.098364641887 mfu:22.279660829692755 +Step 25992 | loss:0.4039141535758972 lr:1.870023341210772e-05 tokens_per_second_per_gpu:4772.144096065649 mfu:22.51101791288331 +Step 25993 | loss:0.4646913409233093 lr:1.8696852027871557e-05 tokens_per_second_per_gpu:4331.861408170304 mfu:20.434129354108283 +Step 25994 | loss:0.44949597120285034 lr:1.8693470879070364e-05 tokens_per_second_per_gpu:5378.909190080638 mfu:25.37323238615219 +Step 25995 | loss:0.6055686473846436 lr:1.869008996572955e-05 tokens_per_second_per_gpu:4856.81230357389 mfu:22.910412293585498 +Step 25996 | loss:0.3510640859603882 lr:1.8686709287874565e-05 tokens_per_second_per_gpu:4288.580867091278 mfu:20.229967657416758 +Step 25997 | loss:0.3520394563674927 lr:1.8683328845530857e-05 tokens_per_second_per_gpu:5188.1300906307115 mfu:24.473294823776044 +Step 25998 | loss:0.4366688132286072 lr:1.8679948638723803e-05 tokens_per_second_per_gpu:3804.2083878888375 mfu:17.94510003055207 +Step 25999 | loss:0.3099346458911896 lr:1.867656866747886e-05 tokens_per_second_per_gpu:4822.779355012005 mfu:22.7498730685993 +Step 26000 | loss:0.39765313267707825 lr:1.867318893182144e-05 tokens_per_second_per_gpu:3746.041082701045 mfu:17.67071492761562 +Step 26001 | loss:0.4048941433429718 lr:1.866980943177699e-05 tokens_per_second_per_gpu:5352.623592584659 mfu:25.249238737978263 +Step 26002 | loss:0.4715013802051544 lr:1.8666430167370875e-05 tokens_per_second_per_gpu:5074.311686841268 mfu:23.936393993679086 +Step 26003 | loss:0.5250102281570435 lr:1.8663051138628533e-05 tokens_per_second_per_gpu:4561.795934592684 mfu:21.51876974611822 +Step 26004 | loss:0.3919900059700012 lr:1.8659672345575385e-05 tokens_per_second_per_gpu:4097.87176061508 mfu:19.330360263839722 +Step 26005 | loss:0.4020087420940399 lr:1.8656293788236857e-05 tokens_per_second_per_gpu:5373.35324322187 mfu:25.3470240368771 +Step 26006 | loss:0.3593444228172302 lr:1.8652915466638328e-05 tokens_per_second_per_gpu:4396.8895029138275 mfu:20.740877971950525 +Step 26007 | loss:0.5490975379943848 lr:1.8649537380805216e-05 tokens_per_second_per_gpu:5362.553918489998 mfu:25.296081779562062 +Step 26008 | loss:0.447855144739151 lr:1.8646159530762953e-05 tokens_per_second_per_gpu:4681.042092582242 mfu:22.08127421884748 +Step 26009 | loss:0.4482833445072174 lr:1.8642781916536922e-05 tokens_per_second_per_gpu:5336.007009893358 mfu:25.170855482342052 +Step 26010 | loss:0.4167819023132324 lr:1.8639404538152518e-05 tokens_per_second_per_gpu:3891.937342108303 mfu:18.35893247570834 +Step 26011 | loss:0.3737199902534485 lr:1.8636027395635154e-05 tokens_per_second_per_gpu:5113.996526169538 mfu:24.123594151722322 +Step 26012 | loss:0.38628971576690674 lr:1.863265048901024e-05 tokens_per_second_per_gpu:4218.965450194618 mfu:19.90157985830031 +Step 26013 | loss:0.3589516580104828 lr:1.8629273818303156e-05 tokens_per_second_per_gpu:5064.805579452751 mfu:23.891552063218285 +Step 26014 | loss:0.37142306566238403 lr:1.86258973835393e-05 tokens_per_second_per_gpu:4346.4224961478585 mfu:20.502816490476203 +Step 26015 | loss:0.42257818579673767 lr:1.862252118474409e-05 tokens_per_second_per_gpu:3550.0701536582997 mfu:16.746286619232677 +Step 26016 | loss:0.4427182972431183 lr:1.8619145221942906e-05 tokens_per_second_per_gpu:5351.743949589431 mfu:25.245089311887206 +Step 26017 | loss:0.33952784538269043 lr:1.8615769495161116e-05 tokens_per_second_per_gpu:5075.375609305744 mfu:23.941412697467907 +Step 26018 | loss:0.4259219765663147 lr:1.8612394004424126e-05 tokens_per_second_per_gpu:5364.48362652315 mfu:25.305184541596578 +Step 26019 | loss:0.27925431728363037 lr:1.860901874975735e-05 tokens_per_second_per_gpu:5334.364776624435 mfu:25.163108787818654 +Step 26020 | loss:0.43483269214630127 lr:1.860564373118613e-05 tokens_per_second_per_gpu:3572.5981989355514 mfu:16.852555252486344 +Step 26021 | loss:0.46088293194770813 lr:1.8602268948735863e-05 tokens_per_second_per_gpu:4244.86531475529 mfu:20.02375393840556 +Step 26022 | loss:0.4132434129714966 lr:1.859889440243196e-05 tokens_per_second_per_gpu:5283.990850764886 mfu:24.925486384861152 +Step 26023 | loss:0.435859739780426 lr:1.8595520092299777e-05 tokens_per_second_per_gpu:5136.762313880628 mfu:24.23098425659945 +Step 26024 | loss:0.5972257852554321 lr:1.859214601836468e-05 tokens_per_second_per_gpu:4250.096584781669 mfu:20.048430731666695 +Step 26025 | loss:0.32146358489990234 lr:1.8588772180652057e-05 tokens_per_second_per_gpu:5326.56247162935 mfu:25.126303983946332 +Step 26026 | loss:0.3985241949558258 lr:1.8585398579187307e-05 tokens_per_second_per_gpu:4186.795365063159 mfu:19.749828077952703 +Step 26027 | loss:0.5046388506889343 lr:1.8582025213995763e-05 tokens_per_second_per_gpu:3506.570585072412 mfu:16.541091732421332 +Step 26028 | loss:0.45078086853027344 lr:1.857865208510282e-05 tokens_per_second_per_gpu:4223.036218178432 mfu:19.920782365424593 +Step 26029 | loss:0.5631422400474548 lr:1.8575279192533856e-05 tokens_per_second_per_gpu:4589.8561784074145 mfu:21.651134703763578 +Step 26030 | loss:0.413793683052063 lr:1.8571906536314232e-05 tokens_per_second_per_gpu:3311.0674991543938 mfu:15.618870319879992 +Step 26031 | loss:0.3055255115032196 lr:1.856853411646929e-05 tokens_per_second_per_gpu:4156.35164245124 mfu:19.606219844157593 +Step 26032 | loss:0.4222624897956848 lr:1.8565161933024416e-05 tokens_per_second_per_gpu:4749.8506508460905 mfu:22.40585592812868 +Step 26033 | loss:0.47172433137893677 lr:1.8561789986004985e-05 tokens_per_second_per_gpu:4572.417384943469 mfu:21.568872939629173 +Step 26034 | loss:0.4828144609928131 lr:1.8558418275436325e-05 tokens_per_second_per_gpu:5074.358487315707 mfu:23.936614759502 +Step 26035 | loss:0.44710731506347656 lr:1.855504680134381e-05 tokens_per_second_per_gpu:4498.708113735201 mfu:21.221173731241294 +Step 26036 | loss:0.4201335608959198 lr:1.8551675563752806e-05 tokens_per_second_per_gpu:4284.457401526427 mfu:20.210516566811584 +Step 26037 | loss:0.40877482295036316 lr:1.854830456268868e-05 tokens_per_second_per_gpu:4196.317981407895 mfu:19.794747883976708 +Step 26038 | loss:0.42968228459358215 lr:1.8544933798176767e-05 tokens_per_second_per_gpu:5328.790589946533 mfu:25.13681439820467 +Step 26039 | loss:0.5028572678565979 lr:1.85415632702424e-05 tokens_per_second_per_gpu:4228.871033858928 mfu:19.948306186511704 +Step 26040 | loss:0.40727975964546204 lr:1.8538192978910956e-05 tokens_per_second_per_gpu:4039.2178492232847 mfu:19.053679756415406 +Step 26041 | loss:0.45068877935409546 lr:1.853482292420779e-05 tokens_per_second_per_gpu:4568.458259304903 mfu:21.550197068495564 +Step 26042 | loss:0.36148297786712646 lr:1.853145310615822e-05 tokens_per_second_per_gpu:5347.7128988697195 mfu:25.226074157874155 +Step 26043 | loss:0.4511924684047699 lr:1.852808352478761e-05 tokens_per_second_per_gpu:4929.933493600323 mfu:23.25533742270166 +Step 26044 | loss:0.35606148838996887 lr:1.852471418012131e-05 tokens_per_second_per_gpu:5275.122932042565 mfu:24.88365489922501 +Step 26045 | loss:0.46915796399116516 lr:1.852134507218466e-05 tokens_per_second_per_gpu:5322.830809882191 mfu:25.108701098797837 +Step 26046 | loss:0.3543862998485565 lr:1.8517976201002964e-05 tokens_per_second_per_gpu:4700.333277178793 mfu:22.172274028004903 +Step 26047 | loss:0.4637395143508911 lr:1.851460756660159e-05 tokens_per_second_per_gpu:5332.79914614325 mfu:25.155723441714798 +Step 26048 | loss:0.4169922173023224 lr:1.851123916900589e-05 tokens_per_second_per_gpu:4581.821827127554 mfu:21.613235297974796 +Step 26049 | loss:0.47149786353111267 lr:1.8507871008241156e-05 tokens_per_second_per_gpu:4549.945759959331 mfu:21.462870450524413 +Step 26050 | loss:0.41464611887931824 lr:1.850450308433275e-05 tokens_per_second_per_gpu:4245.747159022411 mfu:20.027913748271718 +Step 26051 | loss:0.3551011383533478 lr:1.8501135397306002e-05 tokens_per_second_per_gpu:4457.468352170037 mfu:21.026638739709544 +Step 26052 | loss:0.33943402767181396 lr:1.849776794718624e-05 tokens_per_second_per_gpu:5369.928990944888 mfu:25.330871254648894 +Step 26053 | loss:0.5540269613265991 lr:1.849440073399877e-05 tokens_per_second_per_gpu:3633.2775709808107 mfu:17.13879020339227 +Step 26054 | loss:0.44780343770980835 lr:1.849103375776893e-05 tokens_per_second_per_gpu:4152.939502189782 mfu:19.590124196373136 +Step 26055 | loss:0.46042922139167786 lr:1.8487667018522058e-05 tokens_per_second_per_gpu:4322.03466274451 mfu:20.38777491931916 +Step 26056 | loss:0.465655118227005 lr:1.848430051628345e-05 tokens_per_second_per_gpu:3702.0596496966086 mfu:17.463247004127435 +Step 26057 | loss:0.3474123477935791 lr:1.8480934251078435e-05 tokens_per_second_per_gpu:5346.960246605302 mfu:25.22252376872718 +Step 26058 | loss:0.3366074562072754 lr:1.8477568222932363e-05 tokens_per_second_per_gpu:4557.218069674749 mfu:21.497175176236023 +Step 26059 | loss:0.3624861240386963 lr:1.8474202431870515e-05 tokens_per_second_per_gpu:4781.50727901732 mfu:22.555185644390082 +Step 26060 | loss:0.37552541494369507 lr:1.8470836877918196e-05 tokens_per_second_per_gpu:5074.012666330896 mfu:23.9349834629135 +Step 26061 | loss:0.4405660033226013 lr:1.8467471561100736e-05 tokens_per_second_per_gpu:4576.303580231079 mfu:21.587204785854116 +Step 26062 | loss:0.43203213810920715 lr:1.8464106481443466e-05 tokens_per_second_per_gpu:4279.288221839446 mfu:20.186132664275092 +Step 26063 | loss:0.41662514209747314 lr:1.8460741638971657e-05 tokens_per_second_per_gpu:4291.511974156937 mfu:20.2437941895439 +Step 26064 | loss:0.4356652796268463 lr:1.845737703371063e-05 tokens_per_second_per_gpu:4854.149176571705 mfu:22.897849869139865 +Step 26065 | loss:0.47317078709602356 lr:1.845401266568572e-05 tokens_per_second_per_gpu:3530.9830020089435 mfu:16.65624926829878 +Step 26066 | loss:0.3574804961681366 lr:1.8450648534922188e-05 tokens_per_second_per_gpu:5397.539969052958 mfu:25.4611169493033 +Step 26067 | loss:0.27117547392845154 lr:1.8447284641445367e-05 tokens_per_second_per_gpu:5341.300667908072 mfu:25.195826570388647 +Step 26068 | loss:0.336259663105011 lr:1.8443920985280528e-05 tokens_per_second_per_gpu:5318.754844700725 mfu:25.08947407560549 +Step 26069 | loss:0.3821994364261627 lr:1.8440557566453002e-05 tokens_per_second_per_gpu:5328.208612503522 mfu:25.1340691113104 +Step 26070 | loss:0.3946782946586609 lr:1.843719438498806e-05 tokens_per_second_per_gpu:5331.50543964456 mfu:25.14962080743139 +Step 26071 | loss:0.41999873518943787 lr:1.8433831440910998e-05 tokens_per_second_per_gpu:4299.276239088666 mfu:20.28041955194707 +Step 26072 | loss:0.4723723828792572 lr:1.843046873424712e-05 tokens_per_second_per_gpu:4293.830373658562 mfu:20.25473047555281 +Step 26073 | loss:0.49379703402519226 lr:1.8427106265021733e-05 tokens_per_second_per_gpu:3734.8586801840283 mfu:17.617965627028244 +Step 26074 | loss:0.43929699063301086 lr:1.842374403326011e-05 tokens_per_second_per_gpu:5348.53457587412 mfu:25.229950148496464 +Step 26075 | loss:0.4955500662326813 lr:1.8420382038987515e-05 tokens_per_second_per_gpu:5313.811256650807 mfu:25.066154327310933 +Step 26076 | loss:0.51813805103302 lr:1.8417020282229265e-05 tokens_per_second_per_gpu:4322.045163417047 mfu:20.387824452783114 +Step 26077 | loss:0.301272451877594 lr:1.8413658763010645e-05 tokens_per_second_per_gpu:4490.841076964783 mfu:21.184063576540353 +Step 26078 | loss:0.5310233235359192 lr:1.8410297481356915e-05 tokens_per_second_per_gpu:5370.725538317328 mfu:25.334628704509687 +Step 26079 | loss:0.43484729528427124 lr:1.8406936437293366e-05 tokens_per_second_per_gpu:5318.585329242884 mfu:25.088674442268534 +Step 26080 | loss:0.3610168397426605 lr:1.8403575630845293e-05 tokens_per_second_per_gpu:5337.934539464876 mfu:25.179947968201358 +Step 26081 | loss:0.4343969225883484 lr:1.8400215062037962e-05 tokens_per_second_per_gpu:3730.3460330879807 mfu:17.5966787007392 +Step 26082 | loss:0.4396272599697113 lr:1.8396854730896633e-05 tokens_per_second_per_gpu:5001.423212661607 mfu:23.592566624917566 +Step 26083 | loss:0.38763442635536194 lr:1.839349463744659e-05 tokens_per_second_per_gpu:5358.34152877226 mfu:25.276211218555698 +Step 26084 | loss:0.451969712972641 lr:1.839013478171312e-05 tokens_per_second_per_gpu:5113.473481525445 mfu:24.121126860895515 +Step 26085 | loss:0.39399030804634094 lr:1.8386775163721463e-05 tokens_per_second_per_gpu:5356.339902828856 mfu:25.266769207468055 +Step 26086 | loss:0.49200719594955444 lr:1.838341578349691e-05 tokens_per_second_per_gpu:4459.982883510266 mfu:21.03850021306469 +Step 26087 | loss:0.4661271274089813 lr:1.8380056641064736e-05 tokens_per_second_per_gpu:4859.202206759447 mfu:22.9216858746713 +Step 26088 | loss:0.4336756467819214 lr:1.837669773645019e-05 tokens_per_second_per_gpu:5156.353621560912 mfu:24.323399797549115 +Step 26089 | loss:0.4358372092247009 lr:1.8373339069678515e-05 tokens_per_second_per_gpu:5066.561324353221 mfu:23.899834211474772 +Step 26090 | loss:0.4547075629234314 lr:1.8369980640774997e-05 tokens_per_second_per_gpu:4985.272825303306 mfu:23.51638249220823 +Step 26091 | loss:0.49127665162086487 lr:1.8366622449764905e-05 tokens_per_second_per_gpu:5127.900729891961 mfu:24.18918265298277 +Step 26092 | loss:0.32008054852485657 lr:1.8363264496673465e-05 tokens_per_second_per_gpu:4972.752875615759 mfu:23.457323753408385 +Step 26093 | loss:0.3690425753593445 lr:1.8359906781525953e-05 tokens_per_second_per_gpu:3729.425947882717 mfu:17.592338501843248 +Step 26094 | loss:0.3952765464782715 lr:1.8356549304347632e-05 tokens_per_second_per_gpu:3843.8606332190134 mfu:18.132146437145188 +Step 26095 | loss:0.316010445356369 lr:1.8353192065163726e-05 tokens_per_second_per_gpu:5329.01932094403 mfu:25.137893361345398 +Step 26096 | loss:0.44853588938713074 lr:1.834983506399952e-05 tokens_per_second_per_gpu:5348.411891973344 mfu:25.229371427604473 +Step 26097 | loss:0.40972602367401123 lr:1.8346478300880222e-05 tokens_per_second_per_gpu:5327.388443079906 mfu:25.13020023220387 +Step 26098 | loss:0.4061407148838043 lr:1.834312177583112e-05 tokens_per_second_per_gpu:3445.057245937178 mfu:16.250922816462538 +Step 26099 | loss:0.37109699845314026 lr:1.833976548887742e-05 tokens_per_second_per_gpu:4740.461342678913 mfu:22.361564959522937 +Step 26100 | loss:0.39642518758773804 lr:1.8336409440044384e-05 tokens_per_second_per_gpu:4205.711510595155 mfu:19.83905876385436 +Step 26101 | loss:0.4222317934036255 lr:1.833305362935727e-05 tokens_per_second_per_gpu:5156.332723162892 mfu:24.32330121624002 +Step 26102 | loss:0.4064001142978668 lr:1.8329698056841284e-05 tokens_per_second_per_gpu:4930.8468794447945 mfu:23.259646019569725 +Step 26103 | loss:0.36913052201271057 lr:1.8326342722521695e-05 tokens_per_second_per_gpu:5110.643911235334 mfu:24.10777929506259 +Step 26104 | loss:0.44927576184272766 lr:1.832298762642371e-05 tokens_per_second_per_gpu:3557.5850358617236 mfu:16.781735600757344 +Step 26105 | loss:0.41224202513694763 lr:1.831963276857259e-05 tokens_per_second_per_gpu:4844.515826812319 mfu:22.85240771470589 +Step 26106 | loss:0.3342139720916748 lr:1.8316278148993542e-05 tokens_per_second_per_gpu:4989.613690543411 mfu:23.536859094173792 +Step 26107 | loss:0.43434232473373413 lr:1.8312923767711803e-05 tokens_per_second_per_gpu:3328.1414516792815 mfu:15.69941106705658 +Step 26108 | loss:0.3552105724811554 lr:1.8309569624752613e-05 tokens_per_second_per_gpu:4956.159456794749 mfu:23.379049765700493 +Step 26109 | loss:0.41694119572639465 lr:1.8306215720141214e-05 tokens_per_second_per_gpu:5099.510907468246 mfu:24.055263016806926 +Step 26110 | loss:0.4190741181373596 lr:1.83028620539028e-05 tokens_per_second_per_gpu:4264.22992746672 mfu:20.115100120509 +Step 26111 | loss:0.3917124271392822 lr:1.8299508626062595e-05 tokens_per_second_per_gpu:4996.90225057519 mfu:23.571240475400263 +Step 26112 | loss:0.3538951277732849 lr:1.829615543664583e-05 tokens_per_second_per_gpu:2987.5744296367 mfu:14.092898317355989 +Step 26113 | loss:0.34332412481307983 lr:1.829280248567774e-05 tokens_per_second_per_gpu:4803.271479707822 mfu:22.65785109240273 +Step 26114 | loss:0.32710543274879456 lr:1.8289449773183515e-05 tokens_per_second_per_gpu:5343.19957672855 mfu:25.204784047282097 +Step 26115 | loss:0.48353394865989685 lr:1.828609729918838e-05 tokens_per_second_per_gpu:4591.293204215962 mfu:21.65791339968438 +Step 26116 | loss:0.42211759090423584 lr:1.8282745063717576e-05 tokens_per_second_per_gpu:5000.320292390239 mfu:23.587363961819918 +Step 26117 | loss:0.35207459330558777 lr:1.827939306679629e-05 tokens_per_second_per_gpu:3772.9695150094 mfu:17.797740937278448 +Step 26118 | loss:0.391915887594223 lr:1.827604130844972e-05 tokens_per_second_per_gpu:4629.102159539915 mfu:21.83626469282044 +Step 26119 | loss:0.574429452419281 lr:1.8272689788703096e-05 tokens_per_second_per_gpu:4085.408532640656 mfu:19.271569091038398 +Step 26120 | loss:0.3161834478378296 lr:1.826933850758163e-05 tokens_per_second_per_gpu:4731.487484348307 mfu:22.31923373867935 +Step 26121 | loss:0.3558703660964966 lr:1.8265987465110506e-05 tokens_per_second_per_gpu:4187.692655307705 mfu:19.75406074912973 +Step 26122 | loss:0.3816341161727905 lr:1.8262636661314935e-05 tokens_per_second_per_gpu:5256.771203805686 mfu:24.79708666600395 +Step 26123 | loss:0.2809886038303375 lr:1.8259286096220146e-05 tokens_per_second_per_gpu:4492.658188189778 mfu:21.1926352002198 +Step 26124 | loss:0.5036992430686951 lr:1.8255935769851313e-05 tokens_per_second_per_gpu:5380.721676382348 mfu:25.38178219327941 +Step 26125 | loss:0.4721628725528717 lr:1.825258568223362e-05 tokens_per_second_per_gpu:5331.796510891562 mfu:25.150993840165373 +Step 26126 | loss:0.38000190258026123 lr:1.824923583339228e-05 tokens_per_second_per_gpu:4839.582763209157 mfu:22.829137612022333 +Step 26127 | loss:0.47186046838760376 lr:1.824588622335251e-05 tokens_per_second_per_gpu:3743.3561190662094 mfu:17.65804949604807 +Step 26128 | loss:0.5653385519981384 lr:1.8242536852139464e-05 tokens_per_second_per_gpu:5342.171048969715 mfu:25.199932306358743 +Step 26129 | loss:0.3171197474002838 lr:1.823918771977835e-05 tokens_per_second_per_gpu:5275.553771992816 mfu:24.885687244779223 +Step 26130 | loss:0.5210679769515991 lr:1.8235838826294372e-05 tokens_per_second_per_gpu:5345.611770384643 mfu:25.216162776320562 +Step 26131 | loss:0.4315333664417267 lr:1.8232490171712692e-05 tokens_per_second_per_gpu:3358.0801210138343 mfu:15.840636878371381 +Step 26132 | loss:0.3506733477115631 lr:1.8229141756058516e-05 tokens_per_second_per_gpu:5108.046262215873 mfu:24.095525741433164 +Step 26133 | loss:0.3498546779155731 lr:1.822579357935701e-05 tokens_per_second_per_gpu:4961.75278930477 mfu:23.405434469471196 +Step 26134 | loss:0.40654656291007996 lr:1.8222445641633373e-05 tokens_per_second_per_gpu:4952.926459307397 mfu:23.363799164946403 +Step 26135 | loss:0.49078962206840515 lr:1.8219097942912765e-05 tokens_per_second_per_gpu:4549.551849974077 mfu:21.461012309916 +Step 26136 | loss:0.42882248759269714 lr:1.821575048322038e-05 tokens_per_second_per_gpu:4626.988906773535 mfu:21.82629611896332 +Step 26137 | loss:0.36046522855758667 lr:1.8212403262581406e-05 tokens_per_second_per_gpu:5339.959952531792 mfu:25.189502187209104 +Step 26138 | loss:0.4419701397418976 lr:1.8209056281020986e-05 tokens_per_second_per_gpu:3756.4549958641123 mfu:17.719839132802612 +Step 26139 | loss:0.4721825122833252 lr:1.8205709538564326e-05 tokens_per_second_per_gpu:5355.857472920025 mfu:25.264493503277073 +Step 26140 | loss:0.38274312019348145 lr:1.8202363035236563e-05 tokens_per_second_per_gpu:5356.517612027515 mfu:25.267607492078447 +Step 26141 | loss:0.3269740641117096 lr:1.8199016771062905e-05 tokens_per_second_per_gpu:5336.420376318686 mfu:25.172805402297907 +Step 26142 | loss:0.4502370357513428 lr:1.8195670746068484e-05 tokens_per_second_per_gpu:4296.514591988168 mfu:20.267392391388434 +Step 26143 | loss:0.4248010814189911 lr:1.819232496027848e-05 tokens_per_second_per_gpu:4064.9285527057477 mfu:19.17496153144054 +Step 26144 | loss:0.4535065293312073 lr:1.8188979413718055e-05 tokens_per_second_per_gpu:3682.3746684624903 mfu:17.37038958904197 +Step 26145 | loss:0.3921208679676056 lr:1.8185634106412397e-05 tokens_per_second_per_gpu:5329.683026754507 mfu:25.141024174893307 +Step 26146 | loss:0.4057328999042511 lr:1.818228903838664e-05 tokens_per_second_per_gpu:4228.347645780018 mfu:19.94583727564427 +Step 26147 | loss:0.4305940568447113 lr:1.8178944209665933e-05 tokens_per_second_per_gpu:5212.559163841455 mfu:24.588530930140806 +Step 26148 | loss:0.4358566999435425 lr:1.8175599620275464e-05 tokens_per_second_per_gpu:2306.1982489487586 mfu:10.87873061828673 +Step 26149 | loss:0.42212429642677307 lr:1.817225527024036e-05 tokens_per_second_per_gpu:4509.970083612315 mfu:21.27429836464184 +Step 26150 | loss:0.4358938932418823 lr:1.8168911159585773e-05 tokens_per_second_per_gpu:5304.172998814061 mfu:25.020689020638645 +Step 26151 | loss:0.2917833626270294 lr:1.8165567288336877e-05 tokens_per_second_per_gpu:4374.944507863663 mfu:20.63735968609665 +Step 26152 | loss:0.4465216398239136 lr:1.8162223656518822e-05 tokens_per_second_per_gpu:4573.307959779299 mfu:21.57307393307752 +Step 26153 | loss:0.5561264157295227 lr:1.8158880264156747e-05 tokens_per_second_per_gpu:4673.255113805562 mfu:22.044541711362303 +Step 26154 | loss:0.48416486382484436 lr:1.8155537111275783e-05 tokens_per_second_per_gpu:5370.826336522052 mfu:25.335104186839317 +Step 26155 | loss:0.4481068551540375 lr:1.8152194197901085e-05 tokens_per_second_per_gpu:5340.036558980116 mfu:25.189863552896256 +Step 26156 | loss:0.4011477828025818 lr:1.8148851524057815e-05 tokens_per_second_per_gpu:4335.579513796646 mfu:20.451668292721976 +Step 26157 | loss:0.6006405353546143 lr:1.8145509089771078e-05 tokens_per_second_per_gpu:5345.492850751329 mfu:25.215601812120436 +Step 26158 | loss:0.5259685516357422 lr:1.814216689506604e-05 tokens_per_second_per_gpu:4530.713817968857 mfu:21.372150098847687 +Step 26159 | loss:0.36436396837234497 lr:1.8138824939967836e-05 tokens_per_second_per_gpu:4218.815066959277 mfu:19.900870474921177 +Step 26160 | loss:0.3139837980270386 lr:1.8135483224501583e-05 tokens_per_second_per_gpu:4676.496217096981 mfu:22.059830548577168 +Step 26161 | loss:0.4336247742176056 lr:1.8132141748692443e-05 tokens_per_second_per_gpu:4034.273749596956 mfu:19.030357594927633 +Step 26162 | loss:0.48206081986427307 lr:1.8128800512565513e-05 tokens_per_second_per_gpu:4802.692402210261 mfu:22.655119485045926 +Step 26163 | loss:0.4172396957874298 lr:1.8125459516145966e-05 tokens_per_second_per_gpu:5386.853768970491 mfu:25.410708320260557 +Step 26164 | loss:0.3739814758300781 lr:1.8122118759458878e-05 tokens_per_second_per_gpu:4810.858185854506 mfu:22.693638879722183 +Step 26165 | loss:0.49843108654022217 lr:1.8118778242529405e-05 tokens_per_second_per_gpu:5352.409279444906 mfu:25.24822778633246 +Step 26166 | loss:0.41800448298454285 lr:1.8115437965382692e-05 tokens_per_second_per_gpu:5247.531340449818 mfu:24.753500654071065 +Step 26167 | loss:0.3690910339355469 lr:1.8112097928043813e-05 tokens_per_second_per_gpu:5354.928034249835 mfu:25.260109182491195 +Step 26168 | loss:0.45491302013397217 lr:1.810875813053794e-05 tokens_per_second_per_gpu:5340.203145595305 mfu:25.190649370382708 +Step 26169 | loss:0.37715789675712585 lr:1.810541857289014e-05 tokens_per_second_per_gpu:3623.3023630240355 mfu:17.091735445514182 +Step 26170 | loss:0.40284988284111023 lr:1.810207925512558e-05 tokens_per_second_per_gpu:4358.2580535336365 mfu:20.558646834019235 +Step 26171 | loss:0.40389928221702576 lr:1.8098740177269335e-05 tokens_per_second_per_gpu:5300.424184531395 mfu:25.003005224053638 +Step 26172 | loss:0.4585157334804535 lr:1.8095401339346534e-05 tokens_per_second_per_gpu:4421.268199094188 mfu:20.85587644126776 +Step 26173 | loss:0.36407601833343506 lr:1.809206274138231e-05 tokens_per_second_per_gpu:4638.914770681718 mfu:21.882552453780423 +Step 26174 | loss:0.4533093571662903 lr:1.8088724383401733e-05 tokens_per_second_per_gpu:4558.933589449652 mfu:21.505267575712097 +Step 26175 | loss:0.43176352977752686 lr:1.8085386265429954e-05 tokens_per_second_per_gpu:3277.2010691870014 mfu:15.459116591515022 +Step 26176 | loss:0.33742767572402954 lr:1.8082048387492036e-05 tokens_per_second_per_gpu:4151.882905552755 mfu:19.585140049762305 +Step 26177 | loss:0.3602658808231354 lr:1.807871074961312e-05 tokens_per_second_per_gpu:5337.132152224806 mfu:25.17616297068857 +Step 26178 | loss:0.3145892322063446 lr:1.807537335181828e-05 tokens_per_second_per_gpu:3943.023115824706 mfu:18.59991278646039 +Step 26179 | loss:0.433529257774353 lr:1.807203619413263e-05 tokens_per_second_per_gpu:3156.725435992645 mfu:14.890812474474627 +Step 26180 | loss:0.5261333584785461 lr:1.8068699276581285e-05 tokens_per_second_per_gpu:3288.697588953373 mfu:15.51334763676153 +Step 26181 | loss:0.3996634781360626 lr:1.8065362599189315e-05 tokens_per_second_per_gpu:5387.452010961355 mfu:25.413530329802075 +Step 26182 | loss:0.4128393828868866 lr:1.8062026161981843e-05 tokens_per_second_per_gpu:4010.415761154105 mfu:18.91781539285013 +Step 26183 | loss:0.49168747663497925 lr:1.8058689964983925e-05 tokens_per_second_per_gpu:5047.816047499109 mfu:23.8114095422799 +Step 26184 | loss:0.49205583333969116 lr:1.80553540082207e-05 tokens_per_second_per_gpu:4441.184613818589 mfu:20.949825567613512 +Step 26185 | loss:0.5021626949310303 lr:1.8052018291717214e-05 tokens_per_second_per_gpu:5337.356815313814 mfu:25.177222744810894 +Step 26186 | loss:0.5907819271087646 lr:1.8048682815498574e-05 tokens_per_second_per_gpu:4348.088166699199 mfu:20.510673742659577 +Step 26187 | loss:0.5417359471321106 lr:1.8045347579589867e-05 tokens_per_second_per_gpu:4434.139014957481 mfu:20.91659027568254 +Step 26188 | loss:0.2953912615776062 lr:1.80420125840162e-05 tokens_per_second_per_gpu:4480.920203746164 mfu:21.13726512489261 +Step 26189 | loss:0.4216408133506775 lr:1.8038677828802607e-05 tokens_per_second_per_gpu:5057.513583368593 mfu:23.857154473546483 +Step 26190 | loss:0.36861276626586914 lr:1.8035343313974223e-05 tokens_per_second_per_gpu:5008.986887174547 mfu:23.62824576809104 +Step 26191 | loss:0.42672866582870483 lr:1.803200903955608e-05 tokens_per_second_per_gpu:5150.422072740755 mfu:24.29541966973785 +Step 26192 | loss:0.4069700837135315 lr:1.8028675005573298e-05 tokens_per_second_per_gpu:5318.541683414968 mfu:25.08846855745151 +Step 26193 | loss:0.4724016487598419 lr:1.8025341212050904e-05 tokens_per_second_per_gpu:5311.13921065254 mfu:25.05354983043905 +Step 26194 | loss:0.3162105083465576 lr:1.8022007659014005e-05 tokens_per_second_per_gpu:5345.385825493627 mfu:25.215096955719467 +Step 26195 | loss:0.33439329266548157 lr:1.8018674346487687e-05 tokens_per_second_per_gpu:4781.808220277871 mfu:22.556605235661976 +Step 26196 | loss:0.3509186804294586 lr:1.8015341274496977e-05 tokens_per_second_per_gpu:4914.683754291652 mfu:23.183401800508832 +Step 26197 | loss:0.48187482357025146 lr:1.8012008443066985e-05 tokens_per_second_per_gpu:2964.2601098453506 mfu:13.982920693067054 +Step 26198 | loss:0.41096675395965576 lr:1.8008675852222746e-05 tokens_per_second_per_gpu:5340.053672986944 mfu:25.189944282586413 +Step 26199 | loss:0.40897274017333984 lr:1.8005343501989353e-05 tokens_per_second_per_gpu:3730.524393911947 mfu:17.59752005917734 +Step 26200 | loss:0.32676756381988525 lr:1.8002011392391833e-05 tokens_per_second_per_gpu:5256.984570519191 mfu:24.798093153195666 +Step 26201 | loss:0.47899556159973145 lr:1.799867952345527e-05 tokens_per_second_per_gpu:3971.889841024428 mfu:18.73608205439866 +Step 26202 | loss:0.5606476068496704 lr:1.799534789520474e-05 tokens_per_second_per_gpu:5287.559525296482 mfu:24.942320431505355 +Step 26203 | loss:0.5255308151245117 lr:1.7992016507665267e-05 tokens_per_second_per_gpu:4063.1236506068885 mfu:19.16644750029217 +Step 26204 | loss:0.49374258518218994 lr:1.798868536086194e-05 tokens_per_second_per_gpu:3760.545617979239 mfu:17.739135295251767 +Step 26205 | loss:0.44277045130729675 lr:1.7985354454819775e-05 tokens_per_second_per_gpu:4230.2503881955 mfu:19.95481283625943 +Step 26206 | loss:0.413973331451416 lr:1.7982023789563864e-05 tokens_per_second_per_gpu:4701.1409392974865 mfu:22.176083908020146 +Step 26207 | loss:0.43608689308166504 lr:1.7978693365119216e-05 tokens_per_second_per_gpu:5375.222395806683 mfu:25.355841148529983 +Step 26208 | loss:0.4480423331260681 lr:1.79753631815109e-05 tokens_per_second_per_gpu:5333.9961590487965 mfu:25.161369955821723 +Step 26209 | loss:0.346954345703125 lr:1.7972033238763985e-05 tokens_per_second_per_gpu:4557.294194807978 mfu:21.49753427147775 +Step 26210 | loss:0.40349817276000977 lr:1.796870353690348e-05 tokens_per_second_per_gpu:5370.226861355878 mfu:25.332276360199465 +Step 26211 | loss:0.40885084867477417 lr:1.7965374075954455e-05 tokens_per_second_per_gpu:5344.2838297981725 mfu:25.209898654004153 +Step 26212 | loss:0.393071711063385 lr:1.7962044855941923e-05 tokens_per_second_per_gpu:5306.235298522359 mfu:25.030417240227298 +Step 26213 | loss:0.4226922392845154 lr:1.795871587689096e-05 tokens_per_second_per_gpu:4860.608675074242 mfu:22.928320425680425 +Step 26214 | loss:0.48712223768234253 lr:1.7955387138826564e-05 tokens_per_second_per_gpu:5030.468398456454 mfu:23.729577721931562 +Step 26215 | loss:0.5800467133522034 lr:1.795205864177379e-05 tokens_per_second_per_gpu:4428.699286137034 mfu:20.89093014672305 +Step 26216 | loss:0.3719419538974762 lr:1.7948730385757688e-05 tokens_per_second_per_gpu:4298.696088691283 mfu:20.277682883538745 +Step 26217 | loss:0.5063830018043518 lr:1.794540237080326e-05 tokens_per_second_per_gpu:3875.247601659951 mfu:18.280204122450115 +Step 26218 | loss:0.37832504510879517 lr:1.794207459693557e-05 tokens_per_second_per_gpu:5343.185576289895 mfu:25.20471800482424 +Step 26219 | loss:0.38378146290779114 lr:1.7938747064179605e-05 tokens_per_second_per_gpu:3839.354037356107 mfu:18.110888055555314 +Step 26220 | loss:0.49795523285865784 lr:1.7935419772560435e-05 tokens_per_second_per_gpu:3332.6008472032468 mfu:15.720446796594457 +Step 26221 | loss:0.3838224709033966 lr:1.793209272210305e-05 tokens_per_second_per_gpu:3593.1397919876817 mfu:16.949453451670415 +Step 26222 | loss:0.33626022934913635 lr:1.7928765912832484e-05 tokens_per_second_per_gpu:5358.736778726216 mfu:25.2780756800251 +Step 26223 | loss:0.4290463626384735 lr:1.792543934477376e-05 tokens_per_second_per_gpu:5233.293512961696 mfu:24.686338392585732 +Step 26224 | loss:0.4175594747066498 lr:1.792211301795192e-05 tokens_per_second_per_gpu:3681.864172352978 mfu:17.367981491793785 +Step 26225 | loss:0.39618271589279175 lr:1.7918786932391944e-05 tokens_per_second_per_gpu:3472.2416583187814 mfu:16.379156327805585 +Step 26226 | loss:0.5010896921157837 lr:1.791546108811888e-05 tokens_per_second_per_gpu:3945.7667442335464 mfu:18.612854949776168 +Step 26227 | loss:0.37885716557502747 lr:1.791213548515771e-05 tokens_per_second_per_gpu:4894.603013283382 mfu:23.08867751904512 +Step 26228 | loss:0.4450654685497284 lr:1.790881012353348e-05 tokens_per_second_per_gpu:5365.53668094012 mfu:25.31015197895104 +Step 26229 | loss:0.41189372539520264 lr:1.7905485003271172e-05 tokens_per_second_per_gpu:5349.459087880706 mfu:25.234311229369546 +Step 26230 | loss:0.35348039865493774 lr:1.79021601243958e-05 tokens_per_second_per_gpu:5297.860679420397 mfu:24.990912733065986 +Step 26231 | loss:0.4953545928001404 lr:1.78988354869324e-05 tokens_per_second_per_gpu:5334.23475421188 mfu:25.162495449914516 +Step 26232 | loss:0.35187581181526184 lr:1.7895511090905936e-05 tokens_per_second_per_gpu:4978.653767856039 mfu:23.48515926890182 +Step 26233 | loss:0.3364719748497009 lr:1.7892186936341445e-05 tokens_per_second_per_gpu:5354.287136485237 mfu:25.257085958387155 +Step 26234 | loss:0.4089128375053406 lr:1.7888863023263895e-05 tokens_per_second_per_gpu:3610.4878627930616 mfu:17.031287261545486 +Step 26235 | loss:0.5224948525428772 lr:1.7885539351698322e-05 tokens_per_second_per_gpu:3909.4692474765043 mfu:18.44163346458165 +Step 26236 | loss:0.3590937852859497 lr:1.7882215921669686e-05 tokens_per_second_per_gpu:3713.4874601972 mfu:17.517153936044252 +Step 26237 | loss:0.4725862145423889 lr:1.7878892733203006e-05 tokens_per_second_per_gpu:4369.113599971892 mfu:20.609854298715693 +Step 26238 | loss:0.5115736126899719 lr:1.7875569786323288e-05 tokens_per_second_per_gpu:5190.985568476172 mfu:24.48676459996784 +Step 26239 | loss:0.4669409692287445 lr:1.7872247081055493e-05 tokens_per_second_per_gpu:4232.127634622553 mfu:19.96366812794711 +Step 26240 | loss:0.35635077953338623 lr:1.7868924617424643e-05 tokens_per_second_per_gpu:5385.556528957664 mfu:25.404589017788084 +Step 26241 | loss:0.3830476999282837 lr:1.7865602395455694e-05 tokens_per_second_per_gpu:4066.74301530368 mfu:19.183520661094228 +Step 26242 | loss:0.4538608193397522 lr:1.7862280415173665e-05 tokens_per_second_per_gpu:3534.8164965669016 mfu:16.674332516190287 +Step 26243 | loss:0.39674532413482666 lr:1.7858958676603516e-05 tokens_per_second_per_gpu:4659.914914532394 mfu:21.981613715318066 +Step 26244 | loss:0.4629863202571869 lr:1.785563717977023e-05 tokens_per_second_per_gpu:4295.60226493998 mfu:20.263088789043085 +Step 26245 | loss:0.43399447202682495 lr:1.785231592469882e-05 tokens_per_second_per_gpu:4459.970433223317 mfu:21.03844148293691 +Step 26246 | loss:0.4477507770061493 lr:1.784899491141423e-05 tokens_per_second_per_gpu:5380.211408754101 mfu:25.379375173816335 +Step 26247 | loss:0.40314939618110657 lr:1.7845674139941464e-05 tokens_per_second_per_gpu:3638.1074926275996 mfu:17.16157376786978 +Step 26248 | loss:0.4684969484806061 lr:1.7842353610305472e-05 tokens_per_second_per_gpu:4113.5903156290815 mfu:19.404507369702795 +Step 26249 | loss:0.48624399304389954 lr:1.7839033322531258e-05 tokens_per_second_per_gpu:3715.6091123203246 mfu:17.52716213110021 +Step 26250 | loss:0.4368610978126526 lr:1.783571327664376e-05 tokens_per_second_per_gpu:4353.81063514577 mfu:20.537667602675846 +Step 26251 | loss:0.36873579025268555 lr:1.783239347266797e-05 tokens_per_second_per_gpu:5065.765320211638 mfu:23.896079324126614 +Step 26252 | loss:0.357891708612442 lr:1.7829073910628863e-05 tokens_per_second_per_gpu:3633.820020245985 mfu:17.141349029127483 +Step 26253 | loss:0.39019134640693665 lr:1.7825754590551387e-05 tokens_per_second_per_gpu:5129.196210048558 mfu:24.195293653913144 +Step 26254 | loss:0.34612762928009033 lr:1.782243551246051e-05 tokens_per_second_per_gpu:4852.574506266573 mfu:22.8904218806441 +Step 26255 | loss:0.4584748446941376 lr:1.781911667638122e-05 tokens_per_second_per_gpu:5168.59735023452 mfu:24.381155554698047 +Step 26256 | loss:0.391920268535614 lr:1.7815798082338464e-05 tokens_per_second_per_gpu:5366.224283687906 mfu:25.313395518429473 +Step 26257 | loss:0.37691909074783325 lr:1.781247973035718e-05 tokens_per_second_per_gpu:4708.976365997875 mfu:22.213044952627527 +Step 26258 | loss:0.4504174590110779 lr:1.780916162046234e-05 tokens_per_second_per_gpu:4415.114255462733 mfu:20.826847239187135 +Step 26259 | loss:0.4085897207260132 lr:1.7805843752678904e-05 tokens_per_second_per_gpu:3699.548353048426 mfu:17.451400789366566 +Step 26260 | loss:0.5384088158607483 lr:1.7802526127031837e-05 tokens_per_second_per_gpu:4195.236343995919 mfu:19.789645615758047 +Step 26261 | loss:0.3406725823879242 lr:1.779920874354607e-05 tokens_per_second_per_gpu:4193.068164728309 mfu:19.779417944223205 +Step 26262 | loss:0.3714865744113922 lr:1.7795891602246574e-05 tokens_per_second_per_gpu:3964.0609952222458 mfu:18.69915205301094 +Step 26263 | loss:0.36087825894355774 lr:1.779257470315827e-05 tokens_per_second_per_gpu:5352.956664628472 mfu:25.250809895636884 +Step 26264 | loss:0.35202187299728394 lr:1.7789258046306144e-05 tokens_per_second_per_gpu:4572.857720233173 mfu:21.57095007631965 +Step 26265 | loss:0.4017116129398346 lr:1.7785941631715096e-05 tokens_per_second_per_gpu:5306.752417896305 mfu:25.032856580543868 +Step 26266 | loss:0.4281775653362274 lr:1.7782625459410098e-05 tokens_per_second_per_gpu:5293.061240541478 mfu:24.96827296100541 +Step 26267 | loss:0.47408050298690796 lr:1.7779309529416093e-05 tokens_per_second_per_gpu:4384.357846243663 mfu:20.68176400931625 +Step 26268 | loss:0.3067723512649536 lr:1.7775993841758004e-05 tokens_per_second_per_gpu:4251.048267839613 mfu:20.052919983024886 +Step 26269 | loss:0.5031595230102539 lr:1.777267839646079e-05 tokens_per_second_per_gpu:5128.334406011531 mfu:24.19122837723156 +Step 26270 | loss:0.45876070857048035 lr:1.7769363193549355e-05 tokens_per_second_per_gpu:5322.916705011262 mfu:25.109106280776885 +Step 26271 | loss:0.3536677360534668 lr:1.776604823304867e-05 tokens_per_second_per_gpu:5350.521523648835 mfu:25.239322920158042 +Step 26272 | loss:0.4976893663406372 lr:1.7762733514983637e-05 tokens_per_second_per_gpu:3829.8840458768987 mfu:18.06621649000073 +Step 26273 | loss:0.4130702316761017 lr:1.7759419039379195e-05 tokens_per_second_per_gpu:5376.8495977639295 mfu:25.36351693779159 +Step 26274 | loss:0.45863279700279236 lr:1.7756104806260293e-05 tokens_per_second_per_gpu:4441.326538506669 mfu:20.950495050582035 +Step 26275 | loss:0.3943113684654236 lr:1.7752790815651826e-05 tokens_per_second_per_gpu:5403.0939130157385 mfu:25.487315850576287 +Step 26276 | loss:0.49981334805488586 lr:1.7749477067578752e-05 tokens_per_second_per_gpu:4176.249968549111 mfu:19.70008364336577 +Step 26277 | loss:0.42635565996170044 lr:1.7746163562065953e-05 tokens_per_second_per_gpu:4767.128363469967 mfu:22.48735784645725 +Step 26278 | loss:0.3735608160495758 lr:1.77428502991384e-05 tokens_per_second_per_gpu:5153.127063117111 mfu:24.30817957086207 +Step 26279 | loss:0.5173914432525635 lr:1.7739537278820955e-05 tokens_per_second_per_gpu:4155.851325676816 mfu:19.603859764569787 +Step 26280 | loss:0.4107432961463928 lr:1.7736224501138577e-05 tokens_per_second_per_gpu:4658.959512275409 mfu:21.977106919863438 +Step 26281 | loss:0.438626229763031 lr:1.773291196611619e-05 tokens_per_second_per_gpu:5336.683429973888 mfu:25.174046271271077 +Step 26282 | loss:0.397410124540329 lr:1.7729599673778668e-05 tokens_per_second_per_gpu:3572.1920148337686 mfu:16.85063921277608 +Step 26283 | loss:0.3132189214229584 lr:1.7726287624150943e-05 tokens_per_second_per_gpu:3274.9588715140803 mfu:15.44853976253348 +Step 26284 | loss:0.32468393445014954 lr:1.7722975817257948e-05 tokens_per_second_per_gpu:5138.200210205457 mfu:24.237767058894075 +Step 26285 | loss:0.4313001036643982 lr:1.771966425312457e-05 tokens_per_second_per_gpu:4661.4665227399155 mfu:21.988932915964607 +Step 26286 | loss:0.38486817479133606 lr:1.7716352931775697e-05 tokens_per_second_per_gpu:3285.3779893548112 mfu:15.497688519073021 +Step 26287 | loss:0.44235730171203613 lr:1.771304185323625e-05 tokens_per_second_per_gpu:3165.968643913194 mfu:14.934414263289748 +Step 26288 | loss:0.43928900361061096 lr:1.7709731017531155e-05 tokens_per_second_per_gpu:3461.6236015004 mfu:16.329069142165977 +Step 26289 | loss:0.41713955998420715 lr:1.7706420424685283e-05 tokens_per_second_per_gpu:5106.894480683703 mfu:24.090092591431723 +Step 26290 | loss:0.3451194167137146 lr:1.7703110074723533e-05 tokens_per_second_per_gpu:4408.02746312236 mfu:20.793417630586138 +Step 26291 | loss:0.4157462418079376 lr:1.769979996767084e-05 tokens_per_second_per_gpu:5359.4318548114925 mfu:25.28135447251156 +Step 26292 | loss:0.47018399834632874 lr:1.7696490103552067e-05 tokens_per_second_per_gpu:4453.834131421334 mfu:21.009495500375913 +Step 26293 | loss:0.47472476959228516 lr:1.7693180482392098e-05 tokens_per_second_per_gpu:3569.9713854141414 mfu:16.840164124925252 +Step 26294 | loss:0.538042426109314 lr:1.768987110421584e-05 tokens_per_second_per_gpu:4762.405115114077 mfu:22.46507747809325 +Step 26295 | loss:0.33353477716445923 lr:1.7686561969048187e-05 tokens_per_second_per_gpu:5301.609174374298 mfu:25.008595023322727 +Step 26296 | loss:0.2777388095855713 lr:1.7683253076914036e-05 tokens_per_second_per_gpu:4510.556099780311 mfu:21.277062702890948 +Step 26297 | loss:0.4775639772415161 lr:1.7679944427838253e-05 tokens_per_second_per_gpu:3467.7896076304346 mfu:16.358155245110478 +Step 26298 | loss:0.3704012632369995 lr:1.7676636021845743e-05 tokens_per_second_per_gpu:4904.053037931953 mfu:23.133254897652893 +Step 26299 | loss:0.46994301676750183 lr:1.7673327858961363e-05 tokens_per_second_per_gpu:3743.254482008516 mfu:17.657570056705858 +Step 26300 | loss:0.36121150851249695 lr:1.7670019939210024e-05 tokens_per_second_per_gpu:5159.266097699877 mfu:24.337138444416283 +Step 26301 | loss:0.44619816541671753 lr:1.766671226261658e-05 tokens_per_second_per_gpu:3736.1352661056494 mfu:17.62398750062855 +Step 26302 | loss:0.3765069842338562 lr:1.7663404829205914e-05 tokens_per_second_per_gpu:4249.793119067586 mfu:20.046999232116775 +Step 26303 | loss:0.38248223066329956 lr:1.766009763900292e-05 tokens_per_second_per_gpu:5181.783451781725 mfu:24.443356645477124 +Step 26304 | loss:0.5344383120536804 lr:1.765679069203245e-05 tokens_per_second_per_gpu:4173.4011495920095 mfu:19.686645278286484 +Step 26305 | loss:0.3648197650909424 lr:1.765348398831939e-05 tokens_per_second_per_gpu:4176.296183863356 mfu:19.700301648888207 +Step 26306 | loss:0.37359949946403503 lr:1.7650177527888595e-05 tokens_per_second_per_gpu:3549.238273742976 mfu:16.742362499738743 +Step 26307 | loss:0.4289715886116028 lr:1.7646871310764963e-05 tokens_per_second_per_gpu:5365.310293843144 mfu:25.30908407238898 +Step 26308 | loss:0.4288296699523926 lr:1.7643565336973317e-05 tokens_per_second_per_gpu:4257.049893612372 mfu:20.08123067577794 +Step 26309 | loss:0.34993842244148254 lr:1.7640259606538544e-05 tokens_per_second_per_gpu:5329.269077681061 mfu:25.139071506486868 +Step 26310 | loss:0.5878493785858154 lr:1.763695411948553e-05 tokens_per_second_per_gpu:3746.6560080938757 mfu:17.673615635610695 +Step 26311 | loss:0.41685524582862854 lr:1.7633648875839092e-05 tokens_per_second_per_gpu:4004.8090071646416 mfu:18.89136737767096 +Step 26312 | loss:0.37999680638313293 lr:1.763034387562411e-05 tokens_per_second_per_gpu:5058.956742086726 mfu:23.86396209944819 +Step 26313 | loss:0.3608761131763458 lr:1.7627039118865464e-05 tokens_per_second_per_gpu:4085.467198635123 mfu:19.27184582854375 +Step 26314 | loss:0.4826084077358246 lr:1.7623734605587983e-05 tokens_per_second_per_gpu:4194.368283445989 mfu:19.785550826037625 +Step 26315 | loss:0.40311095118522644 lr:1.7620430335816512e-05 tokens_per_second_per_gpu:5289.06155388116 mfu:24.94940575661178 +Step 26316 | loss:0.4487561881542206 lr:1.7617126309575915e-05 tokens_per_second_per_gpu:4492.995686986786 mfu:21.194227239628518 +Step 26317 | loss:0.33172184228897095 lr:1.7613822526891062e-05 tokens_per_second_per_gpu:3168.8451952861074 mfu:14.947983446906052 +Step 26318 | loss:0.4744042456150055 lr:1.7610518987786763e-05 tokens_per_second_per_gpu:4490.624861836136 mfu:21.183043652888422 +Step 26319 | loss:0.39797788858413696 lr:1.7607215692287882e-05 tokens_per_second_per_gpu:5352.566066275287 mfu:25.24896737656196 +Step 26320 | loss:0.4553130269050598 lr:1.7603912640419286e-05 tokens_per_second_per_gpu:4649.7497544209755 mfu:21.93366292071735 +Step 26321 | loss:0.45072460174560547 lr:1.7600609832205788e-05 tokens_per_second_per_gpu:5347.6285846861065 mfu:25.22567643348457 +Step 26322 | loss:0.4046157896518707 lr:1.759730726767223e-05 tokens_per_second_per_gpu:4107.572138986858 mfu:19.37611860367461 +Step 26323 | loss:0.29026418924331665 lr:1.7594004946843456e-05 tokens_per_second_per_gpu:5022.093665734137 mfu:23.69007267880365 +Step 26324 | loss:0.5173365473747253 lr:1.7590702869744323e-05 tokens_per_second_per_gpu:5337.378457612663 mfu:25.17732483522712 +Step 26325 | loss:0.5177244544029236 lr:1.758740103639963e-05 tokens_per_second_per_gpu:4161.31042078078 mfu:19.629611247592862 +Step 26326 | loss:0.36132630705833435 lr:1.7584099446834234e-05 tokens_per_second_per_gpu:4832.727756606143 mfu:22.79680137629119 +Step 26327 | loss:0.4356309771537781 lr:1.7580798101072976e-05 tokens_per_second_per_gpu:5357.665323331669 mfu:25.273021442119884 +Step 26328 | loss:0.4270503520965576 lr:1.757749699914067e-05 tokens_per_second_per_gpu:4377.388241575476 mfu:20.64888719495924 +Step 26329 | loss:0.4181984066963196 lr:1.7574196141062134e-05 tokens_per_second_per_gpu:5008.9430261976795 mfu:23.628038868379708 +Step 26330 | loss:0.445842444896698 lr:1.7570895526862202e-05 tokens_per_second_per_gpu:4622.847457006669 mfu:21.806760193808824 +Step 26331 | loss:0.48481354117393494 lr:1.756759515656571e-05 tokens_per_second_per_gpu:5270.494637702388 mfu:24.861822445153955 +Step 26332 | loss:0.4480780363082886 lr:1.756429503019748e-05 tokens_per_second_per_gpu:4198.936525168323 mfu:19.807100001664285 +Step 26333 | loss:0.41862398386001587 lr:1.7560995147782316e-05 tokens_per_second_per_gpu:5359.962613200733 mfu:25.28385815039033 +Step 26334 | loss:0.49379265308380127 lr:1.755769550934506e-05 tokens_per_second_per_gpu:4186.505713856275 mfu:19.748461743790244 +Step 26335 | loss:0.3468445837497711 lr:1.7554396114910504e-05 tokens_per_second_per_gpu:5345.204357041468 mfu:25.214240938068922 +Step 26336 | loss:0.4723955988883972 lr:1.755109696450349e-05 tokens_per_second_per_gpu:3835.7256502503524 mfu:18.093772334510795 +Step 26337 | loss:0.3833439350128174 lr:1.7547798058148794e-05 tokens_per_second_per_gpu:5357.208135176563 mfu:25.270864807587902 +Step 26338 | loss:0.4723994731903076 lr:1.7544499395871255e-05 tokens_per_second_per_gpu:3267.0070864377167 mfu:15.411029835613839 +Step 26339 | loss:0.4239455461502075 lr:1.7541200977695695e-05 tokens_per_second_per_gpu:4130.249059656493 mfu:19.48308950755491 +Step 26340 | loss:0.4249267876148224 lr:1.7537902803646882e-05 tokens_per_second_per_gpu:5247.837056890708 mfu:24.754942770683346 +Step 26341 | loss:0.5026692152023315 lr:1.7534604873749665e-05 tokens_per_second_per_gpu:5013.985641610951 mfu:23.651825745643013 +Step 26342 | loss:0.4760843813419342 lr:1.753130718802881e-05 tokens_per_second_per_gpu:5159.382874062025 mfu:24.337689298440633 +Step 26343 | loss:0.29995429515838623 lr:1.7528009746509154e-05 tokens_per_second_per_gpu:4617.395043151742 mfu:21.781040227377403 +Step 26344 | loss:0.4514285922050476 lr:1.7524712549215466e-05 tokens_per_second_per_gpu:5336.649263068146 mfu:25.173885100147217 +Step 26345 | loss:0.34823301434516907 lr:1.752141559617255e-05 tokens_per_second_per_gpu:5324.4943099620905 mfu:25.116548112497103 +Step 26346 | loss:0.3615071177482605 lr:1.7518118887405238e-05 tokens_per_second_per_gpu:4002.56483515521 mfu:18.880781235407245 +Step 26347 | loss:0.4689755141735077 lr:1.7514822422938276e-05 tokens_per_second_per_gpu:3063.884098707356 mfu:14.452864046134371 +Step 26348 | loss:0.39675647020339966 lr:1.7511526202796487e-05 tokens_per_second_per_gpu:5336.515591378342 mfu:25.173254547979667 +Step 26349 | loss:0.46559226512908936 lr:1.7508230227004662e-05 tokens_per_second_per_gpu:5339.207476386698 mfu:25.185952628847026 +Step 26350 | loss:0.3433854579925537 lr:1.750493449558759e-05 tokens_per_second_per_gpu:5280.184347825994 mfu:24.90753046104203 +Step 26351 | loss:0.41233429312705994 lr:1.750163900857003e-05 tokens_per_second_per_gpu:5179.617943880099 mfu:24.43314157523141 +Step 26352 | loss:0.37370941042900085 lr:1.7498343765976793e-05 tokens_per_second_per_gpu:4836.70536686733 mfu:22.815564442543497 +Step 26353 | loss:0.3864925801753998 lr:1.749504876783268e-05 tokens_per_second_per_gpu:5343.325989714091 mfu:25.20538035890317 +Step 26354 | loss:0.36647549271583557 lr:1.749175401416243e-05 tokens_per_second_per_gpu:5155.6274865784435 mfu:24.319974494945406 +Step 26355 | loss:0.49303537607192993 lr:1.748845950499085e-05 tokens_per_second_per_gpu:4450.445948206527 mfu:20.993512862068 +Step 26356 | loss:0.4269842505455017 lr:1.7485165240342726e-05 tokens_per_second_per_gpu:5358.141125220896 mfu:25.275265880811542 +Step 26357 | loss:0.21401074528694153 lr:1.748187122024283e-05 tokens_per_second_per_gpu:4628.046521213946 mfu:21.831285066726373 +Step 26358 | loss:0.4476391673088074 lr:1.747857744471591e-05 tokens_per_second_per_gpu:4907.509170990023 mfu:23.149558066964914 +Step 26359 | loss:0.49694812297821045 lr:1.7475283913786762e-05 tokens_per_second_per_gpu:5353.390921531765 mfu:25.25285835954159 +Step 26360 | loss:0.39277002215385437 lr:1.747199062748016e-05 tokens_per_second_per_gpu:5331.863531552346 mfu:25.151309988057193 +Step 26361 | loss:0.33134955167770386 lr:1.746869758582086e-05 tokens_per_second_per_gpu:4533.96013954729 mfu:21.387463551612733 +Step 26362 | loss:0.4503469169139862 lr:1.7465404788833633e-05 tokens_per_second_per_gpu:4008.0860593371153 mfu:18.906825791891272 +Step 26363 | loss:0.2302456647157669 lr:1.746211223654327e-05 tokens_per_second_per_gpu:3771.0794443641107 mfu:17.788825151565316 +Step 26364 | loss:0.43780264258384705 lr:1.7458819928974506e-05 tokens_per_second_per_gpu:4619.705712414629 mfu:21.791940048531266 +Step 26365 | loss:0.4265854060649872 lr:1.74555278661521e-05 tokens_per_second_per_gpu:4342.904497150629 mfu:20.486221488973747 +Step 26366 | loss:0.4234381914138794 lr:1.7452236048100816e-05 tokens_per_second_per_gpu:3949.919355811688 mfu:18.632443527098854 +Step 26367 | loss:0.36907777190208435 lr:1.744894447484542e-05 tokens_per_second_per_gpu:3904.840276704452 mfu:18.419797819666027 +Step 26368 | loss:0.4768388569355011 lr:1.7445653146410678e-05 tokens_per_second_per_gpu:5296.2127674928 mfu:24.983139251340745 +Step 26369 | loss:0.4683361053466797 lr:1.744236206282132e-05 tokens_per_second_per_gpu:5364.923024664284 mfu:25.307257257597364 +Step 26370 | loss:0.37308469414711 lr:1.7439071224102134e-05 tokens_per_second_per_gpu:4174.972571043042 mfu:19.69405794138315 +Step 26371 | loss:0.41446927189826965 lr:1.7435780630277838e-05 tokens_per_second_per_gpu:5367.737645977944 mfu:25.320534306557818 +Step 26372 | loss:0.3838832676410675 lr:1.743249028137318e-05 tokens_per_second_per_gpu:4730.795073290018 mfu:22.3159675175799 +Step 26373 | loss:0.527409017086029 lr:1.7429200177412913e-05 tokens_per_second_per_gpu:4656.899902549504 mfu:21.967391389380687 +Step 26374 | loss:0.4263971149921417 lr:1.7425910318421795e-05 tokens_per_second_per_gpu:5349.0857423391535 mfu:25.232550094750724 +Step 26375 | loss:0.3412456214427948 lr:1.7422620704424576e-05 tokens_per_second_per_gpu:4493.264611492462 mfu:21.195495802405045 +Step 26376 | loss:0.33205196261405945 lr:1.7419331335445965e-05 tokens_per_second_per_gpu:4752.5824581440365 mfu:22.418742329247475 +Step 26377 | loss:0.41935303807258606 lr:1.741604221151072e-05 tokens_per_second_per_gpu:4302.656481160266 mfu:20.29636473052757 +Step 26378 | loss:0.44342121481895447 lr:1.74127533326436e-05 tokens_per_second_per_gpu:5325.561976918141 mfu:25.121584479691563 +Step 26379 | loss:0.4565185308456421 lr:1.7409464698869315e-05 tokens_per_second_per_gpu:4228.528054086088 mfu:19.946688292404833 +Step 26380 | loss:0.3664827346801758 lr:1.740617631021259e-05 tokens_per_second_per_gpu:3871.7078399215757 mfu:18.263506462384566 +Step 26381 | loss:0.42543795704841614 lr:1.7402888166698174e-05 tokens_per_second_per_gpu:3795.1104738804383 mfu:17.902183617910257 +Step 26382 | loss:0.4174770414829254 lr:1.7399600268350807e-05 tokens_per_second_per_gpu:4359.540517160305 mfu:20.56469643375709 +Step 26383 | loss:0.4838062822818756 lr:1.7396312615195192e-05 tokens_per_second_per_gpu:4459.702107810488 mfu:21.037175746183564 +Step 26384 | loss:0.36141636967658997 lr:1.7393025207256077e-05 tokens_per_second_per_gpu:4916.181561739865 mfu:23.190467213794356 +Step 26385 | loss:0.3937232494354248 lr:1.738973804455819e-05 tokens_per_second_per_gpu:5309.660207876449 mfu:25.046573122000332 +Step 26386 | loss:0.4501999318599701 lr:1.738645112712624e-05 tokens_per_second_per_gpu:5318.445137413708 mfu:25.088013133491124 +Step 26387 | loss:0.3131839334964752 lr:1.7383164454984945e-05 tokens_per_second_per_gpu:4340.325874156239 mfu:20.474057684350758 +Step 26388 | loss:0.2822220027446747 lr:1.737987802815903e-05 tokens_per_second_per_gpu:5340.466796054398 mfu:25.191893054581662 +Step 26389 | loss:0.3186364471912384 lr:1.737659184667323e-05 tokens_per_second_per_gpu:3492.158201102286 mfu:16.473106058227977 +Step 26390 | loss:0.3557089865207672 lr:1.7373305910552225e-05 tokens_per_second_per_gpu:4852.599080863615 mfu:22.89053780321121 +Step 26391 | loss:0.39756593108177185 lr:1.7370020219820753e-05 tokens_per_second_per_gpu:4745.3747640001075 mfu:22.384742406211135 +Step 26392 | loss:0.40491437911987305 lr:1.736673477450354e-05 tokens_per_second_per_gpu:5272.714239877635 mfu:24.872292687317774 +Step 26393 | loss:0.36020520329475403 lr:1.736344957462528e-05 tokens_per_second_per_gpu:4230.506953965202 mfu:19.95602310076975 +Step 26394 | loss:0.37173721194267273 lr:1.7360164620210655e-05 tokens_per_second_per_gpu:5318.040373758276 mfu:25.08610379426865 +Step 26395 | loss:0.43948879837989807 lr:1.7356879911284407e-05 tokens_per_second_per_gpu:4902.752222927058 mfu:23.127118731332878 +Step 26396 | loss:0.49087095260620117 lr:1.7353595447871245e-05 tokens_per_second_per_gpu:5339.716399119807 mfu:25.188353304210274 +Step 26397 | loss:0.2965741753578186 lr:1.7350311229995834e-05 tokens_per_second_per_gpu:4406.264068799295 mfu:20.785099398698147 +Step 26398 | loss:0.29419705271720886 lr:1.734702725768291e-05 tokens_per_second_per_gpu:5328.643445419149 mfu:25.136120292363557 +Step 26399 | loss:0.35927319526672363 lr:1.7343743530957167e-05 tokens_per_second_per_gpu:4081.903132287189 mfu:19.255033519488116 +Step 26400 | loss:0.3779457211494446 lr:1.73404600498433e-05 tokens_per_second_per_gpu:4248.92435637742 mfu:20.04290113028106 +Step 26401 | loss:0.4860759973526001 lr:1.7337176814365978e-05 tokens_per_second_per_gpu:4108.191863742191 mfu:19.379041951081465 +Step 26402 | loss:0.40280959010124207 lr:1.7333893824549922e-05 tokens_per_second_per_gpu:5342.745136291206 mfu:25.20264037420361 +Step 26403 | loss:0.48296189308166504 lr:1.7330611080419822e-05 tokens_per_second_per_gpu:3910.1392803101658 mfu:18.44479412377742 +Step 26404 | loss:0.3890185058116913 lr:1.732732858200038e-05 tokens_per_second_per_gpu:5396.131341238001 mfu:25.45447221156346 +Step 26405 | loss:0.3710252046585083 lr:1.732404632931625e-05 tokens_per_second_per_gpu:4418.561540358718 mfu:20.843108670208945 +Step 26406 | loss:0.41215065121650696 lr:1.7320764322392134e-05 tokens_per_second_per_gpu:5331.714404233185 mfu:25.150606529048968 +Step 26407 | loss:0.41098329424858093 lr:1.7317482561252758e-05 tokens_per_second_per_gpu:5325.223448998721 mfu:25.11998758573684 +Step 26408 | loss:0.27417680621147156 lr:1.7314201045922735e-05 tokens_per_second_per_gpu:5306.022005392011 mfu:25.029411100139885 +Step 26409 | loss:0.3733780086040497 lr:1.731091977642677e-05 tokens_per_second_per_gpu:4237.249785295188 mfu:19.987830186598952 +Step 26410 | loss:0.3937718868255615 lr:1.730763875278955e-05 tokens_per_second_per_gpu:4753.9282937133985 mfu:22.42509086525138 +Step 26411 | loss:0.4095991849899292 lr:1.730435797503577e-05 tokens_per_second_per_gpu:5245.776272864628 mfu:24.745221700826313 +Step 26412 | loss:0.340335875749588 lr:1.7301077443190067e-05 tokens_per_second_per_gpu:4397.690219845074 mfu:20.744655090331612 +Step 26413 | loss:0.3852202296257019 lr:1.729779715727713e-05 tokens_per_second_per_gpu:3875.3885756643194 mfu:18.280869120881363 +Step 26414 | loss:0.45550042390823364 lr:1.7294517117321658e-05 tokens_per_second_per_gpu:4442.316065035887 mfu:20.95516281605099 +Step 26415 | loss:0.40602582693099976 lr:1.7291237323348284e-05 tokens_per_second_per_gpu:5327.8179852635 mfu:25.132226455971338 +Step 26416 | loss:0.4116334915161133 lr:1.728795777538168e-05 tokens_per_second_per_gpu:4676.88160489553 mfu:22.061648488576857 +Step 26417 | loss:0.33078551292419434 lr:1.728467847344652e-05 tokens_per_second_per_gpu:5336.021754478265 mfu:25.170925035064744 +Step 26418 | loss:0.5302417278289795 lr:1.7281399417567484e-05 tokens_per_second_per_gpu:4413.309657173253 mfu:20.81833464115944 +Step 26419 | loss:0.3873557150363922 lr:1.72781206077692e-05 tokens_per_second_per_gpu:4292.618144899744 mfu:20.249012185669912 +Step 26420 | loss:0.46179646253585815 lr:1.7274842044076346e-05 tokens_per_second_per_gpu:5336.799547299542 mfu:25.174594016507843 +Step 26421 | loss:0.4276760518550873 lr:1.7271563726513594e-05 tokens_per_second_per_gpu:4507.203174508181 mfu:21.261246382314273 +Step 26422 | loss:0.3617929518222809 lr:1.7268285655105592e-05 tokens_per_second_per_gpu:5354.200881022673 mfu:25.256679076653537 +Step 26423 | loss:0.49093860387802124 lr:1.7265007829876973e-05 tokens_per_second_per_gpu:4967.847406561733 mfu:23.43418381892122 +Step 26424 | loss:0.4469975233078003 lr:1.7261730250852405e-05 tokens_per_second_per_gpu:5115.099886963601 mfu:24.12879889284068 +Step 26425 | loss:0.4928356409072876 lr:1.725845291805656e-05 tokens_per_second_per_gpu:3702.492894243784 mfu:17.465290692575604 +Step 26426 | loss:0.35462650656700134 lr:1.725517583151405e-05 tokens_per_second_per_gpu:5374.5351173793815 mfu:25.352599138926603 +Step 26427 | loss:0.4752074182033539 lr:1.7251898991249544e-05 tokens_per_second_per_gpu:5350.7284044452945 mfu:25.240298812172444 +Step 26428 | loss:0.46167340874671936 lr:1.7248622397287695e-05 tokens_per_second_per_gpu:5299.590745651639 mfu:24.999073750658493 +Step 26429 | loss:0.44705337285995483 lr:1.7245346049653137e-05 tokens_per_second_per_gpu:5368.572372788798 mfu:25.324471855343692 +Step 26430 | loss:0.3836669623851776 lr:1.72420699483705e-05 tokens_per_second_per_gpu:4483.365551594156 mfu:21.14880024791096 +Step 26431 | loss:0.39267152547836304 lr:1.7238794093464433e-05 tokens_per_second_per_gpu:5346.85707091883 mfu:25.222037071410504 +Step 26432 | loss:0.44076311588287354 lr:1.723551848495959e-05 tokens_per_second_per_gpu:4741.42322210201 mfu:22.366102308875238 +Step 26433 | loss:0.43558797240257263 lr:1.7232243122880583e-05 tokens_per_second_per_gpu:5020.231645774116 mfu:23.681289212958866 +Step 26434 | loss:0.4133526384830475 lr:1.722896800725205e-05 tokens_per_second_per_gpu:4666.266638747812 mfu:22.011575882158095 +Step 26435 | loss:0.4292103350162506 lr:1.7225693138098647e-05 tokens_per_second_per_gpu:4009.8557180375747 mfu:18.91517357391625 +Step 26436 | loss:0.46778807044029236 lr:1.722241851544499e-05 tokens_per_second_per_gpu:5364.129582766845 mfu:25.303514456791994 +Step 26437 | loss:0.4716677963733673 lr:1.721914413931569e-05 tokens_per_second_per_gpu:5025.163621304159 mfu:23.704554183015876 +Step 26438 | loss:0.4718599319458008 lr:1.7215870009735387e-05 tokens_per_second_per_gpu:3975.95752356145 mfu:18.755270006944006 +Step 26439 | loss:0.4414416551589966 lr:1.7212596126728724e-05 tokens_per_second_per_gpu:4292.992777112536 mfu:20.250779389736497 +Step 26440 | loss:0.3722720146179199 lr:1.72093224903203e-05 tokens_per_second_per_gpu:4189.0662988048025 mfu:19.760540459873287 +Step 26441 | loss:0.42165830731391907 lr:1.720604910053474e-05 tokens_per_second_per_gpu:5354.056111848456 mfu:25.25599617575859 +Step 26442 | loss:0.45520925521850586 lr:1.720277595739667e-05 tokens_per_second_per_gpu:4082.248089585278 mfu:19.25666074191899 +Step 26443 | loss:0.3654269278049469 lr:1.7199503060930744e-05 tokens_per_second_per_gpu:5088.623045713076 mfu:24.003903115247013 +Step 26444 | loss:0.45877620577812195 lr:1.71962304111615e-05 tokens_per_second_per_gpu:4500.451477107018 mfu:21.22939747371462 +Step 26445 | loss:0.39926883578300476 lr:1.7192958008113598e-05 tokens_per_second_per_gpu:3906.7341579748345 mfu:18.428731580747886 +Step 26446 | loss:0.5159711837768555 lr:1.7189685851811644e-05 tokens_per_second_per_gpu:3382.945858365193 mfu:15.957932803990008 +Step 26447 | loss:0.44010257720947266 lr:1.7186413942280265e-05 tokens_per_second_per_gpu:3843.18571827066 mfu:18.128962748181436 +Step 26448 | loss:0.4205693006515503 lr:1.718314227954404e-05 tokens_per_second_per_gpu:4211.997770020268 mfu:19.868712122109077 +Step 26449 | loss:0.3944382071495056 lr:1.717987086362759e-05 tokens_per_second_per_gpu:4958.126427703206 mfu:23.388328303075532 +Step 26450 | loss:0.3511422276496887 lr:1.7176599694555532e-05 tokens_per_second_per_gpu:5315.73693034932 mfu:25.07523805870063 +Step 26451 | loss:0.3110615611076355 lr:1.7173328772352458e-05 tokens_per_second_per_gpu:2804.5140336912323 mfu:13.229371196356988 +Step 26452 | loss:0.4365170896053314 lr:1.717005809704295e-05 tokens_per_second_per_gpu:4498.1213098845 mfu:21.21840567735848 +Step 26453 | loss:0.4295167922973633 lr:1.7166787668651624e-05 tokens_per_second_per_gpu:3334.35602359569 mfu:15.728726263101585 +Step 26454 | loss:0.43862152099609375 lr:1.7163517487203094e-05 tokens_per_second_per_gpu:4609.1092205779005 mfu:21.741954588589483 +Step 26455 | loss:0.5030987858772278 lr:1.7160247552721925e-05 tokens_per_second_per_gpu:5248.0373671645675 mfu:24.7558876684216 +Step 26456 | loss:0.3180374503135681 lr:1.715697786523272e-05 tokens_per_second_per_gpu:4211.929475136274 mfu:19.868389963488966 +Step 26457 | loss:0.39287665486335754 lr:1.7153708424760094e-05 tokens_per_second_per_gpu:5190.815251108639 mfu:24.485961183886047 +Step 26458 | loss:0.44565650820732117 lr:1.715043923132862e-05 tokens_per_second_per_gpu:4131.3326347245775 mfu:19.488200916028035 +Step 26459 | loss:0.40102532505989075 lr:1.7147170284962865e-05 tokens_per_second_per_gpu:3673.1429750825005 mfu:17.32684211627925 +Step 26460 | loss:0.4407324194908142 lr:1.7143901585687433e-05 tokens_per_second_per_gpu:4338.290918699884 mfu:20.46445845226393 +Step 26461 | loss:0.6164948344230652 lr:1.7140633133526928e-05 tokens_per_second_per_gpu:4695.50034237373 mfu:22.149476249945714 +Step 26462 | loss:0.2881320118904114 lr:1.71373649285059e-05 tokens_per_second_per_gpu:3651.647595856599 mfu:17.2254447449808 +Step 26463 | loss:0.44303181767463684 lr:1.7134096970648937e-05 tokens_per_second_per_gpu:4762.710047469267 mfu:22.466515896039507 +Step 26464 | loss:0.5248064994812012 lr:1.7130829259980647e-05 tokens_per_second_per_gpu:5325.037748938077 mfu:25.119111606867083 +Step 26465 | loss:0.5090746879577637 lr:1.712756179652558e-05 tokens_per_second_per_gpu:4346.298866387264 mfu:20.50223330780206 +Step 26466 | loss:0.35586613416671753 lr:1.7124294580308305e-05 tokens_per_second_per_gpu:4818.366653520128 mfu:22.72905760277761 +Step 26467 | loss:0.3002597689628601 lr:1.71210276113534e-05 tokens_per_second_per_gpu:5331.2484030030555 mfu:25.14840832173844 +Step 26468 | loss:0.3790614604949951 lr:1.7117760889685462e-05 tokens_per_second_per_gpu:5340.682514543944 mfu:25.192910635502216 +Step 26469 | loss:0.3517603278160095 lr:1.7114494415329023e-05 tokens_per_second_per_gpu:3805.6993872941234 mfu:17.952133329137535 +Step 26470 | loss:0.5470489263534546 lr:1.7111228188308665e-05 tokens_per_second_per_gpu:4903.56249800256 mfu:23.1309409371112 +Step 26471 | loss:0.4816945493221283 lr:1.7107962208648957e-05 tokens_per_second_per_gpu:5291.093420265379 mfu:24.958990417018292 +Step 26472 | loss:0.38880449533462524 lr:1.71046964763745e-05 tokens_per_second_per_gpu:5339.609342490977 mfu:25.18784829982634 +Step 26473 | loss:0.5249618291854858 lr:1.7101430991509782e-05 tokens_per_second_per_gpu:3998.1568439905673 mfu:18.85998798900278 +Step 26474 | loss:0.34383103251457214 lr:1.70981657540794e-05 tokens_per_second_per_gpu:4747.705199806193 mfu:22.39573546952179 +Step 26475 | loss:0.35458269715309143 lr:1.7094900764107925e-05 tokens_per_second_per_gpu:5379.902166699044 mfu:25.377916426280493 +Step 26476 | loss:0.33275216817855835 lr:1.7091636021619885e-05 tokens_per_second_per_gpu:5366.050321325071 mfu:25.31257490828294 +Step 26477 | loss:0.3653380870819092 lr:1.7088371526639853e-05 tokens_per_second_per_gpu:4777.432324948816 mfu:22.535963390790094 +Step 26478 | loss:0.44385120272636414 lr:1.708510727919237e-05 tokens_per_second_per_gpu:5355.116913842499 mfu:25.261000159755977 +Step 26479 | loss:0.34844690561294556 lr:1.7081843279302034e-05 tokens_per_second_per_gpu:4680.031737334563 mfu:22.076508200759797 +Step 26480 | loss:0.3985726535320282 lr:1.707857952699332e-05 tokens_per_second_per_gpu:4436.263112230276 mfu:20.926610004926683 +Step 26481 | loss:0.37599265575408936 lr:1.7075316022290817e-05 tokens_per_second_per_gpu:4823.012570416738 mfu:22.750973185454075 +Step 26482 | loss:0.44456395506858826 lr:1.7072052765219055e-05 tokens_per_second_per_gpu:5081.963057880891 mfu:23.972486816331504 +Step 26483 | loss:0.3884725868701935 lr:1.7068789755802605e-05 tokens_per_second_per_gpu:5370.6427136369275 mfu:25.33423800636089 +Step 26484 | loss:0.41743984818458557 lr:1.7065526994065973e-05 tokens_per_second_per_gpu:5129.696776477584 mfu:24.197654911164296 +Step 26485 | loss:0.4823438823223114 lr:1.706226448003372e-05 tokens_per_second_per_gpu:5167.812345716865 mfu:24.377452554452688 +Step 26486 | loss:0.4837471842765808 lr:1.7059002213730386e-05 tokens_per_second_per_gpu:4591.597303313142 mfu:21.659347887010433 +Step 26487 | loss:0.3884791135787964 lr:1.7055740195180507e-05 tokens_per_second_per_gpu:3409.543328016405 mfu:16.08339760633104 +Step 26488 | loss:0.41300156712532043 lr:1.7052478424408592e-05 tokens_per_second_per_gpu:4081.7958993603543 mfu:19.254527683476418 +Step 26489 | loss:0.47823309898376465 lr:1.7049216901439184e-05 tokens_per_second_per_gpu:4783.373589450515 mfu:22.56398934912886 +Step 26490 | loss:0.3961554169654846 lr:1.7045955626296845e-05 tokens_per_second_per_gpu:5026.2153252053495 mfu:23.70951524975698 +Step 26491 | loss:0.393826961517334 lr:1.704269459900606e-05 tokens_per_second_per_gpu:4917.053551774312 mfu:23.194580539563685 +Step 26492 | loss:0.45625004172325134 lr:1.7039433819591372e-05 tokens_per_second_per_gpu:4970.416821042451 mfu:23.446304185414082 +Step 26493 | loss:0.3758055865764618 lr:1.703617328807733e-05 tokens_per_second_per_gpu:4707.1822487981135 mfu:22.20458179568829 +Step 26494 | loss:0.4968530237674713 lr:1.7032913004488427e-05 tokens_per_second_per_gpu:5050.1461915814125 mfu:23.822401229479702 +Step 26495 | loss:0.30295830965042114 lr:1.702965296884918e-05 tokens_per_second_per_gpu:4484.1001095300935 mfu:21.15226528302328 +Step 26496 | loss:0.39066410064697266 lr:1.702639318118412e-05 tokens_per_second_per_gpu:5337.131381960486 mfu:25.176159337220348 +Step 26497 | loss:0.34215524792671204 lr:1.702313364151778e-05 tokens_per_second_per_gpu:4326.052838806523 mfu:20.406729341375208 +Step 26498 | loss:0.40085721015930176 lr:1.701987434987464e-05 tokens_per_second_per_gpu:4964.1450534303885 mfu:23.416719187514087 +Step 26499 | loss:0.38547372817993164 lr:1.7016615306279236e-05 tokens_per_second_per_gpu:5333.413708536499 mfu:25.158622437378913 +Step 26500 | loss:0.5630062818527222 lr:1.7013356510756095e-05 tokens_per_second_per_gpu:5345.609931640521 mfu:25.216154102650815 +Step 26501 | loss:0.4591577351093292 lr:1.7010097963329714e-05 tokens_per_second_per_gpu:5325.728516155344 mfu:25.12237007368797 +Step 26502 | loss:0.37705424427986145 lr:1.7006839664024572e-05 tokens_per_second_per_gpu:4882.175883935765 mfu:23.03005662962551 +Step 26503 | loss:0.49950721859931946 lr:1.70035816128652e-05 tokens_per_second_per_gpu:4048.873105036194 mfu:19.09922524544141 +Step 26504 | loss:0.4191805422306061 lr:1.700032380987612e-05 tokens_per_second_per_gpu:4197.795012397833 mfu:19.801715291163717 +Step 26505 | loss:0.49810925126075745 lr:1.6997066255081794e-05 tokens_per_second_per_gpu:5338.859342026516 mfu:25.1843104196722 +Step 26506 | loss:0.4137713611125946 lr:1.6993808948506744e-05 tokens_per_second_per_gpu:5349.110687238303 mfu:25.232667764096053 +Step 26507 | loss:0.3509703576564789 lr:1.699055189017549e-05 tokens_per_second_per_gpu:4512.675326340118 mfu:21.287059456150892 +Step 26508 | loss:0.4633277952671051 lr:1.6987295080112504e-05 tokens_per_second_per_gpu:4476.068049132114 mfu:21.114376683715196 +Step 26509 | loss:0.43354782462120056 lr:1.6984038518342272e-05 tokens_per_second_per_gpu:4514.930998328918 mfu:21.297699845779388 +Step 26510 | loss:0.46499210596084595 lr:1.6980782204889293e-05 tokens_per_second_per_gpu:5321.922854030494 mfu:25.10441811613214 +Step 26511 | loss:0.40350788831710815 lr:1.6977526139778084e-05 tokens_per_second_per_gpu:5299.005845980678 mfu:24.99631467911639 +Step 26512 | loss:0.4312122166156769 lr:1.6974270323033096e-05 tokens_per_second_per_gpu:4590.088736055922 mfu:21.652231717870013 +Step 26513 | loss:0.27801066637039185 lr:1.6971014754678837e-05 tokens_per_second_per_gpu:4723.305440235636 mfu:22.280637640599206 +Step 26514 | loss:0.4201929271221161 lr:1.6967759434739787e-05 tokens_per_second_per_gpu:4848.310238370335 mfu:22.870306601418424 +Step 26515 | loss:0.47601354122161865 lr:1.6964504363240475e-05 tokens_per_second_per_gpu:4832.488017021854 mfu:22.79567048376831 +Step 26516 | loss:0.41685476899147034 lr:1.6961249540205298e-05 tokens_per_second_per_gpu:4553.586110165997 mfu:21.48004260355705 +Step 26517 | loss:0.450592964887619 lr:1.6957994965658786e-05 tokens_per_second_per_gpu:5111.280412416983 mfu:24.11078177973473 +Step 26518 | loss:0.44860514998435974 lr:1.6954740639625405e-05 tokens_per_second_per_gpu:4479.428305943095 mfu:21.130227588411525 +Step 26519 | loss:0.352090060710907 lr:1.6951486562129658e-05 tokens_per_second_per_gpu:3927.523817063147 mfu:18.52679994974906 +Step 26520 | loss:0.42849045991897583 lr:1.694823273319598e-05 tokens_per_second_per_gpu:4291.001927022152 mfu:20.241388210186106 +Step 26521 | loss:0.37111446261405945 lr:1.6944979152848854e-05 tokens_per_second_per_gpu:5398.927855716806 mfu:25.467663847494553 +Step 26522 | loss:0.4260321855545044 lr:1.694172582111278e-05 tokens_per_second_per_gpu:4445.608295010227 mfu:20.97069282655231 +Step 26523 | loss:0.4303685426712036 lr:1.6938472738012208e-05 tokens_per_second_per_gpu:5377.617617538521 mfu:25.367139818106025 +Step 26524 | loss:0.5925835967063904 lr:1.6935219903571582e-05 tokens_per_second_per_gpu:4310.007415092026 mfu:20.331040340082343 +Step 26525 | loss:0.3686940670013428 lr:1.6931967317815384e-05 tokens_per_second_per_gpu:4840.546297919914 mfu:22.833682769648952 +Step 26526 | loss:0.406948983669281 lr:1.6928714980768097e-05 tokens_per_second_per_gpu:4820.951419339844 mfu:22.74125038415573 +Step 26527 | loss:0.3216235041618347 lr:1.6925462892454146e-05 tokens_per_second_per_gpu:2987.8789147088883 mfu:14.094334625392008 +Step 26528 | loss:0.4578922986984253 lr:1.6922211052898013e-05 tokens_per_second_per_gpu:5375.3612772192655 mfu:25.356496275848745 +Step 26529 | loss:0.30410242080688477 lr:1.691895946212416e-05 tokens_per_second_per_gpu:5351.884314977258 mfu:25.245751439370395 +Step 26530 | loss:0.38729992508888245 lr:1.691570812015704e-05 tokens_per_second_per_gpu:5315.3490160848105 mfu:25.07340820092994 +Step 26531 | loss:0.4511362910270691 lr:1.691245702702109e-05 tokens_per_second_per_gpu:5361.040744123602 mfu:25.288943877901602 +Step 26532 | loss:0.35485121607780457 lr:1.6909206182740762e-05 tokens_per_second_per_gpu:5355.251954288152 mfu:25.261637168578396 +Step 26533 | loss:0.3583317697048187 lr:1.6905955587340537e-05 tokens_per_second_per_gpu:4792.470208712058 mfu:22.606899654228844 +Step 26534 | loss:0.4501234292984009 lr:1.690270524084483e-05 tokens_per_second_per_gpu:5174.227463426175 mfu:24.407713759219448 +Step 26535 | loss:0.4166046679019928 lr:1.6899455143278096e-05 tokens_per_second_per_gpu:3213.0247992546306 mfu:15.15638617664329 +Step 26536 | loss:0.4489360749721527 lr:1.68962052946648e-05 tokens_per_second_per_gpu:4671.680804469371 mfu:22.037115425618616 +Step 26537 | loss:0.39921802282333374 lr:1.689295569502937e-05 tokens_per_second_per_gpu:4761.483332060145 mfu:22.460729270155447 +Step 26538 | loss:0.5194857716560364 lr:1.688970634439623e-05 tokens_per_second_per_gpu:4878.284490978063 mfu:23.011700265103922 +Step 26539 | loss:0.4040433466434479 lr:1.6886457242789828e-05 tokens_per_second_per_gpu:5332.383258224282 mfu:25.15376162744323 +Step 26540 | loss:0.45550140738487244 lr:1.6883208390234628e-05 tokens_per_second_per_gpu:4456.365566119202 mfu:21.021436709753054 +Step 26541 | loss:0.3530404269695282 lr:1.6879959786755022e-05 tokens_per_second_per_gpu:4351.115934856544 mfu:20.52495624164829 +Step 26542 | loss:0.4598158895969391 lr:1.687671143237547e-05 tokens_per_second_per_gpu:5368.230017594008 mfu:25.322856907478048 +Step 26543 | loss:0.36219412088394165 lr:1.687346332712041e-05 tokens_per_second_per_gpu:4871.694952133438 mfu:22.98061628608257 +Step 26544 | loss:0.4110410511493683 lr:1.687021547101426e-05 tokens_per_second_per_gpu:5299.392431176922 mfu:24.998138267445313 +Step 26545 | loss:0.48138192296028137 lr:1.6866967864081433e-05 tokens_per_second_per_gpu:4852.647509728866 mfu:22.890766250431412 +Step 26546 | loss:0.3891030550003052 lr:1.686372050634637e-05 tokens_per_second_per_gpu:4703.648515428998 mfu:22.187912572469088 +Step 26547 | loss:0.442142516374588 lr:1.6860473397833505e-05 tokens_per_second_per_gpu:5223.705731113291 mfu:24.64111118976254 +Step 26548 | loss:0.41520756483078003 lr:1.6857226538567234e-05 tokens_per_second_per_gpu:4666.1754601489165 mfu:22.011145777150436 +Step 26549 | loss:0.45396000146865845 lr:1.685397992857199e-05 tokens_per_second_per_gpu:3772.582177927458 mfu:17.795913802176408 +Step 26550 | loss:0.49593138694763184 lr:1.6850733567872197e-05 tokens_per_second_per_gpu:4652.192343561526 mfu:21.94518503044201 +Step 26551 | loss:0.42147091031074524 lr:1.6847487456492295e-05 tokens_per_second_per_gpu:4737.155450734795 mfu:22.345970503179203 +Step 26552 | loss:0.3647724688053131 lr:1.6844241594456632e-05 tokens_per_second_per_gpu:5347.470800554794 mfu:25.224932138816417 +Step 26553 | loss:0.4107213616371155 lr:1.684099598178967e-05 tokens_per_second_per_gpu:5340.069482225214 mfu:25.19001885746067 +Step 26554 | loss:0.38027751445770264 lr:1.68377506185158e-05 tokens_per_second_per_gpu:5272.622578662538 mfu:24.87186030572715 +Step 26555 | loss:0.37306705117225647 lr:1.683450550465946e-05 tokens_per_second_per_gpu:4381.397149858503 mfu:20.66779789932056 +Step 26556 | loss:0.36776891350746155 lr:1.683126064024502e-05 tokens_per_second_per_gpu:4401.124257684186 mfu:20.760854032725973 +Step 26557 | loss:0.3516296148300171 lr:1.6828016025296896e-05 tokens_per_second_per_gpu:5370.236989134013 mfu:25.332324134656908 +Step 26558 | loss:0.5422744750976562 lr:1.6824771659839516e-05 tokens_per_second_per_gpu:4321.710270957508 mfu:20.386244707912926 +Step 26559 | loss:0.42358437180519104 lr:1.6821527543897266e-05 tokens_per_second_per_gpu:4379.93250507052 mfu:20.660888919961476 +Step 26560 | loss:0.47871965169906616 lr:1.6818283677494517e-05 tokens_per_second_per_gpu:4605.438482632237 mfu:21.72463909140731 +Step 26561 | loss:0.3145166337490082 lr:1.681504006065569e-05 tokens_per_second_per_gpu:4466.247694439349 mfu:21.068052395104136 +Step 26562 | loss:0.5892332196235657 lr:1.6811796693405202e-05 tokens_per_second_per_gpu:4825.979790684546 mfu:22.76497006971731 +Step 26563 | loss:0.3790416717529297 lr:1.6808553575767412e-05 tokens_per_second_per_gpu:5305.747675211898 mfu:25.02811703787474 +Step 26564 | loss:0.43192800879478455 lr:1.6805310707766718e-05 tokens_per_second_per_gpu:4865.051426855406 mfu:22.949277643843793 +Step 26565 | loss:0.43881094455718994 lr:1.6802068089427542e-05 tokens_per_second_per_gpu:5265.701606605713 mfu:24.839212899695507 +Step 26566 | loss:0.5183326005935669 lr:1.679882572077424e-05 tokens_per_second_per_gpu:4921.227545398275 mfu:23.214269979644666 +Step 26567 | loss:0.4942954182624817 lr:1.67955836018312e-05 tokens_per_second_per_gpu:4317.085956374883 mfu:20.364431027037252 +Step 26568 | loss:0.4900086522102356 lr:1.679234173262281e-05 tokens_per_second_per_gpu:5322.022086805652 mfu:25.104886214063416 +Step 26569 | loss:0.4168233573436737 lr:1.6789100113173463e-05 tokens_per_second_per_gpu:4519.756619245062 mfu:21.320463122976673 +Step 26570 | loss:0.466318815946579 lr:1.678585874350752e-05 tokens_per_second_per_gpu:5124.935478792037 mfu:24.17519505762042 +Step 26571 | loss:0.49087053537368774 lr:1.678261762364937e-05 tokens_per_second_per_gpu:5070.498630367829 mfu:23.918407155719354 +Step 26572 | loss:0.4290350079536438 lr:1.6779376753623416e-05 tokens_per_second_per_gpu:3975.356252263776 mfu:18.752433707645412 +Step 26573 | loss:0.5160393714904785 lr:1.6776136133454e-05 tokens_per_second_per_gpu:4709.581117543189 mfu:22.21589766884803 +Step 26574 | loss:0.368980348110199 lr:1.677289576316549e-05 tokens_per_second_per_gpu:5356.685230155377 mfu:25.268398175386174 +Step 26575 | loss:0.46456676721572876 lr:1.6769655642782272e-05 tokens_per_second_per_gpu:4411.798669010435 mfu:20.811207052194266 +Step 26576 | loss:0.35601669549942017 lr:1.676641577232873e-05 tokens_per_second_per_gpu:4217.237321797494 mfu:19.893427981801963 +Step 26577 | loss:0.37310677766799927 lr:1.6763176151829196e-05 tokens_per_second_per_gpu:5374.5855965632445 mfu:25.352837257849526 +Step 26578 | loss:0.341628760099411 lr:1.675993678130806e-05 tokens_per_second_per_gpu:5328.1721995627895 mfu:25.133897345256276 +Step 26579 | loss:0.4991194009780884 lr:1.6756697660789694e-05 tokens_per_second_per_gpu:4932.12377353321 mfu:23.265669346845375 +Step 26580 | loss:0.4542014002799988 lr:1.675345879029845e-05 tokens_per_second_per_gpu:4706.47587195424 mfu:22.201249695595422 +Step 26581 | loss:0.4234810173511505 lr:1.6750220169858667e-05 tokens_per_second_per_gpu:5346.107637474295 mfu:25.218501865986102 +Step 26582 | loss:0.41277849674224854 lr:1.6746981799494716e-05 tokens_per_second_per_gpu:4251.308057310903 mfu:20.054145454285617 +Step 26583 | loss:0.42097049951553345 lr:1.6743743679230978e-05 tokens_per_second_per_gpu:5376.524403911033 mfu:25.361982942903687 +Step 26584 | loss:0.428148478269577 lr:1.6740505809091774e-05 tokens_per_second_per_gpu:4761.101631618881 mfu:22.458928724049194 +Step 26585 | loss:0.4589773118495941 lr:1.6737268189101463e-05 tokens_per_second_per_gpu:4010.340125876478 mfu:18.91745860834086 +Step 26586 | loss:0.5826311111450195 lr:1.6734030819284407e-05 tokens_per_second_per_gpu:5389.241424264751 mfu:25.421971297659585 +Step 26587 | loss:0.3129228949546814 lr:1.6730793699664983e-05 tokens_per_second_per_gpu:4380.047401569935 mfu:20.661430906352578 +Step 26588 | loss:0.3793436586856842 lr:1.672755683026747e-05 tokens_per_second_per_gpu:4474.590790178791 mfu:21.107408201186008 +Step 26589 | loss:0.35125070810317993 lr:1.672432021111625e-05 tokens_per_second_per_gpu:3759.121257506092 mfu:17.732416343878974 +Step 26590 | loss:0.4495832920074463 lr:1.6721083842235663e-05 tokens_per_second_per_gpu:3866.886964621922 mfu:18.240765571069957 +Step 26591 | loss:0.5460535287857056 lr:1.6717847723650065e-05 tokens_per_second_per_gpu:3780.3947063652267 mfu:17.832766831772165 +Step 26592 | loss:0.3226657807826996 lr:1.6714611855383765e-05 tokens_per_second_per_gpu:4774.453466381668 mfu:22.521911606683176 +Step 26593 | loss:0.4196203947067261 lr:1.6711376237461118e-05 tokens_per_second_per_gpu:4799.180231261112 mfu:22.63855197544063 +Step 26594 | loss:0.35661235451698303 lr:1.6708140869906475e-05 tokens_per_second_per_gpu:4112.935383547589 mfu:19.401417943331452 +Step 26595 | loss:0.40439435839653015 lr:1.670490575274415e-05 tokens_per_second_per_gpu:4676.849997110664 mfu:22.061499389262803 +Step 26596 | loss:0.4016503691673279 lr:1.6701670885998466e-05 tokens_per_second_per_gpu:3766.2657668782563 mfu:17.766118213566468 +Step 26597 | loss:0.3913191258907318 lr:1.6698436269693757e-05 tokens_per_second_per_gpu:3500.3261228901124 mfu:16.511635538892314 +Step 26598 | loss:0.48526135087013245 lr:1.669520190385438e-05 tokens_per_second_per_gpu:5353.697817285486 mfu:25.254306039173702 +Step 26599 | loss:0.36401045322418213 lr:1.669196778850462e-05 tokens_per_second_per_gpu:5332.875838260021 mfu:25.15608521151463 +Step 26600 | loss:0.43948838114738464 lr:1.668873392366882e-05 tokens_per_second_per_gpu:4708.084328930645 mfu:22.208837061584187 +Step 26601 | loss:0.3840494155883789 lr:1.668550030937132e-05 tokens_per_second_per_gpu:4739.743357735494 mfu:22.35817810204397 +Step 26602 | loss:0.4681451916694641 lr:1.6682266945636427e-05 tokens_per_second_per_gpu:5167.546156532331 mfu:24.376196894652683 +Step 26603 | loss:0.48187458515167236 lr:1.6679033832488434e-05 tokens_per_second_per_gpu:5136.294469843632 mfu:24.22877735645451 +Step 26604 | loss:0.3117862939834595 lr:1.6675800969951682e-05 tokens_per_second_per_gpu:4701.123175471218 mfu:22.176000113020816 +Step 26605 | loss:0.36594074964523315 lr:1.66725683580505e-05 tokens_per_second_per_gpu:4394.878692742988 mfu:20.73139263729508 +Step 26606 | loss:0.5212678909301758 lr:1.6669335996809164e-05 tokens_per_second_per_gpu:5077.645670281031 mfu:23.95212096240068 +Step 26607 | loss:0.44558072090148926 lr:1.666610388625201e-05 tokens_per_second_per_gpu:4569.212875764331 mfu:21.553756723085534 +Step 26608 | loss:0.30304986238479614 lr:1.6662872026403355e-05 tokens_per_second_per_gpu:4532.863925186208 mfu:21.38229252143805 +Step 26609 | loss:0.3232984244823456 lr:1.6659640417287493e-05 tokens_per_second_per_gpu:4549.239850963485 mfu:21.459540557350515 +Step 26610 | loss:0.5133464336395264 lr:1.665640905892872e-05 tokens_per_second_per_gpu:4544.89772709461 mfu:21.439058018218304 +Step 26611 | loss:0.4700712263584137 lr:1.665317795135134e-05 tokens_per_second_per_gpu:4661.532320254024 mfu:21.98924329406487 +Step 26612 | loss:0.46997207403182983 lr:1.6649947094579682e-05 tokens_per_second_per_gpu:5332.128308241088 mfu:25.152558984873554 +Step 26613 | loss:0.3771517276763916 lr:1.6646716488638015e-05 tokens_per_second_per_gpu:3818.384674860156 mfu:18.011971994919747 +Step 26614 | loss:0.3529719114303589 lr:1.6643486133550644e-05 tokens_per_second_per_gpu:3646.5441561567986 mfu:17.201370949180237 +Step 26615 | loss:0.438946932554245 lr:1.6640256029341896e-05 tokens_per_second_per_gpu:4210.2450371485465 mfu:19.8604441821061 +Step 26616 | loss:0.4787427484989166 lr:1.663702617603603e-05 tokens_per_second_per_gpu:4179.03302104977 mfu:19.713211776849093 +Step 26617 | loss:0.32807475328445435 lr:1.6633796573657335e-05 tokens_per_second_per_gpu:5368.024955652366 mfu:25.321889595312147 +Step 26618 | loss:0.3343020975589752 lr:1.6630567222230108e-05 tokens_per_second_per_gpu:4842.455081574633 mfu:22.842686827820337 +Step 26619 | loss:0.35546356439590454 lr:1.6627338121778662e-05 tokens_per_second_per_gpu:4262.821634301239 mfu:20.108456961367832 +Step 26620 | loss:0.319928377866745 lr:1.662410927232725e-05 tokens_per_second_per_gpu:3399.1748230372273 mfu:16.03448759929489 +Step 26621 | loss:0.49013376235961914 lr:1.662088067390017e-05 tokens_per_second_per_gpu:5105.282437357465 mfu:24.08248830801891 +Step 26622 | loss:0.3906278610229492 lr:1.6617652326521705e-05 tokens_per_second_per_gpu:3621.796579594104 mfu:17.08463240816205 +Step 26623 | loss:0.39765480160713196 lr:1.661442423021616e-05 tokens_per_second_per_gpu:4791.904589305383 mfu:22.604231530982936 +Step 26624 | loss:0.29906395077705383 lr:1.661119638500778e-05 tokens_per_second_per_gpu:3698.2187287518054 mfu:17.44512872470211 +Step 26625 | loss:0.374324768781662 lr:1.6607968790920843e-05 tokens_per_second_per_gpu:5275.183610058914 mfu:24.88394112778087 +Step 26626 | loss:0.3977559804916382 lr:1.6604741447979628e-05 tokens_per_second_per_gpu:5341.649196968084 mfu:25.197470641429913 +Step 26627 | loss:0.4052601158618927 lr:1.6601514356208432e-05 tokens_per_second_per_gpu:3926.7086974147633 mfu:18.52295489129388 +Step 26628 | loss:0.4477916955947876 lr:1.6598287515631494e-05 tokens_per_second_per_gpu:3379.145230316763 mfu:15.940004593031551 +Step 26629 | loss:0.3608325123786926 lr:1.6595060926273097e-05 tokens_per_second_per_gpu:5342.801898378269 mfu:25.202908130652894 +Step 26630 | loss:0.41504597663879395 lr:1.6591834588157523e-05 tokens_per_second_per_gpu:3915.140765311427 mfu:18.468386981869706 +Step 26631 | loss:0.32885462045669556 lr:1.6588608501309027e-05 tokens_per_second_per_gpu:5340.401304934387 mfu:25.19158412179463 +Step 26632 | loss:0.3685513138771057 lr:1.6585382665751853e-05 tokens_per_second_per_gpu:5327.87631870209 mfu:25.13250162475388 +Step 26633 | loss:0.4144742786884308 lr:1.658215708151028e-05 tokens_per_second_per_gpu:3838.599643767094 mfu:18.107329452283125 +Step 26634 | loss:0.4467228055000305 lr:1.6578931748608582e-05 tokens_per_second_per_gpu:5329.477892830449 mfu:25.140056523174067 +Step 26635 | loss:0.35228991508483887 lr:1.657570666707099e-05 tokens_per_second_per_gpu:5324.247292436046 mfu:25.115382888681218 +Step 26636 | loss:0.3450673222541809 lr:1.6572481836921765e-05 tokens_per_second_per_gpu:5367.562239292651 mfu:25.31970688329582 +Step 26637 | loss:0.44310927391052246 lr:1.6569257258185193e-05 tokens_per_second_per_gpu:3341.0531633546057 mfu:15.76031781399426 +Step 26638 | loss:0.27854883670806885 lr:1.6566032930885507e-05 tokens_per_second_per_gpu:5388.179583503299 mfu:25.4169624136191 +Step 26639 | loss:0.4168075919151306 lr:1.656280885504693e-05 tokens_per_second_per_gpu:3645.423130952503 mfu:17.196082882024953 +Step 26640 | loss:0.32783177495002747 lr:1.6559585030693735e-05 tokens_per_second_per_gpu:5389.09764139979 mfu:25.421293049353675 +Step 26641 | loss:0.3798760175704956 lr:1.6556361457850185e-05 tokens_per_second_per_gpu:5336.768419317619 mfu:25.174447180506135 +Step 26642 | loss:0.40865659713745117 lr:1.655313813654049e-05 tokens_per_second_per_gpu:4315.341169552633 mfu:20.356200569905376 +Step 26643 | loss:0.48942315578460693 lr:1.6549915066788917e-05 tokens_per_second_per_gpu:4412.4378758060775 mfu:20.814222299709233 +Step 26644 | loss:0.35402339696884155 lr:1.654669224861971e-05 tokens_per_second_per_gpu:4353.10465940908 mfu:20.534337394673997 +Step 26645 | loss:0.4307451844215393 lr:1.6543469682057106e-05 tokens_per_second_per_gpu:4576.216038630682 mfu:21.586791837188905 +Step 26646 | loss:0.3847849369049072 lr:1.6540247367125316e-05 tokens_per_second_per_gpu:4560.6482020783815 mfu:21.513355696024895 +Step 26647 | loss:0.41713884472846985 lr:1.653702530384859e-05 tokens_per_second_per_gpu:4371.9864716902875 mfu:20.62340612477383 +Step 26648 | loss:0.4365982115268707 lr:1.6533803492251194e-05 tokens_per_second_per_gpu:5327.388697977672 mfu:25.130201434600117 +Step 26649 | loss:0.4768564999103546 lr:1.653058193235731e-05 tokens_per_second_per_gpu:4770.3907277260805 mfu:22.502746975270078 +Step 26650 | loss:0.4329372048377991 lr:1.6527360624191184e-05 tokens_per_second_per_gpu:4236.186519680735 mfu:19.982814581283755 +Step 26651 | loss:0.4109271764755249 lr:1.6524139567777074e-05 tokens_per_second_per_gpu:3797.0722357904383 mfu:17.911437583550825 +Step 26652 | loss:0.44981449842453003 lr:1.652091876313918e-05 tokens_per_second_per_gpu:5353.277220177071 mfu:25.252322010852044 +Step 26653 | loss:0.49602729082107544 lr:1.6517698210301713e-05 tokens_per_second_per_gpu:4415.884854290408 mfu:20.830482285334995 +Step 26654 | loss:0.49517184495925903 lr:1.6514477909288906e-05 tokens_per_second_per_gpu:4646.670277852667 mfu:21.919136504331757 +Step 26655 | loss:0.47673532366752625 lr:1.6511257860125e-05 tokens_per_second_per_gpu:3787.8911993146585 mfu:17.86812906804822 +Step 26656 | loss:0.5224282741546631 lr:1.6508038062834187e-05 tokens_per_second_per_gpu:4876.0306087307235 mfu:23.001068317991255 +Step 26657 | loss:0.4653354585170746 lr:1.650481851744069e-05 tokens_per_second_per_gpu:3436.6883483330994 mfu:16.21144529858293 +Step 26658 | loss:0.45387887954711914 lr:1.6501599223968724e-05 tokens_per_second_per_gpu:4681.78503390368 mfu:22.084778799819563 +Step 26659 | loss:0.3309086561203003 lr:1.6498380182442525e-05 tokens_per_second_per_gpu:5054.759346741176 mfu:23.844162269453737 +Step 26660 | loss:0.34207257628440857 lr:1.6495161392886283e-05 tokens_per_second_per_gpu:4889.987481988398 mfu:23.06690527043584 +Step 26661 | loss:0.287039190530777 lr:1.649194285532419e-05 tokens_per_second_per_gpu:5157.198492952214 mfu:24.32738520005181 +Step 26662 | loss:0.4842102527618408 lr:1.6488724569780482e-05 tokens_per_second_per_gpu:3519.193856916014 mfu:16.600637859459866 +Step 26663 | loss:0.42441263794898987 lr:1.6485506536279344e-05 tokens_per_second_per_gpu:4982.8920260034165 mfu:23.505151855704405 +Step 26664 | loss:0.5157425999641418 lr:1.648228875484498e-05 tokens_per_second_per_gpu:4829.067470175643 mfu:22.77953518068926 +Step 26665 | loss:0.38061121106147766 lr:1.6479071225501607e-05 tokens_per_second_per_gpu:3530.318172053788 mfu:16.653113151968626 +Step 26666 | loss:0.39299070835113525 lr:1.647585394827343e-05 tokens_per_second_per_gpu:5356.967821315058 mfu:25.269731206101415 +Step 26667 | loss:0.41615715622901917 lr:1.6472636923184632e-05 tokens_per_second_per_gpu:2991.9848293682066 mfu:14.113702925381292 +Step 26668 | loss:0.340800017118454 lr:1.6469420150259396e-05 tokens_per_second_per_gpu:4565.146953547899 mfu:21.53457707427276 +Step 26669 | loss:0.33028101921081543 lr:1.6466203629521927e-05 tokens_per_second_per_gpu:3978.025082988472 mfu:18.76502303752328 +Step 26670 | loss:0.40808799862861633 lr:1.6462987360996436e-05 tokens_per_second_per_gpu:5320.5196514661575 mfu:25.09779897022512 +Step 26671 | loss:0.5019363760948181 lr:1.645977134470708e-05 tokens_per_second_per_gpu:5359.824638171196 mfu:25.283207298635034 +Step 26672 | loss:0.38101622462272644 lr:1.645655558067806e-05 tokens_per_second_per_gpu:5333.121251366828 mfu:25.15724286701123 +Step 26673 | loss:0.3566170334815979 lr:1.645334006893359e-05 tokens_per_second_per_gpu:5365.849150895546 mfu:25.311625953043652 +Step 26674 | loss:0.5207066535949707 lr:1.6450124809497823e-05 tokens_per_second_per_gpu:4890.414720456792 mfu:23.06892062718565 +Step 26675 | loss:0.3898693323135376 lr:1.644690980239494e-05 tokens_per_second_per_gpu:5353.575906494955 mfu:25.253730965174643 +Step 26676 | loss:0.31558454036712646 lr:1.6443695047649123e-05 tokens_per_second_per_gpu:4816.947474910414 mfu:22.722363095138746 +Step 26677 | loss:0.45455220341682434 lr:1.6440480545284576e-05 tokens_per_second_per_gpu:5361.1705302156815 mfu:25.289556101041732 +Step 26678 | loss:0.4060460925102234 lr:1.6437266295325436e-05 tokens_per_second_per_gpu:4219.132794806385 mfu:19.90236925138595 +Step 26679 | loss:0.4011359214782715 lr:1.6434052297795904e-05 tokens_per_second_per_gpu:3843.1504739969664 mfu:18.128796494929134 +Step 26680 | loss:0.39671191573143005 lr:1.6430838552720168e-05 tokens_per_second_per_gpu:5337.071930607678 mfu:25.175878895045706 +Step 26681 | loss:0.3734249472618103 lr:1.642762506012237e-05 tokens_per_second_per_gpu:3338.374036964999 mfu:15.747679917708385 +Step 26682 | loss:0.42405399680137634 lr:1.6424411820026676e-05 tokens_per_second_per_gpu:4109.692096791056 mfu:19.38611880633911 +Step 26683 | loss:0.46444952487945557 lr:1.6421198832457264e-05 tokens_per_second_per_gpu:5332.566221972297 mfu:25.15462469865945 +Step 26684 | loss:0.4082178771495819 lr:1.641798609743832e-05 tokens_per_second_per_gpu:4945.121308327263 mfu:23.326980936077035 +Step 26685 | loss:0.4029013514518738 lr:1.6414773614993967e-05 tokens_per_second_per_gpu:4553.31742726467 mfu:21.47877518046946 +Step 26686 | loss:0.42629092931747437 lr:1.641156138514839e-05 tokens_per_second_per_gpu:5264.8572357476705 mfu:24.835229858293598 +Step 26687 | loss:0.37105074524879456 lr:1.6408349407925756e-05 tokens_per_second_per_gpu:5329.178318366369 mfu:25.13864337931433 +Step 26688 | loss:0.41335320472717285 lr:1.64051376833502e-05 tokens_per_second_per_gpu:4311.489619112539 mfu:20.338032149337185 +Step 26689 | loss:0.34686794877052307 lr:1.6401926211445906e-05 tokens_per_second_per_gpu:5348.779986581606 mfu:25.23110779267506 +Step 26690 | loss:0.33376777172088623 lr:1.6398714992237e-05 tokens_per_second_per_gpu:5371.153420984493 mfu:25.336647100055018 +Step 26691 | loss:0.46522390842437744 lr:1.639550402574766e-05 tokens_per_second_per_gpu:4378.646953831181 mfu:20.654824755427857 +Step 26692 | loss:0.32013651728630066 lr:1.6392293312002004e-05 tokens_per_second_per_gpu:5333.914594233013 mfu:25.160985200669277 +Step 26693 | loss:0.3789586126804352 lr:1.63890828510242e-05 tokens_per_second_per_gpu:5272.769157661707 mfu:24.87255174387519 +Step 26694 | loss:0.4661295711994171 lr:1.638587264283839e-05 tokens_per_second_per_gpu:4482.122379836387 mfu:21.14293599462259 +Step 26695 | loss:0.3158363997936249 lr:1.6382662687468743e-05 tokens_per_second_per_gpu:4677.623627916736 mfu:22.06514873777007 +Step 26696 | loss:0.4167174994945526 lr:1.6379452984939374e-05 tokens_per_second_per_gpu:5330.720645774657 mfu:25.145918800847415 +Step 26697 | loss:0.3541731834411621 lr:1.6376243535274416e-05 tokens_per_second_per_gpu:4064.6780311192624 mfu:19.1737797783737 +Step 26698 | loss:0.5267999768257141 lr:1.6373034338498032e-05 tokens_per_second_per_gpu:5048.585894036107 mfu:23.81504103974428 +Step 26699 | loss:0.4267689287662506 lr:1.6369825394634336e-05 tokens_per_second_per_gpu:5327.441744669177 mfu:25.13045166489465 +Step 26700 | loss:0.51735919713974 lr:1.6366616703707475e-05 tokens_per_second_per_gpu:5021.428265299011 mfu:23.686933871422138 +Step 26701 | loss:0.43678390979766846 lr:1.636340826574158e-05 tokens_per_second_per_gpu:3779.7308915397743 mfu:17.829635503982015 +Step 26702 | loss:0.3822031021118164 lr:1.63602000807608e-05 tokens_per_second_per_gpu:5190.929081090913 mfu:24.486498139333833 +Step 26703 | loss:0.41284799575805664 lr:1.6356992148789246e-05 tokens_per_second_per_gpu:4612.154336522866 mfu:21.756318920052244 +Step 26704 | loss:0.431130051612854 lr:1.6353784469851035e-05 tokens_per_second_per_gpu:5053.471158683056 mfu:23.838085666596037 +Step 26705 | loss:0.39442360401153564 lr:1.63505770439703e-05 tokens_per_second_per_gpu:5330.775622931051 mfu:25.14617813747419 +Step 26706 | loss:0.34473687410354614 lr:1.6347369871171188e-05 tokens_per_second_per_gpu:3579.8954180026703 mfu:16.8869774798599 +Step 26707 | loss:0.4285127520561218 lr:1.6344162951477783e-05 tokens_per_second_per_gpu:5327.905172552786 mfu:25.132637733290583 +Step 26708 | loss:0.4761463701725006 lr:1.634095628491422e-05 tokens_per_second_per_gpu:5328.336221486771 mfu:25.134671064656725 +Step 26709 | loss:0.3515416979789734 lr:1.6337749871504642e-05 tokens_per_second_per_gpu:4347.2038909954135 mfu:20.506502463291827 +Step 26710 | loss:0.29360219836235046 lr:1.6334543711273138e-05 tokens_per_second_per_gpu:5337.911023572171 mfu:25.179837039723242 +Step 26711 | loss:0.3175048828125 lr:1.6331337804243813e-05 tokens_per_second_per_gpu:4839.801181634568 mfu:22.830167929001025 +Step 26712 | loss:0.2780952751636505 lr:1.6328132150440788e-05 tokens_per_second_per_gpu:5356.611327419644 mfu:25.268049563572614 +Step 26713 | loss:0.4326505959033966 lr:1.6324926749888204e-05 tokens_per_second_per_gpu:5333.608409969056 mfu:25.159540877255772 +Step 26714 | loss:0.3311448395252228 lr:1.632172160261012e-05 tokens_per_second_per_gpu:5366.122278306826 mfu:25.312914341643005 +Step 26715 | loss:0.34399157762527466 lr:1.631851670863066e-05 tokens_per_second_per_gpu:5315.38832948869 mfu:25.073593648963598 +Step 26716 | loss:0.5228249430656433 lr:1.631531206797396e-05 tokens_per_second_per_gpu:5356.657608137216 mfu:25.268267877613706 +Step 26717 | loss:0.34571248292922974 lr:1.6312107680664074e-05 tokens_per_second_per_gpu:4059.535481159399 mfu:19.149521492803522 +Step 26718 | loss:0.4651198387145996 lr:1.630890354672514e-05 tokens_per_second_per_gpu:4250.096524829059 mfu:20.04843044886 +Step 26719 | loss:0.3204782009124756 lr:1.6305699666181228e-05 tokens_per_second_per_gpu:4331.416579028389 mfu:20.43203101915019 +Step 26720 | loss:0.4211310148239136 lr:1.6302496039056463e-05 tokens_per_second_per_gpu:4099.966498439792 mfu:19.340241499558015 +Step 26721 | loss:0.4004410207271576 lr:1.629929266537491e-05 tokens_per_second_per_gpu:4303.302232996023 mfu:20.29941085211346 +Step 26722 | loss:0.5457373857498169 lr:1.6296089545160676e-05 tokens_per_second_per_gpu:5154.5696600658375 mfu:24.314984546801632 +Step 26723 | loss:0.5198155045509338 lr:1.629288667843787e-05 tokens_per_second_per_gpu:5360.065940062345 mfu:25.284345560825933 +Step 26724 | loss:0.4274984300136566 lr:1.628968406523055e-05 tokens_per_second_per_gpu:4924.944952776929 mfu:23.231805624504883 +Step 26725 | loss:0.41187718510627747 lr:1.6286481705562835e-05 tokens_per_second_per_gpu:3342.6696609240817 mfu:15.767943108832402 +Step 26726 | loss:0.34577369689941406 lr:1.6283279599458772e-05 tokens_per_second_per_gpu:5331.823631481152 mfu:25.15112177261372 +Step 26727 | loss:0.30227118730545044 lr:1.6280077746942483e-05 tokens_per_second_per_gpu:4563.524763568398 mfu:21.526924927365304 +Step 26728 | loss:0.3192104399204254 lr:1.6276876148038016e-05 tokens_per_second_per_gpu:4762.950126195476 mfu:22.46764838835269 +Step 26729 | loss:0.39912042021751404 lr:1.6273674802769467e-05 tokens_per_second_per_gpu:4817.242832253873 mfu:22.723756346121 +Step 26730 | loss:0.4234836995601654 lr:1.6270473711160926e-05 tokens_per_second_per_gpu:5366.064851039168 mfu:25.312643447423362 +Step 26731 | loss:0.39376288652420044 lr:1.6267272873236443e-05 tokens_per_second_per_gpu:5061.340163109825 mfu:23.875205103067703 +Step 26732 | loss:0.5527467727661133 lr:1.6264072289020126e-05 tokens_per_second_per_gpu:5336.828847793416 mfu:25.174732231936627 +Step 26733 | loss:0.4356430470943451 lr:1.6260871958536005e-05 tokens_per_second_per_gpu:5347.819256518073 mfu:25.226575865047927 +Step 26734 | loss:0.42231959104537964 lr:1.6257671881808186e-05 tokens_per_second_per_gpu:3694.300986740657 mfu:17.426648067199807 +Step 26735 | loss:0.3397245407104492 lr:1.6254472058860707e-05 tokens_per_second_per_gpu:5358.926736791572 mfu:25.278971744629573 +Step 26736 | loss:0.3105839192867279 lr:1.625127248971765e-05 tokens_per_second_per_gpu:5350.269854488937 mfu:25.23813575379162 +Step 26737 | loss:0.3750961124897003 lr:1.6248073174403085e-05 tokens_per_second_per_gpu:4384.725991996453 mfu:20.683500615644302 +Step 26738 | loss:0.39493924379348755 lr:1.6244874112941076e-05 tokens_per_second_per_gpu:5190.800717729085 mfu:24.485892627455037 +Step 26739 | loss:0.410539835691452 lr:1.624167530535568e-05 tokens_per_second_per_gpu:4229.793838776515 mfu:19.952659214754718 +Step 26740 | loss:0.35940754413604736 lr:1.6238476751670934e-05 tokens_per_second_per_gpu:5362.29716723569 mfu:25.294870640089208 +Step 26741 | loss:0.37887048721313477 lr:1.6235278451910912e-05 tokens_per_second_per_gpu:4647.74730098162 mfu:21.924217006190126 +Step 26742 | loss:0.436023086309433 lr:1.6232080406099686e-05 tokens_per_second_per_gpu:4515.931759823952 mfu:21.302420608498984 +Step 26743 | loss:0.4842909872531891 lr:1.6228882614261277e-05 tokens_per_second_per_gpu:4054.728014302491 mfu:19.126843851401063 +Step 26744 | loss:0.38571614027023315 lr:1.622568507641975e-05 tokens_per_second_per_gpu:5086.000014616002 mfu:23.991529830026863 +Step 26745 | loss:0.27084895968437195 lr:1.622248779259917e-05 tokens_per_second_per_gpu:5358.290798733741 mfu:25.275971916308567 +Step 26746 | loss:0.33604881167411804 lr:1.6219290762823552e-05 tokens_per_second_per_gpu:5331.644987774191 mfu:25.15027907976822 +Step 26747 | loss:0.38319873809814453 lr:1.621609398711698e-05 tokens_per_second_per_gpu:4422.091273235309 mfu:20.85975902242236 +Step 26748 | loss:0.39371925592422485 lr:1.6212897465503457e-05 tokens_per_second_per_gpu:5229.315393351496 mfu:24.667572923647224 +Step 26749 | loss:0.39711713790893555 lr:1.620970119800706e-05 tokens_per_second_per_gpu:5381.902107025402 mfu:25.387350486024932 +Step 26750 | loss:0.5384762287139893 lr:1.6206505184651792e-05 tokens_per_second_per_gpu:3398.526102750464 mfu:16.031427475019136 +Step 26751 | loss:0.4449268877506256 lr:1.6203309425461714e-05 tokens_per_second_per_gpu:5231.607657262769 mfu:24.67838592361725 +Step 26752 | loss:0.39590248465538025 lr:1.620011392046088e-05 tokens_per_second_per_gpu:5373.188563389029 mfu:25.34624721400831 +Step 26753 | loss:0.4365704357624054 lr:1.6196918669673277e-05 tokens_per_second_per_gpu:3197.9134313655813 mfu:15.08510327604542 +Step 26754 | loss:0.3691633343696594 lr:1.6193723673122985e-05 tokens_per_second_per_gpu:3451.681264134472 mfu:16.28216944047322 +Step 26755 | loss:0.4182911813259125 lr:1.6190528930833993e-05 tokens_per_second_per_gpu:5237.030169899945 mfu:24.703964840902863 +Step 26756 | loss:0.3767542839050293 lr:1.618733444283036e-05 tokens_per_second_per_gpu:4921.848175678284 mfu:23.217197598565914 +Step 26757 | loss:0.35759079456329346 lr:1.6184140209136083e-05 tokens_per_second_per_gpu:4124.470615375302 mfu:19.455831599976275 +Step 26758 | loss:0.2751072943210602 lr:1.6180946229775206e-05 tokens_per_second_per_gpu:3430.4978323308032 mfu:16.182243578389155 +Step 26759 | loss:0.46088579297065735 lr:1.617775250477176e-05 tokens_per_second_per_gpu:4172.947722517155 mfu:19.68450638541176 +Step 26760 | loss:0.38596072793006897 lr:1.617455903414974e-05 tokens_per_second_per_gpu:4958.43028155204 mfu:23.38976163352329 +Step 26761 | loss:0.36255407333374023 lr:1.617136581793319e-05 tokens_per_second_per_gpu:5316.022028594775 mfu:25.07658291576722 +Step 26762 | loss:0.3834342360496521 lr:1.6168172856146102e-05 tokens_per_second_per_gpu:5344.173037817153 mfu:25.209376029326624 +Step 26763 | loss:0.5064051151275635 lr:1.616498014881252e-05 tokens_per_second_per_gpu:5350.035100693304 mfu:25.23702838008449 +Step 26764 | loss:0.3462586998939514 lr:1.616178769595642e-05 tokens_per_second_per_gpu:4215.472560871775 mfu:19.885103303416088 +Step 26765 | loss:0.2849368453025818 lr:1.6158595497601835e-05 tokens_per_second_per_gpu:5334.559349322755 mfu:25.164026620434157 +Step 26766 | loss:0.4578673839569092 lr:1.6155403553772786e-05 tokens_per_second_per_gpu:4065.1786084321966 mfu:19.17614108696611 +Step 26767 | loss:0.5017362236976624 lr:1.615221186449325e-05 tokens_per_second_per_gpu:4399.432326343613 mfu:20.752872903919236 +Step 26768 | loss:0.4781478941440582 lr:1.6149020429787264e-05 tokens_per_second_per_gpu:3400.535357787257 mfu:16.040905473841946 +Step 26769 | loss:0.29617395997047424 lr:1.61458292496788e-05 tokens_per_second_per_gpu:5125.319877661421 mfu:24.177008332672514 +Step 26770 | loss:0.38222160935401917 lr:1.614263832419189e-05 tokens_per_second_per_gpu:4080.3601047811558 mfu:19.24775479547452 +Step 26771 | loss:0.325850248336792 lr:1.6139447653350497e-05 tokens_per_second_per_gpu:5043.23515293965 mfu:23.789800681061724 +Step 26772 | loss:0.524970531463623 lr:1.6136257237178643e-05 tokens_per_second_per_gpu:3808.686446214169 mfu:17.966223795708196 +Step 26773 | loss:0.42594510316848755 lr:1.613306707570032e-05 tokens_per_second_per_gpu:3522.4362027466073 mfu:16.615932557944017 +Step 26774 | loss:0.5088642239570618 lr:1.6129877168939532e-05 tokens_per_second_per_gpu:4404.2189375824355 mfu:20.775452165812855 +Step 26775 | loss:0.4292461574077606 lr:1.6126687516920264e-05 tokens_per_second_per_gpu:5340.618162770267 mfu:25.192607077206237 +Step 26776 | loss:0.4608570635318756 lr:1.6123498119666487e-05 tokens_per_second_per_gpu:4300.207787956158 mfu:20.284813826893654 +Step 26777 | loss:0.45021817088127136 lr:1.6120308977202196e-05 tokens_per_second_per_gpu:4741.501604978627 mfu:22.36647205427774 +Step 26778 | loss:0.4091492295265198 lr:1.6117120089551406e-05 tokens_per_second_per_gpu:5263.539008626086 mfu:24.829011556048343 +Step 26779 | loss:0.3853146731853485 lr:1.6113931456738064e-05 tokens_per_second_per_gpu:4735.748409333084 mfu:22.339333248821315 +Step 26780 | loss:0.3645913898944855 lr:1.6110743078786162e-05 tokens_per_second_per_gpu:4161.639887396743 mfu:19.631165397833158 +Step 26781 | loss:0.38974639773368835 lr:1.6107554955719704e-05 tokens_per_second_per_gpu:5223.860699165441 mfu:24.641842200504804 +Step 26782 | loss:0.31014424562454224 lr:1.610436708756264e-05 tokens_per_second_per_gpu:3420.9156762503335 mfu:16.13704291327317 +Step 26783 | loss:0.3878258466720581 lr:1.610117947433897e-05 tokens_per_second_per_gpu:4435.659658861502 mfu:20.923763412426414 +Step 26784 | loss:0.2681412398815155 lr:1.609799211607263e-05 tokens_per_second_per_gpu:4177.997031475608 mfu:19.70832483726973 +Step 26785 | loss:0.42496469616889954 lr:1.609480501278764e-05 tokens_per_second_per_gpu:3952.742086143695 mfu:18.645758827682386 +Step 26786 | loss:0.40119418501853943 lr:1.6091618164507933e-05 tokens_per_second_per_gpu:3092.8917288820317 mfu:14.58969798688037 +Step 26787 | loss:0.3168170154094696 lr:1.608843157125749e-05 tokens_per_second_per_gpu:3663.0969017563016 mfu:17.279453074362717 +Step 26788 | loss:0.3507269322872162 lr:1.608524523306029e-05 tokens_per_second_per_gpu:3961.7801993412468 mfu:18.688393149696363 +Step 26789 | loss:0.413967102766037 lr:1.6082059149940282e-05 tokens_per_second_per_gpu:5368.394018105452 mfu:25.323630525871902 +Step 26790 | loss:0.5143190026283264 lr:1.607887332192145e-05 tokens_per_second_per_gpu:4516.326574273016 mfu:21.304283015617514 +Step 26791 | loss:0.31323912739753723 lr:1.6075687749027722e-05 tokens_per_second_per_gpu:5208.7021720946295 mfu:24.57033684200038 +Step 26792 | loss:0.4794425666332245 lr:1.6072502431283093e-05 tokens_per_second_per_gpu:4599.963288110368 mfu:21.69881166468314 +Step 26793 | loss:0.418412983417511 lr:1.6069317368711486e-05 tokens_per_second_per_gpu:5328.695390481549 mfu:25.136365326085638 +Step 26794 | loss:0.371305376291275 lr:1.6066132561336865e-05 tokens_per_second_per_gpu:4458.0995294188715 mfu:21.029616110482237 +Step 26795 | loss:0.3835340142250061 lr:1.6062948009183216e-05 tokens_per_second_per_gpu:4642.582073678661 mfu:21.899851747723414 +Step 26796 | loss:0.5006053447723389 lr:1.6059763712274444e-05 tokens_per_second_per_gpu:5278.49884753555 mfu:24.899579668596004 +Step 26797 | loss:0.42325207591056824 lr:1.6056579670634542e-05 tokens_per_second_per_gpu:3830.7928579634136 mfu:18.070503511672154 +Step 26798 | loss:0.5094830393791199 lr:1.6053395884287415e-05 tokens_per_second_per_gpu:5049.628902724537 mfu:23.819961089683236 +Step 26799 | loss:0.42738035321235657 lr:1.605021235325705e-05 tokens_per_second_per_gpu:3934.5452568973974 mfu:18.559921279428384 +Step 26800 | loss:0.3588317334651947 lr:1.604702907756735e-05 tokens_per_second_per_gpu:4778.66811218566 mfu:22.541792810033332 +Step 26801 | loss:0.44305145740509033 lr:1.6043846057242273e-05 tokens_per_second_per_gpu:4667.657264844619 mfu:22.018135702722013 +Step 26802 | loss:0.3099234700202942 lr:1.6040663292305784e-05 tokens_per_second_per_gpu:5332.01578786389 mfu:25.15202820705635 +Step 26803 | loss:0.4403030574321747 lr:1.6037480782781784e-05 tokens_per_second_per_gpu:5347.3088710309 mfu:25.224168289624185 +Step 26804 | loss:0.4319131374359131 lr:1.6034298528694236e-05 tokens_per_second_per_gpu:5342.778664102781 mfu:25.202798530611243 +Step 26805 | loss:0.29837504029273987 lr:1.6031116530067052e-05 tokens_per_second_per_gpu:5318.650966165924 mfu:25.08898406283333 +Step 26806 | loss:0.34390202164649963 lr:1.602793478692419e-05 tokens_per_second_per_gpu:4340.975530997216 mfu:20.477122226512535 +Step 26807 | loss:0.44939857721328735 lr:1.6024753299289545e-05 tokens_per_second_per_gpu:4357.787266605316 mfu:20.556426051753817 +Step 26808 | loss:0.39126938581466675 lr:1.602157206718707e-05 tokens_per_second_per_gpu:5319.971532422124 mfu:25.095213399175666 +Step 26809 | loss:0.4338943660259247 lr:1.601839109064068e-05 tokens_per_second_per_gpu:3936.2910436112074 mfu:18.568156453219355 +Step 26810 | loss:0.39570218324661255 lr:1.6015210369674332e-05 tokens_per_second_per_gpu:3814.1412930722636 mfu:17.99195523903067 +Step 26811 | loss:0.4417346119880676 lr:1.6012029904311905e-05 tokens_per_second_per_gpu:4558.388513557494 mfu:21.50269635972925 +Step 26812 | loss:0.46329620480537415 lr:1.6008849694577356e-05 tokens_per_second_per_gpu:4096.524077144175 mfu:19.32400301096895 +Step 26813 | loss:0.363654226064682 lr:1.6005669740494566e-05 tokens_per_second_per_gpu:4518.063169262939 mfu:21.31247483047863 +Step 26814 | loss:0.354904443025589 lr:1.6002490042087486e-05 tokens_per_second_per_gpu:5365.090040954846 mfu:25.308045101935864 +Step 26815 | loss:0.43612146377563477 lr:1.599931059938e-05 tokens_per_second_per_gpu:4280.922481858963 mfu:20.193841747620667 +Step 26816 | loss:0.5092917680740356 lr:1.599613141239604e-05 tokens_per_second_per_gpu:5235.999138972052 mfu:24.69910129210398 +Step 26817 | loss:0.35982462763786316 lr:1.5992952481159535e-05 tokens_per_second_per_gpu:5339.489482409462 mfu:25.187282899371592 +Step 26818 | loss:0.3062640428543091 lr:1.598977380569435e-05 tokens_per_second_per_gpu:4760.654380867126 mfu:22.45681896594432 +Step 26819 | loss:0.4255083203315735 lr:1.5986595386024437e-05 tokens_per_second_per_gpu:5383.304266908568 mfu:25.393964713426968 +Step 26820 | loss:0.3455381691455841 lr:1.5983417222173662e-05 tokens_per_second_per_gpu:4041.8045433784514 mfu:19.06588163408086 +Step 26821 | loss:0.4702141284942627 lr:1.5980239314165967e-05 tokens_per_second_per_gpu:5350.275124881216 mfu:25.238160615131218 +Step 26822 | loss:0.4219815731048584 lr:1.597706166202521e-05 tokens_per_second_per_gpu:5341.695065416693 mfu:25.197687010727968 +Step 26823 | loss:0.6290056705474854 lr:1.597388426577532e-05 tokens_per_second_per_gpu:4719.202186553705 mfu:22.261281892894274 +Step 26824 | loss:0.413629412651062 lr:1.5970707125440198e-05 tokens_per_second_per_gpu:5357.017710038956 mfu:25.26996653972393 +Step 26825 | loss:0.3766278922557831 lr:1.596753024104372e-05 tokens_per_second_per_gpu:5306.544486558526 mfu:25.03187573294682 +Step 26826 | loss:0.37223249673843384 lr:1.59643536126098e-05 tokens_per_second_per_gpu:5325.04101616717 mfu:25.119127018944162 +Step 26827 | loss:0.44275903701782227 lr:1.5961177240162307e-05 tokens_per_second_per_gpu:4692.981588068531 mfu:22.13759485614429 +Step 26828 | loss:0.359773725271225 lr:1.595800112372516e-05 tokens_per_second_per_gpu:4862.494499844534 mfu:22.93721618287678 +Step 26829 | loss:0.4565214514732361 lr:1.5954825263322215e-05 tokens_per_second_per_gpu:3615.5619506467074 mfu:17.05522259967979 +Step 26830 | loss:0.4686289131641388 lr:1.5951649658977368e-05 tokens_per_second_per_gpu:3233.5862301310412 mfu:15.253377954229174 +Step 26831 | loss:0.45165130496025085 lr:1.5948474310714528e-05 tokens_per_second_per_gpu:4156.8100556306445 mfu:19.608382257338196 +Step 26832 | loss:0.45081841945648193 lr:1.594529921855754e-05 tokens_per_second_per_gpu:4637.811410957387 mfu:21.877347717708172 +Step 26833 | loss:0.46229976415634155 lr:1.594212438253032e-05 tokens_per_second_per_gpu:4414.036878920946 mfu:20.82176507022013 +Step 26834 | loss:0.311897337436676 lr:1.5938949802656717e-05 tokens_per_second_per_gpu:4522.238091275768 mfu:21.332168649928295 +Step 26835 | loss:0.4621738791465759 lr:1.5935775478960624e-05 tokens_per_second_per_gpu:5187.368985620159 mfu:24.46970456158336 +Step 26836 | loss:0.4168374538421631 lr:1.5932601411465902e-05 tokens_per_second_per_gpu:3950.465934345254 mfu:18.635021831297596 +Step 26837 | loss:0.40409034490585327 lr:1.5929427600196427e-05 tokens_per_second_per_gpu:3739.3978194541774 mfu:17.639377521422556 +Step 26838 | loss:0.4454576373100281 lr:1.5926254045176097e-05 tokens_per_second_per_gpu:5016.471937620312 mfu:23.663554028125017 +Step 26839 | loss:0.4817233681678772 lr:1.5923080746428735e-05 tokens_per_second_per_gpu:5308.4743741738885 mfu:25.040979342101263 +Step 26840 | loss:0.5032235383987427 lr:1.5919907703978236e-05 tokens_per_second_per_gpu:5344.018549093925 mfu:25.208647279660664 +Step 26841 | loss:0.4278465211391449 lr:1.5916734917848475e-05 tokens_per_second_per_gpu:4778.953606092717 mfu:22.543139533503204 +Step 26842 | loss:0.48236146569252014 lr:1.5913562388063295e-05 tokens_per_second_per_gpu:3150.879951033022 mfu:14.863238324577507 +Step 26843 | loss:0.5357574224472046 lr:1.5910390114646552e-05 tokens_per_second_per_gpu:3067.492401244466 mfu:14.469885024841814 +Step 26844 | loss:0.4409727156162262 lr:1.590721809762211e-05 tokens_per_second_per_gpu:3541.563058722979 mfu:16.7061572009627 +Step 26845 | loss:0.4844851791858673 lr:1.5904046337013835e-05 tokens_per_second_per_gpu:5104.841339824958 mfu:24.080407575699073 +Step 26846 | loss:0.38608163595199585 lr:1.590087483284559e-05 tokens_per_second_per_gpu:4279.675397340018 mfu:20.187959037170245 +Step 26847 | loss:0.39778563380241394 lr:1.5897703585141206e-05 tokens_per_second_per_gpu:4946.042236810995 mfu:23.3313251128614 +Step 26848 | loss:0.355214923620224 lr:1.589453259392456e-05 tokens_per_second_per_gpu:5325.338658697205 mfu:25.120531049541388 +Step 26849 | loss:0.3357097804546356 lr:1.5891361859219466e-05 tokens_per_second_per_gpu:5349.2260545271065 mfu:25.233211971281026 +Step 26850 | loss:0.39108964800834656 lr:1.588819138104981e-05 tokens_per_second_per_gpu:5170.397895076023 mfu:24.389649031920044 +Step 26851 | loss:0.4150111675262451 lr:1.58850211594394e-05 tokens_per_second_per_gpu:4318.099862328347 mfu:20.369213794409937 +Step 26852 | loss:0.5227470397949219 lr:1.5881851194412107e-05 tokens_per_second_per_gpu:4238.44001812877 mfu:19.993444717949618 +Step 26853 | loss:0.3623785078525543 lr:1.5878681485991776e-05 tokens_per_second_per_gpu:4206.8726635105795 mfu:19.844536120269535 +Step 26854 | loss:0.4180888533592224 lr:1.5875512034202218e-05 tokens_per_second_per_gpu:5079.318894219718 mfu:23.960013845200905 +Step 26855 | loss:0.3916829824447632 lr:1.5872342839067306e-05 tokens_per_second_per_gpu:5309.240506333943 mfu:25.044593318215867 +Step 26856 | loss:0.5175004005432129 lr:1.586917390061084e-05 tokens_per_second_per_gpu:5317.173824431921 mfu:25.082016133247674 +Step 26857 | loss:0.28143075108528137 lr:1.586600521885669e-05 tokens_per_second_per_gpu:4048.21010741057 mfu:19.096097772522285 +Step 26858 | loss:0.442158043384552 lr:1.586283679382865e-05 tokens_per_second_per_gpu:3463.0197579622873 mfu:16.335655050405556 +Step 26859 | loss:0.4483287036418915 lr:1.5859668625550578e-05 tokens_per_second_per_gpu:4571.185330192354 mfu:21.563061127157944 +Step 26860 | loss:0.431012898683548 lr:1.5856500714046308e-05 tokens_per_second_per_gpu:4684.441312302517 mfu:22.097308918234706 +Step 26861 | loss:0.39722922444343567 lr:1.5853333059339638e-05 tokens_per_second_per_gpu:4291.045871986457 mfu:20.241595506080518 +Step 26862 | loss:0.4345685839653015 lr:1.585016566145442e-05 tokens_per_second_per_gpu:4687.049632933716 mfu:22.109612811677454 +Step 26863 | loss:0.39204007387161255 lr:1.5846998520414448e-05 tokens_per_second_per_gpu:5322.170487732692 mfu:25.105586246556076 +Step 26864 | loss:0.4381761848926544 lr:1.584383163624357e-05 tokens_per_second_per_gpu:5300.296557981756 mfu:25.002403187843655 +Step 26865 | loss:0.3309876620769501 lr:1.5840665008965583e-05 tokens_per_second_per_gpu:3383.333743702532 mfu:15.9597625253063 +Step 26866 | loss:0.47037628293037415 lr:1.5837498638604316e-05 tokens_per_second_per_gpu:4437.570431317917 mfu:20.932776852114717 +Step 26867 | loss:0.4327133893966675 lr:1.583433252518359e-05 tokens_per_second_per_gpu:5342.563212697194 mfu:25.20178220957129 +Step 26868 | loss:0.33499088883399963 lr:1.5831166668727195e-05 tokens_per_second_per_gpu:3954.6828062528866 mfu:18.65491353555844 +Step 26869 | loss:0.36194005608558655 lr:1.5828001069258975e-05 tokens_per_second_per_gpu:4944.944041453602 mfu:23.32614473798913 +Step 26870 | loss:0.39151811599731445 lr:1.5824835726802696e-05 tokens_per_second_per_gpu:4227.930673433837 mfu:19.943870345944518 +Step 26871 | loss:0.4338516891002655 lr:1.5821670641382207e-05 tokens_per_second_per_gpu:5336.54231643893 mfu:25.173380614650227 +Step 26872 | loss:0.3534647524356842 lr:1.5818505813021283e-05 tokens_per_second_per_gpu:4171.734421439909 mfu:19.67878303721988 +Step 26873 | loss:0.4436948597431183 lr:1.5815341241743735e-05 tokens_per_second_per_gpu:3578.20781126311 mfu:16.879016750933992 +Step 26874 | loss:0.3877739906311035 lr:1.581217692757338e-05 tokens_per_second_per_gpu:5249.467511366269 mfu:24.76263390262879 +Step 26875 | loss:0.36292213201522827 lr:1.5809012870533995e-05 tokens_per_second_per_gpu:4878.8158294237755 mfu:23.014206679208037 +Step 26876 | loss:0.488452672958374 lr:1.5805849070649383e-05 tokens_per_second_per_gpu:4774.6882210078975 mfu:22.523018984308372 +Step 26877 | loss:0.39154714345932007 lr:1.580268552794336e-05 tokens_per_second_per_gpu:4275.0088852435 mfu:20.165946303421904 +Step 26878 | loss:0.38157713413238525 lr:1.579952224243971e-05 tokens_per_second_per_gpu:4407.442964412496 mfu:20.790660450445944 +Step 26879 | loss:0.3732074201107025 lr:1.579635921416219e-05 tokens_per_second_per_gpu:3900.4586971473796 mfu:18.399129161320758 +Step 26880 | loss:0.5095394849777222 lr:1.5793196443134623e-05 tokens_per_second_per_gpu:4343.606073397298 mfu:20.489530943830538 +Step 26881 | loss:0.3490445911884308 lr:1.579003392938079e-05 tokens_per_second_per_gpu:4430.579059057285 mfu:20.89979735631021 +Step 26882 | loss:0.5293467044830322 lr:1.5786871672924485e-05 tokens_per_second_per_gpu:5362.92583099518 mfu:25.297836154303823 +Step 26883 | loss:0.29952701926231384 lr:1.5783709673789475e-05 tokens_per_second_per_gpu:3553.441151432655 mfu:16.762188190885837 +Step 26884 | loss:0.43615803122520447 lr:1.578054793199957e-05 tokens_per_second_per_gpu:5296.392186311371 mfu:24.98398560051995 +Step 26885 | loss:0.3077322244644165 lr:1.5777386447578522e-05 tokens_per_second_per_gpu:5038.188660054424 mfu:23.765995513101643 +Step 26886 | loss:0.49393802881240845 lr:1.577422522055011e-05 tokens_per_second_per_gpu:4502.582004828753 mfu:21.23944753648356 +Step 26887 | loss:0.36019614338874817 lr:1.5771064250938112e-05 tokens_per_second_per_gpu:4908.881594774013 mfu:23.15603202411364 +Step 26888 | loss:0.461914986371994 lr:1.576790353876631e-05 tokens_per_second_per_gpu:4145.7002889832065 mfu:19.555975592540584 +Step 26889 | loss:0.43141165375709534 lr:1.5764743084058485e-05 tokens_per_second_per_gpu:5222.712711439608 mfu:24.63642694653515 +Step 26890 | loss:0.49705976247787476 lr:1.5761582886838382e-05 tokens_per_second_per_gpu:4090.3209971075826 mfu:19.294742024081756 +Step 26891 | loss:0.23974286019802094 lr:1.5758422947129804e-05 tokens_per_second_per_gpu:5360.988676245565 mfu:25.28869826483721 +Step 26892 | loss:0.35341688990592957 lr:1.5755263264956477e-05 tokens_per_second_per_gpu:5339.068857560307 mfu:25.18529874019213 +Step 26893 | loss:0.39823800325393677 lr:1.5752103840342196e-05 tokens_per_second_per_gpu:3726.4958275457634 mfu:17.578516624283758 +Step 26894 | loss:0.3947349488735199 lr:1.57489446733107e-05 tokens_per_second_per_gpu:4292.044936214361 mfu:20.246308262501476 +Step 26895 | loss:0.5075006484985352 lr:1.574578576388576e-05 tokens_per_second_per_gpu:3559.97457941113 mfu:16.793007485378062 +Step 26896 | loss:0.3963301181793213 lr:1.5742627112091158e-05 tokens_per_second_per_gpu:5051.501853264146 mfu:23.828796116936886 +Step 26897 | loss:0.5631264448165894 lr:1.5739468717950604e-05 tokens_per_second_per_gpu:4435.02679199596 mfu:20.92077807144343 +Step 26898 | loss:0.4040844142436981 lr:1.57363105814879e-05 tokens_per_second_per_gpu:4728.832816976083 mfu:22.30671121975162 +Step 26899 | loss:0.4235137701034546 lr:1.5733152702726755e-05 tokens_per_second_per_gpu:3890.131555894165 mfu:18.350414274037355 +Step 26900 | loss:0.4590309262275696 lr:1.5729995081690964e-05 tokens_per_second_per_gpu:3840.640193814611 mfu:18.116955075011 +Step 26901 | loss:0.2676984369754791 lr:1.572683771840423e-05 tokens_per_second_per_gpu:4703.20651914826 mfu:22.185827600600504 +Step 26902 | loss:0.4189969301223755 lr:1.572368061289033e-05 tokens_per_second_per_gpu:5371.834979430201 mfu:25.339862127529283 +Step 26903 | loss:0.557655394077301 lr:1.5720523765173013e-05 tokens_per_second_per_gpu:4610.918963310537 mfu:21.750491453833277 +Step 26904 | loss:0.38254496455192566 lr:1.5717367175275992e-05 tokens_per_second_per_gpu:5291.209058731354 mfu:24.959535903392254 +Step 26905 | loss:0.35975322127342224 lr:1.571421084322303e-05 tokens_per_second_per_gpu:4714.791709426966 mfu:22.240476919782452 +Step 26906 | loss:0.39397063851356506 lr:1.571105476903788e-05 tokens_per_second_per_gpu:4738.257530760662 mfu:22.351169202695463 +Step 26907 | loss:0.4240291118621826 lr:1.570789895274426e-05 tokens_per_second_per_gpu:4102.063834718141 mfu:19.350134992625453 +Step 26908 | loss:0.48856624960899353 lr:1.5704743394365896e-05 tokens_per_second_per_gpu:4415.482832379225 mfu:20.828585879387937 +Step 26909 | loss:0.36300522089004517 lr:1.5701588093926527e-05 tokens_per_second_per_gpu:5136.085578446602 mfu:24.227791980094175 +Step 26910 | loss:0.3319748342037201 lr:1.5698433051449912e-05 tokens_per_second_per_gpu:4970.873329029469 mfu:23.448457611477142 +Step 26911 | loss:0.36731433868408203 lr:1.569527826695974e-05 tokens_per_second_per_gpu:4859.458601010048 mfu:22.922895330096758 +Step 26912 | loss:0.44298455119132996 lr:1.5692123740479758e-05 tokens_per_second_per_gpu:4648.417190370251 mfu:21.92737698873121 +Step 26913 | loss:0.3664399981498718 lr:1.568896947203371e-05 tokens_per_second_per_gpu:3801.212965982266 mfu:17.930970114347094 +Step 26914 | loss:0.4149560332298279 lr:1.56858154616453e-05 tokens_per_second_per_gpu:5396.033088235396 mfu:25.45400873538648 +Step 26915 | loss:0.4687401354312897 lr:1.5682661709338233e-05 tokens_per_second_per_gpu:5355.497799247636 mfu:25.262796861197756 +Step 26916 | loss:0.44784843921661377 lr:1.567950821513624e-05 tokens_per_second_per_gpu:5354.421435959682 mfu:25.257719471921963 +Step 26917 | loss:0.5813286304473877 lr:1.567635497906305e-05 tokens_per_second_per_gpu:4399.968446732126 mfu:20.75540187526342 +Step 26918 | loss:0.42328742146492004 lr:1.5673202001142396e-05 tokens_per_second_per_gpu:3956.7718999524973 mfu:18.66476814697564 +Step 26919 | loss:0.47994187474250793 lr:1.5670049281397946e-05 tokens_per_second_per_gpu:5349.880025576425 mfu:25.236296864299632 +Step 26920 | loss:0.41809433698654175 lr:1.5666896819853455e-05 tokens_per_second_per_gpu:4862.721827179719 mfu:22.938288524702887 +Step 26921 | loss:0.41063404083251953 lr:1.566374461653261e-05 tokens_per_second_per_gpu:4287.1972972549665 mfu:20.223441122437517 +Step 26922 | loss:0.23109519481658936 lr:1.5660592671459113e-05 tokens_per_second_per_gpu:5380.789348419994 mfu:25.382101413827026 +Step 26923 | loss:0.4896169900894165 lr:1.5657440984656678e-05 tokens_per_second_per_gpu:4905.751659013288 mfu:23.14126758310819 +Step 26924 | loss:0.49525704979896545 lr:1.565428955614901e-05 tokens_per_second_per_gpu:4126.401501207002 mfu:19.4649399178875 +Step 26925 | loss:0.4057339131832123 lr:1.565113838595983e-05 tokens_per_second_per_gpu:5386.406666379336 mfu:25.408599261054512 +Step 26926 | loss:0.37321561574935913 lr:1.564798747411281e-05 tokens_per_second_per_gpu:5003.617616432494 mfu:23.602918001908726 +Step 26927 | loss:0.4954446852207184 lr:1.5644836820631664e-05 tokens_per_second_per_gpu:5377.959315060171 mfu:25.36875166361526 +Step 26928 | loss:0.4906035363674164 lr:1.5641686425540074e-05 tokens_per_second_per_gpu:5332.284885345191 mfu:25.153297585788412 +Step 26929 | loss:0.4191926419734955 lr:1.5638536288861767e-05 tokens_per_second_per_gpu:4882.168782575263 mfu:23.0300231312964 +Step 26930 | loss:0.4027484357357025 lr:1.5635386410620388e-05 tokens_per_second_per_gpu:4869.847091108112 mfu:22.9718996103485 +Step 26931 | loss:0.3581311106681824 lr:1.5632236790839654e-05 tokens_per_second_per_gpu:5317.858553308814 mfu:25.08524611618677 +Step 26932 | loss:0.4578004777431488 lr:1.5629087429543273e-05 tokens_per_second_per_gpu:4083.256662916212 mfu:19.26141835440136 +Step 26933 | loss:0.5688766837120056 lr:1.5625938326754892e-05 tokens_per_second_per_gpu:5008.939070299132 mfu:23.628020207731314 +Step 26934 | loss:0.3307524621486664 lr:1.562278948249822e-05 tokens_per_second_per_gpu:4174.394203008761 mfu:19.691329680685683 +Step 26935 | loss:0.42331528663635254 lr:1.5619640896796944e-05 tokens_per_second_per_gpu:5319.175399676729 mfu:25.091457905181382 +Step 26936 | loss:0.4102020263671875 lr:1.561649256967474e-05 tokens_per_second_per_gpu:4581.281665710767 mfu:21.61068726441077 +Step 26937 | loss:0.36289989948272705 lr:1.5613344501155265e-05 tokens_per_second_per_gpu:5339.496724944506 mfu:25.1873170636451 +Step 26938 | loss:0.4738888740539551 lr:1.5610196691262213e-05 tokens_per_second_per_gpu:3656.3297992749144 mfu:17.247531497371273 +Step 26939 | loss:0.46918433904647827 lr:1.5607049140019275e-05 tokens_per_second_per_gpu:3088.296738507657 mfu:14.568022633298247 +Step 26940 | loss:0.4263886511325836 lr:1.56039018474501e-05 tokens_per_second_per_gpu:4332.335283878629 mfu:20.436364706675988 +Step 26941 | loss:0.48501667380332947 lr:1.560075481357836e-05 tokens_per_second_per_gpu:5383.228787751672 mfu:25.393608665365814 +Step 26942 | loss:0.33445876836776733 lr:1.5597608038427753e-05 tokens_per_second_per_gpu:4690.78337010474 mfu:22.12722548695443 +Step 26943 | loss:0.3300700783729553 lr:1.559446152202192e-05 tokens_per_second_per_gpu:4681.2644322772885 mfu:22.082323033124077 +Step 26944 | loss:0.4241783618927002 lr:1.559131526438452e-05 tokens_per_second_per_gpu:5329.94024107841 mfu:25.142237498744446 +Step 26945 | loss:0.43833214044570923 lr:1.558816926553923e-05 tokens_per_second_per_gpu:4777.307980398729 mfu:22.535376836331967 +Step 26946 | loss:0.40754643082618713 lr:1.5585023525509724e-05 tokens_per_second_per_gpu:4016.459914741073 mfu:18.9463266965588 +Step 26947 | loss:0.3697640299797058 lr:1.558187804431963e-05 tokens_per_second_per_gpu:4384.146520043137 mfu:20.680767147572404 +Step 26948 | loss:0.4643406569957733 lr:1.5578732821992625e-05 tokens_per_second_per_gpu:4261.307087284324 mfu:20.101312584680674 +Step 26949 | loss:0.5180772542953491 lr:1.5575587858552383e-05 tokens_per_second_per_gpu:5307.561210428988 mfu:25.036671792914767 +Step 26950 | loss:0.41218680143356323 lr:1.557244315402253e-05 tokens_per_second_per_gpu:3792.881578865623 mfu:17.89166953983667 +Step 26951 | loss:0.5444594025611877 lr:1.556929870842671e-05 tokens_per_second_per_gpu:4047.380108343617 mfu:19.09218252531102 +Step 26952 | loss:0.3848833441734314 lr:1.5566154521788597e-05 tokens_per_second_per_gpu:5335.183807067858 mfu:25.166972294161283 +Step 26953 | loss:0.4649522602558136 lr:1.5563010594131834e-05 tokens_per_second_per_gpu:5337.131230909482 mfu:25.176158624686995 +Step 26954 | loss:0.4309728145599365 lr:1.555986692548005e-05 tokens_per_second_per_gpu:4991.157444382397 mfu:23.544141244423855 +Step 26955 | loss:0.4052821695804596 lr:1.55567235158569e-05 tokens_per_second_per_gpu:3767.61368702125 mfu:17.772476582859863 +Step 26956 | loss:0.4370833933353424 lr:1.5553580365286047e-05 tokens_per_second_per_gpu:4825.144780724315 mfu:22.761031185267505 +Step 26957 | loss:0.3386370837688446 lr:1.5550437473791112e-05 tokens_per_second_per_gpu:5349.8821666142485 mfu:25.23630696394034 +Step 26958 | loss:0.4298572242259979 lr:1.5547294841395716e-05 tokens_per_second_per_gpu:5268.552369202103 mfu:24.852660433252936 +Step 26959 | loss:0.29149800539016724 lr:1.5544152468123507e-05 tokens_per_second_per_gpu:4319.022216269964 mfu:20.373564695322358 +Step 26960 | loss:0.4122582972049713 lr:1.554101035399813e-05 tokens_per_second_per_gpu:4091.4202376080625 mfu:19.299927329072286 +Step 26961 | loss:0.3092094659805298 lr:1.553786849904322e-05 tokens_per_second_per_gpu:5351.593169280586 mfu:25.244378055444614 +Step 26962 | loss:0.3388362228870392 lr:1.5534726903282393e-05 tokens_per_second_per_gpu:4575.47543225927 mfu:21.583298270574478 +Step 26963 | loss:0.4509110748767853 lr:1.5531585566739277e-05 tokens_per_second_per_gpu:5338.314369509728 mfu:25.181739691328612 +Step 26964 | loss:0.44784632325172424 lr:1.5528444489437516e-05 tokens_per_second_per_gpu:4708.32193567879 mfu:22.209957893155078 +Step 26965 | loss:0.3889799118041992 lr:1.5525303671400727e-05 tokens_per_second_per_gpu:4227.35795820099 mfu:19.94116875047588 +Step 26966 | loss:0.4606683552265167 lr:1.552216311265251e-05 tokens_per_second_per_gpu:3712.7492477623045 mfu:17.51367166203682 +Step 26967 | loss:0.32490456104278564 lr:1.551902281321651e-05 tokens_per_second_per_gpu:4219.097823118285 mfu:19.902204283966395 +Step 26968 | loss:0.3807239830493927 lr:1.551588277311635e-05 tokens_per_second_per_gpu:3548.7826075212847 mfu:16.74021304442635 +Step 26969 | loss:0.33045607805252075 lr:1.5512742992375618e-05 tokens_per_second_per_gpu:5320.660845793192 mfu:25.098465007955976 +Step 26970 | loss:0.5171202421188354 lr:1.550960347101795e-05 tokens_per_second_per_gpu:5311.729929790893 mfu:25.05633635340174 +Step 26971 | loss:0.49451136589050293 lr:1.550646420906697e-05 tokens_per_second_per_gpu:3866.205578647147 mfu:18.237551357170506 +Step 26972 | loss:0.396025151014328 lr:1.5503325206546277e-05 tokens_per_second_per_gpu:5237.5620600723505 mfu:24.706473857595924 +Step 26973 | loss:0.456502228975296 lr:1.5500186463479454e-05 tokens_per_second_per_gpu:3874.521721041231 mfu:18.276780019723606 +Step 26974 | loss:0.3753524720668793 lr:1.5497047979890132e-05 tokens_per_second_per_gpu:5385.298437501969 mfu:25.40337155635663 +Step 26975 | loss:0.3750082850456238 lr:1.5493909755801938e-05 tokens_per_second_per_gpu:3257.039904423973 mfu:15.3640129374784 +Step 26976 | loss:0.3586595356464386 lr:1.5490771791238428e-05 tokens_per_second_per_gpu:5307.043775204704 mfu:25.034230962678265 +Step 26977 | loss:0.3574451506137848 lr:1.5487634086223235e-05 tokens_per_second_per_gpu:4147.919304389902 mfu:19.566443066817037 +Step 26978 | loss:0.3774033188819885 lr:1.548449664077996e-05 tokens_per_second_per_gpu:5345.882154258305 mfu:25.21743822318487 +Step 26979 | loss:0.4571131765842438 lr:1.548135945493219e-05 tokens_per_second_per_gpu:3017.406508846142 mfu:14.233621324864488 +Step 26980 | loss:0.34256836771965027 lr:1.5478222528703506e-05 tokens_per_second_per_gpu:4272.01663149154 mfu:20.15183133194277 +Step 26981 | loss:0.44385969638824463 lr:1.5475085862117512e-05 tokens_per_second_per_gpu:4283.926793024715 mfu:20.20801359597786 +Step 26982 | loss:0.4628177285194397 lr:1.547194945519782e-05 tokens_per_second_per_gpu:5374.089819188687 mfu:25.350498591385396 +Step 26983 | loss:0.3028240203857422 lr:1.546881330796799e-05 tokens_per_second_per_gpu:4502.941407311311 mfu:21.241142899314173 +Step 26984 | loss:0.432132363319397 lr:1.5465677420451617e-05 tokens_per_second_per_gpu:4102.229981060694 mfu:19.350918733270763 +Step 26985 | loss:0.3366104066371918 lr:1.546254179267231e-05 tokens_per_second_per_gpu:5335.725598161892 mfu:25.16952801519093 +Step 26986 | loss:0.5219681262969971 lr:1.5459406424653632e-05 tokens_per_second_per_gpu:5327.6515750876315 mfu:25.13144147077926 +Step 26987 | loss:0.48721927404403687 lr:1.5456271316419146e-05 tokens_per_second_per_gpu:5204.7088782028995 mfu:24.551499793386895 +Step 26988 | loss:0.44878682494163513 lr:1.5453136467992457e-05 tokens_per_second_per_gpu:5283.367801207259 mfu:24.92254735379461 +Step 26989 | loss:0.3752020001411438 lr:1.5450001879397147e-05 tokens_per_second_per_gpu:5361.412837276852 mfu:25.290699104791123 +Step 26990 | loss:0.3947666585445404 lr:1.544686755065677e-05 tokens_per_second_per_gpu:4303.095913123083 mfu:20.298437606071 +Step 26991 | loss:0.48081129789352417 lr:1.5443733481794908e-05 tokens_per_second_per_gpu:3850.6938425586927 mfu:18.164379851466187 +Step 26992 | loss:0.4100615680217743 lr:1.5440599672835153e-05 tokens_per_second_per_gpu:4092.0648435564513 mfu:19.302968045311953 +Step 26993 | loss:0.40993908047676086 lr:1.5437466123801058e-05 tokens_per_second_per_gpu:5246.368831360575 mfu:24.748016900352606 +Step 26994 | loss:0.44163191318511963 lr:1.5434332834716177e-05 tokens_per_second_per_gpu:3981.4077963571076 mfu:18.780979873633548 +Step 26995 | loss:0.4958375096321106 lr:1.5431199805604084e-05 tokens_per_second_per_gpu:5401.1551071523945 mfu:25.4781701725245 +Step 26996 | loss:0.36944031715393066 lr:1.5428067036488352e-05 tokens_per_second_per_gpu:4778.37717412905 mfu:22.540420405581152 +Step 26997 | loss:0.4726963937282562 lr:1.542493452739256e-05 tokens_per_second_per_gpu:5351.926202644122 mfu:25.245949030641764 +Step 26998 | loss:0.44490349292755127 lr:1.5421802278340225e-05 tokens_per_second_per_gpu:5105.707291872269 mfu:24.084492419252893 +Step 26999 | loss:0.37790900468826294 lr:1.5418670289354925e-05 tokens_per_second_per_gpu:4064.1228820740303 mfu:19.171161045609146 +Step 27000 | loss:0.39115890860557556 lr:1.541553856046024e-05 tokens_per_second_per_gpu:5323.725652827059 mfu:25.11292222564473 +Step 27001 | loss:0.4692162573337555 lr:1.54124070916797e-05 tokens_per_second_per_gpu:4414.870756402462 mfu:20.82569861257461 +Step 27002 | loss:0.30563002824783325 lr:1.5409275883036845e-05 tokens_per_second_per_gpu:4191.474792283232 mfu:19.771901734542404 +Step 27003 | loss:0.4226725697517395 lr:1.5406144934555237e-05 tokens_per_second_per_gpu:4635.593384616455 mfu:21.866884909023682 +Step 27004 | loss:0.5336478352546692 lr:1.5403014246258448e-05 tokens_per_second_per_gpu:3858.3340678603254 mfu:18.200420097770845 +Step 27005 | loss:0.4218084216117859 lr:1.539988381816998e-05 tokens_per_second_per_gpu:3430.7403465393895 mfu:16.18338755928825 +Step 27006 | loss:0.4372105002403259 lr:1.5396753650313407e-05 tokens_per_second_per_gpu:3940.7459936841165 mfu:18.58917121280613 +Step 27007 | loss:0.4586420953273773 lr:1.5393623742712275e-05 tokens_per_second_per_gpu:4962.969275562465 mfu:23.411172842702697 +Step 27008 | loss:0.2728780508041382 lr:1.5390494095390124e-05 tokens_per_second_per_gpu:5333.515643255599 mfu:25.15910328084076 +Step 27009 | loss:0.39660394191741943 lr:1.538736470837046e-05 tokens_per_second_per_gpu:4366.771468510659 mfu:20.598806065004226 +Step 27010 | loss:0.3573410212993622 lr:1.5384235581676844e-05 tokens_per_second_per_gpu:5200.497280872133 mfu:24.531632970205067 +Step 27011 | loss:0.49352768063545227 lr:1.5381106715332826e-05 tokens_per_second_per_gpu:5320.031664769397 mfu:25.095497053717388 +Step 27012 | loss:0.40372011065483093 lr:1.537797810936191e-05 tokens_per_second_per_gpu:5212.413155102324 mfu:24.58784218200666 +Step 27013 | loss:0.3839874863624573 lr:1.537484976378763e-05 tokens_per_second_per_gpu:5346.536032597312 mfu:25.22052267887312 +Step 27014 | loss:0.5543724894523621 lr:1.5371721678633542e-05 tokens_per_second_per_gpu:4507.314079996632 mfu:21.26176954242572 +Step 27015 | loss:0.44207844138145447 lr:1.5368593853923152e-05 tokens_per_second_per_gpu:3993.8303222549475 mfu:18.839579047795134 +Step 27016 | loss:0.43856245279312134 lr:1.5365466289679977e-05 tokens_per_second_per_gpu:3727.693810536482 mfu:17.58416771444795 +Step 27017 | loss:0.4304201900959015 lr:1.5362338985927544e-05 tokens_per_second_per_gpu:3428.24191379797 mfu:16.171602025770394 +Step 27018 | loss:0.5609493255615234 lr:1.53592119426894e-05 tokens_per_second_per_gpu:4102.644497873504 mfu:19.35287407980065 +Step 27019 | loss:0.4715500771999359 lr:1.535608515998902e-05 tokens_per_second_per_gpu:4445.055791470133 mfu:20.96808657308686 +Step 27020 | loss:0.44088637828826904 lr:1.535295863784995e-05 tokens_per_second_per_gpu:5347.068245040975 mfu:25.22303321577468 +Step 27021 | loss:0.4840869903564453 lr:1.5349832376295716e-05 tokens_per_second_per_gpu:5327.453744944402 mfu:25.130508272240345 +Step 27022 | loss:0.5117260813713074 lr:1.5346706375349813e-05 tokens_per_second_per_gpu:4561.919956103679 mfu:21.519354776745836 +Step 27023 | loss:0.35903051495552063 lr:1.534358063503573e-05 tokens_per_second_per_gpu:5318.190785257395 mfu:25.08681331097284 +Step 27024 | loss:0.5075541138648987 lr:1.534045515537701e-05 tokens_per_second_per_gpu:3795.8852233113166 mfu:17.905838243161593 +Step 27025 | loss:0.3191033601760864 lr:1.5337329936397162e-05 tokens_per_second_per_gpu:4356.224683072823 mfu:20.549055078626804 +Step 27026 | loss:0.25980663299560547 lr:1.533420497811966e-05 tokens_per_second_per_gpu:4641.199098295082 mfu:21.893328016879142 +Step 27027 | loss:0.41507476568222046 lr:1.533108028056802e-05 tokens_per_second_per_gpu:4649.661699876308 mfu:21.933247552406502 +Step 27028 | loss:0.4181090295314789 lr:1.532795584376576e-05 tokens_per_second_per_gpu:3781.6906010548178 mfu:17.838879788125396 +Step 27029 | loss:0.4241465628147125 lr:1.5324831667736395e-05 tokens_per_second_per_gpu:5352.372748305934 mfu:25.248055462717076 +Step 27030 | loss:0.43182674050331116 lr:1.5321707752503367e-05 tokens_per_second_per_gpu:4756.892265034526 mfu:22.439072423678795 +Step 27031 | loss:0.3946975767612457 lr:1.5318584098090195e-05 tokens_per_second_per_gpu:5014.743116369889 mfu:23.65539888331945 +Step 27032 | loss:0.45175155997276306 lr:1.531546070452038e-05 tokens_per_second_per_gpu:5360.203433442303 mfu:25.28499414055798 +Step 27033 | loss:0.3790507912635803 lr:1.5312337571817425e-05 tokens_per_second_per_gpu:4267.519249442602 mfu:20.13061641348543 +Step 27034 | loss:0.4046972095966339 lr:1.5309214700004793e-05 tokens_per_second_per_gpu:4264.184081703325 mfu:20.11488385822097 +Step 27035 | loss:0.3202524781227112 lr:1.5306092089105983e-05 tokens_per_second_per_gpu:4853.5337048493875 mfu:22.894946584015393 +Step 27036 | loss:0.4528869688510895 lr:1.5302969739144495e-05 tokens_per_second_per_gpu:3820.6755674351507 mfu:18.022778526062577 +Step 27037 | loss:0.3572582006454468 lr:1.5299847650143802e-05 tokens_per_second_per_gpu:4644.676007787405 mfu:21.90972919217636 +Step 27038 | loss:0.3780685365200043 lr:1.529672582212736e-05 tokens_per_second_per_gpu:5337.852577296649 mfu:25.179561338669643 +Step 27039 | loss:0.40237662196159363 lr:1.5293604255118683e-05 tokens_per_second_per_gpu:5351.194847717974 mfu:25.242499104673104 +Step 27040 | loss:0.35720759630203247 lr:1.5290482949141244e-05 tokens_per_second_per_gpu:5212.253694517918 mfu:24.58708997922324 +Step 27041 | loss:0.41900530457496643 lr:1.5287361904218495e-05 tokens_per_second_per_gpu:3766.790773450347 mfu:17.768594759142378 +Step 27042 | loss:0.48158642649650574 lr:1.5284241120373933e-05 tokens_per_second_per_gpu:5309.616055333601 mfu:25.04636484692252 +Step 27043 | loss:0.40899744629859924 lr:1.528112059763103e-05 tokens_per_second_per_gpu:4730.34905407455 mfu:22.313863568841807 +Step 27044 | loss:0.46163052320480347 lr:1.527800033601326e-05 tokens_per_second_per_gpu:3698.7705494746106 mfu:17.447731757201904 +Step 27045 | loss:0.3470698595046997 lr:1.5274880335544056e-05 tokens_per_second_per_gpu:5362.477920002358 mfu:25.295723281729185 +Step 27046 | loss:0.38986340165138245 lr:1.5271760596246907e-05 tokens_per_second_per_gpu:5213.923111774338 mfu:24.594964905253732 +Step 27047 | loss:0.39563658833503723 lr:1.52686411181453e-05 tokens_per_second_per_gpu:4209.404429962935 mfu:19.856478894589195 +Step 27048 | loss:0.4036150872707367 lr:1.526552190126266e-05 tokens_per_second_per_gpu:5365.815409830743 mfu:25.31146679068416 +Step 27049 | loss:0.42514151334762573 lr:1.5262402945622456e-05 tokens_per_second_per_gpu:4534.366373096428 mfu:21.389379824574608 +Step 27050 | loss:0.31418678164482117 lr:1.5259284251248167e-05 tokens_per_second_per_gpu:3623.2986878866022 mfu:17.091718109263866 +Step 27051 | loss:0.41468575596809387 lr:1.5256165818163243e-05 tokens_per_second_per_gpu:4888.868365493455 mfu:23.061626206988226 +Step 27052 | loss:0.31764671206474304 lr:1.5253047646391106e-05 tokens_per_second_per_gpu:4033.4445663758484 mfu:19.026446196200425 +Step 27053 | loss:0.5139453411102295 lr:1.5249929735955232e-05 tokens_per_second_per_gpu:5346.940957740708 mfu:25.22243277986187 +Step 27054 | loss:0.38496166467666626 lr:1.5246812086879086e-05 tokens_per_second_per_gpu:4594.31057902693 mfu:21.672146871484948 +Step 27055 | loss:0.4292154014110565 lr:1.524369469918609e-05 tokens_per_second_per_gpu:4745.133538373641 mfu:22.383604503773757 +Step 27056 | loss:0.417005330324173 lr:1.5240577572899695e-05 tokens_per_second_per_gpu:4644.848884787884 mfu:21.91054468248348 +Step 27057 | loss:0.4545404613018036 lr:1.5237460708043372e-05 tokens_per_second_per_gpu:5354.455661321105 mfu:25.257880918791244 +Step 27058 | loss:0.2985881268978119 lr:1.523434410464054e-05 tokens_per_second_per_gpu:5349.214062147877 mfu:25.23315540118209 +Step 27059 | loss:0.4071090519428253 lr:1.5231227762714629e-05 tokens_per_second_per_gpu:5367.009926098733 mfu:25.31710152772576 +Step 27060 | loss:0.3841075599193573 lr:1.5228111682289087e-05 tokens_per_second_per_gpu:3793.0477072619265 mfu:17.892453195826462 +Step 27061 | loss:0.4792940616607666 lr:1.522499586338737e-05 tokens_per_second_per_gpu:4877.902016508096 mfu:23.009896067772214 +Step 27062 | loss:0.48592743277549744 lr:1.5221880306032881e-05 tokens_per_second_per_gpu:4267.252135816078 mfu:20.129356392934834 +Step 27063 | loss:0.4841049313545227 lr:1.521876501024907e-05 tokens_per_second_per_gpu:5341.503474176164 mfu:25.196783242157554 +Step 27064 | loss:0.4494999051094055 lr:1.5215649976059365e-05 tokens_per_second_per_gpu:3805.230645513654 mfu:17.949922194183465 +Step 27065 | loss:0.4243375062942505 lr:1.5212535203487227e-05 tokens_per_second_per_gpu:4894.663357853044 mfu:23.088962174676055 +Step 27066 | loss:0.390400767326355 lr:1.5209420692556015e-05 tokens_per_second_per_gpu:5344.937380508553 mfu:25.21298156421227 +Step 27067 | loss:0.4582890570163727 lr:1.5206306443289192e-05 tokens_per_second_per_gpu:4691.934891028646 mfu:22.132657407622226 +Step 27068 | loss:0.513976514339447 lr:1.520319245571018e-05 tokens_per_second_per_gpu:4121.8970322157875 mfu:19.443691569114282 +Step 27069 | loss:0.508964478969574 lr:1.520007872984241e-05 tokens_per_second_per_gpu:5332.479266762222 mfu:25.15421451609744 +Step 27070 | loss:0.4343738257884979 lr:1.5196965265709273e-05 tokens_per_second_per_gpu:3457.3049064670417 mfu:16.308697120848272 +Step 27071 | loss:0.4319438934326172 lr:1.5193852063334203e-05 tokens_per_second_per_gpu:5373.80059270732 mfu:25.34913426072566 +Step 27072 | loss:0.45689550042152405 lr:1.5190739122740627e-05 tokens_per_second_per_gpu:4161.17882281678 mfu:19.62899047754374 +Step 27073 | loss:0.44926396012306213 lr:1.5187626443951946e-05 tokens_per_second_per_gpu:4710.239323375353 mfu:22.21900253805984 +Step 27074 | loss:0.32564908266067505 lr:1.5184514026991554e-05 tokens_per_second_per_gpu:4524.350555694932 mfu:21.342133505017088 +Step 27075 | loss:0.37633663415908813 lr:1.5181401871882872e-05 tokens_per_second_per_gpu:4320.089107924272 mfu:20.378597405286218 +Step 27076 | loss:0.2597382366657257 lr:1.5178289978649324e-05 tokens_per_second_per_gpu:5290.657490294659 mfu:24.956934060969083 +Step 27077 | loss:0.42034903168678284 lr:1.5175178347314284e-05 tokens_per_second_per_gpu:4868.481452768448 mfu:22.965457661298384 +Step 27078 | loss:0.3470565974712372 lr:1.5172066977901173e-05 tokens_per_second_per_gpu:4023.406774379828 mfu:18.97909621873128 +Step 27079 | loss:0.356894314289093 lr:1.516895587043341e-05 tokens_per_second_per_gpu:5320.872376139514 mfu:25.099462832690378 +Step 27080 | loss:0.3529343903064728 lr:1.5165845024934366e-05 tokens_per_second_per_gpu:5337.103353580208 mfu:25.176027122568446 +Step 27081 | loss:0.5272745490074158 lr:1.516273444142744e-05 tokens_per_second_per_gpu:5321.4556281719015 mfu:25.10221413204021 +Step 27082 | loss:0.4392912983894348 lr:1.5159624119936029e-05 tokens_per_second_per_gpu:5362.756600919155 mfu:25.297037867161585 +Step 27083 | loss:0.3614460825920105 lr:1.5156514060483546e-05 tokens_per_second_per_gpu:3489.7375171287003 mfu:16.4616872789133 +Step 27084 | loss:0.42799893021583557 lr:1.5153404263093352e-05 tokens_per_second_per_gpu:4090.363410951447 mfu:19.29494209741981 +Step 27085 | loss:0.4848323166370392 lr:1.515029472778885e-05 tokens_per_second_per_gpu:4432.989354598417 mfu:20.911167131616555 +Step 27086 | loss:0.4965309798717499 lr:1.5147185454593444e-05 tokens_per_second_per_gpu:4404.963584181839 mfu:20.77896479087205 +Step 27087 | loss:0.3729843199253082 lr:1.5144076443530502e-05 tokens_per_second_per_gpu:4341.699739628848 mfu:20.480538442190678 +Step 27088 | loss:0.46376094222068787 lr:1.5140967694623393e-05 tokens_per_second_per_gpu:5337.532585087077 mfu:25.17805188082079 +Step 27089 | loss:0.35028380155563354 lr:1.5137859207895517e-05 tokens_per_second_per_gpu:4872.329510944689 mfu:22.983609608262135 +Step 27090 | loss:0.45055076479911804 lr:1.5134750983370265e-05 tokens_per_second_per_gpu:5345.437369835233 mfu:25.215340099173194 +Step 27091 | loss:0.40431714057922363 lr:1.5131643021070985e-05 tokens_per_second_per_gpu:4954.583440216986 mfu:23.37161542660703 +Step 27092 | loss:0.43057510256767273 lr:1.5128535321021064e-05 tokens_per_second_per_gpu:5365.690499899237 mfu:25.31087756922541 +Step 27093 | loss:0.4702718257904053 lr:1.5125427883243899e-05 tokens_per_second_per_gpu:4179.278119992405 mfu:19.714367950379692 +Step 27094 | loss:0.3778507113456726 lr:1.5122320707762832e-05 tokens_per_second_per_gpu:3642.6135122142173 mfu:17.18282943106607 +Step 27095 | loss:0.4530903697013855 lr:1.5119213794601233e-05 tokens_per_second_per_gpu:3790.0038274101166 mfu:17.878094695226018 +Step 27096 | loss:0.36780041456222534 lr:1.5116107143782476e-05 tokens_per_second_per_gpu:5331.315192588182 mfu:25.148723379606956 +Step 27097 | loss:0.5464913845062256 lr:1.5113000755329948e-05 tokens_per_second_per_gpu:4409.009149594761 mfu:20.798048413169436 +Step 27098 | loss:0.4223070740699768 lr:1.5109894629266974e-05 tokens_per_second_per_gpu:5349.72075896851 mfu:25.23554557653641 +Step 27099 | loss:0.3424821197986603 lr:1.5106788765616931e-05 tokens_per_second_per_gpu:5233.751534630438 mfu:24.68849895894374 +Step 27100 | loss:0.35672685503959656 lr:1.5103683164403187e-05 tokens_per_second_per_gpu:5361.46288725821 mfu:25.290935199092594 +Step 27101 | loss:0.39849844574928284 lr:1.5100577825649126e-05 tokens_per_second_per_gpu:4788.927377432049 mfu:22.590187514611696 +Step 27102 | loss:0.5253407955169678 lr:1.5097472749378038e-05 tokens_per_second_per_gpu:4321.123916978359 mfu:20.383478776150977 +Step 27103 | loss:0.39928364753723145 lr:1.5094367935613312e-05 tokens_per_second_per_gpu:3484.5707105164993 mfu:16.437314570575435 +Step 27104 | loss:0.4037168025970459 lr:1.5091263384378301e-05 tokens_per_second_per_gpu:5389.742786959567 mfu:25.424336311032327 +Step 27105 | loss:0.3727513551712036 lr:1.5088159095696363e-05 tokens_per_second_per_gpu:4174.601361941603 mfu:19.692306884716857 +Step 27106 | loss:0.34930551052093506 lr:1.5085055069590825e-05 tokens_per_second_per_gpu:5045.65162231602 mfu:23.801199579402248 +Step 27107 | loss:0.5079741477966309 lr:1.508195130608504e-05 tokens_per_second_per_gpu:4409.072819808748 mfu:20.798348756432073 +Step 27108 | loss:0.49411964416503906 lr:1.5078847805202373e-05 tokens_per_second_per_gpu:3793.3109695120797 mfu:17.893695048777605 +Step 27109 | loss:0.39044737815856934 lr:1.5075744566966144e-05 tokens_per_second_per_gpu:5311.331269279001 mfu:25.05445580374865 +Step 27110 | loss:0.37259072065353394 lr:1.5072641591399684e-05 tokens_per_second_per_gpu:3886.149136485833 mfu:18.331628522217436 +Step 27111 | loss:0.38864850997924805 lr:1.506953887852634e-05 tokens_per_second_per_gpu:5352.203721131823 mfu:25.247258132697784 +Step 27112 | loss:0.3423062264919281 lr:1.5066436428369468e-05 tokens_per_second_per_gpu:5316.118538048317 mfu:25.077038167325785 +Step 27113 | loss:0.29108044505119324 lr:1.506333424095237e-05 tokens_per_second_per_gpu:3394.3790542482075 mfu:16.011865139674555 +Step 27114 | loss:0.46496766805648804 lr:1.5060232316298395e-05 tokens_per_second_per_gpu:5286.667338249577 mfu:24.938111832982297 +Step 27115 | loss:0.4112425446510315 lr:1.5057130654430885e-05 tokens_per_second_per_gpu:5273.430744289709 mfu:24.87567256087779 +Step 27116 | loss:0.42334452271461487 lr:1.5054029255373159e-05 tokens_per_second_per_gpu:3695.1680443065266 mfu:17.43073812567329 +Step 27117 | loss:0.36172768473625183 lr:1.5050928119148516e-05 tokens_per_second_per_gpu:3340.1067921818294 mfu:15.7558536197047 +Step 27118 | loss:0.4914151728153229 lr:1.5047827245780305e-05 tokens_per_second_per_gpu:5215.514142171076 mfu:24.602470067092916 +Step 27119 | loss:0.4232877194881439 lr:1.504472663529186e-05 tokens_per_second_per_gpu:4491.643180901791 mfu:21.18784723766427 +Step 27120 | loss:0.3900747001171112 lr:1.5041626287706468e-05 tokens_per_second_per_gpu:4241.8136244374955 mfu:20.009358594504466 +Step 27121 | loss:0.3944598436355591 lr:1.503852620304747e-05 tokens_per_second_per_gpu:3670.105986548241 mfu:17.312516123199476 +Step 27122 | loss:0.3706701695919037 lr:1.5035426381338191e-05 tokens_per_second_per_gpu:5104.492476988187 mfu:24.07876193017589 +Step 27123 | loss:0.46088796854019165 lr:1.5032326822601933e-05 tokens_per_second_per_gpu:4044.23578828393 mfu:19.077350231113158 +Step 27124 | loss:0.3665432631969452 lr:1.5029227526861988e-05 tokens_per_second_per_gpu:5361.162261556308 mfu:25.28951709636469 +Step 27125 | loss:0.5000112652778625 lr:1.502612849414169e-05 tokens_per_second_per_gpu:5155.987663188647 mfu:24.321673509468205 +Step 27126 | loss:0.4906274974346161 lr:1.5023029724464355e-05 tokens_per_second_per_gpu:4474.775839923529 mfu:21.108281112404676 +Step 27127 | loss:0.38010475039482117 lr:1.5019931217853262e-05 tokens_per_second_per_gpu:5286.292180046244 mfu:24.93634214772858 +Step 27128 | loss:0.48092710971832275 lr:1.5016832974331724e-05 tokens_per_second_per_gpu:4419.926110276296 mfu:20.849545579329796 +Step 27129 | loss:0.571071207523346 lr:1.5013734993923073e-05 tokens_per_second_per_gpu:4659.775567034262 mfu:21.980956389394244 +Step 27130 | loss:0.34464719891548157 lr:1.501063727665058e-05 tokens_per_second_per_gpu:4104.955218250246 mfu:19.363774142067292 +Step 27131 | loss:0.45348718762397766 lr:1.5007539822537537e-05 tokens_per_second_per_gpu:5402.699273892758 mfu:25.485454270500814 +Step 27132 | loss:0.49621281027793884 lr:1.5004442631607246e-05 tokens_per_second_per_gpu:5253.107677849686 mfu:24.779805189007515 +Step 27133 | loss:0.3424711525440216 lr:1.500134570388303e-05 tokens_per_second_per_gpu:5311.427100319921 mfu:25.054907855119126 +Step 27134 | loss:0.28912997245788574 lr:1.499824903938814e-05 tokens_per_second_per_gpu:4095.6055340072103 mfu:19.319670086272268 +Step 27135 | loss:0.49161213636398315 lr:1.4995152638145887e-05 tokens_per_second_per_gpu:3030.3547070672307 mfu:14.294700185063785 +Step 27136 | loss:0.34547504782676697 lr:1.4992056500179552e-05 tokens_per_second_per_gpu:4709.887416867452 mfu:22.217342535020627 +Step 27137 | loss:0.4394061863422394 lr:1.4988960625512467e-05 tokens_per_second_per_gpu:5264.87831989737 mfu:24.835329315825057 +Step 27138 | loss:0.4255487620830536 lr:1.4985865014167838e-05 tokens_per_second_per_gpu:5303.836042536582 mfu:25.019099540387106 +Step 27139 | loss:0.456228643655777 lr:1.4982769666168989e-05 tokens_per_second_per_gpu:4937.221090703975 mfu:23.289714261631868 +Step 27140 | loss:0.45684802532196045 lr:1.49796745815392e-05 tokens_per_second_per_gpu:3796.435706936192 mfu:17.908434968342323 +Step 27141 | loss:0.35347801446914673 lr:1.4976579760301756e-05 tokens_per_second_per_gpu:4366.190628810495 mfu:20.59606614503708 +Step 27142 | loss:0.5035057067871094 lr:1.4973485202479915e-05 tokens_per_second_per_gpu:5303.586790843135 mfu:25.01792377762239 +Step 27143 | loss:0.43051573634147644 lr:1.497039090809696e-05 tokens_per_second_per_gpu:3843.23302620418 mfu:18.12918590777542 +Step 27144 | loss:0.3838266134262085 lr:1.4967296877176196e-05 tokens_per_second_per_gpu:4989.602148215787 mfu:23.53680464704506 +Step 27145 | loss:0.424449622631073 lr:1.4964203109740833e-05 tokens_per_second_per_gpu:4330.507602735194 mfu:20.427743222887845 +Step 27146 | loss:0.34398597478866577 lr:1.4961109605814167e-05 tokens_per_second_per_gpu:4070.8822856664015 mfu:19.203046305627534 +Step 27147 | loss:0.34389710426330566 lr:1.4958016365419463e-05 tokens_per_second_per_gpu:4355.380439028168 mfu:20.545072635425278 +Step 27148 | loss:0.49377623200416565 lr:1.4954923388580011e-05 tokens_per_second_per_gpu:3807.079964876225 mfu:17.958645749143134 +Step 27149 | loss:0.3986654579639435 lr:1.4951830675319033e-05 tokens_per_second_per_gpu:4616.461210405154 mfu:21.77663517898345 +Step 27150 | loss:0.3848835527896881 lr:1.4948738225659808e-05 tokens_per_second_per_gpu:4531.558375097895 mfu:21.376134018921665 +Step 27151 | loss:0.34959617257118225 lr:1.494564603962561e-05 tokens_per_second_per_gpu:4206.8597964430755 mfu:19.84447542411677 +Step 27152 | loss:0.29865413904190063 lr:1.4942554117239688e-05 tokens_per_second_per_gpu:3207.652502648638 mfu:15.131044136944503 +Step 27153 | loss:0.4317520260810852 lr:1.4939462458525266e-05 tokens_per_second_per_gpu:4117.170791676115 mfu:19.421397086108858 +Step 27154 | loss:0.34272530674934387 lr:1.4936371063505628e-05 tokens_per_second_per_gpu:5032.132073636903 mfu:23.73742556161 +Step 27155 | loss:0.3993070721626282 lr:1.4933279932204025e-05 tokens_per_second_per_gpu:4887.704704201205 mfu:23.05613701813159 +Step 27156 | loss:0.4553171396255493 lr:1.493018906464369e-05 tokens_per_second_per_gpu:5328.677238728249 mfu:25.136279701168316 +Step 27157 | loss:0.32932448387145996 lr:1.4927098460847871e-05 tokens_per_second_per_gpu:4568.381921496011 mfu:21.54983697002645 +Step 27158 | loss:0.39434266090393066 lr:1.4924008120839845e-05 tokens_per_second_per_gpu:3967.280558300369 mfu:18.714339281364555 +Step 27159 | loss:0.47975262999534607 lr:1.4920918044642818e-05 tokens_per_second_per_gpu:4570.5926795700125 mfu:21.560265493055738 +Step 27160 | loss:0.4563126862049103 lr:1.4917828232280034e-05 tokens_per_second_per_gpu:3976.269032484173 mfu:18.756739447681387 +Step 27161 | loss:0.41779065132141113 lr:1.4914738683774743e-05 tokens_per_second_per_gpu:5360.923290186346 mfu:25.288389827639858 +Step 27162 | loss:0.3963417112827301 lr:1.491164939915019e-05 tokens_per_second_per_gpu:4312.862711816774 mfu:20.34450926189683 +Step 27163 | loss:0.3906524181365967 lr:1.4908560378429581e-05 tokens_per_second_per_gpu:4507.291774580817 mfu:21.261664323973626 +Step 27164 | loss:0.4639146029949188 lr:1.4905471621636174e-05 tokens_per_second_per_gpu:4649.1520685889745 mfu:21.930843534672086 +Step 27165 | loss:0.4043378233909607 lr:1.4902383128793201e-05 tokens_per_second_per_gpu:5349.011583993377 mfu:25.23220027718098 +Step 27166 | loss:0.33372196555137634 lr:1.4899294899923889e-05 tokens_per_second_per_gpu:4647.714243175825 mfu:21.924061066879947 +Step 27167 | loss:0.2971523404121399 lr:1.4896206935051444e-05 tokens_per_second_per_gpu:3866.7577474805903 mfu:18.24015603176758 +Step 27168 | loss:0.31750568747520447 lr:1.48931192341991e-05 tokens_per_second_per_gpu:4373.7408135133 mfu:20.631681654474555 +Step 27169 | loss:0.4724467694759369 lr:1.4890031797390108e-05 tokens_per_second_per_gpu:5334.931498113339 mfu:25.165782109774472 +Step 27170 | loss:0.3018723130226135 lr:1.4886944624647647e-05 tokens_per_second_per_gpu:3879.0377450568267 mfu:18.298082875518322 +Step 27171 | loss:0.39881837368011475 lr:1.4883857715994959e-05 tokens_per_second_per_gpu:4929.177844867158 mfu:23.25177289870035 +Step 27172 | loss:0.4082321226596832 lr:1.4880771071455251e-05 tokens_per_second_per_gpu:4832.99419703277 mfu:22.798058221243014 +Step 27173 | loss:0.39800095558166504 lr:1.4877684691051785e-05 tokens_per_second_per_gpu:4283.255016673054 mfu:20.204844712310212 +Step 27174 | loss:0.34044983983039856 lr:1.4874598574807696e-05 tokens_per_second_per_gpu:4788.084884334262 mfu:22.58621333092529 +Step 27175 | loss:0.42780211567878723 lr:1.4871512722746239e-05 tokens_per_second_per_gpu:5295.35007149401 mfu:24.979069767123338 +Step 27176 | loss:0.3767254650592804 lr:1.4868427134890627e-05 tokens_per_second_per_gpu:4538.371837509685 mfu:21.40827428361464 +Step 27177 | loss:0.4386622905731201 lr:1.4865341811264049e-05 tokens_per_second_per_gpu:4097.934972144188 mfu:19.33065844340752 +Step 27178 | loss:0.3997062146663666 lr:1.4862256751889714e-05 tokens_per_second_per_gpu:5078.370209735521 mfu:23.955538738626686 +Step 27179 | loss:0.43422695994377136 lr:1.485917195679083e-05 tokens_per_second_per_gpu:5294.302888964359 mfu:24.97413002846321 +Step 27180 | loss:0.411226749420166 lr:1.485608742599061e-05 tokens_per_second_per_gpu:3287.3538956026255 mfu:15.50700920596833 +Step 27181 | loss:0.4009664058685303 lr:1.4853003159512247e-05 tokens_per_second_per_gpu:5191.628506032773 mfu:24.489797446118782 +Step 27182 | loss:0.3963443338871002 lr:1.4849919157378917e-05 tokens_per_second_per_gpu:4792.323861264272 mfu:22.60620930834819 +Step 27183 | loss:0.3528728485107422 lr:1.4846835419613824e-05 tokens_per_second_per_gpu:4521.335654802882 mfu:21.327911703114157 +Step 27184 | loss:0.3701155185699463 lr:1.4843751946240186e-05 tokens_per_second_per_gpu:5349.61767030652 mfu:25.235059289729065 +Step 27185 | loss:0.4207228422164917 lr:1.4840668737281161e-05 tokens_per_second_per_gpu:5338.02590664287 mfu:25.1803789627688 +Step 27186 | loss:0.3789394497871399 lr:1.483758579275995e-05 tokens_per_second_per_gpu:5312.181845729425 mfu:25.058468117988603 +Step 27187 | loss:0.4532841145992279 lr:1.483450311269976e-05 tokens_per_second_per_gpu:4988.538710701657 mfu:23.53178823084976 +Step 27188 | loss:0.49804890155792236 lr:1.483142069712376e-05 tokens_per_second_per_gpu:4006.91962577815 mfu:18.901323525779546 +Step 27189 | loss:0.4114462435245514 lr:1.4828338546055115e-05 tokens_per_second_per_gpu:4830.748825390391 mfu:22.787466420106007 +Step 27190 | loss:0.4222915470600128 lr:1.4825256659517024e-05 tokens_per_second_per_gpu:3873.466904313975 mfu:18.271804269250847 +Step 27191 | loss:0.4038195013999939 lr:1.4822175037532682e-05 tokens_per_second_per_gpu:4793.935576449944 mfu:22.613812043865973 +Step 27192 | loss:0.41909560561180115 lr:1.481909368012524e-05 tokens_per_second_per_gpu:3973.434553699087 mfu:18.74336872764019 +Step 27193 | loss:0.37522101402282715 lr:1.4816012587317873e-05 tokens_per_second_per_gpu:4330.078317512746 mfu:20.42571821124995 +Step 27194 | loss:0.4263947606086731 lr:1.4812931759133792e-05 tokens_per_second_per_gpu:5051.9207795447 mfu:23.830772263678455 +Step 27195 | loss:0.44828876852989197 lr:1.4809851195596131e-05 tokens_per_second_per_gpu:4685.713894511039 mfu:22.103311905637216 +Step 27196 | loss:0.33929628133773804 lr:1.4806770896728062e-05 tokens_per_second_per_gpu:4529.649018286865 mfu:21.367127257074173 +Step 27197 | loss:0.3649389147758484 lr:1.4803690862552755e-05 tokens_per_second_per_gpu:4467.089754488904 mfu:21.07202453602815 +Step 27198 | loss:0.41470062732696533 lr:1.4800611093093397e-05 tokens_per_second_per_gpu:4087.8568591495937 mfu:19.28311826490863 +Step 27199 | loss:0.339262455701828 lr:1.4797531588373115e-05 tokens_per_second_per_gpu:5346.1925129609635 mfu:25.218902238137968 +Step 27200 | loss:0.4315997362136841 lr:1.4794452348415094e-05 tokens_per_second_per_gpu:5358.329578722079 mfu:25.276154848130098 +Step 27201 | loss:0.37601515650749207 lr:1.4791373373242507e-05 tokens_per_second_per_gpu:5325.2407712411 mfu:25.120069297709623 +Step 27202 | loss:0.3518666923046112 lr:1.4788294662878488e-05 tokens_per_second_per_gpu:5222.1100980347865 mfu:24.633584316288157 +Step 27203 | loss:0.4682740271091461 lr:1.4785216217346187e-05 tokens_per_second_per_gpu:5179.15157892971 mfu:24.430941652190363 +Step 27204 | loss:0.4210273027420044 lr:1.4782138036668764e-05 tokens_per_second_per_gpu:4693.400846021242 mfu:22.139572567440013 +Step 27205 | loss:0.34522828459739685 lr:1.4779060120869392e-05 tokens_per_second_per_gpu:5345.314702918475 mfu:25.21476145839771 +Step 27206 | loss:0.4572756290435791 lr:1.4775982469971189e-05 tokens_per_second_per_gpu:4264.647890641633 mfu:20.117071724116933 +Step 27207 | loss:0.4534463882446289 lr:1.4772905083997319e-05 tokens_per_second_per_gpu:4544.511931164024 mfu:21.437238152990158 +Step 27208 | loss:0.41207167506217957 lr:1.4769827962970917e-05 tokens_per_second_per_gpu:4644.1766807302565 mfu:21.907373781253806 +Step 27209 | loss:0.2842084765434265 lr:1.4766751106915172e-05 tokens_per_second_per_gpu:4444.198283828717 mfu:20.964041563236407 +Step 27210 | loss:0.3421842157840729 lr:1.4763674515853154e-05 tokens_per_second_per_gpu:5243.236305293887 mfu:24.73324024043198 +Step 27211 | loss:0.38465234637260437 lr:1.4760598189808034e-05 tokens_per_second_per_gpu:3799.9691254145678 mfu:17.92510270616818 +Step 27212 | loss:0.4056653082370758 lr:1.475752212880297e-05 tokens_per_second_per_gpu:5348.4342402914 mfu:25.22947684843376 +Step 27213 | loss:0.41843271255493164 lr:1.4754446332861066e-05 tokens_per_second_per_gpu:4752.150519222991 mfu:22.416704799660522 +Step 27214 | loss:0.492133766412735 lr:1.4751370802005466e-05 tokens_per_second_per_gpu:5295.5892516577305 mfu:24.98019802076395 +Step 27215 | loss:0.5037412643432617 lr:1.4748295536259299e-05 tokens_per_second_per_gpu:4747.910680231791 mfu:22.396704755747912 +Step 27216 | loss:0.4215982258319855 lr:1.4745220535645722e-05 tokens_per_second_per_gpu:5336.5770032438895 mfu:25.1735442382274 +Step 27217 | loss:0.31931039690971375 lr:1.4742145800187835e-05 tokens_per_second_per_gpu:4986.828077971655 mfu:23.523718884398303 +Step 27218 | loss:0.4071645140647888 lr:1.4739071329908755e-05 tokens_per_second_per_gpu:5347.659103099736 mfu:25.22582039404864 +Step 27219 | loss:0.36486873030662537 lr:1.4735997124831619e-05 tokens_per_second_per_gpu:3040.2831841916486 mfu:14.34153450563254 +Step 27220 | loss:0.42805948853492737 lr:1.4732923184979563e-05 tokens_per_second_per_gpu:4815.416163154897 mfu:22.715139636319243 +Step 27221 | loss:0.3960811495780945 lr:1.4729849510375682e-05 tokens_per_second_per_gpu:5364.179057035386 mfu:25.303747835357886 +Step 27222 | loss:0.3260093629360199 lr:1.4726776101043094e-05 tokens_per_second_per_gpu:4864.762920149447 mfu:22.947916708487465 +Step 27223 | loss:0.44920477271080017 lr:1.4723702957004947e-05 tokens_per_second_per_gpu:4454.404582930373 mfu:21.012186417473114 +Step 27224 | loss:0.3876911997795105 lr:1.4720630078284325e-05 tokens_per_second_per_gpu:5102.4517965546165 mfu:24.06913569239475 +Step 27225 | loss:0.5591915845870972 lr:1.4717557464904335e-05 tokens_per_second_per_gpu:5346.146958784734 mfu:25.21868735131635 +Step 27226 | loss:0.49093061685562134 lr:1.4714485116888094e-05 tokens_per_second_per_gpu:4219.401680995167 mfu:19.903637633415148 +Step 27227 | loss:0.4825400710105896 lr:1.471141303425873e-05 tokens_per_second_per_gpu:5059.531024531894 mfu:23.866671087724736 +Step 27228 | loss:0.4645644426345825 lr:1.4708341217039318e-05 tokens_per_second_per_gpu:4666.185553127821 mfu:22.01119338745396 +Step 27229 | loss:0.45831209421157837 lr:1.4705269665252975e-05 tokens_per_second_per_gpu:3894.493824805768 mfu:18.370991840773208 +Step 27230 | loss:0.35847607254981995 lr:1.4702198378922816e-05 tokens_per_second_per_gpu:4223.979137809794 mfu:19.92523027820441 +Step 27231 | loss:0.3824734389781952 lr:1.4699127358071923e-05 tokens_per_second_per_gpu:5326.627112618246 mfu:25.126608906519852 +Step 27232 | loss:0.44335830211639404 lr:1.469605660272339e-05 tokens_per_second_per_gpu:5010.734674697909 mfu:23.636490380042048 +Step 27233 | loss:0.3719567358493805 lr:1.4692986112900315e-05 tokens_per_second_per_gpu:4490.28978555761 mfu:21.181463040912707 +Step 27234 | loss:0.4468763768672943 lr:1.4689915888625817e-05 tokens_per_second_per_gpu:3818.497837350909 mfu:18.012505801696143 +Step 27235 | loss:0.4444858431816101 lr:1.4686845929922944e-05 tokens_per_second_per_gpu:5345.225129172356 mfu:25.21433892375441 +Step 27236 | loss:0.35027822852134705 lr:1.468377623681481e-05 tokens_per_second_per_gpu:4667.683042742193 mfu:22.018257301463034 +Step 27237 | loss:0.45715925097465515 lr:1.4680706809324524e-05 tokens_per_second_per_gpu:3509.459076056215 mfu:16.55471723721931 +Step 27238 | loss:0.3801293671131134 lr:1.4677637647475145e-05 tokens_per_second_per_gpu:5332.736083438787 mfu:25.155425964178242 +Step 27239 | loss:0.4471542537212372 lr:1.4674568751289741e-05 tokens_per_second_per_gpu:4372.776303091344 mfu:20.62713189425174 +Step 27240 | loss:0.4625927209854126 lr:1.467150012079142e-05 tokens_per_second_per_gpu:3915.499230037732 mfu:18.470077921144078 +Step 27241 | loss:0.4930005967617035 lr:1.4668431756003265e-05 tokens_per_second_per_gpu:4666.637737474492 mfu:22.013326418167814 +Step 27242 | loss:0.49337247014045715 lr:1.4665363656948328e-05 tokens_per_second_per_gpu:5329.729967150466 mfu:25.1412456007499 +Step 27243 | loss:0.33158671855926514 lr:1.4662295823649701e-05 tokens_per_second_per_gpu:5337.030594102701 mfu:25.175683903698996 +Step 27244 | loss:0.42856258153915405 lr:1.4659228256130475e-05 tokens_per_second_per_gpu:3770.7428950924646 mfu:17.78723759122973 +Step 27245 | loss:0.37154412269592285 lr:1.4656160954413683e-05 tokens_per_second_per_gpu:3988.053228511984 mfu:18.81232751093574 +Step 27246 | loss:0.3178471028804779 lr:1.4653093918522431e-05 tokens_per_second_per_gpu:5377.545887813361 mfu:25.366801456754096 +Step 27247 | loss:0.3048253357410431 lr:1.465002714847975e-05 tokens_per_second_per_gpu:3914.005162500206 mfu:18.463030149655356 +Step 27248 | loss:0.4509248733520508 lr:1.464696064430875e-05 tokens_per_second_per_gpu:4049.624682816125 mfu:19.10277056606174 +Step 27249 | loss:0.37240007519721985 lr:1.4643894406032449e-05 tokens_per_second_per_gpu:5364.295354928704 mfu:25.304296432363135 +Step 27250 | loss:0.3873971998691559 lr:1.464082843367393e-05 tokens_per_second_per_gpu:5017.792453309158 mfu:23.66978312593251 +Step 27251 | loss:0.49539342522621155 lr:1.463776272725625e-05 tokens_per_second_per_gpu:4985.636649185772 mfu:23.518098711536577 +Step 27252 | loss:0.4045746624469757 lr:1.463469728680249e-05 tokens_per_second_per_gpu:3949.446189408682 mfu:18.630211520445364 +Step 27253 | loss:0.47600284218788147 lr:1.4631632112335681e-05 tokens_per_second_per_gpu:3619.7194224658106 mfu:17.07483410358834 +Step 27254 | loss:0.48280301690101624 lr:1.4628567203878862e-05 tokens_per_second_per_gpu:3798.502686999469 mfu:17.918185265963963 +Step 27255 | loss:0.5372681617736816 lr:1.4625502561455106e-05 tokens_per_second_per_gpu:3457.972039290313 mfu:16.31184410020024 +Step 27256 | loss:0.4115409255027771 lr:1.4622438185087473e-05 tokens_per_second_per_gpu:5356.282511880284 mfu:25.266498484571738 +Step 27257 | loss:0.4762878715991974 lr:1.461937407479898e-05 tokens_per_second_per_gpu:5353.929527934439 mfu:25.255399057838726 +Step 27258 | loss:0.558985710144043 lr:1.4616310230612696e-05 tokens_per_second_per_gpu:5273.816442136361 mfu:24.877491963427286 +Step 27259 | loss:0.428134948015213 lr:1.4613246652551665e-05 tokens_per_second_per_gpu:4346.837108914221 mfu:20.504772289635365 +Step 27260 | loss:0.3258150517940521 lr:1.4610183340638928e-05 tokens_per_second_per_gpu:4794.851115876624 mfu:22.618130799548755 +Step 27261 | loss:0.3576970100402832 lr:1.4607120294897498e-05 tokens_per_second_per_gpu:4352.18312913382 mfu:20.52999037913632 +Step 27262 | loss:0.3405110836029053 lr:1.4604057515350433e-05 tokens_per_second_per_gpu:4986.742850591357 mfu:23.523316852304145 +Step 27263 | loss:0.41117891669273376 lr:1.4600995002020784e-05 tokens_per_second_per_gpu:4968.955915003027 mfu:23.439412842369848 +Step 27264 | loss:0.5230230093002319 lr:1.4597932754931549e-05 tokens_per_second_per_gpu:4788.870561249593 mfu:22.589919502985214 +Step 27265 | loss:0.589331328868866 lr:1.4594870774105785e-05 tokens_per_second_per_gpu:5011.5572975564055 mfu:23.640370832419507 +Step 27266 | loss:0.34960418939590454 lr:1.4591809059566531e-05 tokens_per_second_per_gpu:4539.472714775304 mfu:21.413467309505595 +Step 27267 | loss:0.4153546392917633 lr:1.4588747611336795e-05 tokens_per_second_per_gpu:5330.282453687419 mfu:25.14385177400863 +Step 27268 | loss:0.39581090211868286 lr:1.4585686429439594e-05 tokens_per_second_per_gpu:4691.885904160277 mfu:22.132426328206108 +Step 27269 | loss:0.419575572013855 lr:1.4582625513897962e-05 tokens_per_second_per_gpu:3966.7287627123314 mfu:18.711736367429744 +Step 27270 | loss:0.4621434807777405 lr:1.457956486473494e-05 tokens_per_second_per_gpu:4570.651236927293 mfu:21.560541718100946 +Step 27271 | loss:0.47183728218078613 lr:1.4576504481973513e-05 tokens_per_second_per_gpu:4940.5124056005425 mfu:23.305239955555205 +Step 27272 | loss:0.3675154745578766 lr:1.4573444365636719e-05 tokens_per_second_per_gpu:4484.317910873956 mfu:21.15329268912302 +Step 27273 | loss:0.43230336904525757 lr:1.4570384515747587e-05 tokens_per_second_per_gpu:4382.700099615155 mfu:20.673944135628837 +Step 27274 | loss:0.36638718843460083 lr:1.4567324932329097e-05 tokens_per_second_per_gpu:4437.978141299464 mfu:20.934700089660176 +Step 27275 | loss:0.40536683797836304 lr:1.4564265615404288e-05 tokens_per_second_per_gpu:5378.85797674201 mfu:25.372990804095984 +Step 27276 | loss:0.36697810888290405 lr:1.456120656499615e-05 tokens_per_second_per_gpu:3966.173308086708 mfu:18.709116193190347 +Step 27277 | loss:0.47228315472602844 lr:1.4558147781127713e-05 tokens_per_second_per_gpu:4936.514081924523 mfu:23.28637918059106 +Step 27278 | loss:0.3565022647380829 lr:1.4555089263821953e-05 tokens_per_second_per_gpu:4787.916469175322 mfu:22.585418887885982 +Step 27279 | loss:0.3910331428050995 lr:1.4552031013101886e-05 tokens_per_second_per_gpu:4608.048556841626 mfu:21.73695125677758 +Step 27280 | loss:0.353933721780777 lr:1.4548973028990536e-05 tokens_per_second_per_gpu:4191.763302962233 mfu:19.773262688640248 +Step 27281 | loss:0.353380411863327 lr:1.4545915311510866e-05 tokens_per_second_per_gpu:5298.541853568716 mfu:24.9941259477439 +Step 27282 | loss:0.4119158089160919 lr:1.454285786068591e-05 tokens_per_second_per_gpu:5317.847197880573 mfu:25.085192550694316 +Step 27283 | loss:0.4670681953430176 lr:1.4539800676538628e-05 tokens_per_second_per_gpu:4537.129791582287 mfu:21.402415341060216 +Step 27284 | loss:0.41626572608947754 lr:1.4536743759092042e-05 tokens_per_second_per_gpu:4174.1590306399885 mfu:19.690220332497642 +Step 27285 | loss:0.3111381232738495 lr:1.4533687108369121e-05 tokens_per_second_per_gpu:4602.354215110697 mfu:21.710090075277904 +Step 27286 | loss:0.3330167233943939 lr:1.4530630724392858e-05 tokens_per_second_per_gpu:5337.647448870315 mfu:25.17859371288411 +Step 27287 | loss:0.5050042271614075 lr:1.4527574607186251e-05 tokens_per_second_per_gpu:5155.324635625988 mfu:24.318545895330978 +Step 27288 | loss:0.46362268924713135 lr:1.4524518756772299e-05 tokens_per_second_per_gpu:3943.7031078553428 mfu:18.603120424887507 +Step 27289 | loss:0.4289878010749817 lr:1.4521463173173965e-05 tokens_per_second_per_gpu:3751.3665360067203 mfu:17.695835999474518 +Step 27290 | loss:0.3947901725769043 lr:1.4518407856414217e-05 tokens_per_second_per_gpu:5368.699604189666 mfu:25.3250720275695 +Step 27291 | loss:0.5301513075828552 lr:1.4515352806516058e-05 tokens_per_second_per_gpu:4538.902177888916 mfu:21.410775989668874 +Step 27292 | loss:0.39112216234207153 lr:1.4512298023502468e-05 tokens_per_second_per_gpu:4525.81762196162 mfu:21.349053906904366 +Step 27293 | loss:0.4058677554130554 lr:1.4509243507396397e-05 tokens_per_second_per_gpu:4559.903665214679 mfu:21.509843588607723 +Step 27294 | loss:0.37220636010169983 lr:1.4506189258220837e-05 tokens_per_second_per_gpu:3492.045542124588 mfu:16.472574626608573 +Step 27295 | loss:0.3800889253616333 lr:1.450313527599877e-05 tokens_per_second_per_gpu:5304.728296318697 mfu:25.023308453711536 +Step 27296 | loss:0.3377971947193146 lr:1.4500081560753153e-05 tokens_per_second_per_gpu:3018.374314163775 mfu:14.238186627672379 +Step 27297 | loss:0.3009241223335266 lr:1.4497028112506938e-05 tokens_per_second_per_gpu:4701.807394228041 mfu:22.179227689636505 +Step 27298 | loss:0.43656203150749207 lr:1.4493974931283106e-05 tokens_per_second_per_gpu:4159.272749890233 mfu:19.61999920633965 +Step 27299 | loss:0.3581252098083496 lr:1.449092201710463e-05 tokens_per_second_per_gpu:4804.114146256393 mfu:22.661826094286077 +Step 27300 | loss:0.2969112992286682 lr:1.448786936999445e-05 tokens_per_second_per_gpu:3661.145796459795 mfu:17.270249377786204 +Step 27301 | loss:0.32293710112571716 lr:1.448481698997553e-05 tokens_per_second_per_gpu:5293.315002989978 mfu:24.969470001771324 +Step 27302 | loss:0.38047096133232117 lr:1.4481764877070852e-05 tokens_per_second_per_gpu:3167.501951211826 mfu:14.941647135426596 +Step 27303 | loss:0.45257505774497986 lr:1.4478713031303337e-05 tokens_per_second_per_gpu:3366.5870476018695 mfu:15.880765502518486 +Step 27304 | loss:0.4510281980037689 lr:1.4475661452695972e-05 tokens_per_second_per_gpu:4151.075685685457 mfu:19.581332255922344 +Step 27305 | loss:0.4860627055168152 lr:1.447261014127167e-05 tokens_per_second_per_gpu:4642.024581702689 mfu:21.897221962954816 +Step 27306 | loss:0.4505907893180847 lr:1.446955909705342e-05 tokens_per_second_per_gpu:3950.511887667367 mfu:18.635238600958452 +Step 27307 | loss:0.5284914374351501 lr:1.4466508320064131e-05 tokens_per_second_per_gpu:5345.253843376744 mfu:25.21447437355566 +Step 27308 | loss:0.47180265188217163 lr:1.446345781032677e-05 tokens_per_second_per_gpu:5069.906025078761 mfu:23.91561173546193 +Step 27309 | loss:0.3408295810222626 lr:1.4460407567864287e-05 tokens_per_second_per_gpu:4089.126574392536 mfu:19.28910772834535 +Step 27310 | loss:0.21892711520195007 lr:1.44573575926996e-05 tokens_per_second_per_gpu:3450.4195913885683 mfu:16.276217914867225 +Step 27311 | loss:0.4223559498786926 lr:1.4454307884855683e-05 tokens_per_second_per_gpu:5358.906852177704 mfu:25.278877945513294 +Step 27312 | loss:0.4168274700641632 lr:1.445125844435543e-05 tokens_per_second_per_gpu:5337.184886318218 mfu:25.176411726404552 +Step 27313 | loss:0.45575541257858276 lr:1.4448209271221818e-05 tokens_per_second_per_gpu:3707.081497493134 mfu:17.486935916999233 +Step 27314 | loss:0.47049662470817566 lr:1.4445160365477745e-05 tokens_per_second_per_gpu:4819.213000100427 mfu:22.733049963998496 +Step 27315 | loss:0.5601497888565063 lr:1.4442111727146157e-05 tokens_per_second_per_gpu:4700.249928423806 mfu:22.171880857706864 +Step 27316 | loss:0.34570249915122986 lr:1.4439063356249998e-05 tokens_per_second_per_gpu:4722.869960010772 mfu:22.27858340607794 +Step 27317 | loss:0.4216572940349579 lr:1.4436015252812163e-05 tokens_per_second_per_gpu:5098.219516435877 mfu:24.04917130497321 +Step 27318 | loss:0.3559630811214447 lr:1.4432967416855613e-05 tokens_per_second_per_gpu:5370.439723222528 mfu:25.333280465942433 +Step 27319 | loss:0.3712095022201538 lr:1.4429919848403244e-05 tokens_per_second_per_gpu:4793.1901134596865 mfu:22.610295567751827 +Step 27320 | loss:0.41777321696281433 lr:1.4426872547477993e-05 tokens_per_second_per_gpu:4986.482094788148 mfu:23.52208682269088 +Step 27321 | loss:0.3492395877838135 lr:1.4423825514102762e-05 tokens_per_second_per_gpu:3792.1297383422716 mfu:17.888122979810802 +Step 27322 | loss:0.49099743366241455 lr:1.4420778748300479e-05 tokens_per_second_per_gpu:5382.7137459574715 mfu:25.39117912534738 +Step 27323 | loss:0.39771220088005066 lr:1.4417732250094057e-05 tokens_per_second_per_gpu:4901.271105590216 mfu:23.120132048150317 +Step 27324 | loss:0.4473436176776886 lr:1.441468601950643e-05 tokens_per_second_per_gpu:5329.826435642584 mfu:25.141700659086446 +Step 27325 | loss:0.30060112476348877 lr:1.4411640056560488e-05 tokens_per_second_per_gpu:5025.883046971819 mfu:23.7079478366369 +Step 27326 | loss:0.3686702847480774 lr:1.4408594361279127e-05 tokens_per_second_per_gpu:5313.945995586205 mfu:25.066789913858038 +Step 27327 | loss:0.5265968441963196 lr:1.440554893368527e-05 tokens_per_second_per_gpu:4417.056496888108 mfu:20.836009123371188 +Step 27328 | loss:0.46790939569473267 lr:1.4402503773801834e-05 tokens_per_second_per_gpu:4726.66152988937 mfu:22.296468888103814 +Step 27329 | loss:0.33023780584335327 lr:1.4399458881651695e-05 tokens_per_second_per_gpu:5349.9501519639825 mfu:25.236627662434532 +Step 27330 | loss:0.4127535820007324 lr:1.4396414257257774e-05 tokens_per_second_per_gpu:4995.878264048712 mfu:23.56641015624475 +Step 27331 | loss:0.44569236040115356 lr:1.4393369900642972e-05 tokens_per_second_per_gpu:5350.53508900587 mfu:25.239386910261665 +Step 27332 | loss:0.29002389311790466 lr:1.439032581183018e-05 tokens_per_second_per_gpu:5171.705109896519 mfu:24.39581538726217 +Step 27333 | loss:0.3231239318847656 lr:1.4387281990842277e-05 tokens_per_second_per_gpu:4832.269224622506 mfu:22.79463840269075 +Step 27334 | loss:0.4240492880344391 lr:1.4384238437702174e-05 tokens_per_second_per_gpu:4878.73235766321 mfu:23.013812928671697 +Step 27335 | loss:0.4032641053199768 lr:1.438119515243277e-05 tokens_per_second_per_gpu:4071.6653394047394 mfu:19.20674010371402 +Step 27336 | loss:0.4247848093509674 lr:1.4378152135056928e-05 tokens_per_second_per_gpu:5178.656491482197 mfu:24.428606240229815 +Step 27337 | loss:0.3641796410083771 lr:1.4375109385597546e-05 tokens_per_second_per_gpu:5319.249209102195 mfu:25.09180607683449 +Step 27338 | loss:0.3407007157802582 lr:1.4372066904077531e-05 tokens_per_second_per_gpu:4036.2782197839956 mfu:19.03981302279778 +Step 27339 | loss:0.40861785411834717 lr:1.4369024690519727e-05 tokens_per_second_per_gpu:4439.191834076017 mfu:20.940425285564512 +Step 27340 | loss:0.45199263095855713 lr:1.436598274494706e-05 tokens_per_second_per_gpu:5360.114279479654 mfu:25.284573586104933 +Step 27341 | loss:0.3892112374305725 lr:1.4362941067382357e-05 tokens_per_second_per_gpu:4694.803756725661 mfu:22.146190336594003 +Step 27342 | loss:0.35744473338127136 lr:1.4359899657848547e-05 tokens_per_second_per_gpu:4928.665191935089 mfu:23.24935462735217 +Step 27343 | loss:0.47800159454345703 lr:1.4356858516368454e-05 tokens_per_second_per_gpu:5237.185003249236 mfu:24.704695216228764 +Step 27344 | loss:0.291947603225708 lr:1.4353817642964985e-05 tokens_per_second_per_gpu:4806.02830135146 mfu:22.670855490458045 +Step 27345 | loss:0.5078390836715698 lr:1.4350777037661011e-05 tokens_per_second_per_gpu:4002.7113259273774 mfu:18.88147225737346 +Step 27346 | loss:0.39294302463531494 lr:1.4347736700479381e-05 tokens_per_second_per_gpu:5315.66532407027 mfu:25.07490027966506 +Step 27347 | loss:0.43572816252708435 lr:1.4344696631442983e-05 tokens_per_second_per_gpu:3920.0442433330295 mfu:18.491517524317224 +Step 27348 | loss:0.45858174562454224 lr:1.4341656830574662e-05 tokens_per_second_per_gpu:5358.578107325086 mfu:25.27732719995384 +Step 27349 | loss:0.4407152533531189 lr:1.43386172978973e-05 tokens_per_second_per_gpu:4288.9502579612545 mfu:20.231710137175018 +Step 27350 | loss:0.43274176120758057 lr:1.4335578033433732e-05 tokens_per_second_per_gpu:3793.20669747965 mfu:17.89320317980999 +Step 27351 | loss:0.37246811389923096 lr:1.4332539037206837e-05 tokens_per_second_per_gpu:3718.4200536856674 mfu:17.54042182111658 +Step 27352 | loss:0.4947121739387512 lr:1.432950030923948e-05 tokens_per_second_per_gpu:4659.069596479636 mfu:21.977626205837076 +Step 27353 | loss:0.38521721959114075 lr:1.432646184955449e-05 tokens_per_second_per_gpu:3765.082549895601 mfu:17.76053677718165 +Step 27354 | loss:0.4205794632434845 lr:1.4323423658174745e-05 tokens_per_second_per_gpu:4722.392841667102 mfu:22.276332757446067 +Step 27355 | loss:0.44991952180862427 lr:1.4320385735123071e-05 tokens_per_second_per_gpu:4730.480607317478 mfu:22.314484127933998 +Step 27356 | loss:0.4325707256793976 lr:1.4317348080422338e-05 tokens_per_second_per_gpu:5372.084209613671 mfu:25.341037788827766 +Step 27357 | loss:0.44171178340911865 lr:1.4314310694095374e-05 tokens_per_second_per_gpu:4183.4584510647355 mfu:19.734087285285906 +Step 27358 | loss:0.34905901551246643 lr:1.431127357616503e-05 tokens_per_second_per_gpu:4968.211350858178 mfu:23.43590060626285 +Step 27359 | loss:0.43033385276794434 lr:1.4308236726654151e-05 tokens_per_second_per_gpu:5347.185281674711 mfu:25.223585297544524 +Step 27360 | loss:0.4651119112968445 lr:1.430520014558559e-05 tokens_per_second_per_gpu:5118.856379752465 mfu:24.146518910249508 +Step 27361 | loss:0.38880693912506104 lr:1.4302163832982179e-05 tokens_per_second_per_gpu:4431.160914388738 mfu:20.90254206718343 +Step 27362 | loss:0.40263107419013977 lr:1.4299127788866735e-05 tokens_per_second_per_gpu:3862.7812685266863 mfu:18.221398302084744 +Step 27363 | loss:0.39043116569519043 lr:1.4296092013262097e-05 tokens_per_second_per_gpu:3742.710470242671 mfu:17.655003860388387 +Step 27364 | loss:0.3519420921802521 lr:1.4293056506191133e-05 tokens_per_second_per_gpu:4171.658035246592 mfu:19.67842271051317 +Step 27365 | loss:0.3360808789730072 lr:1.4290021267676622e-05 tokens_per_second_per_gpu:5107.141163908486 mfu:24.09125623829153 +Step 27366 | loss:0.3636433780193329 lr:1.4286986297741423e-05 tokens_per_second_per_gpu:4384.487248407083 mfu:20.682374421399388 +Step 27367 | loss:0.4222911596298218 lr:1.4283951596408373e-05 tokens_per_second_per_gpu:5346.919228072668 mfu:25.222330277310267 +Step 27368 | loss:0.3996419906616211 lr:1.4280917163700263e-05 tokens_per_second_per_gpu:4983.432166604597 mfu:23.507699791077666 +Step 27369 | loss:0.38196346163749695 lr:1.4277882999639953e-05 tokens_per_second_per_gpu:5314.157066184814 mfu:25.067785569884034 +Step 27370 | loss:0.4279269278049469 lr:1.4274849104250222e-05 tokens_per_second_per_gpu:3954.502621722733 mfu:18.654063574387052 +Step 27371 | loss:0.3611464202404022 lr:1.4271815477553924e-05 tokens_per_second_per_gpu:4426.245988942767 mfu:20.879357525282035 +Step 27372 | loss:0.4276338219642639 lr:1.4268782119573848e-05 tokens_per_second_per_gpu:5360.12445734166 mfu:25.284621596816535 +Step 27373 | loss:0.4310947358608246 lr:1.4265749030332826e-05 tokens_per_second_per_gpu:4600.938056653435 mfu:21.70340981421345 +Step 27374 | loss:0.4602193534374237 lr:1.4262716209853672e-05 tokens_per_second_per_gpu:5368.786191442397 mfu:25.325480474413936 +Step 27375 | loss:0.5028277635574341 lr:1.425968365815918e-05 tokens_per_second_per_gpu:5311.071970806609 mfu:25.05323264861349 +Step 27376 | loss:0.32299113273620605 lr:1.4256651375272178e-05 tokens_per_second_per_gpu:3795.158087296041 mfu:17.90240821851534 +Step 27377 | loss:0.566839337348938 lr:1.4253619361215453e-05 tokens_per_second_per_gpu:5379.08856451212 mfu:25.374078525949926 +Step 27378 | loss:0.4943818747997284 lr:1.425058761601183e-05 tokens_per_second_per_gpu:4108.614410945331 mfu:19.38103517833266 +Step 27379 | loss:0.30268892645835876 lr:1.4247556139684082e-05 tokens_per_second_per_gpu:5359.590119582628 mfu:25.282101034439762 +Step 27380 | loss:0.4562262296676636 lr:1.4244524932255027e-05 tokens_per_second_per_gpu:4973.224225505463 mfu:23.459547191258707 +Step 27381 | loss:0.45447927713394165 lr:1.4241493993747478e-05 tokens_per_second_per_gpu:4475.318750105006 mfu:21.110842112360732 +Step 27382 | loss:0.4513702094554901 lr:1.4238463324184203e-05 tokens_per_second_per_gpu:4765.717188485911 mfu:22.48070109328618 +Step 27383 | loss:0.353572279214859 lr:1.423543292358802e-05 tokens_per_second_per_gpu:3236.724128756159 mfu:15.268179957424568 +Step 27384 | loss:0.339881956577301 lr:1.4232402791981698e-05 tokens_per_second_per_gpu:4099.648112272633 mfu:19.33873961768524 +Step 27385 | loss:0.24269942939281464 lr:1.4229372929388051e-05 tokens_per_second_per_gpu:4144.364440511074 mfu:19.54967416737828 +Step 27386 | loss:0.40647396445274353 lr:1.422634333582984e-05 tokens_per_second_per_gpu:3816.896233468781 mfu:18.00495076292234 +Step 27387 | loss:0.2975429892539978 lr:1.4223314011329868e-05 tokens_per_second_per_gpu:3947.3074166898737 mfu:18.620122564618487 +Step 27388 | loss:0.3853060007095337 lr:1.4220284955910928e-05 tokens_per_second_per_gpu:5360.0293124983855 mfu:25.284172782357427 +Step 27389 | loss:0.37371864914894104 lr:1.4217256169595778e-05 tokens_per_second_per_gpu:5072.847907042032 mfu:23.929489094619502 +Step 27390 | loss:0.5353450179100037 lr:1.4214227652407225e-05 tokens_per_second_per_gpu:3935.890287403687 mfu:18.56626601781242 +Step 27391 | loss:0.4090742766857147 lr:1.4211199404368014e-05 tokens_per_second_per_gpu:5193.926790662188 mfu:24.500638846843525 +Step 27392 | loss:0.4080902934074402 lr:1.4208171425500955e-05 tokens_per_second_per_gpu:2973.3850442483617 mfu:14.025964565520406 +Step 27393 | loss:0.3792042136192322 lr:1.4205143715828796e-05 tokens_per_second_per_gpu:5313.890039190673 mfu:25.066525957993164 +Step 27394 | loss:0.3959570825099945 lr:1.420211627537431e-05 tokens_per_second_per_gpu:5084.000438588683 mfu:23.98209748874315 +Step 27395 | loss:0.4220239222049713 lr:1.419908910416028e-05 tokens_per_second_per_gpu:3733.6327108044006 mfu:17.612182520292595 +Step 27396 | loss:0.4572286307811737 lr:1.4196062202209488e-05 tokens_per_second_per_gpu:3605.7665511886876 mfu:17.00901602916883 +Step 27397 | loss:0.34393110871315 lr:1.4193035569544661e-05 tokens_per_second_per_gpu:4324.415416395513 mfu:20.3990053404894 +Step 27398 | loss:0.5359929800033569 lr:1.4190009206188598e-05 tokens_per_second_per_gpu:4642.135763786587 mfu:21.89774642781418 +Step 27399 | loss:0.3978220224380493 lr:1.4186983112164027e-05 tokens_per_second_per_gpu:5123.267204568061 mfu:24.167325523468993 +Step 27400 | loss:0.3884643614292145 lr:1.4183957287493744e-05 tokens_per_second_per_gpu:5312.070529713757 mfu:25.05794302135023 +Step 27401 | loss:0.38458937406539917 lr:1.4180931732200475e-05 tokens_per_second_per_gpu:5350.813818942762 mfu:25.240701726930215 +Step 27402 | loss:0.3931683897972107 lr:1.4177906446306988e-05 tokens_per_second_per_gpu:3701.602684686971 mfu:17.461091422210703 +Step 27403 | loss:0.297066330909729 lr:1.4174881429836051e-05 tokens_per_second_per_gpu:3456.5645774338172 mfu:16.305204862485397 +Step 27404 | loss:0.2747357487678528 lr:1.4171856682810386e-05 tokens_per_second_per_gpu:5126.684836136428 mfu:24.183447074685056 +Step 27405 | loss:0.39390406012535095 lr:1.416883220525278e-05 tokens_per_second_per_gpu:4428.067849492196 mfu:20.88795155233515 +Step 27406 | loss:0.3820056617259979 lr:1.4165807997185937e-05 tokens_per_second_per_gpu:5155.443360497929 mfu:24.319105940808473 +Step 27407 | loss:0.3689970374107361 lr:1.4162784058632644e-05 tokens_per_second_per_gpu:5304.698069191859 mfu:25.023165867196962 +Step 27408 | loss:0.4706583023071289 lr:1.4159760389615606e-05 tokens_per_second_per_gpu:5350.85399886026 mfu:25.240891262456454 +Step 27409 | loss:0.4456959366798401 lr:1.4156736990157588e-05 tokens_per_second_per_gpu:4648.020540532458 mfu:21.9255059237709 +Step 27410 | loss:0.40543246269226074 lr:1.4153713860281337e-05 tokens_per_second_per_gpu:5321.715822760849 mfu:25.103441514310045 +Step 27411 | loss:0.384780615568161 lr:1.4150691000009563e-05 tokens_per_second_per_gpu:4098.869227014946 mfu:19.3350654830332 +Step 27412 | loss:0.3982560932636261 lr:1.4147668409365027e-05 tokens_per_second_per_gpu:4320.241630265697 mfu:20.37931687919371 +Step 27413 | loss:0.4602762758731842 lr:1.4144646088370445e-05 tokens_per_second_per_gpu:3803.9457670015063 mfu:17.943861203018784 +Step 27414 | loss:0.35079774260520935 lr:1.4141624037048568e-05 tokens_per_second_per_gpu:4701.125577391569 mfu:22.17601144328892 +Step 27415 | loss:0.3756067752838135 lr:1.4138602255422095e-05 tokens_per_second_per_gpu:4519.369543768058 mfu:21.318637221909757 +Step 27416 | loss:0.4999680519104004 lr:1.413558074351377e-05 tokens_per_second_per_gpu:3831.9618051101907 mfu:18.07601763480645 +Step 27417 | loss:0.34364211559295654 lr:1.4132559501346332e-05 tokens_per_second_per_gpu:5344.9143225252965 mfu:25.21287279577115 +Step 27418 | loss:0.4748159348964691 lr:1.4129538528942482e-05 tokens_per_second_per_gpu:4639.786817104229 mfu:21.88666604554187 +Step 27419 | loss:0.3332436680793762 lr:1.412651782632496e-05 tokens_per_second_per_gpu:4209.785279810581 mfu:19.858275428300228 +Step 27420 | loss:0.40382495522499084 lr:1.412349739351646e-05 tokens_per_second_per_gpu:2605.8533407136324 mfu:12.292254812572285 +Step 27421 | loss:0.4274940490722656 lr:1.412047723053973e-05 tokens_per_second_per_gpu:4468.529304047327 mfu:21.078815136907714 +Step 27422 | loss:0.3613497018814087 lr:1.4117457337417456e-05 tokens_per_second_per_gpu:3702.655397482092 mfu:17.466057248076616 +Step 27423 | loss:0.4897964894771576 lr:1.4114437714172358e-05 tokens_per_second_per_gpu:5123.237406729453 mfu:24.167184961980244 +Step 27424 | loss:0.45910313725471497 lr:1.4111418360827172e-05 tokens_per_second_per_gpu:5301.059594917101 mfu:25.006002563255205 +Step 27425 | loss:0.30669718980789185 lr:1.4108399277404572e-05 tokens_per_second_per_gpu:5309.449197041991 mfu:25.045577747892036 +Step 27426 | loss:0.4149639904499054 lr:1.4105380463927298e-05 tokens_per_second_per_gpu:5318.14096927905 mfu:25.086578320503506 +Step 27427 | loss:0.3367326259613037 lr:1.4102361920418022e-05 tokens_per_second_per_gpu:4662.467245734443 mfu:21.993653497070742 +Step 27428 | loss:0.41234931349754333 lr:1.4099343646899477e-05 tokens_per_second_per_gpu:5342.888681052285 mfu:25.203317499332883 +Step 27429 | loss:0.48703041672706604 lr:1.4096325643394332e-05 tokens_per_second_per_gpu:4917.442988866313 mfu:23.19641758077147 +Step 27430 | loss:0.36545005440711975 lr:1.4093307909925308e-05 tokens_per_second_per_gpu:4814.220805519374 mfu:22.709500930403443 +Step 27431 | loss:0.41376569867134094 lr:1.4090290446515092e-05 tokens_per_second_per_gpu:5364.7583818034345 mfu:25.306480609131768 +Step 27432 | loss:0.406972199678421 lr:1.40872732531864e-05 tokens_per_second_per_gpu:3603.299748294806 mfu:16.997379699038415 +Step 27433 | loss:0.34897178411483765 lr:1.4084256329961898e-05 tokens_per_second_per_gpu:4100.605045219341 mfu:19.34325363366446 +Step 27434 | loss:0.34455356001853943 lr:1.40812396768643e-05 tokens_per_second_per_gpu:4150.107407698974 mfu:19.57676472345485 +Step 27435 | loss:0.42307034134864807 lr:1.4078223293916277e-05 tokens_per_second_per_gpu:4653.861346539582 mfu:21.95305800225077 +Step 27436 | loss:0.34055253863334656 lr:1.407520718114051e-05 tokens_per_second_per_gpu:5360.3917462103145 mfu:25.285882444014163 +Step 27437 | loss:0.2870616018772125 lr:1.4072191338559693e-05 tokens_per_second_per_gpu:4526.845407171527 mfu:21.353902145097738 +Step 27438 | loss:0.442405104637146 lr:1.4069175766196508e-05 tokens_per_second_per_gpu:3981.56782435997 mfu:18.78173475302664 +Step 27439 | loss:0.47764065861701965 lr:1.4066160464073653e-05 tokens_per_second_per_gpu:3523.5457186875688 mfu:16.621166333940472 +Step 27440 | loss:0.4068466126918793 lr:1.406314543221377e-05 tokens_per_second_per_gpu:3527.8262027064725 mfu:16.641358107383525 +Step 27441 | loss:0.34690093994140625 lr:1.4060130670639576e-05 tokens_per_second_per_gpu:5359.157804833339 mfu:25.28006173200686 +Step 27442 | loss:0.3778226375579834 lr:1.4057116179373714e-05 tokens_per_second_per_gpu:5102.717547114684 mfu:24.07038928312831 +Step 27443 | loss:0.5043124556541443 lr:1.4054101958438875e-05 tokens_per_second_per_gpu:5023.290985740001 mfu:23.695720641555166 +Step 27444 | loss:0.4903920590877533 lr:1.405108800785771e-05 tokens_per_second_per_gpu:5354.721759240724 mfu:25.259136148079072 +Step 27445 | loss:0.41048794984817505 lr:1.4048074327652899e-05 tokens_per_second_per_gpu:4793.5886935462495 mfu:22.612175738024916 +Step 27446 | loss:0.37186723947525024 lr:1.4045060917847125e-05 tokens_per_second_per_gpu:4011.002893150387 mfu:18.920584994651644 +Step 27447 | loss:0.3871893584728241 lr:1.4042047778463019e-05 tokens_per_second_per_gpu:5181.340726447539 mfu:24.44126823453904 +Step 27448 | loss:0.3524262309074402 lr:1.4039034909523274e-05 tokens_per_second_per_gpu:4508.879295586912 mfu:21.26915293141839 +Step 27449 | loss:0.43620097637176514 lr:1.4036022311050523e-05 tokens_per_second_per_gpu:4069.610186355968 mfu:19.1970455961377 +Step 27450 | loss:0.2851123511791229 lr:1.4033009983067452e-05 tokens_per_second_per_gpu:4798.899092475191 mfu:22.637225795820218 +Step 27451 | loss:0.4539353549480438 lr:1.4029997925596688e-05 tokens_per_second_per_gpu:4909.043194994692 mfu:23.156794319926423 +Step 27452 | loss:0.44710707664489746 lr:1.4026986138660897e-05 tokens_per_second_per_gpu:4511.937890221968 mfu:21.283580844162003 +Step 27453 | loss:0.49487826228141785 lr:1.4023974622282754e-05 tokens_per_second_per_gpu:4573.372110058324 mfu:21.573376540888844 +Step 27454 | loss:0.41913649439811707 lr:1.402096337648487e-05 tokens_per_second_per_gpu:5359.384469777639 mfu:25.281130949222156 +Step 27455 | loss:0.34262165427207947 lr:1.4017952401289925e-05 tokens_per_second_per_gpu:5316.22542109226 mfu:25.077542352880194 +Step 27456 | loss:0.38642922043800354 lr:1.4014941696720534e-05 tokens_per_second_per_gpu:5312.5187062704 mfu:25.06005714663472 +Step 27457 | loss:0.41432175040245056 lr:1.4011931262799377e-05 tokens_per_second_per_gpu:4427.320410426338 mfu:20.884425754735243 +Step 27458 | loss:0.2387431412935257 lr:1.400892109954906e-05 tokens_per_second_per_gpu:3791.0336487877826 mfu:17.882952538370112 +Step 27459 | loss:0.38176244497299194 lr:1.4005911206992244e-05 tokens_per_second_per_gpu:5326.177476552404 mfu:25.124487896481448 +Step 27460 | loss:0.4134886562824249 lr:1.4002901585151574e-05 tokens_per_second_per_gpu:5232.6930895209225 mfu:24.683506092772802 +Step 27461 | loss:0.3327397406101227 lr:1.3999892234049662e-05 tokens_per_second_per_gpu:5321.798693944523 mfu:25.103832431822717 +Step 27462 | loss:0.38928472995758057 lr:1.3996883153709151e-05 tokens_per_second_per_gpu:3176.7988240905584 mfu:14.985502071005657 +Step 27463 | loss:0.39107745885849 lr:1.399387434415269e-05 tokens_per_second_per_gpu:4455.478168037933 mfu:21.01725070159798 +Step 27464 | loss:0.44986048340797424 lr:1.3990865805402904e-05 tokens_per_second_per_gpu:4937.240053142793 mfu:23.289803710691093 +Step 27465 | loss:0.39813563227653503 lr:1.3987857537482384e-05 tokens_per_second_per_gpu:5346.778918246783 mfu:25.221668411922238 +Step 27466 | loss:0.4576936364173889 lr:1.3984849540413786e-05 tokens_per_second_per_gpu:3683.279080816579 mfu:17.374655856424866 +Step 27467 | loss:0.40000417828559875 lr:1.3981841814219732e-05 tokens_per_second_per_gpu:4743.296594736117 mfu:22.374939327220652 +Step 27468 | loss:0.3306442201137543 lr:1.397883435892286e-05 tokens_per_second_per_gpu:3410.6304078061917 mfu:16.088525547174402 +Step 27469 | loss:0.37066593766212463 lr:1.3975827174545753e-05 tokens_per_second_per_gpu:5362.9123790253725 mfu:25.2977726990671 +Step 27470 | loss:0.43826499581336975 lr:1.3972820261111058e-05 tokens_per_second_per_gpu:5038.022598741819 mfu:23.7652121735571 +Step 27471 | loss:0.5054605603218079 lr:1.3969813618641386e-05 tokens_per_second_per_gpu:4812.179119322664 mfu:22.699869948265974 +Step 27472 | loss:0.30866360664367676 lr:1.3966807247159319e-05 tokens_per_second_per_gpu:5310.154513860774 mfu:25.048904847665696 +Step 27473 | loss:0.3964683711528778 lr:1.39638011466875e-05 tokens_per_second_per_gpu:5340.581199018903 mfu:25.19243271288439 +Step 27474 | loss:0.42572495341300964 lr:1.3960795317248526e-05 tokens_per_second_per_gpu:5086.586669753273 mfu:23.99429718240306 +Step 27475 | loss:0.42848122119903564 lr:1.395778975886503e-05 tokens_per_second_per_gpu:4190.654547219103 mfu:19.768032498626667 +Step 27476 | loss:0.3964816927909851 lr:1.395478447155958e-05 tokens_per_second_per_gpu:4858.113591615688 mfu:22.91655069130129 +Step 27477 | loss:0.3930716812610626 lr:1.3951779455354807e-05 tokens_per_second_per_gpu:5301.65373384996 mfu:25.008805217972057 +Step 27478 | loss:0.45916247367858887 lr:1.3948774710273288e-05 tokens_per_second_per_gpu:5357.668314177729 mfu:25.27303555045104 +Step 27479 | loss:0.4046311676502228 lr:1.3945770236337646e-05 tokens_per_second_per_gpu:4784.063750697289 mfu:22.567244957483542 +Step 27480 | loss:0.38888001441955566 lr:1.3942766033570453e-05 tokens_per_second_per_gpu:3260.9281947912164 mfu:15.382354666551434 +Step 27481 | loss:0.4211515188217163 lr:1.393976210199432e-05 tokens_per_second_per_gpu:5335.56664402231 mfu:25.16877820139415 +Step 27482 | loss:0.3565899729728699 lr:1.3936758441631852e-05 tokens_per_second_per_gpu:5048.167904790174 mfu:23.813069313154895 +Step 27483 | loss:0.434640109539032 lr:1.393375505250561e-05 tokens_per_second_per_gpu:3776.0301278633806 mfu:17.812178370305073 +Step 27484 | loss:0.31508952379226685 lr:1.3930751934638214e-05 tokens_per_second_per_gpu:5302.804058170477 mfu:25.0142314940575 +Step 27485 | loss:0.418058842420578 lr:1.3927749088052217e-05 tokens_per_second_per_gpu:5322.843832560553 mfu:25.10876252899357 +Step 27486 | loss:0.45378655195236206 lr:1.3924746512770238e-05 tokens_per_second_per_gpu:4655.921682725866 mfu:21.962776959571073 +Step 27487 | loss:0.3224269449710846 lr:1.3921744208814829e-05 tokens_per_second_per_gpu:4982.690022419151 mfu:23.504198970332062 +Step 27488 | loss:0.40286943316459656 lr:1.3918742176208589e-05 tokens_per_second_per_gpu:5347.861740611067 mfu:25.22677626976311 +Step 27489 | loss:0.44292473793029785 lr:1.3915740414974105e-05 tokens_per_second_per_gpu:2844.2656616895783 mfu:13.416886407952296 +Step 27490 | loss:0.4235946536064148 lr:1.3912738925133927e-05 tokens_per_second_per_gpu:3176.958217251339 mfu:14.986253955740336 +Step 27491 | loss:0.34604912996292114 lr:1.390973770671064e-05 tokens_per_second_per_gpu:4967.921941255051 mfu:23.43453541178653 +Step 27492 | loss:0.47289392352104187 lr:1.3906736759726846e-05 tokens_per_second_per_gpu:4686.9869084554075 mfu:22.109316929611378 +Step 27493 | loss:0.31830278038978577 lr:1.3903736084205082e-05 tokens_per_second_per_gpu:5334.616338787229 mfu:25.164295449461484 +Step 27494 | loss:0.3951866626739502 lr:1.3900735680167914e-05 tokens_per_second_per_gpu:4686.143875152455 mfu:22.105340197685035 +Step 27495 | loss:0.5081823468208313 lr:1.3897735547637913e-05 tokens_per_second_per_gpu:5344.632745045035 mfu:25.211544546754684 +Step 27496 | loss:0.3871264159679413 lr:1.3894735686637672e-05 tokens_per_second_per_gpu:3422.1627175742447 mfu:16.14292541996525 +Step 27497 | loss:0.3300628662109375 lr:1.3891736097189712e-05 tokens_per_second_per_gpu:5318.430467119061 mfu:25.08794393120826 +Step 27498 | loss:0.3847913444042206 lr:1.3888736779316614e-05 tokens_per_second_per_gpu:4762.204577510144 mfu:22.46413150800026 +Step 27499 | loss:0.3540755808353424 lr:1.3885737733040948e-05 tokens_per_second_per_gpu:4229.757834029627 mfu:19.9524893742206 +Step 27500 | loss:0.36547616124153137 lr:1.3882738958385261e-05 tokens_per_second_per_gpu:5371.713322149625 mfu:25.339288249379937 +Step 27501 | loss:0.4813002347946167 lr:1.3879740455372081e-05 tokens_per_second_per_gpu:5212.828678795287 mfu:24.589802278162022 +Step 27502 | loss:0.39742109179496765 lr:1.3876742224023986e-05 tokens_per_second_per_gpu:3717.2322249378994 mfu:17.53481863024888 +Step 27503 | loss:0.3330076336860657 lr:1.3873744264363536e-05 tokens_per_second_per_gpu:5341.7348603401115 mfu:25.197874730171446 +Step 27504 | loss:0.35336974263191223 lr:1.3870746576413251e-05 tokens_per_second_per_gpu:4389.12162545263 mfu:20.704235568629496 +Step 27505 | loss:0.40293481945991516 lr:1.3867749160195692e-05 tokens_per_second_per_gpu:5363.675918736297 mfu:25.30137444615684 +Step 27506 | loss:0.3650986850261688 lr:1.3864752015733417e-05 tokens_per_second_per_gpu:4561.701607396732 mfu:21.51832478864103 +Step 27507 | loss:0.3352013826370239 lr:1.3861755143048954e-05 tokens_per_second_per_gpu:5301.175151192545 mfu:25.006547661922408 +Step 27508 | loss:0.4744076728820801 lr:1.3858758542164824e-05 tokens_per_second_per_gpu:5338.854995113042 mfu:25.184289914539683 +Step 27509 | loss:0.4076618552207947 lr:1.3855762213103584e-05 tokens_per_second_per_gpu:5352.137248176855 mfu:25.246944568427747 +Step 27510 | loss:0.4368833005428314 lr:1.3852766155887765e-05 tokens_per_second_per_gpu:4504.198491722351 mfu:21.247072780961762 +Step 27511 | loss:0.32165148854255676 lr:1.3849770370539921e-05 tokens_per_second_per_gpu:4570.410453661613 mfu:21.55940590235487 +Step 27512 | loss:0.4892040193080902 lr:1.3846774857082556e-05 tokens_per_second_per_gpu:4426.62129785107 mfu:20.8811279214368 +Step 27513 | loss:0.45071133971214294 lr:1.3843779615538221e-05 tokens_per_second_per_gpu:5323.476591490557 mfu:25.111747360826207 +Step 27514 | loss:0.4278561472892761 lr:1.3840784645929422e-05 tokens_per_second_per_gpu:5349.49759473219 mfu:25.23449287275794 +Step 27515 | loss:0.4265860319137573 lr:1.3837789948278707e-05 tokens_per_second_per_gpu:4100.434603083997 mfu:19.342449629031616 +Step 27516 | loss:0.43289974331855774 lr:1.3834795522608574e-05 tokens_per_second_per_gpu:5357.52501838627 mfu:25.27235959975394 +Step 27517 | loss:0.4482344686985016 lr:1.3831801368941561e-05 tokens_per_second_per_gpu:5241.453109594892 mfu:24.72482860207526 +Step 27518 | loss:0.4630756378173828 lr:1.3828807487300194e-05 tokens_per_second_per_gpu:5328.055418144125 mfu:25.13334646737947 +Step 27519 | loss:0.43043839931488037 lr:1.3825813877706973e-05 tokens_per_second_per_gpu:4858.503054977005 mfu:22.918387856425927 +Step 27520 | loss:0.4361129403114319 lr:1.3822820540184417e-05 tokens_per_second_per_gpu:5324.078785003305 mfu:25.114588010370646 +Step 27521 | loss:0.2984060049057007 lr:1.381982747475506e-05 tokens_per_second_per_gpu:4259.8943395182205 mfu:20.094648412428338 +Step 27522 | loss:0.43654289841651917 lr:1.3816834681441399e-05 tokens_per_second_per_gpu:4336.0114502845545 mfu:20.453705810831405 +Step 27523 | loss:0.40839266777038574 lr:1.3813842160265928e-05 tokens_per_second_per_gpu:4603.258572516795 mfu:21.714356083461755 +Step 27524 | loss:0.43508562445640564 lr:1.381084991125116e-05 tokens_per_second_per_gpu:4447.994774614185 mfu:20.981950258019584 +Step 27525 | loss:0.36534827947616577 lr:1.3807857934419627e-05 tokens_per_second_per_gpu:3898.13383093573 mfu:18.388162370737327 +Step 27526 | loss:0.24205973744392395 lr:1.38048662297938e-05 tokens_per_second_per_gpu:5313.581593109063 mfu:25.065070965199887 +Step 27527 | loss:0.45209771394729614 lr:1.3801874797396191e-05 tokens_per_second_per_gpu:4622.429188432572 mfu:21.80478714958083 +Step 27528 | loss:0.30496537685394287 lr:1.3798883637249316e-05 tokens_per_second_per_gpu:4738.500949678687 mfu:22.352317451263335 +Step 27529 | loss:0.3934437334537506 lr:1.3795892749375655e-05 tokens_per_second_per_gpu:4647.881739890153 mfu:21.924851177459317 +Step 27530 | loss:0.4563984274864197 lr:1.3792902133797692e-05 tokens_per_second_per_gpu:4632.79077957772 mfu:21.85366454288224 +Step 27531 | loss:0.4613611102104187 lr:1.378991179053793e-05 tokens_per_second_per_gpu:5337.134765395361 mfu:25.176175297459928 +Step 27532 | loss:0.32662689685821533 lr:1.3786921719618873e-05 tokens_per_second_per_gpu:5098.974495308267 mfu:24.052732669126954 +Step 27533 | loss:0.5236330628395081 lr:1.3783931921062987e-05 tokens_per_second_per_gpu:5320.5327835226935 mfu:25.09786091637636 +Step 27534 | loss:0.41444042325019836 lr:1.3780942394892771e-05 tokens_per_second_per_gpu:3430.080676782001 mfu:16.180275784491382 +Step 27535 | loss:0.49320274591445923 lr:1.3777953141130723e-05 tokens_per_second_per_gpu:5337.7055052053 mfu:25.178867574521668 +Step 27536 | loss:0.3537867069244385 lr:1.3774964159799314e-05 tokens_per_second_per_gpu:3867.830610509034 mfu:18.245216909722053 +Step 27537 | loss:0.5137201547622681 lr:1.3771975450921004e-05 tokens_per_second_per_gpu:5378.911410334137 mfu:25.373242859466764 +Step 27538 | loss:0.33785927295684814 lr:1.3768987014518291e-05 tokens_per_second_per_gpu:4317.71091928978 mfu:20.367379083736495 +Step 27539 | loss:0.4558470547199249 lr:1.3765998850613665e-05 tokens_per_second_per_gpu:4773.07617848094 mfu:22.515414704665897 +Step 27540 | loss:0.5176423192024231 lr:1.3763010959229567e-05 tokens_per_second_per_gpu:4490.798561369592 mfu:21.183863023223928 +Step 27541 | loss:0.39857226610183716 lr:1.3760023340388489e-05 tokens_per_second_per_gpu:4385.823337415438 mfu:20.688676981212094 +Step 27542 | loss:0.4010980725288391 lr:1.3757035994112915e-05 tokens_per_second_per_gpu:4807.514595612057 mfu:22.67786659407323 +Step 27543 | loss:0.45672258734703064 lr:1.37540489204253e-05 tokens_per_second_per_gpu:5334.480387846075 mfu:25.163654145676436 +Step 27544 | loss:0.37729933857917786 lr:1.375106211934809e-05 tokens_per_second_per_gpu:3720.13012996409 mfu:17.548488542691 +Step 27545 | loss:0.45429810881614685 lr:1.374807559090377e-05 tokens_per_second_per_gpu:4546.285067127472 mfu:21.445602337857935 +Step 27546 | loss:0.4564681053161621 lr:1.3745089335114797e-05 tokens_per_second_per_gpu:5011.5236954615875 mfu:23.64021232560514 +Step 27547 | loss:0.39087584614753723 lr:1.3742103352003649e-05 tokens_per_second_per_gpu:3103.370288213809 mfu:14.63912713907494 +Step 27548 | loss:0.41407957673072815 lr:1.3739117641592753e-05 tokens_per_second_per_gpu:4598.834048424784 mfu:21.693484848418088 +Step 27549 | loss:0.3966500461101532 lr:1.3736132203904594e-05 tokens_per_second_per_gpu:5379.071587546683 mfu:25.37399844270739 +Step 27550 | loss:0.40165379643440247 lr:1.3733147038961602e-05 tokens_per_second_per_gpu:5353.975012789312 mfu:25.25561361765968 +Step 27551 | loss:0.41703882813453674 lr:1.3730162146786252e-05 tokens_per_second_per_gpu:4072.891816462734 mfu:19.21252560525514 +Step 27552 | loss:0.31919169425964355 lr:1.3727177527400958e-05 tokens_per_second_per_gpu:3702.272980641123 mfu:17.46425332259072 +Step 27553 | loss:0.42334699630737305 lr:1.37241931808282e-05 tokens_per_second_per_gpu:3412.8883320392424 mfu:16.099176561023302 +Step 27554 | loss:0.4999181032180786 lr:1.3721209107090421e-05 tokens_per_second_per_gpu:5355.82098071493 mfu:25.264321363319425 +Step 27555 | loss:0.46444663405418396 lr:1.3718225306210048e-05 tokens_per_second_per_gpu:3548.0759764490303 mfu:16.736879745094836 +Step 27556 | loss:0.45208635926246643 lr:1.371524177820953e-05 tokens_per_second_per_gpu:5362.4414458087085 mfu:25.295551226734602 +Step 27557 | loss:0.3914780914783478 lr:1.3712258523111327e-05 tokens_per_second_per_gpu:5368.890536331709 mfu:25.3259726870599 +Step 27558 | loss:0.32887914776802063 lr:1.3709275540937855e-05 tokens_per_second_per_gpu:3734.858692274181 mfu:17.617965684059556 +Step 27559 | loss:0.3884827792644501 lr:1.3706292831711537e-05 tokens_per_second_per_gpu:5346.49018983993 mfu:25.220306430764968 +Step 27560 | loss:0.36913251876831055 lr:1.3703310395454821e-05 tokens_per_second_per_gpu:3246.2748891296537 mfu:15.313232523634094 +Step 27561 | loss:0.38481515645980835 lr:1.3700328232190157e-05 tokens_per_second_per_gpu:4766.509572415982 mfu:22.484438903478964 +Step 27562 | loss:0.4664624035358429 lr:1.3697346341939938e-05 tokens_per_second_per_gpu:3991.241095414484 mfu:18.827365223021015 +Step 27563 | loss:0.32991087436676025 lr:1.3694364724726611e-05 tokens_per_second_per_gpu:5227.970521444316 mfu:24.661228933402246 +Step 27564 | loss:0.4453270435333252 lr:1.3691383380572614e-05 tokens_per_second_per_gpu:5355.800813003549 mfu:25.264226228783876 +Step 27565 | loss:0.3584376871585846 lr:1.3688402309500353e-05 tokens_per_second_per_gpu:5153.707160110487 mfu:24.310915987354576 +Step 27566 | loss:0.4157711863517761 lr:1.3685421511532243e-05 tokens_per_second_per_gpu:5318.553850598328 mfu:25.088525952131352 +Step 27567 | loss:0.4688011109828949 lr:1.3682440986690709e-05 tokens_per_second_per_gpu:4765.732079207335 mfu:22.48077133535976 +Step 27568 | loss:0.43054431676864624 lr:1.3679460734998184e-05 tokens_per_second_per_gpu:4160.551785671609 mfu:19.6260326363464 +Step 27569 | loss:0.34170469641685486 lr:1.3676480756477055e-05 tokens_per_second_per_gpu:5351.732605279955 mfu:25.24503579884387 +Step 27570 | loss:0.4519260823726654 lr:1.3673501051149745e-05 tokens_per_second_per_gpu:5340.045691407692 mfu:25.189906632115285 +Step 27571 | loss:0.36605122685432434 lr:1.367052161903869e-05 tokens_per_second_per_gpu:5325.147418486583 mfu:25.11962893683894 +Step 27572 | loss:0.398325115442276 lr:1.3667542460166277e-05 tokens_per_second_per_gpu:5170.244925716478 mfu:24.388927449351804 +Step 27573 | loss:0.362217515707016 lr:1.3664563574554894e-05 tokens_per_second_per_gpu:4642.514242238313 mfu:21.899531775246057 +Step 27574 | loss:0.41793927550315857 lr:1.366158496222697e-05 tokens_per_second_per_gpu:4662.983463143232 mfu:21.996088582663774 +Step 27575 | loss:0.3805758059024811 lr:1.365860662320491e-05 tokens_per_second_per_gpu:4966.862413887353 mfu:23.42953744041909 +Step 27576 | loss:0.3838634788990021 lr:1.3655628557511103e-05 tokens_per_second_per_gpu:5343.791540920852 mfu:25.207576443376688 +Step 27577 | loss:0.3253788948059082 lr:1.365265076516794e-05 tokens_per_second_per_gpu:4290.484157550452 mfu:20.238945803247518 +Step 27578 | loss:0.45311903953552246 lr:1.3649673246197853e-05 tokens_per_second_per_gpu:5327.228223688541 mfu:25.12944444999855 +Step 27579 | loss:0.4706980586051941 lr:1.3646696000623205e-05 tokens_per_second_per_gpu:5022.192870350298 mfu:23.690540643904132 +Step 27580 | loss:0.4201628863811493 lr:1.3643719028466383e-05 tokens_per_second_per_gpu:4830.134942098796 mfu:22.784570627880264 +Step 27581 | loss:0.4267291724681854 lr:1.3640742329749789e-05 tokens_per_second_per_gpu:4628.489990768471 mfu:21.833376988278776 +Step 27582 | loss:0.49647361040115356 lr:1.3637765904495809e-05 tokens_per_second_per_gpu:4788.639471535444 mfu:22.5888294133756 +Step 27583 | loss:0.3974161744117737 lr:1.363478975272685e-05 tokens_per_second_per_gpu:4349.432706793685 mfu:20.51701616768746 +Step 27584 | loss:0.46862873435020447 lr:1.363181387446526e-05 tokens_per_second_per_gpu:4940.535064469307 mfu:23.305346841305166 +Step 27585 | loss:0.3962492346763611 lr:1.362883826973344e-05 tokens_per_second_per_gpu:5195.110916434615 mfu:24.506224570144756 +Step 27586 | loss:0.3608677387237549 lr:1.362586293855378e-05 tokens_per_second_per_gpu:5341.532995017555 mfu:25.19692249700317 +Step 27587 | loss:0.40864765644073486 lr:1.3622887880948654e-05 tokens_per_second_per_gpu:5007.944579664206 mfu:23.62332902572877 +Step 27588 | loss:0.49350887537002563 lr:1.3619913096940407e-05 tokens_per_second_per_gpu:3392.611897376734 mfu:16.00352915920376 +Step 27589 | loss:0.44210442900657654 lr:1.3616938586551436e-05 tokens_per_second_per_gpu:4833.937989436604 mfu:22.802510251039486 +Step 27590 | loss:0.40096789598464966 lr:1.3613964349804131e-05 tokens_per_second_per_gpu:4684.996484799083 mfu:22.099927761623203 +Step 27591 | loss:0.4628150463104248 lr:1.3610990386720829e-05 tokens_per_second_per_gpu:4419.622285681334 mfu:20.848112386877542 +Step 27592 | loss:0.45505982637405396 lr:1.360801669732391e-05 tokens_per_second_per_gpu:5315.895880984923 mfu:25.075987855968545 +Step 27593 | loss:0.40132075548171997 lr:1.3605043281635755e-05 tokens_per_second_per_gpu:3476.057587403786 mfu:16.397156716364186 +Step 27594 | loss:0.4263400137424469 lr:1.3602070139678714e-05 tokens_per_second_per_gpu:5244.4345556964145 mfu:24.7388925920233 +Step 27595 | loss:0.5908854603767395 lr:1.3599097271475137e-05 tokens_per_second_per_gpu:5321.779604601635 mfu:25.103742384135337 +Step 27596 | loss:0.28164535760879517 lr:1.359612467704739e-05 tokens_per_second_per_gpu:5326.1563877910785 mfu:25.124388417196162 +Step 27597 | loss:0.28665778040885925 lr:1.3593152356417848e-05 tokens_per_second_per_gpu:5337.588002681532 mfu:25.178313295069053 +Step 27598 | loss:0.5083027482032776 lr:1.3590180309608841e-05 tokens_per_second_per_gpu:4990.65497833113 mfu:23.541771026330476 +Step 27599 | loss:0.4471520185470581 lr:1.3587208536642731e-05 tokens_per_second_per_gpu:3792.863328480606 mfu:17.891583449656707 +Step 27600 | loss:0.3574341833591461 lr:1.3584237037541887e-05 tokens_per_second_per_gpu:4582.978849503906 mfu:21.61869316993242 +Step 27601 | loss:0.3613196611404419 lr:1.3581265812328641e-05 tokens_per_second_per_gpu:5318.172444644084 mfu:25.08672679517062 +Step 27602 | loss:0.38294100761413574 lr:1.357829486102533e-05 tokens_per_second_per_gpu:3183.5537112945867 mfu:15.017366026449656 +Step 27603 | loss:0.34076809883117676 lr:1.3575324183654315e-05 tokens_per_second_per_gpu:4092.9942434209615 mfu:19.307352183044816 +Step 27604 | loss:0.41226232051849365 lr:1.3572353780237945e-05 tokens_per_second_per_gpu:5383.3455358925785 mfu:25.39415938626576 +Step 27605 | loss:0.3365338146686554 lr:1.3569383650798533e-05 tokens_per_second_per_gpu:5333.189037742591 mfu:25.157562626911258 +Step 27606 | loss:0.5468800067901611 lr:1.356641379535844e-05 tokens_per_second_per_gpu:5294.121409996818 mfu:24.973273961209816 +Step 27607 | loss:0.4533044695854187 lr:1.3563444213940012e-05 tokens_per_second_per_gpu:5208.799676169751 mfu:24.570796785358638 +Step 27608 | loss:0.5142045021057129 lr:1.3560474906565568e-05 tokens_per_second_per_gpu:4543.46570575511 mfu:21.43230292923185 +Step 27609 | loss:0.3229013979434967 lr:1.3557505873257425e-05 tokens_per_second_per_gpu:5369.822324087483 mfu:25.330368088883272 +Step 27610 | loss:0.4179401695728302 lr:1.3554537114037935e-05 tokens_per_second_per_gpu:4309.254795710328 mfu:20.327490106048764 +Step 27611 | loss:0.46262261271476746 lr:1.3551568628929434e-05 tokens_per_second_per_gpu:3380.8812215901567 mfu:15.948193560058803 +Step 27612 | loss:0.4663343131542206 lr:1.3548600417954221e-05 tokens_per_second_per_gpu:4008.8874786131923 mfu:18.910606223352758 +Step 27613 | loss:0.3430556356906891 lr:1.3545632481134635e-05 tokens_per_second_per_gpu:4869.4450339838395 mfu:22.97000303829529 +Step 27614 | loss:0.47035372257232666 lr:1.3542664818493e-05 tokens_per_second_per_gpu:4690.1435740605275 mfu:22.12420745985342 +Step 27615 | loss:0.42313152551651 lr:1.3539697430051669e-05 tokens_per_second_per_gpu:4451.175370241396 mfu:20.99695367025871 +Step 27616 | loss:0.4219370186328888 lr:1.3536730315832891e-05 tokens_per_second_per_gpu:4096.037664586397 mfu:19.321708519943368 +Step 27617 | loss:0.5689234137535095 lr:1.3533763475859013e-05 tokens_per_second_per_gpu:3737.7053824608997 mfu:17.63139400736574 +Step 27618 | loss:0.4186764061450958 lr:1.353079691015236e-05 tokens_per_second_per_gpu:4806.951183950604 mfu:22.675208885138215 +Step 27619 | loss:0.5062744617462158 lr:1.3527830618735249e-05 tokens_per_second_per_gpu:5336.321389864885 mfu:25.172338466306496 +Step 27620 | loss:0.5016647577285767 lr:1.3524864601629967e-05 tokens_per_second_per_gpu:5097.075957765503 mfu:24.043776943613825 +Step 27621 | loss:0.4494197964668274 lr:1.3521898858858828e-05 tokens_per_second_per_gpu:4181.767337290554 mfu:19.726110013079328 +Step 27622 | loss:0.4465680420398712 lr:1.3518933390444156e-05 tokens_per_second_per_gpu:4570.198959194437 mfu:21.55840824686852 +Step 27623 | loss:0.37701645493507385 lr:1.3515968196408246e-05 tokens_per_second_per_gpu:5330.012037066813 mfu:25.1425761726714 +Step 27624 | loss:0.4796203374862671 lr:1.3513003276773373e-05 tokens_per_second_per_gpu:4641.559004243663 mfu:21.895025754644887 +Step 27625 | loss:0.44801461696624756 lr:1.3510038631561867e-05 tokens_per_second_per_gpu:3989.8605149388763 mfu:18.820852789366253 +Step 27626 | loss:0.3348667621612549 lr:1.350707426079602e-05 tokens_per_second_per_gpu:4436.614005201211 mfu:20.928265226939093 +Step 27627 | loss:0.4004005789756775 lr:1.3504110164498119e-05 tokens_per_second_per_gpu:4639.128135679254 mfu:21.883558932877623 +Step 27628 | loss:0.3889847993850708 lr:1.3501146342690458e-05 tokens_per_second_per_gpu:3114.2763960017683 mfu:14.69057311028461 +Step 27629 | loss:0.4396015405654907 lr:1.3498182795395347e-05 tokens_per_second_per_gpu:4296.702582540061 mfu:20.268279174896207 +Step 27630 | loss:0.4544391632080078 lr:1.3495219522635066e-05 tokens_per_second_per_gpu:3826.870489395143 mfu:18.052001029910535 +Step 27631 | loss:0.4505281150341034 lr:1.3492256524431878e-05 tokens_per_second_per_gpu:4582.072266622121 mfu:21.61441666380005 +Step 27632 | loss:0.3792845904827118 lr:1.348929380080809e-05 tokens_per_second_per_gpu:5341.305147085649 mfu:25.195847699433504 +Step 27633 | loss:0.3170050084590912 lr:1.3486331351785991e-05 tokens_per_second_per_gpu:4688.132189647025 mfu:22.114719416397364 +Step 27634 | loss:0.4969191253185272 lr:1.3483369177387844e-05 tokens_per_second_per_gpu:4888.728816073331 mfu:23.060967928563915 +Step 27635 | loss:0.41788265109062195 lr:1.3480407277635938e-05 tokens_per_second_per_gpu:5317.161525352222 mfu:25.081958116390293 +Step 27636 | loss:0.4095274806022644 lr:1.3477445652552557e-05 tokens_per_second_per_gpu:4912.945570357466 mfu:23.17520249032652 +Step 27637 | loss:0.3652535676956177 lr:1.3474484302159974e-05 tokens_per_second_per_gpu:3899.6389461811955 mfu:18.395262256149326 +Step 27638 | loss:0.3315683603286743 lr:1.3471523226480438e-05 tokens_per_second_per_gpu:5029.5335386287015 mfu:23.725167828620517 +Step 27639 | loss:0.378570556640625 lr:1.3468562425536235e-05 tokens_per_second_per_gpu:5197.697322094411 mfu:24.518425086928325 +Step 27640 | loss:0.4975827932357788 lr:1.346560189934965e-05 tokens_per_second_per_gpu:3594.7292175005537 mfu:16.95695104299859 +Step 27641 | loss:0.4474692642688751 lr:1.346264164794292e-05 tokens_per_second_per_gpu:3793.7381441032985 mfu:17.895710104207886 +Step 27642 | loss:0.651850163936615 lr:1.3459681671338326e-05 tokens_per_second_per_gpu:5341.640837576846 mfu:25.197431208755187 +Step 27643 | loss:0.39399024844169617 lr:1.3456721969558145e-05 tokens_per_second_per_gpu:3633.671682543479 mfu:17.14064929487574 +Step 27644 | loss:0.4108332097530365 lr:1.3453762542624621e-05 tokens_per_second_per_gpu:3490.981081205802 mfu:16.467553382838897 +Step 27645 | loss:0.2793913781642914 lr:1.3450803390559995e-05 tokens_per_second_per_gpu:3398.107488709412 mfu:16.029452801164574 +Step 27646 | loss:0.3771595060825348 lr:1.3447844513386543e-05 tokens_per_second_per_gpu:5360.145930274881 mfu:25.2847228883062 +Step 27647 | loss:0.5204008221626282 lr:1.344488591112653e-05 tokens_per_second_per_gpu:4346.8223869871945 mfu:20.50470284379376 +Step 27648 | loss:0.5158236026763916 lr:1.3441927583802178e-05 tokens_per_second_per_gpu:5352.941608047604 mfu:25.25073887117601 +Step 27649 | loss:0.3835594356060028 lr:1.3438969531435752e-05 tokens_per_second_per_gpu:5355.3696391560725 mfu:25.262192308179483 +Step 27650 | loss:0.46110183000564575 lr:1.3436011754049504e-05 tokens_per_second_per_gpu:5363.885275238385 mfu:25.302362016497085 +Step 27651 | loss:0.43645381927490234 lr:1.343305425166571e-05 tokens_per_second_per_gpu:4864.966724833103 mfu:22.948878089952938 +Step 27652 | loss:0.36891642212867737 lr:1.3430097024306548e-05 tokens_per_second_per_gpu:4659.492509988571 mfu:21.97962116101521 +Step 27653 | loss:0.3976426422595978 lr:1.3427140071994294e-05 tokens_per_second_per_gpu:4267.016570001295 mfu:20.128245188794832 +Step 27654 | loss:0.37404245138168335 lr:1.3424183394751183e-05 tokens_per_second_per_gpu:4931.579994650981 mfu:23.263104249079657 +Step 27655 | loss:0.49607163667678833 lr:1.3421226992599478e-05 tokens_per_second_per_gpu:4028.6786973715843 mfu:19.003964778966207 +Step 27656 | loss:0.4705371558666229 lr:1.3418270865561372e-05 tokens_per_second_per_gpu:4034.36989924473 mfu:19.030811149221 +Step 27657 | loss:0.4605279862880707 lr:1.3415315013659119e-05 tokens_per_second_per_gpu:4166.126269950421 mfu:19.652328429794654 +Step 27658 | loss:0.34884336590766907 lr:1.3412359436914978e-05 tokens_per_second_per_gpu:5305.953084782094 mfu:25.02908598986384 +Step 27659 | loss:0.5669928193092346 lr:1.3409404135351122e-05 tokens_per_second_per_gpu:4383.7718536124075 mfu:20.67899978209397 +Step 27660 | loss:0.510970950126648 lr:1.340644910898981e-05 tokens_per_second_per_gpu:3847.9963115760825 mfu:18.15165513757497 +Step 27661 | loss:0.4294441342353821 lr:1.3403494357853258e-05 tokens_per_second_per_gpu:4214.810745169457 mfu:19.881981405830846 +Step 27662 | loss:0.34811681509017944 lr:1.3400539881963713e-05 tokens_per_second_per_gpu:3840.4931334577523 mfu:18.116261366216612 +Step 27663 | loss:0.38358408212661743 lr:1.3397585681343361e-05 tokens_per_second_per_gpu:4954.344647546259 mfu:23.37048900083698 +Step 27664 | loss:0.379170298576355 lr:1.3394631756014436e-05 tokens_per_second_per_gpu:4709.424360248072 mfu:22.215158217941408 +Step 27665 | loss:0.47176504135131836 lr:1.3391678105999178e-05 tokens_per_second_per_gpu:5083.204586838381 mfu:23.978343320250666 +Step 27666 | loss:0.30553552508354187 lr:1.3388724731319775e-05 tokens_per_second_per_gpu:5178.34911715629 mfu:24.427156303091042 +Step 27667 | loss:0.39870426058769226 lr:1.3385771631998428e-05 tokens_per_second_per_gpu:5137.475486753698 mfu:24.234348414721552 +Step 27668 | loss:0.4300333559513092 lr:1.3382818808057367e-05 tokens_per_second_per_gpu:5359.577616964101 mfu:25.282042057455 +Step 27669 | loss:0.525041937828064 lr:1.3379866259518819e-05 tokens_per_second_per_gpu:4768.5202811421805 mfu:22.493923759602726 +Step 27670 | loss:0.4236249625682831 lr:1.3376913986404949e-05 tokens_per_second_per_gpu:4563.481156377766 mfu:21.526719224806936 +Step 27671 | loss:0.2931494116783142 lr:1.3373961988737982e-05 tokens_per_second_per_gpu:3456.830729074874 mfu:16.306460345186412 +Step 27672 | loss:0.4838777184486389 lr:1.3371010266540135e-05 tokens_per_second_per_gpu:4329.886003853563 mfu:20.424811034907492 +Step 27673 | loss:0.5380856394767761 lr:1.3368058819833596e-05 tokens_per_second_per_gpu:5083.073665042461 mfu:23.97772573980175 +Step 27674 | loss:0.41495123505592346 lr:1.336510764864054e-05 tokens_per_second_per_gpu:5355.242748596043 mfu:25.261593743758066 +Step 27675 | loss:0.518018364906311 lr:1.336215675298319e-05 tokens_per_second_per_gpu:5392.924425370736 mfu:25.439344642261297 +Step 27676 | loss:0.4149141013622284 lr:1.335920613288375e-05 tokens_per_second_per_gpu:3866.623767866906 mfu:18.239524027070548 +Step 27677 | loss:0.4533306956291199 lr:1.3356255788364375e-05 tokens_per_second_per_gpu:5348.90038123036 mfu:25.231675714774678 +Step 27678 | loss:0.4257200062274933 lr:1.3353305719447278e-05 tokens_per_second_per_gpu:5011.0604097007745 mfu:23.638026927626687 +Step 27679 | loss:0.45807749032974243 lr:1.3350355926154661e-05 tokens_per_second_per_gpu:5365.025069536373 mfu:25.307738620669657 +Step 27680 | loss:0.3619876205921173 lr:1.3347406408508695e-05 tokens_per_second_per_gpu:5336.821397167663 mfu:25.174697086063745 +Step 27681 | loss:0.3879532217979431 lr:1.334445716653155e-05 tokens_per_second_per_gpu:5323.572537322468 mfu:25.112199953685376 +Step 27682 | loss:0.49942827224731445 lr:1.3341508200245417e-05 tokens_per_second_per_gpu:3548.831062838905 mfu:16.740441616426896 +Step 27683 | loss:0.48281538486480713 lr:1.3338559509672488e-05 tokens_per_second_per_gpu:5300.998187927836 mfu:25.005712896009726 +Step 27684 | loss:0.43001291155815125 lr:1.3335611094834922e-05 tokens_per_second_per_gpu:3749.3304161749174 mfu:17.68623127483004 +Step 27685 | loss:0.3873696029186249 lr:1.3332662955754904e-05 tokens_per_second_per_gpu:5344.092956053067 mfu:25.20899826998207 +Step 27686 | loss:0.402347594499588 lr:1.3329715092454603e-05 tokens_per_second_per_gpu:5323.338111727992 mfu:25.111094128159102 +Step 27687 | loss:0.34104540944099426 lr:1.3326767504956227e-05 tokens_per_second_per_gpu:3937.256790544893 mfu:18.57271204627897 +Step 27688 | loss:0.338358998298645 lr:1.3323820193281877e-05 tokens_per_second_per_gpu:5332.855291326637 mfu:25.155988288124192 +Step 27689 | loss:0.31094369292259216 lr:1.3320873157453761e-05 tokens_per_second_per_gpu:5342.2270217884525 mfu:25.200196339694674 +Step 27690 | loss:0.34467488527297974 lr:1.3317926397494035e-05 tokens_per_second_per_gpu:5358.443117775877 mfu:25.27669043121863 +Step 27691 | loss:0.3804170489311218 lr:1.3314979913424885e-05 tokens_per_second_per_gpu:4994.14490023469 mfu:23.558233583391644 +Step 27692 | loss:0.5347428917884827 lr:1.3312033705268434e-05 tokens_per_second_per_gpu:4352.361992873587 mfu:20.5308341099141 +Step 27693 | loss:0.3350761830806732 lr:1.3309087773046858e-05 tokens_per_second_per_gpu:4129.475993913116 mfu:19.479442824544783 +Step 27694 | loss:0.45984551310539246 lr:1.3306142116782344e-05 tokens_per_second_per_gpu:4505.272139570561 mfu:21.252137361044984 +Step 27695 | loss:0.46348991990089417 lr:1.3303196736496986e-05 tokens_per_second_per_gpu:5211.244297926287 mfu:24.582328483279696 +Step 27696 | loss:0.3091297745704651 lr:1.3300251632212974e-05 tokens_per_second_per_gpu:4402.294059473762 mfu:20.76637218735617 +Step 27697 | loss:0.4159715473651886 lr:1.3297306803952447e-05 tokens_per_second_per_gpu:5385.361044869826 mfu:25.40366688599273 +Step 27698 | loss:0.33986616134643555 lr:1.3294362251737574e-05 tokens_per_second_per_gpu:4156.58284944616 mfu:19.607310487000614 +Step 27699 | loss:0.5602121353149414 lr:1.3291417975590476e-05 tokens_per_second_per_gpu:4428.384917801377 mfu:20.88944721764699 +Step 27700 | loss:0.4532413184642792 lr:1.3288473975533305e-05 tokens_per_second_per_gpu:4271.804135138845 mfu:20.150828950391496 +Step 27701 | loss:0.42820146679878235 lr:1.3285530251588218e-05 tokens_per_second_per_gpu:5333.987212169536 mfu:25.161327751865755 +Step 27702 | loss:0.4567113220691681 lr:1.3282586803777347e-05 tokens_per_second_per_gpu:4412.675861464665 mfu:20.815344918665332 +Step 27703 | loss:0.42140859365463257 lr:1.3279643632122806e-05 tokens_per_second_per_gpu:5333.574537916133 mfu:25.159381097002964 +Step 27704 | loss:0.38335803151130676 lr:1.3276700736646752e-05 tokens_per_second_per_gpu:4276.812368217838 mfu:20.174453640317783 +Step 27705 | loss:0.38310158252716064 lr:1.3273758117371338e-05 tokens_per_second_per_gpu:4657.6298864975 mfu:21.970834848211485 +Step 27706 | loss:0.43338045477867126 lr:1.3270815774318651e-05 tokens_per_second_per_gpu:5311.1083508233005 mfu:25.053404259359297 +Step 27707 | loss:0.37996017932891846 lr:1.3267873707510852e-05 tokens_per_second_per_gpu:4153.3474127666095 mfu:19.59204838016075 +Step 27708 | loss:0.40713366866111755 lr:1.3264931916970075e-05 tokens_per_second_per_gpu:5099.9792324988575 mfu:24.057472185880805 +Step 27709 | loss:0.40008965134620667 lr:1.3261990402718432e-05 tokens_per_second_per_gpu:5037.367302278836 mfu:23.762121028336104 +Step 27710 | loss:0.4686814546585083 lr:1.3259049164778038e-05 tokens_per_second_per_gpu:5109.171339169154 mfu:24.100832921379713 +Step 27711 | loss:0.49674126505851746 lr:1.3256108203171014e-05 tokens_per_second_per_gpu:4301.120912037765 mfu:20.289121188981625 +Step 27712 | loss:0.51138836145401 lr:1.3253167517919512e-05 tokens_per_second_per_gpu:4645.2472866039 mfu:21.91242401182398 +Step 27713 | loss:0.40959033370018005 lr:1.3250227109045605e-05 tokens_per_second_per_gpu:4697.255688851661 mfu:22.157756518775106 +Step 27714 | loss:0.46210429072380066 lr:1.3247286976571432e-05 tokens_per_second_per_gpu:5333.938615086968 mfu:25.161098511135762 +Step 27715 | loss:0.4216688573360443 lr:1.3244347120519119e-05 tokens_per_second_per_gpu:4171.79376311516 mfu:19.67906296202677 +Step 27716 | loss:0.31567928194999695 lr:1.324140754091076e-05 tokens_per_second_per_gpu:4975.953277231991 mfu:23.472420593876976 +Step 27717 | loss:0.42893654108047485 lr:1.3238468237768448e-05 tokens_per_second_per_gpu:4560.619277950315 mfu:21.513219255977845 +Step 27718 | loss:0.4179914891719818 lr:1.3235529211114312e-05 tokens_per_second_per_gpu:4842.113604765102 mfu:22.841076023447823 +Step 27719 | loss:0.4124586284160614 lr:1.3232590460970462e-05 tokens_per_second_per_gpu:3006.0206392547925 mfu:14.179912235372425 +Step 27720 | loss:0.43230175971984863 lr:1.3229651987358982e-05 tokens_per_second_per_gpu:4521.8695931350685 mfu:21.33043038132635 +Step 27721 | loss:0.4270886182785034 lr:1.3226713790301976e-05 tokens_per_second_per_gpu:5372.087126354271 mfu:25.341051547590823 +Step 27722 | loss:0.3630249500274658 lr:1.3223775869821547e-05 tokens_per_second_per_gpu:5289.1346466929335 mfu:24.949750547875833 +Step 27723 | loss:0.4294734597206116 lr:1.3220838225939824e-05 tokens_per_second_per_gpu:5294.487920666125 mfu:24.975002854570103 +Step 27724 | loss:0.34698542952537537 lr:1.3217900858678845e-05 tokens_per_second_per_gpu:4259.0628451058665 mfu:20.09072611141769 +Step 27725 | loss:0.4013834297657013 lr:1.3214963768060723e-05 tokens_per_second_per_gpu:4101.689811577712 mfu:19.34837066165712 +Step 27726 | loss:0.4033546447753906 lr:1.3212026954107564e-05 tokens_per_second_per_gpu:4649.552299306479 mfu:21.932731491252955 +Step 27727 | loss:0.42177510261535645 lr:1.3209090416841435e-05 tokens_per_second_per_gpu:5380.707806120665 mfu:25.381716764888537 +Step 27728 | loss:0.26898568868637085 lr:1.3206154156284428e-05 tokens_per_second_per_gpu:5275.079209302231 mfu:24.88344865159899 +Step 27729 | loss:0.4014069437980652 lr:1.320321817245863e-05 tokens_per_second_per_gpu:5077.984034222906 mfu:23.953717082845348 +Step 27730 | loss:0.38539552688598633 lr:1.3200282465386155e-05 tokens_per_second_per_gpu:4127.26845958567 mfu:19.469029508478027 +Step 27731 | loss:0.41235148906707764 lr:1.319734703508902e-05 tokens_per_second_per_gpu:4608.50488266136 mfu:21.739103823526484 +Step 27732 | loss:0.4007492661476135 lr:1.3194411881589336e-05 tokens_per_second_per_gpu:5391.197716153442 mfu:25.431199460276357 +Step 27733 | loss:0.4495500326156616 lr:1.3191477004909175e-05 tokens_per_second_per_gpu:4127.352035851831 mfu:19.469423751985087 +Step 27734 | loss:0.38233837485313416 lr:1.3188542405070631e-05 tokens_per_second_per_gpu:4327.266059396283 mfu:20.41245230989388 +Step 27735 | loss:0.38563501834869385 lr:1.3185608082095736e-05 tokens_per_second_per_gpu:5324.45327602877 mfu:25.116354548432522 +Step 27736 | loss:0.445083349943161 lr:1.3182674036006582e-05 tokens_per_second_per_gpu:5272.137811615646 mfu:24.869573576857167 +Step 27737 | loss:0.3369942307472229 lr:1.3179740266825253e-05 tokens_per_second_per_gpu:3413.375605038451 mfu:16.101475110897915 +Step 27738 | loss:0.4204436242580414 lr:1.317680677457379e-05 tokens_per_second_per_gpu:4056.3035418093973 mfu:19.134275883463854 +Step 27739 | loss:0.39936888217926025 lr:1.3173873559274253e-05 tokens_per_second_per_gpu:5138.919048558602 mfu:24.241157942052332 +Step 27740 | loss:0.3203219175338745 lr:1.3170940620948707e-05 tokens_per_second_per_gpu:5355.238909498785 mfu:25.26157563408128 +Step 27741 | loss:0.4578992426395416 lr:1.3168007959619232e-05 tokens_per_second_per_gpu:5319.340880262518 mfu:25.092238505338447 +Step 27742 | loss:0.4921918213367462 lr:1.3165075575307856e-05 tokens_per_second_per_gpu:3747.4918109097134 mfu:17.677558260095665 +Step 27743 | loss:0.33117642998695374 lr:1.3162143468036647e-05 tokens_per_second_per_gpu:5377.8644254141345 mfu:25.36830405296506 +Step 27744 | loss:0.4201940596103668 lr:1.3159211637827674e-05 tokens_per_second_per_gpu:5346.886876855373 mfu:25.222177671098592 +Step 27745 | loss:0.41262614727020264 lr:1.3156280084702966e-05 tokens_per_second_per_gpu:4826.6065323477105 mfu:22.76792651707561 +Step 27746 | loss:0.2742469608783722 lr:1.3153348808684562e-05 tokens_per_second_per_gpu:5371.422229170963 mfu:25.337915114134002 +Step 27747 | loss:0.4826485514640808 lr:1.315041780979453e-05 tokens_per_second_per_gpu:4431.555130423866 mfu:20.90440165148132 +Step 27748 | loss:0.49385473132133484 lr:1.3147487088054916e-05 tokens_per_second_per_gpu:5183.110488441236 mfu:24.449616503815843 +Step 27749 | loss:0.5077090263366699 lr:1.3144556643487743e-05 tokens_per_second_per_gpu:4413.932049723313 mfu:20.82127057300919 +Step 27750 | loss:0.25221720337867737 lr:1.314162647611506e-05 tokens_per_second_per_gpu:4432.865716267156 mfu:20.91058390851312 +Step 27751 | loss:0.5247617363929749 lr:1.313869658595892e-05 tokens_per_second_per_gpu:5352.142103774672 mfu:25.2469674731111 +Step 27752 | loss:0.3927259147167206 lr:1.3135766973041352e-05 tokens_per_second_per_gpu:3683.4320981907435 mfu:17.37537766548626 +Step 27753 | loss:0.49414220452308655 lr:1.3132837637384365e-05 tokens_per_second_per_gpu:5336.411726784682 mfu:25.17276460097025 +Step 27754 | loss:0.38061290979385376 lr:1.3129908579010015e-05 tokens_per_second_per_gpu:5353.82761468427 mfu:25.254918315649487 +Step 27755 | loss:0.3803984820842743 lr:1.3126979797940336e-05 tokens_per_second_per_gpu:3763.950229909886 mfu:17.75519543061532 +Step 27756 | loss:0.4307716488838196 lr:1.3124051294197337e-05 tokens_per_second_per_gpu:5179.847798722168 mfu:24.43422583973873 +Step 27757 | loss:0.44676199555397034 lr:1.3121123067803055e-05 tokens_per_second_per_gpu:3245.779007164615 mfu:15.310893363798817 +Step 27758 | loss:0.41601279377937317 lr:1.3118195118779509e-05 tokens_per_second_per_gpu:5160.818137516155 mfu:24.344459681034966 +Step 27759 | loss:0.45319122076034546 lr:1.3115267447148755e-05 tokens_per_second_per_gpu:5328.3999306302485 mfu:25.134971591556404 +Step 27760 | loss:0.4633170962333679 lr:1.3112340052932748e-05 tokens_per_second_per_gpu:4358.739685494372 mfu:20.56091877415291 +Step 27761 | loss:0.40786615014076233 lr:1.3109412936153542e-05 tokens_per_second_per_gpu:5352.762632769219 mfu:25.249894614251538 +Step 27762 | loss:0.42484137415885925 lr:1.310648609683316e-05 tokens_per_second_per_gpu:4143.296785141821 mfu:19.544637854840722 +Step 27763 | loss:0.36714795231819153 lr:1.3103559534993593e-05 tokens_per_second_per_gpu:2541.2425704637208 mfu:11.987474785569317 +Step 27764 | loss:0.41724810004234314 lr:1.3100633250656858e-05 tokens_per_second_per_gpu:4684.754833619041 mfu:22.098787851777004 +Step 27765 | loss:0.34718072414398193 lr:1.309770724384497e-05 tokens_per_second_per_gpu:3295.747357510443 mfu:15.546602597859778 +Step 27766 | loss:0.5734211206436157 lr:1.3094781514579968e-05 tokens_per_second_per_gpu:4249.631794887752 mfu:20.046238238435752 +Step 27767 | loss:0.3781595528125763 lr:1.3091856062883789e-05 tokens_per_second_per_gpu:4456.2312261459265 mfu:21.02080300517835 +Step 27768 | loss:0.41260290145874023 lr:1.3088930888778473e-05 tokens_per_second_per_gpu:5260.829810528311 mfu:24.81623180638428 +Step 27769 | loss:0.40941140055656433 lr:1.3086005992286016e-05 tokens_per_second_per_gpu:4353.824119474869 mfu:20.53773121055687 +Step 27770 | loss:0.40506571531295776 lr:1.3083081373428435e-05 tokens_per_second_per_gpu:5168.702515751633 mfu:24.381651638385353 +Step 27771 | loss:0.44981423020362854 lr:1.3080157032227697e-05 tokens_per_second_per_gpu:4134.362710015349 mfu:19.50249429815876 +Step 27772 | loss:0.3249174654483795 lr:1.3077232968705805e-05 tokens_per_second_per_gpu:4484.077064226883 mfu:21.152156574396102 +Step 27773 | loss:0.4453263282775879 lr:1.307430918288477e-05 tokens_per_second_per_gpu:5340.681517420911 mfu:25.192905931902747 +Step 27774 | loss:0.4430352747440338 lr:1.3071385674786569e-05 tokens_per_second_per_gpu:4037.0331824735495 mfu:19.043374310614322 +Step 27775 | loss:0.3687358796596527 lr:1.3068462444433171e-05 tokens_per_second_per_gpu:4403.282730986853 mfu:20.77103591956817 +Step 27776 | loss:0.45570069551467896 lr:1.3065539491846585e-05 tokens_per_second_per_gpu:5338.82144507397 mfu:25.184131653281153 +Step 27777 | loss:0.48362264037132263 lr:1.3062616817048795e-05 tokens_per_second_per_gpu:5353.355409804429 mfu:25.252690844664816 +Step 27778 | loss:0.3391762375831604 lr:1.3059694420061768e-05 tokens_per_second_per_gpu:4217.370563678981 mfu:19.894056506490347 +Step 27779 | loss:0.31560978293418884 lr:1.3056772300907494e-05 tokens_per_second_per_gpu:3670.4248092172315 mfu:17.31402006412586 +Step 27780 | loss:0.3355244994163513 lr:1.3053850459607963e-05 tokens_per_second_per_gpu:5380.935583370346 mfu:25.38279122903809 +Step 27781 | loss:0.3463572561740875 lr:1.3050928896185138e-05 tokens_per_second_per_gpu:4855.284248379993 mfu:22.903204196522786 +Step 27782 | loss:0.44925814867019653 lr:1.3048007610660978e-05 tokens_per_second_per_gpu:4907.687446101113 mfu:23.150399021080798 +Step 27783 | loss:0.4016638994216919 lr:1.3045086603057467e-05 tokens_per_second_per_gpu:3279.7170544071714 mfu:15.470984923069905 +Step 27784 | loss:0.4465872347354889 lr:1.3042165873396583e-05 tokens_per_second_per_gpu:3268.746564828133 mfu:15.419235251967052 +Step 27785 | loss:0.37810075283050537 lr:1.303924542170028e-05 tokens_per_second_per_gpu:3548.1886781903354 mfu:16.73741137843747 +Step 27786 | loss:0.4879101514816284 lr:1.3036325247990521e-05 tokens_per_second_per_gpu:5348.538732213414 mfu:25.229969754658057 +Step 27787 | loss:0.3732445538043976 lr:1.3033405352289297e-05 tokens_per_second_per_gpu:5254.130225695496 mfu:24.784628721660848 +Step 27788 | loss:0.46758854389190674 lr:1.3030485734618547e-05 tokens_per_second_per_gpu:4737.8208092253235 mfu:22.34910911270089 +Step 27789 | loss:0.412031352519989 lr:1.3027566395000212e-05 tokens_per_second_per_gpu:4579.451821790236 mfu:21.602055578434058 +Step 27790 | loss:0.4498078525066376 lr:1.3024647333456274e-05 tokens_per_second_per_gpu:3552.956754596396 mfu:16.75990320836258 +Step 27791 | loss:0.4677464962005615 lr:1.3021728550008694e-05 tokens_per_second_per_gpu:4189.540762819675 mfu:19.76277858758335 +Step 27792 | loss:0.31196120381355286 lr:1.3018810044679398e-05 tokens_per_second_per_gpu:5352.863788830885 mfu:25.250371784653424 +Step 27793 | loss:0.4822147488594055 lr:1.3015891817490344e-05 tokens_per_second_per_gpu:4020.3750019690724 mfu:18.96479483099603 +Step 27794 | loss:0.3927207589149475 lr:1.3012973868463508e-05 tokens_per_second_per_gpu:5345.441586331424 mfu:25.215359989105135 +Step 27795 | loss:0.3570328950881958 lr:1.3010056197620812e-05 tokens_per_second_per_gpu:4456.7667867133005 mfu:21.023329335750706 +Step 27796 | loss:0.5109516978263855 lr:1.3007138804984192e-05 tokens_per_second_per_gpu:4703.786801457754 mfu:22.188564891260764 +Step 27797 | loss:0.40073859691619873 lr:1.3004221690575602e-05 tokens_per_second_per_gpu:2083.5473661390893 mfu:9.828448415915698 +Step 27798 | loss:0.4414841830730438 lr:1.300130485441699e-05 tokens_per_second_per_gpu:5355.366169947288 mfu:25.26217594332974 +Step 27799 | loss:0.5007382035255432 lr:1.2998388296530278e-05 tokens_per_second_per_gpu:4106.2189487339665 mfu:19.369735374373654 +Step 27800 | loss:0.3754505217075348 lr:1.2995472016937404e-05 tokens_per_second_per_gpu:4076.7580898483648 mfu:19.230763476469424 +Step 27801 | loss:0.35715436935424805 lr:1.2992556015660307e-05 tokens_per_second_per_gpu:4945.746225889853 mfu:23.3299287788417 +Step 27802 | loss:0.3948444724082947 lr:1.2989640292720934e-05 tokens_per_second_per_gpu:4058.888672688494 mfu:19.146470386889362 +Step 27803 | loss:0.38096022605895996 lr:1.2986724848141201e-05 tokens_per_second_per_gpu:4647.915619071965 mfu:21.92501099134031 +Step 27804 | loss:0.45725175738334656 lr:1.2983809681943016e-05 tokens_per_second_per_gpu:3619.2951388372517 mfu:17.072832685322176 +Step 27805 | loss:0.5332629084587097 lr:1.2980894794148323e-05 tokens_per_second_per_gpu:5332.353091707122 mfu:25.153619326834846 +Step 27806 | loss:0.36986440420150757 lr:1.2977980184779059e-05 tokens_per_second_per_gpu:5013.629886356383 mfu:23.650147587409116 +Step 27807 | loss:0.4305570423603058 lr:1.2975065853857115e-05 tokens_per_second_per_gpu:3438.7974502639454 mfu:16.22139429224063 +Step 27808 | loss:0.6422185897827148 lr:1.2972151801404424e-05 tokens_per_second_per_gpu:3590.932082222263 mfu:16.939039308032807 +Step 27809 | loss:0.4455178380012512 lr:1.296923802744292e-05 tokens_per_second_per_gpu:5090.955976671714 mfu:24.014907948617974 +Step 27810 | loss:0.351419061422348 lr:1.29663245319945e-05 tokens_per_second_per_gpu:3728.409390281061 mfu:17.587543226193624 +Step 27811 | loss:0.37047436833381653 lr:1.2963411315081064e-05 tokens_per_second_per_gpu:3710.3457021867416 mfu:17.502333727469725 +Step 27812 | loss:0.4804425835609436 lr:1.2960498376724539e-05 tokens_per_second_per_gpu:4178.95115534112 mfu:19.712825602333663 +Step 27813 | loss:0.3894486427307129 lr:1.2957585716946852e-05 tokens_per_second_per_gpu:4004.199038188695 mfu:18.88849004993998 +Step 27814 | loss:0.48208633065223694 lr:1.2954673335769868e-05 tokens_per_second_per_gpu:3767.0550878092126 mfu:17.76984157506983 +Step 27815 | loss:0.41470083594322205 lr:1.2951761233215515e-05 tokens_per_second_per_gpu:5355.4081691952715 mfu:25.262374060948286 +Step 27816 | loss:0.2946406900882721 lr:1.2948849409305703e-05 tokens_per_second_per_gpu:5113.962747375626 mfu:24.123434811388183 +Step 27817 | loss:0.49770388007164 lr:1.294593786406233e-05 tokens_per_second_per_gpu:5347.542153500219 mfu:25.225268722834056 +Step 27818 | loss:0.3990269899368286 lr:1.2943026597507269e-05 tokens_per_second_per_gpu:4143.466678691322 mfu:19.54543927170055 +Step 27819 | loss:0.298515260219574 lr:1.2940115609662429e-05 tokens_per_second_per_gpu:4597.474272765211 mfu:21.68707055463023 +Step 27820 | loss:0.47488898038864136 lr:1.2937204900549727e-05 tokens_per_second_per_gpu:4196.1940187889595 mfu:19.794163131152988 +Step 27821 | loss:0.41397157311439514 lr:1.2934294470191022e-05 tokens_per_second_per_gpu:4244.048579540405 mfu:20.01990125904755 +Step 27822 | loss:0.38564619421958923 lr:1.2931384318608213e-05 tokens_per_second_per_gpu:4406.3065127858645 mfu:20.785299614224485 +Step 27823 | loss:0.4368962049484253 lr:1.2928474445823214e-05 tokens_per_second_per_gpu:4118.073486774745 mfu:19.42565525290476 +Step 27824 | loss:0.36479583382606506 lr:1.2925564851857886e-05 tokens_per_second_per_gpu:5339.6474191838215 mfu:25.18802791408304 +Step 27825 | loss:0.3853130638599396 lr:1.2922655536734102e-05 tokens_per_second_per_gpu:4146.960250692607 mfu:19.561919046895458 +Step 27826 | loss:0.37693464756011963 lr:1.2919746500473756e-05 tokens_per_second_per_gpu:5057.4859033294515 mfu:23.857023902079156 +Step 27827 | loss:0.440109521150589 lr:1.2916837743098741e-05 tokens_per_second_per_gpu:3527.8730503286934 mfu:16.641579095610567 +Step 27828 | loss:0.35858774185180664 lr:1.2913929264630909e-05 tokens_per_second_per_gpu:5297.433853071111 mfu:24.98889932035232 +Step 27829 | loss:0.5027979612350464 lr:1.2911021065092149e-05 tokens_per_second_per_gpu:4892.088877535247 mfu:23.076817911765374 +Step 27830 | loss:0.4628658592700958 lr:1.290811314450434e-05 tokens_per_second_per_gpu:5130.741922074031 mfu:24.202585041262967 +Step 27831 | loss:0.36031270027160645 lr:1.290520550288934e-05 tokens_per_second_per_gpu:4148.188877800379 mfu:19.567714690588218 +Step 27832 | loss:0.4326332211494446 lr:1.2902298140269032e-05 tokens_per_second_per_gpu:4659.938525398928 mfu:21.981725091803934 +Step 27833 | loss:0.33989641070365906 lr:1.2899391056665255e-05 tokens_per_second_per_gpu:3346.4256758830065 mfu:15.785660872235885 +Step 27834 | loss:0.47852709889411926 lr:1.2896484252099911e-05 tokens_per_second_per_gpu:4456.061299520611 mfu:21.020001432294265 +Step 27835 | loss:0.3447588384151459 lr:1.289357772659483e-05 tokens_per_second_per_gpu:4083.9865358454363 mfu:19.264861289537674 +Step 27836 | loss:0.3745294213294983 lr:1.2890671480171879e-05 tokens_per_second_per_gpu:4377.274801917311 mfu:20.648352080736874 +Step 27837 | loss:0.46254074573516846 lr:1.2887765512852928e-05 tokens_per_second_per_gpu:4604.029091202279 mfu:21.717990751565228 +Step 27838 | loss:0.4564697742462158 lr:1.2884859824659845e-05 tokens_per_second_per_gpu:5343.780868225892 mfu:25.207526098453613 +Step 27839 | loss:0.42916056513786316 lr:1.2881954415614466e-05 tokens_per_second_per_gpu:5326.373817419909 mfu:25.12541406985267 +Step 27840 | loss:0.36763691902160645 lr:1.2879049285738625e-05 tokens_per_second_per_gpu:5020.318558772957 mfu:23.681699196403283 +Step 27841 | loss:0.4682939946651459 lr:1.2876144435054193e-05 tokens_per_second_per_gpu:5320.341098296508 mfu:25.096956704459572 +Step 27842 | loss:0.4486493468284607 lr:1.2873239863583037e-05 tokens_per_second_per_gpu:5340.694360069792 mfu:25.1929665128687 +Step 27843 | loss:0.3472140431404114 lr:1.2870335571346959e-05 tokens_per_second_per_gpu:4129.328324160821 mfu:19.47874624112815 +Step 27844 | loss:0.33894577622413635 lr:1.2867431558367827e-05 tokens_per_second_per_gpu:3881.7116054497183 mfu:18.310695931199216 +Step 27845 | loss:0.43419116735458374 lr:1.2864527824667494e-05 tokens_per_second_per_gpu:4095.5234618076297 mfu:19.319282937703882 +Step 27846 | loss:0.40854498744010925 lr:1.286162437026779e-05 tokens_per_second_per_gpu:4698.768837513663 mfu:22.164894299183555 +Step 27847 | loss:0.46711239218711853 lr:1.2858721195190527e-05 tokens_per_second_per_gpu:4346.886393998072 mfu:20.505004775784794 +Step 27848 | loss:0.35583144426345825 lr:1.2855818299457567e-05 tokens_per_second_per_gpu:5346.3482306302685 mfu:25.219636784952254 +Step 27849 | loss:0.5112809538841248 lr:1.285291568309075e-05 tokens_per_second_per_gpu:4708.929500386116 mfu:22.212823879540764 +Step 27850 | loss:0.44272589683532715 lr:1.2850013346111878e-05 tokens_per_second_per_gpu:5352.163822180185 mfu:25.247069922535445 +Step 27851 | loss:0.485946387052536 lr:1.2847111288542796e-05 tokens_per_second_per_gpu:4071.800444824158 mfu:19.20737741902878 +Step 27852 | loss:0.5201403498649597 lr:1.2844209510405347e-05 tokens_per_second_per_gpu:5349.21495836207 mfu:25.23315962877735 +Step 27853 | loss:0.49320149421691895 lr:1.2841308011721337e-05 tokens_per_second_per_gpu:3814.5370133016077 mfu:17.99382191886932 +Step 27854 | loss:0.43059012293815613 lr:1.2838406792512576e-05 tokens_per_second_per_gpu:4870.522512443984 mfu:22.975085688028575 +Step 27855 | loss:0.32986223697662354 lr:1.2835505852800906e-05 tokens_per_second_per_gpu:5153.362064277754 mfu:24.309288111432156 +Step 27856 | loss:0.4561750888824463 lr:1.2832605192608143e-05 tokens_per_second_per_gpu:3861.831557462535 mfu:18.216918352957723 +Step 27857 | loss:0.4306854009628296 lr:1.2829704811956094e-05 tokens_per_second_per_gpu:4449.69216282865 mfu:20.989957127831506 +Step 27858 | loss:0.3115258812904358 lr:1.2826804710866575e-05 tokens_per_second_per_gpu:3670.02201461756 mfu:17.31212001327536 +Step 27859 | loss:0.3719710111618042 lr:1.282390488936141e-05 tokens_per_second_per_gpu:4715.743715215854 mfu:22.24496769351757 +Step 27860 | loss:0.44450414180755615 lr:1.2821005347462405e-05 tokens_per_second_per_gpu:4497.230753589868 mfu:21.21420477137213 +Step 27861 | loss:0.40948665142059326 lr:1.281810608519135e-05 tokens_per_second_per_gpu:4606.953591925269 mfu:21.73178612044685 +Step 27862 | loss:0.3661442995071411 lr:1.2815207102570059e-05 tokens_per_second_per_gpu:5349.355737659426 mfu:25.233823708742403 +Step 27863 | loss:0.403608113527298 lr:1.2812308399620359e-05 tokens_per_second_per_gpu:5322.625810142159 mfu:25.107734080047543 +Step 27864 | loss:0.3428259789943695 lr:1.2809409976364017e-05 tokens_per_second_per_gpu:4274.686204994619 mfu:20.164424165633008 +Step 27865 | loss:0.4713652431964874 lr:1.2806511832822848e-05 tokens_per_second_per_gpu:5334.036468381033 mfu:25.161560101819365 +Step 27866 | loss:0.4157920479774475 lr:1.2803613969018663e-05 tokens_per_second_per_gpu:4164.719793814378 mfu:19.645693841891365 +Step 27867 | loss:0.47083228826522827 lr:1.2800716384973227e-05 tokens_per_second_per_gpu:4525.541761817245 mfu:21.34775262753676 +Step 27868 | loss:0.4278985559940338 lr:1.2797819080708367e-05 tokens_per_second_per_gpu:5314.322175271574 mfu:25.0685644176168 +Step 27869 | loss:0.48721325397491455 lr:1.2794922056245844e-05 tokens_per_second_per_gpu:4087.398584561202 mfu:19.28095650548548 +Step 27870 | loss:0.3640387952327728 lr:1.2792025311607476e-05 tokens_per_second_per_gpu:5315.420093486373 mfu:25.073743485160165 +Step 27871 | loss:0.3549163043498993 lr:1.2789128846815013e-05 tokens_per_second_per_gpu:4089.6304513717996 mfu:19.291484602026816 +Step 27872 | loss:0.4473850727081299 lr:1.2786232661890263e-05 tokens_per_second_per_gpu:3539.1708671004453 mfu:16.69487282492923 +Step 27873 | loss:0.43492165207862854 lr:1.2783336756855003e-05 tokens_per_second_per_gpu:4477.067329531569 mfu:21.119090459854245 +Step 27874 | loss:0.348658949136734 lr:1.278044113173103e-05 tokens_per_second_per_gpu:4658.568629527581 mfu:21.97526305925046 +Step 27875 | loss:0.35013359785079956 lr:1.2777545786540113e-05 tokens_per_second_per_gpu:5346.613864555258 mfu:25.220889825498254 +Step 27876 | loss:0.47585615515708923 lr:1.2774650721304004e-05 tokens_per_second_per_gpu:4248.197904215716 mfu:20.039474331488783 +Step 27877 | loss:0.3816925883293152 lr:1.2771755936044504e-05 tokens_per_second_per_gpu:4180.728844818578 mfu:19.721211266903932 +Step 27878 | loss:0.43639546632766724 lr:1.2768861430783385e-05 tokens_per_second_per_gpu:5346.926479401383 mfu:25.222364483065014 +Step 27879 | loss:0.48640310764312744 lr:1.27659672055424e-05 tokens_per_second_per_gpu:5350.472737100779 mfu:25.23909278568702 +Step 27880 | loss:0.33556854724884033 lr:1.2763073260343328e-05 tokens_per_second_per_gpu:4812.504324709006 mfu:22.70140399755911 +Step 27881 | loss:0.46766769886016846 lr:1.2760179595207944e-05 tokens_per_second_per_gpu:5034.339677235155 mfu:23.747839204438964 +Step 27882 | loss:0.5608237981796265 lr:1.2757286210158004e-05 tokens_per_second_per_gpu:4818.003819039793 mfu:22.72734605062792 +Step 27883 | loss:0.4325478672981262 lr:1.2754393105215252e-05 tokens_per_second_per_gpu:4177.005693645927 mfu:19.703648527587443 +Step 27884 | loss:0.54495769739151 lr:1.2751500280401458e-05 tokens_per_second_per_gpu:5371.625366034482 mfu:25.33887334537794 +Step 27885 | loss:0.47714248299598694 lr:1.2748607735738404e-05 tokens_per_second_per_gpu:4072.8460069302964 mfu:19.212309513874725 +Step 27886 | loss:0.4297935962677002 lr:1.2745715471247805e-05 tokens_per_second_per_gpu:4103.9603133363125 mfu:19.359081005839982 +Step 27887 | loss:0.3837187886238098 lr:1.2742823486951432e-05 tokens_per_second_per_gpu:4025.2830687277074 mfu:18.987947019299053 +Step 27888 | loss:0.34897634387016296 lr:1.2739931782871057e-05 tokens_per_second_per_gpu:4231.73546991268 mfu:19.961818220100437 +Step 27889 | loss:0.48561978340148926 lr:1.2737040359028413e-05 tokens_per_second_per_gpu:3805.5169838078773 mfu:17.95127290077167 +Step 27890 | loss:0.37927401065826416 lr:1.2734149215445218e-05 tokens_per_second_per_gpu:3587.253128076779 mfu:16.921685053634256 +Step 27891 | loss:0.28116342425346375 lr:1.273125835214325e-05 tokens_per_second_per_gpu:4096.01632242921 mfu:19.321607845346573 +Step 27892 | loss:0.5432879328727722 lr:1.2728367769144256e-05 tokens_per_second_per_gpu:3780.935596453103 mfu:17.83531830260204 +Step 27893 | loss:0.36646613478660583 lr:1.2725477466469949e-05 tokens_per_second_per_gpu:5350.249807892572 mfu:25.23804119057626 +Step 27894 | loss:0.557643473148346 lr:1.2722587444142082e-05 tokens_per_second_per_gpu:4886.236480907321 mfu:23.049211158349618 +Step 27895 | loss:0.47201576828956604 lr:1.2719697702182403e-05 tokens_per_second_per_gpu:4022.1765218708274 mfu:18.973292907743602 +Step 27896 | loss:0.40444591641426086 lr:1.2716808240612626e-05 tokens_per_second_per_gpu:4536.445253561898 mfu:21.399186258423814 +Step 27897 | loss:0.3555251359939575 lr:1.2713919059454504e-05 tokens_per_second_per_gpu:3774.175370472492 mfu:17.80342916323801 +Step 27898 | loss:0.40182507038116455 lr:1.2711030158729737e-05 tokens_per_second_per_gpu:4231.321680968624 mfu:19.95986630705165 +Step 27899 | loss:0.47512754797935486 lr:1.2708141538460088e-05 tokens_per_second_per_gpu:4993.72219124589 mfu:23.556239592969884 +Step 27900 | loss:0.35072076320648193 lr:1.270525319866725e-05 tokens_per_second_per_gpu:3243.6786960322793 mfu:15.300985837837558 +Step 27901 | loss:0.4255789816379547 lr:1.2702365139372963e-05 tokens_per_second_per_gpu:5174.8806770588435 mfu:24.410795079374484 +Step 27902 | loss:0.4202823042869568 lr:1.2699477360598961e-05 tokens_per_second_per_gpu:4690.361745284378 mfu:22.125236610740174 +Step 27903 | loss:0.4142373502254486 lr:1.2696589862366932e-05 tokens_per_second_per_gpu:4216.125682957591 mfu:19.888184191729 +Step 27904 | loss:0.5196831226348877 lr:1.2693702644698623e-05 tokens_per_second_per_gpu:5337.348791595538 mfu:25.1771848955628 +Step 27905 | loss:0.5066800117492676 lr:1.2690815707615727e-05 tokens_per_second_per_gpu:4818.136649163233 mfu:22.727972632983345 +Step 27906 | loss:0.4757387042045593 lr:1.268792905113998e-05 tokens_per_second_per_gpu:5320.414972257575 mfu:25.09730518053812 +Step 27907 | loss:0.428979754447937 lr:1.2685042675293068e-05 tokens_per_second_per_gpu:3683.5940603509375 mfu:17.376141668629487 +Step 27908 | loss:0.3731231093406677 lr:1.2682156580096704e-05 tokens_per_second_per_gpu:4052.65169359236 mfu:19.117049492354532 +Step 27909 | loss:0.4111780524253845 lr:1.267927076557261e-05 tokens_per_second_per_gpu:4360.832683122755 mfu:20.57079180106821 +Step 27910 | loss:0.36595889925956726 lr:1.2676385231742494e-05 tokens_per_second_per_gpu:5102.570466128067 mfu:24.069695477019835 +Step 27911 | loss:0.4803275763988495 lr:1.2673499978628046e-05 tokens_per_second_per_gpu:4834.485380156224 mfu:22.805092386458423 +Step 27912 | loss:0.43821048736572266 lr:1.2670615006250957e-05 tokens_per_second_per_gpu:5109.200414062509 mfu:24.100970072612355 +Step 27913 | loss:0.3147359788417816 lr:1.266773031463293e-05 tokens_per_second_per_gpu:4426.038379465716 mfu:20.878378195956955 +Step 27914 | loss:0.3956315517425537 lr:1.2664845903795685e-05 tokens_per_second_per_gpu:4152.4788559291455 mfu:19.587951249367357 +Step 27915 | loss:0.3771798312664032 lr:1.2661961773760882e-05 tokens_per_second_per_gpu:5344.225678980284 mfu:25.20962434667462 +Step 27916 | loss:0.4289575517177582 lr:1.2659077924550234e-05 tokens_per_second_per_gpu:4333.708088693277 mfu:20.442840461211336 +Step 27917 | loss:0.37810787558555603 lr:1.2656194356185436e-05 tokens_per_second_per_gpu:4716.3961027793375 mfu:22.248045116963322 +Step 27918 | loss:0.38603222370147705 lr:1.2653311068688161e-05 tokens_per_second_per_gpu:2451.4262352391797 mfu:11.56379580806818 +Step 27919 | loss:0.4782642424106598 lr:1.2650428062080088e-05 tokens_per_second_per_gpu:5354.712625442891 mfu:25.25909306239631 +Step 27920 | loss:0.43309804797172546 lr:1.2647545336382916e-05 tokens_per_second_per_gpu:4882.462659795304 mfu:23.031409400283955 +Step 27921 | loss:0.39953547716140747 lr:1.2644662891618331e-05 tokens_per_second_per_gpu:3879.1684920913704 mfu:18.298699631588008 +Step 27922 | loss:0.42165058851242065 lr:1.264178072780799e-05 tokens_per_second_per_gpu:5192.330331161836 mfu:24.493108074996286 +Step 27923 | loss:0.39419570565223694 lr:1.2638898844973585e-05 tokens_per_second_per_gpu:4248.232502016893 mfu:20.039637535219928 +Step 27924 | loss:0.46348103880882263 lr:1.26360172431368e-05 tokens_per_second_per_gpu:5328.471397847359 mfu:25.13530871461308 +Step 27925 | loss:0.421111136674881 lr:1.2633135922319284e-05 tokens_per_second_per_gpu:5342.919096708462 mfu:25.20346097517252 +Step 27926 | loss:0.44644156098365784 lr:1.2630254882542736e-05 tokens_per_second_per_gpu:4610.701494011041 mfu:21.74946561404347 +Step 27927 | loss:0.42533421516418457 lr:1.2627374123828794e-05 tokens_per_second_per_gpu:3254.4325648679956 mfu:15.351713671934855 +Step 27928 | loss:0.38340863585472107 lr:1.2624493646199154e-05 tokens_per_second_per_gpu:4370.027510021384 mfu:20.61416536834828 +Step 27929 | loss:0.3398621380329132 lr:1.2621613449675452e-05 tokens_per_second_per_gpu:4851.7245008735545 mfu:22.886412260179362 +Step 27930 | loss:0.5195015668869019 lr:1.2618733534279364e-05 tokens_per_second_per_gpu:3498.898713283834 mfu:16.504902204238306 +Step 27931 | loss:0.3298274576663971 lr:1.261585390003257e-05 tokens_per_second_per_gpu:5356.086308087442 mfu:25.265572957805084 +Step 27932 | loss:0.3963563144207001 lr:1.261297454695669e-05 tokens_per_second_per_gpu:4117.03062194972 mfu:19.42073588159403 +Step 27933 | loss:0.3541978895664215 lr:1.2610095475073414e-05 tokens_per_second_per_gpu:3080.28816939129 mfu:14.53024484637355 +Step 27934 | loss:0.4377744197845459 lr:1.2607216684404365e-05 tokens_per_second_per_gpu:4906.048063711222 mfu:23.142665774640204 +Step 27935 | loss:0.5053020119667053 lr:1.2604338174971231e-05 tokens_per_second_per_gpu:4169.102354189982 mfu:19.666367126924996 +Step 27936 | loss:0.41222772002220154 lr:1.2601459946795625e-05 tokens_per_second_per_gpu:5354.416815789417 mfu:25.25769767779054 +Step 27937 | loss:0.3742334842681885 lr:1.2598581999899206e-05 tokens_per_second_per_gpu:4712.043295289393 mfu:22.227512181367715 +Step 27938 | loss:0.36179280281066895 lr:1.2595704334303643e-05 tokens_per_second_per_gpu:5332.762838551841 mfu:25.155552172611408 +Step 27939 | loss:0.4775058627128601 lr:1.2592826950030545e-05 tokens_per_second_per_gpu:5349.934511942887 mfu:25.236553885786563 +Step 27940 | loss:0.26781949400901794 lr:1.2589949847101584e-05 tokens_per_second_per_gpu:3750.3649124463914 mfu:17.691111170245843 +Step 27941 | loss:0.3680952191352844 lr:1.2587073025538365e-05 tokens_per_second_per_gpu:5103.609369347383 mfu:24.074596160760514 +Step 27942 | loss:0.34116679430007935 lr:1.2584196485362565e-05 tokens_per_second_per_gpu:4780.409295008084 mfu:22.55000626648313 +Step 27943 | loss:0.4734829366207123 lr:1.2581320226595778e-05 tokens_per_second_per_gpu:5289.829583205301 mfu:24.953028681973066 +Step 27944 | loss:0.38822659850120544 lr:1.2578444249259658e-05 tokens_per_second_per_gpu:4138.416421323395 mfu:19.521616346033284 +Step 27945 | loss:0.38347211480140686 lr:1.2575568553375833e-05 tokens_per_second_per_gpu:5364.395455531901 mfu:25.304768623987332 +Step 27946 | loss:0.4103623926639557 lr:1.2572693138965946e-05 tokens_per_second_per_gpu:4633.936002721791 mfu:21.859066755848037 +Step 27947 | loss:0.4037737250328064 lr:1.2569818006051615e-05 tokens_per_second_per_gpu:4689.592366145256 mfu:22.121607318071316 +Step 27948 | loss:0.42566707730293274 lr:1.2566943154654438e-05 tokens_per_second_per_gpu:3397.550588093139 mfu:16.02682580594057 +Step 27949 | loss:0.4274648427963257 lr:1.2564068584796073e-05 tokens_per_second_per_gpu:5109.03590459172 mfu:24.100194053370437 +Step 27950 | loss:0.5069437026977539 lr:1.2561194296498113e-05 tokens_per_second_per_gpu:5344.032883442892 mfu:25.208714897230436 +Step 27951 | loss:0.48105141520500183 lr:1.2558320289782188e-05 tokens_per_second_per_gpu:5301.250381458146 mfu:25.006902535920855 +Step 27952 | loss:0.5169278383255005 lr:1.2555446564669914e-05 tokens_per_second_per_gpu:4389.584503525639 mfu:20.70641904347502 +Step 27953 | loss:0.28992754220962524 lr:1.255257312118292e-05 tokens_per_second_per_gpu:4043.2392088556608 mfu:19.07264919591603 +Step 27954 | loss:0.3611239492893219 lr:1.2549699959342787e-05 tokens_per_second_per_gpu:4303.065161915853 mfu:20.298292547379805 +Step 27955 | loss:0.47891223430633545 lr:1.2546827079171147e-05 tokens_per_second_per_gpu:4430.057338386511 mfu:20.897336310891443 +Step 27956 | loss:0.3249904215335846 lr:1.254395448068959e-05 tokens_per_second_per_gpu:4695.592218657405 mfu:22.14990964605152 +Step 27957 | loss:0.34324324131011963 lr:1.2541082163919749e-05 tokens_per_second_per_gpu:4986.1858015594735 mfu:23.520689156978385 +Step 27958 | loss:0.6063579320907593 lr:1.2538210128883189e-05 tokens_per_second_per_gpu:4640.283483503697 mfu:21.889008905688186 +Step 27959 | loss:0.5650920271873474 lr:1.2535338375601535e-05 tokens_per_second_per_gpu:4409.0473057488625 mfu:20.798228402258445 +Step 27960 | loss:0.4984021484851837 lr:1.253246690409639e-05 tokens_per_second_per_gpu:5040.19852372697 mfu:23.77547638296314 +Step 27961 | loss:0.39099428057670593 lr:1.2529595714389337e-05 tokens_per_second_per_gpu:3923.18833462701 mfu:18.506348739388223 +Step 27962 | loss:0.44942981004714966 lr:1.2526724806501982e-05 tokens_per_second_per_gpu:4954.239930806899 mfu:23.3699950341109 +Step 27963 | loss:0.4881531894207001 lr:1.2523854180455902e-05 tokens_per_second_per_gpu:5295.287349302895 mfu:24.978773895846338 +Step 27964 | loss:0.4877142906188965 lr:1.252098383627271e-05 tokens_per_second_per_gpu:5348.234406389797 mfu:25.228534197824718 +Step 27965 | loss:0.5102894306182861 lr:1.251811377397396e-05 tokens_per_second_per_gpu:3649.963965970467 mfu:17.217502775551033 +Step 27966 | loss:0.5465654134750366 lr:1.2515243993581266e-05 tokens_per_second_per_gpu:5077.437823182353 mfu:23.951140512173843 +Step 27967 | loss:0.4775841534137726 lr:1.2512374495116219e-05 tokens_per_second_per_gpu:4028.846661278092 mfu:19.004757093370802 +Step 27968 | loss:0.5070304274559021 lr:1.250950527860037e-05 tokens_per_second_per_gpu:4702.240123229601 mfu:22.181268946168043 +Step 27969 | loss:0.42579278349876404 lr:1.2506636344055334e-05 tokens_per_second_per_gpu:5345.858308371077 mfu:25.217325738066293 +Step 27970 | loss:0.3052574396133423 lr:1.250376769150265e-05 tokens_per_second_per_gpu:5318.082916551609 mfu:25.08630447588335 +Step 27971 | loss:0.3417750895023346 lr:1.2500899320963933e-05 tokens_per_second_per_gpu:4504.523811508515 mfu:21.248607369898377 +Step 27972 | loss:0.46007490158081055 lr:1.2498031232460723e-05 tokens_per_second_per_gpu:4509.371367546471 mfu:21.271474118807767 +Step 27973 | loss:0.37856942415237427 lr:1.24951634260146e-05 tokens_per_second_per_gpu:4241.559930073968 mfu:20.008161874906985 +Step 27974 | loss:0.33125847578048706 lr:1.249229590164716e-05 tokens_per_second_per_gpu:4880.196320418774 mfu:23.020718690767495 +Step 27975 | loss:0.5097303986549377 lr:1.2489428659379926e-05 tokens_per_second_per_gpu:5352.772585660318 mfu:25.249941563736996 +Step 27976 | loss:0.3700764775276184 lr:1.2486561699234505e-05 tokens_per_second_per_gpu:4433.891540208007 mfu:20.91542289506573 +Step 27977 | loss:0.38320741057395935 lr:1.2483695021232428e-05 tokens_per_second_per_gpu:4514.9223923225245 mfu:21.297659249778896 +Step 27978 | loss:0.35671165585517883 lr:1.248082862539528e-05 tokens_per_second_per_gpu:5360.62993513004 mfu:25.28700602178773 +Step 27979 | loss:0.3545505404472351 lr:1.247796251174459e-05 tokens_per_second_per_gpu:4898.757313737509 mfu:23.108274063083982 +Step 27980 | loss:0.31546759605407715 lr:1.2475096680301934e-05 tokens_per_second_per_gpu:4019.5746367551865 mfu:18.961019371725747 +Step 27981 | loss:0.416362464427948 lr:1.2472231131088862e-05 tokens_per_second_per_gpu:5359.304750455161 mfu:25.280754899576504 +Step 27982 | loss:0.46369555592536926 lr:1.2469365864126947e-05 tokens_per_second_per_gpu:3563.8302320017024 mfu:16.81119525648996 +Step 27983 | loss:0.4758727252483368 lr:1.246650087943772e-05 tokens_per_second_per_gpu:4289.931989644701 mfu:20.23634113302653 +Step 27984 | loss:0.44998830556869507 lr:1.2463636177042714e-05 tokens_per_second_per_gpu:4156.494186917939 mfu:19.606892250727608 +Step 27985 | loss:0.4169577360153198 lr:1.2460771756963502e-05 tokens_per_second_per_gpu:5259.0940937584355 mfu:24.808044134236958 +Step 27986 | loss:0.4709809720516205 lr:1.2457907619221604e-05 tokens_per_second_per_gpu:4368.668778127286 mfu:20.607755998180256 +Step 27987 | loss:0.31329095363616943 lr:1.245504376383857e-05 tokens_per_second_per_gpu:4654.93217655028 mfu:21.958109290113676 +Step 27988 | loss:0.46923860907554626 lr:1.2452180190835943e-05 tokens_per_second_per_gpu:4098.718979036471 mfu:19.334356737683784 +Step 27989 | loss:0.4583253264427185 lr:1.2449316900235275e-05 tokens_per_second_per_gpu:5304.635432617861 mfu:25.022870399790513 +Step 27990 | loss:0.3766528367996216 lr:1.2446453892058074e-05 tokens_per_second_per_gpu:5393.437664646322 mfu:25.44176567949189 +Step 27991 | loss:0.390970915555954 lr:1.2443591166325901e-05 tokens_per_second_per_gpu:5332.973091582014 mfu:25.1565439720276 +Step 27992 | loss:0.4238128364086151 lr:1.2440728723060252e-05 tokens_per_second_per_gpu:5327.222332362139 mfu:25.12941665960675 +Step 27993 | loss:0.3204018771648407 lr:1.243786656228269e-05 tokens_per_second_per_gpu:4639.38584752682 mfu:21.884774603632764 +Step 27994 | loss:0.4151977598667145 lr:1.2435004684014712e-05 tokens_per_second_per_gpu:3290.2571201892183 mfu:15.520704211684482 +Step 27995 | loss:0.36677759885787964 lr:1.2432143088277859e-05 tokens_per_second_per_gpu:4702.2544711654255 mfu:22.181336627829342 +Step 27996 | loss:0.33963435888290405 lr:1.2429281775093666e-05 tokens_per_second_per_gpu:5219.171094205626 mfu:24.619720533780246 +Step 27997 | loss:0.3731673061847687 lr:1.2426420744483624e-05 tokens_per_second_per_gpu:4923.6938715402885 mfu:23.225904060855175 +Step 27998 | loss:0.5263491868972778 lr:1.242355999646928e-05 tokens_per_second_per_gpu:4143.77145719186 mfu:19.546876963888163 +Step 27999 | loss:0.4891919493675232 lr:1.2420699531072121e-05 tokens_per_second_per_gpu:3663.400805369094 mfu:17.28088663955556 +Step 28000 | loss:0.4241676330566406 lr:1.2417839348313692e-05 tokens_per_second_per_gpu:3770.826450132191 mfu:17.78763173460808 +Step 28001 | loss:0.5019065141677856 lr:1.2414979448215475e-05 tokens_per_second_per_gpu:3888.4908672240476 mfu:18.342674865650057 +Step 28002 | loss:0.4311322271823883 lr:1.2412119830798991e-05 tokens_per_second_per_gpu:4434.11156715373 mfu:20.916460799709167 +Step 28003 | loss:0.41307640075683594 lr:1.2409260496085761e-05 tokens_per_second_per_gpu:4864.53867994715 mfu:22.946858929193905 +Step 28004 | loss:0.312981516122818 lr:1.2406401444097272e-05 tokens_per_second_per_gpu:5356.794519686252 mfu:25.268913712750027 +Step 28005 | loss:0.4911029636859894 lr:1.2403542674855046e-05 tokens_per_second_per_gpu:3184.1152939373546 mfu:15.020015107591473 +Step 28006 | loss:0.4155605435371399 lr:1.2400684188380557e-05 tokens_per_second_per_gpu:4345.817812822812 mfu:20.499964096062342 +Step 28007 | loss:0.33427485823631287 lr:1.2397825984695339e-05 tokens_per_second_per_gpu:5077.479381430921 mfu:23.951336549522907 +Step 28008 | loss:0.4076678454875946 lr:1.239496806382085e-05 tokens_per_second_per_gpu:4595.515374922528 mfu:21.67783009928386 +Step 28009 | loss:0.3694249391555786 lr:1.2392110425778607e-05 tokens_per_second_per_gpu:5033.690295845368 mfu:23.74477596162745 +Step 28010 | loss:0.3024418354034424 lr:1.2389253070590118e-05 tokens_per_second_per_gpu:3376.3409693214044 mfu:15.926776415460182 +Step 28011 | loss:0.4360876679420471 lr:1.2386395998276835e-05 tokens_per_second_per_gpu:5371.063772852783 mfu:25.336224214522186 +Step 28012 | loss:0.3450523316860199 lr:1.2383539208860284e-05 tokens_per_second_per_gpu:5223.797225944858 mfu:24.641542786493666 +Step 28013 | loss:0.499941349029541 lr:1.2380682702361923e-05 tokens_per_second_per_gpu:4267.254055609427 mfu:20.12936544892761 +Step 28014 | loss:0.5578166246414185 lr:1.237782647880326e-05 tokens_per_second_per_gpu:4392.222065264784 mfu:20.718860872213007 +Step 28015 | loss:0.4397377669811249 lr:1.2374970538205748e-05 tokens_per_second_per_gpu:4933.919917256547 mfu:23.274142063242078 +Step 28016 | loss:0.4897548258304596 lr:1.237211488059089e-05 tokens_per_second_per_gpu:3905.0102563004752 mfu:18.420599642421976 +Step 28017 | loss:0.297524094581604 lr:1.2369259505980169e-05 tokens_per_second_per_gpu:4115.557785458507 mfu:19.41378826057316 +Step 28018 | loss:0.4219208061695099 lr:1.2366404414395033e-05 tokens_per_second_per_gpu:5351.6794888931945 mfu:25.24478523978493 +Step 28019 | loss:0.323682963848114 lr:1.2363549605856967e-05 tokens_per_second_per_gpu:4135.1162033828 mfu:19.506048654932254 +Step 28020 | loss:0.4188723862171173 lr:1.2360695080387468e-05 tokens_per_second_per_gpu:4118.859069094156 mfu:19.429360978739588 +Step 28021 | loss:0.3759705126285553 lr:1.2357840838007984e-05 tokens_per_second_per_gpu:4723.966131953158 mfu:22.283754236156394 +Step 28022 | loss:0.5604080557823181 lr:1.2354986878739967e-05 tokens_per_second_per_gpu:5013.009405964158 mfu:23.647220675534115 +Step 28023 | loss:0.3394133746623993 lr:1.2352133202604892e-05 tokens_per_second_per_gpu:5318.500263633626 mfu:25.088273173275976 +Step 28024 | loss:0.3935258686542511 lr:1.2349279809624232e-05 tokens_per_second_per_gpu:5374.3280104410505 mfu:25.351622180162412 +Step 28025 | loss:0.49369415640830994 lr:1.2346426699819458e-05 tokens_per_second_per_gpu:4910.633452893674 mfu:23.164295837761898 +Step 28026 | loss:0.39756709337234497 lr:1.2343573873211994e-05 tokens_per_second_per_gpu:4699.642553502307 mfu:22.169015766573306 +Step 28027 | loss:0.29681098461151123 lr:1.2340721329823335e-05 tokens_per_second_per_gpu:5318.674475467271 mfu:25.089094960218883 +Step 28028 | loss:0.36505553126335144 lr:1.2337869069674902e-05 tokens_per_second_per_gpu:3531.8192944927405 mfu:16.66019420263106 +Step 28029 | loss:0.406035453081131 lr:1.2335017092788176e-05 tokens_per_second_per_gpu:3868.4246534696263 mfu:18.24801911172135 +Step 28030 | loss:0.36117690801620483 lr:1.2332165399184576e-05 tokens_per_second_per_gpu:5254.860647041548 mfu:24.78807424377291 +Step 28031 | loss:0.3237917125225067 lr:1.2329313988885566e-05 tokens_per_second_per_gpu:4194.437920720287 mfu:19.785879316941838 +Step 28032 | loss:0.33887431025505066 lr:1.2326462861912608e-05 tokens_per_second_per_gpu:4412.357428330306 mfu:20.813842815240264 +Step 28033 | loss:0.2846488952636719 lr:1.2323612018287112e-05 tokens_per_second_per_gpu:5206.4416540409275 mfu:24.559673592656914 +Step 28034 | loss:0.40737882256507874 lr:1.2320761458030555e-05 tokens_per_second_per_gpu:5337.458052259131 mfu:25.177700296754757 +Step 28035 | loss:0.4358839690685272 lr:1.2317911181164348e-05 tokens_per_second_per_gpu:5229.2630746926925 mfu:24.66732612760716 +Step 28036 | loss:0.4175415635108948 lr:1.2315061187709948e-05 tokens_per_second_per_gpu:4627.928026705958 mfu:21.830726107914945 +Step 28037 | loss:0.42860177159309387 lr:1.2312211477688773e-05 tokens_per_second_per_gpu:4845.641965512459 mfu:22.85771990309353 +Step 28038 | loss:0.4069869816303253 lr:1.2309362051122259e-05 tokens_per_second_per_gpu:5374.907371718408 mfu:25.354355126156936 +Step 28039 | loss:0.4060148596763611 lr:1.2306512908031858e-05 tokens_per_second_per_gpu:3500.19851622194 mfu:16.5110335964666 +Step 28040 | loss:0.3018649220466614 lr:1.2303664048438974e-05 tokens_per_second_per_gpu:5323.4112828370935 mfu:25.111439288765013 +Step 28041 | loss:0.38757753372192383 lr:1.2300815472365052e-05 tokens_per_second_per_gpu:5315.111018515799 mfu:25.07228552579034 +Step 28042 | loss:0.374778151512146 lr:1.2297967179831493e-05 tokens_per_second_per_gpu:4629.715718890052 mfu:21.839158956959054 +Step 28043 | loss:0.38419148325920105 lr:1.2295119170859748e-05 tokens_per_second_per_gpu:4536.762308481388 mfu:21.400681860574192 +Step 28044 | loss:0.3506777584552765 lr:1.229227144547121e-05 tokens_per_second_per_gpu:5341.111349249805 mfu:25.19493352197651 +Step 28045 | loss:0.4655010402202606 lr:1.2289424003687306e-05 tokens_per_second_per_gpu:4399.388215893373 mfu:20.752664827399435 +Step 28046 | loss:0.33373507857322693 lr:1.2286576845529474e-05 tokens_per_second_per_gpu:4279.859139106029 mfu:20.188825778430093 +Step 28047 | loss:0.4600333273410797 lr:1.2283729971019097e-05 tokens_per_second_per_gpu:4184.5815882484 mfu:19.739385315008214 +Step 28048 | loss:0.30191341042518616 lr:1.2280883380177594e-05 tokens_per_second_per_gpu:5329.519981060531 mfu:25.140255060536354 +Step 28049 | loss:0.391579270362854 lr:1.22780370730264e-05 tokens_per_second_per_gpu:4333.921455904683 mfu:20.44384695075174 +Step 28050 | loss:0.5315473079681396 lr:1.2275191049586899e-05 tokens_per_second_per_gpu:5345.442858103346 mfu:25.215365988270282 +Step 28051 | loss:0.32842034101486206 lr:1.227234530988049e-05 tokens_per_second_per_gpu:3977.448751975992 mfu:18.7623043858052 +Step 28052 | loss:0.5225338339805603 lr:1.2269499853928579e-05 tokens_per_second_per_gpu:5355.474128353298 mfu:25.262685201550493 +Step 28053 | loss:0.3629230260848999 lr:1.2266654681752592e-05 tokens_per_second_per_gpu:3740.1407119114738 mfu:17.642881871894147 +Step 28054 | loss:0.3688352704048157 lr:1.2263809793373899e-05 tokens_per_second_per_gpu:4503.068819386374 mfu:21.241743923810986 +Step 28055 | loss:0.4941689968109131 lr:1.2260965188813906e-05 tokens_per_second_per_gpu:4814.880962679612 mfu:22.71261500436271 +Step 28056 | loss:0.3232784867286682 lr:1.225812086809402e-05 tokens_per_second_per_gpu:2731.84349657425 mfu:12.88657187390402 +Step 28057 | loss:0.30134817957878113 lr:1.2255276831235624e-05 tokens_per_second_per_gpu:4327.553474808326 mfu:20.413808097430618 +Step 28058 | loss:0.3620549142360687 lr:1.2252433078260095e-05 tokens_per_second_per_gpu:4025.03627026402 mfu:18.986782828837683 +Step 28059 | loss:0.37936297059059143 lr:1.2249589609188834e-05 tokens_per_second_per_gpu:3280.2581787609943 mfu:15.473537498972222 +Step 28060 | loss:0.46407562494277954 lr:1.2246746424043221e-05 tokens_per_second_per_gpu:4391.51657372187 mfu:20.715532948235815 +Step 28061 | loss:0.42551034688949585 lr:1.2243903522844663e-05 tokens_per_second_per_gpu:4030.3729577364716 mfu:19.01195689417801 +Step 28062 | loss:0.48107704520225525 lr:1.2241060905614516e-05 tokens_per_second_per_gpu:4769.621187753265 mfu:22.49911692392125 +Step 28063 | loss:0.5005394816398621 lr:1.2238218572374178e-05 tokens_per_second_per_gpu:3922.6627473767444 mfu:18.503869454654286 +Step 28064 | loss:0.33685770630836487 lr:1.2235376523144998e-05 tokens_per_second_per_gpu:4216.165697216353 mfu:19.888372945814712 +Step 28065 | loss:0.4893570840358734 lr:1.2232534757948388e-05 tokens_per_second_per_gpu:4418.063885657381 mfu:20.840761148074094 +Step 28066 | loss:0.49828678369522095 lr:1.2229693276805686e-05 tokens_per_second_per_gpu:5336.358186218905 mfu:25.172512040986515 +Step 28067 | loss:0.387626975774765 lr:1.2226852079738283e-05 tokens_per_second_per_gpu:5322.521589113277 mfu:25.107242451672363 +Step 28068 | loss:0.4115086495876312 lr:1.2224011166767562e-05 tokens_per_second_per_gpu:4768.722915684166 mfu:22.49487962131029 +Step 28069 | loss:0.4138650596141815 lr:1.2221170537914855e-05 tokens_per_second_per_gpu:4311.6712740026605 mfu:20.338889046447406 +Step 28070 | loss:0.4337504506111145 lr:1.2218330193201561e-05 tokens_per_second_per_gpu:4612.600739322975 mfu:21.758424678224525 +Step 28071 | loss:0.3811590075492859 lr:1.2215490132649015e-05 tokens_per_second_per_gpu:4472.306036546507 mfu:21.096630628483354 +Step 28072 | loss:0.6249457597732544 lr:1.22126503562786e-05 tokens_per_second_per_gpu:4541.526265293735 mfu:21.42315425767288 +Step 28073 | loss:0.43067115545272827 lr:1.2209810864111647e-05 tokens_per_second_per_gpu:3737.8860148143403 mfu:17.632246080995976 +Step 28074 | loss:0.35394617915153503 lr:1.220697165616953e-05 tokens_per_second_per_gpu:4718.270321939952 mfu:22.2568861285187 +Step 28075 | loss:0.5001642107963562 lr:1.2204132732473621e-05 tokens_per_second_per_gpu:5367.554095022539 mfu:25.31966846538414 +Step 28076 | loss:0.4013034403324127 lr:1.2201294093045234e-05 tokens_per_second_per_gpu:5337.303483544263 mfu:25.176971169754584 +Step 28077 | loss:0.3183085024356842 lr:1.2198455737905751e-05 tokens_per_second_per_gpu:4244.803751807378 mfu:20.023463535476537 +Step 28078 | loss:0.3996102213859558 lr:1.2195617667076492e-05 tokens_per_second_per_gpu:4148.124929136109 mfu:19.567413033828092 +Step 28079 | loss:0.4550601840019226 lr:1.219277988057883e-05 tokens_per_second_per_gpu:4029.326088295315 mfu:19.00701863240973 +Step 28080 | loss:0.3495904207229614 lr:1.2189942378434083e-05 tokens_per_second_per_gpu:4707.473274393953 mfu:22.205954613077953 +Step 28081 | loss:0.39852607250213623 lr:1.2187105160663603e-05 tokens_per_second_per_gpu:3076.4913852231175 mfu:14.512334767654105 +Step 28082 | loss:0.40481701493263245 lr:1.2184268227288742e-05 tokens_per_second_per_gpu:4238.631880683803 mfu:19.994349766356834 +Step 28083 | loss:0.4272686839103699 lr:1.2181431578330815e-05 tokens_per_second_per_gpu:4335.4201684891505 mfu:20.450916633719405 +Step 28084 | loss:0.4779762327671051 lr:1.2178595213811162e-05 tokens_per_second_per_gpu:5321.483303963717 mfu:25.10234468347218 +Step 28085 | loss:0.40936174988746643 lr:1.2175759133751136e-05 tokens_per_second_per_gpu:4160.943061882988 mfu:19.627878352994085 +Step 28086 | loss:0.415621280670166 lr:1.2172923338172054e-05 tokens_per_second_per_gpu:4700.422946367358 mfu:22.172697012866507 +Step 28087 | loss:0.3148282766342163 lr:1.2170087827095228e-05 tokens_per_second_per_gpu:5314.8560186202285 mfu:25.071082647775466 +Step 28088 | loss:0.45252934098243713 lr:1.2167252600541996e-05 tokens_per_second_per_gpu:5301.592596676665 mfu:25.008516823494546 +Step 28089 | loss:0.4012059271335602 lr:1.21644176585337e-05 tokens_per_second_per_gpu:5296.4365816437185 mfu:24.98419502087726 +Step 28090 | loss:0.4449217915534973 lr:1.2161583001091636e-05 tokens_per_second_per_gpu:5065.765743801789 mfu:23.896081322273627 +Step 28091 | loss:0.3787788450717926 lr:1.2158748628237133e-05 tokens_per_second_per_gpu:5322.958105028049 mfu:25.109301571719644 +Step 28092 | loss:0.37487542629241943 lr:1.215591453999152e-05 tokens_per_second_per_gpu:5337.817674696993 mfu:25.179396697152068 +Step 28093 | loss:0.414314329624176 lr:1.2153080736376105e-05 tokens_per_second_per_gpu:5317.505900197655 mfu:25.083582591292608 +Step 28094 | loss:0.41913196444511414 lr:1.2150247217412186e-05 tokens_per_second_per_gpu:4977.96006359183 mfu:23.481886947529393 +Step 28095 | loss:0.49668359756469727 lr:1.2147413983121086e-05 tokens_per_second_per_gpu:5078.955661905992 mfu:23.95830041640324 +Step 28096 | loss:0.2870486080646515 lr:1.2144581033524116e-05 tokens_per_second_per_gpu:4327.784763988416 mfu:20.4148991279551 +Step 28097 | loss:0.42748361825942993 lr:1.2141748368642597e-05 tokens_per_second_per_gpu:5155.50781419809 mfu:24.319409979909068 +Step 28098 | loss:0.3929917812347412 lr:1.2138915988497806e-05 tokens_per_second_per_gpu:4223.446422036948 mfu:19.92271736701321 +Step 28099 | loss:0.4123440086841583 lr:1.2136083893111073e-05 tokens_per_second_per_gpu:3766.861286859706 mfu:17.768927382925156 +Step 28100 | loss:0.47795429825782776 lr:1.2133252082503665e-05 tokens_per_second_per_gpu:5329.103343646675 mfu:25.138289710769524 +Step 28101 | loss:0.31069833040237427 lr:1.213042055669692e-05 tokens_per_second_per_gpu:4677.256658195501 mfu:22.063417678983516 +Step 28102 | loss:0.33987295627593994 lr:1.21275893157121e-05 tokens_per_second_per_gpu:5301.957986716573 mfu:25.010240430654758 +Step 28103 | loss:0.40481096506118774 lr:1.2124758359570509e-05 tokens_per_second_per_gpu:4756.474470384273 mfu:22.437101615030436 +Step 28104 | loss:0.33702215552330017 lr:1.2121927688293461e-05 tokens_per_second_per_gpu:4746.383560598403 mfu:22.389501071883743 +Step 28105 | loss:0.49197322130203247 lr:1.2119097301902216e-05 tokens_per_second_per_gpu:3945.5302110720672 mfu:18.61173918249675 +Step 28106 | loss:0.44554993510246277 lr:1.2116267200418086e-05 tokens_per_second_per_gpu:4421.890303304236 mfu:20.85881101296848 +Step 28107 | loss:0.36833521723747253 lr:1.2113437383862324e-05 tokens_per_second_per_gpu:4726.140487697122 mfu:22.29401104318401 +Step 28108 | loss:0.2399350106716156 lr:1.2110607852256255e-05 tokens_per_second_per_gpu:4204.300649709118 mfu:19.83240349233724 +Step 28109 | loss:0.3525625467300415 lr:1.2107778605621123e-05 tokens_per_second_per_gpu:4021.3194832426407 mfu:18.96925011030858 +Step 28110 | loss:0.3786677420139313 lr:1.2104949643978225e-05 tokens_per_second_per_gpu:3117.0281729926987 mfu:14.703553711852399 +Step 28111 | loss:0.45131489634513855 lr:1.2102120967348851e-05 tokens_per_second_per_gpu:4312.279328465178 mfu:20.341757343091892 +Step 28112 | loss:0.4176066815853119 lr:1.2099292575754245e-05 tokens_per_second_per_gpu:4921.430697352906 mfu:23.215228282075955 +Step 28113 | loss:0.4894300401210785 lr:1.209646446921569e-05 tokens_per_second_per_gpu:5326.037993573843 mfu:25.123829931473324 +Step 28114 | loss:0.4932597577571869 lr:1.2093636647754486e-05 tokens_per_second_per_gpu:4708.124869625705 mfu:22.20902829895974 +Step 28115 | loss:0.43341168761253357 lr:1.209080911139187e-05 tokens_per_second_per_gpu:5354.910039115956 mfu:25.260024296374404 +Step 28116 | loss:0.44611918926239014 lr:1.2087981860149105e-05 tokens_per_second_per_gpu:5345.169506023627 mfu:25.21407653987157 +Step 28117 | loss:0.4805034101009369 lr:1.2085154894047469e-05 tokens_per_second_per_gpu:4969.814792472509 mfu:23.443464313933 +Step 28118 | loss:0.43282830715179443 lr:1.2082328213108229e-05 tokens_per_second_per_gpu:4436.921988198545 mfu:20.929718035284836 +Step 28119 | loss:0.42763790488243103 lr:1.207950181735263e-05 tokens_per_second_per_gpu:5369.7759729579375 mfu:25.330149442696886 +Step 28120 | loss:0.38667067885398865 lr:1.2076675706801932e-05 tokens_per_second_per_gpu:4293.2037179898225 mfu:20.251774433844886 +Step 28121 | loss:0.4002964496612549 lr:1.2073849881477412e-05 tokens_per_second_per_gpu:5354.937198122641 mfu:25.260152410042743 +Step 28122 | loss:0.3462035059928894 lr:1.20710243414003e-05 tokens_per_second_per_gpu:3685.2069150945313 mfu:17.383749779636414 +Step 28123 | loss:0.46973270177841187 lr:1.2068199086591848e-05 tokens_per_second_per_gpu:5092.636097904183 mfu:24.02283336712977 +Step 28124 | loss:0.28828030824661255 lr:1.2065374117073303e-05 tokens_per_second_per_gpu:4346.811316329059 mfu:20.504650621610487 +Step 28125 | loss:0.5335277915000916 lr:1.2062549432865938e-05 tokens_per_second_per_gpu:4259.721384314055 mfu:20.09383255322109 +Step 28126 | loss:0.3030218482017517 lr:1.2059725033990965e-05 tokens_per_second_per_gpu:3568.8948811807363 mfu:16.835086070785717 +Step 28127 | loss:0.4284837245941162 lr:1.2056900920469638e-05 tokens_per_second_per_gpu:4576.8081208250005 mfu:21.589584789919147 +Step 28128 | loss:0.5118028521537781 lr:1.2054077092323217e-05 tokens_per_second_per_gpu:5359.635493294961 mfu:25.282315069982385 +Step 28129 | loss:0.34239253401756287 lr:1.2051253549572927e-05 tokens_per_second_per_gpu:5195.50654865032 mfu:24.50809083480805 +Step 28130 | loss:0.3813472092151642 lr:1.2048430292239987e-05 tokens_per_second_per_gpu:4185.066783885579 mfu:19.7416740656117 +Step 28131 | loss:0.42899850010871887 lr:1.2045607320345643e-05 tokens_per_second_per_gpu:5358.118359529244 mfu:25.27515849115979 +Step 28132 | loss:0.3316909670829773 lr:1.204278463391113e-05 tokens_per_second_per_gpu:5330.866094199197 mfu:25.146604905882025 +Step 28133 | loss:0.4463670253753662 lr:1.203996223295769e-05 tokens_per_second_per_gpu:4637.263382543093 mfu:21.874762574174657 +Step 28134 | loss:0.4692566394805908 lr:1.203714011750653e-05 tokens_per_second_per_gpu:5368.037640218399 mfu:25.321949430574335 +Step 28135 | loss:0.4574899673461914 lr:1.2034318287578889e-05 tokens_per_second_per_gpu:3720.0442674583724 mfu:17.548083514601878 +Step 28136 | loss:0.47293856739997864 lr:1.2031496743195975e-05 tokens_per_second_per_gpu:4684.011130903925 mfu:22.095279679181033 +Step 28137 | loss:0.5788901448249817 lr:1.2028675484379032e-05 tokens_per_second_per_gpu:4630.082225594346 mfu:21.8408878316157 +Step 28138 | loss:0.31162089109420776 lr:1.2025854511149254e-05 tokens_per_second_per_gpu:3818.0526527121965 mfu:18.010405789799 +Step 28139 | loss:0.4374430179595947 lr:1.2023033823527863e-05 tokens_per_second_per_gpu:3951.3711643251245 mfu:18.639291955560807 +Step 28140 | loss:0.45485806465148926 lr:1.2020213421536103e-05 tokens_per_second_per_gpu:5241.268368574518 mfu:24.723957147162086 +Step 28141 | loss:0.4239415228366852 lr:1.2017393305195141e-05 tokens_per_second_per_gpu:4524.674728403506 mfu:21.343662682993 +Step 28142 | loss:0.4044133722782135 lr:1.201457347452622e-05 tokens_per_second_per_gpu:3482.8849198312555 mfu:16.42936240828759 +Step 28143 | loss:0.3806288540363312 lr:1.2011753929550545e-05 tokens_per_second_per_gpu:3283.0393317308617 mfu:15.486656672044717 +Step 28144 | loss:0.36567872762680054 lr:1.2008934670289324e-05 tokens_per_second_per_gpu:4188.782313918374 mfu:19.75920085471119 +Step 28145 | loss:0.4268818199634552 lr:1.2006115696763732e-05 tokens_per_second_per_gpu:4671.13293162602 mfu:22.03453101593978 +Step 28146 | loss:0.5367539525032043 lr:1.2003297008994996e-05 tokens_per_second_per_gpu:3738.1790910336467 mfu:17.633628571526405 +Step 28147 | loss:0.37993431091308594 lr:1.2000478607004323e-05 tokens_per_second_per_gpu:4023.310691183775 mfu:18.978642977902012 +Step 28148 | loss:0.47360074520111084 lr:1.1997660490812885e-05 tokens_per_second_per_gpu:4147.551999262369 mfu:19.564710425861744 +Step 28149 | loss:0.5224899053573608 lr:1.1994842660441891e-05 tokens_per_second_per_gpu:5378.328439648157 mfu:25.370492887275937 +Step 28150 | loss:0.3407689332962036 lr:1.1992025115912554e-05 tokens_per_second_per_gpu:5326.96088641695 mfu:25.128183374476016 +Step 28151 | loss:0.34842509031295776 lr:1.1989207857246043e-05 tokens_per_second_per_gpu:5325.566195972723 mfu:25.12160438169187 +Step 28152 | loss:0.31962230801582336 lr:1.1986390884463533e-05 tokens_per_second_per_gpu:5089.462662350545 mfu:24.007863730179324 +Step 28153 | loss:0.29866358637809753 lr:1.1983574197586228e-05 tokens_per_second_per_gpu:4640.108502591926 mfu:21.88818349087254 +Step 28154 | loss:0.440324604511261 lr:1.1980757796635328e-05 tokens_per_second_per_gpu:4379.85812705912 mfu:20.660538066191506 +Step 28155 | loss:0.438605397939682 lr:1.1977941681631988e-05 tokens_per_second_per_gpu:5075.48082992453 mfu:23.94190904107921 +Step 28156 | loss:0.4631975293159485 lr:1.1975125852597401e-05 tokens_per_second_per_gpu:4099.025944073528 mfu:19.33580474413722 +Step 28157 | loss:0.40658360719680786 lr:1.1972310309552754e-05 tokens_per_second_per_gpu:4999.21086947461 mfu:23.58213062463983 +Step 28158 | loss:0.42905551195144653 lr:1.1969495052519213e-05 tokens_per_second_per_gpu:5264.624967339828 mfu:24.83413420858473 +Step 28159 | loss:0.38666123151779175 lr:1.1966680081517945e-05 tokens_per_second_per_gpu:5323.887780272101 mfu:25.11368700846479 +Step 28160 | loss:0.43858641386032104 lr:1.1963865396570123e-05 tokens_per_second_per_gpu:5270.569544447461 mfu:24.862175793050877 +Step 28161 | loss:0.3208064138889313 lr:1.1961050997696944e-05 tokens_per_second_per_gpu:4599.765565584481 mfu:21.697878973793593 +Step 28162 | loss:0.36504778265953064 lr:1.1958236884919532e-05 tokens_per_second_per_gpu:3628.6742139155053 mfu:17.11707538270152 +Step 28163 | loss:0.49677392840385437 lr:1.195542305825908e-05 tokens_per_second_per_gpu:4107.211153407757 mfu:19.374415773107497 +Step 28164 | loss:0.39194968342781067 lr:1.1952609517736757e-05 tokens_per_second_per_gpu:4120.553990082499 mfu:19.437356210224937 +Step 28165 | loss:0.38472190499305725 lr:1.1949796263373714e-05 tokens_per_second_per_gpu:4582.243794480206 mfu:21.615225789972307 +Step 28166 | loss:0.36941009759902954 lr:1.1946983295191095e-05 tokens_per_second_per_gpu:4609.657050765445 mfu:21.744538797053597 +Step 28167 | loss:0.4338696300983429 lr:1.1944170613210066e-05 tokens_per_second_per_gpu:5354.980144210568 mfu:25.26035499406685 +Step 28168 | loss:0.329459011554718 lr:1.1941358217451786e-05 tokens_per_second_per_gpu:4526.675352421931 mfu:21.353099967829564 +Step 28169 | loss:0.4803869128227234 lr:1.1938546107937426e-05 tokens_per_second_per_gpu:5033.765518441421 mfu:23.745130799447324 +Step 28170 | loss:0.48446792364120483 lr:1.19357342846881e-05 tokens_per_second_per_gpu:4455.25733786968 mfu:21.016209008016816 +Step 28171 | loss:0.3461795747280121 lr:1.1932922747724979e-05 tokens_per_second_per_gpu:4137.618401035743 mfu:19.517851948180127 +Step 28172 | loss:0.43626347184181213 lr:1.1930111497069213e-05 tokens_per_second_per_gpu:3597.4822452735466 mfu:16.96993754471877 +Step 28173 | loss:0.30140018463134766 lr:1.1927300532741936e-05 tokens_per_second_per_gpu:5071.724278399931 mfu:23.924188746603896 +Step 28174 | loss:0.3687894940376282 lr:1.1924489854764275e-05 tokens_per_second_per_gpu:5214.895643528013 mfu:24.599552503466843 +Step 28175 | loss:0.41468706727027893 lr:1.1921679463157387e-05 tokens_per_second_per_gpu:4150.968211386514 mfu:19.580825281317225 +Step 28176 | loss:0.4483262002468109 lr:1.1918869357942419e-05 tokens_per_second_per_gpu:5335.963621471196 mfu:25.170650811751887 +Step 28177 | loss:0.4396193325519562 lr:1.1916059539140478e-05 tokens_per_second_per_gpu:3687.451889584713 mfu:17.394339707338744 +Step 28178 | loss:0.41617220640182495 lr:1.1913250006772714e-05 tokens_per_second_per_gpu:4744.126833563545 mfu:22.378855705422982 +Step 28179 | loss:0.4193885624408722 lr:1.191044076086027e-05 tokens_per_second_per_gpu:4139.8750404653665 mfu:19.52849690139205 +Step 28180 | loss:0.42692267894744873 lr:1.1907631801424262e-05 tokens_per_second_per_gpu:5311.014884962591 mfu:25.05296336494741 +Step 28181 | loss:0.6968510150909424 lr:1.1904823128485804e-05 tokens_per_second_per_gpu:4638.929906002349 mfu:21.88262384967019 +Step 28182 | loss:0.3406999707221985 lr:1.1902014742066031e-05 tokens_per_second_per_gpu:4705.285963861773 mfu:22.195636696104 +Step 28183 | loss:0.3617843687534332 lr:1.1899206642186078e-05 tokens_per_second_per_gpu:3676.3104944325364 mfu:17.3417838454879 +Step 28184 | loss:0.4325619637966156 lr:1.1896398828867045e-05 tokens_per_second_per_gpu:4829.1175772252545 mfu:22.77977154419142 +Step 28185 | loss:0.45754167437553406 lr:1.1893591302130053e-05 tokens_per_second_per_gpu:4291.794641566804 mfu:20.245127579944278 +Step 28186 | loss:0.4282418191432953 lr:1.1890784061996241e-05 tokens_per_second_per_gpu:5143.029758188714 mfu:24.260548860736954 +Step 28187 | loss:0.4440862536430359 lr:1.1887977108486703e-05 tokens_per_second_per_gpu:4582.488506875553 mfu:21.616380140093398 +Step 28188 | loss:0.3844493329524994 lr:1.188517044162254e-05 tokens_per_second_per_gpu:5015.383675647039 mfu:23.658420510720443 +Step 28189 | loss:0.3967369496822357 lr:1.1882364061424872e-05 tokens_per_second_per_gpu:3940.283952879613 mfu:18.586991687498998 +Step 28190 | loss:0.41400235891342163 lr:1.1879557967914818e-05 tokens_per_second_per_gpu:4367.519978372995 mfu:20.602336913733815 +Step 28191 | loss:0.34970828890800476 lr:1.1876752161113464e-05 tokens_per_second_per_gpu:4226.9837553459665 mfu:19.939403571763147 +Step 28192 | loss:0.5283381938934326 lr:1.1873946641041921e-05 tokens_per_second_per_gpu:4696.535398307508 mfu:22.154358785384098 +Step 28193 | loss:0.3892267048358917 lr:1.1871141407721304e-05 tokens_per_second_per_gpu:5327.233623007067 mfu:25.129469919505258 +Step 28194 | loss:0.4425109922885895 lr:1.1868336461172697e-05 tokens_per_second_per_gpu:4605.686063902432 mfu:21.725806974500543 +Step 28195 | loss:0.43299704790115356 lr:1.1865531801417179e-05 tokens_per_second_per_gpu:5338.850053253829 mfu:25.184266602946607 +Step 28196 | loss:0.3291478753089905 lr:1.1862727428475866e-05 tokens_per_second_per_gpu:4009.6479782137317 mfu:18.914193629723115 +Step 28197 | loss:0.4836689531803131 lr:1.1859923342369866e-05 tokens_per_second_per_gpu:3587.587932664484 mfu:16.923264383998017 +Step 28198 | loss:0.3933505713939667 lr:1.1857119543120227e-05 tokens_per_second_per_gpu:5372.281925206361 mfu:25.341970447012244 +Step 28199 | loss:0.36753904819488525 lr:1.1854316030748063e-05 tokens_per_second_per_gpu:4653.496615813445 mfu:21.951337505186103 +Step 28200 | loss:0.4502263069152832 lr:1.1851512805274468e-05 tokens_per_second_per_gpu:5349.348061702929 mfu:25.233787499945954 +Step 28201 | loss:0.44834640622138977 lr:1.1848709866720514e-05 tokens_per_second_per_gpu:4759.155509008373 mfu:22.44974853165267 +Step 28202 | loss:0.3628394901752472 lr:1.1845907215107267e-05 tokens_per_second_per_gpu:5323.7814121732235 mfu:25.113185251994114 +Step 28203 | loss:0.28458166122436523 lr:1.1843104850455822e-05 tokens_per_second_per_gpu:5358.345024686119 mfu:25.27622770937783 +Step 28204 | loss:0.3425154983997345 lr:1.1840302772787248e-05 tokens_per_second_per_gpu:5371.178550810484 mfu:25.336765641731805 +Step 28205 | loss:0.4770585894584656 lr:1.1837500982122645e-05 tokens_per_second_per_gpu:5339.848443432049 mfu:25.188976179760203 +Step 28206 | loss:0.4247974157333374 lr:1.1834699478483053e-05 tokens_per_second_per_gpu:5373.674403019983 mfu:25.34853900244031 +Step 28207 | loss:0.5383400321006775 lr:1.1831898261889557e-05 tokens_per_second_per_gpu:3863.6345266444887 mfu:18.225423266207265 +Step 28208 | loss:0.3915925920009613 lr:1.1829097332363248e-05 tokens_per_second_per_gpu:3968.260496866576 mfu:18.71896181877611 +Step 28209 | loss:0.4761459231376648 lr:1.1826296689925142e-05 tokens_per_second_per_gpu:5348.5507947401675 mfu:25.230026655654836 +Step 28210 | loss:0.47679153084754944 lr:1.1823496334596334e-05 tokens_per_second_per_gpu:4467.956204972993 mfu:21.076111730793293 +Step 28211 | loss:0.4101979434490204 lr:1.182069626639788e-05 tokens_per_second_per_gpu:5025.363817056805 mfu:23.705498540538024 +Step 28212 | loss:0.42654091119766235 lr:1.181789648535085e-05 tokens_per_second_per_gpu:4505.291386838264 mfu:21.25222815369066 +Step 28213 | loss:0.40162786841392517 lr:1.1815096991476282e-05 tokens_per_second_per_gpu:4157.883255801306 mfu:19.61344472565004 +Step 28214 | loss:0.5645228624343872 lr:1.1812297784795245e-05 tokens_per_second_per_gpu:5353.322579917518 mfu:25.2525359804869 +Step 28215 | loss:0.449319064617157 lr:1.18094988653288e-05 tokens_per_second_per_gpu:4085.873076840948 mfu:19.27376042529059 +Step 28216 | loss:0.4285420775413513 lr:1.1806700233097983e-05 tokens_per_second_per_gpu:5199.964674394101 mfu:24.529120574576154 +Step 28217 | loss:0.47217148542404175 lr:1.1803901888123842e-05 tokens_per_second_per_gpu:4633.058493784799 mfu:21.854927396473784 +Step 28218 | loss:0.36013299226760864 lr:1.1801103830427423e-05 tokens_per_second_per_gpu:4883.030880227369 mfu:23.034089793010416 +Step 28219 | loss:0.49728119373321533 lr:1.179830606002979e-05 tokens_per_second_per_gpu:5241.379281377594 mfu:24.724480341777873 +Step 28220 | loss:0.493632048368454 lr:1.1795508576951958e-05 tokens_per_second_per_gpu:4709.988267245016 mfu:22.217818263458447 +Step 28221 | loss:0.3569071590900421 lr:1.1792711381214982e-05 tokens_per_second_per_gpu:3834.944078568169 mfu:18.090085527536058 +Step 28222 | loss:0.3903685510158539 lr:1.1789914472839914e-05 tokens_per_second_per_gpu:4179.306201735447 mfu:19.714500416752863 +Step 28223 | loss:0.42588913440704346 lr:1.1787117851847773e-05 tokens_per_second_per_gpu:4116.195374796375 mfu:19.416795878263848 +Step 28224 | loss:0.38028237223625183 lr:1.1784321518259583e-05 tokens_per_second_per_gpu:3954.241914699359 mfu:18.652833774876594 +Step 28225 | loss:0.3746601343154907 lr:1.1781525472096384e-05 tokens_per_second_per_gpu:3714.0197229272567 mfu:17.519664710155315 +Step 28226 | loss:0.40461161732673645 lr:1.1778729713379228e-05 tokens_per_second_per_gpu:4687.574129146311 mfu:22.11208694980055 +Step 28227 | loss:0.4505123198032379 lr:1.17759342421291e-05 tokens_per_second_per_gpu:5328.419457151886 mfu:25.135063701490537 +Step 28228 | loss:0.4359097480773926 lr:1.1773139058367056e-05 tokens_per_second_per_gpu:4120.278882969918 mfu:19.436058483036785 +Step 28229 | loss:0.49509209394454956 lr:1.1770344162114122e-05 tokens_per_second_per_gpu:4690.836840167015 mfu:22.127477714361117 +Step 28230 | loss:0.34896770119667053 lr:1.1767549553391305e-05 tokens_per_second_per_gpu:5296.8503200718305 mfu:24.986146695633717 +Step 28231 | loss:0.4189819395542145 lr:1.1764755232219615e-05 tokens_per_second_per_gpu:4007.777322482709 mfu:18.9053694274731 +Step 28232 | loss:0.4252099394798279 lr:1.176196119862008e-05 tokens_per_second_per_gpu:4409.778404345167 mfu:20.801677119075496 +Step 28233 | loss:0.432407021522522 lr:1.1759167452613728e-05 tokens_per_second_per_gpu:5358.765971278579 mfu:25.27821338627536 +Step 28234 | loss:0.45203977823257446 lr:1.1756373994221548e-05 tokens_per_second_per_gpu:4850.321265937869 mfu:22.879792961571162 +Step 28235 | loss:0.37613049149513245 lr:1.175358082346455e-05 tokens_per_second_per_gpu:4228.522186100732 mfu:19.946660612116727 +Step 28236 | loss:0.31406837701797485 lr:1.1750787940363756e-05 tokens_per_second_per_gpu:5381.399475874058 mfu:25.384979489126323 +Step 28237 | loss:0.400895357131958 lr:1.1747995344940193e-05 tokens_per_second_per_gpu:3983.5531079857474 mfu:18.791099674613747 +Step 28238 | loss:0.3616047203540802 lr:1.1745203037214813e-05 tokens_per_second_per_gpu:4086.9528711672137 mfu:19.27885399935949 +Step 28239 | loss:0.43040135502815247 lr:1.1742411017208638e-05 tokens_per_second_per_gpu:4569.405005636249 mfu:21.55466303247211 +Step 28240 | loss:0.46295735239982605 lr:1.1739619284942687e-05 tokens_per_second_per_gpu:5317.805443307697 mfu:25.08499558725 +Step 28241 | loss:0.38571226596832275 lr:1.173682784043793e-05 tokens_per_second_per_gpu:5068.860015520864 mfu:23.910677529909698 +Step 28242 | loss:0.4329535961151123 lr:1.1734036683715371e-05 tokens_per_second_per_gpu:4038.40364491078 mfu:19.04983901575574 +Step 28243 | loss:0.5174281597137451 lr:1.1731245814796005e-05 tokens_per_second_per_gpu:4879.241081076823 mfu:23.016212663810876 +Step 28244 | loss:0.401254802942276 lr:1.1728455233700847e-05 tokens_per_second_per_gpu:4700.242217978126 mfu:22.171844486219054 +Step 28245 | loss:0.3921523988246918 lr:1.1725664940450837e-05 tokens_per_second_per_gpu:4713.954591696004 mfu:22.236528092618492 +Step 28246 | loss:0.41045328974723816 lr:1.1722874935066985e-05 tokens_per_second_per_gpu:4676.369889934168 mfu:22.059234641796543 +Step 28247 | loss:0.460986465215683 lr:1.172008521757027e-05 tokens_per_second_per_gpu:3311.2221852913676 mfu:15.619600000780276 +Step 28248 | loss:0.5427013635635376 lr:1.1717295787981696e-05 tokens_per_second_per_gpu:4614.151042293395 mfu:21.765737721844957 +Step 28249 | loss:0.2820129990577698 lr:1.1714506646322215e-05 tokens_per_second_per_gpu:5345.0872484450065 mfu:25.213688516838626 +Step 28250 | loss:0.44578778743743896 lr:1.1711717792612808e-05 tokens_per_second_per_gpu:4578.388580155147 mfu:21.597040086233612 +Step 28251 | loss:0.36432990431785583 lr:1.1708929226874477e-05 tokens_per_second_per_gpu:3904.995357491278 mfu:18.420529362196994 +Step 28252 | loss:0.35380735993385315 lr:1.1706140949128169e-05 tokens_per_second_per_gpu:4591.7817637980315 mfu:21.660218018589866 +Step 28253 | loss:0.4163235127925873 lr:1.1703352959394853e-05 tokens_per_second_per_gpu:4697.316581764236 mfu:22.158043761033124 +Step 28254 | loss:0.3253322243690491 lr:1.1700565257695505e-05 tokens_per_second_per_gpu:4685.3227030400785 mfu:22.10146658872455 +Step 28255 | loss:0.3705003559589386 lr:1.1697777844051105e-05 tokens_per_second_per_gpu:3690.977749128098 mfu:17.41097178837856 +Step 28256 | loss:0.41652151942253113 lr:1.169499071848259e-05 tokens_per_second_per_gpu:4822.190613772682 mfu:22.747095875723797 +Step 28257 | loss:0.47722917795181274 lr:1.1692203881010943e-05 tokens_per_second_per_gpu:4409.256710334485 mfu:20.799216199416936 +Step 28258 | loss:0.42311736941337585 lr:1.1689417331657126e-05 tokens_per_second_per_gpu:5058.425721522979 mfu:23.86145718484754 +Step 28259 | loss:0.42619892954826355 lr:1.1686631070442095e-05 tokens_per_second_per_gpu:3728.1104979551815 mfu:17.58613330009616 +Step 28260 | loss:0.4729621112346649 lr:1.1683845097386781e-05 tokens_per_second_per_gpu:3814.154270256508 mfu:17.99201645462289 +Step 28261 | loss:0.3727526068687439 lr:1.1681059412512162e-05 tokens_per_second_per_gpu:4956.79879365264 mfu:23.382065626741284 +Step 28262 | loss:0.46311241388320923 lr:1.1678274015839196e-05 tokens_per_second_per_gpu:5252.292912265132 mfu:24.775961800731825 +Step 28263 | loss:0.38144028186798096 lr:1.167548890738881e-05 tokens_per_second_per_gpu:5313.759737795866 mfu:25.065911304082157 +Step 28264 | loss:0.4511955976486206 lr:1.1672704087181962e-05 tokens_per_second_per_gpu:5313.140989305167 mfu:25.062992561882147 +Step 28265 | loss:0.4552222788333893 lr:1.1669919555239606e-05 tokens_per_second_per_gpu:5313.28332319988 mfu:25.063663975148177 +Step 28266 | loss:0.34942105412483215 lr:1.1667135311582684e-05 tokens_per_second_per_gpu:4341.819090668046 mfu:20.481101441405958 +Step 28267 | loss:0.37983301281929016 lr:1.1664351356232105e-05 tokens_per_second_per_gpu:5343.724627039868 mfu:25.207260799183405 +Step 28268 | loss:0.372018963098526 lr:1.166156768920884e-05 tokens_per_second_per_gpu:4136.219467370173 mfu:19.51125293939694 +Step 28269 | loss:0.39640507102012634 lr:1.1658784310533827e-05 tokens_per_second_per_gpu:4182.4435784286825 mfu:19.729299957906516 +Step 28270 | loss:0.4872147738933563 lr:1.1656001220227974e-05 tokens_per_second_per_gpu:5356.630056538462 mfu:25.268137912021594 +Step 28271 | loss:0.37268370389938354 lr:1.1653218418312228e-05 tokens_per_second_per_gpu:4340.76096838879 mfu:20.476110098080433 +Step 28272 | loss:0.3618236184120178 lr:1.1650435904807521e-05 tokens_per_second_per_gpu:5341.676599540938 mfu:25.19759990404143 +Step 28273 | loss:0.48416081070899963 lr:1.1647653679734811e-05 tokens_per_second_per_gpu:4515.716072278541 mfu:21.30140317354719 +Step 28274 | loss:0.3505014479160309 lr:1.164487174311496e-05 tokens_per_second_per_gpu:5366.898933189808 mfu:25.31657795523702 +Step 28275 | loss:0.513814389705658 lr:1.1642090094968922e-05 tokens_per_second_per_gpu:4305.852408999506 mfu:20.31144046742678 +Step 28276 | loss:0.4328087568283081 lr:1.1639308735317634e-05 tokens_per_second_per_gpu:5315.550012156503 mfu:25.074356333693583 +Step 28277 | loss:0.42123377323150635 lr:1.1636527664181985e-05 tokens_per_second_per_gpu:5325.3211681673765 mfu:25.120448543727996 +Step 28278 | loss:0.3179195523262024 lr:1.1633746881582902e-05 tokens_per_second_per_gpu:4086.4899761264264 mfu:19.27667044447406 +Step 28279 | loss:0.5238986611366272 lr:1.1630966387541309e-05 tokens_per_second_per_gpu:4057.5719706366863 mfu:19.14025927865845 +Step 28280 | loss:0.42342430353164673 lr:1.1628186182078143e-05 tokens_per_second_per_gpu:4521.5750391864585 mfu:21.32904092009467 +Step 28281 | loss:0.36762380599975586 lr:1.1625406265214256e-05 tokens_per_second_per_gpu:5330.528736170182 mfu:25.145013530499952 +Step 28282 | loss:0.3628767132759094 lr:1.1622626636970585e-05 tokens_per_second_per_gpu:5231.156284373038 mfu:24.676256720684236 +Step 28283 | loss:0.5203951001167297 lr:1.1619847297368037e-05 tokens_per_second_per_gpu:5253.626560268484 mfu:24.782252845907454 +Step 28284 | loss:0.3074435591697693 lr:1.1617068246427526e-05 tokens_per_second_per_gpu:3821.5530265424104 mfu:18.02691765038148 +Step 28285 | loss:0.5217111706733704 lr:1.1614289484169933e-05 tokens_per_second_per_gpu:4623.805449360221 mfu:21.81127920719088 +Step 28286 | loss:0.44012176990509033 lr:1.1611511010616161e-05 tokens_per_second_per_gpu:5312.39061533737 mfu:25.059452919849047 +Step 28287 | loss:0.4574859142303467 lr:1.1608732825787134e-05 tokens_per_second_per_gpu:4300.958001409861 mfu:20.28835271175438 +Step 28288 | loss:0.4894363582134247 lr:1.1605954929703722e-05 tokens_per_second_per_gpu:5287.509479947919 mfu:24.942084359057564 +Step 28289 | loss:0.3491315543651581 lr:1.1603177322386804e-05 tokens_per_second_per_gpu:3814.8347602497306 mfu:17.99522644202448 +Step 28290 | loss:0.42068690061569214 lr:1.1600400003857298e-05 tokens_per_second_per_gpu:5342.363606359069 mfu:25.200840632418004 +Step 28291 | loss:0.4209452271461487 lr:1.1597622974136091e-05 tokens_per_second_per_gpu:3423.1813888437623 mfu:16.147730666146806 +Step 28292 | loss:0.5537506937980652 lr:1.159484623324405e-05 tokens_per_second_per_gpu:4123.7849928800615 mfu:19.452597401686965 +Step 28293 | loss:0.4304855763912201 lr:1.1592069781202069e-05 tokens_per_second_per_gpu:3725.615427433478 mfu:17.574363626742358 +Step 28294 | loss:0.41119399666786194 lr:1.158929361803105e-05 tokens_per_second_per_gpu:4898.8617946920585 mfu:23.108766917572883 +Step 28295 | loss:0.4170686900615692 lr:1.1586517743751851e-05 tokens_per_second_per_gpu:4467.56771858946 mfu:21.074279174239557 +Step 28296 | loss:0.38408541679382324 lr:1.1583742158385341e-05 tokens_per_second_per_gpu:4542.443837363701 mfu:21.42748260168085 +Step 28297 | loss:0.3546113967895508 lr:1.1580966861952408e-05 tokens_per_second_per_gpu:4920.9990073741155 mfu:23.213191926792096 +Step 28298 | loss:0.46070462465286255 lr:1.1578191854473941e-05 tokens_per_second_per_gpu:3177.235907204978 mfu:14.987563866630444 +Step 28299 | loss:0.37567469477653503 lr:1.1575417135970778e-05 tokens_per_second_per_gpu:4796.28795349823 mfu:22.624908607759647 +Step 28300 | loss:0.4376571774482727 lr:1.157264270646381e-05 tokens_per_second_per_gpu:5358.450413452148 mfu:25.276724846168367 +Step 28301 | loss:0.40086227655410767 lr:1.1569868565973912e-05 tokens_per_second_per_gpu:5384.607077086501 mfu:25.400110291318615 +Step 28302 | loss:0.46085813641548157 lr:1.1567094714521937e-05 tokens_per_second_per_gpu:4051.48361500185 mfu:19.11153946634834 +Step 28303 | loss:0.4645645320415497 lr:1.1564321152128726e-05 tokens_per_second_per_gpu:4247.34316150056 mfu:20.035442364267247 +Step 28304 | loss:0.4045465886592865 lr:1.1561547878815171e-05 tokens_per_second_per_gpu:3728.1172025369633 mfu:17.586164926752346 +Step 28305 | loss:0.323024719953537 lr:1.1558774894602125e-05 tokens_per_second_per_gpu:3543.3656186900907 mfu:16.714660183876013 +Step 28306 | loss:0.4191686809062958 lr:1.155600219951043e-05 tokens_per_second_per_gpu:5026.624187039423 mfu:23.711443920787325 +Step 28307 | loss:0.45500364899635315 lr:1.1553229793560944e-05 tokens_per_second_per_gpu:5279.51476762231 mfu:24.904371936979086 +Step 28308 | loss:0.46239814162254333 lr:1.1550457676774534e-05 tokens_per_second_per_gpu:3684.859582044435 mfu:17.38211135037798 +Step 28309 | loss:0.4743102192878723 lr:1.1547685849172046e-05 tokens_per_second_per_gpu:5334.920451465768 mfu:25.165730000853124 +Step 28310 | loss:0.3964708149433136 lr:1.1544914310774296e-05 tokens_per_second_per_gpu:4125.458415016311 mfu:19.460491219418913 +Step 28311 | loss:0.46412450075149536 lr:1.1542143061602162e-05 tokens_per_second_per_gpu:5317.970692374309 mfu:25.085775095291602 +Step 28312 | loss:0.36133500933647156 lr:1.1539372101676482e-05 tokens_per_second_per_gpu:5044.8354463809565 mfu:23.79734953826258 +Step 28313 | loss:0.45508652925491333 lr:1.1536601431018085e-05 tokens_per_second_per_gpu:3921.6565794231187 mfu:18.499123188747078 +Step 28314 | loss:0.4763261377811432 lr:1.1533831049647814e-05 tokens_per_second_per_gpu:4525.571343492792 mfu:21.34789216934747 +Step 28315 | loss:0.43570423126220703 lr:1.153106095758651e-05 tokens_per_second_per_gpu:3826.0419758788057 mfu:18.04809279029499 +Step 28316 | loss:0.37692776322364807 lr:1.152829115485503e-05 tokens_per_second_per_gpu:3701.5191629331666 mfu:17.460697435847727 +Step 28317 | loss:0.3603546917438507 lr:1.1525521641474152e-05 tokens_per_second_per_gpu:5342.641232441708 mfu:25.2021502420177 +Step 28318 | loss:0.38143855333328247 lr:1.1522752417464738e-05 tokens_per_second_per_gpu:4806.57036650674 mfu:22.673412504281355 +Step 28319 | loss:0.5901361107826233 lr:1.1519983482847613e-05 tokens_per_second_per_gpu:5355.903870027917 mfu:25.26471236635116 +Step 28320 | loss:0.2953120470046997 lr:1.151721483764362e-05 tokens_per_second_per_gpu:4250.261126668715 mfu:20.04920690382163 +Step 28321 | loss:0.5261816382408142 lr:1.1514446481873542e-05 tokens_per_second_per_gpu:3659.5678674053725 mfu:17.262806017214036 +Step 28322 | loss:0.4533478021621704 lr:1.151167841555823e-05 tokens_per_second_per_gpu:3544.853158403025 mfu:16.72167716250237 +Step 28323 | loss:0.3475540578365326 lr:1.1508910638718506e-05 tokens_per_second_per_gpu:3335.6582139044067 mfu:15.734868916964379 +Step 28324 | loss:0.4047817885875702 lr:1.1506143151375176e-05 tokens_per_second_per_gpu:4121.036923068782 mfu:19.439634287518103 +Step 28325 | loss:0.3576129376888275 lr:1.1503375953549045e-05 tokens_per_second_per_gpu:5249.318151636978 mfu:24.76192934730232 +Step 28326 | loss:0.3847159147262573 lr:1.1500609045260935e-05 tokens_per_second_per_gpu:5327.551277709158 mfu:25.130968350932502 +Step 28327 | loss:0.3549543023109436 lr:1.1497842426531669e-05 tokens_per_second_per_gpu:5349.522307599764 mfu:25.234609447570737 +Step 28328 | loss:0.4004676342010498 lr:1.149507609738203e-05 tokens_per_second_per_gpu:4299.502130886123 mfu:20.28148512209679 +Step 28329 | loss:0.4325837194919586 lr:1.1492310057832834e-05 tokens_per_second_per_gpu:4519.546598637856 mfu:21.319472419939267 +Step 28330 | loss:0.46186861395835876 lr:1.1489544307904904e-05 tokens_per_second_per_gpu:5264.76141763115 mfu:24.834777867889716 +Step 28331 | loss:0.3815379738807678 lr:1.1486778847619023e-05 tokens_per_second_per_gpu:3832.204205863454 mfu:18.077161080517428 +Step 28332 | loss:0.4300520420074463 lr:1.148401367699598e-05 tokens_per_second_per_gpu:4356.210810995683 mfu:20.548989641672108 +Step 28333 | loss:0.43400418758392334 lr:1.1481248796056588e-05 tokens_per_second_per_gpu:4422.27750442532 mfu:20.86063750671996 +Step 28334 | loss:0.4413483440876007 lr:1.1478484204821648e-05 tokens_per_second_per_gpu:3869.070105661352 mfu:18.25106381983522 +Step 28335 | loss:0.3554389774799347 lr:1.1475719903311938e-05 tokens_per_second_per_gpu:5017.46267092463 mfu:23.668227486158013 +Step 28336 | loss:0.5183528065681458 lr:1.1472955891548247e-05 tokens_per_second_per_gpu:4948.576132649182 mfu:23.34327793185734 +Step 28337 | loss:0.4335647523403168 lr:1.1470192169551385e-05 tokens_per_second_per_gpu:4224.165243174687 mfu:19.92610816896346 +Step 28338 | loss:0.5229780673980713 lr:1.1467428737342123e-05 tokens_per_second_per_gpu:5001.772003269851 mfu:23.594211929726463 +Step 28339 | loss:0.3214685618877411 lr:1.1464665594941237e-05 tokens_per_second_per_gpu:4831.594926558826 mfu:22.791457623677257 +Step 28340 | loss:0.34515759348869324 lr:1.1461902742369518e-05 tokens_per_second_per_gpu:4765.203676419356 mfu:22.47827876925405 +Step 28341 | loss:0.4198185205459595 lr:1.1459140179647754e-05 tokens_per_second_per_gpu:4136.719480848178 mfu:19.51361158828378 +Step 28342 | loss:0.41253799200057983 lr:1.1456377906796705e-05 tokens_per_second_per_gpu:5294.216152434053 mfu:24.973720877450976 +Step 28343 | loss:0.49831661581993103 lr:1.1453615923837158e-05 tokens_per_second_per_gpu:3416.9350691382597 mfu:16.11826571036391 +Step 28344 | loss:0.443651407957077 lr:1.1450854230789893e-05 tokens_per_second_per_gpu:5346.610895788097 mfu:25.220875821317026 +Step 28345 | loss:0.5195959806442261 lr:1.1448092827675677e-05 tokens_per_second_per_gpu:5313.029504721487 mfu:25.062466670079697 +Step 28346 | loss:0.4420664608478546 lr:1.1445331714515256e-05 tokens_per_second_per_gpu:4973.85462709003 mfu:23.462520903090706 +Step 28347 | loss:0.47283390164375305 lr:1.1442570891329419e-05 tokens_per_second_per_gpu:5204.37242180315 mfu:24.549912671143893 +Step 28348 | loss:0.4048656225204468 lr:1.1439810358138936e-05 tokens_per_second_per_gpu:4534.8453076729875 mfu:21.391639040686904 +Step 28349 | loss:0.3824771046638489 lr:1.1437050114964549e-05 tokens_per_second_per_gpu:5324.788924230726 mfu:25.117937858268636 +Step 28350 | loss:0.39511337876319885 lr:1.1434290161827028e-05 tokens_per_second_per_gpu:3848.565087504076 mfu:18.154338150670494 +Step 28351 | loss:0.3004181683063507 lr:1.1431530498747129e-05 tokens_per_second_per_gpu:5310.118632568806 mfu:25.0487355894895 +Step 28352 | loss:0.3611890375614166 lr:1.1428771125745641e-05 tokens_per_second_per_gpu:4043.5563306759473 mfu:19.074145113648882 +Step 28353 | loss:0.3639768064022064 lr:1.142601204284326e-05 tokens_per_second_per_gpu:5362.5428290791015 mfu:25.296029468918814 +Step 28354 | loss:0.5016692280769348 lr:1.1423253250060761e-05 tokens_per_second_per_gpu:4895.1108855319035 mfu:23.091073239093333 +Step 28355 | loss:0.44658616185188293 lr:1.1420494747418897e-05 tokens_per_second_per_gpu:4008.9617889106667 mfu:18.910956757704586 +Step 28356 | loss:0.3226813077926636 lr:1.1417736534938434e-05 tokens_per_second_per_gpu:5340.49255205958 mfu:25.19201455005253 +Step 28357 | loss:0.43244320154190063 lr:1.141497861264008e-05 tokens_per_second_per_gpu:3483.743991036148 mfu:16.433414793733732 +Step 28358 | loss:0.34341055154800415 lr:1.1412220980544597e-05 tokens_per_second_per_gpu:3744.4890898733415 mfu:17.66339391265006 +Step 28359 | loss:0.4829636812210083 lr:1.1409463638672734e-05 tokens_per_second_per_gpu:5329.639266410471 mfu:25.140817749884203 +Step 28360 | loss:0.4203588366508484 lr:1.1406706587045224e-05 tokens_per_second_per_gpu:5350.088345521608 mfu:25.237279545024105 +Step 28361 | loss:0.5016449093818665 lr:1.140394982568278e-05 tokens_per_second_per_gpu:5323.825743378503 mfu:25.1133943698531 +Step 28362 | loss:0.38968387246131897 lr:1.1401193354606155e-05 tokens_per_second_per_gpu:3806.6348849301344 mfu:17.95654623109895 +Step 28363 | loss:0.49090734124183655 lr:1.1398437173836096e-05 tokens_per_second_per_gpu:5393.89620375483 mfu:25.44392868670156 +Step 28364 | loss:0.4256769120693207 lr:1.13956812833933e-05 tokens_per_second_per_gpu:3157.3003452746952 mfu:14.893524419647697 +Step 28365 | loss:0.2669070065021515 lr:1.1392925683298516e-05 tokens_per_second_per_gpu:5203.24673607334 mfu:24.54460261949459 +Step 28366 | loss:0.3458041548728943 lr:1.1390170373572473e-05 tokens_per_second_per_gpu:3920.6953918184863 mfu:18.49458910282051 +Step 28367 | loss:0.46229466795921326 lr:1.1387415354235887e-05 tokens_per_second_per_gpu:3368.234058074276 mfu:15.888534731925791 +Step 28368 | loss:0.37914830446243286 lr:1.1384660625309463e-05 tokens_per_second_per_gpu:5361.502817130597 mfu:25.291123555113384 +Step 28369 | loss:0.5206282138824463 lr:1.1381906186813935e-05 tokens_per_second_per_gpu:5345.638502163569 mfu:25.216288874682668 +Step 28370 | loss:0.45227742195129395 lr:1.137915203877003e-05 tokens_per_second_per_gpu:5238.689700973804 mfu:24.711793132123457 +Step 28371 | loss:0.4667940139770508 lr:1.137639818119844e-05 tokens_per_second_per_gpu:4511.192430753879 mfu:21.280064384662527 +Step 28372 | loss:0.48595109581947327 lr:1.1373644614119887e-05 tokens_per_second_per_gpu:4988.906555988758 mfu:23.533523419829894 +Step 28373 | loss:0.4019278585910797 lr:1.1370891337555095e-05 tokens_per_second_per_gpu:5333.500925453127 mfu:25.159033854455384 +Step 28374 | loss:0.5269784331321716 lr:1.1368138351524754e-05 tokens_per_second_per_gpu:3425.5289033870995 mfu:16.158804292775006 +Step 28375 | loss:0.35732513666152954 lr:1.1365385656049566e-05 tokens_per_second_per_gpu:4442.777490893466 mfu:20.95733944054856 +Step 28376 | loss:0.4002346992492676 lr:1.136263325115024e-05 tokens_per_second_per_gpu:3403.20938487805 mfu:16.053519315865284 +Step 28377 | loss:0.47298112511634827 lr:1.1359881136847494e-05 tokens_per_second_per_gpu:5079.528232412482 mfu:23.961001329172955 +Step 28378 | loss:0.4080562889575958 lr:1.1357129313161997e-05 tokens_per_second_per_gpu:5362.72750382565 mfu:25.296900611207047 +Step 28379 | loss:0.33294758200645447 lr:1.135437778011446e-05 tokens_per_second_per_gpu:4354.686621287891 mfu:20.54179977876688 +Step 28380 | loss:0.3436638116836548 lr:1.1351626537725596e-05 tokens_per_second_per_gpu:5316.224910217973 mfu:25.077539942999017 +Step 28381 | loss:0.4358411431312561 lr:1.1348875586016083e-05 tokens_per_second_per_gpu:4028.889554838618 mfu:19.0049594296142 +Step 28382 | loss:0.3332977592945099 lr:1.1346124925006591e-05 tokens_per_second_per_gpu:4158.979707533949 mfu:19.61861687554687 +Step 28383 | loss:0.47512924671173096 lr:1.1343374554717823e-05 tokens_per_second_per_gpu:3063.629630351509 mfu:14.451663675483063 +Step 28384 | loss:0.40460240840911865 lr:1.1340624475170486e-05 tokens_per_second_per_gpu:4084.870332757179 mfu:19.269030310361668 +Step 28385 | loss:0.3883828818798065 lr:1.133787468638523e-05 tokens_per_second_per_gpu:5369.909737621039 mfu:25.33078043343534 +Step 28386 | loss:0.43111979961395264 lr:1.1335125188382751e-05 tokens_per_second_per_gpu:3693.942121319236 mfu:17.42495523777821 +Step 28387 | loss:0.36440297961235046 lr:1.133237598118373e-05 tokens_per_second_per_gpu:4333.488815556316 mfu:20.441806112412436 +Step 28388 | loss:0.4004928767681122 lr:1.1329627064808857e-05 tokens_per_second_per_gpu:4965.191869649008 mfu:23.42165719822258 +Step 28389 | loss:0.42372480034828186 lr:1.1326878439278793e-05 tokens_per_second_per_gpu:5340.09094848719 mfu:25.19012011748093 +Step 28390 | loss:0.49279338121414185 lr:1.1324130104614194e-05 tokens_per_second_per_gpu:3985.0950862337927 mfu:18.7983734490984 +Step 28391 | loss:0.4296046197414398 lr:1.132138206083575e-05 tokens_per_second_per_gpu:5326.669193411618 mfu:25.126807408801906 +Step 28392 | loss:0.269907146692276 lr:1.1318634307964142e-05 tokens_per_second_per_gpu:4529.83681056543 mfu:21.368013105292828 +Step 28393 | loss:0.3520115911960602 lr:1.1315886846020007e-05 tokens_per_second_per_gpu:4012.9601738794727 mfu:18.929817821797265 +Step 28394 | loss:0.5208162069320679 lr:1.1313139675024025e-05 tokens_per_second_per_gpu:4547.346855692903 mfu:21.45061097568088 +Step 28395 | loss:0.49291884899139404 lr:1.1310392794996866e-05 tokens_per_second_per_gpu:4614.430695574104 mfu:21.767056894083833 +Step 28396 | loss:0.428964763879776 lr:1.1307646205959182e-05 tokens_per_second_per_gpu:4448.275211988096 mfu:20.98327312895987 +Step 28397 | loss:0.36250272393226624 lr:1.1304899907931616e-05 tokens_per_second_per_gpu:4066.2456362316793 mfu:19.18117443914998 +Step 28398 | loss:0.4040672183036804 lr:1.130215390093483e-05 tokens_per_second_per_gpu:5394.46005776877 mfu:25.446588482288732 +Step 28399 | loss:0.40312904119491577 lr:1.1299408184989501e-05 tokens_per_second_per_gpu:4374.925735445395 mfu:20.637271133396776 +Step 28400 | loss:0.3981652855873108 lr:1.1296662760116244e-05 tokens_per_second_per_gpu:4184.486621325876 mfu:19.73893733983164 +Step 28401 | loss:0.43265247344970703 lr:1.1293917626335726e-05 tokens_per_second_per_gpu:5308.303664968314 mfu:25.04017407765257 +Step 28402 | loss:0.4413990080356598 lr:1.1291172783668608e-05 tokens_per_second_per_gpu:5327.672191656268 mfu:25.13153872265107 +Step 28403 | loss:0.39406701922416687 lr:1.1288428232135517e-05 tokens_per_second_per_gpu:5347.913010818451 mfu:25.22701812007894 +Step 28404 | loss:0.5979399681091309 lr:1.1285683971757083e-05 tokens_per_second_per_gpu:4069.6876525742355 mfu:19.197411017506365 +Step 28405 | loss:0.4154704511165619 lr:1.128294000255396e-05 tokens_per_second_per_gpu:4823.3921011584625 mfu:22.75276349671741 +Step 28406 | loss:0.5900477170944214 lr:1.1280196324546804e-05 tokens_per_second_per_gpu:5289.4979073360255 mfu:24.95146411030796 +Step 28407 | loss:0.43098026514053345 lr:1.127745293775621e-05 tokens_per_second_per_gpu:5354.9482267596395 mfu:25.26020443400514 +Step 28408 | loss:0.49496397376060486 lr:1.127470984220284e-05 tokens_per_second_per_gpu:4071.6878801198254 mfu:19.206846432112922 +Step 28409 | loss:0.4374552071094513 lr:1.1271967037907328e-05 tokens_per_second_per_gpu:5335.745710510929 mfu:25.16962288857286 +Step 28410 | loss:0.406984806060791 lr:1.1269224524890298e-05 tokens_per_second_per_gpu:5062.036097582205 mfu:23.878487944712585 +Step 28411 | loss:0.4307864308357239 lr:1.1266482303172354e-05 tokens_per_second_per_gpu:5321.877542853822 mfu:25.104204375580895 +Step 28412 | loss:0.24639587104320526 lr:1.1263740372774145e-05 tokens_per_second_per_gpu:5331.474210198771 mfu:25.14947349280731 +Step 28413 | loss:0.35611316561698914 lr:1.1260998733716294e-05 tokens_per_second_per_gpu:5312.470040774157 mfu:25.059827583185747 +Step 28414 | loss:0.37958425283432007 lr:1.1258257386019405e-05 tokens_per_second_per_gpu:4751.224276543606 mfu:22.412335554908566 +Step 28415 | loss:0.45057278871536255 lr:1.1255516329704108e-05 tokens_per_second_per_gpu:3917.1370496777045 mfu:18.47780379582667 +Step 28416 | loss:0.5127120614051819 lr:1.1252775564791024e-05 tokens_per_second_per_gpu:5327.618557932104 mfu:25.131285723223325 +Step 28417 | loss:0.5357431769371033 lr:1.1250035091300765e-05 tokens_per_second_per_gpu:5157.426158508958 mfu:24.32845913732672 +Step 28418 | loss:0.38094794750213623 lr:1.124729490925392e-05 tokens_per_second_per_gpu:5353.997192537456 mfu:25.255718243315933 +Step 28419 | loss:0.5213288068771362 lr:1.1244555018671116e-05 tokens_per_second_per_gpu:3719.228636411314 mfu:17.544236043791894 +Step 28420 | loss:0.34736669063568115 lr:1.124181541957297e-05 tokens_per_second_per_gpu:5343.176368205523 mfu:25.2046745687192 +Step 28421 | loss:0.33302441239356995 lr:1.1239076111980068e-05 tokens_per_second_per_gpu:5349.353193644279 mfu:25.2338117081889 +Step 28422 | loss:0.36479651927948 lr:1.1236337095913012e-05 tokens_per_second_per_gpu:3462.1612915699225 mfu:16.3316055179633 +Step 28423 | loss:0.3653451204299927 lr:1.1233598371392417e-05 tokens_per_second_per_gpu:5056.07351737756 mfu:23.850361436566192 +Step 28424 | loss:0.32861167192459106 lr:1.1230859938438887e-05 tokens_per_second_per_gpu:4575.280730633517 mfu:21.58237982978628 +Step 28425 | loss:0.4307485520839691 lr:1.122812179707301e-05 tokens_per_second_per_gpu:5355.215625614103 mfu:25.261465800024595 +Step 28426 | loss:0.42067596316337585 lr:1.122538394731536e-05 tokens_per_second_per_gpu:3781.308919675589 mfu:17.837079331938018 +Step 28427 | loss:0.3382342457771301 lr:1.1222646389186542e-05 tokens_per_second_per_gpu:5168.75527608027 mfu:24.381900517857378 +Step 28428 | loss:0.3541865944862366 lr:1.1219909122707168e-05 tokens_per_second_per_gpu:5345.0103753493895 mfu:25.213325893330698 +Step 28429 | loss:0.402059406042099 lr:1.1217172147897787e-05 tokens_per_second_per_gpu:5357.645133168534 mfu:25.272926201675574 +Step 28430 | loss:0.32461267709732056 lr:1.1214435464779006e-05 tokens_per_second_per_gpu:4587.1116957186505 mfu:21.638188510663642 +Step 28431 | loss:0.4654427170753479 lr:1.1211699073371417e-05 tokens_per_second_per_gpu:4223.802933874573 mfu:19.92439909417848 +Step 28432 | loss:0.47809743881225586 lr:1.1208962973695586e-05 tokens_per_second_per_gpu:3917.07345498157 mfu:18.477503808794633 +Step 28433 | loss:0.521439790725708 lr:1.1206227165772088e-05 tokens_per_second_per_gpu:4988.932033499033 mfu:23.533643601592775 +Step 28434 | loss:0.45183417201042175 lr:1.1203491649621501e-05 tokens_per_second_per_gpu:5133.568009151102 mfu:24.215916176185686 +Step 28435 | loss:0.3898315727710724 lr:1.1200756425264413e-05 tokens_per_second_per_gpu:5170.492565697012 mfu:24.39009560939172 +Step 28436 | loss:0.43868136405944824 lr:1.1198021492721384e-05 tokens_per_second_per_gpu:4473.068830762864 mfu:21.100228858948025 +Step 28437 | loss:0.4182312488555908 lr:1.1195286852012977e-05 tokens_per_second_per_gpu:5269.604172099635 mfu:24.857621966976673 +Step 28438 | loss:0.5278169512748718 lr:1.1192552503159787e-05 tokens_per_second_per_gpu:5324.702335289349 mfu:25.117529403458565 +Step 28439 | loss:0.41835060715675354 lr:1.118981844618236e-05 tokens_per_second_per_gpu:3936.4059241950263 mfu:18.568698364533876 +Step 28440 | loss:0.5578206181526184 lr:1.118708468110125e-05 tokens_per_second_per_gpu:5366.427264947705 mfu:25.314353015664118 +Step 28441 | loss:0.4624270498752594 lr:1.1184351207937022e-05 tokens_per_second_per_gpu:4045.783909246622 mfu:19.084652981830846 +Step 28442 | loss:0.4624357223510742 lr:1.1181618026710261e-05 tokens_per_second_per_gpu:4940.01914848364 mfu:23.30291317757643 +Step 28443 | loss:0.3806769847869873 lr:1.1178885137441492e-05 tokens_per_second_per_gpu:4036.5189236679394 mfu:19.04094846408666 +Step 28444 | loss:0.3215082585811615 lr:1.1176152540151274e-05 tokens_per_second_per_gpu:5329.271260372562 mfu:25.13908180261508 +Step 28445 | loss:0.3597370684146881 lr:1.1173420234860188e-05 tokens_per_second_per_gpu:4868.525021935183 mfu:22.965663184491536 +Step 28446 | loss:0.4288429021835327 lr:1.1170688221588759e-05 tokens_per_second_per_gpu:3762.266401208671 mfu:17.74725252334089 +Step 28447 | loss:0.41768547892570496 lr:1.1167956500357524e-05 tokens_per_second_per_gpu:5376.8066452847215 mfu:25.363314323618717 +Step 28448 | loss:0.45443227887153625 lr:1.1165225071187047e-05 tokens_per_second_per_gpu:4981.437219007224 mfu:23.498289282887697 +Step 28449 | loss:0.5801761746406555 lr:1.1162493934097878e-05 tokens_per_second_per_gpu:4081.816428877124 mfu:19.254624524709715 +Step 28450 | loss:0.4512157738208771 lr:1.1159763089110536e-05 tokens_per_second_per_gpu:5332.711090581024 mfu:25.15530806860395 +Step 28451 | loss:0.4443332850933075 lr:1.115703253624557e-05 tokens_per_second_per_gpu:5209.7071139720865 mfu:24.57507732429711 +Step 28452 | loss:0.3713819980621338 lr:1.1154302275523532e-05 tokens_per_second_per_gpu:5352.10787704563 mfu:25.246806019790522 +Step 28453 | loss:0.41799238324165344 lr:1.115157230696493e-05 tokens_per_second_per_gpu:3838.703602675737 mfu:18.10781984419241 +Step 28454 | loss:0.375613272190094 lr:1.1148842630590322e-05 tokens_per_second_per_gpu:3308.3132262835256 mfu:15.605877944820698 +Step 28455 | loss:0.37583667039871216 lr:1.114611324642022e-05 tokens_per_second_per_gpu:5354.282809875928 mfu:25.257065549032845 +Step 28456 | loss:0.4556322991847992 lr:1.1143384154475162e-05 tokens_per_second_per_gpu:4273.448314201734 mfu:20.158584823557018 +Step 28457 | loss:0.3992450535297394 lr:1.1140655354775664e-05 tokens_per_second_per_gpu:5344.277470454796 mfu:25.20986865589643 +Step 28458 | loss:0.5072561502456665 lr:1.113792684734225e-05 tokens_per_second_per_gpu:4393.395731131048 mfu:20.72439725435222 +Step 28459 | loss:0.3954453766345978 lr:1.1135198632195448e-05 tokens_per_second_per_gpu:4007.515497875095 mfu:18.904134356127102 +Step 28460 | loss:0.3492374122142792 lr:1.113247070935579e-05 tokens_per_second_per_gpu:4009.9669699665333 mfu:18.91569836824668 +Step 28461 | loss:0.42947208881378174 lr:1.1129743078843786e-05 tokens_per_second_per_gpu:4047.2565773970864 mfu:19.091599808759685 +Step 28462 | loss:0.3536425530910492 lr:1.1127015740679925e-05 tokens_per_second_per_gpu:5316.997605732215 mfu:25.081184879575147 +Step 28463 | loss:0.46789926290512085 lr:1.1124288694884739e-05 tokens_per_second_per_gpu:3935.360045496568 mfu:18.563764776267945 +Step 28464 | loss:0.3475433588027954 lr:1.1121561941478748e-05 tokens_per_second_per_gpu:4940.047183647346 mfu:23.303045424226923 +Step 28465 | loss:0.37124180793762207 lr:1.1118835480482443e-05 tokens_per_second_per_gpu:4270.878055470313 mfu:20.146460474589546 +Step 28466 | loss:0.5891218781471252 lr:1.1116109311916334e-05 tokens_per_second_per_gpu:4782.626969030207 mfu:22.560467413219687 +Step 28467 | loss:0.3866531252861023 lr:1.1113383435800945e-05 tokens_per_second_per_gpu:3490.3037687782976 mfu:16.46435838455694 +Step 28468 | loss:0.43831121921539307 lr:1.1110657852156759e-05 tokens_per_second_per_gpu:5383.280260617191 mfu:25.393851471654646 +Step 28469 | loss:0.4897209405899048 lr:1.1107932561004265e-05 tokens_per_second_per_gpu:3986.8065711879285 mfu:18.8064468156365 +Step 28470 | loss:0.36535677313804626 lr:1.1105207562363974e-05 tokens_per_second_per_gpu:3899.803963721793 mfu:18.396040672043224 +Step 28471 | loss:0.3527509570121765 lr:1.110248285625639e-05 tokens_per_second_per_gpu:5336.710661821021 mfu:25.174174728540233 +Step 28472 | loss:0.45560726523399353 lr:1.1099758442701985e-05 tokens_per_second_per_gpu:3458.0518468331293 mfu:16.312220565996274 +Step 28473 | loss:0.3992534279823303 lr:1.1097034321721255e-05 tokens_per_second_per_gpu:5382.657440315792 mfu:25.39091352202885 +Step 28474 | loss:0.45820170640945435 lr:1.1094310493334715e-05 tokens_per_second_per_gpu:5260.171621484832 mfu:24.81312701636756 +Step 28475 | loss:0.4817999005317688 lr:1.1091586957562828e-05 tokens_per_second_per_gpu:4577.301902161776 mfu:21.591914040732497 +Step 28476 | loss:0.4232337772846222 lr:1.1088863714426062e-05 tokens_per_second_per_gpu:5229.883223427363 mfu:24.67025147499689 +Step 28477 | loss:0.45875242352485657 lr:1.1086140763944924e-05 tokens_per_second_per_gpu:5313.985243293952 mfu:25.06697505199183 +Step 28478 | loss:0.35214707255363464 lr:1.1083418106139898e-05 tokens_per_second_per_gpu:4940.322539064581 mfu:23.3043443227105 +Step 28479 | loss:0.31086161732673645 lr:1.1080695741031433e-05 tokens_per_second_per_gpu:3614.9186378274235 mfu:17.052187983350745 +Step 28480 | loss:0.39806562662124634 lr:1.1077973668640023e-05 tokens_per_second_per_gpu:5327.959510746478 mfu:25.132894055821282 +Step 28481 | loss:0.4814881384372711 lr:1.107525188898616e-05 tokens_per_second_per_gpu:4506.304305155304 mfu:21.257006262213824 +Step 28482 | loss:0.36894556879997253 lr:1.1072530402090275e-05 tokens_per_second_per_gpu:4989.110131991056 mfu:23.53448372256638 +Step 28483 | loss:0.5162513852119446 lr:1.106980920797287e-05 tokens_per_second_per_gpu:4992.417815096463 mfu:23.550086628123516 +Step 28484 | loss:0.454872727394104 lr:1.1067088306654382e-05 tokens_per_second_per_gpu:5336.404247124724 mfu:25.172729318138064 +Step 28485 | loss:0.2903180420398712 lr:1.1064367698155304e-05 tokens_per_second_per_gpu:5340.494528575503 mfu:25.192023873615366 +Step 28486 | loss:0.3773459494113922 lr:1.1061647382496077e-05 tokens_per_second_per_gpu:3896.751214097088 mfu:18.381640331210694 +Step 28487 | loss:0.3773220181465149 lr:1.1058927359697163e-05 tokens_per_second_per_gpu:4736.551551757437 mfu:22.343121808667217 +Step 28488 | loss:0.4912571907043457 lr:1.1056207629779042e-05 tokens_per_second_per_gpu:5381.035120902443 mfu:25.383260764559846 +Step 28489 | loss:0.4390332102775574 lr:1.1053488192762134e-05 tokens_per_second_per_gpu:3943.1297259971143 mfu:18.60041568483321 +Step 28490 | loss:0.41253718733787537 lr:1.1050769048666931e-05 tokens_per_second_per_gpu:4521.771212597543 mfu:21.32996630354539 +Step 28491 | loss:0.43489962816238403 lr:1.1048050197513848e-05 tokens_per_second_per_gpu:5175.263621636807 mfu:24.41260149428582 +Step 28492 | loss:0.46958935260772705 lr:1.1045331639323364e-05 tokens_per_second_per_gpu:3308.969196118902 mfu:15.608972266454163 +Step 28493 | loss:0.3260532021522522 lr:1.1042613374115895e-05 tokens_per_second_per_gpu:5362.183219636505 mfu:25.29433312982239 +Step 28494 | loss:0.5018007159233093 lr:1.1039895401911904e-05 tokens_per_second_per_gpu:4081.805662782231 mfu:19.25457373920322 +Step 28495 | loss:0.49739870429039 lr:1.1037177722731828e-05 tokens_per_second_per_gpu:5381.835755414597 mfu:25.387037494156008 +Step 28496 | loss:0.4101344347000122 lr:1.1034460336596131e-05 tokens_per_second_per_gpu:5112.068324465914 mfu:24.11449849530119 +Step 28497 | loss:0.4565468728542328 lr:1.1031743243525222e-05 tokens_per_second_per_gpu:5328.245229280225 mfu:25.134241838893573 +Step 28498 | loss:0.31409621238708496 lr:1.1029026443539536e-05 tokens_per_second_per_gpu:5040.755181209729 mfu:23.778102231284876 +Step 28499 | loss:0.4895815849304199 lr:1.1026309936659524e-05 tokens_per_second_per_gpu:4374.327726282507 mfu:20.63445022214455 +Step 28500 | loss:0.3433994650840759 lr:1.1023593722905602e-05 tokens_per_second_per_gpu:5341.334102993059 mfu:25.195984289389145 +Step 28501 | loss:0.29905471205711365 lr:1.1020877802298197e-05 tokens_per_second_per_gpu:4751.33645972151 mfu:22.412864742099167 +Step 28502 | loss:0.39986443519592285 lr:1.1018162174857743e-05 tokens_per_second_per_gpu:4141.508963726582 mfu:19.536204396191255 +Step 28503 | loss:0.3877766728401184 lr:1.1015446840604688e-05 tokens_per_second_per_gpu:5326.811409009832 mfu:25.127478264043162 +Step 28504 | loss:0.44797757267951965 lr:1.1012731799559422e-05 tokens_per_second_per_gpu:5153.200390842876 mfu:24.308525470255503 +Step 28505 | loss:0.4805983006954193 lr:1.1010017051742366e-05 tokens_per_second_per_gpu:3991.7444451138085 mfu:18.829739609433044 +Step 28506 | loss:0.342317134141922 lr:1.1007302597173946e-05 tokens_per_second_per_gpu:4462.419098223824 mfu:21.049992253529634 +Step 28507 | loss:0.36435338854789734 lr:1.1004588435874596e-05 tokens_per_second_per_gpu:5162.619930185778 mfu:24.352959044475107 +Step 28508 | loss:0.4451088607311249 lr:1.1001874567864696e-05 tokens_per_second_per_gpu:4736.273542376599 mfu:22.341810390984513 +Step 28509 | loss:0.465945303440094 lr:1.099916099316467e-05 tokens_per_second_per_gpu:5270.810203898122 mfu:24.86331102474038 +Step 28510 | loss:0.44289523363113403 lr:1.0996447711794949e-05 tokens_per_second_per_gpu:4747.202953560095 mfu:22.393366288286895 +Step 28511 | loss:0.44758355617523193 lr:1.0993734723775922e-05 tokens_per_second_per_gpu:4176.161550632294 mfu:19.699666560966488 +Step 28512 | loss:0.46963807940483093 lr:1.0991022029127978e-05 tokens_per_second_per_gpu:5188.229395845 mfu:24.47376326341506 +Step 28513 | loss:0.37547340989112854 lr:1.0988309627871535e-05 tokens_per_second_per_gpu:4619.601595543058 mfu:21.791448911483894 +Step 28514 | loss:0.3221126198768616 lr:1.0985597520027008e-05 tokens_per_second_per_gpu:5303.449705354659 mfu:25.01727712198406 +Step 28515 | loss:0.3224007785320282 lr:1.0982885705614764e-05 tokens_per_second_per_gpu:5045.701377677097 mfu:23.801434283929595 +Step 28516 | loss:0.4434128701686859 lr:1.0980174184655218e-05 tokens_per_second_per_gpu:4929.774694819858 mfu:23.25458834176034 +Step 28517 | loss:0.396882027387619 lr:1.0977462957168771e-05 tokens_per_second_per_gpu:4411.550265760814 mfu:20.810035291683835 +Step 28518 | loss:0.4169045686721802 lr:1.0974752023175795e-05 tokens_per_second_per_gpu:5304.963608452414 mfu:25.024418461194664 +Step 28519 | loss:0.36681121587753296 lr:1.0972041382696695e-05 tokens_per_second_per_gpu:4775.679662788003 mfu:22.52769578434254 +Step 28520 | loss:0.30524781346321106 lr:1.0969331035751846e-05 tokens_per_second_per_gpu:5235.7691648591845 mfu:24.698016464975957 +Step 28521 | loss:0.45546048879623413 lr:1.0966620982361647e-05 tokens_per_second_per_gpu:4539.52507685951 mfu:21.413714310390684 +Step 28522 | loss:0.40509718656539917 lr:1.0963911222546464e-05 tokens_per_second_per_gpu:4321.346385991275 mfu:20.384528200442027 +Step 28523 | loss:0.44617611169815063 lr:1.0961201756326683e-05 tokens_per_second_per_gpu:5325.120193298414 mfu:25.11950051098125 +Step 28524 | loss:0.4056500494480133 lr:1.0958492583722696e-05 tokens_per_second_per_gpu:5063.276855999956 mfu:23.88434081386433 +Step 28525 | loss:0.3760330379009247 lr:1.0955783704754858e-05 tokens_per_second_per_gpu:4896.028711196157 mfu:23.095402779348824 +Step 28526 | loss:0.3680649995803833 lr:1.0953075119443567e-05 tokens_per_second_per_gpu:5349.337441624828 mfu:25.23373740322558 +Step 28527 | loss:0.3212733566761017 lr:1.0950366827809167e-05 tokens_per_second_per_gpu:5321.336623873313 mfu:25.101652768459545 +Step 28528 | loss:0.3952792286872864 lr:1.0947658829872048e-05 tokens_per_second_per_gpu:4819.532673130827 mfu:22.73455791622473 +Step 28529 | loss:0.4731450080871582 lr:1.0944951125652564e-05 tokens_per_second_per_gpu:5100.590392583178 mfu:24.06035513227321 +Step 28530 | loss:0.40811190009117126 lr:1.0942243715171086e-05 tokens_per_second_per_gpu:5336.623713858354 mfu:25.17376458016548 +Step 28531 | loss:0.3499843180179596 lr:1.0939536598447986e-05 tokens_per_second_per_gpu:5358.814594136122 mfu:25.278442748590304 +Step 28532 | loss:0.46773213148117065 lr:1.0936829775503604e-05 tokens_per_second_per_gpu:5305.760664605148 mfu:25.02817831105892 +Step 28533 | loss:0.4018370509147644 lr:1.0934123246358319e-05 tokens_per_second_per_gpu:5114.295394531296 mfu:24.12500396477693 +Step 28534 | loss:0.47484666109085083 lr:1.0931417011032469e-05 tokens_per_second_per_gpu:3141.027495953343 mfu:14.816762612964487 +Step 28535 | loss:0.40740320086479187 lr:1.0928711069546432e-05 tokens_per_second_per_gpu:3844.4685141824493 mfu:18.1350139153649 +Step 28536 | loss:0.36455368995666504 lr:1.0926005421920527e-05 tokens_per_second_per_gpu:4877.5200109589005 mfu:23.008094082419795 +Step 28537 | loss:0.5051164031028748 lr:1.0923300068175125e-05 tokens_per_second_per_gpu:4001.232501985849 mfu:18.874496392527888 +Step 28538 | loss:0.3182277977466583 lr:1.0920595008330564e-05 tokens_per_second_per_gpu:4616.495711939839 mfu:21.776797928609124 +Step 28539 | loss:0.41348177194595337 lr:1.0917890242407215e-05 tokens_per_second_per_gpu:5315.3565353566755 mfu:25.07344367061825 +Step 28540 | loss:0.4770061671733856 lr:1.0915185770425396e-05 tokens_per_second_per_gpu:3601.1549219376384 mfu:16.987262187166547 +Step 28541 | loss:0.3803004324436188 lr:1.0912481592405439e-05 tokens_per_second_per_gpu:3868.8728864232676 mfu:18.250133503040104 +Step 28542 | loss:0.419742226600647 lr:1.0909777708367691e-05 tokens_per_second_per_gpu:4378.883611617099 mfu:20.65594111058201 +Step 28543 | loss:0.4214761555194855 lr:1.0907074118332512e-05 tokens_per_second_per_gpu:4122.155324553854 mfu:19.44490997814216 +Step 28544 | loss:0.3825518488883972 lr:1.0904370822320203e-05 tokens_per_second_per_gpu:4789.68042828281 mfu:22.59373978395846 +Step 28545 | loss:0.513377845287323 lr:1.0901667820351109e-05 tokens_per_second_per_gpu:4955.3156706735435 mfu:23.37506948260201 +Step 28546 | loss:0.4762651324272156 lr:1.0898965112445575e-05 tokens_per_second_per_gpu:4819.797816779609 mfu:22.735808644054213 +Step 28547 | loss:0.39029306173324585 lr:1.0896262698623898e-05 tokens_per_second_per_gpu:4920.645534209107 mfu:23.211524533564013 +Step 28548 | loss:0.3068001866340637 lr:1.089356057890643e-05 tokens_per_second_per_gpu:3540.8078623271635 mfu:16.702594810713848 +Step 28549 | loss:0.36023589968681335 lr:1.0890858753313476e-05 tokens_per_second_per_gpu:4042.6427789520208 mfu:19.069835734299122 +Step 28550 | loss:0.3440715968608856 lr:1.0888157221865375e-05 tokens_per_second_per_gpu:4202.732623741897 mfu:19.825006846317038 +Step 28551 | loss:0.6221956014633179 lr:1.0885455984582415e-05 tokens_per_second_per_gpu:5333.891204310253 mfu:25.160874866412886 +Step 28552 | loss:0.4385448098182678 lr:1.0882755041484938e-05 tokens_per_second_per_gpu:4268.144217835034 mfu:20.133564496023137 +Step 28553 | loss:0.5385099053382874 lr:1.0880054392593264e-05 tokens_per_second_per_gpu:3975.078995550416 mfu:18.75112584042362 +Step 28554 | loss:0.4040774405002594 lr:1.0877354037927678e-05 tokens_per_second_per_gpu:4465.809458713996 mfu:21.065985162416922 +Step 28555 | loss:0.4624089300632477 lr:1.087465397750852e-05 tokens_per_second_per_gpu:5272.359369700102 mfu:24.87061870414326 +Step 28556 | loss:0.34662002325057983 lr:1.0871954211356067e-05 tokens_per_second_per_gpu:5326.703051398509 mfu:25.12696712270284 +Step 28557 | loss:0.556644082069397 lr:1.0869254739490653e-05 tokens_per_second_per_gpu:4601.743957515165 mfu:21.7072113860791 +Step 28558 | loss:0.4022340476512909 lr:1.086655556193255e-05 tokens_per_second_per_gpu:4300.654291114181 mfu:20.286920058471345 +Step 28559 | loss:0.42252257466316223 lr:1.0863856678702084e-05 tokens_per_second_per_gpu:4718.385100918077 mfu:22.25742756054183 +Step 28560 | loss:0.4355086088180542 lr:1.086115808981955e-05 tokens_per_second_per_gpu:4573.261413301166 mfu:21.572854365398516 +Step 28561 | loss:0.4644906222820282 lr:1.0858459795305232e-05 tokens_per_second_per_gpu:4212.252450775202 mfu:19.869913494683118 +Step 28562 | loss:0.42394551634788513 lr:1.0855761795179447e-05 tokens_per_second_per_gpu:3602.2879164042643 mfu:16.992606715374407 +Step 28563 | loss:0.27775201201438904 lr:1.0853064089462455e-05 tokens_per_second_per_gpu:5321.542995618638 mfu:25.10262625919289 +Step 28564 | loss:0.2550577223300934 lr:1.0850366678174579e-05 tokens_per_second_per_gpu:5366.764468767469 mfu:25.315943663614853 +Step 28565 | loss:0.39439791440963745 lr:1.0847669561336078e-05 tokens_per_second_per_gpu:4108.752130071737 mfu:19.38168482294922 +Step 28566 | loss:0.4568309783935547 lr:1.0844972738967246e-05 tokens_per_second_per_gpu:5079.565117281052 mfu:23.961175321391206 +Step 28567 | loss:0.45529910922050476 lr:1.0842276211088387e-05 tokens_per_second_per_gpu:5372.165115069899 mfu:25.341419433668687 +Step 28568 | loss:0.3410365581512451 lr:1.0839579977719754e-05 tokens_per_second_per_gpu:5334.735468980249 mfu:25.164857406907664 +Step 28569 | loss:0.39123594760894775 lr:1.0836884038881646e-05 tokens_per_second_per_gpu:3907.0453638362424 mfu:18.43019959189334 +Step 28570 | loss:0.3180498480796814 lr:1.083418839459432e-05 tokens_per_second_per_gpu:5343.928557762499 mfu:25.208222775196223 +Step 28571 | loss:0.47009339928627014 lr:1.0831493044878071e-05 tokens_per_second_per_gpu:4085.7659225868806 mfu:19.27325496039175 +Step 28572 | loss:0.5358081459999084 lr:1.0828797989753148e-05 tokens_per_second_per_gpu:4940.833037807611 mfu:23.30675243238144 +Step 28573 | loss:0.3889678716659546 lr:1.0826103229239838e-05 tokens_per_second_per_gpu:5135.804531925804 mfu:24.226466235704095 +Step 28574 | loss:0.4483867585659027 lr:1.08234087633584e-05 tokens_per_second_per_gpu:4901.105891707904 mfu:23.11935270607906 +Step 28575 | loss:0.39901694655418396 lr:1.0820714592129117e-05 tokens_per_second_per_gpu:4991.370019190523 mfu:23.545143996062784 +Step 28576 | loss:0.37306612730026245 lr:1.0818020715572224e-05 tokens_per_second_per_gpu:5313.399642507307 mfu:25.06421267316743 +Step 28577 | loss:0.3335301876068115 lr:1.0815327133708015e-05 tokens_per_second_per_gpu:3272.358173513495 mfu:15.436271826339885 +Step 28578 | loss:0.3343478739261627 lr:1.0812633846556714e-05 tokens_per_second_per_gpu:5320.2392015507185 mfu:25.096476040124305 +Step 28579 | loss:0.4183111786842346 lr:1.080994085413861e-05 tokens_per_second_per_gpu:4338.876677710961 mfu:20.467221577459977 +Step 28580 | loss:0.45880213379859924 lr:1.0807248156473926e-05 tokens_per_second_per_gpu:5356.984228392481 mfu:25.26980860108495 +Step 28581 | loss:0.3979785144329071 lr:1.080455575358293e-05 tokens_per_second_per_gpu:4578.845136510924 mfu:21.59919374046036 +Step 28582 | loss:0.5252547264099121 lr:1.0801863645485889e-05 tokens_per_second_per_gpu:4543.821148720745 mfu:21.433979614345578 +Step 28583 | loss:0.47694259881973267 lr:1.0799171832203014e-05 tokens_per_second_per_gpu:4680.766910259801 mfu:22.07997613688158 +Step 28584 | loss:0.40903085470199585 lr:1.0796480313754586e-05 tokens_per_second_per_gpu:4192.473456644742 mfu:19.776612604725354 +Step 28585 | loss:0.3718595504760742 lr:1.0793789090160816e-05 tokens_per_second_per_gpu:5345.931004958352 mfu:25.21766866027178 +Step 28586 | loss:0.3994266092777252 lr:1.0791098161441976e-05 tokens_per_second_per_gpu:4519.387160924313 mfu:21.31872032504161 +Step 28587 | loss:0.2783215343952179 lr:1.0788407527618277e-05 tokens_per_second_per_gpu:3489.0400297301235 mfu:16.45839711185059 +Step 28588 | loss:0.3909505009651184 lr:1.0785717188709964e-05 tokens_per_second_per_gpu:4731.473659387127 mfu:22.319168523978686 +Step 28589 | loss:0.4159446656703949 lr:1.0783027144737296e-05 tokens_per_second_per_gpu:4239.316784703734 mfu:19.997580575475528 +Step 28590 | loss:0.39019298553466797 lr:1.078033739572047e-05 tokens_per_second_per_gpu:4752.116297939736 mfu:22.41654337202866 +Step 28591 | loss:0.4059300124645233 lr:1.0777647941679742e-05 tokens_per_second_per_gpu:4186.753491561719 mfu:19.749630553502055 +Step 28592 | loss:0.4159241020679474 lr:1.0774958782635314e-05 tokens_per_second_per_gpu:5367.933772356489 mfu:25.32145946814802 +Step 28593 | loss:0.46449241042137146 lr:1.0772269918607447e-05 tokens_per_second_per_gpu:4530.5489998901685 mfu:21.37137262384965 +Step 28594 | loss:0.49532440304756165 lr:1.076958134961632e-05 tokens_per_second_per_gpu:4284.291231494648 mfu:20.20973271442015 +Step 28595 | loss:0.4648587107658386 lr:1.0766893075682183e-05 tokens_per_second_per_gpu:4506.843312072444 mfu:21.259548849806034 +Step 28596 | loss:0.3762994408607483 lr:1.0764205096825264e-05 tokens_per_second_per_gpu:4277.202857849219 mfu:20.176295646533408 +Step 28597 | loss:0.47619372606277466 lr:1.0761517413065747e-05 tokens_per_second_per_gpu:4136.635907981709 mfu:19.513217360813645 +Step 28598 | loss:0.2743052840232849 lr:1.0758830024423883e-05 tokens_per_second_per_gpu:4186.078053777943 mfu:19.746444398234413 +Step 28599 | loss:0.37511059641838074 lr:1.075614293091985e-05 tokens_per_second_per_gpu:5391.451958403189 mfu:25.43239876434661 +Step 28600 | loss:0.40463146567344666 lr:1.0753456132573886e-05 tokens_per_second_per_gpu:4341.917252090047 mfu:20.481564485581604 +Step 28601 | loss:0.4427914321422577 lr:1.075076962940617e-05 tokens_per_second_per_gpu:4771.52768145785 mfu:22.508110180007147 +Step 28602 | loss:0.3701176643371582 lr:1.0748083421436933e-05 tokens_per_second_per_gpu:3460.156032025104 mfu:16.322146366557835 +Step 28603 | loss:0.3701251149177551 lr:1.0745397508686372e-05 tokens_per_second_per_gpu:4221.576262439543 mfu:19.91389550511012 +Step 28604 | loss:0.3554688096046448 lr:1.0742711891174678e-05 tokens_per_second_per_gpu:5123.624872494808 mfu:24.1690127041039 +Step 28605 | loss:0.4019874632358551 lr:1.0740026568922058e-05 tokens_per_second_per_gpu:5371.570807627176 mfu:25.338615984062056 +Step 28606 | loss:0.4398798942565918 lr:1.073734154194872e-05 tokens_per_second_per_gpu:5337.704599887326 mfu:25.178863303982318 +Step 28607 | loss:0.32205918431282043 lr:1.0734656810274845e-05 tokens_per_second_per_gpu:4502.225393125003 mfu:21.23776533823274 +Step 28608 | loss:0.4167273938655853 lr:1.0731972373920613e-05 tokens_per_second_per_gpu:5354.0495187276965 mfu:25.255965074883065 +Step 28609 | loss:0.38847094774246216 lr:1.0729288232906227e-05 tokens_per_second_per_gpu:4095.3734298136615 mfu:19.318575211190236 +Step 28610 | loss:0.4198274612426758 lr:1.0726604387251877e-05 tokens_per_second_per_gpu:4462.395906275603 mfu:21.049882853152923 +Step 28611 | loss:0.35102078318595886 lr:1.0723920836977757e-05 tokens_per_second_per_gpu:4456.2341974448 mfu:21.02081702130209 +Step 28612 | loss:0.3173050582408905 lr:1.0721237582104026e-05 tokens_per_second_per_gpu:3850.1942588286665 mfu:18.16202322977388 +Step 28613 | loss:0.47420695424079895 lr:1.0718554622650901e-05 tokens_per_second_per_gpu:5349.604727769855 mfu:25.23499823757539 +Step 28614 | loss:0.40114158391952515 lr:1.0715871958638517e-05 tokens_per_second_per_gpu:4829.625260092911 mfu:22.782166370897784 +Step 28615 | loss:0.3932873606681824 lr:1.0713189590087091e-05 tokens_per_second_per_gpu:5293.859413785373 mfu:24.972038080379267 +Step 28616 | loss:0.3682275712490082 lr:1.071050751701676e-05 tokens_per_second_per_gpu:3876.192299478776 mfu:18.28466042324358 +Step 28617 | loss:0.46654075384140015 lr:1.070782573944772e-05 tokens_per_second_per_gpu:5319.398711149024 mfu:25.092511303497062 +Step 28618 | loss:0.3890116214752197 lr:1.0705144257400152e-05 tokens_per_second_per_gpu:3448.3193876129353 mfu:16.266310895325848 +Step 28619 | loss:0.42131659388542175 lr:1.0702463070894191e-05 tokens_per_second_per_gpu:4219.791829504114 mfu:19.905478030498045 +Step 28620 | loss:0.326290100812912 lr:1.069978217995003e-05 tokens_per_second_per_gpu:5350.737752498928 mfu:25.240342908536377 +Step 28621 | loss:0.41876861453056335 lr:1.0697101584587816e-05 tokens_per_second_per_gpu:5341.9709658318525 mfu:25.19898848005974 +Step 28622 | loss:0.38536912202835083 lr:1.0694421284827722e-05 tokens_per_second_per_gpu:5338.847857848296 mfu:25.184256246844136 +Step 28623 | loss:0.39764103293418884 lr:1.0691741280689893e-05 tokens_per_second_per_gpu:5321.4649147867485 mfu:25.1022579385869 +Step 28624 | loss:0.34778228402137756 lr:1.0689061572194492e-05 tokens_per_second_per_gpu:4934.974766311129 mfu:23.279117966208354 +Step 28625 | loss:0.44728323817253113 lr:1.0686382159361686e-05 tokens_per_second_per_gpu:3464.4346955888236 mfu:16.342329552603182 +Step 28626 | loss:0.43492546677589417 lr:1.0683703042211606e-05 tokens_per_second_per_gpu:5361.539191953194 mfu:25.29129514135776 +Step 28627 | loss:0.3696245551109314 lr:1.0681024220764424e-05 tokens_per_second_per_gpu:4746.248723969078 mfu:22.388865024497715 +Step 28628 | loss:0.35501936078071594 lr:1.0678345695040265e-05 tokens_per_second_per_gpu:5356.398943841711 mfu:25.26704771399913 +Step 28629 | loss:0.3777686059474945 lr:1.0675667465059298e-05 tokens_per_second_per_gpu:4443.299953589906 mfu:20.959803986228994 +Step 28630 | loss:0.430798202753067 lr:1.0672989530841637e-05 tokens_per_second_per_gpu:5313.635556069778 mfu:25.06532551769261 +Step 28631 | loss:0.4770226776599884 lr:1.0670311892407442e-05 tokens_per_second_per_gpu:4204.274086715415 mfu:19.832278190163827 +Step 28632 | loss:0.4996548593044281 lr:1.0667634549776862e-05 tokens_per_second_per_gpu:4776.425721063708 mfu:22.53121506851971 +Step 28633 | loss:0.4258364737033844 lr:1.0664957502970013e-05 tokens_per_second_per_gpu:5130.568314422133 mfu:24.201766104348444 +Step 28634 | loss:0.43660736083984375 lr:1.0662280752007048e-05 tokens_per_second_per_gpu:3988.973305914149 mfu:18.816667673023147 +Step 28635 | loss:0.3410118818283081 lr:1.0659604296908072e-05 tokens_per_second_per_gpu:5256.7205666117825 mfu:24.79684780172071 +Step 28636 | loss:0.4465380907058716 lr:1.065692813769325e-05 tokens_per_second_per_gpu:4267.152351351897 mfu:20.128885692592306 +Step 28637 | loss:0.47240325808525085 lr:1.0654252274382675e-05 tokens_per_second_per_gpu:3963.8343670754152 mfu:18.69808300937559 +Step 28638 | loss:0.3244134783744812 lr:1.0651576706996491e-05 tokens_per_second_per_gpu:3110.0472985433776 mfu:14.670623736014951 +Step 28639 | loss:0.28943321108818054 lr:1.0648901435554836e-05 tokens_per_second_per_gpu:5319.301989276293 mfu:25.092055049920937 +Step 28640 | loss:0.41787633299827576 lr:1.0646226460077796e-05 tokens_per_second_per_gpu:5315.273879138451 mfu:25.073053767134276 +Step 28641 | loss:0.5215972065925598 lr:1.0643551780585508e-05 tokens_per_second_per_gpu:5375.112762755176 mfu:25.35532399072069 +Step 28642 | loss:0.4323091506958008 lr:1.064087739709811e-05 tokens_per_second_per_gpu:5067.624912150978 mfu:23.904851336583583 +Step 28643 | loss:0.37129831314086914 lr:1.0638203309635686e-05 tokens_per_second_per_gpu:5185.310651499405 mfu:24.459995048347754 +Step 28644 | loss:0.26728975772857666 lr:1.0635529518218351e-05 tokens_per_second_per_gpu:4217.7287331005155 mfu:19.89574605276173 +Step 28645 | loss:0.40864768624305725 lr:1.0632856022866216e-05 tokens_per_second_per_gpu:5357.361733117305 mfu:25.2715893552809 +Step 28646 | loss:0.4591355621814728 lr:1.0630182823599399e-05 tokens_per_second_per_gpu:5319.382197331096 mfu:25.09243340500035 +Step 28647 | loss:0.4112081825733185 lr:1.062750992043801e-05 tokens_per_second_per_gpu:4262.110365819401 mfu:20.105101786584086 +Step 28648 | loss:0.49087080359458923 lr:1.0624837313402136e-05 tokens_per_second_per_gpu:4935.571256236834 mfu:23.281931710959576 +Step 28649 | loss:0.3754318654537201 lr:1.0622165002511892e-05 tokens_per_second_per_gpu:4979.668056623867 mfu:23.489943842075174 +Step 28650 | loss:0.4165151119232178 lr:1.0619492987787355e-05 tokens_per_second_per_gpu:4400.208956027413 mfu:20.756536398644403 +Step 28651 | loss:0.42811518907546997 lr:1.0616821269248656e-05 tokens_per_second_per_gpu:5049.115585276576 mfu:23.81753968370022 +Step 28652 | loss:0.44677597284317017 lr:1.061414984691585e-05 tokens_per_second_per_gpu:3828.624123037925 mfu:18.06027321900429 +Step 28653 | loss:0.31103283166885376 lr:1.0611478720809054e-05 tokens_per_second_per_gpu:4947.320286373697 mfu:23.337353890706407 +Step 28654 | loss:0.46908533573150635 lr:1.0608807890948363e-05 tokens_per_second_per_gpu:5375.380627141635 mfu:25.356587552734 +Step 28655 | loss:0.45751094818115234 lr:1.060613735735384e-05 tokens_per_second_per_gpu:4659.951599582911 mfu:21.981786764960468 +Step 28656 | loss:0.4800035357475281 lr:1.0603467120045601e-05 tokens_per_second_per_gpu:5056.082897043899 mfu:23.850405682052646 +Step 28657 | loss:0.3688582181930542 lr:1.0600797179043697e-05 tokens_per_second_per_gpu:4003.519111706025 mfu:18.885282720713803 +Step 28658 | loss:0.40169328451156616 lr:1.0598127534368234e-05 tokens_per_second_per_gpu:4263.496112588004 mfu:20.111638590524365 +Step 28659 | loss:0.46855980157852173 lr:1.0595458186039276e-05 tokens_per_second_per_gpu:4706.247998513595 mfu:22.200174777696684 +Step 28660 | loss:0.3692915141582489 lr:1.0592789134076902e-05 tokens_per_second_per_gpu:4413.57122795669 mfu:20.81956851517349 +Step 28661 | loss:0.43809109926223755 lr:1.0590120378501201e-05 tokens_per_second_per_gpu:3405.5834371055885 mfu:16.06471812527766 +Step 28662 | loss:0.40859484672546387 lr:1.0587451919332225e-05 tokens_per_second_per_gpu:5146.529975552336 mfu:24.27705998323982 +Step 28663 | loss:0.39865225553512573 lr:1.0584783756590062e-05 tokens_per_second_per_gpu:4440.839198071328 mfu:20.948196182599677 +Step 28664 | loss:0.48645395040512085 lr:1.0582115890294753e-05 tokens_per_second_per_gpu:5087.937217542945 mfu:24.000667946754156 +Step 28665 | loss:0.49132463335990906 lr:1.0579448320466402e-05 tokens_per_second_per_gpu:4269.228434058412 mfu:20.138678928935175 +Step 28666 | loss:0.35166260600090027 lr:1.0576781047125028e-05 tokens_per_second_per_gpu:5170.800782719613 mfu:24.39154952167458 +Step 28667 | loss:0.31864529848098755 lr:1.057411407029072e-05 tokens_per_second_per_gpu:5311.7264896266515 mfu:25.056320125560056 +Step 28668 | loss:0.45225051045417786 lr:1.057144738998354e-05 tokens_per_second_per_gpu:5312.483593548013 mfu:25.05989151393237 +Step 28669 | loss:0.3910696506500244 lr:1.0568781006223526e-05 tokens_per_second_per_gpu:3322.002390893776 mfu:15.670452069899573 +Step 28670 | loss:0.36767980456352234 lr:1.056611491903074e-05 tokens_per_second_per_gpu:4306.041558263495 mfu:20.312332716777213 +Step 28671 | loss:0.34387052059173584 lr:1.0563449128425245e-05 tokens_per_second_per_gpu:4481.505790671522 mfu:21.14002743833096 +Step 28672 | loss:0.4269934296607971 lr:1.0560783634427085e-05 tokens_per_second_per_gpu:5366.636633286924 mfu:25.315340641842134 +Step 28673 | loss:0.4486256539821625 lr:1.0558118437056291e-05 tokens_per_second_per_gpu:5218.109141420276 mfu:24.61471112130392 +Step 28674 | loss:0.37905165553092957 lr:1.0555453536332921e-05 tokens_per_second_per_gpu:5277.341884806469 mfu:24.89412207800473 +Step 28675 | loss:0.40723174810409546 lr:1.055278893227703e-05 tokens_per_second_per_gpu:5294.31354395545 mfu:24.974180289874116 +Step 28676 | loss:0.42565444111824036 lr:1.0550124624908636e-05 tokens_per_second_per_gpu:4187.141571626411 mfu:19.751461193380496 +Step 28677 | loss:0.48879337310791016 lr:1.054746061424779e-05 tokens_per_second_per_gpu:4102.541024556517 mfu:19.352385978510053 +Step 28678 | loss:0.3798831105232239 lr:1.0544796900314535e-05 tokens_per_second_per_gpu:4438.748312332593 mfu:20.93833311782817 +Step 28679 | loss:0.36788108944892883 lr:1.0542133483128896e-05 tokens_per_second_per_gpu:4075.6304638405786 mfu:19.225444272197866 +Step 28680 | loss:0.47953277826309204 lr:1.0539470362710896e-05 tokens_per_second_per_gpu:3525.7432966729543 mfu:16.631532684243005 +Step 28681 | loss:0.4414788782596588 lr:1.0536807539080578e-05 tokens_per_second_per_gpu:4400.03317480525 mfu:20.755707208628095 +Step 28682 | loss:0.4224119484424591 lr:1.0534145012257962e-05 tokens_per_second_per_gpu:3751.253368939305 mfu:17.695302171109226 +Step 28683 | loss:0.3401787579059601 lr:1.053148278226309e-05 tokens_per_second_per_gpu:4076.6455028408145 mfu:19.230232384345488 +Step 28684 | loss:0.5144479274749756 lr:1.0528820849115962e-05 tokens_per_second_per_gpu:4677.466835331201 mfu:22.06440912039253 +Step 28685 | loss:0.2818921208381653 lr:1.0526159212836617e-05 tokens_per_second_per_gpu:5363.102713637376 mfu:25.29867053990635 +Step 28686 | loss:0.4363601505756378 lr:1.0523497873445054e-05 tokens_per_second_per_gpu:4307.154430977192 mfu:20.317582327243382 +Step 28687 | loss:0.3198108375072479 lr:1.0520836830961316e-05 tokens_per_second_per_gpu:5372.9440486575995 mfu:25.345093796301644 +Step 28688 | loss:0.3903176486492157 lr:1.051817608540539e-05 tokens_per_second_per_gpu:5392.7499938772 mfu:25.43852181914521 +Step 28689 | loss:0.4729920029640198 lr:1.0515515636797297e-05 tokens_per_second_per_gpu:5338.3023822690175 mfu:25.181683145468945 +Step 28690 | loss:0.43404486775398254 lr:1.0512855485157064e-05 tokens_per_second_per_gpu:5346.838748095127 mfu:25.221950639524902 +Step 28691 | loss:0.3807653486728668 lr:1.0510195630504666e-05 tokens_per_second_per_gpu:3778.2518162674214 mfu:17.822658453566063 +Step 28692 | loss:0.37999093532562256 lr:1.0507536072860142e-05 tokens_per_second_per_gpu:5367.362907480039 mfu:25.318766601126217 +Step 28693 | loss:0.5275479555130005 lr:1.0504876812243464e-05 tokens_per_second_per_gpu:5326.988867586717 mfu:25.128315366427888 +Step 28694 | loss:0.42845988273620605 lr:1.0502217848674662e-05 tokens_per_second_per_gpu:5320.590490676445 mfu:25.09813313086646 +Step 28695 | loss:0.44137218594551086 lr:1.0499559182173707e-05 tokens_per_second_per_gpu:4152.021369518005 mfu:19.585793207910775 +Step 28696 | loss:0.37152788043022156 lr:1.0496900812760608e-05 tokens_per_second_per_gpu:2937.3828392723317 mfu:13.856136022038898 +Step 28697 | loss:0.45384472608566284 lr:1.0494242740455367e-05 tokens_per_second_per_gpu:5103.631555623453 mfu:24.074700817210108 +Step 28698 | loss:0.4266771376132965 lr:1.0491584965277956e-05 tokens_per_second_per_gpu:4370.414326356463 mfu:20.615990046998522 +Step 28699 | loss:0.4077325761318207 lr:1.0488927487248379e-05 tokens_per_second_per_gpu:4935.495733878354 mfu:23.281575459109185 +Step 28700 | loss:0.3232768177986145 lr:1.0486270306386631e-05 tokens_per_second_per_gpu:4958.7630414328 mfu:23.391331318655524 +Step 28701 | loss:0.3908376097679138 lr:1.048361342271269e-05 tokens_per_second_per_gpu:5305.085769375017 mfu:25.024994715109386 +Step 28702 | loss:0.46312054991722107 lr:1.0480956836246514e-05 tokens_per_second_per_gpu:5340.024883328801 mfu:25.189808476856932 +Step 28703 | loss:0.4223436117172241 lr:1.0478300547008108e-05 tokens_per_second_per_gpu:3769.871573787125 mfu:17.783127419969702 +Step 28704 | loss:0.4236959218978882 lr:1.0475644555017462e-05 tokens_per_second_per_gpu:3954.5337287098628 mfu:18.654210311352745 +Step 28705 | loss:0.43361109495162964 lr:1.0472988860294519e-05 tokens_per_second_per_gpu:4422.313936541473 mfu:20.86080936322781 +Step 28706 | loss:0.3581182658672333 lr:1.0470333462859278e-05 tokens_per_second_per_gpu:5329.977831188453 mfu:25.142414817707053 +Step 28707 | loss:0.4697273075580597 lr:1.0467678362731708e-05 tokens_per_second_per_gpu:3726.24631059765 mfu:17.577339610267526 +Step 28708 | loss:0.40876176953315735 lr:1.0465023559931775e-05 tokens_per_second_per_gpu:5325.882308747563 mfu:25.12309553958519 +Step 28709 | loss:0.547096312046051 lr:1.0462369054479436e-05 tokens_per_second_per_gpu:4866.742050782454 mfu:22.957252605357866 +Step 28710 | loss:0.35309353470802307 lr:1.0459714846394658e-05 tokens_per_second_per_gpu:5342.26067288133 mfu:25.20035507764124 +Step 28711 | loss:0.43250352144241333 lr:1.0457060935697426e-05 tokens_per_second_per_gpu:3998.883365873364 mfu:18.86341511668134 +Step 28712 | loss:0.422839492559433 lr:1.0454407322407672e-05 tokens_per_second_per_gpu:5307.59567711379 mfu:25.0368343781476 +Step 28713 | loss:0.4155616760253906 lr:1.0451754006545367e-05 tokens_per_second_per_gpu:4678.631368333769 mfu:22.069902421255463 +Step 28714 | loss:0.3900480568408966 lr:1.0449100988130477e-05 tokens_per_second_per_gpu:3485.733491160737 mfu:16.44279960526514 +Step 28715 | loss:0.46456101536750793 lr:1.0446448267182952e-05 tokens_per_second_per_gpu:5378.444870093809 mfu:25.371042109553162 +Step 28716 | loss:0.27100270986557007 lr:1.0443795843722715e-05 tokens_per_second_per_gpu:4724.370326845921 mfu:22.285660892428314 +Step 28717 | loss:0.2998697757720947 lr:1.0441143717769741e-05 tokens_per_second_per_gpu:3497.525833427941 mfu:16.498426095720603 +Step 28718 | loss:0.36710062623023987 lr:1.0438491889343971e-05 tokens_per_second_per_gpu:4431.867298288809 mfu:20.905874200561584 +Step 28719 | loss:0.3475579023361206 lr:1.043584035846537e-05 tokens_per_second_per_gpu:5359.7799503630395 mfu:25.282996498619433 +Step 28720 | loss:0.4396986663341522 lr:1.0433189125153841e-05 tokens_per_second_per_gpu:5349.477651523443 mfu:25.234398797239617 +Step 28721 | loss:0.4062367379665375 lr:1.0430538189429368e-05 tokens_per_second_per_gpu:5353.8481622392 mfu:25.255015241971865 +Step 28722 | loss:0.30763643980026245 lr:1.042788755131186e-05 tokens_per_second_per_gpu:5371.300759502933 mfu:25.33734212098504 +Step 28723 | loss:0.39283594489097595 lr:1.0425237210821243e-05 tokens_per_second_per_gpu:5322.712199503745 mfu:25.10814159340569 +Step 28724 | loss:0.39336881041526794 lr:1.042258716797747e-05 tokens_per_second_per_gpu:5315.081223749113 mfu:25.072144978792373 +Step 28725 | loss:0.4658670425415039 lr:1.041993742280047e-05 tokens_per_second_per_gpu:4917.5920358846215 mfu:23.19712066098688 +Step 28726 | loss:0.5005396008491516 lr:1.0417287975310185e-05 tokens_per_second_per_gpu:4351.232997479513 mfu:20.525508446013774 +Step 28727 | loss:0.3161657750606537 lr:1.0414638825526512e-05 tokens_per_second_per_gpu:5309.360342170193 mfu:25.04515860430156 +Step 28728 | loss:0.3937319815158844 lr:1.041198997346941e-05 tokens_per_second_per_gpu:4807.899618114435 mfu:22.67968281090856 +Step 28729 | loss:0.4756797254085541 lr:1.0409341419158763e-05 tokens_per_second_per_gpu:3774.684865373283 mfu:17.805832537613835 +Step 28730 | loss:0.44275861978530884 lr:1.0406693162614528e-05 tokens_per_second_per_gpu:4033.1790439415404 mfu:19.025193681574994 +Step 28731 | loss:0.3307991325855255 lr:1.0404045203856594e-05 tokens_per_second_per_gpu:3809.925873174463 mfu:17.972070384147944 +Step 28732 | loss:0.43898072838783264 lr:1.0401397542904883e-05 tokens_per_second_per_gpu:5321.717798039064 mfu:25.103450832034408 +Step 28733 | loss:0.4030705392360687 lr:1.0398750179779332e-05 tokens_per_second_per_gpu:3856.5124760493363 mfu:18.191827338402796 +Step 28734 | loss:0.35872411727905273 lr:1.0396103114499817e-05 tokens_per_second_per_gpu:3238.0273145483243 mfu:15.274327307152635 +Step 28735 | loss:0.3753117024898529 lr:1.0393456347086266e-05 tokens_per_second_per_gpu:5256.338911284458 mfu:24.795047468424602 +Step 28736 | loss:0.35006842017173767 lr:1.0390809877558599e-05 tokens_per_second_per_gpu:3462.3618095676047 mfu:16.332551395570267 +Step 28737 | loss:0.37783282995224 lr:1.0388163705936705e-05 tokens_per_second_per_gpu:5053.922822443019 mfu:23.840216241613565 +Step 28738 | loss:0.3747933804988861 lr:1.0385517832240471e-05 tokens_per_second_per_gpu:4610.550175826268 mfu:21.748751820348346 +Step 28739 | loss:0.4353898763656616 lr:1.0382872256489812e-05 tokens_per_second_per_gpu:3855.9392376890864 mfu:18.18912327525295 +Step 28740 | loss:0.504667341709137 lr:1.0380226978704638e-05 tokens_per_second_per_gpu:5368.486201998721 mfu:25.32406537302465 +Step 28741 | loss:0.4045895040035248 lr:1.0377581998904823e-05 tokens_per_second_per_gpu:4646.070129975019 mfu:21.91630550439711 +Step 28742 | loss:0.2723945677280426 lr:1.0374937317110268e-05 tokens_per_second_per_gpu:5302.948536871672 mfu:25.014913024741645 +Step 28743 | loss:0.4129806458950043 lr:1.0372292933340877e-05 tokens_per_second_per_gpu:5328.701654689032 mfu:25.13639487542112 +Step 28744 | loss:0.44471773505210876 lr:1.0369648847616525e-05 tokens_per_second_per_gpu:4113.813006194613 mfu:19.405557839095334 +Step 28745 | loss:0.41538360714912415 lr:1.0367005059957096e-05 tokens_per_second_per_gpu:4897.120946038979 mfu:23.100555037456896 +Step 28746 | loss:0.37526506185531616 lr:1.0364361570382475e-05 tokens_per_second_per_gpu:3859.4727783622056 mfu:18.205791589492257 +Step 28747 | loss:0.42389291524887085 lr:1.0361718378912561e-05 tokens_per_second_per_gpu:5331.5924374253445 mfu:25.1500311908067 +Step 28748 | loss:0.39042583107948303 lr:1.0359075485567204e-05 tokens_per_second_per_gpu:4118.467031695747 mfu:19.427511671442815 +Step 28749 | loss:0.3271258473396301 lr:1.0356432890366303e-05 tokens_per_second_per_gpu:3988.77610560432 mfu:18.815737445515843 +Step 28750 | loss:0.4581149220466614 lr:1.0353790593329742e-05 tokens_per_second_per_gpu:4472.469137464747 mfu:21.09740000334258 +Step 28751 | loss:0.3853057622909546 lr:1.0351148594477378e-05 tokens_per_second_per_gpu:5086.766498012839 mfu:23.9951454629853 +Step 28752 | loss:0.5022565722465515 lr:1.0348506893829074e-05 tokens_per_second_per_gpu:4449.137729048947 mfu:20.987341769096492 +Step 28753 | loss:0.4966578781604767 lr:1.034586549140471e-05 tokens_per_second_per_gpu:3986.4215361861347 mfu:18.80463053983929 +Step 28754 | loss:0.36568722128868103 lr:1.0343224387224148e-05 tokens_per_second_per_gpu:3764.5378735933646 mfu:17.757967446132643 +Step 28755 | loss:0.5250576138496399 lr:1.0340583581307272e-05 tokens_per_second_per_gpu:4068.576755789174 mfu:19.192170727832696 +Step 28756 | loss:0.5169933438301086 lr:1.0337943073673912e-05 tokens_per_second_per_gpu:4598.9912122888445 mfu:21.694226217179782 +Step 28757 | loss:0.4991886019706726 lr:1.0335302864343959e-05 tokens_per_second_per_gpu:5343.858178458819 mfu:25.207890784012818 +Step 28758 | loss:0.60728919506073 lr:1.0332662953337252e-05 tokens_per_second_per_gpu:2727.7005182955395 mfu:12.867028738498359 +Step 28759 | loss:0.26163583993911743 lr:1.0330023340673638e-05 tokens_per_second_per_gpu:4190.647301582827 mfu:19.767998319724118 +Step 28760 | loss:0.5066114664077759 lr:1.0327384026372977e-05 tokens_per_second_per_gpu:3704.694598329794 mfu:17.475676506398766 +Step 28761 | loss:0.3235073387622833 lr:1.0324745010455122e-05 tokens_per_second_per_gpu:4355.901196116851 mfu:20.547529135462916 +Step 28762 | loss:0.3258785605430603 lr:1.0322106292939943e-05 tokens_per_second_per_gpu:5093.384054727426 mfu:24.026361607079604 +Step 28763 | loss:0.3949821889400482 lr:1.0319467873847244e-05 tokens_per_second_per_gpu:3659.631464955746 mfu:17.263106017710008 +Step 28764 | loss:0.3363186717033386 lr:1.0316829753196894e-05 tokens_per_second_per_gpu:5320.385209712114 mfu:25.097164785533174 +Step 28765 | loss:0.5164480805397034 lr:1.0314191931008743e-05 tokens_per_second_per_gpu:3135.484523801935 mfu:14.790615467597727 +Step 28766 | loss:0.3814292252063751 lr:1.031155440730262e-05 tokens_per_second_per_gpu:4651.973946132518 mfu:21.944154812506937 +Step 28767 | loss:0.33105069398880005 lr:1.0308917182098349e-05 tokens_per_second_per_gpu:3742.8563394726584 mfu:17.655691950434097 +Step 28768 | loss:0.5463686585426331 lr:1.0306280255415774e-05 tokens_per_second_per_gpu:5145.08440710135 mfu:24.270240990217108 +Step 28769 | loss:0.4185357093811035 lr:1.0303643627274745e-05 tokens_per_second_per_gpu:5319.389829349722 mfu:25.092469406534367 +Step 28770 | loss:0.382752001285553 lr:1.0301007297695058e-05 tokens_per_second_per_gpu:4699.680013786121 mfu:22.16919247312344 +Step 28771 | loss:0.533535897731781 lr:1.0298371266696565e-05 tokens_per_second_per_gpu:5313.34694409314 mfu:25.06396408575653 +Step 28772 | loss:0.44920721650123596 lr:1.0295735534299106e-05 tokens_per_second_per_gpu:5305.698658598359 mfu:25.027885818145577 +Step 28773 | loss:0.3476197421550751 lr:1.0293100100522478e-05 tokens_per_second_per_gpu:3561.0283103519023 mfu:16.797978114010725 +Step 28774 | loss:0.3758019506931305 lr:1.0290464965386504e-05 tokens_per_second_per_gpu:5326.156486155459 mfu:25.124388881197728 +Step 28775 | loss:0.4876610338687897 lr:1.0287830128911009e-05 tokens_per_second_per_gpu:3208.591631741732 mfu:15.135474169108575 +Step 28776 | loss:0.3728523552417755 lr:1.028519559111582e-05 tokens_per_second_per_gpu:3545.51124042392 mfu:16.7247814476753 +Step 28777 | loss:0.31225284934043884 lr:1.0282561352020737e-05 tokens_per_second_per_gpu:4762.274292286291 mfu:22.46446036449391 +Step 28778 | loss:0.4463188648223877 lr:1.0279927411645568e-05 tokens_per_second_per_gpu:5340.08651215228 mfu:25.190099190532248 +Step 28779 | loss:0.35579976439476013 lr:1.0277293770010154e-05 tokens_per_second_per_gpu:4745.19235244805 mfu:22.383881939797455 +Step 28780 | loss:0.47570955753326416 lr:1.027466042713428e-05 tokens_per_second_per_gpu:3807.6290288365703 mfu:17.961235777524053 +Step 28781 | loss:0.3425377309322357 lr:1.0272027383037741e-05 tokens_per_second_per_gpu:4093.354797448525 mfu:19.309052977909694 +Step 28782 | loss:0.46782752871513367 lr:1.0269394637740354e-05 tokens_per_second_per_gpu:5161.213949636832 mfu:24.346326794340712 +Step 28783 | loss:0.46416905522346497 lr:1.026676219126193e-05 tokens_per_second_per_gpu:5330.052346049267 mfu:25.142766317019056 +Step 28784 | loss:0.42868730425834656 lr:1.0264130043622243e-05 tokens_per_second_per_gpu:5334.101431495467 mfu:25.161866543913998 +Step 28785 | loss:0.5065117478370667 lr:1.0261498194841108e-05 tokens_per_second_per_gpu:4677.778077319648 mfu:22.065877301955393 +Step 28786 | loss:0.4425947368144989 lr:1.0258866644938331e-05 tokens_per_second_per_gpu:4906.200415117202 mfu:23.14338444221604 +Step 28787 | loss:0.6304453611373901 lr:1.0256235393933683e-05 tokens_per_second_per_gpu:5325.9034114623 mfu:25.123195084691098 +Step 28788 | loss:0.3801959455013275 lr:1.0253604441846953e-05 tokens_per_second_per_gpu:5325.995750705041 mfu:25.123630664653806 +Step 28789 | loss:0.39842578768730164 lr:1.0250973788697932e-05 tokens_per_second_per_gpu:5328.970361250658 mfu:25.137662410118377 +Step 28790 | loss:0.45818763971328735 lr:1.0248343434506424e-05 tokens_per_second_per_gpu:3646.8242646768695 mfu:17.202692268860854 +Step 28791 | loss:0.3711753785610199 lr:1.0245713379292182e-05 tokens_per_second_per_gpu:5180.475148933146 mfu:24.437185157720894 +Step 28792 | loss:0.4190422594547272 lr:1.0243083623075e-05 tokens_per_second_per_gpu:5341.105313665328 mfu:25.194905051094754 +Step 28793 | loss:0.39205479621887207 lr:1.0240454165874664e-05 tokens_per_second_per_gpu:5209.777392367604 mfu:24.575408839479493 +Step 28794 | loss:0.5043143033981323 lr:1.0237825007710972e-05 tokens_per_second_per_gpu:4611.173723252529 mfu:21.751693199946185 +Step 28795 | loss:0.46895697712898254 lr:1.0235196148603642e-05 tokens_per_second_per_gpu:5372.482188598468 mfu:25.34291512360127 +Step 28796 | loss:0.33176618814468384 lr:1.0232567588572479e-05 tokens_per_second_per_gpu:4034.47219050483 mfu:19.031293674547587 +Step 28797 | loss:0.4094526171684265 lr:1.022993932763725e-05 tokens_per_second_per_gpu:4082.411970208905 mfu:19.25743379478156 +Step 28798 | loss:0.3900164067745209 lr:1.0227311365817732e-05 tokens_per_second_per_gpu:5166.029752763581 mfu:24.369043759350326 +Step 28799 | loss:0.541675329208374 lr:1.022468370313367e-05 tokens_per_second_per_gpu:5331.938703544928 mfu:25.151664587171865 +Step 28800 | loss:0.35877859592437744 lr:1.0222056339604836e-05 tokens_per_second_per_gpu:4598.63809156181 mfu:21.69256048646175 +Step 28801 | loss:0.4012179970741272 lr:1.0219429275251002e-05 tokens_per_second_per_gpu:5311.886648832528 mfu:25.057075623860158 +Step 28802 | loss:0.36297711730003357 lr:1.0216802510091922e-05 tokens_per_second_per_gpu:4100.164496665904 mfu:19.341175490972564 +Step 28803 | loss:0.3832741677761078 lr:1.0214176044147327e-05 tokens_per_second_per_gpu:5332.450363139591 mfu:25.15407817277802 +Step 28804 | loss:0.42105796933174133 lr:1.0211549877436993e-05 tokens_per_second_per_gpu:4495.254437972902 mfu:21.204882153412687 +Step 28805 | loss:0.4026884138584137 lr:1.0208924009980686e-05 tokens_per_second_per_gpu:5163.749899560597 mfu:24.358289302808785 +Step 28806 | loss:0.47475793957710266 lr:1.0206298441798123e-05 tokens_per_second_per_gpu:5262.745170343027 mfu:24.825266885424377 +Step 28807 | loss:0.3651637136936188 lr:1.0203673172909067e-05 tokens_per_second_per_gpu:5370.998799998078 mfu:25.335917726481036 +Step 28808 | loss:0.34894615411758423 lr:1.020104820333328e-05 tokens_per_second_per_gpu:4912.756630321918 mfu:23.174311227942347 +Step 28809 | loss:0.5159972906112671 lr:1.0198423533090484e-05 tokens_per_second_per_gpu:4888.528615477688 mfu:23.060023548196522 +Step 28810 | loss:0.4181346893310547 lr:1.0195799162200415e-05 tokens_per_second_per_gpu:5355.8492418387095 mfu:25.26445467586214 +Step 28811 | loss:0.4091322124004364 lr:1.0193175090682811e-05 tokens_per_second_per_gpu:5109.848702771757 mfu:24.104028161063344 +Step 28812 | loss:0.3180031180381775 lr:1.0190551318557434e-05 tokens_per_second_per_gpu:4568.401034570379 mfu:21.54992712965927 +Step 28813 | loss:0.40824761986732483 lr:1.0187927845843993e-05 tokens_per_second_per_gpu:4951.858988361771 mfu:23.35876372236656 +Step 28814 | loss:0.4070310890674591 lr:1.0185304672562219e-05 tokens_per_second_per_gpu:3815.35419312737 mfu:17.997676695532697 +Step 28815 | loss:0.4745415151119232 lr:1.0182681798731864e-05 tokens_per_second_per_gpu:4646.403419842112 mfu:21.9178776895653 +Step 28816 | loss:0.39210906624794006 lr:1.018005922437264e-05 tokens_per_second_per_gpu:4911.592170978092 mfu:23.168818274540268 +Step 28817 | loss:0.4251159727573395 lr:1.017743694950426e-05 tokens_per_second_per_gpu:5356.420223845241 mfu:25.267148095406522 +Step 28818 | loss:0.20608735084533691 lr:1.0174814974146458e-05 tokens_per_second_per_gpu:4895.0120049243105 mfu:23.090606802396547 +Step 28819 | loss:0.2789627015590668 lr:1.0172193298318966e-05 tokens_per_second_per_gpu:5334.050734673912 mfu:25.161627398356956 +Step 28820 | loss:0.40644973516464233 lr:1.0169571922041476e-05 tokens_per_second_per_gpu:4655.887205072075 mfu:21.96261432259565 +Step 28821 | loss:0.3873901963233948 lr:1.0166950845333717e-05 tokens_per_second_per_gpu:4872.772218871863 mfu:22.98569793708837 +Step 28822 | loss:0.4164375960826874 lr:1.0164330068215421e-05 tokens_per_second_per_gpu:4630.4559317684 mfu:21.842650667399564 +Step 28823 | loss:0.337091863155365 lr:1.016170959070628e-05 tokens_per_second_per_gpu:3930.6801664404247 mfu:18.541688988290414 +Step 28824 | loss:0.4225645065307617 lr:1.015908941282599e-05 tokens_per_second_per_gpu:4049.0981022519713 mfu:19.100286595696716 +Step 28825 | loss:0.46064433455467224 lr:1.0156469534594276e-05 tokens_per_second_per_gpu:5333.608563588455 mfu:25.159541601904685 +Step 28826 | loss:0.4603228271007538 lr:1.0153849956030848e-05 tokens_per_second_per_gpu:4574.540129086481 mfu:21.578886285929258 +Step 28827 | loss:0.3125120997428894 lr:1.0151230677155388e-05 tokens_per_second_per_gpu:4357.709822993642 mfu:20.556060737024303 +Step 28828 | loss:0.2783353626728058 lr:1.0148611697987603e-05 tokens_per_second_per_gpu:3844.3113513134335 mfu:18.134272551297002 +Step 28829 | loss:0.4263520836830139 lr:1.0145993018547196e-05 tokens_per_second_per_gpu:4279.726396019997 mfu:20.18819960664498 +Step 28830 | loss:0.34859588742256165 lr:1.0143374638853891e-05 tokens_per_second_per_gpu:5242.306577464075 mfu:24.728854555630832 +Step 28831 | loss:0.4627598822116852 lr:1.0140756558927322e-05 tokens_per_second_per_gpu:5324.0152903242615 mfu:25.114288495136286 +Step 28832 | loss:0.3442307710647583 lr:1.0138138778787209e-05 tokens_per_second_per_gpu:5069.432286335984 mfu:23.91337702898409 +Step 28833 | loss:0.38939616084098816 lr:1.0135521298453237e-05 tokens_per_second_per_gpu:5336.976830078974 mfu:25.175430289625986 +Step 28834 | loss:0.5382842421531677 lr:1.0132904117945113e-05 tokens_per_second_per_gpu:5237.030613664895 mfu:24.703966934217853 +Step 28835 | loss:0.44245263934135437 lr:1.0130287237282494e-05 tokens_per_second_per_gpu:4531.939091574654 mfu:21.37792992349947 +Step 28836 | loss:0.489067405462265 lr:1.0127670656485067e-05 tokens_per_second_per_gpu:4256.356918711519 mfu:20.077961794936908 +Step 28837 | loss:0.35788407921791077 lr:1.0125054375572535e-05 tokens_per_second_per_gpu:3583.8816077943084 mfu:16.905781017221166 +Step 28838 | loss:0.478801429271698 lr:1.0122438394564554e-05 tokens_per_second_per_gpu:5357.201725461271 mfu:25.270834571867265 +Step 28839 | loss:0.38708776235580444 lr:1.0119822713480786e-05 tokens_per_second_per_gpu:3847.820754731832 mfu:18.150827005987555 +Step 28840 | loss:0.30030930042266846 lr:1.0117207332340928e-05 tokens_per_second_per_gpu:4243.655120630673 mfu:20.018045246239332 +Step 28841 | loss:0.4920736253261566 lr:1.011459225116465e-05 tokens_per_second_per_gpu:5393.1837740876945 mfu:25.440568034408646 +Step 28842 | loss:0.3369467854499817 lr:1.01119774699716e-05 tokens_per_second_per_gpu:5366.404535193978 mfu:25.31424579553769 +Step 28843 | loss:0.42435020208358765 lr:1.0109362988781457e-05 tokens_per_second_per_gpu:5311.92296462653 mfu:25.05724693165658 +Step 28844 | loss:0.4224293529987335 lr:1.0106748807613903e-05 tokens_per_second_per_gpu:5138.299442423362 mfu:24.238235154196694 +Step 28845 | loss:0.47427070140838623 lr:1.0104134926488578e-05 tokens_per_second_per_gpu:3933.4323150165746 mfu:18.55467134268908 +Step 28846 | loss:0.4091705083847046 lr:1.010152134542513e-05 tokens_per_second_per_gpu:4421.476480254465 mfu:20.856858939037217 +Step 28847 | loss:0.4303623139858246 lr:1.0098908064443235e-05 tokens_per_second_per_gpu:3806.5553379962867 mfu:17.956170994639887 +Step 28848 | loss:0.37773600220680237 lr:1.0096295083562551e-05 tokens_per_second_per_gpu:4739.296638314871 mfu:22.356070850318698 +Step 28849 | loss:0.5369178652763367 lr:1.0093682402802717e-05 tokens_per_second_per_gpu:3505.2011313734533 mfu:16.53463178567045 +Step 28850 | loss:0.5058133006095886 lr:1.0091070022183385e-05 tokens_per_second_per_gpu:4844.565114976799 mfu:22.852640215387172 +Step 28851 | loss:0.5013779401779175 lr:1.0088457941724221e-05 tokens_per_second_per_gpu:4094.0974672690795 mfu:19.312556278165786 +Step 28852 | loss:0.33714577555656433 lr:1.0085846161444863e-05 tokens_per_second_per_gpu:3322.9086013095102 mfu:15.674726819046032 +Step 28853 | loss:0.43222376704216003 lr:1.0083234681364934e-05 tokens_per_second_per_gpu:3971.397175413695 mfu:18.73375806665542 +Step 28854 | loss:0.41955095529556274 lr:1.0080623501504089e-05 tokens_per_second_per_gpu:4883.827026309558 mfu:23.037845349916626 +Step 28855 | loss:0.4657818377017975 lr:1.0078012621881983e-05 tokens_per_second_per_gpu:3797.92511592351 mfu:17.915460764654878 +Step 28856 | loss:0.4485076069831848 lr:1.0075402042518222e-05 tokens_per_second_per_gpu:5217.628179081007 mfu:24.612442339888954 +Step 28857 | loss:0.2739788889884949 lr:1.0072791763432454e-05 tokens_per_second_per_gpu:3049.2127358506023 mfu:14.383656724346439 +Step 28858 | loss:0.3521558344364166 lr:1.0070181784644334e-05 tokens_per_second_per_gpu:5129.01499649401 mfu:24.19443883865812 +Step 28859 | loss:0.46395787596702576 lr:1.0067572106173467e-05 tokens_per_second_per_gpu:3886.027876443953 mfu:18.331056517910792 +Step 28860 | loss:0.3882674276828766 lr:1.0064962728039479e-05 tokens_per_second_per_gpu:4030.0157192746046 mfu:19.010271739402217 +Step 28861 | loss:0.3763801157474518 lr:1.0062353650261996e-05 tokens_per_second_per_gpu:4787.843898040862 mfu:22.58507655746282 +Step 28862 | loss:0.44674891233444214 lr:1.005974487286066e-05 tokens_per_second_per_gpu:5322.046513130336 mfu:25.105001437204514 +Step 28863 | loss:0.37681588530540466 lr:1.0057136395855065e-05 tokens_per_second_per_gpu:3709.364274455909 mfu:17.497704165414554 +Step 28864 | loss:0.48555073142051697 lr:1.0054528219264853e-05 tokens_per_second_per_gpu:4868.753138163598 mfu:22.96673924766161 +Step 28865 | loss:0.48968371748924255 lr:1.0051920343109622e-05 tokens_per_second_per_gpu:3762.2597207155736 mfu:17.74722101031514 +Step 28866 | loss:0.31542396545410156 lr:1.0049312767409024e-05 tokens_per_second_per_gpu:5316.9386925319195 mfu:25.080906975957724 +Step 28867 | loss:0.4239423871040344 lr:1.0046705492182618e-05 tokens_per_second_per_gpu:5314.423033137935 mfu:25.069040181380547 +Step 28868 | loss:0.5271579623222351 lr:1.004409851745004e-05 tokens_per_second_per_gpu:3833.5395375579424 mfu:18.08346006795141 +Step 28869 | loss:0.4371296763420105 lr:1.0041491843230888e-05 tokens_per_second_per_gpu:3695.368006032919 mfu:17.43168137925367 +Step 28870 | loss:0.37704020738601685 lr:1.0038885469544795e-05 tokens_per_second_per_gpu:4340.534799576419 mfu:20.475043221204096 +Step 28871 | loss:0.2826591730117798 lr:1.003627939641133e-05 tokens_per_second_per_gpu:4314.998333522361 mfu:20.35458335385672 +Step 28872 | loss:0.4942740797996521 lr:1.0033673623850104e-05 tokens_per_second_per_gpu:5333.274984555729 mfu:25.157968052692866 +Step 28873 | loss:0.3095840513706207 lr:1.003106815188073e-05 tokens_per_second_per_gpu:3457.6362605792438 mfu:16.31026017473037 +Step 28874 | loss:0.4878746569156647 lr:1.0028462980522795e-05 tokens_per_second_per_gpu:4817.002714937747 mfu:22.7226236717726 +Step 28875 | loss:0.3303193747997284 lr:1.0025858109795877e-05 tokens_per_second_per_gpu:5346.876266547411 mfu:25.222127620465635 +Step 28876 | loss:0.28881195187568665 lr:1.002325353971958e-05 tokens_per_second_per_gpu:4598.560179244413 mfu:21.69219296076738 +Step 28877 | loss:0.4478926658630371 lr:1.0020649270313509e-05 tokens_per_second_per_gpu:4876.460195476845 mfu:23.00309475196777 +Step 28878 | loss:0.24744093418121338 lr:1.0018045301597224e-05 tokens_per_second_per_gpu:5307.847040950736 mfu:25.03802010425357 +Step 28879 | loss:0.32435911893844604 lr:1.0015441633590322e-05 tokens_per_second_per_gpu:3094.5616820645546 mfu:14.597575440965725 +Step 28880 | loss:0.33794066309928894 lr:1.0012838266312396e-05 tokens_per_second_per_gpu:4379.4558258814595 mfu:20.65864034289682 +Step 28881 | loss:0.4593738913536072 lr:1.0010235199783014e-05 tokens_per_second_per_gpu:4057.687422998951 mfu:19.140803887149456 +Step 28882 | loss:0.33570411801338196 lr:1.0007632434021747e-05 tokens_per_second_per_gpu:4319.534802768712 mfu:20.375982653293228 +Step 28883 | loss:0.32402822375297546 lr:1.000502996904818e-05 tokens_per_second_per_gpu:4377.242119749838 mfu:20.648197913376748 +Step 28884 | loss:0.38376739621162415 lr:1.0002427804881897e-05 tokens_per_second_per_gpu:5350.72680629594 mfu:25.24029127342927 +Step 28885 | loss:0.4710899293422699 lr:9.999825941542446e-06 tokens_per_second_per_gpu:4283.242541893048 mfu:20.204785866644507 +Step 28886 | loss:0.45626547932624817 lr:9.997224379049403e-06 tokens_per_second_per_gpu:5357.18853575434 mfu:25.27077235376927 +Step 28887 | loss:0.22253118455410004 lr:9.994623117422358e-06 tokens_per_second_per_gpu:4259.958366626312 mfu:20.094950439221364 +Step 28888 | loss:0.3384459316730499 lr:9.992022156680858e-06 tokens_per_second_per_gpu:3594.863625940219 mfu:16.957585070540937 +Step 28889 | loss:0.4657917618751526 lr:9.989421496844448e-06 tokens_per_second_per_gpu:3851.3698592527267 mfu:18.167568737552287 +Step 28890 | loss:0.4464220702648163 lr:9.986821137932707e-06 tokens_per_second_per_gpu:4946.939922630382 mfu:23.335559650032444 +Step 28891 | loss:0.4981148838996887 lr:9.984221079965205e-06 tokens_per_second_per_gpu:5283.483765833811 mfu:24.92309437872366 +Step 28892 | loss:0.4677780866622925 lr:9.981621322961466e-06 tokens_per_second_per_gpu:3600.338282320803 mfu:16.98340995876085 +Step 28893 | loss:0.37561434507369995 lr:9.979021866941057e-06 tokens_per_second_per_gpu:4459.454942091792 mfu:21.036009823317656 +Step 28894 | loss:0.5124908685684204 lr:9.97642271192355e-06 tokens_per_second_per_gpu:4289.267012769057 mfu:20.233204323647364 +Step 28895 | loss:0.35028305649757385 lr:9.973823857928472e-06 tokens_per_second_per_gpu:3932.8630308585834 mfu:18.551985932185943 +Step 28896 | loss:0.46512365341186523 lr:9.971225304975357e-06 tokens_per_second_per_gpu:4363.996955524719 mfu:20.585718213868816 +Step 28897 | loss:0.34517496824264526 lr:9.968627053083769e-06 tokens_per_second_per_gpu:3901.5880314459414 mfu:18.40445642389137 +Step 28898 | loss:0.4068070650100708 lr:9.966029102273255e-06 tokens_per_second_per_gpu:4488.001170693488 mfu:21.170667254120684 +Step 28899 | loss:0.36140528321266174 lr:9.963431452563332e-06 tokens_per_second_per_gpu:5320.845358796872 mfu:25.0993353872755 +Step 28900 | loss:0.354004830121994 lr:9.96083410397355e-06 tokens_per_second_per_gpu:5323.1427629039445 mfu:25.110172634426124 +Step 28901 | loss:0.2830996811389923 lr:9.958237056523439e-06 tokens_per_second_per_gpu:4879.981090609255 mfu:23.019703415034 +Step 28902 | loss:0.4552363455295563 lr:9.955640310232567e-06 tokens_per_second_per_gpu:4749.053617364378 mfu:22.4020961852075 +Step 28903 | loss:0.29015621542930603 lr:9.95304386512041e-06 tokens_per_second_per_gpu:4803.435169416737 mfu:22.658623244688027 +Step 28904 | loss:0.412358820438385 lr:9.950447721206519e-06 tokens_per_second_per_gpu:4844.708591974582 mfu:22.853317020865415 +Step 28905 | loss:0.4518648386001587 lr:9.947851878510417e-06 tokens_per_second_per_gpu:5362.811877929223 mfu:25.297298618249304 +Step 28906 | loss:0.3776898682117462 lr:9.945256337051644e-06 tokens_per_second_per_gpu:5310.253713266044 mfu:25.04937278818551 +Step 28907 | loss:0.417937308549881 lr:9.942661096849704e-06 tokens_per_second_per_gpu:4220.393189031908 mfu:19.90831474599317 +Step 28908 | loss:0.4605238735675812 lr:9.940066157924116e-06 tokens_per_second_per_gpu:4873.402100969028 mfu:22.98866919840967 +Step 28909 | loss:0.41022181510925293 lr:9.937471520294416e-06 tokens_per_second_per_gpu:4460.977935386679 mfu:21.043194042539085 +Step 28910 | loss:0.3437005281448364 lr:9.934877183980107e-06 tokens_per_second_per_gpu:5300.0069905377695 mfu:25.001037248805073 +Step 28911 | loss:0.3306172490119934 lr:9.932283149000681e-06 tokens_per_second_per_gpu:3898.545359237853 mfu:18.39010361995241 +Step 28912 | loss:0.4847719669342041 lr:9.92968941537567e-06 tokens_per_second_per_gpu:4985.580237122434 mfu:23.517832606208923 +Step 28913 | loss:0.37780895829200745 lr:9.92709598312459e-06 tokens_per_second_per_gpu:4642.4487008214655 mfu:21.89922260520027 +Step 28914 | loss:0.40719476342201233 lr:9.92450285226692e-06 tokens_per_second_per_gpu:3901.2215542086556 mfu:18.402727688235633 +Step 28915 | loss:0.40889230370521545 lr:9.921910022822178e-06 tokens_per_second_per_gpu:5281.795431024053 mfu:24.91513021536641 +Step 28916 | loss:0.4026588499546051 lr:9.919317494809876e-06 tokens_per_second_per_gpu:4191.900372727523 mfu:19.773909270109634 +Step 28917 | loss:0.360623836517334 lr:9.916725268249505e-06 tokens_per_second_per_gpu:3375.114713116856 mfu:15.920991955722444 +Step 28918 | loss:0.3005780279636383 lr:9.914133343160547e-06 tokens_per_second_per_gpu:4938.474274941766 mfu:23.29562574549681 +Step 28919 | loss:0.4208902418613434 lr:9.9115417195625e-06 tokens_per_second_per_gpu:5376.942655332087 mfu:25.363955906217853 +Step 28920 | loss:0.23047925531864166 lr:9.90895039747488e-06 tokens_per_second_per_gpu:5366.517632560866 mfu:25.314779295113702 +Step 28921 | loss:0.537807047367096 lr:9.906359376917151e-06 tokens_per_second_per_gpu:5353.47790192215 mfu:25.253268660883414 +Step 28922 | loss:0.38981524109840393 lr:9.903768657908802e-06 tokens_per_second_per_gpu:3639.519644090006 mfu:17.168235127255834 +Step 28923 | loss:0.4991558790206909 lr:9.901178240469344e-06 tokens_per_second_per_gpu:4536.054099342867 mfu:21.39734111723485 +Step 28924 | loss:0.4238847494125366 lr:9.898588124618236e-06 tokens_per_second_per_gpu:5332.416045141608 mfu:25.153916288926283 +Step 28925 | loss:0.3580973148345947 lr:9.895998310374955e-06 tokens_per_second_per_gpu:5329.079103192115 mfu:25.13817536440952 +Step 28926 | loss:0.3716529905796051 lr:9.893408797758985e-06 tokens_per_second_per_gpu:5341.222633796472 mfu:25.19545847017024 +Step 28927 | loss:0.43721598386764526 lr:9.890819586789818e-06 tokens_per_second_per_gpu:3681.6026280368383 mfu:17.36674774263049 +Step 28928 | loss:0.4141301214694977 lr:9.888230677486904e-06 tokens_per_second_per_gpu:4457.290792796075 mfu:21.02580116184781 +Step 28929 | loss:0.44295039772987366 lr:9.885642069869727e-06 tokens_per_second_per_gpu:4506.383039193098 mfu:21.25737766410351 +Step 28930 | loss:0.34632566571235657 lr:9.883053763957762e-06 tokens_per_second_per_gpu:5369.439575492348 mfu:25.328562598456376 +Step 28931 | loss:0.45342937111854553 lr:9.880465759770474e-06 tokens_per_second_per_gpu:4725.37965608779 mfu:22.29042207067154 +Step 28932 | loss:0.3047039210796356 lr:9.877878057327306e-06 tokens_per_second_per_gpu:3473.810984853115 mfu:16.386559108822752 +Step 28933 | loss:0.4121065139770508 lr:9.875290656647735e-06 tokens_per_second_per_gpu:4852.898419317437 mfu:22.891949833770298 +Step 28934 | loss:0.36334550380706787 lr:9.872703557751239e-06 tokens_per_second_per_gpu:4304.74438054722 mfu:20.3062137081682 +Step 28935 | loss:0.32813164591789246 lr:9.870116760657245e-06 tokens_per_second_per_gpu:5382.241834793642 mfu:25.38895303987125 +Step 28936 | loss:0.3201432526111603 lr:9.867530265385228e-06 tokens_per_second_per_gpu:4544.0266520398045 mfu:21.43494900856406 +Step 28937 | loss:0.40887150168418884 lr:9.864944071954629e-06 tokens_per_second_per_gpu:3723.88775515013 mfu:17.566213901810517 +Step 28938 | loss:0.2683335244655609 lr:9.86235818038494e-06 tokens_per_second_per_gpu:5335.513195911028 mfu:25.168526077534047 +Step 28939 | loss:0.43010732531547546 lr:9.859772590695548e-06 tokens_per_second_per_gpu:4458.639818684551 mfu:21.0321647471313 +Step 28940 | loss:0.3855636715888977 lr:9.85718730290593e-06 tokens_per_second_per_gpu:5163.098259208307 mfu:24.355215404085722 +Step 28941 | loss:0.30670514702796936 lr:9.854602317035527e-06 tokens_per_second_per_gpu:4619.6475081357785 mfu:21.79166548901726 +Step 28942 | loss:0.3573426902294159 lr:9.852017633103793e-06 tokens_per_second_per_gpu:5358.9237984106385 mfu:25.27895788378537 +Step 28943 | loss:0.5265339612960815 lr:9.849433251130147e-06 tokens_per_second_per_gpu:4655.883869502883 mfu:21.962598588146662 +Step 28944 | loss:0.32781288027763367 lr:9.846849171134038e-06 tokens_per_second_per_gpu:3987.7604168869234 mfu:18.810946268541517 +Step 28945 | loss:0.3873484134674072 lr:9.844265393134927e-06 tokens_per_second_per_gpu:4615.134573465903 mfu:21.770377206193398 +Step 28946 | loss:0.2924598455429077 lr:9.84168191715219e-06 tokens_per_second_per_gpu:4292.529855769079 mfu:20.24859571077689 +Step 28947 | loss:0.41513797640800476 lr:9.839098743205288e-06 tokens_per_second_per_gpu:5129.019199836149 mfu:24.194458666540154 +Step 28948 | loss:0.41113126277923584 lr:9.83651587131365e-06 tokens_per_second_per_gpu:5324.66356409623 mfu:25.11734651312558 +Step 28949 | loss:0.363603413105011 lr:9.833933301496712e-06 tokens_per_second_per_gpu:4566.776702444439 mfu:21.542264877881575 +Step 28950 | loss:0.5328779816627502 lr:9.831351033773878e-06 tokens_per_second_per_gpu:5283.496780928236 mfu:24.923155773144668 +Step 28951 | loss:0.38064858317375183 lr:9.828769068164573e-06 tokens_per_second_per_gpu:3240.036996761778 mfu:15.283807321040639 +Step 28952 | loss:0.47941669821739197 lr:9.826187404688241e-06 tokens_per_second_per_gpu:3682.4811771897425 mfu:17.370892008880684 +Step 28953 | loss:0.36251726746559143 lr:9.823606043364275e-06 tokens_per_second_per_gpu:4932.471392161299 mfu:23.267309123223928 +Step 28954 | loss:0.4104234278202057 lr:9.821024984212085e-06 tokens_per_second_per_gpu:5353.087600513331 mfu:25.25142754254573 +Step 28955 | loss:0.4936388432979584 lr:9.818444227251089e-06 tokens_per_second_per_gpu:4534.252333664219 mfu:21.38884188111574 +Step 28956 | loss:0.3646085262298584 lr:9.815863772500717e-06 tokens_per_second_per_gpu:4609.184039101407 mfu:21.742307520330215 +Step 28957 | loss:0.3833879828453064 lr:9.813283619980346e-06 tokens_per_second_per_gpu:3404.448693949815 mfu:16.059365348204732 +Step 28958 | loss:0.5302339196205139 lr:9.8107037697094e-06 tokens_per_second_per_gpu:4712.212878382618 mfu:22.228312133752315 +Step 28959 | loss:0.4401957392692566 lr:9.808124221707288e-06 tokens_per_second_per_gpu:5327.845042804191 mfu:25.132354091007265 +Step 28960 | loss:0.4259783625602722 lr:9.805544975993408e-06 tokens_per_second_per_gpu:5342.370365117001 mfu:25.200872514632326 +Step 28961 | loss:0.2955634295940399 lr:9.802966032587136e-06 tokens_per_second_per_gpu:5332.206530995837 mfu:25.15292797495388 +Step 28962 | loss:0.360795259475708 lr:9.800387391507888e-06 tokens_per_second_per_gpu:3348.1076117395955 mfu:15.793594850639263 +Step 28963 | loss:0.6081231236457825 lr:9.797809052775065e-06 tokens_per_second_per_gpu:5336.332897169173 mfu:25.17239274822417 +Step 28964 | loss:0.4288077652454376 lr:9.79523101640804e-06 tokens_per_second_per_gpu:5068.264915461358 mfu:23.90787034139447 +Step 28965 | loss:0.34456002712249756 lr:9.792653282426213e-06 tokens_per_second_per_gpu:5309.799378383604 mfu:25.04722961302752 +Step 28966 | loss:0.48205187916755676 lr:9.79007585084898e-06 tokens_per_second_per_gpu:5338.546799789405 mfu:25.182836104614545 +Step 28967 | loss:0.306484580039978 lr:9.78749872169572e-06 tokens_per_second_per_gpu:4341.519072829501 mfu:20.47968620607322 +Step 28968 | loss:0.3924356997013092 lr:9.784921894985799e-06 tokens_per_second_per_gpu:3827.902057656475 mfu:18.056867113402934 +Step 28969 | loss:0.408793181180954 lr:9.782345370738616e-06 tokens_per_second_per_gpu:4264.368411581679 mfu:20.115753373707083 +Step 28970 | loss:0.4693130850791931 lr:9.779769148973555e-06 tokens_per_second_per_gpu:4398.717931741472 mfu:20.74950298269265 +Step 28971 | loss:0.511570394039154 lr:9.77719322970997e-06 tokens_per_second_per_gpu:5093.2234995207455 mfu:24.02560424077615 +Step 28972 | loss:0.486050009727478 lr:9.774617612967247e-06 tokens_per_second_per_gpu:5032.084982797946 mfu:23.737203426088108 +Step 28973 | loss:0.38190388679504395 lr:9.77204229876476e-06 tokens_per_second_per_gpu:5165.81492862407 mfu:24.368030397231475 +Step 28974 | loss:0.34441831707954407 lr:9.769467287121897e-06 tokens_per_second_per_gpu:4737.645491043761 mfu:22.348282106926096 +Step 28975 | loss:0.3649154007434845 lr:9.766892578057985e-06 tokens_per_second_per_gpu:3875.36412162783 mfu:18.280753767018957 +Step 28976 | loss:0.4430837035179138 lr:9.764318171592402e-06 tokens_per_second_per_gpu:5328.3809981588065 mfu:25.134882283858403 +Step 28977 | loss:0.3415991961956024 lr:9.761744067744516e-06 tokens_per_second_per_gpu:5004.661484782992 mfu:23.607842107020335 +Step 28978 | loss:0.2718486487865448 lr:9.759170266533702e-06 tokens_per_second_per_gpu:3947.1547673217565 mfu:18.619402491504413 +Step 28979 | loss:0.37630394101142883 lr:9.756596767979293e-06 tokens_per_second_per_gpu:4726.236598935404 mfu:22.29446441629326 +Step 28980 | loss:0.38045936822891235 lr:9.754023572100652e-06 tokens_per_second_per_gpu:3535.3773192083386 mfu:16.676978012276155 +Step 28981 | loss:0.38887518644332886 lr:9.751450678917146e-06 tokens_per_second_per_gpu:4520.420764257265 mfu:21.323596008314052 +Step 28982 | loss:0.4440447986125946 lr:9.748878088448121e-06 tokens_per_second_per_gpu:5001.0108929507305 mfu:23.590621642492366 +Step 28983 | loss:0.331026554107666 lr:9.746305800712902e-06 tokens_per_second_per_gpu:4647.93618487991 mfu:21.925108003765274 +Step 28984 | loss:0.44116470217704773 lr:9.743733815730855e-06 tokens_per_second_per_gpu:4985.995296703402 mfu:23.519790513068816 +Step 28985 | loss:0.37716227769851685 lr:9.741162133521336e-06 tokens_per_second_per_gpu:5331.504415245719 mfu:25.149615975167276 +Step 28986 | loss:0.32964807748794556 lr:9.738590754103661e-06 tokens_per_second_per_gpu:4434.628135563589 mfu:20.918897541034134 +Step 28987 | loss:0.4578173756599426 lr:9.736019677497182e-06 tokens_per_second_per_gpu:5315.8300851657505 mfu:25.075677485863583 +Step 28988 | loss:0.3861331641674042 lr:9.733448903721248e-06 tokens_per_second_per_gpu:5344.1732602309885 mfu:25.209377078490633 +Step 28989 | loss:0.5772368907928467 lr:9.730878432795183e-06 tokens_per_second_per_gpu:3851.9779802783423 mfu:18.17043734818626 +Step 28990 | loss:0.3701932430267334 lr:9.728308264738306e-06 tokens_per_second_per_gpu:5344.576796493988 mfu:25.21128062789359 +Step 28991 | loss:0.38233068585395813 lr:9.725738399569968e-06 tokens_per_second_per_gpu:5123.752280254281 mfu:24.169613708243336 +Step 28992 | loss:0.3149985373020172 lr:9.723168837309498e-06 tokens_per_second_per_gpu:5134.247108449447 mfu:24.21911960344227 +Step 28993 | loss:0.3442895710468292 lr:9.720599577976203e-06 tokens_per_second_per_gpu:5359.459079298214 mfu:25.281482895060403 +Step 28994 | loss:0.38640832901000977 lr:9.718030621589414e-06 tokens_per_second_per_gpu:4414.838977545318 mfu:20.82554870628343 +Step 28995 | loss:0.4290815591812134 lr:9.715461968168471e-06 tokens_per_second_per_gpu:4289.793756631814 mfu:20.235689064319757 +Step 28996 | loss:0.3592415750026703 lr:9.712893617732682e-06 tokens_per_second_per_gpu:5329.949500488381 mfu:25.142281176961095 +Step 28997 | loss:0.4947872757911682 lr:9.710325570301348e-06 tokens_per_second_per_gpu:3642.0876146381565 mfu:17.180348682472413 +Step 28998 | loss:0.4067017436027527 lr:9.707757825893793e-06 tokens_per_second_per_gpu:4356.012006918414 mfu:20.54805184892016 +Step 28999 | loss:0.3979416489601135 lr:9.70519038452935e-06 tokens_per_second_per_gpu:3511.5739794540677 mfu:16.56469359738671 +Step 29000 | loss:0.37595298886299133 lr:9.7026232462273e-06 tokens_per_second_per_gpu:3591.191023514216 mfu:16.94026077828689 +Step 29001 | loss:0.38162875175476074 lr:9.700056411006958e-06 tokens_per_second_per_gpu:4712.862100659409 mfu:22.231374626000683 +Step 29002 | loss:0.33562955260276794 lr:9.697489878887644e-06 tokens_per_second_per_gpu:4025.5660761589816 mfu:18.989282013638423 +Step 29003 | loss:0.3650205731391907 lr:9.69492364988866e-06 tokens_per_second_per_gpu:4157.112257162181 mfu:19.609807793523352 +Step 29004 | loss:0.42305588722229004 lr:9.692357724029277e-06 tokens_per_second_per_gpu:3841.1789581463872 mfu:18.11949651828666 +Step 29005 | loss:0.4153271019458771 lr:9.68979210132882e-06 tokens_per_second_per_gpu:4528.053105980014 mfu:21.35959906643204 +Step 29006 | loss:0.3783646821975708 lr:9.687226781806591e-06 tokens_per_second_per_gpu:5339.558254062219 mfu:25.1876073069913 +Step 29007 | loss:0.4808810353279114 lr:9.684661765481861e-06 tokens_per_second_per_gpu:5231.325948834518 mfu:24.677057056896775 +Step 29008 | loss:0.33294516801834106 lr:9.682097052373934e-06 tokens_per_second_per_gpu:4374.401391843246 mfu:20.634797715163195 +Step 29009 | loss:0.4005298614501953 lr:9.679532642502099e-06 tokens_per_second_per_gpu:5342.588059481626 mfu:25.201899416092758 +Step 29010 | loss:0.3550967574119568 lr:9.676968535885662e-06 tokens_per_second_per_gpu:5024.417905158666 mfu:23.70103651272529 +Step 29011 | loss:0.39487358927726746 lr:9.674404732543884e-06 tokens_per_second_per_gpu:4054.1439745212356 mfu:19.124088836105418 +Step 29012 | loss:0.5459676384925842 lr:9.671841232496043e-06 tokens_per_second_per_gpu:5366.407710963625 mfu:25.314260776184955 +Step 29013 | loss:0.3744989037513733 lr:9.669278035761442e-06 tokens_per_second_per_gpu:5314.999122058865 mfu:25.071757691111483 +Step 29014 | loss:0.4581928253173828 lr:9.666715142359334e-06 tokens_per_second_per_gpu:4395.626857829518 mfu:20.734921859658368 +Step 29015 | loss:0.48593711853027344 lr:9.664152552309007e-06 tokens_per_second_per_gpu:5278.972499538494 mfu:24.90181396590725 +Step 29016 | loss:0.3550698459148407 lr:9.661590265629739e-06 tokens_per_second_per_gpu:3928.058852076893 mfu:18.529323801195286 +Step 29017 | loss:0.46052658557891846 lr:9.659028282340805e-06 tokens_per_second_per_gpu:5377.417245617595 mfu:25.36619462956804 +Step 29018 | loss:0.3937996029853821 lr:9.656466602461472e-06 tokens_per_second_per_gpu:5337.48296169972 mfu:25.177817798836006 +Step 29019 | loss:0.37851014733314514 lr:9.65390522601099e-06 tokens_per_second_per_gpu:3960.854912766909 mfu:18.684028415055042 +Step 29020 | loss:0.46741020679473877 lr:9.651344153008629e-06 tokens_per_second_per_gpu:4404.17157108371 mfu:20.775228729956805 +Step 29021 | loss:0.43257734179496765 lr:9.648783383473675e-06 tokens_per_second_per_gpu:5101.95684332621 mfu:24.06680091356863 +Step 29022 | loss:0.4090519845485687 lr:9.646222917425352e-06 tokens_per_second_per_gpu:4112.320071187547 mfu:19.398515409946025 +Step 29023 | loss:0.48775190114974976 lr:9.643662754882944e-06 tokens_per_second_per_gpu:4288.0536875950875 mfu:20.227480861787125 +Step 29024 | loss:0.3339836299419403 lr:9.6411028958657e-06 tokens_per_second_per_gpu:5334.087414001246 mfu:25.161800421002123 +Step 29025 | loss:0.35061827301979065 lr:9.63854334039288e-06 tokens_per_second_per_gpu:4673.804615135542 mfu:22.04713380289041 +Step 29026 | loss:0.4797878563404083 lr:9.635984088483708e-06 tokens_per_second_per_gpu:4728.134512593648 mfu:22.303417198836655 +Step 29027 | loss:0.44923263788223267 lr:9.633425140157454e-06 tokens_per_second_per_gpu:4656.279145052446 mfu:21.964463170354914 +Step 29028 | loss:0.46987229585647583 lr:9.630866495433372e-06 tokens_per_second_per_gpu:5228.0404749702775 mfu:24.66155891611951 +Step 29029 | loss:0.4893743395805359 lr:9.628308154330678e-06 tokens_per_second_per_gpu:4187.9356832794065 mfu:19.755207153537096 +Step 29030 | loss:0.416972815990448 lr:9.62575011686863e-06 tokens_per_second_per_gpu:5292.385800713499 mfu:24.965086796094916 +Step 29031 | loss:0.4954291582107544 lr:9.623192383066482e-06 tokens_per_second_per_gpu:3953.7104300306733 mfu:18.65032667101377 +Step 29032 | loss:0.48169568181037903 lr:9.620634952943459e-06 tokens_per_second_per_gpu:4322.846334947173 mfu:20.39160371558468 +Step 29033 | loss:0.40378493070602417 lr:9.618077826518774e-06 tokens_per_second_per_gpu:4951.742908961452 mfu:23.358216156030068 +Step 29034 | loss:0.46579664945602417 lr:9.615521003811684e-06 tokens_per_second_per_gpu:5061.6055745548665 mfu:23.876457094137827 +Step 29035 | loss:0.48360466957092285 lr:9.612964484841424e-06 tokens_per_second_per_gpu:4472.185162995458 mfu:21.096060447319903 +Step 29036 | loss:0.4160858988761902 lr:9.610408269627196e-06 tokens_per_second_per_gpu:5146.657137720825 mfu:24.277659828884687 +Step 29037 | loss:0.42921411991119385 lr:9.607852358188246e-06 tokens_per_second_per_gpu:3927.233750824167 mfu:18.52543165780939 +Step 29038 | loss:0.4349335730075836 lr:9.605296750543802e-06 tokens_per_second_per_gpu:5346.045385466412 mfu:25.218208212643628 +Step 29039 | loss:0.44108593463897705 lr:9.602741446713065e-06 tokens_per_second_per_gpu:4643.251622055869 mfu:21.903010121655438 +Step 29040 | loss:0.440614253282547 lr:9.600186446715276e-06 tokens_per_second_per_gpu:4481.725667117135 mfu:21.141064633039708 +Step 29041 | loss:0.36374756693840027 lr:9.59763175056963e-06 tokens_per_second_per_gpu:3316.0661104783157 mfu:15.642449622345968 +Step 29042 | loss:0.5391443371772766 lr:9.595077358295363e-06 tokens_per_second_per_gpu:4367.785662735663 mfu:20.603590192202947 +Step 29043 | loss:0.470960795879364 lr:9.592523269911658e-06 tokens_per_second_per_gpu:5381.747923545481 mfu:25.386623176243727 +Step 29044 | loss:0.4428989887237549 lr:9.589969485437745e-06 tokens_per_second_per_gpu:4289.206679051838 mfu:20.23291971920925 +Step 29045 | loss:0.2948537766933441 lr:9.587416004892824e-06 tokens_per_second_per_gpu:264.5043241068813 mfu:1.2477120259034162 +Step 29046 | loss:0.33649370074272156 lr:9.58486282829612e-06 tokens_per_second_per_gpu:3826.2095998593904 mfu:18.04888350121101 +Step 29047 | loss:0.3050433099269867 lr:9.582309955666818e-06 tokens_per_second_per_gpu:4704.140947008028 mfu:22.190235456244867 +Step 29048 | loss:0.25598111748695374 lr:9.579757387024107e-06 tokens_per_second_per_gpu:3493.4698415052612 mfu:16.47929328979774 +Step 29049 | loss:0.3695688247680664 lr:9.577205122387212e-06 tokens_per_second_per_gpu:4651.942049962221 mfu:21.944004352829598 +Step 29050 | loss:0.4319819509983063 lr:9.574653161775305e-06 tokens_per_second_per_gpu:5354.734516294239 mfu:25.259196325277085 +Step 29051 | loss:0.3095265030860901 lr:9.572101505207588e-06 tokens_per_second_per_gpu:4666.130364742407 mfu:22.01093305442386 +Step 29052 | loss:0.2418045550584793 lr:9.569550152703254e-06 tokens_per_second_per_gpu:5372.869069780088 mfu:25.344740108143704 +Step 29053 | loss:0.325111448764801 lr:9.566999104281505e-06 tokens_per_second_per_gpu:5350.757081786543 mfu:25.240434088083973 +Step 29054 | loss:0.3081415295600891 lr:9.564448359961514e-06 tokens_per_second_per_gpu:4599.48912611395 mfu:21.696574961645663 +Step 29055 | loss:0.32687434554100037 lr:9.561897919762453e-06 tokens_per_second_per_gpu:4411.6851385905975 mfu:20.8106715098336 +Step 29056 | loss:0.28573694825172424 lr:9.559347783703521e-06 tokens_per_second_per_gpu:3395.446883284984 mfu:16.01690227143145 +Step 29057 | loss:0.35781365633010864 lr:9.556797951803902e-06 tokens_per_second_per_gpu:4528.367224449349 mfu:21.36108081684603 +Step 29058 | loss:0.31543269753456116 lr:9.554248424082763e-06 tokens_per_second_per_gpu:3973.1041452515487 mfu:18.741810134619655 +Step 29059 | loss:0.27522245049476624 lr:9.551699200559271e-06 tokens_per_second_per_gpu:3498.015693600017 mfu:16.500736849731048 +Step 29060 | loss:0.3512982130050659 lr:9.549150281252633e-06 tokens_per_second_per_gpu:5356.42929170928 mfu:25.267190870068333 +Step 29061 | loss:0.35519033670425415 lr:9.546601666181993e-06 tokens_per_second_per_gpu:5357.630471165485 mfu:25.2728570385056 +Step 29062 | loss:0.3502400815486908 lr:9.54405335536651e-06 tokens_per_second_per_gpu:5327.234757075062 mfu:25.12947526909748 +Step 29063 | loss:0.2804395258426666 lr:9.541505348825363e-06 tokens_per_second_per_gpu:4962.824438688281 mfu:23.410489622455454 +Step 29064 | loss:0.40278327465057373 lr:9.538957646577734e-06 tokens_per_second_per_gpu:4891.148571921845 mfu:23.072382329754276 +Step 29065 | loss:0.3147265315055847 lr:9.536410248642752e-06 tokens_per_second_per_gpu:2984.5279978061017 mfu:14.078527778636186 +Step 29066 | loss:0.2603941261768341 lr:9.53386315503959e-06 tokens_per_second_per_gpu:4341.21672483636 mfu:20.47825997900369 +Step 29067 | loss:0.3008662462234497 lr:9.531316365787424e-06 tokens_per_second_per_gpu:5226.422890520724 mfu:24.653928494282596 +Step 29068 | loss:0.26335057616233826 lr:9.528769880905391e-06 tokens_per_second_per_gpu:3871.8189244280793 mfu:18.264030466954715 +Step 29069 | loss:0.3720315992832184 lr:9.526223700412628e-06 tokens_per_second_per_gpu:5017.826115544257 mfu:23.66994191643884 +Step 29070 | loss:0.40810927748680115 lr:9.523677824328302e-06 tokens_per_second_per_gpu:3866.3249847746706 mfu:18.238114616246822 +Step 29071 | loss:0.39669069647789 lr:9.521132252671572e-06 tokens_per_second_per_gpu:5343.583104003769 mfu:25.206593210875806 +Step 29072 | loss:0.29254353046417236 lr:9.518586985461559e-06 tokens_per_second_per_gpu:5308.638588869186 mfu:25.041753970836897 +Step 29073 | loss:0.25439995527267456 lr:9.51604202271742e-06 tokens_per_second_per_gpu:3367.1443334599494 mfu:15.88339431499382 +Step 29074 | loss:0.27762138843536377 lr:9.513497364458307e-06 tokens_per_second_per_gpu:4694.296629709956 mfu:22.143798131936133 +Step 29075 | loss:0.3208572268486023 lr:9.510953010703333e-06 tokens_per_second_per_gpu:4985.674595613954 mfu:23.518277711312642 +Step 29076 | loss:0.340496689081192 lr:9.50840896147166e-06 tokens_per_second_per_gpu:4044.189097374139 mfu:19.077129982125385 +Step 29077 | loss:0.24116665124893188 lr:9.5058652167824e-06 tokens_per_second_per_gpu:4913.718086806615 mfu:23.178846582222054 +Step 29078 | loss:0.25427064299583435 lr:9.503321776654706e-06 tokens_per_second_per_gpu:5371.028582171003 mfu:25.336058214072086 +Step 29079 | loss:0.18779516220092773 lr:9.500778641107683e-06 tokens_per_second_per_gpu:5348.934479124644 mfu:25.231836560359717 +Step 29080 | loss:0.3247125744819641 lr:9.498235810160472e-06 tokens_per_second_per_gpu:3690.9644690622276 mfu:17.41090914404204 +Step 29081 | loss:0.2886415719985962 lr:9.495693283832213e-06 tokens_per_second_per_gpu:4160.523936404307 mfu:19.625901266600966 +Step 29082 | loss:0.34733015298843384 lr:9.493151062141998e-06 tokens_per_second_per_gpu:5368.09106380002 mfu:25.32220143872384 +Step 29083 | loss:0.3746299743652344 lr:9.490609145108975e-06 tokens_per_second_per_gpu:4767.485143223881 mfu:22.489040837429368 +Step 29084 | loss:0.295986533164978 lr:9.488067532752232e-06 tokens_per_second_per_gpu:5352.608241474484 mfu:25.249166324172403 +Step 29085 | loss:0.29181262850761414 lr:9.485526225090924e-06 tokens_per_second_per_gpu:4606.816738702617 mfu:21.73114056044603 +Step 29086 | loss:0.2757686376571655 lr:9.482985222144126e-06 tokens_per_second_per_gpu:4452.554296061732 mfu:21.003458298622242 +Step 29087 | loss:0.2699403166770935 lr:9.480444523930964e-06 tokens_per_second_per_gpu:3260.5964600195134 mfu:15.380789817033524 +Step 29088 | loss:0.43450024724006653 lr:9.477904130470555e-06 tokens_per_second_per_gpu:4263.140065393311 mfu:20.109959055159077 +Step 29089 | loss:0.23194153606891632 lr:9.475364041782008e-06 tokens_per_second_per_gpu:5316.1958794540005 mfu:25.077402999932204 +Step 29090 | loss:0.3161672055721283 lr:9.472824257884416e-06 tokens_per_second_per_gpu:5364.975466768568 mfu:25.30750463595082 +Step 29091 | loss:0.3815554082393646 lr:9.470284778796879e-06 tokens_per_second_per_gpu:4264.714830128463 mfu:20.117387489097364 +Step 29092 | loss:0.2450423538684845 lr:9.467745604538498e-06 tokens_per_second_per_gpu:3997.693028846689 mfu:18.85780009383408 +Step 29093 | loss:0.23020987212657928 lr:9.465206735128384e-06 tokens_per_second_per_gpu:5328.479630531685 mfu:25.135347549589685 +Step 29094 | loss:0.24885721504688263 lr:9.462668170585614e-06 tokens_per_second_per_gpu:4129.7313778126845 mfu:19.480647513972038 +Step 29095 | loss:0.3998745083808899 lr:9.460129910929288e-06 tokens_per_second_per_gpu:5341.662235659461 mfu:25.197532147161766 +Step 29096 | loss:0.33760184049606323 lr:9.457591956178514e-06 tokens_per_second_per_gpu:4223.667337035012 mfu:19.92375946075106 +Step 29097 | loss:0.2120586782693863 lr:9.455054306352357e-06 tokens_per_second_per_gpu:5361.912895405074 mfu:25.293057964300658 +Step 29098 | loss:0.249037966132164 lr:9.452516961469904e-06 tokens_per_second_per_gpu:4018.5029895825664 mfu:18.955964229170636 +Step 29099 | loss:0.2969604730606079 lr:9.449979921550239e-06 tokens_per_second_per_gpu:5155.090285826352 mfu:24.317440427341893 +Step 29100 | loss:0.3475685119628906 lr:9.447443186612464e-06 tokens_per_second_per_gpu:3832.6148172469193 mfu:18.07909800447076 +Step 29101 | loss:0.3998890817165375 lr:9.444906756675636e-06 tokens_per_second_per_gpu:3484.004936346629 mfu:16.434645717285132 +Step 29102 | loss:0.4113095998764038 lr:9.442370631758835e-06 tokens_per_second_per_gpu:5322.3282872137925 mfu:25.106330613631794 +Step 29103 | loss:0.205545112490654 lr:9.439834811881148e-06 tokens_per_second_per_gpu:5020.289194290248 mfu:23.681560679128246 +Step 29104 | loss:0.22989994287490845 lr:9.43729929706163e-06 tokens_per_second_per_gpu:5311.68494819095 mfu:25.056124167520156 +Step 29105 | loss:0.23923245072364807 lr:9.434764087319376e-06 tokens_per_second_per_gpu:5363.376933408036 mfu:25.299964081351273 +Step 29106 | loss:0.37631797790527344 lr:9.43222918267342e-06 tokens_per_second_per_gpu:3286.2954801478572 mfu:15.50201647968389 +Step 29107 | loss:0.25717949867248535 lr:9.429694583142862e-06 tokens_per_second_per_gpu:5330.8861638037715 mfu:25.146699577631033 +Step 29108 | loss:0.28936800360679626 lr:9.427160288746727e-06 tokens_per_second_per_gpu:4334.360224901681 mfu:20.44591669897229 +Step 29109 | loss:0.24344496428966522 lr:9.424626299504102e-06 tokens_per_second_per_gpu:4037.677749024297 mfu:19.046414841008595 +Step 29110 | loss:0.3416409194469452 lr:9.422092615434054e-06 tokens_per_second_per_gpu:5351.808669078488 mfu:25.24539460475937 +Step 29111 | loss:0.32542794942855835 lr:9.419559236555614e-06 tokens_per_second_per_gpu:4370.3942915285 mfu:20.615895539296723 +Step 29112 | loss:0.23818457126617432 lr:9.417026162887854e-06 tokens_per_second_per_gpu:3647.141923791091 mfu:17.204190721100872 +Step 29113 | loss:0.23411403596401215 lr:9.414493394449808e-06 tokens_per_second_per_gpu:5375.853507580517 mfu:25.358818210445925 +Step 29114 | loss:0.3391611874103546 lr:9.41196093126055e-06 tokens_per_second_per_gpu:5327.428604298178 mfu:25.130389679522665 +Step 29115 | loss:0.2500424385070801 lr:9.409428773339102e-06 tokens_per_second_per_gpu:4939.036949347215 mfu:23.298279976669463 +Step 29116 | loss:0.2625516951084137 lr:9.406896920704516e-06 tokens_per_second_per_gpu:3061.566847835776 mfu:14.44193317840936 +Step 29117 | loss:0.35191610455513 lr:9.40436537337585e-06 tokens_per_second_per_gpu:4764.66100787061 mfu:22.475718909120808 +Step 29118 | loss:0.30838730931282043 lr:9.401834131372123e-06 tokens_per_second_per_gpu:4920.876520184607 mfu:23.21261413382072 +Step 29119 | loss:0.32907387614250183 lr:9.399303194712395e-06 tokens_per_second_per_gpu:4809.443147122556 mfu:22.68696390059307 +Step 29120 | loss:0.3187333643436432 lr:9.396772563415674e-06 tokens_per_second_per_gpu:5308.132839358239 mfu:25.039368264103995 +Step 29121 | loss:0.2798016369342804 lr:9.394242237501017e-06 tokens_per_second_per_gpu:4126.98439455965 mfu:19.467689525283802 +Step 29122 | loss:0.4043677747249603 lr:9.391712216987437e-06 tokens_per_second_per_gpu:4016.403355892194 mfu:18.946059898818703 +Step 29123 | loss:0.3606814742088318 lr:9.389182501893968e-06 tokens_per_second_per_gpu:3756.870294158842 mfu:17.72179816571605 +Step 29124 | loss:0.3282037675380707 lr:9.386653092239639e-06 tokens_per_second_per_gpu:4618.877796803508 mfu:21.788034629336643 +Step 29125 | loss:0.3181636929512024 lr:9.384123988043492e-06 tokens_per_second_per_gpu:4453.843284188443 mfu:21.00953867553999 +Step 29126 | loss:0.2951931655406952 lr:9.381595189324527e-06 tokens_per_second_per_gpu:4424.902907937812 mfu:20.873021983028412 +Step 29127 | loss:0.3477097749710083 lr:9.379066696101757e-06 tokens_per_second_per_gpu:5210.70500280935 mfu:24.57978453619987 +Step 29128 | loss:0.3815079927444458 lr:9.376538508394206e-06 tokens_per_second_per_gpu:4927.7822247108425 mfu:23.245189520305658 +Step 29129 | loss:0.29410433769226074 lr:9.374010626220908e-06 tokens_per_second_per_gpu:3876.20923008858 mfu:18.284740287818686 +Step 29130 | loss:0.3198130130767822 lr:9.371483049600849e-06 tokens_per_second_per_gpu:5330.851328016278 mfu:25.14653525127783 +Step 29131 | loss:0.39617541432380676 lr:9.368955778553045e-06 tokens_per_second_per_gpu:5332.508151301639 mfu:25.15435076939808 +Step 29132 | loss:0.30283811688423157 lr:9.366428813096529e-06 tokens_per_second_per_gpu:4041.7008444126413 mfu:19.065392468366266 +Step 29133 | loss:0.3208841383457184 lr:9.36390215325027e-06 tokens_per_second_per_gpu:4009.644113931217 mfu:18.914175401243174 +Step 29134 | loss:0.3818974494934082 lr:9.361375799033301e-06 tokens_per_second_per_gpu:3841.4442458858853 mfu:18.12074792581663 +Step 29135 | loss:0.3641749620437622 lr:9.3588497504646e-06 tokens_per_second_per_gpu:4599.062505149677 mfu:21.694562517768144 +Step 29136 | loss:0.36118459701538086 lr:9.356324007563184e-06 tokens_per_second_per_gpu:4100.2121634595105 mfu:19.3414003433708 +Step 29137 | loss:0.469110369682312 lr:9.353798570348027e-06 tokens_per_second_per_gpu:4599.8785099953075 mfu:21.69841175184977 +Step 29138 | loss:0.26043668389320374 lr:9.351273438838143e-06 tokens_per_second_per_gpu:4379.379284955447 mfu:20.658279286289588 +Step 29139 | loss:0.32613635063171387 lr:9.348748613052527e-06 tokens_per_second_per_gpu:5311.0659435955895 mfu:25.053204217230768 +Step 29140 | loss:0.35097166895866394 lr:9.346224093010153e-06 tokens_per_second_per_gpu:4419.332427856476 mfu:20.846745078062842 +Step 29141 | loss:0.31829768419265747 lr:9.343699878730022e-06 tokens_per_second_per_gpu:4097.81623716007 mfu:19.330098350229157 +Step 29142 | loss:0.3335717022418976 lr:9.341175970231102e-06 tokens_per_second_per_gpu:3871.045367707186 mfu:18.260381467919874 +Step 29143 | loss:0.24528343975543976 lr:9.338652367532396e-06 tokens_per_second_per_gpu:4476.138152000468 mfu:21.114707370906007 +Step 29144 | loss:0.282666951417923 lr:9.336129070652866e-06 tokens_per_second_per_gpu:4626.055146497339 mfu:21.821891412424574 +Step 29145 | loss:0.2637026011943817 lr:9.333606079611495e-06 tokens_per_second_per_gpu:5323.881917276703 mfu:25.113659351715164 +Step 29146 | loss:0.2551158666610718 lr:9.33108339442727e-06 tokens_per_second_per_gpu:5082.854388120391 mfu:23.976691373147847 +Step 29147 | loss:0.26308050751686096 lr:9.328561015119148e-06 tokens_per_second_per_gpu:4679.480201159891 mfu:22.073906510522097 +Step 29148 | loss:0.2768445312976837 lr:9.326038941706122e-06 tokens_per_second_per_gpu:4496.9755322753035 mfu:21.213000848886082 +Step 29149 | loss:0.3492833375930786 lr:9.323517174207136e-06 tokens_per_second_per_gpu:3934.273241550671 mfu:18.558638136627277 +Step 29150 | loss:0.34233611822128296 lr:9.320995712641172e-06 tokens_per_second_per_gpu:4367.556481777735 mfu:20.6025091065221 +Step 29151 | loss:0.4706691801548004 lr:9.318474557027185e-06 tokens_per_second_per_gpu:4052.2478128878424 mfu:19.115144318162173 +Step 29152 | loss:0.2881161570549011 lr:9.31595370738414e-06 tokens_per_second_per_gpu:4415.431037230494 mfu:20.828341552834207 +Step 29153 | loss:0.39294755458831787 lr:9.313433163731011e-06 tokens_per_second_per_gpu:3352.9113938906967 mfu:15.816255110655757 +Step 29154 | loss:0.214866042137146 lr:9.310912926086729e-06 tokens_per_second_per_gpu:4922.492354145022 mfu:23.220236298301696 +Step 29155 | loss:0.4496874213218689 lr:9.308392994470277e-06 tokens_per_second_per_gpu:5334.212547044331 mfu:25.162390694916255 +Step 29156 | loss:0.2876346707344055 lr:9.30587336890058e-06 tokens_per_second_per_gpu:3796.3982954548655 mfu:17.90825849200185 +Step 29157 | loss:0.4233892261981964 lr:9.303354049396617e-06 tokens_per_second_per_gpu:4707.42226314821 mfu:22.20571398432837 +Step 29158 | loss:0.23745787143707275 lr:9.300835035977306e-06 tokens_per_second_per_gpu:4129.429769196597 mfu:19.47922477467021 +Step 29159 | loss:0.27684712409973145 lr:9.29831632866161e-06 tokens_per_second_per_gpu:4991.3831968890545 mfu:23.545206157515103 +Step 29160 | loss:0.35512304306030273 lr:9.29579792746847e-06 tokens_per_second_per_gpu:5344.346487267771 mfu:25.210194219978717 +Step 29161 | loss:0.2795757055282593 lr:9.293279832416835e-06 tokens_per_second_per_gpu:4649.294019265426 mfu:21.93151314023219 +Step 29162 | loss:0.4062596261501312 lr:9.290762043525642e-06 tokens_per_second_per_gpu:5313.233234753825 mfu:25.063427699402165 +Step 29163 | loss:0.2788313031196594 lr:9.288244560813808e-06 tokens_per_second_per_gpu:5326.487102916546 mfu:25.12594845686881 +Step 29164 | loss:0.3298001289367676 lr:9.28572738430028e-06 tokens_per_second_per_gpu:5281.344246925359 mfu:24.913001902992928 +Step 29165 | loss:0.3665320575237274 lr:9.283210514004009e-06 tokens_per_second_per_gpu:5289.099243582412 mfu:24.94958354536311 +Step 29166 | loss:0.2841971814632416 lr:9.280693949943891e-06 tokens_per_second_per_gpu:4702.933320646101 mfu:22.184538876653352 +Step 29167 | loss:0.2577309310436249 lr:9.278177692138867e-06 tokens_per_second_per_gpu:5324.084801503596 mfu:25.11461639122904 +Step 29168 | loss:0.2863261103630066 lr:9.275661740607882e-06 tokens_per_second_per_gpu:4041.1859763213347 mfu:19.062963747734017 +Step 29169 | loss:0.355251282453537 lr:9.273146095369828e-06 tokens_per_second_per_gpu:4561.181255696104 mfu:21.51587020088802 +Step 29170 | loss:0.27351945638656616 lr:9.270630756443654e-06 tokens_per_second_per_gpu:4192.099501097777 mfu:19.774848592606926 +Step 29171 | loss:0.3207148611545563 lr:9.268115723848247e-06 tokens_per_second_per_gpu:3979.010529007627 mfu:18.769671554530063 +Step 29172 | loss:0.3613312840461731 lr:9.265600997602552e-06 tokens_per_second_per_gpu:5325.662663730805 mfu:25.12205943656584 +Step 29173 | loss:0.2632940411567688 lr:9.263086577725455e-06 tokens_per_second_per_gpu:4203.219892190597 mfu:19.827305374726127 +Step 29174 | loss:0.2968199551105499 lr:9.260572464235884e-06 tokens_per_second_per_gpu:5315.729647722239 mfu:25.075203705306144 +Step 29175 | loss:0.3647134304046631 lr:9.258058657152762e-06 tokens_per_second_per_gpu:5332.632943543509 mfu:25.154939435694732 +Step 29176 | loss:0.4346349239349365 lr:9.25554515649496e-06 tokens_per_second_per_gpu:5342.530868740984 mfu:25.2016296376114 +Step 29177 | loss:0.28401264548301697 lr:9.253031962281421e-06 tokens_per_second_per_gpu:4451.059292416102 mfu:20.996406111351877 +Step 29178 | loss:0.32397836446762085 lr:9.250519074531011e-06 tokens_per_second_per_gpu:3662.4866278269283 mfu:17.27657430811439 +Step 29179 | loss:0.22737590968608856 lr:9.248006493262662e-06 tokens_per_second_per_gpu:2974.5791042898004 mfu:14.031597150463522 +Step 29180 | loss:0.3469054102897644 lr:9.245494218495244e-06 tokens_per_second_per_gpu:4782.946126931564 mfu:22.561972935494346 +Step 29181 | loss:0.3452625274658203 lr:9.242982250247661e-06 tokens_per_second_per_gpu:4695.314730460227 mfu:22.148600686881064 +Step 29182 | loss:0.36576005816459656 lr:9.240470588538824e-06 tokens_per_second_per_gpu:4103.572548060783 mfu:19.357251850875738 +Step 29183 | loss:0.29497525095939636 lr:9.237959233387594e-06 tokens_per_second_per_gpu:5168.767489143321 mfu:24.38195812895955 +Step 29184 | loss:0.43850401043891907 lr:9.23544818481289e-06 tokens_per_second_per_gpu:5013.464580781508 mfu:23.64936781280839 +Step 29185 | loss:0.28805676102638245 lr:9.232937442833567e-06 tokens_per_second_per_gpu:5329.418549361735 mfu:25.139776589907108 +Step 29186 | loss:0.3921273350715637 lr:9.230427007468535e-06 tokens_per_second_per_gpu:3819.677995694345 mfu:18.018072809957907 +Step 29187 | loss:0.3513604402542114 lr:9.227916878736652e-06 tokens_per_second_per_gpu:5243.924569107847 mfu:24.73648689827234 +Step 29188 | loss:0.2721760869026184 lr:9.22540705665681e-06 tokens_per_second_per_gpu:5078.81271855056 mfu:23.957626128207014 +Step 29189 | loss:0.3119656443595886 lr:9.222897541247893e-06 tokens_per_second_per_gpu:3821.0692594290654 mfu:18.02463563836865 +Step 29190 | loss:0.36391550302505493 lr:9.220388332528761e-06 tokens_per_second_per_gpu:4461.269139721693 mfu:21.04456770307193 +Step 29191 | loss:0.3091432452201843 lr:9.2178794305183e-06 tokens_per_second_per_gpu:3902.354004917195 mfu:18.408069651443824 +Step 29192 | loss:0.42749935388565063 lr:9.21537083523536e-06 tokens_per_second_per_gpu:4393.69672929552 mfu:20.725817114049615 +Step 29193 | loss:0.40699130296707153 lr:9.21286254669883e-06 tokens_per_second_per_gpu:4925.736382686101 mfu:23.235538934418308 +Step 29194 | loss:0.3307652175426483 lr:9.210354564927559e-06 tokens_per_second_per_gpu:5087.6916358769395 mfu:23.999509496135055 +Step 29195 | loss:0.30362585186958313 lr:9.207846889940408e-06 tokens_per_second_per_gpu:4483.6776503176625 mfu:21.150272470839273 +Step 29196 | loss:0.2483881413936615 lr:9.20533952175625e-06 tokens_per_second_per_gpu:4436.480458519258 mfu:20.927635264454906 +Step 29197 | loss:0.39406222105026245 lr:9.20283246039395e-06 tokens_per_second_per_gpu:4987.137733060977 mfu:23.52517958028724 +Step 29198 | loss:0.2612042725086212 lr:9.200325705872342e-06 tokens_per_second_per_gpu:5320.269929452497 mfu:25.096620988879717 +Step 29199 | loss:0.32717156410217285 lr:9.197819258210305e-06 tokens_per_second_per_gpu:4321.007959532022 mfu:20.38293178509222 +Step 29200 | loss:0.3057083785533905 lr:9.195313117426663e-06 tokens_per_second_per_gpu:5314.3604080722935 mfu:25.068744768260977 +Step 29201 | loss:0.2848913073539734 lr:9.192807283540283e-06 tokens_per_second_per_gpu:4536.51102064457 mfu:21.399496492973796 +Step 29202 | loss:0.24975033104419708 lr:9.190301756570001e-06 tokens_per_second_per_gpu:4390.890167886064 mfu:20.712578085032458 +Step 29203 | loss:0.21709990501403809 lr:9.187796536534665e-06 tokens_per_second_per_gpu:5085.285538942045 mfu:23.988159526370687 +Step 29204 | loss:0.32183048129081726 lr:9.18529162345313e-06 tokens_per_second_per_gpu:4316.76763809581 mfu:20.36292946540417 +Step 29205 | loss:0.20833121240139008 lr:9.182787017344213e-06 tokens_per_second_per_gpu:5249.794246314932 mfu:24.764175167128364 +Step 29206 | loss:0.2010650336742401 lr:9.18028271822678e-06 tokens_per_second_per_gpu:4358.3694794315 mfu:20.559172448990672 +Step 29207 | loss:0.2739593982696533 lr:9.17777872611963e-06 tokens_per_second_per_gpu:4517.279356648606 mfu:21.308777452644602 +Step 29208 | loss:0.37106722593307495 lr:9.175275041041631e-06 tokens_per_second_per_gpu:3788.8703460761203 mfu:17.8727478703817 +Step 29209 | loss:0.317371129989624 lr:9.172771663011587e-06 tokens_per_second_per_gpu:5346.716263853729 mfu:25.221372860460264 +Step 29210 | loss:0.34410953521728516 lr:9.17026859204833e-06 tokens_per_second_per_gpu:4516.8100069746515 mfu:21.30656345015325 +Step 29211 | loss:0.39998677372932434 lr:9.167765828170711e-06 tokens_per_second_per_gpu:3904.4204794398197 mfu:18.417817564343956 +Step 29212 | loss:0.35217899084091187 lr:9.165263371397526e-06 tokens_per_second_per_gpu:3599.524630158711 mfu:16.979571822688882 +Step 29213 | loss:0.2762729227542877 lr:9.162761221747613e-06 tokens_per_second_per_gpu:5343.806682423437 mfu:25.207647868427795 +Step 29214 | loss:0.3212815821170807 lr:9.160259379239772e-06 tokens_per_second_per_gpu:4157.344459802419 mfu:19.61090313299521 +Step 29215 | loss:0.32566213607788086 lr:9.157757843892845e-06 tokens_per_second_per_gpu:5357.977947630502 mfu:25.274496144276682 +Step 29216 | loss:0.25905877351760864 lr:9.155256615725615e-06 tokens_per_second_per_gpu:4314.92984648743 mfu:20.354260288827863 +Step 29217 | loss:0.3362590968608856 lr:9.15275569475692e-06 tokens_per_second_per_gpu:5314.336321855956 mfu:25.06863114946898 +Step 29218 | loss:0.27351272106170654 lr:9.15025508100557e-06 tokens_per_second_per_gpu:4291.97728403506 mfu:20.24598913562906 +Step 29219 | loss:0.23497125506401062 lr:9.147754774490352e-06 tokens_per_second_per_gpu:4952.570319661068 mfu:23.36211919347098 +Step 29220 | loss:0.3098841905593872 lr:9.145254775230094e-06 tokens_per_second_per_gpu:4375.4262917051365 mfu:20.639632342677682 +Step 29221 | loss:0.32664358615875244 lr:9.142755083243576e-06 tokens_per_second_per_gpu:5134.274098053194 mfu:24.219246918008043 +Step 29222 | loss:0.4351651966571808 lr:9.140255698549627e-06 tokens_per_second_per_gpu:4437.802284713904 mfu:20.933870544141858 +Step 29223 | loss:0.32954490184783936 lr:9.13775662116701e-06 tokens_per_second_per_gpu:3591.2062234480372 mfu:16.940332478968102 +Step 29224 | loss:0.2828884720802307 lr:9.135257851114543e-06 tokens_per_second_per_gpu:5324.346545319555 mfu:25.115851081467017 +Step 29225 | loss:0.30937525629997253 lr:9.132759388411027e-06 tokens_per_second_per_gpu:4978.281222066496 mfu:23.483401906849682 +Step 29226 | loss:0.29693424701690674 lr:9.130261233075232e-06 tokens_per_second_per_gpu:4977.368541627281 mfu:23.479096637498564 +Step 29227 | loss:0.310246080160141 lr:9.127763385125959e-06 tokens_per_second_per_gpu:3905.1454856633086 mfu:18.42123754239897 +Step 29228 | loss:0.2683348059654236 lr:9.125265844581998e-06 tokens_per_second_per_gpu:5277.972377462179 mfu:24.89709621943485 +Step 29229 | loss:0.44059404730796814 lr:9.122768611462135e-06 tokens_per_second_per_gpu:4717.090729483477 mfu:22.25132178964228 +Step 29230 | loss:0.2729014456272125 lr:9.12027168578513e-06 tokens_per_second_per_gpu:5339.620986087093 mfu:25.187903224655823 +Step 29231 | loss:0.3560447096824646 lr:9.117775067569778e-06 tokens_per_second_per_gpu:3713.982706899555 mfu:17.51949009923706 +Step 29232 | loss:0.3187936842441559 lr:9.115278756834855e-06 tokens_per_second_per_gpu:5357.896801518458 mfu:25.27411336422122 +Step 29233 | loss:0.38586875796318054 lr:9.112782753599153e-06 tokens_per_second_per_gpu:3689.252519098197 mfu:17.402833583972924 +Step 29234 | loss:0.30369627475738525 lr:9.110287057881416e-06 tokens_per_second_per_gpu:4908.652042243406 mfu:23.154949185661064 +Step 29235 | loss:0.3959331810474396 lr:9.107791669700438e-06 tokens_per_second_per_gpu:4967.209712360655 mfu:23.431175706572887 +Step 29236 | loss:0.321077436208725 lr:9.10529658907498e-06 tokens_per_second_per_gpu:3094.7604325805305 mfu:14.598512981059917 +Step 29237 | loss:0.3146900534629822 lr:9.102801816023792e-06 tokens_per_second_per_gpu:5334.237875846029 mfu:25.162510175195404 +Step 29238 | loss:0.3673288822174072 lr:9.100307350565645e-06 tokens_per_second_per_gpu:5273.1897083195945 mfu:24.874535553081785 +Step 29239 | loss:0.3322478234767914 lr:9.097813192719313e-06 tokens_per_second_per_gpu:3625.8670972405503 mfu:17.10383373440226 +Step 29240 | loss:0.26638683676719666 lr:9.095319342503555e-06 tokens_per_second_per_gpu:4686.115654101625 mfu:22.105207074173265 +Step 29241 | loss:0.3280274271965027 lr:9.092825799937104e-06 tokens_per_second_per_gpu:5326.004830333757 mfu:25.12367349481161 +Step 29242 | loss:0.30396512150764465 lr:9.090332565038744e-06 tokens_per_second_per_gpu:4936.154355491117 mfu:23.284682289629018 +Step 29243 | loss:0.29302525520324707 lr:9.087839637827205e-06 tokens_per_second_per_gpu:5246.556096043042 mfu:24.74890025981045 +Step 29244 | loss:0.33329057693481445 lr:9.085347018321256e-06 tokens_per_second_per_gpu:3772.797947364459 mfu:17.796931623424804 +Step 29245 | loss:0.3221113681793213 lr:9.082854706539617e-06 tokens_per_second_per_gpu:3520.6312021290014 mfu:16.60741806206591 +Step 29246 | loss:0.3028135597705841 lr:9.08036270250105e-06 tokens_per_second_per_gpu:5058.477270081863 mfu:23.861700348194887 +Step 29247 | loss:0.3203122019767761 lr:9.077871006224309e-06 tokens_per_second_per_gpu:5354.768869530793 mfu:25.259358375355173 +Step 29248 | loss:0.309315949678421 lr:9.075379617728108e-06 tokens_per_second_per_gpu:3486.4134996862285 mfu:16.446007321501302 +Step 29249 | loss:0.24640172719955444 lr:9.072888537031216e-06 tokens_per_second_per_gpu:5309.688242957697 mfu:25.04670536826186 +Step 29250 | loss:0.35249078273773193 lr:9.07039776415234e-06 tokens_per_second_per_gpu:4129.265553655588 mfu:19.478450141945213 +Step 29251 | loss:0.40580591559410095 lr:9.067907299110228e-06 tokens_per_second_per_gpu:4441.248951871982 mfu:20.95012906118857 +Step 29252 | loss:0.324051171541214 lr:9.065417141923604e-06 tokens_per_second_per_gpu:4509.3462504808385 mfu:21.27135563732376 +Step 29253 | loss:0.31384578347206116 lr:9.062927292611196e-06 tokens_per_second_per_gpu:5349.257445114898 mfu:25.233360046039422 +Step 29254 | loss:0.3472673296928406 lr:9.060437751191752e-06 tokens_per_second_per_gpu:4789.045766335803 mfu:22.590745975269932 +Step 29255 | loss:0.38338741660118103 lr:9.057948517683967e-06 tokens_per_second_per_gpu:4367.332344077263 mfu:20.601451810747136 +Step 29256 | loss:0.3093518614768982 lr:9.055459592106574e-06 tokens_per_second_per_gpu:5347.6589681768255 mfu:25.225819757594248 +Step 29257 | loss:0.2013077437877655 lr:9.052970974478303e-06 tokens_per_second_per_gpu:3436.278476103362 mfu:16.209511861343845 +Step 29258 | loss:0.2855437695980072 lr:9.050482664817866e-06 tokens_per_second_per_gpu:5330.472426765382 mfu:25.14474790943013 +Step 29259 | loss:0.24274419248104095 lr:9.04799466314396e-06 tokens_per_second_per_gpu:5358.590213545643 mfu:25.277384307061737 +Step 29260 | loss:0.3571939170360565 lr:9.045506969475315e-06 tokens_per_second_per_gpu:5340.217594030981 mfu:25.19071752611895 +Step 29261 | loss:0.24271197617053986 lr:9.043019583830647e-06 tokens_per_second_per_gpu:5092.8746517592535 mfu:24.023958666365626 +Step 29262 | loss:0.3445088267326355 lr:9.040532506228643e-06 tokens_per_second_per_gpu:5326.152090399959 mfu:25.124368145669038 +Step 29263 | loss:0.32455122470855713 lr:9.038045736688022e-06 tokens_per_second_per_gpu:5328.976385583055 mfu:25.137690827922153 +Step 29264 | loss:0.40407004952430725 lr:9.035559275227501e-06 tokens_per_second_per_gpu:5351.501244282976 mfu:25.243944429546865 +Step 29265 | loss:0.34267228841781616 lr:9.03307312186576e-06 tokens_per_second_per_gpu:4552.318110389241 mfu:21.474061232267022 +Step 29266 | loss:0.3122822046279907 lr:9.030587276621493e-06 tokens_per_second_per_gpu:4997.581521594834 mfu:23.57444471269551 +Step 29267 | loss:0.29246246814727783 lr:9.028101739513406e-06 tokens_per_second_per_gpu:5242.310195267371 mfu:24.728871621426237 +Step 29268 | loss:0.28171899914741516 lr:9.025616510560198e-06 tokens_per_second_per_gpu:4563.2580956951415 mfu:21.525667009508922 +Step 29269 | loss:0.2730369567871094 lr:9.023131589780564e-06 tokens_per_second_per_gpu:5334.821078994083 mfu:25.165261243949498 +Step 29270 | loss:0.2590562701225281 lr:9.020646977193175e-06 tokens_per_second_per_gpu:3845.956067159498 mfu:18.14203095656035 +Step 29271 | loss:0.24158932268619537 lr:9.018162672816743e-06 tokens_per_second_per_gpu:5317.630031433039 mfu:25.084168139508886 +Step 29272 | loss:0.2985919415950775 lr:9.015678676669936e-06 tokens_per_second_per_gpu:3711.0851828151876 mfu:17.505821983762512 +Step 29273 | loss:0.26993194222450256 lr:9.013194988771428e-06 tokens_per_second_per_gpu:3458.242513536701 mfu:16.313119973368117 +Step 29274 | loss:0.3483940660953522 lr:9.01071160913991e-06 tokens_per_second_per_gpu:5374.0790284098275 mfu:25.350447689440415 +Step 29275 | loss:0.3064953684806824 lr:9.00822853779406e-06 tokens_per_second_per_gpu:4936.682311371047 mfu:23.28717274759321 +Step 29276 | loss:0.34239062666893005 lr:9.005745774752572e-06 tokens_per_second_per_gpu:4075.1492282837617 mfu:19.223174201969105 +Step 29277 | loss:0.2895517945289612 lr:9.003263320034084e-06 tokens_per_second_per_gpu:4830.48297580093 mfu:22.7862123622339 +Step 29278 | loss:0.2551867663860321 lr:9.000781173657297e-06 tokens_per_second_per_gpu:4467.194524656862 mfu:21.072518754786042 +Step 29279 | loss:0.2764132618904114 lr:8.998299335640853e-06 tokens_per_second_per_gpu:4361.971160670183 mfu:20.57616219390373 +Step 29280 | loss:0.27209755778312683 lr:8.995817806003443e-06 tokens_per_second_per_gpu:5335.605214905577 mfu:25.16896014683139 +Step 29281 | loss:0.31599557399749756 lr:8.993336584763706e-06 tokens_per_second_per_gpu:4610.959963717201 mfu:21.750684859746908 +Step 29282 | loss:0.29021966457366943 lr:8.990855671940318e-06 tokens_per_second_per_gpu:5367.877090860186 mfu:25.321192091859103 +Step 29283 | loss:0.2559460401535034 lr:8.988375067551952e-06 tokens_per_second_per_gpu:4274.653280560376 mfu:20.16426885545919 +Step 29284 | loss:0.296955943107605 lr:8.985894771617237e-06 tokens_per_second_per_gpu:4913.917984093567 mfu:23.179789531830597 +Step 29285 | loss:0.27105972170829773 lr:8.983414784154853e-06 tokens_per_second_per_gpu:5156.601277960495 mfu:24.32456803504049 +Step 29286 | loss:0.34014105796813965 lr:8.98093510518343e-06 tokens_per_second_per_gpu:5340.828674108985 mfu:25.193600095107907 +Step 29287 | loss:0.30953067541122437 lr:8.978455734721636e-06 tokens_per_second_per_gpu:4124.563716197384 mfu:19.456270772438675 +Step 29288 | loss:0.2539045810699463 lr:8.9759766727881e-06 tokens_per_second_per_gpu:3062.5708738937437 mfu:14.446669340629304 +Step 29289 | loss:0.2197597324848175 lr:8.97349791940148e-06 tokens_per_second_per_gpu:4500.391006933848 mfu:21.229112225590505 +Step 29290 | loss:0.3579912483692169 lr:8.971019474580427e-06 tokens_per_second_per_gpu:4816.849886239549 mfu:22.72190275272793 +Step 29291 | loss:0.30959904193878174 lr:8.968541338343561e-06 tokens_per_second_per_gpu:5027.537057623195 mfu:23.715750087082764 +Step 29292 | loss:0.3617120087146759 lr:8.966063510709533e-06 tokens_per_second_per_gpu:5355.6945501543105 mfu:25.263724968793706 +Step 29293 | loss:0.2565697133541107 lr:8.963585991696988e-06 tokens_per_second_per_gpu:5334.760848045118 mfu:25.164977124286768 +Step 29294 | loss:0.25717219710350037 lr:8.961108781324551e-06 tokens_per_second_per_gpu:3819.677725978225 mfu:18.01807153766095 +Step 29295 | loss:0.21874713897705078 lr:8.958631879610835e-06 tokens_per_second_per_gpu:5352.4270688717415 mfu:25.248311702094036 +Step 29296 | loss:0.348778635263443 lr:8.956155286574486e-06 tokens_per_second_per_gpu:5056.204699352864 mfu:23.85098024432565 +Step 29297 | loss:0.3156149089336395 lr:8.953679002234144e-06 tokens_per_second_per_gpu:4252.407918662797 mfu:20.059333687938196 +Step 29298 | loss:0.2520087659358978 lr:8.951203026608407e-06 tokens_per_second_per_gpu:4507.659065453033 mfu:21.26339689768415 +Step 29299 | loss:0.39866045117378235 lr:8.948727359715914e-06 tokens_per_second_per_gpu:5292.920035347394 mfu:24.96760687201283 +Step 29300 | loss:0.31053221225738525 lr:8.94625200157529e-06 tokens_per_second_per_gpu:5302.034126700741 mfu:25.0105995959509 +Step 29301 | loss:0.27460524439811707 lr:8.94377695220514e-06 tokens_per_second_per_gpu:5316.5273408727935 mfu:25.078966559997795 +Step 29302 | loss:0.31167638301849365 lr:8.941302211624076e-06 tokens_per_second_per_gpu:5356.2500342137055 mfu:25.266345281877225 +Step 29303 | loss:0.2764708697795868 lr:8.93882777985071e-06 tokens_per_second_per_gpu:5329.504390836106 mfu:25.140181518787777 +Step 29304 | loss:0.37375408411026 lr:8.93635365690368e-06 tokens_per_second_per_gpu:4272.710977692585 mfu:20.155106681440785 +Step 29305 | loss:0.3275870978832245 lr:8.933879842801557e-06 tokens_per_second_per_gpu:4163.544508212066 mfu:19.640149819180856 +Step 29306 | loss:0.2665170431137085 lr:8.931406337562959e-06 tokens_per_second_per_gpu:5293.554573136139 mfu:24.97060009502565 +Step 29307 | loss:0.2448427826166153 lr:8.928933141206513e-06 tokens_per_second_per_gpu:5368.201057646117 mfu:25.32272029846277 +Step 29308 | loss:0.3419354557991028 lr:8.926460253750802e-06 tokens_per_second_per_gpu:5317.517729930372 mfu:25.083638394160193 +Step 29309 | loss:0.3728088140487671 lr:8.923987675214412e-06 tokens_per_second_per_gpu:4407.184436365477 mfu:20.789440929538728 +Step 29310 | loss:0.3301629424095154 lr:8.921515405615949e-06 tokens_per_second_per_gpu:5341.262945487212 mfu:25.19564862729334 +Step 29311 | loss:0.27984076738357544 lr:8.919043444974012e-06 tokens_per_second_per_gpu:5343.253823238322 mfu:25.205039937324102 +Step 29312 | loss:0.30365046858787537 lr:8.916571793307205e-06 tokens_per_second_per_gpu:4705.489547263079 mfu:22.196597033742872 +Step 29313 | loss:0.28623610734939575 lr:8.91410045063409e-06 tokens_per_second_per_gpu:4202.254340058363 mfu:19.822750700578194 +Step 29314 | loss:0.4352568984031677 lr:8.911629416973277e-06 tokens_per_second_per_gpu:4626.792127827971 mfu:21.825367879103148 +Step 29315 | loss:0.3091743290424347 lr:8.909158692343333e-06 tokens_per_second_per_gpu:5321.383301470103 mfu:25.101872954647614 +Step 29316 | loss:0.25098541378974915 lr:8.906688276762865e-06 tokens_per_second_per_gpu:5347.440876630465 mfu:25.22479098255981 +Step 29317 | loss:0.2574591338634491 lr:8.904218170250422e-06 tokens_per_second_per_gpu:3760.5957465432166 mfu:17.739371760240864 +Step 29318 | loss:0.2962490916252136 lr:8.901748372824597e-06 tokens_per_second_per_gpu:5358.683900273393 mfu:25.277826243341142 +Step 29319 | loss:0.3571217656135559 lr:8.899278884503982e-06 tokens_per_second_per_gpu:4852.547398858907 mfu:22.890294010377144 +Step 29320 | loss:0.3239941895008087 lr:8.896809705307124e-06 tokens_per_second_per_gpu:5301.681090155614 mfu:25.00893426233329 +Step 29321 | loss:0.29457056522369385 lr:8.894340835252601e-06 tokens_per_second_per_gpu:5300.36936020983 mfu:25.002746608374824 +Step 29322 | loss:0.30054357647895813 lr:8.891872274358998e-06 tokens_per_second_per_gpu:5267.336135738106 mfu:24.846923252492434 +Step 29323 | loss:0.23762710392475128 lr:8.889404022644871e-06 tokens_per_second_per_gpu:4981.951680084176 mfu:23.50071608356379 +Step 29324 | loss:0.34551694989204407 lr:8.886936080128771e-06 tokens_per_second_per_gpu:5220.225563310031 mfu:24.624694644456707 +Step 29325 | loss:0.33335182070732117 lr:8.88446844682927e-06 tokens_per_second_per_gpu:5341.685116372997 mfu:25.197640079391373 +Step 29326 | loss:0.2062370628118515 lr:8.882001122764939e-06 tokens_per_second_per_gpu:3807.088572138661 mfu:17.958686351068593 +Step 29327 | loss:0.227009579539299 lr:8.879534107954312e-06 tokens_per_second_per_gpu:4532.714894691509 mfu:21.381589519167328 +Step 29328 | loss:0.39789384603500366 lr:8.877067402415956e-06 tokens_per_second_per_gpu:5201.191402630315 mfu:24.534907260967795 +Step 29329 | loss:0.2918323576450348 lr:8.874601006168437e-06 tokens_per_second_per_gpu:4635.505020754166 mfu:21.866468081608833 +Step 29330 | loss:0.27304205298423767 lr:8.87213491923029e-06 tokens_per_second_per_gpu:5296.730941587875 mfu:24.985583566956834 +Step 29331 | loss:0.31982746720314026 lr:8.869669141620052e-06 tokens_per_second_per_gpu:5314.07759992714 mfu:25.06741071398824 +Step 29332 | loss:0.2923656105995178 lr:8.867203673356284e-06 tokens_per_second_per_gpu:4429.601157761003 mfu:20.89518442904848 +Step 29333 | loss:0.3621841371059418 lr:8.864738514457533e-06 tokens_per_second_per_gpu:4068.8451729214858 mfu:19.193436897242496 +Step 29334 | loss:0.22489535808563232 lr:8.862273664942322e-06 tokens_per_second_per_gpu:5330.563055723162 mfu:25.145175421686876 +Step 29335 | loss:0.32224220037460327 lr:8.8598091248292e-06 tokens_per_second_per_gpu:5182.959207963971 mfu:24.44890288799347 +Step 29336 | loss:0.28405100107192993 lr:8.857344894136715e-06 tokens_per_second_per_gpu:3578.2476482955685 mfu:16.879204669013 +Step 29337 | loss:0.3261660933494568 lr:8.854880972883389e-06 tokens_per_second_per_gpu:3726.1473879428695 mfu:17.57687297522698 +Step 29338 | loss:0.4243019223213196 lr:8.852417361087745e-06 tokens_per_second_per_gpu:4019.6209872117324 mfu:18.961238014737475 +Step 29339 | loss:0.3312850296497345 lr:8.849954058768317e-06 tokens_per_second_per_gpu:5338.048058107139 mfu:25.180483455005188 +Step 29340 | loss:0.275346964597702 lr:8.847491065943648e-06 tokens_per_second_per_gpu:5182.334321187084 mfu:24.44595519045022 +Step 29341 | loss:0.23146961629390717 lr:8.845028382632237e-06 tokens_per_second_per_gpu:5335.354946975211 mfu:25.167779590303745 +Step 29342 | loss:0.27045658230781555 lr:8.84256600885262e-06 tokens_per_second_per_gpu:5352.327894813662 mfu:25.247843881141232 +Step 29343 | loss:0.26280084252357483 lr:8.840103944623328e-06 tokens_per_second_per_gpu:3426.716615318246 mfu:16.16440693259878 +Step 29344 | loss:0.42559510469436646 lr:8.837642189962864e-06 tokens_per_second_per_gpu:4485.254739692563 mfu:21.15771187049987 +Step 29345 | loss:0.33456072211265564 lr:8.835180744889737e-06 tokens_per_second_per_gpu:2916.1278639926813 mfu:13.755872677172055 +Step 29346 | loss:0.24861204624176025 lr:8.832719609422463e-06 tokens_per_second_per_gpu:5139.278942782892 mfu:24.242855624512593 +Step 29347 | loss:0.3582836389541626 lr:8.830258783579565e-06 tokens_per_second_per_gpu:3793.9463487303346 mfu:17.896692240956988 +Step 29348 | loss:0.3287105858325958 lr:8.82779826737955e-06 tokens_per_second_per_gpu:5357.51263797919 mfu:25.27230119926143 +Step 29349 | loss:0.40597081184387207 lr:8.82533806084091e-06 tokens_per_second_per_gpu:4521.940388013828 mfu:21.33076433284831 +Step 29350 | loss:0.30609244108200073 lr:8.822878163982151e-06 tokens_per_second_per_gpu:4178.5955920768565 mfu:19.711148349750783 +Step 29351 | loss:0.30779457092285156 lr:8.820418576821792e-06 tokens_per_second_per_gpu:4313.009144545435 mfu:20.34520001006193 +Step 29352 | loss:0.3312949538230896 lr:8.817959299378326e-06 tokens_per_second_per_gpu:4888.037167869764 mfu:23.057705305980335 +Step 29353 | loss:0.41239994764328003 lr:8.815500331670229e-06 tokens_per_second_per_gpu:3932.7805976597106 mfu:18.551597080722306 +Step 29354 | loss:0.30429011583328247 lr:8.813041673716005e-06 tokens_per_second_per_gpu:5245.842109380264 mfu:24.745532262903442 +Step 29355 | loss:0.3577299118041992 lr:8.810583325534166e-06 tokens_per_second_per_gpu:5316.350683040863 mfu:25.07813323486365 +Step 29356 | loss:0.22016209363937378 lr:8.808125287143171e-06 tokens_per_second_per_gpu:5039.511012129166 mfu:23.77223327345473 +Step 29357 | loss:0.24393883347511292 lr:8.80566755856152e-06 tokens_per_second_per_gpu:5335.502029081873 mfu:25.168473401695664 +Step 29358 | loss:0.31872427463531494 lr:8.803210139807716e-06 tokens_per_second_per_gpu:3844.116158351242 mfu:18.133351792790993 +Step 29359 | loss:0.24749845266342163 lr:8.800753030900228e-06 tokens_per_second_per_gpu:3821.1071971154406 mfu:18.024814596908122 +Step 29360 | loss:0.2501186430454254 lr:8.798296231857516e-06 tokens_per_second_per_gpu:5209.04372035049 mfu:24.57194798339741 +Step 29361 | loss:0.25494223833084106 lr:8.795839742698075e-06 tokens_per_second_per_gpu:5358.660472163316 mfu:25.27771572894866 +Step 29362 | loss:0.35924074053764343 lr:8.793383563440394e-06 tokens_per_second_per_gpu:3738.033141014639 mfu:17.632940100385103 +Step 29363 | loss:0.28435590863227844 lr:8.790927694102924e-06 tokens_per_second_per_gpu:5326.961901005449 mfu:25.128188160463075 +Step 29364 | loss:0.2134067863225937 lr:8.78847213470414e-06 tokens_per_second_per_gpu:3143.2217209664605 mfu:14.827113146724752 +Step 29365 | loss:0.364810585975647 lr:8.786016885262533e-06 tokens_per_second_per_gpu:3919.848077427449 mfu:18.490592176271246 +Step 29366 | loss:0.29793787002563477 lr:8.783561945796548e-06 tokens_per_second_per_gpu:4367.66913089327 mfu:20.60304049162 +Step 29367 | loss:0.22838377952575684 lr:8.781107316324644e-06 tokens_per_second_per_gpu:5338.248846937167 mfu:25.18143061017454 +Step 29368 | loss:0.2614782154560089 lr:8.778652996865288e-06 tokens_per_second_per_gpu:4162.264006173409 mfu:19.634109472588335 +Step 29369 | loss:0.31443020701408386 lr:8.776198987436963e-06 tokens_per_second_per_gpu:4848.441415497366 mfu:22.87092538630775 +Step 29370 | loss:0.29881417751312256 lr:8.773745288058082e-06 tokens_per_second_per_gpu:4348.179602434813 mfu:20.51110506062501 +Step 29371 | loss:0.3777238726615906 lr:8.77129189874713e-06 tokens_per_second_per_gpu:4233.856809363888 mfu:19.971824940229617 +Step 29372 | loss:0.24448701739311218 lr:8.768838819522563e-06 tokens_per_second_per_gpu:4869.538999010993 mfu:22.97044628736046 +Step 29373 | loss:0.26545676589012146 lr:8.766386050402819e-06 tokens_per_second_per_gpu:5361.096033145446 mfu:25.28920468565092 +Step 29374 | loss:0.30930885672569275 lr:8.763933591406337e-06 tokens_per_second_per_gpu:4135.146509464868 mfu:19.506191613892117 +Step 29375 | loss:0.29016685485839844 lr:8.761481442551573e-06 tokens_per_second_per_gpu:5232.183026555341 mfu:24.681100038738435 +Step 29376 | loss:0.3439542353153229 lr:8.759029603856977e-06 tokens_per_second_per_gpu:4942.381508981637 mfu:23.31405683510575 +Step 29377 | loss:0.29740628600120544 lr:8.756578075340977e-06 tokens_per_second_per_gpu:4631.473577907025 mfu:21.847451077863795 +Step 29378 | loss:0.2327803373336792 lr:8.754126857022011e-06 tokens_per_second_per_gpu:4007.0233795851336 mfu:18.901812950189314 +Step 29379 | loss:0.2806292176246643 lr:8.75167594891853e-06 tokens_per_second_per_gpu:4088.5546050000416 mfu:19.286409651132914 +Step 29380 | loss:0.38084131479263306 lr:8.749225351048962e-06 tokens_per_second_per_gpu:4581.767365564574 mfu:21.612978393490117 +Step 29381 | loss:0.2575317323207855 lr:8.746775063431716e-06 tokens_per_second_per_gpu:5345.394430495303 mfu:25.21513754698054 +Step 29382 | loss:0.30433252453804016 lr:8.744325086085248e-06 tokens_per_second_per_gpu:3300.1295233744754 mfu:15.567274021915011 +Step 29383 | loss:0.19726914167404175 lr:8.741875419027966e-06 tokens_per_second_per_gpu:4778.365749605757 mfu:22.54036651415555 +Step 29384 | loss:0.23268386721611023 lr:8.739426062278322e-06 tokens_per_second_per_gpu:3332.897077235219 mfu:15.721844164197325 +Step 29385 | loss:0.33839452266693115 lr:8.736977015854703e-06 tokens_per_second_per_gpu:4100.025476164309 mfu:19.340519707547195 +Step 29386 | loss:0.4338996410369873 lr:8.734528279775544e-06 tokens_per_second_per_gpu:4059.0913428220465 mfu:19.147426416488422 +Step 29387 | loss:0.4067242741584778 lr:8.732079854059277e-06 tokens_per_second_per_gpu:5307.090333672231 mfu:25.034450586913287 +Step 29388 | loss:0.29959169030189514 lr:8.729631738724308e-06 tokens_per_second_per_gpu:4121.792678745399 mfu:19.44319931598992 +Step 29389 | loss:0.27398067712783813 lr:8.72718393378903e-06 tokens_per_second_per_gpu:5358.99821716526 mfu:25.27930892974807 +Step 29390 | loss:0.31094762682914734 lr:8.724736439271864e-06 tokens_per_second_per_gpu:5335.154988995921 mfu:25.16683635439908 +Step 29391 | loss:0.29918643832206726 lr:8.72228925519124e-06 tokens_per_second_per_gpu:4791.720086546777 mfu:22.603361199991173 +Step 29392 | loss:0.2572060823440552 lr:8.71984238156553e-06 tokens_per_second_per_gpu:3847.6649097549384 mfu:18.15009185864165 +Step 29393 | loss:0.3959392309188843 lr:8.717395818413154e-06 tokens_per_second_per_gpu:5307.738630051931 mfu:25.037508711547098 +Step 29394 | loss:0.25392967462539673 lr:8.714949565752522e-06 tokens_per_second_per_gpu:5237.37785360672 mfu:24.705604924268684 +Step 29395 | loss:0.2730391323566437 lr:8.712503623602025e-06 tokens_per_second_per_gpu:5367.472558697278 mfu:25.31928384462636 +Step 29396 | loss:0.3237067461013794 lr:8.710057991980041e-06 tokens_per_second_per_gpu:4338.328752565816 mfu:20.464636921064656 +Step 29397 | loss:0.3627762794494629 lr:8.707612670904986e-06 tokens_per_second_per_gpu:5348.210815918195 mfu:25.228422917545224 +Step 29398 | loss:0.25020521879196167 lr:8.705167660395252e-06 tokens_per_second_per_gpu:5317.54249382503 mfu:25.08375520967645 +Step 29399 | loss:0.27625566720962524 lr:8.702722960469206e-06 tokens_per_second_per_gpu:5261.940757176138 mfu:24.82147233126912 +Step 29400 | loss:0.2728482186794281 lr:8.700278571145253e-06 tokens_per_second_per_gpu:5319.247608793311 mfu:25.091798527904444 +Step 29401 | loss:0.29031991958618164 lr:8.697834492441792e-06 tokens_per_second_per_gpu:4235.890299413253 mfu:19.981417259741654 +Step 29402 | loss:0.3418046236038208 lr:8.695390724377183e-06 tokens_per_second_per_gpu:4120.242163255944 mfu:19.43588526988094 +Step 29403 | loss:0.3706536293029785 lr:8.692947266969797e-06 tokens_per_second_per_gpu:3052.6811589369054 mfu:14.400017867818532 +Step 29404 | loss:0.33423858880996704 lr:8.69050412023803e-06 tokens_per_second_per_gpu:4433.2273636711925 mfu:20.91228986102137 +Step 29405 | loss:0.3615436255931854 lr:8.688061284200266e-06 tokens_per_second_per_gpu:5370.782735812708 mfu:25.33489851485429 +Step 29406 | loss:0.2867288589477539 lr:8.685618758874847e-06 tokens_per_second_per_gpu:4467.1339292038865 mfu:21.072232915695704 +Step 29407 | loss:0.3676520586013794 lr:8.683176544280164e-06 tokens_per_second_per_gpu:5353.830984925168 mfu:25.254934213650827 +Step 29408 | loss:0.26097846031188965 lr:8.680734640434596e-06 tokens_per_second_per_gpu:4991.168947053576 mfu:23.544195504486357 +Step 29409 | loss:0.33719560503959656 lr:8.67829304735649e-06 tokens_per_second_per_gpu:4899.071114759547 mfu:23.10975431604491 +Step 29410 | loss:0.30196094512939453 lr:8.675851765064207e-06 tokens_per_second_per_gpu:5382.140719621029 mfu:25.38847606235001 +Step 29411 | loss:0.3175629675388336 lr:8.673410793576115e-06 tokens_per_second_per_gpu:3998.6167626862957 mfu:18.862157503960752 +Step 29412 | loss:0.38870206475257874 lr:8.670970132910583e-06 tokens_per_second_per_gpu:4007.912096609817 mfu:18.906005180025506 +Step 29413 | loss:0.4004785418510437 lr:8.668529783085949e-06 tokens_per_second_per_gpu:3642.879791330363 mfu:17.18408551508883 +Step 29414 | loss:0.21530352532863617 lr:8.666089744120575e-06 tokens_per_second_per_gpu:5321.758406671251 mfu:25.103642389880978 +Step 29415 | loss:0.3849587142467499 lr:8.663650016032815e-06 tokens_per_second_per_gpu:4852.766986640269 mfu:22.89132984340737 +Step 29416 | loss:0.3605482280254364 lr:8.66121059884104e-06 tokens_per_second_per_gpu:4135.889499727633 mfu:19.509696425728787 +Step 29417 | loss:0.2868640124797821 lr:8.658771492563555e-06 tokens_per_second_per_gpu:5013.198321738626 mfu:23.648111823474636 +Step 29418 | loss:0.33584681153297424 lr:8.656332697218722e-06 tokens_per_second_per_gpu:4930.811167289554 mfu:23.25947755924033 +Step 29419 | loss:0.2991122901439667 lr:8.653894212824887e-06 tokens_per_second_per_gpu:4563.085774951348 mfu:21.524854143159494 +Step 29420 | loss:0.36939993500709534 lr:8.651456039400401e-06 tokens_per_second_per_gpu:3634.5452492564573 mfu:17.144770058106708 +Step 29421 | loss:0.28483012318611145 lr:8.649018176963581e-06 tokens_per_second_per_gpu:5026.949420924895 mfu:23.71297810451572 +Step 29422 | loss:0.31198474764823914 lr:8.64658062553277e-06 tokens_per_second_per_gpu:5308.000964743523 mfu:25.03874618904625 +Step 29423 | loss:0.40002068877220154 lr:8.644143385126313e-06 tokens_per_second_per_gpu:5316.117150697241 mfu:25.07703162295406 +Step 29424 | loss:0.322806715965271 lr:8.641706455762532e-06 tokens_per_second_per_gpu:5297.574551523955 mfu:24.989563018960016 +Step 29425 | loss:0.3419056534767151 lr:8.639269837459741e-06 tokens_per_second_per_gpu:5278.364208556883 mfu:24.898944553561847 +Step 29426 | loss:0.33295273780822754 lr:8.636833530236278e-06 tokens_per_second_per_gpu:5342.918733224397 mfu:25.203459260556166 +Step 29427 | loss:0.3689221441745758 lr:8.634397534110484e-06 tokens_per_second_per_gpu:5347.874128477377 mfu:25.226834705442084 +Step 29428 | loss:0.2316330224275589 lr:8.631961849100651e-06 tokens_per_second_per_gpu:4987.119394947222 mfu:23.52509307627585 +Step 29429 | loss:0.3323556184768677 lr:8.629526475225108e-06 tokens_per_second_per_gpu:5354.062150666337 mfu:25.256024661892866 +Step 29430 | loss:0.361571341753006 lr:8.627091412502192e-06 tokens_per_second_per_gpu:3853.683568619719 mfu:18.178482899395917 +Step 29431 | loss:0.33711710572242737 lr:8.624656660950197e-06 tokens_per_second_per_gpu:4804.422913102358 mfu:22.663282600179546 +Step 29432 | loss:0.31976819038391113 lr:8.622222220587427e-06 tokens_per_second_per_gpu:5340.184892662569 mfu:25.190563268184725 +Step 29433 | loss:0.3566496968269348 lr:8.619788091432206e-06 tokens_per_second_per_gpu:4605.769633118278 mfu:21.726201184750067 +Step 29434 | loss:0.3334232568740845 lr:8.617354273502853e-06 tokens_per_second_per_gpu:5338.4774726083 mfu:25.182509076472833 +Step 29435 | loss:0.35336223244667053 lr:8.61492076681764e-06 tokens_per_second_per_gpu:5343.393828011234 mfu:25.205700363725207 +Step 29436 | loss:0.35306763648986816 lr:8.612487571394894e-06 tokens_per_second_per_gpu:3418.9526802089513 mfu:16.12778312602431 +Step 29437 | loss:0.2725909650325775 lr:8.610054687252923e-06 tokens_per_second_per_gpu:4368.090322851651 mfu:20.605027325950438 +Step 29438 | loss:0.23449863493442535 lr:8.607622114410013e-06 tokens_per_second_per_gpu:5341.913421710502 mfu:25.19871703462129 +Step 29439 | loss:0.23112516105175018 lr:8.605189852884442e-06 tokens_per_second_per_gpu:4543.054255788309 mfu:21.430362049537848 +Step 29440 | loss:0.23594819009304047 lr:8.602757902694519e-06 tokens_per_second_per_gpu:4654.093751369021 mfu:21.95415429548359 +Step 29441 | loss:0.43195030093193054 lr:8.600326263858555e-06 tokens_per_second_per_gpu:5370.564706814438 mfu:25.33387003486987 +Step 29442 | loss:0.3872990906238556 lr:8.597894936394801e-06 tokens_per_second_per_gpu:4317.632672032919 mfu:20.367009978075522 +Step 29443 | loss:0.3609921336174011 lr:8.595463920321566e-06 tokens_per_second_per_gpu:5040.267857095266 mfu:23.775803440290886 +Step 29444 | loss:0.3115079998970032 lr:8.593033215657142e-06 tokens_per_second_per_gpu:5373.882181870953 mfu:25.34951913073106 +Step 29445 | loss:0.3012196123600006 lr:8.590602822419797e-06 tokens_per_second_per_gpu:4531.132099575098 mfu:21.374123204550532 +Step 29446 | loss:0.22579404711723328 lr:8.588172740627798e-06 tokens_per_second_per_gpu:3145.4694013926282 mfu:14.83771583878881 +Step 29447 | loss:0.2779560387134552 lr:8.585742970299443e-06 tokens_per_second_per_gpu:4951.439047769902 mfu:23.35678279094546 +Step 29448 | loss:0.3837050497531891 lr:8.583313511453007e-06 tokens_per_second_per_gpu:5124.312085393378 mfu:24.172254404597062 +Step 29449 | loss:0.2841545641422272 lr:8.580884364106739e-06 tokens_per_second_per_gpu:5330.6759509171725 mfu:25.14570796757893 +Step 29450 | loss:0.2769981324672699 lr:8.578455528278934e-06 tokens_per_second_per_gpu:4668.979373514431 mfu:22.024372314892517 +Step 29451 | loss:0.2635335922241211 lr:8.57602700398784e-06 tokens_per_second_per_gpu:4823.010141071606 mfu:22.750961725818602 +Step 29452 | loss:0.27673473954200745 lr:8.573598791251769e-06 tokens_per_second_per_gpu:4724.236444968246 mfu:22.28502934876872 +Step 29453 | loss:0.28485414385795593 lr:8.571170890088914e-06 tokens_per_second_per_gpu:4166.127526495951 mfu:19.652334357134304 +Step 29454 | loss:0.32927194237709045 lr:8.568743300517573e-06 tokens_per_second_per_gpu:4190.398838858185 mfu:19.76682627865987 +Step 29455 | loss:0.36969226598739624 lr:8.566316022555998e-06 tokens_per_second_per_gpu:5051.936826199466 mfu:23.830847958486828 +Step 29456 | loss:0.26220300793647766 lr:8.563889056222463e-06 tokens_per_second_per_gpu:5326.617322844061 mfu:25.126562726484707 +Step 29457 | loss:0.3212170898914337 lr:8.561462401535186e-06 tokens_per_second_per_gpu:5226.879663603567 mfu:24.656083170847896 +Step 29458 | loss:0.27869394421577454 lr:8.559036058512444e-06 tokens_per_second_per_gpu:4517.1981544021355 mfu:21.308394407793468 +Step 29459 | loss:0.27052438259124756 lr:8.556610027172491e-06 tokens_per_second_per_gpu:4528.262014071862 mfu:21.36058452154468 +Step 29460 | loss:0.34563398361206055 lr:8.554184307533563e-06 tokens_per_second_per_gpu:4683.482367252101 mfu:22.092785410818994 +Step 29461 | loss:0.35706937313079834 lr:8.551758899613888e-06 tokens_per_second_per_gpu:4313.2702820108525 mfu:20.346431840040946 +Step 29462 | loss:0.43318361043930054 lr:8.549333803431725e-06 tokens_per_second_per_gpu:4581.405604560729 mfu:21.61127190511219 +Step 29463 | loss:0.3048761188983917 lr:8.546909019005328e-06 tokens_per_second_per_gpu:4487.980605792035 mfu:21.170570245971795 +Step 29464 | loss:0.3360212445259094 lr:8.544484546352899e-06 tokens_per_second_per_gpu:5339.224925375014 mfu:25.18603493870202 +Step 29465 | loss:0.2716045081615448 lr:8.542060385492695e-06 tokens_per_second_per_gpu:3681.1433483484857 mfu:17.364581241979337 +Step 29466 | loss:0.3228203058242798 lr:8.539636536442952e-06 tokens_per_second_per_gpu:4821.763883812722 mfu:22.74508291769503 +Step 29467 | loss:0.2471047192811966 lr:8.5372129992219e-06 tokens_per_second_per_gpu:4197.838733269815 mfu:19.801921529976408 +Step 29468 | loss:0.2087647020816803 lr:8.534789773847745e-06 tokens_per_second_per_gpu:4381.388421440999 mfu:20.667756725885972 +Step 29469 | loss:0.22426575422286987 lr:8.532366860338725e-06 tokens_per_second_per_gpu:4646.6663873740235 mfu:21.919118152280216 +Step 29470 | loss:0.31681007146835327 lr:8.52994425871308e-06 tokens_per_second_per_gpu:5337.053219809534 mfu:25.175790633018472 +Step 29471 | loss:0.278470903635025 lr:8.527521968989005e-06 tokens_per_second_per_gpu:4999.651924204772 mfu:23.584211155053698 +Step 29472 | loss:0.35815539956092834 lr:8.525099991184726e-06 tokens_per_second_per_gpu:5351.701117258494 mfu:25.24488726447423 +Step 29473 | loss:0.32373297214508057 lr:8.522678325318472e-06 tokens_per_second_per_gpu:5195.25191883309 mfu:24.50688970251585 +Step 29474 | loss:0.3685857057571411 lr:8.520256971408453e-06 tokens_per_second_per_gpu:4609.5136211158 mfu:21.743862214924597 +Step 29475 | loss:0.2689298093318939 lr:8.517835929472856e-06 tokens_per_second_per_gpu:4191.407802535672 mfu:19.771585732473532 +Step 29476 | loss:0.2708459496498108 lr:8.515415199529914e-06 tokens_per_second_per_gpu:4630.272141201433 mfu:21.841783695937277 +Step 29477 | loss:0.24285349249839783 lr:8.512994781597838e-06 tokens_per_second_per_gpu:5353.699485638658 mfu:25.254313909080267 +Step 29478 | loss:0.23047900199890137 lr:8.510574675694811e-06 tokens_per_second_per_gpu:4759.28577105766 mfu:22.450362999964156 +Step 29479 | loss:0.33694320917129517 lr:8.508154881839042e-06 tokens_per_second_per_gpu:4297.897027823876 mfu:20.273913577093342 +Step 29480 | loss:0.30993038415908813 lr:8.505735400048748e-06 tokens_per_second_per_gpu:5304.284982392264 mfu:25.021217266283397 +Step 29481 | loss:0.26992055773735046 lr:8.503316230342112e-06 tokens_per_second_per_gpu:4230.933517882357 mfu:19.958035275546276 +Step 29482 | loss:0.31489601731300354 lr:8.500897372737321e-06 tokens_per_second_per_gpu:5378.674198946115 mfu:25.372123892877024 +Step 29483 | loss:0.39829006791114807 lr:8.498478827252571e-06 tokens_per_second_per_gpu:5319.687842620357 mfu:25.093875185978252 +Step 29484 | loss:0.3011317253112793 lr:8.496060593906074e-06 tokens_per_second_per_gpu:5333.642427988221 mfu:25.159701346056245 +Step 29485 | loss:0.3068850040435791 lr:8.49364267271599e-06 tokens_per_second_per_gpu:5332.8954865142 mfu:25.156177895681928 +Step 29486 | loss:0.29426905512809753 lr:8.491225063700509e-06 tokens_per_second_per_gpu:4642.521141970091 mfu:21.899564322458065 +Step 29487 | loss:0.20520402491092682 lr:8.488807766877826e-06 tokens_per_second_per_gpu:5371.534850922972 mfu:25.338446370153477 +Step 29488 | loss:0.3519054055213928 lr:8.48639078226614e-06 tokens_per_second_per_gpu:4657.561924506454 mfu:21.97051425990421 +Step 29489 | loss:0.3014904856681824 lr:8.483974109883586e-06 tokens_per_second_per_gpu:4025.149184873192 mfu:18.987315466314367 +Step 29490 | loss:0.31886032223701477 lr:8.481557749748353e-06 tokens_per_second_per_gpu:3141.3905503775145 mfu:14.818475202626775 +Step 29491 | loss:0.33297523856163025 lr:8.479141701878624e-06 tokens_per_second_per_gpu:4792.328965450632 mfu:22.606233385666112 +Step 29492 | loss:0.25092217326164246 lr:8.47672596629258e-06 tokens_per_second_per_gpu:4541.581898961965 mfu:21.423416691178 +Step 29493 | loss:0.2131672203540802 lr:8.474310543008374e-06 tokens_per_second_per_gpu:5350.279536137567 mfu:25.23818142378007 +Step 29494 | loss:0.2713218033313751 lr:8.471895432044168e-06 tokens_per_second_per_gpu:5313.771178497103 mfu:25.065965271821877 +Step 29495 | loss:0.32533276081085205 lr:8.469480633418164e-06 tokens_per_second_per_gpu:4964.807482517684 mfu:23.419843978540914 +Step 29496 | loss:0.2658712863922119 lr:8.467066147148467e-06 tokens_per_second_per_gpu:4894.837420579995 mfu:23.08978325825725 +Step 29497 | loss:0.26661190390586853 lr:8.464651973253269e-06 tokens_per_second_per_gpu:5326.887314457271 mfu:25.127836322989538 +Step 29498 | loss:0.25674352049827576 lr:8.462238111750726e-06 tokens_per_second_per_gpu:5103.090154603364 mfu:24.072146936225806 +Step 29499 | loss:0.3371914327144623 lr:8.459824562658997e-06 tokens_per_second_per_gpu:5361.405875026099 mfu:25.290666262666438 +Step 29500 | loss:0.3854649066925049 lr:8.457411325996217e-06 tokens_per_second_per_gpu:4663.961571161621 mfu:22.000702485068945 +Step 29501 | loss:0.33187535405158997 lr:8.45499840178055e-06 tokens_per_second_per_gpu:4132.069871674755 mfu:19.491678588507614 +Step 29502 | loss:0.33811286091804504 lr:8.452585790030149e-06 tokens_per_second_per_gpu:5379.417833900869 mfu:25.3756317458358 +Step 29503 | loss:0.3407186269760132 lr:8.450173490763158e-06 tokens_per_second_per_gpu:5325.455570928431 mfu:25.121082544483368 +Step 29504 | loss:0.30351361632347107 lr:8.447761503997698e-06 tokens_per_second_per_gpu:5345.838569355248 mfu:25.2172326257608 +Step 29505 | loss:0.36124885082244873 lr:8.445349829751924e-06 tokens_per_second_per_gpu:4873.165454890408 mfu:22.987552898480835 +Step 29506 | loss:0.35584303736686707 lr:8.442938468043993e-06 tokens_per_second_per_gpu:5173.010969876273 mfu:24.401975351589353 +Step 29507 | loss:0.19070123136043549 lr:8.440527418892008e-06 tokens_per_second_per_gpu:4418.36941742114 mfu:20.842202393532837 +Step 29508 | loss:0.33328184485435486 lr:8.438116682314123e-06 tokens_per_second_per_gpu:4975.705888618087 mfu:23.47125361957634 +Step 29509 | loss:0.3891737461090088 lr:8.435706258328473e-06 tokens_per_second_per_gpu:4134.926906057887 mfu:19.505155707153186 +Step 29510 | loss:0.48271217942237854 lr:8.433296146953184e-06 tokens_per_second_per_gpu:5339.566926206707 mfu:25.18764821497647 +Step 29511 | loss:0.2815358638763428 lr:8.430886348206369e-06 tokens_per_second_per_gpu:4304.069143823849 mfu:20.303028501336797 +Step 29512 | loss:0.25487181544303894 lr:8.428476862106154e-06 tokens_per_second_per_gpu:3986.6911083572077 mfu:18.805902157764 +Step 29513 | loss:0.3425125479698181 lr:8.426067688670685e-06 tokens_per_second_per_gpu:5135.44483997142 mfu:24.224769507385485 +Step 29514 | loss:0.40436503291130066 lr:8.423658827918057e-06 tokens_per_second_per_gpu:3815.70783742712 mfu:17.999344896032557 +Step 29515 | loss:0.27390655875205994 lr:8.421250279866394e-06 tokens_per_second_per_gpu:3647.694549110882 mfu:17.20679754902193 +Step 29516 | loss:0.3432708978652954 lr:8.418842044533826e-06 tokens_per_second_per_gpu:4250.957561147951 mfu:20.052492104084163 +Step 29517 | loss:0.3597457706928253 lr:8.416434121938449e-06 tokens_per_second_per_gpu:4173.620774892415 mfu:19.687681288300436 +Step 29518 | loss:0.34170836210250854 lr:8.414026512098367e-06 tokens_per_second_per_gpu:4036.2429432705067 mfu:19.039646617464857 +Step 29519 | loss:0.2482362538576126 lr:8.411619215031697e-06 tokens_per_second_per_gpu:5350.271092761707 mfu:25.23814159493555 +Step 29520 | loss:0.31846287846565247 lr:8.409212230756564e-06 tokens_per_second_per_gpu:5308.128458461272 mfu:25.039347598665547 +Step 29521 | loss:0.3060506284236908 lr:8.406805559291031e-06 tokens_per_second_per_gpu:3985.8035800037997 mfu:18.801715535093024 +Step 29522 | loss:0.5380612015724182 lr:8.40439920065323e-06 tokens_per_second_per_gpu:5331.239194826962 mfu:25.148364885200728 +Step 29523 | loss:0.24414151906967163 lr:8.401993154861242e-06 tokens_per_second_per_gpu:5312.200915102387 mfu:25.058558071474433 +Step 29524 | loss:0.41641944646835327 lr:8.399587421933202e-06 tokens_per_second_per_gpu:4434.540594640014 mfu:20.91848459556161 +Step 29525 | loss:0.34916576743125916 lr:8.397182001887144e-06 tokens_per_second_per_gpu:4019.08525197145 mfu:18.958710860203208 +Step 29526 | loss:0.3112737536430359 lr:8.394776894741191e-06 tokens_per_second_per_gpu:5329.245237704724 mfu:25.138959049250964 +Step 29527 | loss:0.2849280834197998 lr:8.392372100513445e-06 tokens_per_second_per_gpu:5305.072694160809 mfu:25.02493303709311 +Step 29528 | loss:0.2982896566390991 lr:8.389967619221961e-06 tokens_per_second_per_gpu:4256.552331880059 mfu:20.078883592193833 +Step 29529 | loss:0.22382883727550507 lr:8.387563450884845e-06 tokens_per_second_per_gpu:5375.995061666585 mfu:25.359485945221525 +Step 29530 | loss:0.2531491816043854 lr:8.38515959552017e-06 tokens_per_second_per_gpu:4449.303847533523 mfu:20.988125378330974 +Step 29531 | loss:0.2947736084461212 lr:8.38275605314604e-06 tokens_per_second_per_gpu:3758.9847750817416 mfu:17.73177253299701 +Step 29532 | loss:0.26060810685157776 lr:8.380352823780496e-06 tokens_per_second_per_gpu:5347.211810925415 mfu:25.22371044054645 +Step 29533 | loss:0.31731924414634705 lr:8.377949907441623e-06 tokens_per_second_per_gpu:4450.822181408603 mfu:20.995287618274197 +Step 29534 | loss:0.25539806485176086 lr:8.375547304147502e-06 tokens_per_second_per_gpu:4873.33472516835 mfu:22.988351375262376 +Step 29535 | loss:0.31618642807006836 lr:8.373145013916217e-06 tokens_per_second_per_gpu:4156.0654760815405 mfu:19.604869948566705 +Step 29536 | loss:0.2828890383243561 lr:8.370743036765804e-06 tokens_per_second_per_gpu:5020.38293799844 mfu:23.682002884194116 +Step 29537 | loss:0.29707109928131104 lr:8.368341372714344e-06 tokens_per_second_per_gpu:4264.076226974243 mfu:20.114375089061973 +Step 29538 | loss:0.2691088020801544 lr:8.365940021779917e-06 tokens_per_second_per_gpu:5315.820483696638 mfu:25.0756321940956 +Step 29539 | loss:0.26555025577545166 lr:8.363538983980563e-06 tokens_per_second_per_gpu:5348.8277486027555 mfu:25.231333094277865 +Step 29540 | loss:0.407942533493042 lr:8.361138259334333e-06 tokens_per_second_per_gpu:4873.697150707928 mfu:22.990060998368737 +Step 29541 | loss:0.32192832231521606 lr:8.358737847859299e-06 tokens_per_second_per_gpu:4429.325122985733 mfu:20.893882325917144 +Step 29542 | loss:0.25444695353507996 lr:8.356337749573517e-06 tokens_per_second_per_gpu:5049.304864724723 mfu:23.81843254715153 +Step 29543 | loss:0.18365341424942017 lr:8.353937964495029e-06 tokens_per_second_per_gpu:5339.338690129893 mfu:25.186571586461046 +Step 29544 | loss:0.26973050832748413 lr:8.35153849264188e-06 tokens_per_second_per_gpu:5263.339896439845 mfu:24.828072309893805 +Step 29545 | loss:0.3567979335784912 lr:8.349139334032142e-06 tokens_per_second_per_gpu:3841.8456193562615 mfu:18.122641272957768 +Step 29546 | loss:0.33701038360595703 lr:8.346740488683841e-06 tokens_per_second_per_gpu:4936.596872604421 mfu:23.28676971835368 +Step 29547 | loss:0.29333263635635376 lr:8.344341956615004e-06 tokens_per_second_per_gpu:4703.8268815616575 mfu:22.188753955949245 +Step 29548 | loss:0.2775171101093292 lr:8.341943737843688e-06 tokens_per_second_per_gpu:4045.000843933966 mfu:19.08095912914633 +Step 29549 | loss:0.2894839942455292 lr:8.339545832387941e-06 tokens_per_second_per_gpu:4499.476474319722 mfu:21.224798219214318 +Step 29550 | loss:0.29293280839920044 lr:8.337148240265774e-06 tokens_per_second_per_gpu:4960.277224216207 mfu:23.39847397719108 +Step 29551 | loss:0.39333856105804443 lr:8.33475096149523e-06 tokens_per_second_per_gpu:5337.1215384972365 mfu:25.176112903924732 +Step 29552 | loss:0.25567981600761414 lr:8.332353996094356e-06 tokens_per_second_per_gpu:4544.249540593986 mfu:21.436000411902842 +Step 29553 | loss:0.3049933612346649 lr:8.329957344081164e-06 tokens_per_second_per_gpu:3462.4127644375794 mfu:16.33279175838573 +Step 29554 | loss:0.4049006998538971 lr:8.327561005473667e-06 tokens_per_second_per_gpu:4921.635545124448 mfu:23.21619458396496 +Step 29555 | loss:0.3299352526664734 lr:8.325164980289896e-06 tokens_per_second_per_gpu:5348.847808938525 mfu:25.231427722304336 +Step 29556 | loss:0.33336737751960754 lr:8.322769268547897e-06 tokens_per_second_per_gpu:5134.736762834338 mfu:24.22142938671945 +Step 29557 | loss:0.3320402204990387 lr:8.320373870265652e-06 tokens_per_second_per_gpu:5001.966561577691 mfu:23.595129694459615 +Step 29558 | loss:0.22228960692882538 lr:8.317978785461194e-06 tokens_per_second_per_gpu:4881.14995122381 mfu:23.02521712998104 +Step 29559 | loss:0.2720853090286255 lr:8.315584014152533e-06 tokens_per_second_per_gpu:5350.84473828623 mfu:25.240847578748735 +Step 29560 | loss:0.27170059084892273 lr:8.313189556357715e-06 tokens_per_second_per_gpu:4146.981624194682 mfu:19.56201986935144 +Step 29561 | loss:0.3263980448246002 lr:8.310795412094691e-06 tokens_per_second_per_gpu:4672.091071028209 mfu:22.03905072297535 +Step 29562 | loss:0.321152001619339 lr:8.308401581381491e-06 tokens_per_second_per_gpu:3079.212696508957 mfu:14.52517165729299 +Step 29563 | loss:0.2844071090221405 lr:8.306008064236138e-06 tokens_per_second_per_gpu:4304.976641337054 mfu:20.30730932194163 +Step 29564 | loss:0.33077171444892883 lr:8.303614860676601e-06 tokens_per_second_per_gpu:5188.732385698131 mfu:24.476135952371198 +Step 29565 | loss:0.28648072481155396 lr:8.301221970720901e-06 tokens_per_second_per_gpu:4037.5832607792086 mfu:19.045969123835153 +Step 29566 | loss:0.33999013900756836 lr:8.298829394387031e-06 tokens_per_second_per_gpu:4300.362778111095 mfu:20.28554494189854 +Step 29567 | loss:0.33520519733428955 lr:8.296437131693002e-06 tokens_per_second_per_gpu:4992.413729274638 mfu:23.550067354604856 +Step 29568 | loss:0.26628556847572327 lr:8.294045182656784e-06 tokens_per_second_per_gpu:4503.411315586724 mfu:21.243359536823917 +Step 29569 | loss:0.2511819899082184 lr:8.291653547296368e-06 tokens_per_second_per_gpu:5119.840436303919 mfu:24.151160872900082 +Step 29570 | loss:0.3497280776500702 lr:8.289262225629746e-06 tokens_per_second_per_gpu:4478.5327790457995 mfu:21.126003235244337 +Step 29571 | loss:0.4157479703426361 lr:8.286871217674914e-06 tokens_per_second_per_gpu:4128.226121710925 mfu:19.473546964116867 +Step 29572 | loss:0.2684064209461212 lr:8.28448052344984e-06 tokens_per_second_per_gpu:5224.039745493886 mfu:24.642686792584534 +Step 29573 | loss:0.33149832487106323 lr:8.282090142972505e-06 tokens_per_second_per_gpu:5239.689406518771 mfu:24.71650891374642 +Step 29574 | loss:0.2594568133354187 lr:8.27970007626091e-06 tokens_per_second_per_gpu:5342.176491598781 mfu:25.19995798016865 +Step 29575 | loss:0.24228359758853912 lr:8.27731032333301e-06 tokens_per_second_per_gpu:4223.337872104249 mfu:19.92220531846014 +Step 29576 | loss:0.33649224042892456 lr:8.274920884206771e-06 tokens_per_second_per_gpu:3464.735790869673 mfu:16.3437498704151 +Step 29577 | loss:0.31626075506210327 lr:8.272531758900181e-06 tokens_per_second_per_gpu:3923.5043398680923 mfu:18.507839390026458 +Step 29578 | loss:0.2713765799999237 lr:8.270142947431209e-06 tokens_per_second_per_gpu:5327.288241108902 mfu:25.129727562410423 +Step 29579 | loss:0.33246609568595886 lr:8.26775444981781e-06 tokens_per_second_per_gpu:4791.109555987828 mfu:22.600481223177358 +Step 29580 | loss:0.3170706629753113 lr:8.265366266077956e-06 tokens_per_second_per_gpu:4755.869899454587 mfu:22.434249750804668 +Step 29581 | loss:0.24778228998184204 lr:8.262978396229614e-06 tokens_per_second_per_gpu:4234.942434524328 mfu:19.97694601933817 +Step 29582 | loss:0.35261180996894836 lr:8.260590840290744e-06 tokens_per_second_per_gpu:5341.924612505382 mfu:25.198769823510087 +Step 29583 | loss:0.22577087581157684 lr:8.25820359827928e-06 tokens_per_second_per_gpu:5232.273429485609 mfu:24.68152648478484 +Step 29584 | loss:0.2714269161224365 lr:8.25581667021319e-06 tokens_per_second_per_gpu:3781.2488764043765 mfu:17.83679609758327 +Step 29585 | loss:0.26274818181991577 lr:8.25343005611045e-06 tokens_per_second_per_gpu:4926.937142784484 mfu:23.24120312467318 +Step 29586 | loss:0.510479211807251 lr:8.25104375598897e-06 tokens_per_second_per_gpu:4734.577546581758 mfu:22.333810089453525 +Step 29587 | loss:0.3073277771472931 lr:8.24865776986672e-06 tokens_per_second_per_gpu:4870.316758426477 mfu:22.974115111222734 +Step 29588 | loss:0.3213580250740051 lr:8.246272097761654e-06 tokens_per_second_per_gpu:3099.1514410766576 mfu:14.619226117319451 +Step 29589 | loss:0.3626325726509094 lr:8.243886739691703e-06 tokens_per_second_per_gpu:5367.52902530297 mfu:25.319550207239605 +Step 29590 | loss:0.3920072615146637 lr:8.241501695674797e-06 tokens_per_second_per_gpu:4844.980761234093 mfu:22.854600889699388 +Step 29591 | loss:0.3117099404335022 lr:8.239116965728883e-06 tokens_per_second_per_gpu:4273.882205118563 mfu:20.16063156104113 +Step 29592 | loss:0.43602094054222107 lr:8.236732549871912e-06 tokens_per_second_per_gpu:4960.490196962355 mfu:23.39947860597156 +Step 29593 | loss:0.324838787317276 lr:8.234348448121793e-06 tokens_per_second_per_gpu:4547.925292055778 mfu:21.453339558695983 +Step 29594 | loss:0.3335708677768707 lr:8.231964660496472e-06 tokens_per_second_per_gpu:3791.81045027143 mfu:17.886616843504477 +Step 29595 | loss:0.268372118473053 lr:8.22958118701388e-06 tokens_per_second_per_gpu:4854.166582561807 mfu:22.89793197616495 +Step 29596 | loss:0.25470641255378723 lr:8.227198027691946e-06 tokens_per_second_per_gpu:3418.0785852972813 mfu:16.123659871189993 +Step 29597 | loss:0.3128523826599121 lr:8.224815182548567e-06 tokens_per_second_per_gpu:4758.426464379397 mfu:22.446309503749948 +Step 29598 | loss:0.39135122299194336 lr:8.222432651601686e-06 tokens_per_second_per_gpu:3862.1318592751977 mfu:18.21833492784483 +Step 29599 | loss:0.2722063362598419 lr:8.22005043486923e-06 tokens_per_second_per_gpu:5297.523236030166 mfu:24.98932095502049 +Step 29600 | loss:0.3338503837585449 lr:8.217668532369094e-06 tokens_per_second_per_gpu:5320.8345200723 mfu:25.09928425916242 +Step 29601 | loss:0.31193530559539795 lr:8.215286944119205e-06 tokens_per_second_per_gpu:5319.860703719638 mfu:25.094690601276707 +Step 29602 | loss:0.34028905630111694 lr:8.212905670137478e-06 tokens_per_second_per_gpu:4676.402071233721 mfu:22.059386446477227 +Step 29603 | loss:0.3867502510547638 lr:8.21052471044183e-06 tokens_per_second_per_gpu:3846.171987622977 mfu:18.14304949022641 +Step 29604 | loss:0.3480486273765564 lr:8.208144065050156e-06 tokens_per_second_per_gpu:4707.386497818556 mfu:22.205545273166194 +Step 29605 | loss:0.3509884476661682 lr:8.205763733980348e-06 tokens_per_second_per_gpu:5356.384598527017 mfu:25.266980044702155 +Step 29606 | loss:0.2737504839897156 lr:8.203383717250329e-06 tokens_per_second_per_gpu:4143.86295191071 mfu:19.547308560087586 +Step 29607 | loss:0.42437076568603516 lr:8.201004014878006e-06 tokens_per_second_per_gpu:4161.1449258357625 mfu:19.62883057970086 +Step 29608 | loss:0.2648969888687134 lr:8.198624626881251e-06 tokens_per_second_per_gpu:5307.073690577642 mfu:25.034372078596505 +Step 29609 | loss:0.31991255283355713 lr:8.196245553277977e-06 tokens_per_second_per_gpu:3544.4359687710325 mfu:16.71970920782837 +Step 29610 | loss:0.22313863039016724 lr:8.193866794086087e-06 tokens_per_second_per_gpu:4218.719100371487 mfu:19.90041778415298 +Step 29611 | loss:0.373857706785202 lr:8.191488349323457e-06 tokens_per_second_per_gpu:4593.188404660686 mfu:21.666853383536683 +Step 29612 | loss:0.2931058704853058 lr:8.189110219007966e-06 tokens_per_second_per_gpu:3711.7780573892915 mfu:17.509090391345154 +Step 29613 | loss:0.2732105851173401 lr:8.18673240315751e-06 tokens_per_second_per_gpu:3953.4670234188998 mfu:18.649178480496584 +Step 29614 | loss:0.2420785129070282 lr:8.184354901789992e-06 tokens_per_second_per_gpu:5318.022347574222 mfu:25.086018761682883 +Step 29615 | loss:0.39536428451538086 lr:8.181977714923261e-06 tokens_per_second_per_gpu:4010.7753206542648 mfu:18.91951149635956 +Step 29616 | loss:0.26066145300865173 lr:8.179600842575213e-06 tokens_per_second_per_gpu:3747.930786633574 mfu:17.679628983482097 +Step 29617 | loss:0.3011479377746582 lr:8.177224284763735e-06 tokens_per_second_per_gpu:3541.710334859593 mfu:16.70685192762701 +Step 29618 | loss:0.41399627923965454 lr:8.17484804150669e-06 tokens_per_second_per_gpu:3446.063688886691 mfu:16.255670379572265 +Step 29619 | loss:0.24121509492397308 lr:8.172472112821938e-06 tokens_per_second_per_gpu:4138.686661596643 mfu:19.52289111550994 +Step 29620 | loss:0.3016064465045929 lr:8.170096498727359e-06 tokens_per_second_per_gpu:5343.438499429359 mfu:25.20591108642623 +Step 29621 | loss:0.2928798198699951 lr:8.167721199240835e-06 tokens_per_second_per_gpu:3152.2893390133354 mfu:14.869886648147032 +Step 29622 | loss:0.32868507504463196 lr:8.165346214380209e-06 tokens_per_second_per_gpu:4995.819481107185 mfu:23.566132867080167 +Step 29623 | loss:0.33779242634773254 lr:8.162971544163344e-06 tokens_per_second_per_gpu:5341.334450044203 mfu:25.195985926488603 +Step 29624 | loss:0.31062057614326477 lr:8.160597188608126e-06 tokens_per_second_per_gpu:5349.170034165543 mfu:25.232947713677472 +Step 29625 | loss:0.2718800902366638 lr:8.158223147732398e-06 tokens_per_second_per_gpu:3903.416778150861 mfu:18.41308293411484 +Step 29626 | loss:0.342558890581131 lr:8.155849421553996e-06 tokens_per_second_per_gpu:3927.082276200357 mfu:18.52471712616464 +Step 29627 | loss:0.30714598298072815 lr:8.153476010090788e-06 tokens_per_second_per_gpu:3823.744335209628 mfu:18.03725442724056 +Step 29628 | loss:0.37815406918525696 lr:8.151102913360643e-06 tokens_per_second_per_gpu:4387.949014250571 mfu:20.69870416152128 +Step 29629 | loss:0.30294325947761536 lr:8.148730131381382e-06 tokens_per_second_per_gpu:4972.610362013929 mfu:23.45665149243323 +Step 29630 | loss:0.3210551142692566 lr:8.146357664170856e-06 tokens_per_second_per_gpu:5056.865594913319 mfu:23.854097801444727 +Step 29631 | loss:0.23423080146312714 lr:8.143985511746931e-06 tokens_per_second_per_gpu:5315.509278182848 mfu:25.074164184589918 +Step 29632 | loss:0.243986114859581 lr:8.141613674127418e-06 tokens_per_second_per_gpu:4471.262783738377 mfu:21.091709426990118 +Step 29633 | loss:0.3000258207321167 lr:8.139242151330185e-06 tokens_per_second_per_gpu:3846.647919771581 mfu:18.145294543373772 +Step 29634 | loss:0.2565617859363556 lr:8.136870943373038e-06 tokens_per_second_per_gpu:5179.623979148287 mfu:24.43317004462117 +Step 29635 | loss:0.33340156078338623 lr:8.134500050273841e-06 tokens_per_second_per_gpu:4234.545340352197 mfu:19.975072858376972 +Step 29636 | loss:0.3377513885498047 lr:8.132129472050398e-06 tokens_per_second_per_gpu:4876.867732268481 mfu:23.005017172546864 +Step 29637 | loss:0.2580375075340271 lr:8.12975920872055e-06 tokens_per_second_per_gpu:2903.908695032939 mfu:13.698232772383747 +Step 29638 | loss:0.22928321361541748 lr:8.127389260302126e-06 tokens_per_second_per_gpu:4308.973081650008 mfu:20.32616121276086 +Step 29639 | loss:0.31979110836982727 lr:8.125019626812963e-06 tokens_per_second_per_gpu:3679.9345835725794 mfu:17.358879292294848 +Step 29640 | loss:0.31123870611190796 lr:8.122650308270874e-06 tokens_per_second_per_gpu:4878.153762761254 mfu:23.011083597800702 +Step 29641 | loss:0.3370966911315918 lr:8.12028130469366e-06 tokens_per_second_per_gpu:4057.5488190803553 mfu:19.140150068817174 +Step 29642 | loss:0.3551684617996216 lr:8.117912616099155e-06 tokens_per_second_per_gpu:5358.840095085776 mfu:25.278563040920734 +Step 29643 | loss:0.3086966574192047 lr:8.115544242505185e-06 tokens_per_second_per_gpu:4367.385097239132 mfu:20.60170065641229 +Step 29644 | loss:0.30034852027893066 lr:8.113176183929538e-06 tokens_per_second_per_gpu:5218.8603951560735 mfu:24.618254913352786 +Step 29645 | loss:0.4229353368282318 lr:8.110808440390038e-06 tokens_per_second_per_gpu:4185.213862484852 mfu:19.742367860458668 +Step 29646 | loss:0.3747817873954773 lr:8.108441011904505e-06 tokens_per_second_per_gpu:5201.354621667962 mfu:24.535677193016415 +Step 29647 | loss:0.37731024622917175 lr:8.106073898490735e-06 tokens_per_second_per_gpu:3515.004566035025 mfu:16.580876259607514 +Step 29648 | loss:0.2927875518798828 lr:8.103707100166514e-06 tokens_per_second_per_gpu:4538.895438329289 mfu:21.410744198016225 +Step 29649 | loss:0.49102938175201416 lr:8.10134061694966e-06 tokens_per_second_per_gpu:3767.482640212948 mfu:17.77185841270621 +Step 29650 | loss:0.3298346698284149 lr:8.098974448857982e-06 tokens_per_second_per_gpu:5130.363176199086 mfu:24.200798432350144 +Step 29651 | loss:0.3429912328720093 lr:8.096608595909249e-06 tokens_per_second_per_gpu:3708.5897619399166 mfu:17.494050657731155 +Step 29652 | loss:0.3554660975933075 lr:8.09424305812127e-06 tokens_per_second_per_gpu:4835.486536337372 mfu:22.809815010979754 +Step 29653 | loss:0.308986097574234 lr:8.091877835511852e-06 tokens_per_second_per_gpu:5298.206758891194 mfu:24.99254524897698 +Step 29654 | loss:0.3235279321670532 lr:8.089512928098763e-06 tokens_per_second_per_gpu:5349.005839556042 mfu:25.232173179689966 +Step 29655 | loss:0.3025410771369934 lr:8.087148335899786e-06 tokens_per_second_per_gpu:5320.670349774257 mfu:25.09850983985621 +Step 29656 | loss:0.3408159017562866 lr:8.084784058932716e-06 tokens_per_second_per_gpu:4961.813910251943 mfu:23.405722787407317 +Step 29657 | loss:0.2777824401855469 lr:8.082420097215342e-06 tokens_per_second_per_gpu:5370.680304559511 mfu:25.33441532915639 +Step 29658 | loss:0.26532256603240967 lr:8.080056450765427e-06 tokens_per_second_per_gpu:5128.916102434254 mfu:24.193972338505183 +Step 29659 | loss:0.3477218747138977 lr:8.077693119600754e-06 tokens_per_second_per_gpu:4169.911557183259 mfu:19.670184275508547 +Step 29660 | loss:0.21116973459720612 lr:8.075330103739114e-06 tokens_per_second_per_gpu:5343.749189645803 mfu:25.207376665186416 +Step 29661 | loss:0.34795692563056946 lr:8.07296740319825e-06 tokens_per_second_per_gpu:2983.2734500953493 mfu:14.072609863036657 +Step 29662 | loss:0.28576576709747314 lr:8.070605017995964e-06 tokens_per_second_per_gpu:4291.432176410143 mfu:20.243417769957375 +Step 29663 | loss:0.24407273530960083 lr:8.068242948149996e-06 tokens_per_second_per_gpu:3946.5892171172973 mfu:18.61673469469716 +Step 29664 | loss:0.35336124897003174 lr:8.065881193678132e-06 tokens_per_second_per_gpu:5368.747229454516 mfu:25.325296684069503 +Step 29665 | loss:0.3935423195362091 lr:8.06351975459812e-06 tokens_per_second_per_gpu:5080.214844913566 mfu:23.964240197488735 +Step 29666 | loss:0.3058697283267975 lr:8.061158630927724e-06 tokens_per_second_per_gpu:3884.6131262491995 mfu:18.324382899860865 +Step 29667 | loss:0.342865526676178 lr:8.058797822684716e-06 tokens_per_second_per_gpu:5076.181759088178 mfu:23.945215443534988 +Step 29668 | loss:0.34688350558280945 lr:8.05643732988683e-06 tokens_per_second_per_gpu:5319.311716794702 mfu:25.092100936284957 +Step 29669 | loss:0.29486924409866333 lr:8.054077152551842e-06 tokens_per_second_per_gpu:5270.866349500713 mfu:24.86357587312723 +Step 29670 | loss:0.3144569396972656 lr:8.05171729069748e-06 tokens_per_second_per_gpu:5164.796654469757 mfu:24.363227024310095 +Step 29671 | loss:0.3257862329483032 lr:8.049357744341517e-06 tokens_per_second_per_gpu:4069.4339520522085 mfu:19.196214268858192 +Step 29672 | loss:0.27217230200767517 lr:8.046998513501675e-06 tokens_per_second_per_gpu:4965.467237064816 mfu:23.42295615330547 +Step 29673 | loss:0.3465596139431 lr:8.044639598195714e-06 tokens_per_second_per_gpu:4149.293783244068 mfu:19.57292671808241 +Step 29674 | loss:0.25753429532051086 lr:8.04228099844137e-06 tokens_per_second_per_gpu:4121.506812613155 mfu:19.44185083667033 +Step 29675 | loss:0.28786101937294006 lr:8.039922714256393e-06 tokens_per_second_per_gpu:5334.849702558573 mfu:25.165396266186875 +Step 29676 | loss:0.2862852215766907 lr:8.037564745658515e-06 tokens_per_second_per_gpu:5355.3938308172765 mfu:25.262306424372905 +Step 29677 | loss:0.2870633602142334 lr:8.035207092665453e-06 tokens_per_second_per_gpu:5342.332247957737 mfu:25.200692709488624 +Step 29678 | loss:0.2503119707107544 lr:8.032849755294957e-06 tokens_per_second_per_gpu:3186.4668836660444 mfu:15.03110795756428 +Step 29679 | loss:0.2685931324958801 lr:8.030492733564765e-06 tokens_per_second_per_gpu:5325.2004511799205 mfu:25.119879101101724 +Step 29680 | loss:0.2709125876426697 lr:8.028136027492577e-06 tokens_per_second_per_gpu:3326.1009358553347 mfu:15.689785605767122 +Step 29681 | loss:0.3277100622653961 lr:8.025779637096137e-06 tokens_per_second_per_gpu:4794.5599608258835 mfu:22.616757371498238 +Step 29682 | loss:0.30195799469947815 lr:8.023423562393173e-06 tokens_per_second_per_gpu:4259.566148489423 mfu:20.093100279350338 +Step 29683 | loss:0.30866456031799316 lr:8.021067803401394e-06 tokens_per_second_per_gpu:5352.392380901707 mfu:25.248148073021003 +Step 29684 | loss:0.269485741853714 lr:8.018712360138514e-06 tokens_per_second_per_gpu:3672.951088110467 mfu:17.325936952692874 +Step 29685 | loss:0.35936224460601807 lr:8.016357232622257e-06 tokens_per_second_per_gpu:5370.888614405081 mfu:25.335397962239266 +Step 29686 | loss:0.24358487129211426 lr:8.014002420870343e-06 tokens_per_second_per_gpu:4254.195235958956 mfu:20.067764768571955 +Step 29687 | loss:0.2336881011724472 lr:8.011647924900462e-06 tokens_per_second_per_gpu:4021.152811299108 mfu:18.968463890313767 +Step 29688 | loss:0.3645527958869934 lr:8.009293744730335e-06 tokens_per_second_per_gpu:5171.600679041914 mfu:24.39532276910326 +Step 29689 | loss:0.35348471999168396 lr:8.006939880377678e-06 tokens_per_second_per_gpu:4055.4421102417805 mfu:19.130212363784025 +Step 29690 | loss:0.20310334861278534 lr:8.004586331860175e-06 tokens_per_second_per_gpu:4490.646850436415 mfu:21.18314737686758 +Step 29691 | loss:0.30198779702186584 lr:8.00223309919555e-06 tokens_per_second_per_gpu:5323.963643112161 mfu:25.114044866424887 +Step 29692 | loss:0.3773084580898285 lr:7.99988018240147e-06 tokens_per_second_per_gpu:4898.642622168017 mfu:23.107733043382286 +Step 29693 | loss:0.30060601234436035 lr:7.997527581495667e-06 tokens_per_second_per_gpu:4730.064747123337 mfu:22.312522444445168 +Step 29694 | loss:0.37703508138656616 lr:7.99517529649581e-06 tokens_per_second_per_gpu:4109.013482064269 mfu:19.38291766488924 +Step 29695 | loss:0.28801220655441284 lr:7.992823327419597e-06 tokens_per_second_per_gpu:3641.3030681814066 mfu:17.176647842978497 +Step 29696 | loss:0.21363142132759094 lr:7.990471674284728e-06 tokens_per_second_per_gpu:5141.788288271589 mfu:24.254692635321266 +Step 29697 | loss:0.24544471502304077 lr:7.988120337108873e-06 tokens_per_second_per_gpu:4333.301608372472 mfu:20.44092302418541 +Step 29698 | loss:0.32138141989707947 lr:7.985769315909741e-06 tokens_per_second_per_gpu:5244.113007917609 mfu:24.737375796288966 +Step 29699 | loss:0.1993347853422165 lr:7.983418610704985e-06 tokens_per_second_per_gpu:4606.433791607928 mfu:21.729334133662867 +Step 29700 | loss:0.3537680506706238 lr:7.981068221512305e-06 tokens_per_second_per_gpu:4363.93822520274 mfu:20.585441172919616 +Step 29701 | loss:0.2873966097831726 lr:7.978718148349369e-06 tokens_per_second_per_gpu:3032.573113137395 mfu:14.305164784995856 +Step 29702 | loss:0.3225141167640686 lr:7.976368391233852e-06 tokens_per_second_per_gpu:3492.2645397310075 mfu:16.4736076756829 +Step 29703 | loss:0.23879580199718475 lr:7.974018950183443e-06 tokens_per_second_per_gpu:3650.9519274491345 mfu:17.22216315840952 +Step 29704 | loss:0.41513338685035706 lr:7.971669825215788e-06 tokens_per_second_per_gpu:4176.016549501919 mfu:19.698982565894767 +Step 29705 | loss:0.398491233587265 lr:7.96932101634858e-06 tokens_per_second_per_gpu:5354.468684380793 mfu:25.257942350785754 +Step 29706 | loss:0.2004915177822113 lr:7.96697252359946e-06 tokens_per_second_per_gpu:3931.9804149215447 mfu:18.547822482221125 +Step 29707 | loss:0.3337090015411377 lr:7.96462434698611e-06 tokens_per_second_per_gpu:4490.093682159429 mfu:21.180537987724794 +Step 29708 | loss:0.20267821848392487 lr:7.962276486526176e-06 tokens_per_second_per_gpu:3670.981044095155 mfu:17.31664391894848 +Step 29709 | loss:0.2267833650112152 lr:7.959928942237328e-06 tokens_per_second_per_gpu:4147.024756922842 mfu:19.56222333378938 +Step 29710 | loss:0.40310969948768616 lr:7.957581714137213e-06 tokens_per_second_per_gpu:5169.338347888999 mfu:24.38465096706146 +Step 29711 | loss:0.3911669850349426 lr:7.955234802243505e-06 tokens_per_second_per_gpu:4577.338582723741 mfu:21.592087069201632 +Step 29712 | loss:0.3428226411342621 lr:7.952888206573838e-06 tokens_per_second_per_gpu:4187.176566151961 mfu:19.7516262685282 +Step 29713 | loss:0.2862571179866791 lr:7.95054192714586e-06 tokens_per_second_per_gpu:5310.103937824249 mfu:25.048666271872243 +Step 29714 | loss:0.28154975175857544 lr:7.948195963977212e-06 tokens_per_second_per_gpu:4711.071737940129 mfu:22.22292917958646 +Step 29715 | loss:0.3195323348045349 lr:7.945850317085569e-06 tokens_per_second_per_gpu:5027.349127135846 mfu:23.714863586909686 +Step 29716 | loss:0.3638269007205963 lr:7.943504986488532e-06 tokens_per_second_per_gpu:5290.645771946403 mfu:24.956878783521006 +Step 29717 | loss:0.3207586705684662 lr:7.941159972203765e-06 tokens_per_second_per_gpu:5352.213393497157 mfu:25.24730375889534 +Step 29718 | loss:0.35020503401756287 lr:7.938815274248913e-06 tokens_per_second_per_gpu:5100.023356209438 mfu:24.0576803249519 +Step 29719 | loss:0.2510297894477844 lr:7.936470892641596e-06 tokens_per_second_per_gpu:4622.108196987526 mfu:21.803272978167758 +Step 29720 | loss:0.33773621916770935 lr:7.934126827399435e-06 tokens_per_second_per_gpu:5375.538251828497 mfu:25.35733109527547 +Step 29721 | loss:0.34815001487731934 lr:7.931783078540077e-06 tokens_per_second_per_gpu:5339.7737387008065 mfu:25.18862378479699 +Step 29722 | loss:0.4285329580307007 lr:7.929439646081161e-06 tokens_per_second_per_gpu:3713.77444515103 mfu:17.51850769303615 +Step 29723 | loss:0.2789420783519745 lr:7.927096530040279e-06 tokens_per_second_per_gpu:5364.76554090325 mfu:25.306514379827064 +Step 29724 | loss:0.3395976424217224 lr:7.92475373043508e-06 tokens_per_second_per_gpu:2829.679871966282 mfu:13.348082749235164 +Step 29725 | loss:0.27253618836402893 lr:7.922411247283184e-06 tokens_per_second_per_gpu:5325.876363965006 mfu:25.123067497031585 +Step 29726 | loss:0.30569589138031006 lr:7.920069080602189e-06 tokens_per_second_per_gpu:4265.9574463572435 mfu:20.123249121860713 +Step 29727 | loss:0.39922410249710083 lr:7.917727230409739e-06 tokens_per_second_per_gpu:5378.785838965201 mfu:25.372650517894805 +Step 29728 | loss:0.2991762161254883 lr:7.915385696723421e-06 tokens_per_second_per_gpu:3986.5923860523594 mfu:18.805436467808267 +Step 29729 | loss:0.3319041430950165 lr:7.913044479560867e-06 tokens_per_second_per_gpu:4289.877314560132 mfu:20.236083221324083 +Step 29730 | loss:0.3271664083003998 lr:7.910703578939666e-06 tokens_per_second_per_gpu:5223.588907531065 mfu:24.64056011299302 +Step 29731 | loss:0.43008509278297424 lr:7.908362994877434e-06 tokens_per_second_per_gpu:4288.584524436735 mfu:20.22998490973928 +Step 29732 | loss:0.2957013249397278 lr:7.906022727391782e-06 tokens_per_second_per_gpu:5176.700978810166 mfu:24.419381753310976 +Step 29733 | loss:0.23929528892040253 lr:7.903682776500293e-06 tokens_per_second_per_gpu:4779.07030548831 mfu:22.54369002446318 +Step 29734 | loss:0.23139667510986328 lr:7.901343142220597e-06 tokens_per_second_per_gpu:4402.855819597026 mfu:20.769022105703765 +Step 29735 | loss:0.3339690566062927 lr:7.89900382457025e-06 tokens_per_second_per_gpu:4303.836991429974 mfu:20.301933398885605 +Step 29736 | loss:0.3440634310245514 lr:7.896664823566885e-06 tokens_per_second_per_gpu:4562.039548698388 mfu:21.519918915419677 +Step 29737 | loss:0.40591883659362793 lr:7.894326139228064e-06 tokens_per_second_per_gpu:5347.888992366988 mfu:25.226904820945414 +Step 29738 | loss:0.3219431936740875 lr:7.891987771571384e-06 tokens_per_second_per_gpu:3669.314825306671 mfu:17.30878408063696 +Step 29739 | loss:0.321061909198761 lr:7.889649720614456e-06 tokens_per_second_per_gpu:4811.179773030253 mfu:22.695155861298613 +Step 29740 | loss:0.3584268093109131 lr:7.887311986374829e-06 tokens_per_second_per_gpu:5305.5948897880435 mfu:25.027396322962534 +Step 29741 | loss:0.35153403878211975 lr:7.884974568870113e-06 tokens_per_second_per_gpu:3977.7061611943673 mfu:18.763518629006818 +Step 29742 | loss:0.25880035758018494 lr:7.88263746811787e-06 tokens_per_second_per_gpu:4665.302911554218 mfu:22.007029816557097 +Step 29743 | loss:0.2760041058063507 lr:7.880300684135694e-06 tokens_per_second_per_gpu:5311.223664130431 mfu:25.05394821190351 +Step 29744 | loss:0.3796895444393158 lr:7.877964216941136e-06 tokens_per_second_per_gpu:4603.344226780701 mfu:21.714760129238726 +Step 29745 | loss:0.3687070608139038 lr:7.875628066551788e-06 tokens_per_second_per_gpu:5149.450455507771 mfu:24.290836385474872 +Step 29746 | loss:0.3841875195503235 lr:7.873292232985219e-06 tokens_per_second_per_gpu:4939.989878352227 mfu:23.302775105372554 +Step 29747 | loss:0.25390204787254333 lr:7.870956716259003e-06 tokens_per_second_per_gpu:4615.2999312649135 mfu:21.771157227144283 +Step 29748 | loss:0.3648999035358429 lr:7.868621516390696e-06 tokens_per_second_per_gpu:4478.171782903226 mfu:21.124300354847524 +Step 29749 | loss:0.32064342498779297 lr:7.866286633397851e-06 tokens_per_second_per_gpu:4869.8814264505945 mfu:22.972061576017204 +Step 29750 | loss:0.3291694223880768 lr:7.863952067298042e-06 tokens_per_second_per_gpu:4827.8063693516715 mfu:22.773586352936594 +Step 29751 | loss:0.23435313999652863 lr:7.861617818108835e-06 tokens_per_second_per_gpu:4430.819503132069 mfu:20.900931572034956 +Step 29752 | loss:0.34557783603668213 lr:7.859283885847767e-06 tokens_per_second_per_gpu:4197.234630070747 mfu:19.799071872115768 +Step 29753 | loss:0.32169875502586365 lr:7.856950270532409e-06 tokens_per_second_per_gpu:5309.1692207742735 mfu:25.044257052068335 +Step 29754 | loss:0.3224465548992157 lr:7.854616972180307e-06 tokens_per_second_per_gpu:4292.251540438016 mfu:20.24728284987477 +Step 29755 | loss:0.21561820805072784 lr:7.852283990809e-06 tokens_per_second_per_gpu:4767.963110036267 mfu:22.49129548843294 +Step 29756 | loss:0.3879567086696625 lr:7.849951326436061e-06 tokens_per_second_per_gpu:5103.6118069214135 mfu:24.074607659213108 +Step 29757 | loss:0.2598443627357483 lr:7.847618979078997e-06 tokens_per_second_per_gpu:3911.6366153315444 mfu:18.451857308544962 +Step 29758 | loss:0.32544586062431335 lr:7.845286948755386e-06 tokens_per_second_per_gpu:5343.766009308343 mfu:25.207456006404342 +Step 29759 | loss:0.3373432755470276 lr:7.842955235482735e-06 tokens_per_second_per_gpu:3642.888884304438 mfu:17.184128408198966 +Step 29760 | loss:0.3165495991706848 lr:7.8406238392786e-06 tokens_per_second_per_gpu:5148.978663898408 mfu:24.288610863957558 +Step 29761 | loss:0.27105700969696045 lr:7.838292760160525e-06 tokens_per_second_per_gpu:5347.609781914472 mfu:25.225587737602684 +Step 29762 | loss:0.29219669103622437 lr:7.835961998146013e-06 tokens_per_second_per_gpu:5321.117564716657 mfu:25.100619429042247 +Step 29763 | loss:0.32986217737197876 lr:7.83363155325263e-06 tokens_per_second_per_gpu:5091.467077802634 mfu:24.01731889985537 +Step 29764 | loss:0.41467347741127014 lr:7.831301425497867e-06 tokens_per_second_per_gpu:3790.0850320562568 mfu:17.878477751396805 +Step 29765 | loss:0.35829219222068787 lr:7.828971614899284e-06 tokens_per_second_per_gpu:5390.083122210598 mfu:25.425941730478623 +Step 29766 | loss:0.31648141145706177 lr:7.82664212147437e-06 tokens_per_second_per_gpu:4705.760887321718 mfu:22.19787699109486 +Step 29767 | loss:0.36962834000587463 lr:7.824312945240665e-06 tokens_per_second_per_gpu:4762.082546325182 mfu:22.463555866080124 +Step 29768 | loss:0.21922877430915833 lr:7.82198408621569e-06 tokens_per_second_per_gpu:4100.873639296465 mfu:19.344520637753128 +Step 29769 | loss:0.31900158524513245 lr:7.819655544416948e-06 tokens_per_second_per_gpu:4094.2493042779556 mfu:19.31327251924272 +Step 29770 | loss:0.3368326425552368 lr:7.817327319861973e-06 tokens_per_second_per_gpu:4852.998044113053 mfu:22.892419780928957 +Step 29771 | loss:0.3958856165409088 lr:7.814999412568242e-06 tokens_per_second_per_gpu:4615.517972207659 mfu:21.772185763472834 +Step 29772 | loss:0.21189327538013458 lr:7.812671822553302e-06 tokens_per_second_per_gpu:5351.784403970914 mfu:25.245280142106893 +Step 29773 | loss:0.3237953782081604 lr:7.810344549834625e-06 tokens_per_second_per_gpu:4319.203147556222 mfu:20.374418179069696 +Step 29774 | loss:0.4573802351951599 lr:7.808017594429733e-06 tokens_per_second_per_gpu:3907.284618993228 mfu:18.431328199289982 +Step 29775 | loss:0.41377294063568115 lr:7.805690956356132e-06 tokens_per_second_per_gpu:4039.6115961344553 mfu:19.05553712777555 +Step 29776 | loss:0.3074265420436859 lr:7.803364635631305e-06 tokens_per_second_per_gpu:3686.283439926881 mfu:17.388827930944682 +Step 29777 | loss:0.3894544243812561 lr:7.801038632272767e-06 tokens_per_second_per_gpu:4114.063846883026 mfu:19.40674109742041 +Step 29778 | loss:0.24022142589092255 lr:7.798712946297992e-06 tokens_per_second_per_gpu:5262.376195374663 mfu:24.823526367544986 +Step 29779 | loss:0.28954702615737915 lr:7.796387577724495e-06 tokens_per_second_per_gpu:5316.263168403423 mfu:25.077720413387304 +Step 29780 | loss:0.22686904668807983 lr:7.794062526569734e-06 tokens_per_second_per_gpu:4817.908197840408 mfu:22.726894989115767 +Step 29781 | loss:0.3104460835456848 lr:7.791737792851223e-06 tokens_per_second_per_gpu:4072.2177465357727 mfu:19.20934590240148 +Step 29782 | loss:0.22465728223323822 lr:7.78941337658643e-06 tokens_per_second_per_gpu:5342.241719653748 mfu:25.200265672032987 +Step 29783 | loss:0.277762234210968 lr:7.787089277792865e-06 tokens_per_second_per_gpu:4365.409112972272 mfu:20.59237960148783 +Step 29784 | loss:0.29280760884284973 lr:7.784765496487967e-06 tokens_per_second_per_gpu:4774.728148114056 mfu:22.52320732728039 +Step 29785 | loss:0.33449214696884155 lr:7.782442032689253e-06 tokens_per_second_per_gpu:5092.868540153 mfu:24.02392983687618 +Step 29786 | loss:0.3038005232810974 lr:7.780118886414178e-06 tokens_per_second_per_gpu:5047.169862447327 mfu:23.808361377140105 +Step 29787 | loss:0.2574288249015808 lr:7.777796057680204e-06 tokens_per_second_per_gpu:5318.9445147865135 mfu:25.09036878176121 +Step 29788 | loss:0.29457399249076843 lr:7.775473546504809e-06 tokens_per_second_per_gpu:5014.922847730783 mfu:23.65624670681412 +Step 29789 | loss:0.303989440202713 lr:7.773151352905472e-06 tokens_per_second_per_gpu:3410.4697933713414 mfu:16.087767901481566 +Step 29790 | loss:0.25858891010284424 lr:7.77082947689966e-06 tokens_per_second_per_gpu:5364.734345952685 mfu:25.306367227922845 +Step 29791 | loss:0.292330265045166 lr:7.768507918504814e-06 tokens_per_second_per_gpu:5286.5727256641685 mfu:24.93766552927435 +Step 29792 | loss:0.22386206686496735 lr:7.766186677738424e-06 tokens_per_second_per_gpu:5346.662874876473 mfu:25.22112101554545 +Step 29793 | loss:0.23974496126174927 lr:7.763865754617922e-06 tokens_per_second_per_gpu:5354.437431425549 mfu:25.257794925263568 +Step 29794 | loss:0.23089827597141266 lr:7.761545149160788e-06 tokens_per_second_per_gpu:4000.0679755786114 mfu:18.869003122776256 +Step 29795 | loss:0.3185487389564514 lr:7.759224861384446e-06 tokens_per_second_per_gpu:4905.94791266468 mfu:23.142193345066133 +Step 29796 | loss:0.22641848027706146 lr:7.756904891306366e-06 tokens_per_second_per_gpu:4026.375363953921 mfu:18.993099562245884 +Step 29797 | loss:0.28679344058036804 lr:7.754585238944006e-06 tokens_per_second_per_gpu:4163.203793408846 mfu:19.638542609322126 +Step 29798 | loss:0.31944170594215393 lr:7.752265904314793e-06 tokens_per_second_per_gpu:4240.431834496517 mfu:20.002840455595194 +Step 29799 | loss:0.335446298122406 lr:7.749946887436183e-06 tokens_per_second_per_gpu:3860.9795115231614 mfu:18.212899106887672 +Step 29800 | loss:0.31918656826019287 lr:7.747628188325606e-06 tokens_per_second_per_gpu:5163.289457873748 mfu:24.356117320812466 +Step 29801 | loss:0.3189610540866852 lr:7.745309807000523e-06 tokens_per_second_per_gpu:4331.305529160853 mfu:20.43150717797797 +Step 29802 | loss:0.2481139600276947 lr:7.742991743478345e-06 tokens_per_second_per_gpu:4606.358653597534 mfu:21.728979694847975 +Step 29803 | loss:0.2760544717311859 lr:7.740673997776516e-06 tokens_per_second_per_gpu:4946.9209272633625 mfu:23.335470045645266 +Step 29804 | loss:0.44432902336120605 lr:7.738356569912481e-06 tokens_per_second_per_gpu:5316.476489128508 mfu:25.078726683643797 +Step 29805 | loss:0.40338921546936035 lr:7.736039459903649e-06 tokens_per_second_per_gpu:4689.789800678246 mfu:22.122538650449233 +Step 29806 | loss:0.3477143347263336 lr:7.733722667767467e-06 tokens_per_second_per_gpu:4763.940581588437 mfu:22.472320535430402 +Step 29807 | loss:0.29239049553871155 lr:7.73140619352134e-06 tokens_per_second_per_gpu:5346.131618807103 mfu:25.218614990024594 +Step 29808 | loss:0.32919082045555115 lr:7.729090037182706e-06 tokens_per_second_per_gpu:4990.413509286782 mfu:23.54063197564942 +Step 29809 | loss:0.263614296913147 lr:7.726774198768977e-06 tokens_per_second_per_gpu:4038.37220348277 mfu:19.049690701175358 +Step 29810 | loss:0.1791457235813141 lr:7.72445867829757e-06 tokens_per_second_per_gpu:3954.860856443276 mfu:18.655753428685003 +Step 29811 | loss:0.24723272025585175 lr:7.722143475785914e-06 tokens_per_second_per_gpu:5344.506166307641 mfu:25.21094745325384 +Step 29812 | loss:0.31812337040901184 lr:7.7198285912514e-06 tokens_per_second_per_gpu:3511.410777407441 mfu:16.56392374548763 +Step 29813 | loss:0.3253580331802368 lr:7.717514024711465e-06 tokens_per_second_per_gpu:4233.697893752583 mfu:19.971075308177294 +Step 29814 | loss:0.2100478857755661 lr:7.715199776183485e-06 tokens_per_second_per_gpu:5357.8106597944825 mfu:25.27370701901193 +Step 29815 | loss:0.29422831535339355 lr:7.712885845684904e-06 tokens_per_second_per_gpu:5383.864743614955 mfu:25.396608577678183 +Step 29816 | loss:0.32505789399147034 lr:7.710572233233082e-06 tokens_per_second_per_gpu:4500.531352419677 mfu:21.229774259192567 +Step 29817 | loss:0.3300223648548126 lr:7.708258938845448e-06 tokens_per_second_per_gpu:3856.4716645416925 mfu:18.19163482355808 +Step 29818 | loss:0.2715320289134979 lr:7.705945962539407e-06 tokens_per_second_per_gpu:3828.6173715015425 mfu:18.060241370855245 +Step 29819 | loss:0.29798591136932373 lr:7.70363330433233e-06 tokens_per_second_per_gpu:4351.025847170117 mfu:20.52453128266087 +Step 29820 | loss:0.3827536702156067 lr:7.701320964241626e-06 tokens_per_second_per_gpu:5366.630785446009 mfu:25.315313056578795 +Step 29821 | loss:0.3833167850971222 lr:7.699008942284691e-06 tokens_per_second_per_gpu:4159.296251399657 mfu:19.62011006696937 +Step 29822 | loss:0.31444868445396423 lr:7.696697238478911e-06 tokens_per_second_per_gpu:5350.390672351713 mfu:25.238705672263983 +Step 29823 | loss:0.27571648359298706 lr:7.694385852841656e-06 tokens_per_second_per_gpu:4702.26974508993 mfu:22.181408677537235 +Step 29824 | loss:0.35045650601387024 lr:7.69207478539033e-06 tokens_per_second_per_gpu:4640.070539290749 mfu:21.88800441150368 +Step 29825 | loss:0.285525381565094 lr:7.6897640361423e-06 tokens_per_second_per_gpu:5332.844621856414 mfu:25.15593795841275 +Step 29826 | loss:0.343227356672287 lr:7.68745360511497e-06 tokens_per_second_per_gpu:4623.716796858389 mfu:21.81086101821406 +Step 29827 | loss:0.3438858389854431 lr:7.685143492325686e-06 tokens_per_second_per_gpu:4331.994006689838 mfu:20.43475484394857 +Step 29828 | loss:0.21748627722263336 lr:7.682833697791846e-06 tokens_per_second_per_gpu:4616.833915426356 mfu:21.77839329215849 +Step 29829 | loss:0.37183892726898193 lr:7.680524221530804e-06 tokens_per_second_per_gpu:4000.587691418263 mfu:18.87145471106456 +Step 29830 | loss:0.26941362023353577 lr:7.678215063559956e-06 tokens_per_second_per_gpu:5329.929297446823 mfu:25.14218587576707 +Step 29831 | loss:0.3290143311023712 lr:7.67590622389664e-06 tokens_per_second_per_gpu:5117.478531686733 mfu:24.14001936583833 +Step 29832 | loss:0.3365420699119568 lr:7.673597702558228e-06 tokens_per_second_per_gpu:4329.005680311762 mfu:20.420658398562132 +Step 29833 | loss:0.3651776611804962 lr:7.671289499562101e-06 tokens_per_second_per_gpu:4149.953311453477 mfu:19.576037825173305 +Step 29834 | loss:0.217306450009346 lr:7.668981614925601e-06 tokens_per_second_per_gpu:4771.925005349018 mfu:22.50998442459223 +Step 29835 | loss:0.216311976313591 lr:7.6666740486661e-06 tokens_per_second_per_gpu:5367.546795825384 mfu:25.319634033825793 +Step 29836 | loss:0.3048853278160095 lr:7.664366800800931e-06 tokens_per_second_per_gpu:5299.638991142359 mfu:24.999301332869752 +Step 29837 | loss:0.30757200717926025 lr:7.662059871347477e-06 tokens_per_second_per_gpu:3809.0883325801983 mfu:17.968119562265638 +Step 29838 | loss:0.39594465494155884 lr:7.659753260323061e-06 tokens_per_second_per_gpu:5281.655739560604 mfu:24.914471266899504 +Step 29839 | loss:0.35172945261001587 lr:7.65744696774504e-06 tokens_per_second_per_gpu:5067.590660607993 mfu:23.904689766211405 +Step 29840 | loss:0.3736772835254669 lr:7.655140993630778e-06 tokens_per_second_per_gpu:5082.698658784482 mfu:23.975956771300197 +Step 29841 | loss:0.3027730882167816 lr:7.652835337997594e-06 tokens_per_second_per_gpu:5095.525498642809 mfu:24.036463163396473 +Step 29842 | loss:0.2608611285686493 lr:7.650530000862848e-06 tokens_per_second_per_gpu:5116.571850748366 mfu:24.135742397156335 +Step 29843 | loss:0.21682685613632202 lr:7.648224982243857e-06 tokens_per_second_per_gpu:3772.804087983542 mfu:17.796960589772716 +Step 29844 | loss:0.2337779551744461 lr:7.645920282157981e-06 tokens_per_second_per_gpu:3308.0491170495025 mfu:15.60463209650213 +Step 29845 | loss:0.3564469516277313 lr:7.643615900622537e-06 tokens_per_second_per_gpu:5377.119094750716 mfu:25.36478820105919 +Step 29846 | loss:0.31173911690711975 lr:7.641311837654857e-06 tokens_per_second_per_gpu:4740.8111458224075 mfu:22.36321504063337 +Step 29847 | loss:0.3420734107494354 lr:7.639008093272288e-06 tokens_per_second_per_gpu:5309.478396686275 mfu:25.045715487596098 +Step 29848 | loss:0.25598472356796265 lr:7.636704667492134e-06 tokens_per_second_per_gpu:5356.393449361827 mfu:25.267021795600062 +Step 29849 | loss:0.262754887342453 lr:7.634401560331733e-06 tokens_per_second_per_gpu:5329.893263067832 mfu:25.142015895453273 +Step 29850 | loss:0.24143557250499725 lr:7.63209877180841e-06 tokens_per_second_per_gpu:5358.5741279699405 mfu:25.277308428656667 +Step 29851 | loss:0.2753405272960663 lr:7.629796301939478e-06 tokens_per_second_per_gpu:4416.40472596293 mfu:20.832934608713533 +Step 29852 | loss:0.3084166646003723 lr:7.627494150742248e-06 tokens_per_second_per_gpu:3962.949977348826 mfu:18.693911192142824 +Step 29853 | loss:0.2548503279685974 lr:7.625192318234037e-06 tokens_per_second_per_gpu:4582.658282483588 mfu:21.617181000603207 +Step 29854 | loss:0.305999755859375 lr:7.622890804432175e-06 tokens_per_second_per_gpu:5327.819203484273 mfu:25.132232202526573 +Step 29855 | loss:0.3776099979877472 lr:7.620589609353946e-06 tokens_per_second_per_gpu:5322.162451644954 mfu:25.105548338959114 +Step 29856 | loss:0.401436984539032 lr:7.6182887330166705e-06 tokens_per_second_per_gpu:3680.35689649242 mfu:17.360871414391752 +Step 29857 | loss:0.3302103281021118 lr:7.615988175437666e-06 tokens_per_second_per_gpu:3712.488542597624 mfu:17.512441871294133 +Step 29858 | loss:0.30265432596206665 lr:7.613687936634223e-06 tokens_per_second_per_gpu:5398.317620327059 mfu:25.464785263044217 +Step 29859 | loss:0.3017203211784363 lr:7.611388016623627e-06 tokens_per_second_per_gpu:5333.497824085204 mfu:25.159019224773743 +Step 29860 | loss:0.370114803314209 lr:7.6090884154231935e-06 tokens_per_second_per_gpu:4309.726736280506 mfu:20.329716330239656 +Step 29861 | loss:0.2976568341255188 lr:7.606789133050218e-06 tokens_per_second_per_gpu:3458.4234649363957 mfu:16.31397355199397 +Step 29862 | loss:0.36535802483558655 lr:7.604490169521999e-06 tokens_per_second_per_gpu:4007.467209217915 mfu:18.903906570292285 +Step 29863 | loss:0.3009387254714966 lr:7.6021915248558065e-06 tokens_per_second_per_gpu:4326.126683998044 mfu:20.4070776817429 +Step 29864 | loss:0.28090792894363403 lr:7.599893199068953e-06 tokens_per_second_per_gpu:4108.562091969375 mfu:19.380788380796538 +Step 29865 | loss:0.3522820472717285 lr:7.597595192178702e-06 tokens_per_second_per_gpu:4493.274357911356 mfu:21.195541777925875 +Step 29866 | loss:0.30096685886383057 lr:7.595297504202359e-06 tokens_per_second_per_gpu:4065.5439154162173 mfu:19.177864302337532 +Step 29867 | loss:0.3166663646697998 lr:7.593000135157186e-06 tokens_per_second_per_gpu:5075.968566960019 mfu:23.944209779892134 +Step 29868 | loss:0.38023850321769714 lr:7.590703085060469e-06 tokens_per_second_per_gpu:3375.9578481205863 mfu:15.924969167388706 +Step 29869 | loss:0.4285055696964264 lr:7.588406353929495e-06 tokens_per_second_per_gpu:4185.2335368525655 mfu:19.742460667807965 +Step 29870 | loss:0.4631038308143616 lr:7.586109941781516e-06 tokens_per_second_per_gpu:4862.542141939417 mfu:22.93744091876692 +Step 29871 | loss:0.32140567898750305 lr:7.583813848633831e-06 tokens_per_second_per_gpu:4635.7021799409085 mfu:21.86739811513152 +Step 29872 | loss:0.4235408902168274 lr:7.58151807450368e-06 tokens_per_second_per_gpu:4522.514154349896 mfu:21.33347088655017 +Step 29873 | loss:0.2999485433101654 lr:7.579222619408361e-06 tokens_per_second_per_gpu:5319.15194049198 mfu:25.09134724420437 +Step 29874 | loss:0.21670828759670258 lr:7.5769274833651035e-06 tokens_per_second_per_gpu:3999.6235977913625 mfu:18.866906916935136 +Step 29875 | loss:0.37437474727630615 lr:7.57463266639119e-06 tokens_per_second_per_gpu:4208.610583517723 mfu:19.85273418546296 +Step 29876 | loss:0.3283368945121765 lr:7.57233816850389e-06 tokens_per_second_per_gpu:5354.259491935076 mfu:25.256955554327398 +Step 29877 | loss:0.39655518531799316 lr:7.570043989720438e-06 tokens_per_second_per_gpu:5156.186520682567 mfu:24.32261155419561 +Step 29878 | loss:0.36319398880004883 lr:7.567750130058099e-06 tokens_per_second_per_gpu:4476.366024118918 mfu:21.115782282567725 +Step 29879 | loss:0.22223487496376038 lr:7.565456589534137e-06 tokens_per_second_per_gpu:3866.1106658488065 mfu:18.237103637306944 +Step 29880 | loss:0.25936388969421387 lr:7.563163368165788e-06 tokens_per_second_per_gpu:5363.3153790016195 mfu:25.29967371871394 +Step 29881 | loss:0.31747928261756897 lr:7.560870465970293e-06 tokens_per_second_per_gpu:4971.085237468829 mfu:23.44945721974091 +Step 29882 | loss:0.2925280034542084 lr:7.558577882964907e-06 tokens_per_second_per_gpu:5346.8600887121265 mfu:25.222051306856372 +Step 29883 | loss:0.4381319582462311 lr:7.55628561916688e-06 tokens_per_second_per_gpu:4972.739760147419 mfu:23.457261885506494 +Step 29884 | loss:0.271282821893692 lr:7.55399367459344e-06 tokens_per_second_per_gpu:4618.371840399454 mfu:21.785647946653658 +Step 29885 | loss:0.3081966042518616 lr:7.5517020492618214e-06 tokens_per_second_per_gpu:4841.980883667712 mfu:22.840449955386884 +Step 29886 | loss:0.3502919673919678 lr:7.549410743189284e-06 tokens_per_second_per_gpu:4809.574679106965 mfu:22.6875843594052 +Step 29887 | loss:0.2681722342967987 lr:7.5471197563930455e-06 tokens_per_second_per_gpu:5295.955608031805 mfu:24.981926186287303 +Step 29888 | loss:0.3572557866573334 lr:7.544829088890326e-06 tokens_per_second_per_gpu:4604.874553089412 mfu:21.721978939537507 +Step 29889 | loss:0.30697423219680786 lr:7.542538740698363e-06 tokens_per_second_per_gpu:4222.755606592567 mfu:19.9194586726968 +Step 29890 | loss:0.31301647424697876 lr:7.5402487118343965e-06 tokens_per_second_per_gpu:5376.210196554837 mfu:25.360500773195206 +Step 29891 | loss:0.2732268273830414 lr:7.5379590023156255e-06 tokens_per_second_per_gpu:4991.126967243243 mfu:23.543997478558524 +Step 29892 | loss:0.35259518027305603 lr:7.5356696121592795e-06 tokens_per_second_per_gpu:4259.11246071701 mfu:20.090960156720758 +Step 29893 | loss:0.31506723165512085 lr:7.533380541382596e-06 tokens_per_second_per_gpu:5344.805212683571 mfu:25.212358106031672 +Step 29894 | loss:0.3673620820045471 lr:7.531091790002776e-06 tokens_per_second_per_gpu:5241.285751788892 mfu:24.724039146750172 +Step 29895 | loss:0.3168156147003174 lr:7.528803358037017e-06 tokens_per_second_per_gpu:3939.2349804229334 mfu:18.58204350544768 +Step 29896 | loss:0.2634386122226715 lr:7.526515245502552e-06 tokens_per_second_per_gpu:3700.8192544587596 mfu:17.45739584815725 +Step 29897 | loss:0.285499632358551 lr:7.524227452416588e-06 tokens_per_second_per_gpu:4276.517328511015 mfu:20.17306188768173 +Step 29898 | loss:0.3968689441680908 lr:7.521939978796339e-06 tokens_per_second_per_gpu:4285.514813893446 mfu:20.215504561359964 +Step 29899 | loss:0.3118891716003418 lr:7.519652824658985e-06 tokens_per_second_per_gpu:3593.1419388180775 mfu:16.94946357863568 +Step 29900 | loss:0.28657734394073486 lr:7.517365990021758e-06 tokens_per_second_per_gpu:3879.711451951011 mfu:18.301260865884437 +Step 29901 | loss:0.39244967699050903 lr:7.515079474901826e-06 tokens_per_second_per_gpu:3993.064833242914 mfu:18.835968105519477 +Step 29902 | loss:0.31237420439720154 lr:7.512793279316416e-06 tokens_per_second_per_gpu:5349.521674809561 mfu:25.234606462591387 +Step 29903 | loss:0.36019304394721985 lr:7.510507403282696e-06 tokens_per_second_per_gpu:4587.562049610456 mfu:21.64031290680959 +Step 29904 | loss:0.364702433347702 lr:7.508221846817876e-06 tokens_per_second_per_gpu:5343.327895051016 mfu:25.205389346702532 +Step 29905 | loss:0.3660100996494293 lr:7.505936609939152e-06 tokens_per_second_per_gpu:4683.081648293737 mfu:22.09089515112269 +Step 29906 | loss:0.36787149310112 lr:7.503651692663688e-06 tokens_per_second_per_gpu:5362.258954794539 mfu:25.294690385484586 +Step 29907 | loss:0.26596248149871826 lr:7.5013670950086826e-06 tokens_per_second_per_gpu:4090.8686767431004 mfu:19.297325522366272 +Step 29908 | loss:0.2894817888736725 lr:7.499082816991326e-06 tokens_per_second_per_gpu:4578.04475583653 mfu:21.595418208260224 +Step 29909 | loss:0.4354874789714813 lr:7.496798858628796e-06 tokens_per_second_per_gpu:4637.961428821367 mfu:21.878055377568256 +Step 29910 | loss:0.3663230538368225 lr:7.494515219938253e-06 tokens_per_second_per_gpu:4783.906597882663 mfu:22.56650364084392 +Step 29911 | loss:0.34261372685432434 lr:7.4922319009368855e-06 tokens_per_second_per_gpu:3769.8773056659134 mfu:17.783154458219904 +Step 29912 | loss:0.2708780765533447 lr:7.489948901641874e-06 tokens_per_second_per_gpu:5205.329836810174 mfu:24.55442896108387 +Step 29913 | loss:0.34267786145210266 lr:7.487666222070372e-06 tokens_per_second_per_gpu:3941.6295691978366 mfu:18.593339189257872 +Step 29914 | loss:0.30368727445602417 lr:7.485383862239553e-06 tokens_per_second_per_gpu:4536.509239412289 mfu:21.399488090597217 +Step 29915 | loss:0.4413509666919708 lr:7.483101822166605e-06 tokens_per_second_per_gpu:4557.23195088484 mfu:21.497240656272403 +Step 29916 | loss:0.3892621099948883 lr:7.480820101868668e-06 tokens_per_second_per_gpu:5347.481490233875 mfu:25.22498256385649 +Step 29917 | loss:0.21272552013397217 lr:7.478538701362897e-06 tokens_per_second_per_gpu:4806.627712225742 mfu:22.67368301382207 +Step 29918 | loss:0.3330360949039459 lr:7.476257620666466e-06 tokens_per_second_per_gpu:4276.566198216159 mfu:20.173292414418928 +Step 29919 | loss:0.23358415067195892 lr:7.473976859796539e-06 tokens_per_second_per_gpu:5359.332610396499 mfu:25.280886319673193 +Step 29920 | loss:0.3819673955440521 lr:7.471696418770246e-06 tokens_per_second_per_gpu:3956.901139966735 mfu:18.665377794173423 +Step 29921 | loss:0.30009880661964417 lr:7.469416297604748e-06 tokens_per_second_per_gpu:4742.563428192503 mfu:22.37148085554331 +Step 29922 | loss:0.25914937257766724 lr:7.467136496317212e-06 tokens_per_second_per_gpu:5108.976061715306 mfu:24.099911764312147 +Step 29923 | loss:0.3048715889453888 lr:7.464857014924764e-06 tokens_per_second_per_gpu:3309.2308266582004 mfu:15.610206422346806 +Step 29924 | loss:0.2762596905231476 lr:7.462577853444547e-06 tokens_per_second_per_gpu:4316.562934243573 mfu:20.36196384240709 +Step 29925 | loss:0.2740999758243561 lr:7.460299011893712e-06 tokens_per_second_per_gpu:4985.692323135528 mfu:23.518361335056532 +Step 29926 | loss:0.3127184212207794 lr:7.458020490289402e-06 tokens_per_second_per_gpu:5368.602191953718 mfu:25.324612517432293 +Step 29927 | loss:0.42498573660850525 lr:7.455742288648743e-06 tokens_per_second_per_gpu:3802.5299834929656 mfu:17.937182710651424 +Step 29928 | loss:0.34772375226020813 lr:7.453464406988869e-06 tokens_per_second_per_gpu:5336.417446379577 mfu:25.17279158127523 +Step 29929 | loss:0.16914860904216766 lr:7.451186845326935e-06 tokens_per_second_per_gpu:4052.6518504730343 mfu:19.11705023238744 +Step 29930 | loss:0.34634652733802795 lr:7.448909603680049e-06 tokens_per_second_per_gpu:4126.835602343619 mfu:19.46698764701421 +Step 29931 | loss:0.27673205733299255 lr:7.446632682065336e-06 tokens_per_second_per_gpu:5343.996625643259 mfu:25.208543863003346 +Step 29932 | loss:0.2796870172023773 lr:7.444356080499926e-06 tokens_per_second_per_gpu:5391.864107547189 mfu:25.434342942178755 +Step 29933 | loss:0.32283923029899597 lr:7.442079799000945e-06 tokens_per_second_per_gpu:4272.184930870972 mfu:20.152625228830388 +Step 29934 | loss:0.4186498820781708 lr:7.439803837585524e-06 tokens_per_second_per_gpu:4503.778751435778 mfu:21.24509279441492 +Step 29935 | loss:0.24150827527046204 lr:7.437528196270755e-06 tokens_per_second_per_gpu:4078.2540578229555 mfu:19.237820212644024 +Step 29936 | loss:0.3918284475803375 lr:7.435252875073783e-06 tokens_per_second_per_gpu:3382.152734920619 mfu:15.95419150538114 +Step 29937 | loss:0.2990310788154602 lr:7.432977874011693e-06 tokens_per_second_per_gpu:5319.315843923877 mfu:25.092120404657436 +Step 29938 | loss:0.29749009013175964 lr:7.430703193101618e-06 tokens_per_second_per_gpu:4703.760000798011 mfu:22.188438467976088 +Step 29939 | loss:0.3318241238594055 lr:7.428428832360646e-06 tokens_per_second_per_gpu:3983.5264953157157 mfu:18.79097413810862 +Step 29940 | loss:0.3023286759853363 lr:7.426154791805889e-06 tokens_per_second_per_gpu:4175.593612646457 mfu:19.69698750058708 +Step 29941 | loss:0.2924518883228302 lr:7.423881071454469e-06 tokens_per_second_per_gpu:4378.969854794638 mfu:20.65634793436507 +Step 29942 | loss:0.2765929400920868 lr:7.4216076713234605e-06 tokens_per_second_per_gpu:4850.993834010265 mfu:22.8829655799206 +Step 29943 | loss:0.34422567486763 lr:7.419334591429972e-06 tokens_per_second_per_gpu:5346.703306102278 mfu:25.221311736535842 +Step 29944 | loss:0.268119215965271 lr:7.417061831791117e-06 tokens_per_second_per_gpu:4020.5352953681104 mfu:18.96555096230817 +Step 29945 | loss:0.2514529526233673 lr:7.414789392423971e-06 tokens_per_second_per_gpu:4529.738886691933 mfu:21.367551181674006 +Step 29946 | loss:0.3227151334285736 lr:7.412517273345615e-06 tokens_per_second_per_gpu:4155.530681473682 mfu:19.60234723115607 +Step 29947 | loss:0.27468276023864746 lr:7.410245474573152e-06 tokens_per_second_per_gpu:4580.443083034373 mfu:21.60673152684926 +Step 29948 | loss:0.28981804847717285 lr:7.40797399612368e-06 tokens_per_second_per_gpu:4410.992810121411 mfu:20.807405678321032 +Step 29949 | loss:0.18308766186237335 lr:7.405702838014261e-06 tokens_per_second_per_gpu:3751.4653606466063 mfu:17.69630217216208 +Step 29950 | loss:0.3084685504436493 lr:7.403432000261978e-06 tokens_per_second_per_gpu:3817.1742237099843 mfu:18.006262090320124 +Step 29951 | loss:0.25655409693717957 lr:7.401161482883939e-06 tokens_per_second_per_gpu:5283.846991425515 mfu:24.924807775812404 +Step 29952 | loss:0.263814777135849 lr:7.398891285897197e-06 tokens_per_second_per_gpu:5306.6621389549 mfu:25.032430719374084 +Step 29953 | loss:0.40489718317985535 lr:7.396621409318816e-06 tokens_per_second_per_gpu:4995.858766926139 mfu:23.566318184991022 +Step 29954 | loss:0.23771168291568756 lr:7.394351853165882e-06 tokens_per_second_per_gpu:4453.803641680191 mfu:21.009351675064853 +Step 29955 | loss:0.37759116291999817 lr:7.392082617455476e-06 tokens_per_second_per_gpu:4735.27162187131 mfu:22.337084161015802 +Step 29956 | loss:0.3393666744232178 lr:7.38981370220464e-06 tokens_per_second_per_gpu:5329.243613617124 mfu:25.138951388152655 +Step 29957 | loss:0.2878468632698059 lr:7.387545107430455e-06 tokens_per_second_per_gpu:4695.02782933838 mfu:22.14724732533893 +Step 29958 | loss:0.4169420897960663 lr:7.38527683314999e-06 tokens_per_second_per_gpu:4259.156632462147 mfu:20.091168522379043 +Step 29959 | loss:0.31883347034454346 lr:7.383008879380298e-06 tokens_per_second_per_gpu:5320.345326782052 mfu:25.096976650947337 +Step 29960 | loss:0.3119748830795288 lr:7.380741246138423e-06 tokens_per_second_per_gpu:4344.955445604003 mfu:20.495896162756203 +Step 29961 | loss:0.2760903835296631 lr:7.378473933441432e-06 tokens_per_second_per_gpu:5347.0093022814235 mfu:25.222755172721204 +Step 29962 | loss:0.26941296458244324 lr:7.376206941306385e-06 tokens_per_second_per_gpu:5255.6147213904605 mfu:24.791631341135083 +Step 29963 | loss:0.2883845567703247 lr:7.373940269750318e-06 tokens_per_second_per_gpu:4054.830469705716 mfu:19.127327151018772 +Step 29964 | loss:0.3221115171909332 lr:7.371673918790278e-06 tokens_per_second_per_gpu:3925.8298731530795 mfu:18.518809327308634 +Step 29965 | loss:0.2698995769023895 lr:7.369407888443336e-06 tokens_per_second_per_gpu:5324.909398524031 mfu:25.118506156065187 +Step 29966 | loss:0.40092933177948 lr:7.367142178726516e-06 tokens_per_second_per_gpu:4409.369484390511 mfu:20.79974817388045 +Step 29967 | loss:0.2977864146232605 lr:7.3648767896568435e-06 tokens_per_second_per_gpu:4794.872583716618 mfu:22.618232067012798 +Step 29968 | loss:0.32931676506996155 lr:7.362611721251378e-06 tokens_per_second_per_gpu:4449.873216631476 mfu:20.990811189510403 +Step 29969 | loss:0.48059314489364624 lr:7.360346973527149e-06 tokens_per_second_per_gpu:3735.664254677585 mfu:17.621765659359536 +Step 29970 | loss:0.22263751924037933 lr:7.3580825465011995e-06 tokens_per_second_per_gpu:4711.289266635462 mfu:22.223955299556565 +Step 29971 | loss:0.2981679141521454 lr:7.3558184401905425e-06 tokens_per_second_per_gpu:5054.787449045368 mfu:23.844294832817354 +Step 29972 | loss:0.22641688585281372 lr:7.353554654612221e-06 tokens_per_second_per_gpu:4208.738841524673 mfu:19.853339200364818 +Step 29973 | loss:0.28884026408195496 lr:7.35129118978326e-06 tokens_per_second_per_gpu:5175.142887719554 mfu:24.412031971798807 +Step 29974 | loss:0.26987648010253906 lr:7.349028045720685e-06 tokens_per_second_per_gpu:4134.312644737627 mfu:19.502258131701726 +Step 29975 | loss:0.3364810049533844 lr:7.346765222441504e-06 tokens_per_second_per_gpu:5362.129738700685 mfu:25.29408085112334 +Step 29976 | loss:0.32413655519485474 lr:7.344502719962743e-06 tokens_per_second_per_gpu:3772.2124045309083 mfu:17.794169517975035 +Step 29977 | loss:0.30787569284439087 lr:7.342240538301426e-06 tokens_per_second_per_gpu:5367.299075431584 mfu:25.31846549446278 +Step 29978 | loss:0.34026867151260376 lr:7.3399786774745516e-06 tokens_per_second_per_gpu:5055.037034272206 mfu:23.845472168915983 +Step 29979 | loss:0.36283114552497864 lr:7.3377171374991445e-06 tokens_per_second_per_gpu:4489.29081358526 mfu:21.17675071967694 +Step 29980 | loss:0.34143030643463135 lr:7.33545591839222e-06 tokens_per_second_per_gpu:4883.2368788108415 mfu:23.035061523475633 +Step 29981 | loss:0.1932581216096878 lr:7.333195020170774e-06 tokens_per_second_per_gpu:5310.580798780928 mfu:25.050915706365764 +Step 29982 | loss:0.3215301036834717 lr:7.3309344428518e-06 tokens_per_second_per_gpu:4798.585806844925 mfu:22.635747974050215 +Step 29983 | loss:0.28874415159225464 lr:7.3286741864523164e-06 tokens_per_second_per_gpu:4055.6448388008257 mfu:19.1311686689861 +Step 29984 | loss:0.26499903202056885 lr:7.326414250989333e-06 tokens_per_second_per_gpu:3021.616642683802 mfu:14.253481244499635 +Step 29985 | loss:0.31785425543785095 lr:7.324154636479813e-06 tokens_per_second_per_gpu:5377.989708797877 mfu:25.368895036061723 +Step 29986 | loss:0.293039470911026 lr:7.321895342940777e-06 tokens_per_second_per_gpu:3205.4809702853377 mfu:15.120800648285192 +Step 29987 | loss:0.34557434916496277 lr:7.319636370389227e-06 tokens_per_second_per_gpu:4731.115780499826 mfu:22.31748034820685 +Step 29988 | loss:0.2196524292230606 lr:7.317377718842133e-06 tokens_per_second_per_gpu:4924.334505175453 mfu:23.228926039015615 +Step 29989 | loss:0.2914753556251526 lr:7.315119388316477e-06 tokens_per_second_per_gpu:5329.145434032829 mfu:25.138488258302374 +Step 29990 | loss:0.2521524429321289 lr:7.31286137882925e-06 tokens_per_second_per_gpu:4268.023941756819 mfu:20.132997133240984 +Step 29991 | loss:0.2259046584367752 lr:7.3106036903974605e-06 tokens_per_second_per_gpu:5361.581657376593 mfu:25.291495458005283 +Step 29992 | loss:0.2795456349849701 lr:7.308346323038051e-06 tokens_per_second_per_gpu:4739.066130358424 mfu:22.35498350495942 +Step 29993 | loss:0.3571378290653229 lr:7.306089276768013e-06 tokens_per_second_per_gpu:5343.84116991825 mfu:25.207810551824995 +Step 29994 | loss:0.2166697382926941 lr:7.30383255160434e-06 tokens_per_second_per_gpu:5369.469402652779 mfu:25.328703298261164 +Step 29995 | loss:0.2464134395122528 lr:7.301576147563993e-06 tokens_per_second_per_gpu:4865.198784224362 mfu:22.949972753694915 +Step 29996 | loss:0.2737816274166107 lr:7.299320064663928e-06 tokens_per_second_per_gpu:5275.789939518805 mfu:24.886801287293647 +Step 29997 | loss:0.3288205564022064 lr:7.29706430292112e-06 tokens_per_second_per_gpu:4485.784569676706 mfu:21.16021116893339 +Step 29998 | loss:0.2687280774116516 lr:7.294808862352559e-06 tokens_per_second_per_gpu:5038.817895948862 mfu:23.76896372617442 +Step 29999 | loss:0.23335354030132294 lr:7.2925537429751725e-06 tokens_per_second_per_gpu:4702.952783109397 mfu:22.184630684413364 +Step 30000 | loss:0.39659056067466736 lr:7.290298944805946e-06 tokens_per_second_per_gpu:5369.936434310992 mfu:25.33090636627677 +Step 30001 | loss:0.25985389947891235 lr:7.288044467861821e-06 tokens_per_second_per_gpu:5306.877537994733 mfu:25.03344679339584 +Step 30002 | loss:0.2917807996273041 lr:7.2857903121597955e-06 tokens_per_second_per_gpu:3812.7056569919196 mfu:17.985183098694915 +Step 30003 | loss:0.3298896253108978 lr:7.2835364777167626e-06 tokens_per_second_per_gpu:4280.231687615774 mfu:20.19058315331389 +Step 30004 | loss:0.2719672620296478 lr:7.281282964549707e-06 tokens_per_second_per_gpu:5352.067632703846 mfu:25.24661618036394 +Step 30005 | loss:0.3363415598869324 lr:7.2790297726755716e-06 tokens_per_second_per_gpu:3443.965591307032 mfu:16.245773295316656 +Step 30006 | loss:0.35416629910469055 lr:7.276776902111315e-06 tokens_per_second_per_gpu:4430.143854570854 mfu:20.89774442249417 +Step 30007 | loss:0.3271321952342987 lr:7.2745243528738574e-06 tokens_per_second_per_gpu:3923.400364524396 mfu:18.50734892059022 +Step 30008 | loss:0.29029911756515503 lr:7.272272124980156e-06 tokens_per_second_per_gpu:4760.418463838104 mfu:22.455706105066827 +Step 30009 | loss:0.30604299902915955 lr:7.270020218447171e-06 tokens_per_second_per_gpu:5372.163348241155 mfu:25.341411099236048 +Step 30010 | loss:0.24219925701618195 lr:7.267768633291794e-06 tokens_per_second_per_gpu:3477.903412944799 mfu:16.40586379037146 +Step 30011 | loss:0.3109978139400482 lr:7.265517369530983e-06 tokens_per_second_per_gpu:4980.3464991724695 mfu:23.493144171331167 +Step 30012 | loss:0.2551635503768921 lr:7.263266427181664e-06 tokens_per_second_per_gpu:5297.678911583897 mfu:24.99005530316843 +Step 30013 | loss:0.3826351761817932 lr:7.26101580626079e-06 tokens_per_second_per_gpu:5317.278766678672 mfu:25.08251116373059 +Step 30014 | loss:0.32981663942337036 lr:7.258765506785254e-06 tokens_per_second_per_gpu:4994.855609536902 mfu:23.56158612042969 +Step 30015 | loss:0.32030776143074036 lr:7.2565155287719955e-06 tokens_per_second_per_gpu:4682.63266850963 mfu:22.088777236877206 +Step 30016 | loss:0.36194488406181335 lr:7.254265872237948e-06 tokens_per_second_per_gpu:4779.34159715599 mfu:22.544969753546763 +Step 30017 | loss:0.33417809009552 lr:7.252016537200024e-06 tokens_per_second_per_gpu:5347.681297198406 mfu:25.22592508739877 +Step 30018 | loss:0.305252343416214 lr:7.249767523675127e-06 tokens_per_second_per_gpu:4270.942392702606 mfu:20.14676396429134 +Step 30019 | loss:0.23570674657821655 lr:7.247518831680183e-06 tokens_per_second_per_gpu:5388.428441243681 mfu:25.41813631803999 +Step 30020 | loss:0.30416351556777954 lr:7.245270461232118e-06 tokens_per_second_per_gpu:5354.963840688019 mfu:25.26027808756932 +Step 30021 | loss:0.4797842800617218 lr:7.243022412347817e-06 tokens_per_second_per_gpu:4931.298387162694 mfu:23.261775858510216 +Step 30022 | loss:0.30952689051628113 lr:7.2407746850442e-06 tokens_per_second_per_gpu:5345.425178416408 mfu:25.215282590170528 +Step 30023 | loss:0.2959973216056824 lr:7.238527279338181e-06 tokens_per_second_per_gpu:4098.515801088111 mfu:19.333398312635644 +Step 30024 | loss:0.3365227282047272 lr:7.2362801952466584e-06 tokens_per_second_per_gpu:4401.783423670938 mfu:20.763963431150795 +Step 30025 | loss:0.3222033381462097 lr:7.234033432786519e-06 tokens_per_second_per_gpu:5360.9970554608435 mfu:25.28873779102532 +Step 30026 | loss:0.34073418378829956 lr:7.2317869919746705e-06 tokens_per_second_per_gpu:4561.408093371425 mfu:21.516940232905007 +Step 30027 | loss:0.26930737495422363 lr:7.229540872828017e-06 tokens_per_second_per_gpu:5064.382861054456 mfu:23.88955802841033 +Step 30028 | loss:0.3166581094264984 lr:7.2272950753634325e-06 tokens_per_second_per_gpu:5336.503361300271 mfu:25.173196856614737 +Step 30029 | loss:0.4255152642726898 lr:7.225049599597828e-06 tokens_per_second_per_gpu:5346.226690311014 mfu:25.219063458529412 +Step 30030 | loss:0.36520346999168396 lr:7.222804445548087e-06 tokens_per_second_per_gpu:5029.929903245936 mfu:23.72703754814679 +Step 30031 | loss:0.4398518204689026 lr:7.2205596132310925e-06 tokens_per_second_per_gpu:5353.424081331864 mfu:25.2530147799763 +Step 30032 | loss:0.2911694645881653 lr:7.21831510266372e-06 tokens_per_second_per_gpu:5042.884751364088 mfu:23.788147777045122 +Step 30033 | loss:0.4120745062828064 lr:7.216070913862855e-06 tokens_per_second_per_gpu:5322.991459044318 mfu:25.109458908305164 +Step 30034 | loss:0.3098697364330292 lr:7.213827046845395e-06 tokens_per_second_per_gpu:4411.8433643033195 mfu:20.8114178875166 +Step 30035 | loss:0.35580065846443176 lr:7.211583501628188e-06 tokens_per_second_per_gpu:4935.163935279217 mfu:23.280010308506373 +Step 30036 | loss:0.310639888048172 lr:7.209340278228122e-06 tokens_per_second_per_gpu:5344.292026993007 mfu:25.209937321570774 +Step 30037 | loss:0.23092898726463318 lr:7.2070973766620645e-06 tokens_per_second_per_gpu:4969.5810119643365 mfu:23.442361531388883 +Step 30038 | loss:0.3058925271034241 lr:7.2048547969469136e-06 tokens_per_second_per_gpu:4373.196816443445 mfu:20.62911552748058 +Step 30039 | loss:0.2710023522377014 lr:7.20261253909949e-06 tokens_per_second_per_gpu:4564.14236226994 mfu:21.529838245813682 +Step 30040 | loss:0.24768921732902527 lr:7.200370603136675e-06 tokens_per_second_per_gpu:3884.6219412971673 mfu:18.32442448194613 +Step 30041 | loss:0.3690946400165558 lr:7.1981289890753366e-06 tokens_per_second_per_gpu:5359.835144796883 mfu:25.283256860181012 +Step 30042 | loss:0.289093017578125 lr:7.1958876969323404e-06 tokens_per_second_per_gpu:4420.678706664709 mfu:20.85309570490013 +Step 30043 | loss:0.37171658873558044 lr:7.1936467267245235e-06 tokens_per_second_per_gpu:4190.055304672957 mfu:19.765205769295175 +Step 30044 | loss:0.2578166127204895 lr:7.1914060784687545e-06 tokens_per_second_per_gpu:5269.765313118155 mfu:24.858382096654523 +Step 30045 | loss:0.30074232816696167 lr:7.189165752181904e-06 tokens_per_second_per_gpu:2905.692184780409 mfu:13.706645797817293 +Step 30046 | loss:0.38642236590385437 lr:7.186925747880774e-06 tokens_per_second_per_gpu:5350.697292782686 mfu:25.24015205315173 +Step 30047 | loss:0.2855547368526459 lr:7.184686065582247e-06 tokens_per_second_per_gpu:5147.935411046507 mfu:24.28368966225785 +Step 30048 | loss:0.30252179503440857 lr:7.182446705303153e-06 tokens_per_second_per_gpu:3815.980963509595 mfu:18.000633278363704 +Step 30049 | loss:0.27386125922203064 lr:7.180207667060351e-06 tokens_per_second_per_gpu:4809.159933212096 mfu:22.685627932256157 +Step 30050 | loss:0.22215935587882996 lr:7.177968950870662e-06 tokens_per_second_per_gpu:3586.005696426364 mfu:16.915800705705465 +Step 30051 | loss:0.3281552493572235 lr:7.175730556750931e-06 tokens_per_second_per_gpu:5360.9838601282245 mfu:25.28867554639 +Step 30052 | loss:0.2793116271495819 lr:7.173492484718008e-06 tokens_per_second_per_gpu:5207.5090532219165 mfu:24.56470869671072 +Step 30053 | loss:0.2891576588153839 lr:7.171254734788718e-06 tokens_per_second_per_gpu:3727.208522156816 mfu:17.58187852636233 +Step 30054 | loss:0.2656382620334625 lr:7.169017306979869e-06 tokens_per_second_per_gpu:3761.0778936871325 mfu:17.741646130581497 +Step 30055 | loss:0.3267917037010193 lr:7.166780201308304e-06 tokens_per_second_per_gpu:5379.153947060368 mfu:25.374386946585428 +Step 30056 | loss:0.4789240062236786 lr:7.1645434177908645e-06 tokens_per_second_per_gpu:4371.952505232063 mfu:20.623245899195105 +Step 30057 | loss:0.3714120388031006 lr:7.162306956444348e-06 tokens_per_second_per_gpu:5365.966747125699 mfu:25.312180674525273 +Step 30058 | loss:0.3149555027484894 lr:7.160070817285591e-06 tokens_per_second_per_gpu:3315.8555849033733 mfu:15.641456537289598 +Step 30059 | loss:0.35339516401290894 lr:7.1578350003314146e-06 tokens_per_second_per_gpu:4518.829155125967 mfu:21.3160881164852 +Step 30060 | loss:0.25909891724586487 lr:7.155599505598631e-06 tokens_per_second_per_gpu:3790.990123866482 mfu:17.88274722389022 +Step 30061 | loss:0.2510135769844055 lr:7.153364333104034e-06 tokens_per_second_per_gpu:3929.5996900397454 mfu:18.536592196760125 +Step 30062 | loss:0.41467487812042236 lr:7.151129482864455e-06 tokens_per_second_per_gpu:5012.46352867398 mfu:23.644645679220076 +Step 30063 | loss:0.3928264379501343 lr:7.148894954896712e-06 tokens_per_second_per_gpu:3438.7044804802017 mfu:16.22095573790843 +Step 30064 | loss:0.19815875589847565 lr:7.146660749217582e-06 tokens_per_second_per_gpu:5374.497768024041 mfu:25.352422955645007 +Step 30065 | loss:0.203008234500885 lr:7.144426865843884e-06 tokens_per_second_per_gpu:5355.734092511694 mfu:25.263911496840105 +Step 30066 | loss:0.31429344415664673 lr:7.1421933047924386e-06 tokens_per_second_per_gpu:5049.934266105016 mfu:23.821401540849322 +Step 30067 | loss:0.26887017488479614 lr:7.13996006608002e-06 tokens_per_second_per_gpu:5136.018299289252 mfu:24.22747461282995 +Step 30068 | loss:0.30474528670310974 lr:7.1377271497234165e-06 tokens_per_second_per_gpu:4628.192001269038 mfu:21.831971320967696 +Step 30069 | loss:0.3368719518184662 lr:7.1354945557394415e-06 tokens_per_second_per_gpu:4719.546138577173 mfu:22.262904373273237 +Step 30070 | loss:0.32527682185173035 lr:7.1332622841448925e-06 tokens_per_second_per_gpu:3327.263877241182 mfu:15.695271398702424 +Step 30071 | loss:0.31621408462524414 lr:7.131030334956534e-06 tokens_per_second_per_gpu:5343.293621915203 mfu:25.205227674473303 +Step 30072 | loss:0.31205257773399353 lr:7.12879870819117e-06 tokens_per_second_per_gpu:4938.17722803867 mfu:23.294224524573476 +Step 30073 | loss:0.2558479905128479 lr:7.1265674038655805e-06 tokens_per_second_per_gpu:5029.535558980557 mfu:23.725177358964928 +Step 30074 | loss:0.25667068362236023 lr:7.124336421996569e-06 tokens_per_second_per_gpu:4951.748628098968 mfu:23.358243134177517 +Step 30075 | loss:0.3822321593761444 lr:7.122105762600873e-06 tokens_per_second_per_gpu:4692.675237877244 mfu:22.13614975002326 +Step 30076 | loss:0.2673005759716034 lr:7.119875425695288e-06 tokens_per_second_per_gpu:5043.178441821921 mfu:23.789533165043483 +Step 30077 | loss:0.3299010396003723 lr:7.117645411296609e-06 tokens_per_second_per_gpu:3755.075485104536 mfu:17.713331745180884 +Step 30078 | loss:0.39812836050987244 lr:7.1154157194215755e-06 tokens_per_second_per_gpu:5327.788623891083 mfu:25.132087953368078 +Step 30079 | loss:0.25499141216278076 lr:7.11318635008697e-06 tokens_per_second_per_gpu:4960.586005575433 mfu:23.399930551546078 +Step 30080 | loss:0.2932590842247009 lr:7.110957303309568e-06 tokens_per_second_per_gpu:4880.542295355305 mfu:23.022350713572525 +Step 30081 | loss:0.3366670310497284 lr:7.108728579106144e-06 tokens_per_second_per_gpu:5326.543729731549 mfu:25.126215575216712 +Step 30082 | loss:0.4236844778060913 lr:7.10650017749343e-06 tokens_per_second_per_gpu:3939.1523406459755 mfu:18.58165367951997 +Step 30083 | loss:0.31666100025177 lr:7.104272098488196e-06 tokens_per_second_per_gpu:4384.698604545748 mfu:20.68337142436655 +Step 30084 | loss:0.2639446258544922 lr:7.102044342107206e-06 tokens_per_second_per_gpu:4558.591980709214 mfu:21.503656148998992 +Step 30085 | loss:0.3108646869659424 lr:7.0998169083672296e-06 tokens_per_second_per_gpu:3722.4373472781085 mfu:17.55937208041309 +Step 30086 | loss:0.3120414912700653 lr:7.0975897972849925e-06 tokens_per_second_per_gpu:4372.828004649818 mfu:20.627375779324257 +Step 30087 | loss:0.241489976644516 lr:7.095363008877254e-06 tokens_per_second_per_gpu:5359.048218963159 mfu:25.279544796759946 +Step 30088 | loss:0.3620240092277527 lr:7.093136543160778e-06 tokens_per_second_per_gpu:5283.011941420895 mfu:24.92086870246639 +Step 30089 | loss:0.2586721181869507 lr:7.0909104001522954e-06 tokens_per_second_per_gpu:4828.760731085778 mfu:22.77808824006728 +Step 30090 | loss:0.39209800958633423 lr:7.088684579868538e-06 tokens_per_second_per_gpu:4764.897272334413 mfu:22.476833408907446 +Step 30091 | loss:0.3571774661540985 lr:7.086459082326263e-06 tokens_per_second_per_gpu:4891.8023798484855 mfu:23.075466453295373 +Step 30092 | loss:0.32157832384109497 lr:7.084233907542215e-06 tokens_per_second_per_gpu:3656.067468354548 mfu:17.24629403766163 +Step 30093 | loss:0.31030818819999695 lr:7.082009055533106e-06 tokens_per_second_per_gpu:5231.989441002496 mfu:24.680186862656566 +Step 30094 | loss:0.3624935746192932 lr:7.079784526315686e-06 tokens_per_second_per_gpu:5353.327437666302 mfu:25.25255889531673 +Step 30095 | loss:0.3246273696422577 lr:7.077560319906695e-06 tokens_per_second_per_gpu:3372.3728794828617 mfu:15.908058258665838 +Step 30096 | loss:0.3002752959728241 lr:7.0753364363228494e-06 tokens_per_second_per_gpu:5112.36909855756 mfu:24.115917298008842 +Step 30097 | loss:0.3267139494419098 lr:7.0731128755808676e-06 tokens_per_second_per_gpu:5131.067180228571 mfu:24.20411933947269 +Step 30098 | loss:0.17951641976833344 lr:7.070889637697476e-06 tokens_per_second_per_gpu:4278.215974001963 mfu:20.181074688281342 +Step 30099 | loss:0.21760061383247375 lr:7.068666722689415e-06 tokens_per_second_per_gpu:5019.584494111335 mfu:23.678236488150066 +Step 30100 | loss:0.2628667950630188 lr:7.06644413057338e-06 tokens_per_second_per_gpu:5373.176704572421 mfu:25.34619127394705 +Step 30101 | loss:0.27932876348495483 lr:7.064221861366094e-06 tokens_per_second_per_gpu:5364.7714468477825 mfu:25.306542239175084 +Step 30102 | loss:0.4798341393470764 lr:7.061999915084283e-06 tokens_per_second_per_gpu:5213.873505740081 mfu:24.594730905126475 +Step 30103 | loss:0.2890731692314148 lr:7.059778291744656e-06 tokens_per_second_per_gpu:5318.775389254717 mfu:25.089570987771936 +Step 30104 | loss:0.35794997215270996 lr:7.057556991363901e-06 tokens_per_second_per_gpu:5308.562816540664 mfu:25.041396539835166 +Step 30105 | loss:0.3541370928287506 lr:7.055336013958741e-06 tokens_per_second_per_gpu:5124.101124327577 mfu:24.171259265256012 +Step 30106 | loss:0.42071184515953064 lr:7.053115359545886e-06 tokens_per_second_per_gpu:5018.047004007293 mfu:23.67098388500644 +Step 30107 | loss:0.39189475774765015 lr:7.050895028142024e-06 tokens_per_second_per_gpu:3872.061431967546 mfu:18.26517441639443 +Step 30108 | loss:0.3339085578918457 lr:7.048675019763856e-06 tokens_per_second_per_gpu:5358.132249026066 mfu:25.27522401028609 +Step 30109 | loss:0.2553179860115051 lr:7.046455334428087e-06 tokens_per_second_per_gpu:5323.761044812911 mfu:25.11308917568049 +Step 30110 | loss:0.28206875920295715 lr:7.044235972151431e-06 tokens_per_second_per_gpu:5219.080482300407 mfu:24.619293101963343 +Step 30111 | loss:0.3077690899372101 lr:7.042016932950535e-06 tokens_per_second_per_gpu:4367.167877615417 mfu:20.600675994385735 +Step 30112 | loss:0.3330249488353729 lr:7.0397982168421085e-06 tokens_per_second_per_gpu:3957.2178783588697 mfu:18.666871903210264 +Step 30113 | loss:0.34944817423820496 lr:7.037579823842855e-06 tokens_per_second_per_gpu:4125.890578003452 mfu:19.462529805963868 +Step 30114 | loss:0.46531349420547485 lr:7.035361753969433e-06 tokens_per_second_per_gpu:4508.630982860579 mfu:21.267981597921754 +Step 30115 | loss:0.3900914192199707 lr:7.03314400723854e-06 tokens_per_second_per_gpu:3286.531141913286 mfu:15.50312813643946 +Step 30116 | loss:0.29858261346817017 lr:7.030926583666847e-06 tokens_per_second_per_gpu:4897.598003097275 mfu:23.10280539699521 +Step 30117 | loss:0.24479080736637115 lr:7.028709483271067e-06 tokens_per_second_per_gpu:4513.441514688694 mfu:21.29067369731634 +Step 30118 | loss:0.3178081214427948 lr:7.026492706067822e-06 tokens_per_second_per_gpu:5360.536389422035 mfu:25.286564750722526 +Step 30119 | loss:0.2919422686100006 lr:7.024276252073808e-06 tokens_per_second_per_gpu:4059.164657057705 mfu:19.147772252246718 +Step 30120 | loss:0.3104914128780365 lr:7.022060121305696e-06 tokens_per_second_per_gpu:4523.566035675377 mfu:21.338432790231927 +Step 30121 | loss:0.3245792090892792 lr:7.019844313780166e-06 tokens_per_second_per_gpu:5004.351931910896 mfu:23.60638189330749 +Step 30122 | loss:0.2849629521369934 lr:7.017628829513862e-06 tokens_per_second_per_gpu:4484.748735866833 mfu:21.155324964120748 +Step 30123 | loss:0.30192920565605164 lr:7.015413668523452e-06 tokens_per_second_per_gpu:5217.749327423957 mfu:24.61301381729269 +Step 30124 | loss:0.3243873715400696 lr:7.013198830825618e-06 tokens_per_second_per_gpu:5343.761405885065 mfu:25.207434291271316 +Step 30125 | loss:0.414588063955307 lr:7.010984316436997e-06 tokens_per_second_per_gpu:4559.159014923975 mfu:21.50633094613603 +Step 30126 | loss:0.25746944546699524 lr:7.008770125374237e-06 tokens_per_second_per_gpu:5352.807618958765 mfu:25.25010682178306 +Step 30127 | loss:0.3036748170852661 lr:7.006556257654001e-06 tokens_per_second_per_gpu:4277.69969913947 mfu:20.178639331669388 +Step 30128 | loss:0.3865397274494171 lr:7.004342713292955e-06 tokens_per_second_per_gpu:3893.8868968427773 mfu:18.368128858019194 +Step 30129 | loss:0.30952852964401245 lr:7.002129492307719e-06 tokens_per_second_per_gpu:5303.93744338707 mfu:25.019577865499627 +Step 30130 | loss:0.3738328814506531 lr:6.9999165947149555e-06 tokens_per_second_per_gpu:5081.641398283906 mfu:23.970969493131737 +Step 30131 | loss:0.29076576232910156 lr:6.997704020531321e-06 tokens_per_second_per_gpu:4706.237607446458 mfu:22.2001257612601 +Step 30132 | loss:0.45083755254745483 lr:6.995491769773438e-06 tokens_per_second_per_gpu:5322.7679955997855 mfu:25.108404793110513 +Step 30133 | loss:0.1958404928445816 lr:6.993279842457939e-06 tokens_per_second_per_gpu:4723.731975090343 mfu:22.282649678283214 +Step 30134 | loss:0.28717172145843506 lr:6.991068238601467e-06 tokens_per_second_per_gpu:5204.404880210813 mfu:24.550065782990814 +Step 30135 | loss:0.2744904160499573 lr:6.988856958220674e-06 tokens_per_second_per_gpu:4748.5131242348725 mfu:22.39954658689799 +Step 30136 | loss:0.21558907628059387 lr:6.9866460013321644e-06 tokens_per_second_per_gpu:4368.040764758449 mfu:20.604793551969323 +Step 30137 | loss:0.3690508008003235 lr:6.98443536795258e-06 tokens_per_second_per_gpu:5165.899817027064 mfu:24.368430830311855 +Step 30138 | loss:0.34443849325180054 lr:6.982225058098551e-06 tokens_per_second_per_gpu:5324.152430867105 mfu:25.11493541047547 +Step 30139 | loss:0.28466489911079407 lr:6.9800150717867e-06 tokens_per_second_per_gpu:5334.3947450930955 mfu:25.163250154198472 +Step 30140 | loss:0.2506725788116455 lr:6.977805409033633e-06 tokens_per_second_per_gpu:4061.498637593894 mfu:19.158782036654063 +Step 30141 | loss:0.2337614744901657 lr:6.975596069855983e-06 tokens_per_second_per_gpu:4417.785473847471 mfu:20.839447832064792 +Step 30142 | loss:0.32535022497177124 lr:6.973387054270375e-06 tokens_per_second_per_gpu:4983.62877087909 mfu:23.508627206984595 +Step 30143 | loss:0.318350613117218 lr:6.971178362293407e-06 tokens_per_second_per_gpu:5296.9403541578295 mfu:24.986571401778775 +Step 30144 | loss:0.24333229660987854 lr:6.968969993941688e-06 tokens_per_second_per_gpu:4765.161279457341 mfu:22.478078775550557 +Step 30145 | loss:0.4340079426765442 lr:6.966761949231854e-06 tokens_per_second_per_gpu:4879.338088244022 mfu:23.016670263169683 +Step 30146 | loss:0.288287490606308 lr:6.964554228180492e-06 tokens_per_second_per_gpu:4013.2010498997984 mfu:18.93095407508283 +Step 30147 | loss:0.2772482931613922 lr:6.9623468308042e-06 tokens_per_second_per_gpu:4297.657321947159 mfu:20.27278284357483 +Step 30148 | loss:0.3250681161880493 lr:6.960139757119588e-06 tokens_per_second_per_gpu:4324.429452998778 mfu:20.399071553541894 +Step 30149 | loss:0.3061034083366394 lr:6.957933007143269e-06 tokens_per_second_per_gpu:3365.262383247878 mfu:15.874516834749665 +Step 30150 | loss:0.2475476861000061 lr:6.9557265808918135e-06 tokens_per_second_per_gpu:4318.675764489389 mfu:20.371930423162617 +Step 30151 | loss:0.2880672812461853 lr:6.953520478381831e-06 tokens_per_second_per_gpu:5007.8505215692185 mfu:23.622885337646746 +Step 30152 | loss:0.48006874322891235 lr:6.951314699629918e-06 tokens_per_second_per_gpu:3848.561356786534 mfu:18.154320552239323 +Step 30153 | loss:0.43703532218933105 lr:6.94910924465268e-06 tokens_per_second_per_gpu:3878.838301037223 mfu:18.297142064049194 +Step 30154 | loss:0.29024991393089294 lr:6.946904113466662e-06 tokens_per_second_per_gpu:4372.065994682931 mfu:20.623781248298148 +Step 30155 | loss:0.37630656361579895 lr:6.9446993060884705e-06 tokens_per_second_per_gpu:5363.799814154778 mfu:25.30195888198455 +Step 30156 | loss:0.2038891613483429 lr:6.94249482253469e-06 tokens_per_second_per_gpu:4862.082654402568 mfu:22.935273437658946 +Step 30157 | loss:0.30041614174842834 lr:6.940290662821913e-06 tokens_per_second_per_gpu:4981.867944266285 mfu:23.500321087423995 +Step 30158 | loss:0.39282089471817017 lr:6.938086826966694e-06 tokens_per_second_per_gpu:5346.054936485426 mfu:25.21825326642989 +Step 30159 | loss:0.22139175236225128 lr:6.93588331498562e-06 tokens_per_second_per_gpu:4120.092793215337 mfu:19.435180665914242 +Step 30160 | loss:0.38585272431373596 lr:6.93368012689527e-06 tokens_per_second_per_gpu:4841.817143635518 mfu:22.83967756571809 +Step 30161 | loss:0.2688881754875183 lr:6.931477262712205e-06 tokens_per_second_per_gpu:4894.904015827748 mfu:23.09009739940322 +Step 30162 | loss:0.3366466760635376 lr:6.9292747224529884e-06 tokens_per_second_per_gpu:5353.910614022757 mfu:25.255309837690277 +Step 30163 | loss:0.2615554630756378 lr:6.927072506134191e-06 tokens_per_second_per_gpu:5337.038417711704 mfu:25.175720808997507 +Step 30164 | loss:0.339224249124527 lr:6.924870613772389e-06 tokens_per_second_per_gpu:4195.803249170662 mfu:19.792319804191305 +Step 30165 | loss:0.30545979738235474 lr:6.922669045384123e-06 tokens_per_second_per_gpu:5333.586714147622 mfu:25.159438534364376 +Step 30166 | loss:0.27418622374534607 lr:6.9204678009859524e-06 tokens_per_second_per_gpu:3892.908851499928 mfu:18.36351525126523 +Step 30167 | loss:0.37089699506759644 lr:6.918266880594459e-06 tokens_per_second_per_gpu:5119.107552724416 mfu:24.147703736012726 +Step 30168 | loss:0.278842031955719 lr:6.9160662842261746e-06 tokens_per_second_per_gpu:5365.962336151095 mfu:25.312159867205477 +Step 30169 | loss:0.3289964199066162 lr:6.91386601189764e-06 tokens_per_second_per_gpu:5107.2964841147805 mfu:24.09198891020498 +Step 30170 | loss:0.28922876715660095 lr:6.911666063625422e-06 tokens_per_second_per_gpu:3708.048981263836 mfu:17.491499703015386 +Step 30171 | loss:0.27525794506073 lr:6.909466439426071e-06 tokens_per_second_per_gpu:4104.15609607022 mfu:19.360004546399807 +Step 30172 | loss:0.295171856880188 lr:6.9072671393161096e-06 tokens_per_second_per_gpu:3903.1682931020805 mfu:18.41191078774382 +Step 30173 | loss:0.36526942253112793 lr:6.905068163312089e-06 tokens_per_second_per_gpu:3348.8874517517625 mfu:15.79727348903108 +Step 30174 | loss:0.27491286396980286 lr:6.90286951143056e-06 tokens_per_second_per_gpu:3902.9707449307407 mfu:18.41097891931442 +Step 30175 | loss:0.15996059775352478 lr:6.900671183688051e-06 tokens_per_second_per_gpu:4598.880495034099 mfu:21.693703944998834 +Step 30176 | loss:0.34947627782821655 lr:6.898473180101084e-06 tokens_per_second_per_gpu:5053.389798195241 mfu:23.837701875293483 +Step 30177 | loss:0.25635430216789246 lr:6.896275500686195e-06 tokens_per_second_per_gpu:3275.413369666526 mfu:15.450683707863947 +Step 30178 | loss:0.24346601963043213 lr:6.894078145459931e-06 tokens_per_second_per_gpu:3742.64209262165 mfu:17.654681311483607 +Step 30179 | loss:0.2914585471153259 lr:6.891881114438798e-06 tokens_per_second_per_gpu:4245.189141182241 mfu:20.02528148291197 +Step 30180 | loss:0.3888995945453644 lr:6.889684407639324e-06 tokens_per_second_per_gpu:5354.044987488628 mfu:25.25594370025527 +Step 30181 | loss:0.2355402112007141 lr:6.887488025078048e-06 tokens_per_second_per_gpu:4617.81736912074 mfu:21.783032411029406 +Step 30182 | loss:0.27892404794692993 lr:6.885291966771479e-06 tokens_per_second_per_gpu:3981.823541121224 mfu:18.782941012619254 +Step 30183 | loss:0.24547524750232697 lr:6.883096232736114e-06 tokens_per_second_per_gpu:3912.7708249758593 mfu:18.457207568953315 +Step 30184 | loss:0.29145547747612 lr:6.880900822988484e-06 tokens_per_second_per_gpu:4834.4384262384465 mfu:22.8048708968171 +Step 30185 | loss:0.274998277425766 lr:6.8787057375451155e-06 tokens_per_second_per_gpu:5326.482343751439 mfu:25.125926007075 +Step 30186 | loss:0.37344902753829956 lr:6.87651097642249e-06 tokens_per_second_per_gpu:4792.386657621365 mfu:22.606505529478763 +Step 30187 | loss:0.3030971586704254 lr:6.874316539637127e-06 tokens_per_second_per_gpu:4433.223899565145 mfu:20.912273520242106 +Step 30188 | loss:0.22668270766735077 lr:6.872122427205535e-06 tokens_per_second_per_gpu:4760.207064386396 mfu:22.454708897784656 +Step 30189 | loss:0.4089429974555969 lr:6.869928639144219e-06 tokens_per_second_per_gpu:3476.496966313866 mfu:16.39922934164881 +Step 30190 | loss:0.3043360412120819 lr:6.86773517546968e-06 tokens_per_second_per_gpu:5360.978573518203 mfu:25.288650608548537 +Step 30191 | loss:0.3140241503715515 lr:6.865542036198391e-06 tokens_per_second_per_gpu:4801.784978255181 mfu:22.65083901142732 +Step 30192 | loss:0.25191834568977356 lr:6.863349221346871e-06 tokens_per_second_per_gpu:5352.509024874932 mfu:25.24869830254422 +Step 30193 | loss:0.29196715354919434 lr:6.861156730931612e-06 tokens_per_second_per_gpu:4160.583004628823 mfu:19.626179901494083 +Step 30194 | loss:0.24760998785495758 lr:6.858964564969089e-06 tokens_per_second_per_gpu:5346.9638089417995 mfu:25.222540572876234 +Step 30195 | loss:0.3497827649116516 lr:6.856772723475796e-06 tokens_per_second_per_gpu:4973.024088045323 mfu:23.45860310871225 +Step 30196 | loss:0.37756815552711487 lr:6.854581206468236e-06 tokens_per_second_per_gpu:4032.755238575134 mfu:19.023194519353 +Step 30197 | loss:0.35536709427833557 lr:6.852390013962873e-06 tokens_per_second_per_gpu:5321.324934884386 mfu:25.101597629504422 +Step 30198 | loss:0.4903601408004761 lr:6.8501991459761824e-06 tokens_per_second_per_gpu:4777.738401697371 mfu:22.537407207035105 +Step 30199 | loss:0.2652825117111206 lr:6.848008602524647e-06 tokens_per_second_per_gpu:5044.50199140217 mfu:23.79577657423436 +Step 30200 | loss:0.21199700236320496 lr:6.845818383624752e-06 tokens_per_second_per_gpu:5361.423477578601 mfu:25.290749296909894 +Step 30201 | loss:0.2692594528198242 lr:6.843628489292958e-06 tokens_per_second_per_gpu:5320.087665316128 mfu:25.095761217850992 +Step 30202 | loss:0.4023994207382202 lr:6.84143891954574e-06 tokens_per_second_per_gpu:5377.445841554821 mfu:25.366329521482907 +Step 30203 | loss:0.3857172429561615 lr:6.8392496743995786e-06 tokens_per_second_per_gpu:4572.810670882315 mfu:21.57072813650421 +Step 30204 | loss:0.21057040989398956 lr:6.837060753870927e-06 tokens_per_second_per_gpu:3758.4380979434254 mfu:17.729193763662835 +Step 30205 | loss:0.31482675671577454 lr:6.834872157976241e-06 tokens_per_second_per_gpu:5373.301137773793 mfu:25.346778246588414 +Step 30206 | loss:0.40150028467178345 lr:6.832683886731983e-06 tokens_per_second_per_gpu:3640.681797462463 mfu:17.173717202997526 +Step 30207 | loss:0.3675278425216675 lr:6.830495940154635e-06 tokens_per_second_per_gpu:4874.420057769251 mfu:22.99347107431766 +Step 30208 | loss:0.3186700642108917 lr:6.828308318260618e-06 tokens_per_second_per_gpu:3988.140338616067 mfu:18.812738424159214 +Step 30209 | loss:0.243750661611557 lr:6.826121021066406e-06 tokens_per_second_per_gpu:5370.895111025291 mfu:25.33542860790524 +Step 30210 | loss:0.24607989192008972 lr:6.82393404858846e-06 tokens_per_second_per_gpu:5064.379159765851 mfu:23.889540568800477 +Step 30211 | loss:0.36973780393600464 lr:6.821747400843215e-06 tokens_per_second_per_gpu:5364.215070715804 mfu:25.303917718032988 +Step 30212 | loss:0.25259068608283997 lr:6.819561077847103e-06 tokens_per_second_per_gpu:5233.778806350914 mfu:24.688627604302265 +Step 30213 | loss:0.2804854214191437 lr:6.817375079616583e-06 tokens_per_second_per_gpu:3061.132151547073 mfu:14.439882641848714 +Step 30214 | loss:0.2833002507686615 lr:6.81518940616811e-06 tokens_per_second_per_gpu:4114.21916679155 mfu:19.40747376792923 +Step 30215 | loss:0.3681254982948303 lr:6.813004057518091e-06 tokens_per_second_per_gpu:4744.223620539592 mfu:22.379312266102556 +Step 30216 | loss:0.2944280803203583 lr:6.8108190336829855e-06 tokens_per_second_per_gpu:3703.9014434383234 mfu:17.471935059449518 +Step 30217 | loss:0.2675935924053192 lr:6.808634334679226e-06 tokens_per_second_per_gpu:4472.300164549741 mfu:21.09660292927274 +Step 30218 | loss:0.2732771039009094 lr:6.8064499605232314e-06 tokens_per_second_per_gpu:5361.954824167402 mfu:25.293255749426347 +Step 30219 | loss:0.28309306502342224 lr:6.80426591123145e-06 tokens_per_second_per_gpu:5379.925522610659 mfu:25.378026600100508 +Step 30220 | loss:0.3543550670146942 lr:6.802082186820285e-06 tokens_per_second_per_gpu:4951.128813710956 mfu:23.355319363958028 +Step 30221 | loss:0.25521546602249146 lr:6.79989878730618e-06 tokens_per_second_per_gpu:5139.773389441835 mfu:24.245188013764125 +Step 30222 | loss:0.3402658700942993 lr:6.797715712705538e-06 tokens_per_second_per_gpu:5369.851968643269 mfu:25.330507927310794 +Step 30223 | loss:0.3438369035720825 lr:6.79553296303479e-06 tokens_per_second_per_gpu:5333.361527540031 mfu:25.158376290715573 +Step 30224 | loss:0.30663689970970154 lr:6.793350538310355e-06 tokens_per_second_per_gpu:3723.536572264991 mfu:17.56455731222334 +Step 30225 | loss:0.24041442573070526 lr:6.791168438548651e-06 tokens_per_second_per_gpu:3997.7647730068147 mfu:18.858138523278207 +Step 30226 | loss:0.32688912749290466 lr:6.788986663766084e-06 tokens_per_second_per_gpu:5382.256498431043 mfu:25.389022210750746 +Step 30227 | loss:0.21869035065174103 lr:6.786805213979047e-06 tokens_per_second_per_gpu:3217.871478400085 mfu:15.17924878910113 +Step 30228 | loss:0.25904661417007446 lr:6.784624089203967e-06 tokens_per_second_per_gpu:4493.311497560532 mfu:21.19571697198772 +Step 30229 | loss:0.37986162304878235 lr:6.782443289457252e-06 tokens_per_second_per_gpu:3991.137576434586 mfu:18.82687690633068 +Step 30230 | loss:0.3035043478012085 lr:6.780262814755284e-06 tokens_per_second_per_gpu:4421.891878742478 mfu:20.858818444579466 +Step 30231 | loss:0.3686304986476898 lr:6.778082665114477e-06 tokens_per_second_per_gpu:5100.6598654957825 mfu:24.060682847855265 +Step 30232 | loss:0.3336748778820038 lr:6.7759028405512346e-06 tokens_per_second_per_gpu:3676.3121354935947 mfu:17.341791586652906 +Step 30233 | loss:0.3167371153831482 lr:6.773723341081944e-06 tokens_per_second_per_gpu:4245.772080601003 mfu:20.028031307609993 +Step 30234 | loss:0.3597502112388611 lr:6.771544166722987e-06 tokens_per_second_per_gpu:3205.5177581620746 mfu:15.120974182976417 +Step 30235 | loss:0.4029812514781952 lr:6.769365317490756e-06 tokens_per_second_per_gpu:5330.354249205163 mfu:25.14419044571541 +Step 30236 | loss:0.33565661311149597 lr:6.767186793401659e-06 tokens_per_second_per_gpu:4721.791340078393 mfu:22.273495371825366 +Step 30237 | loss:0.3090835511684418 lr:6.765008594472056e-06 tokens_per_second_per_gpu:5037.341496325779 mfu:23.761999297252586 +Step 30238 | loss:0.30718883872032166 lr:6.7628307207183385e-06 tokens_per_second_per_gpu:3863.311177671876 mfu:18.223897973934427 +Step 30239 | loss:0.28250375390052795 lr:6.7606531721569e-06 tokens_per_second_per_gpu:5373.286106028922 mfu:25.346707339283178 +Step 30240 | loss:0.3369896709918976 lr:6.7584759488041106e-06 tokens_per_second_per_gpu:4099.722521151226 mfu:19.33909061706103 +Step 30241 | loss:0.311745285987854 lr:6.756299050676323e-06 tokens_per_second_per_gpu:3612.396305053228 mfu:17.040289709300517 +Step 30242 | loss:0.29085177183151245 lr:6.7541224777899314e-06 tokens_per_second_per_gpu:5161.6532568843495 mfu:24.348399081580798 +Step 30243 | loss:0.3065004348754883 lr:6.751946230161316e-06 tokens_per_second_per_gpu:5039.522323663754 mfu:23.772286631893326 +Step 30244 | loss:0.344937264919281 lr:6.7497703078068185e-06 tokens_per_second_per_gpu:5160.431755069377 mfu:24.342637049109868 +Step 30245 | loss:0.3432236313819885 lr:6.747594710742816e-06 tokens_per_second_per_gpu:5297.951335276382 mfu:24.991340372206153 +Step 30246 | loss:0.3346421420574188 lr:6.745419438985684e-06 tokens_per_second_per_gpu:5340.790919440922 mfu:25.193421999897044 +Step 30247 | loss:0.3853387236595154 lr:6.743244492551765e-06 tokens_per_second_per_gpu:4202.475779204574 mfu:19.823795266812287 +Step 30248 | loss:0.31301626563072205 lr:6.7410698714574285e-06 tokens_per_second_per_gpu:4546.184654252059 mfu:21.445128673192407 +Step 30249 | loss:0.283295214176178 lr:6.738895575719017e-06 tokens_per_second_per_gpu:4426.243349542127 mfu:20.879345074778886 +Step 30250 | loss:0.3887227475643158 lr:6.736721605352908e-06 tokens_per_second_per_gpu:3516.4515385169466 mfu:16.5877018756837 +Step 30251 | loss:0.29078564047813416 lr:6.73454796037542e-06 tokens_per_second_per_gpu:4881.6085348714105 mfu:23.027380347289213 +Step 30252 | loss:0.2099892497062683 lr:6.732374640802919e-06 tokens_per_second_per_gpu:4693.814553641298 mfu:22.141524096869613 +Step 30253 | loss:0.3474120497703552 lr:6.730201646651757e-06 tokens_per_second_per_gpu:5333.875472062467 mfu:25.16080065471588 +Step 30254 | loss:0.295732706785202 lr:6.728028977938261e-06 tokens_per_second_per_gpu:4149.339862314331 mfu:19.57314408091874 +Step 30255 | loss:0.21830543875694275 lr:6.72585663467879e-06 tokens_per_second_per_gpu:4321.683071686532 mfu:20.386116404311053 +Step 30256 | loss:0.2769114077091217 lr:6.723684616889664e-06 tokens_per_second_per_gpu:5324.132592398657 mfu:25.114841829035008 +Step 30257 | loss:0.3252992331981659 lr:6.721512924587242e-06 tokens_per_second_per_gpu:3769.621328206339 mfu:17.781946968868834 +Step 30258 | loss:0.32856351137161255 lr:6.719341557787828e-06 tokens_per_second_per_gpu:5146.627953709159 mfu:24.27752216292235 +Step 30259 | loss:0.35014328360557556 lr:6.71717051650777e-06 tokens_per_second_per_gpu:5002.103564457563 mfu:23.595775960419076 +Step 30260 | loss:0.34659543633461 lr:6.714999800763394e-06 tokens_per_second_per_gpu:4349.597360769718 mfu:20.517792868585275 +Step 30261 | loss:0.27855807542800903 lr:6.712829410571042e-06 tokens_per_second_per_gpu:5220.659503935757 mfu:24.62674161642616 +Step 30262 | loss:0.29588305950164795 lr:6.710659345947024e-06 tokens_per_second_per_gpu:4737.88134242111 mfu:22.349394658113447 +Step 30263 | loss:0.359365314245224 lr:6.7084896069076475e-06 tokens_per_second_per_gpu:3739.985903047519 mfu:17.642151612069757 +Step 30264 | loss:0.31944459676742554 lr:6.706320193469246e-06 tokens_per_second_per_gpu:4650.67697318042 mfu:21.938036769805485 +Step 30265 | loss:0.3145805895328522 lr:6.704151105648143e-06 tokens_per_second_per_gpu:5369.112121282556 mfu:25.327017941079333 +Step 30266 | loss:0.33562585711479187 lr:6.701982343460633e-06 tokens_per_second_per_gpu:4512.120486645267 mfu:21.28444218264487 +Step 30267 | loss:0.27957212924957275 lr:6.6998139069230415e-06 tokens_per_second_per_gpu:5324.8787959123965 mfu:25.118361798324763 +Step 30268 | loss:0.3477645218372345 lr:6.697645796051688e-06 tokens_per_second_per_gpu:5340.801742337144 mfu:25.1934730533451 +Step 30269 | loss:0.292706698179245 lr:6.69547801086286e-06 tokens_per_second_per_gpu:5223.457129398695 mfu:24.63993849305896 +Step 30270 | loss:0.3921985626220703 lr:6.693310551372861e-06 tokens_per_second_per_gpu:5352.389255896109 mfu:25.248133331836417 +Step 30271 | loss:0.2033868134021759 lr:6.691143417597995e-06 tokens_per_second_per_gpu:4655.7820186239305 mfu:21.962118140173093 +Step 30272 | loss:0.2841828763484955 lr:6.688976609554581e-06 tokens_per_second_per_gpu:4254.682358818506 mfu:20.070062610212187 +Step 30273 | loss:0.370015412569046 lr:6.6868101272588855e-06 tokens_per_second_per_gpu:3497.9854789502915 mfu:16.50059432207314 +Step 30274 | loss:0.2648945748806 lr:6.684643970727222e-06 tokens_per_second_per_gpu:5309.824955064323 mfu:25.047350262594307 +Step 30275 | loss:0.32384300231933594 lr:6.682478139975884e-06 tokens_per_second_per_gpu:3690.98589687097 mfu:17.411010222671838 +Step 30276 | loss:0.33600592613220215 lr:6.680312635021163e-06 tokens_per_second_per_gpu:5349.594227052402 mfu:25.234948703899555 +Step 30277 | loss:0.24773269891738892 lr:6.678147455879319e-06 tokens_per_second_per_gpu:4653.755173773675 mfu:21.952557167199046 +Step 30278 | loss:0.3574497699737549 lr:6.67598260256666e-06 tokens_per_second_per_gpu:4864.187412837857 mfu:22.945201942307023 +Step 30279 | loss:0.3706642985343933 lr:6.673818075099475e-06 tokens_per_second_per_gpu:4237.274057050614 mfu:19.987944680610482 +Step 30280 | loss:0.27548009157180786 lr:6.671653873494022e-06 tokens_per_second_per_gpu:5341.299375402398 mfu:25.19582047341886 +Step 30281 | loss:0.3835691511631012 lr:6.669489997766587e-06 tokens_per_second_per_gpu:4493.230052425497 mfu:21.195332781389784 +Step 30282 | loss:0.41557690501213074 lr:6.667326447933459e-06 tokens_per_second_per_gpu:5123.14042760055 mfu:24.16672749488286 +Step 30283 | loss:0.31547826528549194 lr:6.66516322401089e-06 tokens_per_second_per_gpu:5337.7185311292205 mfu:25.178929020027248 +Step 30284 | loss:0.3199257254600525 lr:6.663000326015162e-06 tokens_per_second_per_gpu:4953.643842237518 mfu:23.367183182625933 +Step 30285 | loss:0.2747635543346405 lr:6.660837753962534e-06 tokens_per_second_per_gpu:4475.516619698321 mfu:21.111775496992433 +Step 30286 | loss:0.30433204770088196 lr:6.658675507869283e-06 tokens_per_second_per_gpu:4734.415460639114 mfu:22.333045502407362 +Step 30287 | loss:0.3108540177345276 lr:6.656513587751656e-06 tokens_per_second_per_gpu:5164.427650672664 mfu:24.361486370440666 +Step 30288 | loss:0.4010171890258789 lr:6.654351993625918e-06 tokens_per_second_per_gpu:4345.9924338883975 mfu:20.500787813422136 +Step 30289 | loss:0.32309606671333313 lr:6.652190725508339e-06 tokens_per_second_per_gpu:5175.016353647205 mfu:24.411435088990597 +Step 30290 | loss:0.37222981452941895 lr:6.650029783415157e-06 tokens_per_second_per_gpu:5119.480669191553 mfu:24.14946379004845 +Step 30291 | loss:0.31059619784355164 lr:6.647869167362641e-06 tokens_per_second_per_gpu:5386.918182455217 mfu:25.411012169658086 +Step 30292 | loss:0.36571186780929565 lr:6.645708877367024e-06 tokens_per_second_per_gpu:5323.549645585177 mfu:25.112091969454845 +Step 30293 | loss:0.2993429899215698 lr:6.643548913444575e-06 tokens_per_second_per_gpu:5265.243624833754 mfu:24.83705252153743 +Step 30294 | loss:0.2880905866622925 lr:6.641389275611514e-06 tokens_per_second_per_gpu:5227.576887827905 mfu:24.65937209647282 +Step 30295 | loss:0.23966668546199799 lr:6.639229963884097e-06 tokens_per_second_per_gpu:4298.466513813188 mfu:20.276599939669257 +Step 30296 | loss:0.29499873518943787 lr:6.6370709782785645e-06 tokens_per_second_per_gpu:4466.876884312281 mfu:21.071020391085618 +Step 30297 | loss:0.34847843647003174 lr:6.634912318811165e-06 tokens_per_second_per_gpu:4233.291736838014 mfu:19.969159396713756 +Step 30298 | loss:0.31921982765197754 lr:6.6327539854981194e-06 tokens_per_second_per_gpu:4383.654828912068 mfu:20.678447756615995 +Step 30299 | loss:0.25314974784851074 lr:6.630595978355658e-06 tokens_per_second_per_gpu:5209.172537099353 mfu:24.5725556339802 +Step 30300 | loss:0.3061589300632477 lr:6.62843829740003e-06 tokens_per_second_per_gpu:3663.248887683197 mfu:17.280170017911356 +Step 30301 | loss:0.32675597071647644 lr:6.626280942647439e-06 tokens_per_second_per_gpu:5366.564228680939 mfu:25.314999096962204 +Step 30302 | loss:0.26768073439598083 lr:6.624123914114122e-06 tokens_per_second_per_gpu:4657.158858018124 mfu:21.96861292650853 +Step 30303 | loss:0.25761881470680237 lr:6.62196721181631e-06 tokens_per_second_per_gpu:5380.3995634395105 mfu:25.38026273156991 +Step 30304 | loss:0.23610293865203857 lr:6.619810835770224e-06 tokens_per_second_per_gpu:4398.184739577565 mfu:20.746987824283536 +Step 30305 | loss:0.31472665071487427 lr:6.617654785992078e-06 tokens_per_second_per_gpu:4128.36787529603 mfu:19.47421563996346 +Step 30306 | loss:0.21508274972438812 lr:6.61549906249807e-06 tokens_per_second_per_gpu:3136.757490995581 mfu:14.796620271040752 +Step 30307 | loss:0.33358705043792725 lr:6.613343665304433e-06 tokens_per_second_per_gpu:4072.897550047708 mfu:19.212552651553718 +Step 30308 | loss:0.3107856810092926 lr:6.611188594427387e-06 tokens_per_second_per_gpu:5138.652604354759 mfu:24.239901079282735 +Step 30309 | loss:0.2674926817417145 lr:6.6090338498831174e-06 tokens_per_second_per_gpu:4707.717313037121 mfu:22.207105784995065 +Step 30310 | loss:0.27258971333503723 lr:6.606879431687835e-06 tokens_per_second_per_gpu:5369.750841503201 mfu:25.330030893337025 +Step 30311 | loss:0.3077124357223511 lr:6.604725339857765e-06 tokens_per_second_per_gpu:3822.523375297313 mfu:18.031494951016928 +Step 30312 | loss:0.3420710861682892 lr:6.6025715744090836e-06 tokens_per_second_per_gpu:4508.775409574968 mfu:21.268662883375267 +Step 30313 | loss:0.2303839772939682 lr:6.600418135358005e-06 tokens_per_second_per_gpu:3193.7047221941966 mfu:15.065250076804167 +Step 30314 | loss:0.282630980014801 lr:6.598265022720712e-06 tokens_per_second_per_gpu:2791.6265276530926 mfu:13.168578631539818 +Step 30315 | loss:0.24380511045455933 lr:6.596112236513419e-06 tokens_per_second_per_gpu:5309.557598853109 mfu:25.046089097730334 +Step 30316 | loss:0.31390222907066345 lr:6.5939597767522894e-06 tokens_per_second_per_gpu:4020.7698491183564 mfu:18.96665739236708 +Step 30317 | loss:0.3096514344215393 lr:6.591807643453529e-06 tokens_per_second_per_gpu:3558.270987554078 mfu:16.784971351925606 +Step 30318 | loss:0.3613305389881134 lr:6.589655836633329e-06 tokens_per_second_per_gpu:4084.0422689194006 mfu:19.265124191956605 +Step 30319 | loss:0.23609887063503265 lr:6.587504356307861e-06 tokens_per_second_per_gpu:4475.135963950098 mfu:21.109979878881497 +Step 30320 | loss:0.33018526434898376 lr:6.585353202493322e-06 tokens_per_second_per_gpu:4294.879342298378 mfu:20.25967863959933 +Step 30321 | loss:0.29076364636421204 lr:6.583202375205866e-06 tokens_per_second_per_gpu:4455.543330592566 mfu:21.017558084486062 +Step 30322 | loss:0.43821215629577637 lr:6.5810518744617025e-06 tokens_per_second_per_gpu:4570.446891745399 mfu:21.55957778701307 +Step 30323 | loss:0.2568925619125366 lr:6.5789017002769735e-06 tokens_per_second_per_gpu:5344.5292254943115 mfu:25.211056227371664 +Step 30324 | loss:0.305480420589447 lr:6.576751852667867e-06 tokens_per_second_per_gpu:4162.86932967081 mfu:19.636964886804414 +Step 30325 | loss:0.2983522117137909 lr:6.574602331650559e-06 tokens_per_second_per_gpu:4091.906301167759 mfu:19.30222017381437 +Step 30326 | loss:0.22800496220588684 lr:6.572453137241197e-06 tokens_per_second_per_gpu:3703.548811933255 mfu:17.470271636474973 +Step 30327 | loss:0.36564627289772034 lr:6.570304269455968e-06 tokens_per_second_per_gpu:3689.1204370614596 mfu:17.402210530470114 +Step 30328 | loss:0.29747453331947327 lr:6.568155728311015e-06 tokens_per_second_per_gpu:4413.50021595985 mfu:20.8192335394685 +Step 30329 | loss:0.2880151569843292 lr:6.566007513822514e-06 tokens_per_second_per_gpu:5306.294240454134 mfu:25.030695279376012 +Step 30330 | loss:0.34390848875045776 lr:6.5638596260065965e-06 tokens_per_second_per_gpu:4017.187809449006 mfu:18.949760300087764 +Step 30331 | loss:0.38974276185035706 lr:6.561712064879439e-06 tokens_per_second_per_gpu:4483.548411766034 mfu:21.14966283054087 +Step 30332 | loss:0.27807337045669556 lr:6.559564830457188e-06 tokens_per_second_per_gpu:5366.123846382278 mfu:25.312921738522455 +Step 30333 | loss:0.313607782125473 lr:6.557417922755999e-06 tokens_per_second_per_gpu:4420.843311866702 mfu:20.853872175722472 +Step 30334 | loss:0.2754702568054199 lr:6.555271341792019e-06 tokens_per_second_per_gpu:5339.190037235024 mfu:25.185870365393153 +Step 30335 | loss:0.2964303493499756 lr:6.553125087581369e-06 tokens_per_second_per_gpu:4261.945083481495 mfu:20.10432212159611 +Step 30336 | loss:0.27789756655693054 lr:6.550979160140225e-06 tokens_per_second_per_gpu:5307.368660869711 mfu:25.035763503791312 +Step 30337 | loss:0.3244929313659668 lr:6.548833559484702e-06 tokens_per_second_per_gpu:3219.3847998285305 mfu:15.186387384478367 +Step 30338 | loss:0.3152793347835541 lr:6.546688285630942e-06 tokens_per_second_per_gpu:5339.565880880378 mfu:25.18764328399383 +Step 30339 | loss:0.2678985297679901 lr:6.544543338595083e-06 tokens_per_second_per_gpu:3705.2421256914654 mfu:17.47825928638157 +Step 30340 | loss:0.3056836426258087 lr:6.542398718393272e-06 tokens_per_second_per_gpu:3528.973693726394 mfu:16.646771018306648 +Step 30341 | loss:0.3098949193954468 lr:6.540254425041614e-06 tokens_per_second_per_gpu:5326.30041326162 mfu:25.125067809913983 +Step 30342 | loss:0.26100897789001465 lr:6.538110458556262e-06 tokens_per_second_per_gpu:4175.193484782702 mfu:19.695100029184896 +Step 30343 | loss:0.3473377823829651 lr:6.53596681895331e-06 tokens_per_second_per_gpu:3782.4389101970614 mfu:17.842409690024066 +Step 30344 | loss:0.20923225581645966 lr:6.533823506248915e-06 tokens_per_second_per_gpu:3212.6497258415325 mfu:15.154616891359584 +Step 30345 | loss:0.29131555557250977 lr:6.531680520459166e-06 tokens_per_second_per_gpu:4145.124642906334 mfu:19.553260171780625 +Step 30346 | loss:0.3250100016593933 lr:6.529537861600193e-06 tokens_per_second_per_gpu:4599.529408652191 mfu:21.696764981251455 +Step 30347 | loss:0.34024959802627563 lr:6.527395529688124e-06 tokens_per_second_per_gpu:3158.597383644413 mfu:14.899642770935221 +Step 30348 | loss:0.3674609363079071 lr:6.52525352473905e-06 tokens_per_second_per_gpu:4290.083737277957 mfu:20.237056952503412 +Step 30349 | loss:0.36679741740226746 lr:6.523111846769104e-06 tokens_per_second_per_gpu:3508.7568741620125 mfu:16.55140483107636 +Step 30350 | loss:0.3250434398651123 lr:6.520970495794365e-06 tokens_per_second_per_gpu:5043.503284151823 mfu:23.79106550173745 +Step 30351 | loss:0.3033299446105957 lr:6.518829471830973e-06 tokens_per_second_per_gpu:5336.431717317743 mfu:25.172858899725437 +Step 30352 | loss:0.24667656421661377 lr:6.516688774894997e-06 tokens_per_second_per_gpu:4356.697586285886 mfu:20.55128584376843 +Step 30353 | loss:0.26344120502471924 lr:6.5145484050025515e-06 tokens_per_second_per_gpu:5321.573698338923 mfu:25.1027710891629 +Step 30354 | loss:0.40000027418136597 lr:6.512408362169753e-06 tokens_per_second_per_gpu:4780.37614772161 mfu:22.549849905077323 +Step 30355 | loss:0.2984291613101959 lr:6.510268646412665e-06 tokens_per_second_per_gpu:4830.945638077303 mfu:22.788394819129874 +Step 30356 | loss:0.26283955574035645 lr:6.5081292577474085e-06 tokens_per_second_per_gpu:5330.086533505118 mfu:25.142927585081285 +Step 30357 | loss:0.2895854413509369 lr:6.5059901961900535e-06 tokens_per_second_per_gpu:4685.920441417326 mfu:22.10428622263471 +Step 30358 | loss:0.2732089161872864 lr:6.503851461756705e-06 tokens_per_second_per_gpu:4529.974858816511 mfu:21.368664302446643 +Step 30359 | loss:0.4274403750896454 lr:6.501713054463426e-06 tokens_per_second_per_gpu:3895.7561479171472 mfu:18.376946434264116 +Step 30360 | loss:0.27206307649612427 lr:6.4995749743263165e-06 tokens_per_second_per_gpu:3859.200256478121 mfu:18.204506057268002 +Step 30361 | loss:0.36202484369277954 lr:6.497437221361463e-06 tokens_per_second_per_gpu:4142.777646926489 mfu:19.542188991304894 +Step 30362 | loss:0.3384109139442444 lr:6.495299795584925e-06 tokens_per_second_per_gpu:5216.6762466141 mfu:24.607951912026127 +Step 30363 | loss:0.28605329990386963 lr:6.4931626970128e-06 tokens_per_second_per_gpu:3699.046788596764 mfu:17.449034824272356 +Step 30364 | loss:0.3538251519203186 lr:6.491025925661142e-06 tokens_per_second_per_gpu:5017.156276218665 mfu:23.66678217005365 +Step 30365 | loss:0.16833579540252686 lr:6.488889481546034e-06 tokens_per_second_per_gpu:3880.6784317383776 mfu:18.30582227452521 +Step 30366 | loss:0.4412837028503418 lr:6.4867533646835345e-06 tokens_per_second_per_gpu:3947.20674100294 mfu:18.61964766022618 +Step 30367 | loss:0.36319100856781006 lr:6.484617575089713e-06 tokens_per_second_per_gpu:4300.660190397103 mfu:20.286947886395406 +Step 30368 | loss:0.2733665108680725 lr:6.482482112780646e-06 tokens_per_second_per_gpu:5285.720987350436 mfu:24.933647734326932 +Step 30369 | loss:0.37285545468330383 lr:6.480346977772378e-06 tokens_per_second_per_gpu:5352.2106334499495 mfu:25.247290739281752 +Step 30370 | loss:0.3444983661174774 lr:6.478212170080977e-06 tokens_per_second_per_gpu:5342.868522595176 mfu:25.203222408451307 +Step 30371 | loss:0.29897990822792053 lr:6.476077689722487e-06 tokens_per_second_per_gpu:5387.01912750509 mfu:25.411488344681338 +Step 30372 | loss:0.3685857951641083 lr:6.473943536712984e-06 tokens_per_second_per_gpu:5276.32439712692 mfu:24.889322415019 +Step 30373 | loss:0.24004794657230377 lr:6.471809711068494e-06 tokens_per_second_per_gpu:4974.519367896398 mfu:23.465656598890824 +Step 30374 | loss:0.24502018094062805 lr:6.469676212805076e-06 tokens_per_second_per_gpu:4341.737513042116 mfu:20.480716625825867 +Step 30375 | loss:0.25842195749282837 lr:6.4675430419387786e-06 tokens_per_second_per_gpu:4523.897061532297 mfu:21.339994295677858 +Step 30376 | loss:0.2782031297683716 lr:6.4654101984856556e-06 tokens_per_second_per_gpu:5185.300122162375 mfu:24.459945379668493 +Step 30377 | loss:0.29217737913131714 lr:6.463277682461727e-06 tokens_per_second_per_gpu:4935.676670597763 mfu:23.282428968485622 +Step 30378 | loss:0.2558181881904602 lr:6.461145493883053e-06 tokens_per_second_per_gpu:5320.573493149004 mfu:25.098052950629448 +Step 30379 | loss:0.24451501667499542 lr:6.459013632765648e-06 tokens_per_second_per_gpu:3571.8498890519845 mfu:16.849025347090727 +Step 30380 | loss:0.29973024129867554 lr:6.456882099125566e-06 tokens_per_second_per_gpu:4662.575426623155 mfu:21.99416380478036 +Step 30381 | loss:0.2444116324186325 lr:6.4547508929788226e-06 tokens_per_second_per_gpu:5144.032150203096 mfu:24.265277314894504 +Step 30382 | loss:0.17686966061592102 lr:6.452620014341454e-06 tokens_per_second_per_gpu:3999.3849093287376 mfu:18.86578098273248 +Step 30383 | loss:0.35479187965393066 lr:6.4504894632294925e-06 tokens_per_second_per_gpu:4893.388578845952 mfu:23.082948824599878 +Step 30384 | loss:0.32243508100509644 lr:6.448359239658952e-06 tokens_per_second_per_gpu:4205.589574068936 mfu:19.83848356845567 +Step 30385 | loss:0.3265272080898285 lr:6.446229343645866e-06 tokens_per_second_per_gpu:5023.4504955881885 mfu:23.696473076724256 +Step 30386 | loss:0.3551953136920929 lr:6.444099775206236e-06 tokens_per_second_per_gpu:4127.512982443619 mfu:19.470182964518877 +Step 30387 | loss:0.37888023257255554 lr:6.441970534356107e-06 tokens_per_second_per_gpu:3776.6645371182553 mfu:17.8151709870017 +Step 30388 | loss:0.34251073002815247 lr:6.439841621111458e-06 tokens_per_second_per_gpu:5366.490899460224 mfu:25.314653190516836 +Step 30389 | loss:0.34057965874671936 lr:6.437713035488319e-06 tokens_per_second_per_gpu:4278.718841317574 mfu:20.183446799207086 +Step 30390 | loss:0.36389127373695374 lr:6.435584777502712e-06 tokens_per_second_per_gpu:3834.5723700770573 mfu:18.088332115163475 +Step 30391 | loss:0.35963767766952515 lr:6.4334568471706155e-06 tokens_per_second_per_gpu:3971.039997841487 mfu:18.73207319910653 +Step 30392 | loss:0.4028909504413605 lr:6.431329244508067e-06 tokens_per_second_per_gpu:3135.6324830099543 mfu:14.791313416426396 +Step 30393 | loss:0.30549168586730957 lr:6.429201969531035e-06 tokens_per_second_per_gpu:4241.266439226474 mfu:20.006777428504655 +Step 30394 | loss:0.35182932019233704 lr:6.4270750222555465e-06 tokens_per_second_per_gpu:5233.611537140676 mfu:24.6878385668985 +Step 30395 | loss:0.4317341446876526 lr:6.424948402697572e-06 tokens_per_second_per_gpu:4263.584604818688 mfu:20.112056023474832 +Step 30396 | loss:0.23365351557731628 lr:6.422822110873128e-06 tokens_per_second_per_gpu:4763.742329631396 mfu:22.47138534712377 +Step 30397 | loss:0.36389362812042236 lr:6.420696146798205e-06 tokens_per_second_per_gpu:5053.988022176325 mfu:23.840523799879794 +Step 30398 | loss:0.2896120250225067 lr:6.418570510488775e-06 tokens_per_second_per_gpu:3526.903053093741 mfu:16.637003453154705 +Step 30399 | loss:0.22427256405353546 lr:6.416445201960852e-06 tokens_per_second_per_gpu:4009.5372499879127 mfu:18.913671305789745 +Step 30400 | loss:0.3042680025100708 lr:6.4143202212303965e-06 tokens_per_second_per_gpu:4319.307769003271 mfu:20.374911696285686 +Step 30401 | loss:0.30006173253059387 lr:6.412195568313406e-06 tokens_per_second_per_gpu:5334.275688345089 mfu:25.16268854320495 +Step 30402 | loss:0.2727116048336029 lr:6.410071243225846e-06 tokens_per_second_per_gpu:5357.0728459653565 mfu:25.270226625295923 +Step 30403 | loss:0.3423248529434204 lr:6.407947245983703e-06 tokens_per_second_per_gpu:3713.504574357276 mfu:17.517234666457306 +Step 30404 | loss:0.25409695506095886 lr:6.405823576602959e-06 tokens_per_second_per_gpu:4510.833747207019 mfu:21.278372413174264 +Step 30405 | loss:0.30112141370773315 lr:6.403700235099574e-06 tokens_per_second_per_gpu:5313.145936474785 mfu:25.06301589852531 +Step 30406 | loss:0.2463342249393463 lr:6.401577221489519e-06 tokens_per_second_per_gpu:4447.225998196319 mfu:20.978323808489726 +Step 30407 | loss:0.2493343949317932 lr:6.399454535788773e-06 tokens_per_second_per_gpu:4591.397308079469 mfu:21.658404475370496 +Step 30408 | loss:0.21175526082515717 lr:6.3973321780132935e-06 tokens_per_second_per_gpu:4713.748735910401 mfu:22.23555703575518 +Step 30409 | loss:0.3398391902446747 lr:6.3952101481790375e-06 tokens_per_second_per_gpu:5344.499567757535 mfu:25.21091632676716 +Step 30410 | loss:0.39638954401016235 lr:6.393088446301964e-06 tokens_per_second_per_gpu:5172.135144181044 mfu:24.39784393235369 +Step 30411 | loss:0.29581212997436523 lr:6.3909670723980406e-06 tokens_per_second_per_gpu:4497.614397889585 mfu:21.21601448699036 +Step 30412 | loss:0.3207060694694519 lr:6.388846026483225e-06 tokens_per_second_per_gpu:5339.383374798069 mfu:25.186782371664826 +Step 30413 | loss:0.35127556324005127 lr:6.386725308573455e-06 tokens_per_second_per_gpu:4017.925435716113 mfu:18.953239809041726 +Step 30414 | loss:0.353548139333725 lr:6.3846049186847e-06 tokens_per_second_per_gpu:3960.2597887972747 mfu:18.68122111375166 +Step 30415 | loss:0.2499813735485077 lr:6.382484856832882e-06 tokens_per_second_per_gpu:4436.429881421983 mfu:20.92739668365772 +Step 30416 | loss:0.2587336003780365 lr:6.380365123033977e-06 tokens_per_second_per_gpu:4441.700428158514 mfu:20.95225875186194 +Step 30417 | loss:0.28925850987434387 lr:6.378245717303899e-06 tokens_per_second_per_gpu:5251.782877852651 mfu:24.77355588138731 +Step 30418 | loss:0.25111740827560425 lr:6.376126639658602e-06 tokens_per_second_per_gpu:4496.66737742795 mfu:21.211547229894474 +Step 30419 | loss:0.39385679364204407 lr:6.374007890114031e-06 tokens_per_second_per_gpu:4320.207374928912 mfu:20.37915529092553 +Step 30420 | loss:0.23748840391635895 lr:6.371889468686104e-06 tokens_per_second_per_gpu:3488.664129408787 mfu:16.45662392589964 +Step 30421 | loss:0.34689533710479736 lr:6.3697713753907765e-06 tokens_per_second_per_gpu:5220.484470155438 mfu:24.625915952220556 +Step 30422 | loss:0.39483484625816345 lr:6.367653610243951e-06 tokens_per_second_per_gpu:5324.514122118599 mfu:25.116641569819656 +Step 30423 | loss:0.40326806902885437 lr:6.3655361732615825e-06 tokens_per_second_per_gpu:4318.191740219735 mfu:20.369647198099603 +Step 30424 | loss:0.30313557386398315 lr:6.363419064459575e-06 tokens_per_second_per_gpu:5001.862103391555 mfu:23.59463694737321 +Step 30425 | loss:0.2934541702270508 lr:6.3613022838538595e-06 tokens_per_second_per_gpu:3953.5422793160096 mfu:18.649533475403228 +Step 30426 | loss:0.31501027941703796 lr:6.359185831460368e-06 tokens_per_second_per_gpu:5332.258947990419 mfu:25.153175234860587 +Step 30427 | loss:0.30554795265197754 lr:6.3570697072949995e-06 tokens_per_second_per_gpu:5262.912916071156 mfu:24.826058170644334 +Step 30428 | loss:0.3180105984210968 lr:6.35495391137369e-06 tokens_per_second_per_gpu:5358.93675179746 mfu:25.279018987121084 +Step 30429 | loss:0.39322832226753235 lr:6.352838443712328e-06 tokens_per_second_per_gpu:5355.454706279104 mfu:25.26259358431277 +Step 30430 | loss:0.27516064047813416 lr:6.3507233043268495e-06 tokens_per_second_per_gpu:4697.775536181396 mfu:22.160208727324566 +Step 30431 | loss:0.3856588304042816 lr:6.348608493233143e-06 tokens_per_second_per_gpu:5277.847575813326 mfu:24.896507508763857 +Step 30432 | loss:0.30313897132873535 lr:6.346494010447118e-06 tokens_per_second_per_gpu:3995.9790044525434 mfu:18.84971474832418 +Step 30433 | loss:0.3554569184780121 lr:6.3443798559846945e-06 tokens_per_second_per_gpu:5352.519909972638 mfu:25.248749649407287 +Step 30434 | loss:0.35227301716804504 lr:6.342266029861749e-06 tokens_per_second_per_gpu:4935.59788609618 mfu:23.282057328549634 +Step 30435 | loss:0.25317147374153137 lr:6.34015253209419e-06 tokens_per_second_per_gpu:5319.4163155998385 mfu:25.092594346695183 +Step 30436 | loss:0.2550867199897766 lr:6.3380393626979225e-06 tokens_per_second_per_gpu:5042.494440954433 mfu:23.786306616248957 +Step 30437 | loss:0.19472138583660126 lr:6.33592652168884e-06 tokens_per_second_per_gpu:4104.88403479977 mfu:19.363438357586247 +Step 30438 | loss:0.3422900140285492 lr:6.333814009082812e-06 tokens_per_second_per_gpu:5344.904933099997 mfu:25.21282850425001 +Step 30439 | loss:0.35543709993362427 lr:6.331701824895736e-06 tokens_per_second_per_gpu:4680.060102352891 mfu:22.07664200339082 +Step 30440 | loss:0.3332509696483612 lr:6.329589969143518e-06 tokens_per_second_per_gpu:5364.904026454879 mfu:25.307167639802174 +Step 30441 | loss:0.21110288798809052 lr:6.327478441842016e-06 tokens_per_second_per_gpu:5344.8013528122765 mfu:25.212339898360213 +Step 30442 | loss:0.2954294979572296 lr:6.325367243007119e-06 tokens_per_second_per_gpu:5324.596758160267 mfu:25.11703137812737 +Step 30443 | loss:0.359264075756073 lr:6.323256372654723e-06 tokens_per_second_per_gpu:4887.816406439997 mfu:23.05666393665034 +Step 30444 | loss:0.34493422508239746 lr:6.321145830800684e-06 tokens_per_second_per_gpu:5345.131962798506 mfu:25.213899442073142 +Step 30445 | loss:0.27153053879737854 lr:6.319035617460872e-06 tokens_per_second_per_gpu:4226.137434570155 mfu:19.93541133226643 +Step 30446 | loss:0.23072762787342072 lr:6.316925732651169e-06 tokens_per_second_per_gpu:3771.283921978837 mfu:17.789789707361457 +Step 30447 | loss:0.23549577593803406 lr:6.314816176387439e-06 tokens_per_second_per_gpu:5337.537292413021 mfu:25.17807408608042 +Step 30448 | loss:0.3190440833568573 lr:6.31270694868556e-06 tokens_per_second_per_gpu:5350.859208557559 mfu:25.24091583748748 +Step 30449 | loss:0.2830158472061157 lr:6.310598049561378e-06 tokens_per_second_per_gpu:4019.2492740710086 mfu:18.959484580431884 +Step 30450 | loss:0.22218938171863556 lr:6.308489479030771e-06 tokens_per_second_per_gpu:5340.353142093651 mfu:25.19135692945745 +Step 30451 | loss:0.26378872990608215 lr:6.306381237109582e-06 tokens_per_second_per_gpu:5362.624621101404 mfu:25.296415295843175 +Step 30452 | loss:0.3424178659915924 lr:6.304273323813687e-06 tokens_per_second_per_gpu:5398.940670180579 mfu:25.467724295506542 +Step 30453 | loss:0.3307059705257416 lr:6.302165739158916e-06 tokens_per_second_per_gpu:5319.529474901845 mfu:25.093128138429716 +Step 30454 | loss:0.27697494626045227 lr:6.3000584831611305e-06 tokens_per_second_per_gpu:4026.658117063167 mfu:18.994433356904857 +Step 30455 | loss:0.38059037923812866 lr:6.29795155583619e-06 tokens_per_second_per_gpu:4470.013116642724 mfu:21.085814534084733 +Step 30456 | loss:0.33565038442611694 lr:6.2958449571999255e-06 tokens_per_second_per_gpu:3416.814662150235 mfu:16.11769773005785 +Step 30457 | loss:0.405581533908844 lr:6.293738687268197e-06 tokens_per_second_per_gpu:3852.1694892359787 mfu:18.17134072861423 +Step 30458 | loss:0.3031010925769806 lr:6.29163274605683e-06 tokens_per_second_per_gpu:5258.336133688305 mfu:24.80446870726483 +Step 30459 | loss:0.3395148813724518 lr:6.289527133581674e-06 tokens_per_second_per_gpu:3539.274997552662 mfu:16.695364026038828 +Step 30460 | loss:0.34112611413002014 lr:6.287421849858554e-06 tokens_per_second_per_gpu:5002.300828947062 mfu:23.596706490672833 +Step 30461 | loss:0.28138384222984314 lr:6.285316894903315e-06 tokens_per_second_per_gpu:3985.676211501538 mfu:18.801114716136546 +Step 30462 | loss:0.38803762197494507 lr:6.283212268731797e-06 tokens_per_second_per_gpu:5305.011862960346 mfu:25.02464608594165 +Step 30463 | loss:0.2575925588607788 lr:6.281107971359801e-06 tokens_per_second_per_gpu:4887.9354678214095 mfu:23.05722556950043 +Step 30464 | loss:0.3615339696407318 lr:6.279004002803174e-06 tokens_per_second_per_gpu:4422.087581050802 mfu:20.85974160575809 +Step 30465 | loss:0.4072275459766388 lr:6.276900363077748e-06 tokens_per_second_per_gpu:5342.3354472148885 mfu:25.200707800930456 +Step 30466 | loss:0.47555774450302124 lr:6.274797052199333e-06 tokens_per_second_per_gpu:5328.903966971851 mfu:25.137349216977213 +Step 30467 | loss:0.33359190821647644 lr:6.2726940701837375e-06 tokens_per_second_per_gpu:5298.26043666444 mfu:24.992798456191757 +Step 30468 | loss:0.33045071363449097 lr:6.2705914170467895e-06 tokens_per_second_per_gpu:5324.773606650523 mfu:25.117865602629372 +Step 30469 | loss:0.3380943238735199 lr:6.268489092804314e-06 tokens_per_second_per_gpu:5372.721518865138 mfu:25.344044085303107 +Step 30470 | loss:0.3786711096763611 lr:6.266387097472104e-06 tokens_per_second_per_gpu:4779.267943172676 mfu:22.544622315140806 +Step 30471 | loss:0.2711491882801056 lr:6.264285431065975e-06 tokens_per_second_per_gpu:4529.535143105627 mfu:21.366590088415162 +Step 30472 | loss:0.30573388934135437 lr:6.262184093601747e-06 tokens_per_second_per_gpu:4216.597251961451 mfu:19.89040866317817 +Step 30473 | loss:0.2543047070503235 lr:6.260083085095214e-06 tokens_per_second_per_gpu:4449.325890506421 mfu:20.988229358795053 +Step 30474 | loss:0.31029537320137024 lr:6.257982405562163e-06 tokens_per_second_per_gpu:5330.256603640048 mfu:25.14372983492471 +Step 30475 | loss:0.2933207154273987 lr:6.255882055018414e-06 tokens_per_second_per_gpu:5329.538153086989 mfu:25.140340781085627 +Step 30476 | loss:0.2838384211063385 lr:6.253782033479766e-06 tokens_per_second_per_gpu:4510.318667685009 mfu:21.275942695187318 +Step 30477 | loss:0.32533958554267883 lr:6.251682340961989e-06 tokens_per_second_per_gpu:3996.6914108378423 mfu:18.853075290792695 +Step 30478 | loss:0.2681637108325958 lr:6.2495829774809e-06 tokens_per_second_per_gpu:5313.523211372793 mfu:25.064795568588952 +Step 30479 | loss:0.2967348098754883 lr:6.247483943052285e-06 tokens_per_second_per_gpu:5354.684287071805 mfu:25.25895938546487 +Step 30480 | loss:0.2886662185192108 lr:6.245385237691931e-06 tokens_per_second_per_gpu:3909.512826343246 mfu:18.441839033531394 +Step 30481 | loss:0.16460144519805908 lr:6.243286861415604e-06 tokens_per_second_per_gpu:5360.82172473871 mfu:25.28791072609451 +Step 30482 | loss:0.28591713309288025 lr:6.241188814239096e-06 tokens_per_second_per_gpu:3954.971612842088 mfu:18.656275885515186 +Step 30483 | loss:0.2780076861381531 lr:6.239091096178195e-06 tokens_per_second_per_gpu:3951.8151252308785 mfu:18.641386194900793 +Step 30484 | loss:0.3029385507106781 lr:6.236993707248684e-06 tokens_per_second_per_gpu:5391.301232241435 mfu:25.431687763325083 +Step 30485 | loss:0.28071120381355286 lr:6.234896647466321e-06 tokens_per_second_per_gpu:4231.857297115238 mfu:19.96239289980077 +Step 30486 | loss:0.2938326597213745 lr:6.232799916846887e-06 tokens_per_second_per_gpu:4860.656186143154 mfu:22.928544543498706 +Step 30487 | loss:0.3977943956851959 lr:6.230703515406144e-06 tokens_per_second_per_gpu:3787.2692805917363 mfu:17.86519537132192 +Step 30488 | loss:0.3047332465648651 lr:6.228607443159878e-06 tokens_per_second_per_gpu:4399.101534510719 mfu:20.75131250240457 +Step 30489 | loss:0.3210179805755615 lr:6.226511700123827e-06 tokens_per_second_per_gpu:3831.7743040816213 mfu:18.075133160463714 +Step 30490 | loss:0.3375665247440338 lr:6.2244162863137665e-06 tokens_per_second_per_gpu:5155.859500831397 mfu:24.321068945762306 +Step 30491 | loss:0.24526941776275635 lr:6.222321201745468e-06 tokens_per_second_per_gpu:4079.987153052397 mfu:19.24599551853784 +Step 30492 | loss:0.29280588030815125 lr:6.220226446434668e-06 tokens_per_second_per_gpu:3342.908198915499 mfu:15.769068333236694 +Step 30493 | loss:0.3468392491340637 lr:6.2181320203971385e-06 tokens_per_second_per_gpu:5152.40695996372 mfu:24.304782721425255 +Step 30494 | loss:0.34759286046028137 lr:6.216037923648621e-06 tokens_per_second_per_gpu:5359.579633511531 mfu:25.282051569853287 +Step 30495 | loss:0.2816237211227417 lr:6.213944156204871e-06 tokens_per_second_per_gpu:4947.488822633766 mfu:23.338148905000406 +Step 30496 | loss:0.3892897963523865 lr:6.21185071808163e-06 tokens_per_second_per_gpu:3787.4286417908093 mfu:17.86594710528777 +Step 30497 | loss:0.2780791223049164 lr:6.209757609294647e-06 tokens_per_second_per_gpu:3667.5899785888123 mfu:17.300647684379623 +Step 30498 | loss:0.26228055357933044 lr:6.207664829859672e-06 tokens_per_second_per_gpu:3915.6980585913593 mfu:18.47101582935518 +Step 30499 | loss:0.348703533411026 lr:6.205572379792429e-06 tokens_per_second_per_gpu:4391.626639325208 mfu:20.716052146465884 +Step 30500 | loss:0.2988060712814331 lr:6.203480259108663e-06 tokens_per_second_per_gpu:5325.101103594495 mfu:25.119410461590824 +Step 30501 | loss:0.292232871055603 lr:6.201388467824121e-06 tokens_per_second_per_gpu:4606.246197868868 mfu:21.72844922198893 +Step 30502 | loss:0.2832218110561371 lr:6.199297005954524e-06 tokens_per_second_per_gpu:4774.397678846726 mfu:22.521648447362523 +Step 30503 | loss:0.32811209559440613 lr:6.197205873515599e-06 tokens_per_second_per_gpu:4268.9157416690605 mfu:20.137203905583778 +Step 30504 | loss:0.25156155228614807 lr:6.195115070523072e-06 tokens_per_second_per_gpu:5321.306448616612 mfu:25.101510426625254 +Step 30505 | loss:0.31536003947257996 lr:6.193024596992686e-06 tokens_per_second_per_gpu:2915.2143690929233 mfu:13.751563565871596 +Step 30506 | loss:0.23660825192928314 lr:6.190934452940139e-06 tokens_per_second_per_gpu:5301.903819663982 mfu:25.009984915425836 +Step 30507 | loss:0.2982759177684784 lr:6.188844638381169e-06 tokens_per_second_per_gpu:4244.412679271445 mfu:20.021618779600686 +Step 30508 | loss:0.3609180450439453 lr:6.186755153331497e-06 tokens_per_second_per_gpu:4866.689730118961 mfu:22.957005799861342 +Step 30509 | loss:0.32162052392959595 lr:6.184665997806832e-06 tokens_per_second_per_gpu:4839.588285692954 mfu:22.82916366252064 +Step 30510 | loss:0.3389052748680115 lr:6.182577171822867e-06 tokens_per_second_per_gpu:4639.023811913951 mfu:21.88306681987722 +Step 30511 | loss:0.24903225898742676 lr:6.180488675395341e-06 tokens_per_second_per_gpu:2710.04283968676 mfu:12.783734455790116 +Step 30512 | loss:0.28049010038375854 lr:6.178400508539955e-06 tokens_per_second_per_gpu:5300.038885073859 mfu:25.001187700773574 +Step 30513 | loss:0.3172083795070648 lr:6.176312671272399e-06 tokens_per_second_per_gpu:4181.573064783791 mfu:19.725193596518622 +Step 30514 | loss:0.21829979121685028 lr:6.174225163608388e-06 tokens_per_second_per_gpu:5126.6777831377285 mfu:24.183413804486822 +Step 30515 | loss:0.27952495217323303 lr:6.172137985563637e-06 tokens_per_second_per_gpu:5304.4411213328585 mfu:25.021953800306232 +Step 30516 | loss:0.3894907236099243 lr:6.170051137153826e-06 tokens_per_second_per_gpu:4347.243043734918 mfu:20.506687153444215 +Step 30517 | loss:0.35302016139030457 lr:6.16796461839464e-06 tokens_per_second_per_gpu:3755.9742323489513 mfu:17.717571289274883 +Step 30518 | loss:0.45785263180732727 lr:6.165878429301786e-06 tokens_per_second_per_gpu:4288.939577006524 mfu:20.231659753289193 +Step 30519 | loss:0.25884556770324707 lr:6.163792569890958e-06 tokens_per_second_per_gpu:5335.513016290672 mfu:25.16852523023418 +Step 30520 | loss:0.3376355469226837 lr:6.161707040177845e-06 tokens_per_second_per_gpu:4840.268058082085 mfu:22.832370264861297 +Step 30521 | loss:0.30226829648017883 lr:6.15962184017812e-06 tokens_per_second_per_gpu:4916.81881982445 mfu:23.193473268906786 +Step 30522 | loss:0.29216042160987854 lr:6.157536969907479e-06 tokens_per_second_per_gpu:3978.1922363925833 mfu:18.765811528649785 +Step 30523 | loss:0.3551192283630371 lr:6.155452429381603e-06 tokens_per_second_per_gpu:3760.46256907752 mfu:17.738743539412756 +Step 30524 | loss:0.31435373425483704 lr:6.153368218616151e-06 tokens_per_second_per_gpu:3431.226417883031 mfu:16.18568044074853 +Step 30525 | loss:0.2531026303768158 lr:6.151284337626811e-06 tokens_per_second_per_gpu:5318.343903342185 mfu:25.08753559510343 +Step 30526 | loss:0.2705858051776886 lr:6.149200786429255e-06 tokens_per_second_per_gpu:5338.335865682967 mfu:25.181841092445406 +Step 30527 | loss:0.30032655596733093 lr:6.147117565039168e-06 tokens_per_second_per_gpu:3902.2174675606825 mfu:18.40742558143748 +Step 30528 | loss:0.24876530468463898 lr:6.145034673472194e-06 tokens_per_second_per_gpu:4134.325672426921 mfu:19.50231958553488 +Step 30529 | loss:0.37751826643943787 lr:6.14295211174401e-06 tokens_per_second_per_gpu:5318.287416974622 mfu:25.08726913927007 +Step 30530 | loss:0.3280816674232483 lr:6.140869879870287e-06 tokens_per_second_per_gpu:3608.7230708012057 mfu:17.022962436615337 +Step 30531 | loss:0.33557549118995667 lr:6.138787977866684e-06 tokens_per_second_per_gpu:5323.833128762246 mfu:25.11342920796829 +Step 30532 | loss:0.29968881607055664 lr:6.1367064057488376e-06 tokens_per_second_per_gpu:5343.162156918703 mfu:25.204607531654567 +Step 30533 | loss:0.37917351722717285 lr:6.134625163532426e-06 tokens_per_second_per_gpu:3834.366859376301 mfu:18.087362686124266 +Step 30534 | loss:0.3104943037033081 lr:6.1325442512331025e-06 tokens_per_second_per_gpu:3891.946779320387 mfu:18.358976992647897 +Step 30535 | loss:0.3048156797885895 lr:6.1304636688664985e-06 tokens_per_second_per_gpu:3851.3802708394733 mfu:18.167617850783373 +Step 30536 | loss:0.2971411943435669 lr:6.128383416448275e-06 tokens_per_second_per_gpu:4648.156171417043 mfu:21.926145717794505 +Step 30537 | loss:0.3734486401081085 lr:6.12630349399409e-06 tokens_per_second_per_gpu:3939.5864366291007 mfu:18.583701384336784 +Step 30538 | loss:0.3548946678638458 lr:6.124223901519577e-06 tokens_per_second_per_gpu:4045.1519908874902 mfu:19.081672115114436 +Step 30539 | loss:0.2911655306816101 lr:6.122144639040361e-06 tokens_per_second_per_gpu:5162.170996088497 mfu:24.35084134574218 +Step 30540 | loss:0.3106844425201416 lr:6.120065706572098e-06 tokens_per_second_per_gpu:4547.131717338295 mfu:21.44959613135581 +Step 30541 | loss:0.40479856729507446 lr:6.117987104130424e-06 tokens_per_second_per_gpu:4522.416471353037 mfu:21.333010099187554 +Step 30542 | loss:0.2846396267414093 lr:6.115908831730965e-06 tokens_per_second_per_gpu:5384.557912536004 mfu:25.399878373745583 +Step 30543 | loss:0.3259708285331726 lr:6.113830889389349e-06 tokens_per_second_per_gpu:3573.2906095489975 mfu:16.855821471487403 +Step 30544 | loss:0.35862529277801514 lr:6.1117532771212234e-06 tokens_per_second_per_gpu:4354.397851330117 mfu:20.540437601606822 +Step 30545 | loss:0.308500736951828 lr:6.109675994942205e-06 tokens_per_second_per_gpu:5358.1445506990585 mfu:25.275282039376478 +Step 30546 | loss:0.3687150180339813 lr:6.107599042867901e-06 tokens_per_second_per_gpu:5294.402020716334 mfu:24.974597649850896 +Step 30547 | loss:0.2906480133533478 lr:6.105522420913945e-06 tokens_per_second_per_gpu:4933.707035624775 mfu:23.27313786426366 +Step 30548 | loss:0.3954101502895355 lr:6.1034461290959636e-06 tokens_per_second_per_gpu:5365.914603159158 mfu:25.311934702538423 +Step 30549 | loss:0.28068289160728455 lr:6.10137016742956e-06 tokens_per_second_per_gpu:4752.137234934754 mfu:22.416642135406377 +Step 30550 | loss:0.27812907099723816 lr:6.09929453593035e-06 tokens_per_second_per_gpu:5335.403760809425 mfu:25.16800985348823 +Step 30551 | loss:0.3069285452365875 lr:6.09721923461396e-06 tokens_per_second_per_gpu:3969.1226601812136 mfu:18.723028790231485 +Step 30552 | loss:0.24270573258399963 lr:6.095144263495983e-06 tokens_per_second_per_gpu:3352.821853810172 mfu:15.815832734818809 +Step 30553 | loss:0.26742634177207947 lr:6.093069622592018e-06 tokens_per_second_per_gpu:5158.395220285673 mfu:24.333030367066474 +Step 30554 | loss:0.36082157492637634 lr:6.09099531191768e-06 tokens_per_second_per_gpu:5327.672633741268 mfu:25.13154080804144 +Step 30555 | loss:0.26978984475135803 lr:6.088921331488568e-06 tokens_per_second_per_gpu:4520.160260448536 mfu:21.322367167401747 +Step 30556 | loss:0.2779141366481781 lr:6.086847681320296e-06 tokens_per_second_per_gpu:4124.8754941814095 mfu:19.457741482384133 +Step 30557 | loss:0.2696225941181183 lr:6.08477436142843e-06 tokens_per_second_per_gpu:4859.178884484577 mfu:22.921575859521546 +Step 30558 | loss:0.35807278752326965 lr:6.0827013718285865e-06 tokens_per_second_per_gpu:3868.574989667009 mfu:18.248728273214404 +Step 30559 | loss:0.28797540068626404 lr:6.080628712536368e-06 tokens_per_second_per_gpu:5385.036652032665 mfu:25.402136669632714 +Step 30560 | loss:0.21068894863128662 lr:6.0785563835673185e-06 tokens_per_second_per_gpu:3471.3282590821427 mfu:16.37484766776438 +Step 30561 | loss:0.41223791241645813 lr:6.076484384937059e-06 tokens_per_second_per_gpu:5171.817364615927 mfu:24.39634491192625 +Step 30562 | loss:0.3529432415962219 lr:6.07441271666116e-06 tokens_per_second_per_gpu:5119.705875761049 mfu:24.150526127860967 +Step 30563 | loss:0.33489036560058594 lr:6.07234137875522e-06 tokens_per_second_per_gpu:4451.756519183124 mfu:20.999695048971073 +Step 30564 | loss:0.3034532368183136 lr:6.070270371234793e-06 tokens_per_second_per_gpu:4362.060783216525 mfu:20.576584958746007 +Step 30565 | loss:0.30467379093170166 lr:6.068199694115473e-06 tokens_per_second_per_gpu:4916.692759407291 mfu:23.192878620410923 +Step 30566 | loss:0.24579079449176788 lr:6.0661293474128355e-06 tokens_per_second_per_gpu:4763.14242508606 mfu:22.46855549502915 +Step 30567 | loss:0.21824151277542114 lr:6.0640593311424464e-06 tokens_per_second_per_gpu:4381.567612583561 mfu:20.6686020010786 +Step 30568 | loss:0.30083727836608887 lr:6.061989645319865e-06 tokens_per_second_per_gpu:3270.0381562621483 mfu:15.425327909129523 +Step 30569 | loss:0.27559858560562134 lr:6.059920289960664e-06 tokens_per_second_per_gpu:5341.786962012062 mfu:25.198120502647498 +Step 30570 | loss:0.3304031491279602 lr:6.057851265080416e-06 tokens_per_second_per_gpu:4712.262377506174 mfu:22.22854562956355 +Step 30571 | loss:0.42553219199180603 lr:6.0557825706946684e-06 tokens_per_second_per_gpu:5318.070971769342 mfu:25.0862481303074 +Step 30572 | loss:0.28911566734313965 lr:6.053714206818989e-06 tokens_per_second_per_gpu:5319.842341842313 mfu:25.094603985168504 +Step 30573 | loss:0.3522171676158905 lr:6.051646173468939e-06 tokens_per_second_per_gpu:4003.7546905476456 mfu:18.886393986303624 +Step 30574 | loss:0.31151899695396423 lr:6.049578470660072e-06 tokens_per_second_per_gpu:3453.846517191456 mfu:16.292383308573278 +Step 30575 | loss:0.30314990878105164 lr:6.047511098407921e-06 tokens_per_second_per_gpu:5191.3676000869345 mfu:24.48856670825712 +Step 30576 | loss:0.37224283814430237 lr:6.045444056728045e-06 tokens_per_second_per_gpu:5345.821475105194 mfu:25.217151989266714 +Step 30577 | loss:0.39716118574142456 lr:6.043377345636003e-06 tokens_per_second_per_gpu:5329.840963622413 mfu:25.141769190046027 +Step 30578 | loss:0.3787150979042053 lr:6.041310965147318e-06 tokens_per_second_per_gpu:5265.741890789417 mfu:24.839402927063233 +Step 30579 | loss:0.3274141848087311 lr:6.039244915277547e-06 tokens_per_second_per_gpu:4802.183273850183 mfu:22.652717839705183 +Step 30580 | loss:0.2494661957025528 lr:6.03717919604223e-06 tokens_per_second_per_gpu:3764.964853364925 mfu:17.759981582565352 +Step 30581 | loss:0.24049656093120575 lr:6.035113807456899e-06 tokens_per_second_per_gpu:4143.907172597697 mfu:19.547517156612933 +Step 30582 | loss:0.2934822142124176 lr:6.033048749537079e-06 tokens_per_second_per_gpu:4608.091444043619 mfu:21.737153563026695 +Step 30583 | loss:0.2703540325164795 lr:6.030984022298308e-06 tokens_per_second_per_gpu:5347.329067884686 mfu:25.22426356162943 +Step 30584 | loss:0.32444682717323303 lr:6.02891962575613e-06 tokens_per_second_per_gpu:4062.0100177856566 mfu:19.161194304269067 +Step 30585 | loss:0.27582618594169617 lr:6.026855559926043e-06 tokens_per_second_per_gpu:5355.463999119994 mfu:25.262637420228774 +Step 30586 | loss:0.3160502314567566 lr:6.024791824823589e-06 tokens_per_second_per_gpu:5349.643680830092 mfu:25.235181985806605 +Step 30587 | loss:0.31333664059638977 lr:6.022728420464296e-06 tokens_per_second_per_gpu:4500.425062880876 mfu:21.229272873303177 +Step 30588 | loss:0.2548670470714569 lr:6.020665346863674e-06 tokens_per_second_per_gpu:4137.941662654109 mfu:19.519376828388076 +Step 30589 | loss:0.3078177273273468 lr:6.018602604037232e-06 tokens_per_second_per_gpu:5312.388597246728 mfu:25.059443400171165 +Step 30590 | loss:0.3605545163154602 lr:6.016540192000491e-06 tokens_per_second_per_gpu:4592.331811030292 mfu:21.66281268523217 +Step 30591 | loss:0.337383508682251 lr:6.014478110768973e-06 tokens_per_second_per_gpu:4159.329509726103 mfu:19.620266952169676 +Step 30592 | loss:0.35447782278060913 lr:6.012416360358164e-06 tokens_per_second_per_gpu:3568.791539945625 mfu:16.834598592547863 +Step 30593 | loss:0.23193424940109253 lr:6.0103549407835916e-06 tokens_per_second_per_gpu:3960.669637388327 mfu:18.683154439483033 +Step 30594 | loss:0.24227353930473328 lr:6.008293852060748e-06 tokens_per_second_per_gpu:3197.93257708823 mfu:15.085193589685751 +Step 30595 | loss:0.31180042028427124 lr:6.00623309420516e-06 tokens_per_second_per_gpu:5044.574969186257 mfu:23.79612082289319 +Step 30596 | loss:0.2677907645702362 lr:6.004172667232289e-06 tokens_per_second_per_gpu:5354.498136840019 mfu:25.25808128306103 +Step 30597 | loss:0.3597952127456665 lr:6.002112571157648e-06 tokens_per_second_per_gpu:4286.0826723466525 mfu:20.218183246570383 +Step 30598 | loss:0.2925269901752472 lr:6.000052805996731e-06 tokens_per_second_per_gpu:4569.085548406212 mfu:21.553156098212998 +Step 30599 | loss:0.28318941593170166 lr:5.9979933717650425e-06 tokens_per_second_per_gpu:5338.340584496844 mfu:25.181863351895576 +Step 30600 | loss:0.27428457140922546 lr:5.995934268478048e-06 tokens_per_second_per_gpu:4184.6350694139 mfu:19.739637594790707 +Step 30601 | loss:0.36495241522789 lr:5.9938754961512525e-06 tokens_per_second_per_gpu:4273.996669454272 mfu:20.161171508842294 +Step 30602 | loss:0.27691149711608887 lr:5.991817054800141e-06 tokens_per_second_per_gpu:4504.184054780202 mfu:21.24700467944245 +Step 30603 | loss:0.4089488685131073 lr:5.989758944440188e-06 tokens_per_second_per_gpu:3217.4071161809384 mfu:15.177058313285052 +Step 30604 | loss:0.28220054507255554 lr:5.987701165086868e-06 tokens_per_second_per_gpu:5325.48063285364 mfu:25.12120076586058 +Step 30605 | loss:0.28297507762908936 lr:5.985643716755656e-06 tokens_per_second_per_gpu:4491.096712375428 mfu:21.185269452388262 +Step 30606 | loss:0.36191001534461975 lr:5.983586599462049e-06 tokens_per_second_per_gpu:4214.333070066766 mfu:19.879728130871676 +Step 30607 | loss:0.2700577676296234 lr:5.981529813221492e-06 tokens_per_second_per_gpu:5363.096641328874 mfu:25.298641895791107 +Step 30608 | loss:0.3501131534576416 lr:5.97947335804947e-06 tokens_per_second_per_gpu:5192.386288559551 mfu:24.493372035588628 +Step 30609 | loss:0.25336042046546936 lr:5.977417233961452e-06 tokens_per_second_per_gpu:5022.651651722037 mfu:23.6927047939108 +Step 30610 | loss:0.36382970213890076 lr:5.975361440972899e-06 tokens_per_second_per_gpu:3454.1137315682968 mfu:16.293643804380313 +Step 30611 | loss:0.28119397163391113 lr:5.9733059790992585e-06 tokens_per_second_per_gpu:4038.384668855502 mfu:19.049749502465346 +Step 30612 | loss:0.34854960441589355 lr:5.971250848356002e-06 tokens_per_second_per_gpu:5316.73645406516 mfu:25.079952982604578 +Step 30613 | loss:0.30013948678970337 lr:5.969196048758596e-06 tokens_per_second_per_gpu:3952.447858548739 mfu:18.644370905916574 +Step 30614 | loss:0.300234854221344 lr:5.967141580322472e-06 tokens_per_second_per_gpu:4623.863021061334 mfu:21.811550782727775 +Step 30615 | loss:0.30222800374031067 lr:5.9650874430631e-06 tokens_per_second_per_gpu:4786.723128025194 mfu:22.57978969407583 +Step 30616 | loss:0.3966898024082184 lr:5.9630336369959294e-06 tokens_per_second_per_gpu:5340.946380975921 mfu:25.19415533848211 +Step 30617 | loss:0.2592235803604126 lr:5.96098016213641e-06 tokens_per_second_per_gpu:5363.686229508096 mfu:25.301423083826613 +Step 30618 | loss:0.2992892861366272 lr:5.9589270184999606e-06 tokens_per_second_per_gpu:5340.11241062287 mfu:25.19022135803675 +Step 30619 | loss:0.31072866916656494 lr:5.956874206102042e-06 tokens_per_second_per_gpu:5331.839272444741 mfu:25.15119555370759 +Step 30620 | loss:0.31093090772628784 lr:5.954821724958104e-06 tokens_per_second_per_gpu:5326.083279228101 mfu:25.12404355163099 +Step 30621 | loss:0.29747840762138367 lr:5.952769575083561e-06 tokens_per_second_per_gpu:4458.708083620751 mfu:21.032486764482556 +Step 30622 | loss:0.2412092238664627 lr:5.950717756493857e-06 tokens_per_second_per_gpu:5320.592172251617 mfu:25.098141063143455 +Step 30623 | loss:0.31374630331993103 lr:5.948666269204428e-06 tokens_per_second_per_gpu:4116.411761238633 mfu:19.417816610031306 +Step 30624 | loss:0.30465036630630493 lr:5.946615113230719e-06 tokens_per_second_per_gpu:4642.7763332099485 mfu:21.90076810307659 +Step 30625 | loss:0.3756996989250183 lr:5.944564288588117e-06 tokens_per_second_per_gpu:3970.1313889828193 mfu:18.72778713609554 +Step 30626 | loss:0.2528195083141327 lr:5.942513795292076e-06 tokens_per_second_per_gpu:4743.639922923105 mfu:22.376558864856502 +Step 30627 | loss:0.24666178226470947 lr:5.940463633358012e-06 tokens_per_second_per_gpu:4441.785591885541 mfu:20.952660483692757 +Step 30628 | loss:0.3219231963157654 lr:5.93841380280134e-06 tokens_per_second_per_gpu:5343.554917394671 mfu:25.206460249831558 +Step 30629 | loss:0.21896420419216156 lr:5.936364303637476e-06 tokens_per_second_per_gpu:5362.123949709156 mfu:25.29405354346259 +Step 30630 | loss:0.35198429226875305 lr:5.9343151358818405e-06 tokens_per_second_per_gpu:5349.377903551077 mfu:25.23392826903521 +Step 30631 | loss:0.22608746588230133 lr:5.9322662995498605e-06 tokens_per_second_per_gpu:4328.883008208437 mfu:20.420079733320723 +Step 30632 | loss:0.38231298327445984 lr:5.930217794656912e-06 tokens_per_second_per_gpu:4472.733810182547 mfu:21.098648509710266 +Step 30633 | loss:0.35853177309036255 lr:5.928169621218421e-06 tokens_per_second_per_gpu:5347.86702901994 mfu:25.22680121609006 +Step 30634 | loss:0.2575657367706299 lr:5.926121779249783e-06 tokens_per_second_per_gpu:4994.692106448711 mfu:23.560814848466187 +Step 30635 | loss:0.3553464412689209 lr:5.924074268766422e-06 tokens_per_second_per_gpu:4514.312564245085 mfu:21.294782586690218 +Step 30636 | loss:0.33020254969596863 lr:5.92202708978371e-06 tokens_per_second_per_gpu:4355.542507092749 mfu:20.545837138137568 +Step 30637 | loss:0.3447762131690979 lr:5.919980242317058e-06 tokens_per_second_per_gpu:5361.695352267316 mfu:25.292031776203167 +Step 30638 | loss:0.3278074562549591 lr:5.917933726381869e-06 tokens_per_second_per_gpu:5057.444467541965 mfu:23.856828442399905 +Step 30639 | loss:0.39498192071914673 lr:5.9158875419935315e-06 tokens_per_second_per_gpu:3874.3268228363254 mfu:18.275860651638002 +Step 30640 | loss:0.2366478145122528 lr:5.91384168916741e-06 tokens_per_second_per_gpu:3216.729811475937 mfu:15.17386335143146 +Step 30641 | loss:0.2597854435443878 lr:5.911796167918915e-06 tokens_per_second_per_gpu:5068.379624771836 mfu:23.908411444783418 +Step 30642 | loss:0.3135472238063812 lr:5.9097509782634344e-06 tokens_per_second_per_gpu:5390.590059372174 mfu:25.428333039562176 +Step 30643 | loss:0.26608869433403015 lr:5.9077061202163334e-06 tokens_per_second_per_gpu:4717.486205536829 mfu:22.25318731766112 +Step 30644 | loss:0.30614688992500305 lr:5.905661593792999e-06 tokens_per_second_per_gpu:4517.141618930662 mfu:21.30812772032858 +Step 30645 | loss:0.33017104864120483 lr:5.903617399008826e-06 tokens_per_second_per_gpu:5337.145389083788 mfu:25.176225411210826 +Step 30646 | loss:0.2689938247203827 lr:5.901573535879168e-06 tokens_per_second_per_gpu:4709.284269331298 mfu:22.214497385185005 +Step 30647 | loss:0.30681532621383667 lr:5.899530004419396e-06 tokens_per_second_per_gpu:5366.341217024956 mfu:25.31394711293122 +Step 30648 | loss:0.34580254554748535 lr:5.897486804644881e-06 tokens_per_second_per_gpu:3986.408041788147 mfu:18.8045668844616 +Step 30649 | loss:0.3746437132358551 lr:5.89544393657101e-06 tokens_per_second_per_gpu:3733.6760439484688 mfu:17.61238693012656 +Step 30650 | loss:0.23793642222881317 lr:5.893401400213122e-06 tokens_per_second_per_gpu:5327.643185629991 mfu:25.13140189627618 +Step 30651 | loss:0.2469492107629776 lr:5.891359195586588e-06 tokens_per_second_per_gpu:3967.0251691501517 mfu:18.713134567169096 +Step 30652 | loss:0.3202002942562103 lr:5.889317322706783e-06 tokens_per_second_per_gpu:3642.6690456604624 mfu:17.183091391806993 +Step 30653 | loss:0.3868623673915863 lr:5.887275781589047e-06 tokens_per_second_per_gpu:5325.337915548854 mfu:25.120527543983822 +Step 30654 | loss:0.3225880265235901 lr:5.8852345722487325e-06 tokens_per_second_per_gpu:5336.0063236490205 mfu:25.170852245210426 +Step 30655 | loss:0.28475072979927063 lr:5.883193694701194e-06 tokens_per_second_per_gpu:5295.2507377085085 mfu:24.978601192709036 +Step 30656 | loss:0.2754317820072174 lr:5.881153148961799e-06 tokens_per_second_per_gpu:5350.9698905751 mfu:25.24143794344852 +Step 30657 | loss:0.3812111020088196 lr:5.879112935045866e-06 tokens_per_second_per_gpu:5338.592219676003 mfu:25.18305035796906 +Step 30658 | loss:0.3420058488845825 lr:5.877073052968763e-06 tokens_per_second_per_gpu:3349.564040720186 mfu:15.800465074633115 +Step 30659 | loss:0.253589004278183 lr:5.875033502745825e-06 tokens_per_second_per_gpu:5153.130000182805 mfu:24.30819342550206 +Step 30660 | loss:0.2393321990966797 lr:5.872994284392397e-06 tokens_per_second_per_gpu:4218.86266791111 mfu:19.901095016732533 +Step 30661 | loss:0.23893235623836517 lr:5.8709553979237945e-06 tokens_per_second_per_gpu:3830.6114477328015 mfu:18.06964776866258 +Step 30662 | loss:0.24594341218471527 lr:5.868916843355365e-06 tokens_per_second_per_gpu:3397.8666810949408 mfu:16.028316870561095 +Step 30663 | loss:0.23004597425460815 lr:5.866878620702459e-06 tokens_per_second_per_gpu:4694.940055484622 mfu:22.146833281094988 +Step 30664 | loss:0.31128406524658203 lr:5.864840729980381e-06 tokens_per_second_per_gpu:5331.062457801164 mfu:25.14753118649559 +Step 30665 | loss:0.3875737190246582 lr:5.862803171204462e-06 tokens_per_second_per_gpu:4750.40663170899 mfu:22.40847858471909 +Step 30666 | loss:0.24869702756404877 lr:5.860765944390034e-06 tokens_per_second_per_gpu:5204.000074182794 mfu:24.548156243889622 +Step 30667 | loss:0.3519662022590637 lr:5.858729049552442e-06 tokens_per_second_per_gpu:4454.250119949439 mfu:21.011457789237944 +Step 30668 | loss:0.22342851758003235 lr:5.85669248670696e-06 tokens_per_second_per_gpu:4905.529948418756 mfu:23.140221736406055 +Step 30669 | loss:0.31837934255599976 lr:5.854656255868923e-06 tokens_per_second_per_gpu:4157.801695772236 mfu:19.61305999307734 +Step 30670 | loss:0.2725440263748169 lr:5.852620357053651e-06 tokens_per_second_per_gpu:3666.4787454799575 mfu:17.29540580820924 +Step 30671 | loss:0.27569666504859924 lr:5.850584790276464e-06 tokens_per_second_per_gpu:3762.5596242709216 mfu:17.74863570655454 +Step 30672 | loss:0.27684029936790466 lr:5.848549555552651e-06 tokens_per_second_per_gpu:3605.6989964646723 mfu:17.008697362009627 +Step 30673 | loss:0.24325305223464966 lr:5.846514652897533e-06 tokens_per_second_per_gpu:4648.052088216972 mfu:21.925654739581347 +Step 30674 | loss:0.32563549280166626 lr:5.844480082326425e-06 tokens_per_second_per_gpu:4097.23060513526 mfu:19.327335824049005 +Step 30675 | loss:0.3431271016597748 lr:5.84244584385461e-06 tokens_per_second_per_gpu:4815.060578551034 mfu:22.713462283073838 +Step 30676 | loss:0.2678850293159485 lr:5.840411937497392e-06 tokens_per_second_per_gpu:4851.28324952622 mfu:22.88433080228874 +Step 30677 | loss:0.37469160556793213 lr:5.838378363270064e-06 tokens_per_second_per_gpu:4665.567315056787 mfu:22.008277052990607 +Step 30678 | loss:0.22539420425891876 lr:5.836345121187941e-06 tokens_per_second_per_gpu:4848.087769130034 mfu:22.869257176054756 +Step 30679 | loss:0.31173205375671387 lr:5.834312211266291e-06 tokens_per_second_per_gpu:5319.66498588833 mfu:25.093767366873525 +Step 30680 | loss:0.26205724477767944 lr:5.83227963352041e-06 tokens_per_second_per_gpu:5316.740485335728 mfu:25.079971998795656 +Step 30681 | loss:0.39253807067871094 lr:5.830247387965604e-06 tokens_per_second_per_gpu:3892.7164030511317 mfu:18.362607439097168 +Step 30682 | loss:0.27758604288101196 lr:5.8282154746171435e-06 tokens_per_second_per_gpu:4546.156973307729 mfu:21.444998097455148 +Step 30683 | loss:0.2472531795501709 lr:5.8261838934903e-06 tokens_per_second_per_gpu:4518.248628703142 mfu:21.313349674300714 +Step 30684 | loss:0.2205611765384674 lr:5.824152644600367e-06 tokens_per_second_per_gpu:3846.4268017749087 mfu:18.144251492056846 +Step 30685 | loss:0.3836866021156311 lr:5.822121727962626e-06 tokens_per_second_per_gpu:4998.430785152401 mfu:23.57845083379586 +Step 30686 | loss:0.2744535207748413 lr:5.820091143592332e-06 tokens_per_second_per_gpu:4545.462150807055 mfu:21.441720501170522 +Step 30687 | loss:0.35773634910583496 lr:5.818060891504773e-06 tokens_per_second_per_gpu:3312.335675075988 mfu:15.624852522075276 +Step 30688 | loss:0.24836519360542297 lr:5.816030971715225e-06 tokens_per_second_per_gpu:5348.757414204737 mfu:25.231001314922125 +Step 30689 | loss:0.21786583960056305 lr:5.8140013842389485e-06 tokens_per_second_per_gpu:4936.943208159576 mfu:23.288403442258286 +Step 30690 | loss:0.2714329957962036 lr:5.811972129091192e-06 tokens_per_second_per_gpu:5359.350103008867 mfu:25.280968835310283 +Step 30691 | loss:0.42584285140037537 lr:5.8099432062872326e-06 tokens_per_second_per_gpu:4176.2579667258 mfu:19.700121372129836 +Step 30692 | loss:0.2891276478767395 lr:5.807914615842341e-06 tokens_per_second_per_gpu:3778.986226323308 mfu:17.826122791102957 +Step 30693 | loss:0.26755839586257935 lr:5.8058863577717505e-06 tokens_per_second_per_gpu:5034.199398964738 mfu:23.747177487903468 +Step 30694 | loss:0.2931193709373474 lr:5.80385843209073e-06 tokens_per_second_per_gpu:4038.6860195254776 mfu:19.051171024991348 +Step 30695 | loss:0.22809165716171265 lr:5.801830838814537e-06 tokens_per_second_per_gpu:4206.691643309254 mfu:19.84368221709465 +Step 30696 | loss:0.3670119047164917 lr:5.799803577958412e-06 tokens_per_second_per_gpu:5359.996234129384 mfu:25.284016746047094 +Step 30697 | loss:0.2595685124397278 lr:5.797776649537601e-06 tokens_per_second_per_gpu:4079.6472953379953 mfu:19.2443923517133 +Step 30698 | loss:0.20627336204051971 lr:5.795750053567351e-06 tokens_per_second_per_gpu:5328.055319910306 mfu:25.133346003993783 +Step 30699 | loss:0.33271634578704834 lr:5.79372379006291e-06 tokens_per_second_per_gpu:5312.4521485875875 mfu:25.05974318268898 +Step 30700 | loss:0.435598760843277 lr:5.791697859039508e-06 tokens_per_second_per_gpu:4160.092946083628 mfu:19.62386821172379 +Step 30701 | loss:0.24108977615833282 lr:5.789672260512385e-06 tokens_per_second_per_gpu:3806.426693566536 mfu:17.955564156915752 +Step 30702 | loss:0.23324254155158997 lr:5.787646994496776e-06 tokens_per_second_per_gpu:4523.67160505457 mfu:21.338930779005576 +Step 30703 | loss:0.2450479120016098 lr:5.785622061007939e-06 tokens_per_second_per_gpu:3457.829169788139 mfu:16.311170160382435 +Step 30704 | loss:0.23165543377399445 lr:5.783597460061058e-06 tokens_per_second_per_gpu:4343.809550426074 mfu:20.49049077969203 +Step 30705 | loss:0.34487155079841614 lr:5.781573191671386e-06 tokens_per_second_per_gpu:3232.6663305988145 mfu:15.249038631184616 +Step 30706 | loss:0.31160423159599304 lr:5.779549255854139e-06 tokens_per_second_per_gpu:5334.322209873546 mfu:25.16290799319207 +Step 30707 | loss:0.23251307010650635 lr:5.77752565262456e-06 tokens_per_second_per_gpu:4744.3891298218705 mfu:22.380093001625855 +Step 30708 | loss:0.28821563720703125 lr:5.775502381997844e-06 tokens_per_second_per_gpu:5350.4506565169695 mfu:25.238988627805856 +Step 30709 | loss:0.39735203981399536 lr:5.773479443989211e-06 tokens_per_second_per_gpu:5081.838868275193 mfu:23.97190099277249 +Step 30710 | loss:0.3883365988731384 lr:5.771456838613898e-06 tokens_per_second_per_gpu:4850.759465933388 mfu:22.88186002571466 +Step 30711 | loss:0.33227014541625977 lr:5.769434565887099e-06 tokens_per_second_per_gpu:5362.193876071697 mfu:25.294383398045362 +Step 30712 | loss:0.33101099729537964 lr:5.76741262582402e-06 tokens_per_second_per_gpu:5341.5249749566665 mfu:25.196884665007595 +Step 30713 | loss:0.3489455580711365 lr:5.765391018439869e-06 tokens_per_second_per_gpu:5365.374760960169 mfu:25.30938817477921 +Step 30714 | loss:0.25774815678596497 lr:5.763369743749869e-06 tokens_per_second_per_gpu:5170.815241571643 mfu:24.39161772654654 +Step 30715 | loss:0.378415584564209 lr:5.7613488017692005e-06 tokens_per_second_per_gpu:4369.152528998846 mfu:20.61003793357781 +Step 30716 | loss:0.3791794776916504 lr:5.759328192513075e-06 tokens_per_second_per_gpu:4838.366529484188 mfu:22.823400430030784 +Step 30717 | loss:0.3054819703102112 lr:5.757307915996696e-06 tokens_per_second_per_gpu:4723.461452007752 mfu:22.281373574746635 +Step 30718 | loss:0.37398943305015564 lr:5.755287972235246e-06 tokens_per_second_per_gpu:5334.483813872206 mfu:25.163670306826237 +Step 30719 | loss:0.31397876143455505 lr:5.753268361243913e-06 tokens_per_second_per_gpu:3773.638550946428 mfu:17.800896893942713 +Step 30720 | loss:0.3374921381473541 lr:5.751249083037897e-06 tokens_per_second_per_gpu:5309.493152231189 mfu:25.04578509201901 +Step 30721 | loss:0.18258202075958252 lr:5.749230137632388e-06 tokens_per_second_per_gpu:3996.3244317010794 mfu:18.85134418759149 +Step 30722 | loss:0.3026149570941925 lr:5.747211525042562e-06 tokens_per_second_per_gpu:4748.574157505179 mfu:22.399834491247375 +Step 30723 | loss:0.4095216393470764 lr:5.745193245283598e-06 tokens_per_second_per_gpu:4407.925722606812 mfu:20.79293770321558 +Step 30724 | loss:0.37456101179122925 lr:5.743175298370696e-06 tokens_per_second_per_gpu:3874.813909011323 mfu:18.27815832023094 +Step 30725 | loss:0.2613372504711151 lr:5.741157684319021e-06 tokens_per_second_per_gpu:3387.811373946687 mfu:15.980884271130476 +Step 30726 | loss:0.29662463068962097 lr:5.739140403143734e-06 tokens_per_second_per_gpu:5360.226780318872 mfu:25.285104271758147 +Step 30727 | loss:0.38657358288764954 lr:5.737123454860022e-06 tokens_per_second_per_gpu:4451.801743895805 mfu:20.999908381656923 +Step 30728 | loss:0.30542370676994324 lr:5.735106839483062e-06 tokens_per_second_per_gpu:5339.111366395684 mfu:25.18549926162136 +Step 30729 | loss:0.24725058674812317 lr:5.733090557027998e-06 tokens_per_second_per_gpu:5307.017291948627 mfu:25.034106036640843 +Step 30730 | loss:0.33270063996315 lr:5.731074607510012e-06 tokens_per_second_per_gpu:4973.203171356325 mfu:23.45944787524501 +Step 30731 | loss:0.4179609417915344 lr:5.729058990944275e-06 tokens_per_second_per_gpu:3351.5472826090354 mfu:15.809820364998776 +Step 30732 | loss:0.32069653272628784 lr:5.727043707345931e-06 tokens_per_second_per_gpu:5347.289703177744 mfu:25.224077871590474 +Step 30733 | loss:0.3300768733024597 lr:5.725028756730127e-06 tokens_per_second_per_gpu:3227.1337731713347 mfu:15.222940613847559 +Step 30734 | loss:0.2759886085987091 lr:5.723014139112038e-06 tokens_per_second_per_gpu:3527.99566948856 mfu:16.642157511108646 +Step 30735 | loss:0.32722190022468567 lr:5.720999854506814e-06 tokens_per_second_per_gpu:3961.9218873952263 mfu:18.68906151642146 +Step 30736 | loss:0.22218698263168335 lr:5.718985902929596e-06 tokens_per_second_per_gpu:4805.567392881242 mfu:22.668681306564935 +Step 30737 | loss:0.43076851963996887 lr:5.716972284395533e-06 tokens_per_second_per_gpu:3936.3764746805978 mfu:18.568559446149713 +Step 30738 | loss:0.39967143535614014 lr:5.714958998919767e-06 tokens_per_second_per_gpu:5276.008911199693 mfu:24.887834214072818 +Step 30739 | loss:0.4183040261268616 lr:5.71294604651747e-06 tokens_per_second_per_gpu:3995.3415240810564 mfu:18.846707644646507 +Step 30740 | loss:0.48837652802467346 lr:5.710933427203735e-06 tokens_per_second_per_gpu:4949.393097030414 mfu:23.347131692232104 +Step 30741 | loss:0.2788380980491638 lr:5.708921140993723e-06 tokens_per_second_per_gpu:3166.1514541115885 mfu:14.935276610186211 +Step 30742 | loss:0.33319351077079773 lr:5.706909187902565e-06 tokens_per_second_per_gpu:4352.8613755812585 mfu:20.533189783349663 +Step 30743 | loss:0.2075575441122055 lr:5.704897567945411e-06 tokens_per_second_per_gpu:4316.821345790028 mfu:20.363182813761277 +Step 30744 | loss:0.21964149177074432 lr:5.702886281137354e-06 tokens_per_second_per_gpu:4718.456512036969 mfu:22.25776441897364 +Step 30745 | loss:0.2579696774482727 lr:5.700875327493549e-06 tokens_per_second_per_gpu:4954.7922940330745 mfu:23.372600625691657 +Step 30746 | loss:0.2983454763889313 lr:5.6988647070291224e-06 tokens_per_second_per_gpu:4092.1549762866935 mfu:19.303393216778733 +Step 30747 | loss:0.2221117466688156 lr:5.696854419759184e-06 tokens_per_second_per_gpu:4958.788508275366 mfu:23.391451450097 +Step 30748 | loss:0.2801385521888733 lr:5.694844465698851e-06 tokens_per_second_per_gpu:4126.008121264969 mfu:19.463084277583388 +Step 30749 | loss:0.280354380607605 lr:5.692834844863243e-06 tokens_per_second_per_gpu:4248.89531700518 mfu:20.042764146607578 +Step 30750 | loss:0.31230413913726807 lr:5.690825557267487e-06 tokens_per_second_per_gpu:5369.237257760757 mfu:25.32760823119749 +Step 30751 | loss:0.27273061871528625 lr:5.688816602926677e-06 tokens_per_second_per_gpu:5162.786541931905 mfu:24.353744980508484 +Step 30752 | loss:0.3378894329071045 lr:5.68680798185593e-06 tokens_per_second_per_gpu:3089.4277328262333 mfu:14.573357726466497 +Step 30753 | loss:0.2898035943508148 lr:5.68479969407037e-06 tokens_per_second_per_gpu:5354.129165186803 mfu:25.25634078081976 +Step 30754 | loss:0.40672481060028076 lr:5.68279173958508e-06 tokens_per_second_per_gpu:5340.424080818384 mfu:25.191691559525413 +Step 30755 | loss:0.3724869191646576 lr:5.6807841184151614e-06 tokens_per_second_per_gpu:4290.027993325237 mfu:20.236793998767535 +Step 30756 | loss:0.2898336350917816 lr:5.678776830575716e-06 tokens_per_second_per_gpu:4413.362538537038 mfu:20.818584091574905 +Step 30757 | loss:0.31074994802474976 lr:5.676769876081861e-06 tokens_per_second_per_gpu:4922.936948923637 mfu:23.222333527728114 +Step 30758 | loss:0.2929244935512543 lr:5.674763254948667e-06 tokens_per_second_per_gpu:4971.487125839705 mfu:23.45135299575556 +Step 30759 | loss:0.33539867401123047 lr:5.672756967191228e-06 tokens_per_second_per_gpu:5354.490657983774 mfu:25.258046004020102 +Step 30760 | loss:0.47090569138526917 lr:5.670751012824654e-06 tokens_per_second_per_gpu:4835.390305538839 mfu:22.809361073884407 +Step 30761 | loss:0.41880980134010315 lr:5.6687453918640164e-06 tokens_per_second_per_gpu:5329.7232613289925 mfu:25.141213968245882 +Step 30762 | loss:0.3471818268299103 lr:5.666740104324392e-06 tokens_per_second_per_gpu:5086.995065707884 mfu:23.99622365580049 +Step 30763 | loss:0.2684469521045685 lr:5.664735150220873e-06 tokens_per_second_per_gpu:5375.602013898983 mfu:25.357631871840887 +Step 30764 | loss:0.4253719747066498 lr:5.66273052956855e-06 tokens_per_second_per_gpu:5372.275546633478 mfu:25.341940358195657 +Step 30765 | loss:0.2505991458892822 lr:5.660726242382474e-06 tokens_per_second_per_gpu:5101.9131520818955 mfu:24.0665948145147 +Step 30766 | loss:0.2487495392560959 lr:5.6587222886777366e-06 tokens_per_second_per_gpu:4131.373820370374 mfu:19.488395195746257 +Step 30767 | loss:0.46438679099082947 lr:5.656718668469413e-06 tokens_per_second_per_gpu:5342.275748929079 mfu:25.200426193930713 +Step 30768 | loss:0.2677786350250244 lr:5.654715381772574e-06 tokens_per_second_per_gpu:4741.539880229709 mfu:22.366652605167523 +Step 30769 | loss:0.31317734718322754 lr:5.652712428602264e-06 tokens_per_second_per_gpu:4427.390946069061 mfu:20.884758483396592 +Step 30770 | loss:0.262098103761673 lr:5.6507098089735585e-06 tokens_per_second_per_gpu:4713.7164958548 mfu:22.23540495391219 +Step 30771 | loss:0.2847253084182739 lr:5.648707522901542e-06 tokens_per_second_per_gpu:3264.2938068449193 mfu:15.39823083284138 +Step 30772 | loss:0.22695386409759521 lr:5.64670557040124e-06 tokens_per_second_per_gpu:4118.507178638569 mfu:19.42770105142188 +Step 30773 | loss:0.33234140276908875 lr:5.644703951487728e-06 tokens_per_second_per_gpu:5334.752961549266 mfu:25.16493992234026 +Step 30774 | loss:0.3898566663265228 lr:5.642702666176053e-06 tokens_per_second_per_gpu:5356.17395337665 mfu:25.265986395588104 +Step 30775 | loss:0.28513792157173157 lr:5.640701714481283e-06 tokens_per_second_per_gpu:4957.234191623568 mfu:23.384119473256703 +Step 30776 | loss:0.343144953250885 lr:5.6387010964184585e-06 tokens_per_second_per_gpu:4562.683934718622 mfu:21.522958594220714 +Step 30777 | loss:0.29221072793006897 lr:5.636700812002615e-06 tokens_per_second_per_gpu:5350.673465760909 mfu:25.240039657024692 +Step 30778 | loss:0.2664392292499542 lr:5.634700861248798e-06 tokens_per_second_per_gpu:5321.485575254809 mfu:25.10235539753978 +Step 30779 | loss:0.2820567488670349 lr:5.632701244172073e-06 tokens_per_second_per_gpu:4325.7042386917465 mfu:20.405084935156488 +Step 30780 | loss:0.3279380202293396 lr:5.630701960787455e-06 tokens_per_second_per_gpu:4156.455791514297 mfu:19.606711133057694 +Step 30781 | loss:0.3719662129878998 lr:5.628703011109981e-06 tokens_per_second_per_gpu:4631.879796204577 mfu:21.849367278881335 +Step 30782 | loss:0.2752961218357086 lr:5.626704395154709e-06 tokens_per_second_per_gpu:5065.563431362135 mfu:23.895126979977718 +Step 30783 | loss:0.2898939251899719 lr:5.6247061129366565e-06 tokens_per_second_per_gpu:3967.470898923015 mfu:18.71523715055703 +Step 30784 | loss:0.3412080705165863 lr:5.622708164470841e-06 tokens_per_second_per_gpu:5324.42310731366 mfu:25.116212237456033 +Step 30785 | loss:0.3154412806034088 lr:5.620710549772295e-06 tokens_per_second_per_gpu:4020.7888316026 mfu:18.966746935984 +Step 30786 | loss:0.30684226751327515 lr:5.618713268856063e-06 tokens_per_second_per_gpu:4814.115233033333 mfu:22.709002926974264 +Step 30787 | loss:0.36752307415008545 lr:5.6167163217371365e-06 tokens_per_second_per_gpu:3682.0287828926726 mfu:17.368757987794975 +Step 30788 | loss:0.2285887897014618 lr:5.614719708430555e-06 tokens_per_second_per_gpu:4296.700544474533 mfu:20.26826956099338 +Step 30789 | loss:0.3086954653263092 lr:5.612723428951339e-06 tokens_per_second_per_gpu:3605.011216305264 mfu:17.0054529856503 +Step 30790 | loss:0.2758578658103943 lr:5.610727483314493e-06 tokens_per_second_per_gpu:5353.335988433314 mfu:25.25259923074364 +Step 30791 | loss:0.36453837156295776 lr:5.608731871535023e-06 tokens_per_second_per_gpu:4790.830031817285 mfu:22.599162659973167 +Step 30792 | loss:0.2569718360900879 lr:5.606736593627943e-06 tokens_per_second_per_gpu:5330.42787820052 mfu:25.14453776624891 +Step 30793 | loss:0.29038184881210327 lr:5.604741649608269e-06 tokens_per_second_per_gpu:4175.154679719554 mfu:19.694916979081203 +Step 30794 | loss:0.2586769461631775 lr:5.602747039490991e-06 tokens_per_second_per_gpu:4037.972495440601 mfu:19.047805210143224 +Step 30795 | loss:0.3481326997280121 lr:5.600752763291111e-06 tokens_per_second_per_gpu:5342.0814799752025 mfu:25.19950979412958 +Step 30796 | loss:0.349538654088974 lr:5.598758821023653e-06 tokens_per_second_per_gpu:5325.197796296266 mfu:25.11986657756255 +Step 30797 | loss:0.24719859659671783 lr:5.596765212703592e-06 tokens_per_second_per_gpu:5059.473049293848 mfu:23.866397608635275 +Step 30798 | loss:0.33710604906082153 lr:5.594771938345916e-06 tokens_per_second_per_gpu:4189.0152812835795 mfu:19.760299801521192 +Step 30799 | loss:0.3976988196372986 lr:5.592778997965631e-06 tokens_per_second_per_gpu:3515.6164581355556 mfu:16.583762659045274 +Step 30800 | loss:0.33133378624916077 lr:5.590786391577723e-06 tokens_per_second_per_gpu:4754.312983158649 mfu:22.42690551099942 +Step 30801 | loss:0.3520171642303467 lr:5.588794119197172e-06 tokens_per_second_per_gpu:4249.370052950396 mfu:20.045003557059466 +Step 30802 | loss:0.261082261800766 lr:5.586802180838968e-06 tokens_per_second_per_gpu:4542.0506959563945 mfu:21.425628086585743 +Step 30803 | loss:0.2667205035686493 lr:5.584810576518101e-06 tokens_per_second_per_gpu:5367.188763213548 mfu:25.3179451329091 +Step 30804 | loss:0.325371116399765 lr:5.582819306249542e-06 tokens_per_second_per_gpu:4381.328670569311 mfu:20.667474870829672 +Step 30805 | loss:0.27258363366127014 lr:5.580828370048252e-06 tokens_per_second_per_gpu:3280.705408064837 mfu:15.475647155903635 +Step 30806 | loss:0.3535950779914856 lr:5.578837767929223e-06 tokens_per_second_per_gpu:4811.138374167032 mfu:22.694960575797428 +Step 30807 | loss:0.36162903904914856 lr:5.576847499907439e-06 tokens_per_second_per_gpu:3787.186384689639 mfu:17.864804337208227 +Step 30808 | loss:0.2717016935348511 lr:5.574857565997838e-06 tokens_per_second_per_gpu:3533.5394784694813 mfu:16.668308603943 +Step 30809 | loss:0.2879074811935425 lr:5.5728679662154e-06 tokens_per_second_per_gpu:4305.700371344476 mfu:20.310723279867705 +Step 30810 | loss:0.41965627670288086 lr:5.5708787005750936e-06 tokens_per_second_per_gpu:4684.464242803325 mfu:22.09741708531936 +Step 30811 | loss:0.27049705386161804 lr:5.568889769091884e-06 tokens_per_second_per_gpu:4545.010302573845 mfu:21.439589055961218 +Step 30812 | loss:0.31467679142951965 lr:5.566901171780731e-06 tokens_per_second_per_gpu:5345.794720378494 mfu:25.217025782656048 +Step 30813 | loss:0.2120576798915863 lr:5.564912908656566e-06 tokens_per_second_per_gpu:4887.505586352743 mfu:23.055197745267392 +Step 30814 | loss:0.2942046523094177 lr:5.562924979734369e-06 tokens_per_second_per_gpu:4631.41468487646 mfu:21.84717326939043 +Step 30815 | loss:0.3011445999145508 lr:5.560937385029085e-06 tokens_per_second_per_gpu:3998.841658786407 mfu:18.863218377236205 +Step 30816 | loss:0.33311986923217773 lr:5.5589501245556565e-06 tokens_per_second_per_gpu:5345.950148451123 mfu:25.2177589633934 +Step 30817 | loss:0.3881235420703888 lr:5.556963198329035e-06 tokens_per_second_per_gpu:5292.006965884499 mfu:24.963299767570636 +Step 30818 | loss:0.2617311477661133 lr:5.554976606364176e-06 tokens_per_second_per_gpu:5353.58730586676 mfu:25.25378473795638 +Step 30819 | loss:0.2763553857803345 lr:5.552990348676007e-06 tokens_per_second_per_gpu:3935.4327826660788 mfu:18.564107889906765 +Step 30820 | loss:0.25671127438545227 lr:5.551004425279455e-06 tokens_per_second_per_gpu:5397.428830902844 mfu:25.46059269168709 +Step 30821 | loss:0.2582354247570038 lr:5.5490188361894735e-06 tokens_per_second_per_gpu:5093.02162469512 mfu:24.02465196278028 +Step 30822 | loss:0.3755393624305725 lr:5.547033581421007e-06 tokens_per_second_per_gpu:3867.958489700959 mfu:18.245820137688806 +Step 30823 | loss:0.34788909554481506 lr:5.54504866098896e-06 tokens_per_second_per_gpu:4136.980275397407 mfu:19.514841800668634 +Step 30824 | loss:0.38247597217559814 lr:5.543064074908272e-06 tokens_per_second_per_gpu:4532.6785765495615 mfu:21.38141820029525 +Step 30825 | loss:0.3821551501750946 lr:5.5410798231938845e-06 tokens_per_second_per_gpu:5336.248342544653 mfu:25.171993889633804 +Step 30826 | loss:0.2801392078399658 lr:5.539095905860708e-06 tokens_per_second_per_gpu:5125.351150308354 mfu:24.17715585108376 +Step 30827 | loss:0.2943592369556427 lr:5.537112322923654e-06 tokens_per_second_per_gpu:4975.982770092778 mfu:23.472559716733308 +Step 30828 | loss:0.26554757356643677 lr:5.535129074397655e-06 tokens_per_second_per_gpu:4622.72765470557 mfu:21.80619506591416 +Step 30829 | loss:0.36008334159851074 lr:5.533146160297631e-06 tokens_per_second_per_gpu:5313.853245465116 mfu:25.066352395712055 +Step 30830 | loss:0.23313310742378235 lr:5.531163580638482e-06 tokens_per_second_per_gpu:5340.71767025952 mfu:25.19307647101077 +Step 30831 | loss:0.2373104691505432 lr:5.529181335435124e-06 tokens_per_second_per_gpu:3825.9704084755313 mfu:18.047755194643088 +Step 30832 | loss:0.2764887511730194 lr:5.527199424702473e-06 tokens_per_second_per_gpu:5358.924443221734 mfu:25.278960925469324 +Step 30833 | loss:0.3536258637905121 lr:5.525217848455433e-06 tokens_per_second_per_gpu:4818.971412527063 mfu:22.731910354195847 +Step 30834 | loss:0.20062093436717987 lr:5.523236606708898e-06 tokens_per_second_per_gpu:5382.458322988827 mfu:25.389974251624636 +Step 30835 | loss:0.3338295519351959 lr:5.5212556994777735e-06 tokens_per_second_per_gpu:4203.195550091393 mfu:19.82719054889099 +Step 30836 | loss:0.4105249345302582 lr:5.5192751267769685e-06 tokens_per_second_per_gpu:5284.4597168886785 mfu:24.92769810636389 +Step 30837 | loss:0.40318769216537476 lr:5.517294888621366e-06 tokens_per_second_per_gpu:3510.791869356515 mfu:16.56100425061481 +Step 30838 | loss:0.2955014109611511 lr:5.51531498502586e-06 tokens_per_second_per_gpu:5187.508102406209 mfu:24.470360799198932 +Step 30839 | loss:0.3678472340106964 lr:5.513335416005355e-06 tokens_per_second_per_gpu:5186.395061499075 mfu:24.46511039533562 +Step 30840 | loss:0.2738954722881317 lr:5.511356181574728e-06 tokens_per_second_per_gpu:4941.232489442905 mfu:23.308636713898597 +Step 30841 | loss:0.3505295217037201 lr:5.509377281748873e-06 tokens_per_second_per_gpu:5348.705202490094 mfu:25.23075502335592 +Step 30842 | loss:0.3491954207420349 lr:5.507398716542656e-06 tokens_per_second_per_gpu:4581.170159662304 mfu:21.610161271355278 +Step 30843 | loss:0.27324166893959045 lr:5.5054204859709856e-06 tokens_per_second_per_gpu:4856.974243322403 mfu:22.911176191009 +Step 30844 | loss:0.39597979187965393 lr:5.503442590048713e-06 tokens_per_second_per_gpu:4409.762470151164 mfu:20.801601954763733 +Step 30845 | loss:0.3575243353843689 lr:5.501465028790725e-06 tokens_per_second_per_gpu:5169.727301183015 mfu:24.38648572611161 +Step 30846 | loss:0.34294843673706055 lr:5.4994878022119e-06 tokens_per_second_per_gpu:4235.169805021139 mfu:19.97801856476503 +Step 30847 | loss:0.31487980484962463 lr:5.497510910327114e-06 tokens_per_second_per_gpu:5079.0733922906975 mfu:23.95885577071477 +Step 30848 | loss:0.27496275305747986 lr:5.495534353151227e-06 tokens_per_second_per_gpu:4407.498038465418 mfu:20.790920244150183 +Step 30849 | loss:0.3131113350391388 lr:5.493558130699095e-06 tokens_per_second_per_gpu:4716.212331922413 mfu:22.247178238476664 +Step 30850 | loss:0.3504471182823181 lr:5.491582242985599e-06 tokens_per_second_per_gpu:3317.111818981448 mfu:15.647382407771511 +Step 30851 | loss:0.26892954111099243 lr:5.489606690025584e-06 tokens_per_second_per_gpu:5156.48579722833 mfu:24.324023292723822 +Step 30852 | loss:0.3121819496154785 lr:5.487631471833921e-06 tokens_per_second_per_gpu:3675.9862565939598 mfu:17.340254360282653 +Step 30853 | loss:0.2878120243549347 lr:5.485656588425453e-06 tokens_per_second_per_gpu:4982.391212302939 mfu:23.502789432031623 +Step 30854 | loss:0.3222704529762268 lr:5.483682039815058e-06 tokens_per_second_per_gpu:3888.2175369656825 mfu:18.3413855201864 +Step 30855 | loss:0.38699376583099365 lr:5.481707826017568e-06 tokens_per_second_per_gpu:5272.0919107285235 mfu:24.869357054541116 +Step 30856 | loss:0.4083590805530548 lr:5.479733947047827e-06 tokens_per_second_per_gpu:4372.487598425572 mfu:20.625770025085203 +Step 30857 | loss:0.2601746916770935 lr:5.477760402920684e-06 tokens_per_second_per_gpu:3572.069211426022 mfu:16.850059928149317 +Step 30858 | loss:0.32970741391181946 lr:5.475787193651e-06 tokens_per_second_per_gpu:4760.235486650901 mfu:22.454842970455665 +Step 30859 | loss:0.4260590374469757 lr:5.47381431925359e-06 tokens_per_second_per_gpu:5270.306152028596 mfu:24.86093332607171 +Step 30860 | loss:0.3352344036102295 lr:5.471841779743303e-06 tokens_per_second_per_gpu:4296.2345449219 mfu:20.266071361595895 +Step 30861 | loss:0.33103638887405396 lr:5.469869575134989e-06 tokens_per_second_per_gpu:4435.967332910882 mfu:20.925214763411944 +Step 30862 | loss:0.40233558416366577 lr:5.467897705443464e-06 tokens_per_second_per_gpu:4205.305982869804 mfu:19.837145820383718 +Step 30863 | loss:0.29188454151153564 lr:5.465926170683555e-06 tokens_per_second_per_gpu:4105.586724676261 mfu:19.36675306562491 +Step 30864 | loss:0.4016052484512329 lr:5.463954970870095e-06 tokens_per_second_per_gpu:4803.188519882214 mfu:22.657459756750928 +Step 30865 | loss:0.3129568099975586 lr:5.461984106017926e-06 tokens_per_second_per_gpu:4336.993422007953 mfu:20.458337938992326 +Step 30866 | loss:0.32763543725013733 lr:5.460013576141848e-06 tokens_per_second_per_gpu:5372.721827344934 mfu:25.344045540454935 +Step 30867 | loss:0.324972927570343 lr:5.458043381256689e-06 tokens_per_second_per_gpu:4704.244854185735 mfu:22.190725604130503 +Step 30868 | loss:0.3963399827480316 lr:5.4560735213772805e-06 tokens_per_second_per_gpu:5215.919153977499 mfu:24.60438057688596 +Step 30869 | loss:0.3267832100391388 lr:5.454103996518417e-06 tokens_per_second_per_gpu:3693.6148637439064 mfu:17.42341150795948 +Step 30870 | loss:0.2820369601249695 lr:5.4521348066949305e-06 tokens_per_second_per_gpu:4643.322830994679 mfu:21.903346026369302 +Step 30871 | loss:0.2552241086959839 lr:5.45016595192161e-06 tokens_per_second_per_gpu:4918.992428972688 mfu:23.203726554114986 +Step 30872 | loss:0.27236080169677734 lr:5.448197432213287e-06 tokens_per_second_per_gpu:4393.020354932556 mfu:20.722626540778148 +Step 30873 | loss:0.22316521406173706 lr:5.4462292475847496e-06 tokens_per_second_per_gpu:5348.776199027546 mfu:25.231089926136335 +Step 30874 | loss:0.394447922706604 lr:5.444261398050804e-06 tokens_per_second_per_gpu:5323.775149861936 mfu:25.113155711603312 +Step 30875 | loss:0.3134565055370331 lr:5.442293883626265e-06 tokens_per_second_per_gpu:4374.907808081164 mfu:20.637186566960974 +Step 30876 | loss:0.20157228410243988 lr:5.44032670432591e-06 tokens_per_second_per_gpu:4808.581465438754 mfu:22.682899201072523 +Step 30877 | loss:0.25202253460884094 lr:5.438359860164555e-06 tokens_per_second_per_gpu:3726.725933220413 mfu:17.579602072011156 +Step 30878 | loss:0.31610745191574097 lr:5.436393351156971e-06 tokens_per_second_per_gpu:4774.631533727991 mfu:22.522751580737204 +Step 30879 | loss:0.24825531244277954 lr:5.434427177317963e-06 tokens_per_second_per_gpu:3844.9225192580484 mfu:18.137155534767754 +Step 30880 | loss:0.37563884258270264 lr:5.432461338662309e-06 tokens_per_second_per_gpu:3296.0721258982926 mfu:15.548134585756465 +Step 30881 | loss:0.3225812613964081 lr:5.430495835204802e-06 tokens_per_second_per_gpu:4028.412321958118 mfu:19.00270824069256 +Step 30882 | loss:0.3776073157787323 lr:5.4285306669602294e-06 tokens_per_second_per_gpu:5128.968640742252 mfu:24.194220170667972 +Step 30883 | loss:0.2566679120063782 lr:5.426565833943359e-06 tokens_per_second_per_gpu:3420.0107109171977 mfu:16.132774037393762 +Step 30884 | loss:0.3653694689273834 lr:5.4246013361689825e-06 tokens_per_second_per_gpu:4318.748279812158 mfu:20.37227249031251 +Step 30885 | loss:0.2427750527858734 lr:5.422637173651857e-06 tokens_per_second_per_gpu:5383.891142606032 mfu:25.396733106223227 +Step 30886 | loss:0.398433119058609 lr:5.420673346406779e-06 tokens_per_second_per_gpu:4334.172333827937 mfu:20.445030384719914 +Step 30887 | loss:0.34339267015457153 lr:5.418709854448489e-06 tokens_per_second_per_gpu:5351.683044496529 mfu:25.24480201217251 +Step 30888 | loss:0.25685015320777893 lr:5.416746697791775e-06 tokens_per_second_per_gpu:5041.600525421559 mfu:23.782089864162675 +Step 30889 | loss:0.26569318771362305 lr:5.4147838764513955e-06 tokens_per_second_per_gpu:5351.670767591287 mfu:25.244744099915827 +Step 30890 | loss:0.3713371157646179 lr:5.41282139044213e-06 tokens_per_second_per_gpu:3835.774515581494 mfu:18.094002840615076 +Step 30891 | loss:0.30203911662101746 lr:5.410859239778721e-06 tokens_per_second_per_gpu:3229.9657263165454 mfu:15.23629942001455 +Step 30892 | loss:0.25936585664749146 lr:5.408897424475917e-06 tokens_per_second_per_gpu:5257.950768242451 mfu:24.802650872706614 +Step 30893 | loss:0.2719007432460785 lr:5.406935944548486e-06 tokens_per_second_per_gpu:5278.914926035852 mfu:24.901542381872236 +Step 30894 | loss:0.35415133833885193 lr:5.404974800011192e-06 tokens_per_second_per_gpu:3050.333238928681 mfu:14.388942328543276 +Step 30895 | loss:0.2503676116466522 lr:5.4030139908787636e-06 tokens_per_second_per_gpu:4927.679591382705 mfu:23.244705381385785 +Step 30896 | loss:0.35944169759750366 lr:5.401053517165955e-06 tokens_per_second_per_gpu:4070.9246754913597 mfu:19.203246265664312 +Step 30897 | loss:0.31979265809059143 lr:5.399093378887526e-06 tokens_per_second_per_gpu:4369.659140780145 mfu:20.612427707787084 +Step 30898 | loss:0.19509366154670715 lr:5.397133576058194e-06 tokens_per_second_per_gpu:5335.76909811678 mfu:25.169733211899995 +Step 30899 | loss:0.38946008682250977 lr:5.395174108692724e-06 tokens_per_second_per_gpu:5063.697599171179 mfu:23.886325531189176 +Step 30900 | loss:0.40233859419822693 lr:5.393214976805833e-06 tokens_per_second_per_gpu:5368.458669476074 mfu:25.323935497417782 +Step 30901 | loss:0.34357723593711853 lr:5.391256180412269e-06 tokens_per_second_per_gpu:5267.111598907046 mfu:24.84586407395039 +Step 30902 | loss:0.28440535068511963 lr:5.389297719526754e-06 tokens_per_second_per_gpu:3942.5412194897544 mfu:18.59763959922843 +Step 30903 | loss:0.36404287815093994 lr:5.387339594164021e-06 tokens_per_second_per_gpu:4967.957829876405 mfu:23.434704704536692 +Step 30904 | loss:0.3834129869937897 lr:5.385381804338807e-06 tokens_per_second_per_gpu:4367.140138441346 mfu:20.600545143968855 +Step 30905 | loss:0.32857346534729004 lr:5.383424350065824e-06 tokens_per_second_per_gpu:5338.457167605293 mfu:25.182413294309264 +Step 30906 | loss:0.2389463186264038 lr:5.381467231359805e-06 tokens_per_second_per_gpu:5323.273009431162 mfu:25.110787029517354 +Step 30907 | loss:0.30429890751838684 lr:5.379510448235459e-06 tokens_per_second_per_gpu:3776.427243266409 mfu:17.81405163141603 +Step 30908 | loss:0.3698371648788452 lr:5.377554000707519e-06 tokens_per_second_per_gpu:5349.572648248105 mfu:25.234846912997963 +Step 30909 | loss:0.2992856204509735 lr:5.375597888790673e-06 tokens_per_second_per_gpu:5332.8732015833175 mfu:25.156072773860757 +Step 30910 | loss:0.3116510808467865 lr:5.373642112499655e-06 tokens_per_second_per_gpu:4513.966496564183 mfu:21.293150126394163 +Step 30911 | loss:0.39028918743133545 lr:5.3716866718491745e-06 tokens_per_second_per_gpu:4595.419828669362 mfu:21.6773793913058 +Step 30912 | loss:0.25679677724838257 lr:5.36973156685393e-06 tokens_per_second_per_gpu:5359.913129284527 mfu:25.283624726314756 +Step 30913 | loss:0.3133031129837036 lr:5.367776797528634e-06 tokens_per_second_per_gpu:5349.402215160017 mfu:25.234042951042568 +Step 30914 | loss:0.34117743372917175 lr:5.365822363887974e-06 tokens_per_second_per_gpu:5350.203600409295 mfu:25.237823221993835 +Step 30915 | loss:0.3099936246871948 lr:5.36386826594667e-06 tokens_per_second_per_gpu:4534.506916275869 mfu:21.390042790731172 +Step 30916 | loss:0.31178638339042664 lr:5.3619145037194e-06 tokens_per_second_per_gpu:4620.035521548449 mfu:21.79349581448674 +Step 30917 | loss:0.3380531370639801 lr:5.359961077220865e-06 tokens_per_second_per_gpu:4847.945844932657 mfu:22.86858769540096 +Step 30918 | loss:0.5204499363899231 lr:5.358007986465768e-06 tokens_per_second_per_gpu:4126.552473435695 mfu:19.46565207964876 +Step 30919 | loss:0.25121983885765076 lr:5.356055231468782e-06 tokens_per_second_per_gpu:3531.8236540757703 mfu:16.660214767528036 +Step 30920 | loss:0.2575508654117584 lr:5.35410281224461e-06 tokens_per_second_per_gpu:5091.575532115148 mfu:24.017830497351714 +Step 30921 | loss:0.27469199895858765 lr:5.352150728807915e-06 tokens_per_second_per_gpu:5392.114988223005 mfu:25.435526389131237 +Step 30922 | loss:0.37526223063468933 lr:5.350198981173404e-06 tokens_per_second_per_gpu:5212.6150027652575 mfu:24.588794331871483 +Step 30923 | loss:0.3759937286376953 lr:5.348247569355735e-06 tokens_per_second_per_gpu:3884.284739778984 mfu:18.32283384485235 +Step 30924 | loss:0.3237159848213196 lr:5.34629649336959e-06 tokens_per_second_per_gpu:3698.2810713358904 mfu:17.44542280530798 +Step 30925 | loss:0.32395559549331665 lr:5.344345753229652e-06 tokens_per_second_per_gpu:3649.5538727633716 mfu:17.215568295924047 +Step 30926 | loss:0.4041728973388672 lr:5.342395348950597e-06 tokens_per_second_per_gpu:4734.407053620444 mfu:22.333005845065905 +Step 30927 | loss:0.2735784649848938 lr:5.34044528054708e-06 tokens_per_second_per_gpu:3848.1364132431754 mfu:18.152316021042456 +Step 30928 | loss:0.20477797091007233 lr:5.338495548033767e-06 tokens_per_second_per_gpu:5364.901997718457 mfu:25.30715806990633 +Step 30929 | loss:0.2987392842769623 lr:5.336546151425326e-06 tokens_per_second_per_gpu:4782.92994347756 mfu:22.561896595380627 +Step 30930 | loss:0.2508370280265808 lr:5.3345970907364316e-06 tokens_per_second_per_gpu:4483.9917902292 mfu:21.15175432239979 +Step 30931 | loss:0.24655798077583313 lr:5.332648365981718e-06 tokens_per_second_per_gpu:3482.31398499246 mfu:16.426669211241467 +Step 30932 | loss:0.23830397427082062 lr:5.3306999771758625e-06 tokens_per_second_per_gpu:5342.025722708411 mfu:25.199246777588954 +Step 30933 | loss:0.3433816432952881 lr:5.3287519243335194e-06 tokens_per_second_per_gpu:4298.957148383004 mfu:20.27891434664597 +Step 30934 | loss:0.34040892124176025 lr:5.3268042074693216e-06 tokens_per_second_per_gpu:5355.136786641841 mfu:25.26109390314112 +Step 30935 | loss:0.3219332695007324 lr:5.32485682659794e-06 tokens_per_second_per_gpu:5329.462596697319 mfu:25.139984368704305 +Step 30936 | loss:0.35758495330810547 lr:5.322909781733998e-06 tokens_per_second_per_gpu:4601.022411379658 mfu:21.703807729849448 +Step 30937 | loss:0.2591034770011902 lr:5.320963072892166e-06 tokens_per_second_per_gpu:5303.968219940197 mfu:25.01972304375175 +Step 30938 | loss:0.36915579438209534 lr:5.319016700087065e-06 tokens_per_second_per_gpu:5356.06078931111 mfu:25.265452581383173 +Step 30939 | loss:0.30708739161491394 lr:5.317070663333335e-06 tokens_per_second_per_gpu:4309.154879891786 mfu:20.3270187860853 +Step 30940 | loss:0.4461117088794708 lr:5.315124962645629e-06 tokens_per_second_per_gpu:5222.463687868431 mfu:24.635252259862106 +Step 30941 | loss:0.3143903315067291 lr:5.313179598038565e-06 tokens_per_second_per_gpu:4191.063945243942 mfu:19.769963698960336 +Step 30942 | loss:0.2938098907470703 lr:5.3112345695267795e-06 tokens_per_second_per_gpu:5357.5480025646775 mfu:25.272468020044997 +Step 30943 | loss:0.29516705870628357 lr:5.3092898771249015e-06 tokens_per_second_per_gpu:5322.429801850173 mfu:25.106809475491787 +Step 30944 | loss:0.2770465910434723 lr:5.307345520847562e-06 tokens_per_second_per_gpu:4261.399394913556 mfu:20.101748015517067 +Step 30945 | loss:0.2707405686378479 lr:5.305401500709367e-06 tokens_per_second_per_gpu:3977.5891213737127 mfu:18.762966532203738 +Step 30946 | loss:0.29657840728759766 lr:5.303457816724955e-06 tokens_per_second_per_gpu:3388.3994366110323 mfu:15.983658263052028 +Step 30947 | loss:0.3627457618713379 lr:5.301514468908947e-06 tokens_per_second_per_gpu:5044.876649663665 mfu:23.797543901177125 +Step 30948 | loss:0.24158842861652374 lr:5.299571457275948e-06 tokens_per_second_per_gpu:2836.706236288305 mfu:13.381227308563517 +Step 30949 | loss:0.2625593841075897 lr:5.29762878184058e-06 tokens_per_second_per_gpu:4901.299944980483 mfu:23.120268088474713 +Step 30950 | loss:0.302772581577301 lr:5.295686442617443e-06 tokens_per_second_per_gpu:5297.917922933612 mfu:24.991182760484573 +Step 30951 | loss:0.3765299320220947 lr:5.293744439621162e-06 tokens_per_second_per_gpu:3493.4129773059262 mfu:16.479025051667445 +Step 30952 | loss:0.26946136355400085 lr:5.291802772866322e-06 tokens_per_second_per_gpu:4088.7121914125473 mfu:19.287153013127718 +Step 30953 | loss:0.25414928793907166 lr:5.289861442367538e-06 tokens_per_second_per_gpu:4766.006412419587 mfu:22.48206541192802 +Step 30954 | loss:0.3079463541507721 lr:5.287920448139422e-06 tokens_per_second_per_gpu:5360.063400548159 mfu:25.28433358150423 +Step 30955 | loss:0.30371925234794617 lr:5.2859797901965545e-06 tokens_per_second_per_gpu:4252.890667986252 mfu:20.06161089886246 +Step 30956 | loss:0.33480286598205566 lr:5.284039468553542e-06 tokens_per_second_per_gpu:4844.384104926917 mfu:22.851786360098377 +Step 30957 | loss:0.3520115315914154 lr:5.2820994832249725e-06 tokens_per_second_per_gpu:3961.794074938479 mfu:18.688458603255928 +Step 30958 | loss:0.20726923644542694 lr:5.28015983422544e-06 tokens_per_second_per_gpu:4050.796800278094 mfu:19.108299644113707 +Step 30959 | loss:0.35318854451179504 lr:5.278220521569527e-06 tokens_per_second_per_gpu:5317.104842324366 mfu:25.08169073287677 +Step 30960 | loss:0.37045788764953613 lr:5.276281545271827e-06 tokens_per_second_per_gpu:4974.2533662401775 mfu:23.46440182369387 +Step 30961 | loss:0.27315574884414673 lr:5.274342905346913e-06 tokens_per_second_per_gpu:5091.128245809401 mfu:24.01572057153259 +Step 30962 | loss:0.469758540391922 lr:5.272404601809389e-06 tokens_per_second_per_gpu:5091.913537419509 mfu:24.01942492604204 +Step 30963 | loss:0.2992352247238159 lr:5.270466634673804e-06 tokens_per_second_per_gpu:4039.727939311829 mfu:19.056085938393412 +Step 30964 | loss:0.275787889957428 lr:5.2685290039547585e-06 tokens_per_second_per_gpu:5381.334504577334 mfu:25.38467300843443 +Step 30965 | loss:0.3311891257762909 lr:5.266591709666801e-06 tokens_per_second_per_gpu:4906.395384264789 mfu:23.144304144950375 +Step 30966 | loss:0.2757083475589752 lr:5.264654751824533e-06 tokens_per_second_per_gpu:4401.2896687016155 mfu:20.761634304717216 +Step 30967 | loss:0.2898891568183899 lr:5.262718130442484e-06 tokens_per_second_per_gpu:5319.9332628494185 mfu:25.095032875071748 +Step 30968 | loss:0.22859632968902588 lr:5.260781845535251e-06 tokens_per_second_per_gpu:4896.907776649106 mfu:23.09954948107442 +Step 30969 | loss:0.24294441938400269 lr:5.258845897117387e-06 tokens_per_second_per_gpu:4349.910880950786 mfu:20.51927179677143 +Step 30970 | loss:0.25614041090011597 lr:5.256910285203448e-06 tokens_per_second_per_gpu:5354.95134334398 mfu:25.260219135465217 +Step 30971 | loss:0.45199334621429443 lr:5.254975009808005e-06 tokens_per_second_per_gpu:4403.215827232407 mfu:20.770720323143493 +Step 30972 | loss:0.28676047921180725 lr:5.253040070945597e-06 tokens_per_second_per_gpu:4119.79110180738 mfu:19.433757536069045 +Step 30973 | loss:0.4049418568611145 lr:5.25110546863079e-06 tokens_per_second_per_gpu:5357.308651763635 mfu:25.27133896147902 +Step 30974 | loss:0.2720869779586792 lr:5.2491712028781224e-06 tokens_per_second_per_gpu:5353.858278843694 mfu:25.255062963721336 +Step 30975 | loss:0.2742438316345215 lr:5.247237273702143e-06 tokens_per_second_per_gpu:3486.9069637522243 mfu:16.448335075694256 +Step 30976 | loss:0.2660723328590393 lr:5.245303681117414e-06 tokens_per_second_per_gpu:3622.5865765092317 mfu:17.08835895839808 +Step 30977 | loss:0.3439581096172333 lr:5.2433704251384605e-06 tokens_per_second_per_gpu:5261.3150482762985 mfu:24.81852075563163 +Step 30978 | loss:0.23837421834468842 lr:5.241437505779839e-06 tokens_per_second_per_gpu:3762.557320650133 mfu:17.74862483998224 +Step 30979 | loss:0.2856811583042145 lr:5.239504923056066e-06 tokens_per_second_per_gpu:3987.037859789056 mfu:18.80753784342992 +Step 30980 | loss:0.3574775755405426 lr:5.237572676981694e-06 tokens_per_second_per_gpu:4903.471462262786 mfu:23.13051150599439 +Step 30981 | loss:0.3360128402709961 lr:5.235640767571243e-06 tokens_per_second_per_gpu:4754.230700345839 mfu:22.426517368932497 +Step 30982 | loss:0.28815704584121704 lr:5.2337091948392534e-06 tokens_per_second_per_gpu:4783.681296978489 mfu:22.565440858038627 +Step 30983 | loss:0.19181868433952332 lr:5.231777958800255e-06 tokens_per_second_per_gpu:4700.368111756176 mfu:22.172438348649813 +Step 30984 | loss:0.3620963990688324 lr:5.2298470594687565e-06 tokens_per_second_per_gpu:4723.494275261737 mfu:22.281528407635907 +Step 30985 | loss:0.3400464355945587 lr:5.227916496859303e-06 tokens_per_second_per_gpu:5005.47687816405 mfu:23.61168845671924 +Step 30986 | loss:0.24081066250801086 lr:5.2259862709863885e-06 tokens_per_second_per_gpu:4428.231487362333 mfu:20.888723460088084 +Step 30987 | loss:0.463612824678421 lr:5.224056381864556e-06 tokens_per_second_per_gpu:5345.163855910743 mfu:25.21404988732509 +Step 30988 | loss:0.24263660609722137 lr:5.2221268295083e-06 tokens_per_second_per_gpu:5333.024157543638 mfu:25.15678485888133 +Step 30989 | loss:0.3027133047580719 lr:5.220197613932143e-06 tokens_per_second_per_gpu:5350.4609832232145 mfu:25.23903734064113 +Step 30990 | loss:0.23374004662036896 lr:5.218268735150605e-06 tokens_per_second_per_gpu:3981.8584915655756 mfu:18.783105879828426 +Step 30991 | loss:0.3787403106689453 lr:5.216340193178171e-06 tokens_per_second_per_gpu:4514.804565617405 mfu:21.297103441107538 +Step 30992 | loss:0.30290091037750244 lr:5.214411988029355e-06 tokens_per_second_per_gpu:4004.080412023315 mfu:18.88793047007817 +Step 30993 | loss:0.2245279997587204 lr:5.2124841197186695e-06 tokens_per_second_per_gpu:3261.2890734417324 mfu:15.384056992718163 +Step 30994 | loss:0.24507847428321838 lr:5.210556588260612e-06 tokens_per_second_per_gpu:4335.756673178622 mfu:20.45250398375326 +Step 30995 | loss:0.3084431290626526 lr:5.208629393669656e-06 tokens_per_second_per_gpu:3863.1017437584105 mfu:18.222910038431497 +Step 30996 | loss:0.2161458283662796 lr:5.206702535960317e-06 tokens_per_second_per_gpu:5342.438550681852 mfu:25.201194157575387 +Step 30997 | loss:0.28085559606552124 lr:5.2047760151470835e-06 tokens_per_second_per_gpu:5029.346604785141 mfu:23.724286029786242 +Step 30998 | loss:0.37120288610458374 lr:5.202849831244461e-06 tokens_per_second_per_gpu:4801.357486402486 mfu:22.648822459420526 +Step 30999 | loss:0.21454255282878876 lr:5.200923984266904e-06 tokens_per_second_per_gpu:4037.6100473150846 mfu:19.046095480495133 +Step 31000 | loss:0.370761901140213 lr:5.1989984742289275e-06 tokens_per_second_per_gpu:5375.454506659804 mfu:25.356936055026345 +Step 31001 | loss:0.20003949105739594 lr:5.197073301144989e-06 tokens_per_second_per_gpu:4744.227192436616 mfu:22.379329115350245 +Step 31002 | loss:0.30014321208000183 lr:5.195148465029586e-06 tokens_per_second_per_gpu:4644.094628993412 mfu:21.906986729211646 +Step 31003 | loss:0.36973822116851807 lr:5.193223965897187e-06 tokens_per_second_per_gpu:4413.0178744971145 mfu:20.81695825249284 +Step 31004 | loss:0.39259856939315796 lr:5.191299803762262e-06 tokens_per_second_per_gpu:4982.704833479727 mfu:23.504268836631827 +Step 31005 | loss:0.2693931460380554 lr:5.1893759786392995e-06 tokens_per_second_per_gpu:4628.875955648734 mfu:21.835197650471372 +Step 31006 | loss:0.3636646568775177 lr:5.187452490542749e-06 tokens_per_second_per_gpu:5349.284850981892 mfu:25.23348932418996 +Step 31007 | loss:0.2406589239835739 lr:5.185529339487094e-06 tokens_per_second_per_gpu:4078.278925479549 mfu:19.23793751762304 +Step 31008 | loss:0.3167775571346283 lr:5.1836065254867836e-06 tokens_per_second_per_gpu:3505.5937310356207 mfu:16.536483745261304 +Step 31009 | loss:0.37369629740715027 lr:5.181684048556295e-06 tokens_per_second_per_gpu:3660.1462140805916 mfu:17.26553417715716 +Step 31010 | loss:0.29703831672668457 lr:5.179761908710068e-06 tokens_per_second_per_gpu:4671.934659407672 mfu:22.0383129026747 +Step 31011 | loss:0.2226226031780243 lr:5.177840105962578e-06 tokens_per_second_per_gpu:5336.678558674019 mfu:25.174023292518463 +Step 31012 | loss:0.3570074439048767 lr:5.175918640328276e-06 tokens_per_second_per_gpu:5347.757657163976 mfu:25.226285290384713 +Step 31013 | loss:0.3389036953449249 lr:5.173997511821599e-06 tokens_per_second_per_gpu:5303.902175361135 mfu:25.019411500203947 +Step 31014 | loss:0.29088443517684937 lr:5.172076720457019e-06 tokens_per_second_per_gpu:2895.093566154995 mfu:13.65665030545041 +Step 31015 | loss:0.27454107999801636 lr:5.17015626624896e-06 tokens_per_second_per_gpu:5372.874522629293 mfu:25.34476583016376 +Step 31016 | loss:0.26760080456733704 lr:5.168236149211886e-06 tokens_per_second_per_gpu:5138.654348370573 mfu:24.239909306102888 +Step 31017 | loss:0.3315599858760834 lr:5.166316369360219e-06 tokens_per_second_per_gpu:4979.1621504925 mfu:23.487557396537028 +Step 31018 | loss:0.23879073560237885 lr:5.164396926708409e-06 tokens_per_second_per_gpu:4589.929106823353 mfu:21.651478719544432 +Step 31019 | loss:0.28463542461395264 lr:5.162477821270895e-06 tokens_per_second_per_gpu:4069.615173975883 mfu:19.197069123591813 +Step 31020 | loss:0.3477199971675873 lr:5.1605590530621025e-06 tokens_per_second_per_gpu:4708.42857818656 mfu:22.210460944059317 +Step 31021 | loss:0.3369295597076416 lr:5.158640622096472e-06 tokens_per_second_per_gpu:4288.509949439872 mfu:20.22963312675527 +Step 31022 | loss:0.3366940915584564 lr:5.156722528388419e-06 tokens_per_second_per_gpu:3965.593704290018 mfu:18.706382103190755 +Step 31023 | loss:0.26117345690727234 lr:5.1548047719523876e-06 tokens_per_second_per_gpu:3545.399251173011 mfu:16.72425317527143 +Step 31024 | loss:0.32369673252105713 lr:5.1528873528027874e-06 tokens_per_second_per_gpu:3140.545740574099 mfu:14.814490090644648 +Step 31025 | loss:0.2822476923465729 lr:5.150970270954036e-06 tokens_per_second_per_gpu:4728.840057954308 mfu:22.306745376681338 +Step 31026 | loss:0.28177520632743835 lr:5.149053526420572e-06 tokens_per_second_per_gpu:3797.3008671109224 mfu:17.912516076498072 +Step 31027 | loss:0.2697940766811371 lr:5.147137119216794e-06 tokens_per_second_per_gpu:4554.223272687288 mfu:21.483048207881115 +Step 31028 | loss:0.3447378873825073 lr:5.145221049357118e-06 tokens_per_second_per_gpu:4597.712212872924 mfu:21.68819295871508 +Step 31029 | loss:0.34750935435295105 lr:5.143305316855967e-06 tokens_per_second_per_gpu:5341.912821538743 mfu:25.1987142035087 +Step 31030 | loss:0.2851528823375702 lr:5.141389921727741e-06 tokens_per_second_per_gpu:5335.221249507084 mfu:25.16714891653545 +Step 31031 | loss:0.3062285780906677 lr:5.139474863986832e-06 tokens_per_second_per_gpu:4300.24361251738 mfu:20.284982817461238 +Step 31032 | loss:0.33705782890319824 lr:5.137560143647663e-06 tokens_per_second_per_gpu:3667.522780374445 mfu:17.30033069893726 +Step 31033 | loss:0.23607580363750458 lr:5.135645760724628e-06 tokens_per_second_per_gpu:5132.990050158656 mfu:24.213189845008348 +Step 31034 | loss:0.29631268978118896 lr:5.1337317152321315e-06 tokens_per_second_per_gpu:4163.321433357782 mfu:19.639097537032704 +Step 31035 | loss:0.22413817048072815 lr:5.131818007184558e-06 tokens_per_second_per_gpu:5360.425077730034 mfu:25.286039674479625 +Step 31036 | loss:0.30973899364471436 lr:5.129904636596311e-06 tokens_per_second_per_gpu:3687.877779059856 mfu:17.39634870065723 +Step 31037 | loss:0.341564804315567 lr:5.1279916034817746e-06 tokens_per_second_per_gpu:5360.208883478469 mfu:25.285019849308448 +Step 31038 | loss:0.3202191889286041 lr:5.126078907855342e-06 tokens_per_second_per_gpu:5322.994633568303 mfu:25.10947388307643 +Step 31039 | loss:0.32560575008392334 lr:5.124166549731391e-06 tokens_per_second_per_gpu:4346.294656650914 mfu:20.502213449757445 +Step 31040 | loss:0.315540075302124 lr:5.122254529124304e-06 tokens_per_second_per_gpu:5359.259477035624 mfu:25.280541337132107 +Step 31041 | loss:0.2083570957183838 lr:5.120342846048481e-06 tokens_per_second_per_gpu:4475.452383308857 mfu:21.11147248298347 +Step 31042 | loss:0.2809280455112457 lr:5.118431500518278e-06 tokens_per_second_per_gpu:4047.541735620954 mfu:19.09294494875468 +Step 31043 | loss:0.31159070134162903 lr:5.116520492548082e-06 tokens_per_second_per_gpu:4679.528994049038 mfu:22.074136674905127 +Step 31044 | loss:0.20198984444141388 lr:5.114609822152261e-06 tokens_per_second_per_gpu:5381.66136220538 mfu:25.386214851633586 +Step 31045 | loss:0.266377329826355 lr:5.112699489345191e-06 tokens_per_second_per_gpu:3681.3488511478376 mfu:17.365550633746274 +Step 31046 | loss:0.23602010309696198 lr:5.110789494141227e-06 tokens_per_second_per_gpu:5391.4836284251205 mfu:25.432548157243527 +Step 31047 | loss:0.26629266142845154 lr:5.108879836554742e-06 tokens_per_second_per_gpu:5330.007631715199 mfu:25.142555391876208 +Step 31048 | loss:0.30365556478500366 lr:5.1069705166001124e-06 tokens_per_second_per_gpu:4627.896360432005 mfu:21.830576732697885 +Step 31049 | loss:0.34320083260536194 lr:5.1050615342916775e-06 tokens_per_second_per_gpu:5368.516803305372 mfu:25.32420972460924 +Step 31050 | loss:0.256317675113678 lr:5.103152889643809e-06 tokens_per_second_per_gpu:3367.482534352793 mfu:15.884989666308385 +Step 31051 | loss:0.28284966945648193 lr:5.10124458267085e-06 tokens_per_second_per_gpu:3852.8425140183667 mfu:18.174515501342622 +Step 31052 | loss:0.29198306798934937 lr:5.099336613387168e-06 tokens_per_second_per_gpu:5383.930000111497 mfu:25.39691640370628 +Step 31053 | loss:0.33068084716796875 lr:5.09742898180709e-06 tokens_per_second_per_gpu:3529.4046865738924 mfu:16.648804085103265 +Step 31054 | loss:0.32641857862472534 lr:5.095521687944987e-06 tokens_per_second_per_gpu:4087.18784394072 mfu:19.27996240602261 +Step 31055 | loss:0.3122449219226837 lr:5.0936147318152e-06 tokens_per_second_per_gpu:5305.361299207817 mfu:25.026294436340944 +Step 31056 | loss:0.23137754201889038 lr:5.091708113432059e-06 tokens_per_second_per_gpu:4552.794317215808 mfu:21.476307581117204 +Step 31057 | loss:0.17761942744255066 lr:5.08980183280991e-06 tokens_per_second_per_gpu:4965.082861658007 mfu:23.421142988930345 +Step 31058 | loss:0.3545021712779999 lr:5.087895889963102e-06 tokens_per_second_per_gpu:4051.549020873524 mfu:19.11184799700442 +Step 31059 | loss:0.35579681396484375 lr:5.085990284905956e-06 tokens_per_second_per_gpu:4442.335555675986 mfu:20.955254756725783 +Step 31060 | loss:0.41022491455078125 lr:5.084085017652807e-06 tokens_per_second_per_gpu:4489.944494399599 mfu:21.17983424360762 +Step 31061 | loss:0.3391333222389221 lr:5.0821800882179805e-06 tokens_per_second_per_gpu:5342.744639778357 mfu:25.202638032067785 +Step 31062 | loss:0.3559889495372772 lr:5.080275496615816e-06 tokens_per_second_per_gpu:4136.321714340076 mfu:19.511735255799103 +Step 31063 | loss:0.2653624415397644 lr:5.07837124286063e-06 tokens_per_second_per_gpu:5014.844535833376 mfu:23.65587729623266 +Step 31064 | loss:0.29569128155708313 lr:5.076467326966738e-06 tokens_per_second_per_gpu:5358.847716687994 mfu:25.278598993318774 +Step 31065 | loss:0.3459441661834717 lr:5.07456374894848e-06 tokens_per_second_per_gpu:5354.35673311607 mfu:25.257414257569202 +Step 31066 | loss:0.2941533923149109 lr:5.072660508820165e-06 tokens_per_second_per_gpu:4850.075643488761 mfu:22.87863431857121 +Step 31067 | loss:0.35941386222839355 lr:5.070757606596088e-06 tokens_per_second_per_gpu:5343.499645495868 mfu:25.206199522854547 +Step 31068 | loss:0.29239749908447266 lr:5.068855042290576e-06 tokens_per_second_per_gpu:4761.045931776804 mfu:22.45866597839074 +Step 31069 | loss:0.35006973147392273 lr:5.066952815917941e-06 tokens_per_second_per_gpu:5327.134891575496 mfu:25.129004186497216 +Step 31070 | loss:0.34687766432762146 lr:5.0650509274924925e-06 tokens_per_second_per_gpu:5023.3088598499635 mfu:23.695804956781497 +Step 31071 | loss:0.2982555627822876 lr:5.063149377028525e-06 tokens_per_second_per_gpu:5259.83807060025 mfu:24.811553599935337 +Step 31072 | loss:0.28686362504959106 lr:5.061248164540355e-06 tokens_per_second_per_gpu:5004.43598668795 mfu:23.60677839403208 +Step 31073 | loss:0.34200266003608704 lr:5.059347290042271e-06 tokens_per_second_per_gpu:5365.419656659716 mfu:25.30959995545397 +Step 31074 | loss:0.3277580738067627 lr:5.057446753548556e-06 tokens_per_second_per_gpu:5317.178576619992 mfu:25.082038550129607 +Step 31075 | loss:0.2509017884731293 lr:5.0555465550735256e-06 tokens_per_second_per_gpu:3861.20161302631 mfu:18.213946797572447 +Step 31076 | loss:0.22310830652713776 lr:5.053646694631458e-06 tokens_per_second_per_gpu:4343.978494050231 mfu:20.491287715591934 +Step 31077 | loss:0.4209733009338379 lr:5.051747172236665e-06 tokens_per_second_per_gpu:4478.941714162848 mfu:21.12793225196298 +Step 31078 | loss:0.26370036602020264 lr:5.0498479879034056e-06 tokens_per_second_per_gpu:5092.317075478234 mfu:24.021328483915735 +Step 31079 | loss:0.2904900014400482 lr:5.04794914164598e-06 tokens_per_second_per_gpu:4332.97291893676 mfu:20.43937254003695 +Step 31080 | loss:0.26867809891700745 lr:5.046050633478661e-06 tokens_per_second_per_gpu:4165.829453800232 mfu:19.650928297372154 +Step 31081 | loss:0.3055878281593323 lr:5.044152463415736e-06 tokens_per_second_per_gpu:3946.1132964138246 mfu:18.61448969553844 +Step 31082 | loss:0.37638282775878906 lr:5.0422546314714715e-06 tokens_per_second_per_gpu:4778.872809324218 mfu:22.542758401360846 +Step 31083 | loss:0.22972488403320312 lr:5.040357137660145e-06 tokens_per_second_per_gpu:3857.9376312060394 mfu:18.19855003843351 +Step 31084 | loss:0.31037989258766174 lr:5.038459981996035e-06 tokens_per_second_per_gpu:3777.711547371049 mfu:17.820109913002497 +Step 31085 | loss:0.2612989842891693 lr:5.036563164493402e-06 tokens_per_second_per_gpu:5376.996696757766 mfu:25.364210828843195 +Step 31086 | loss:0.27674931287765503 lr:5.034666685166511e-06 tokens_per_second_per_gpu:5358.769451422904 mfu:25.278229802709916 +Step 31087 | loss:0.35197994112968445 lr:5.032770544029641e-06 tokens_per_second_per_gpu:5109.642999900472 mfu:24.10305782552296 +Step 31088 | loss:0.2564471662044525 lr:5.030874741097036e-06 tokens_per_second_per_gpu:4921.397741411447 mfu:23.215072823277232 +Step 31089 | loss:0.33530473709106445 lr:5.028979276382956e-06 tokens_per_second_per_gpu:5242.226849378724 mfu:24.728478464649214 +Step 31090 | loss:0.2690409719944 lr:5.027084149901662e-06 tokens_per_second_per_gpu:4889.456172842695 mfu:23.06439899454487 +Step 31091 | loss:0.27606555819511414 lr:5.025189361667415e-06 tokens_per_second_per_gpu:3253.3403691605067 mfu:15.346561598434826 +Step 31092 | loss:0.292781800031662 lr:5.023294911694443e-06 tokens_per_second_per_gpu:3779.412042140207 mfu:17.82813143696294 +Step 31093 | loss:0.2520051598548889 lr:5.021400799997017e-06 tokens_per_second_per_gpu:3706.7179031800197 mfu:17.485220780588683 +Step 31094 | loss:0.3838043212890625 lr:5.019507026589376e-06 tokens_per_second_per_gpu:5181.626866065995 mfu:24.442618003939753 +Step 31095 | loss:0.33047640323638916 lr:5.01761359148577e-06 tokens_per_second_per_gpu:4822.759000237494 mfu:22.749777051655045 +Step 31096 | loss:0.2938989996910095 lr:5.015720494700421e-06 tokens_per_second_per_gpu:5303.888942976373 mfu:25.019349080787343 +Step 31097 | loss:0.2507714629173279 lr:5.0138277362475725e-06 tokens_per_second_per_gpu:5347.174253617064 mfu:25.22353327631502 +Step 31098 | loss:0.3134247362613678 lr:5.011935316141481e-06 tokens_per_second_per_gpu:5322.412262791671 mfu:25.10672674076036 +Step 31099 | loss:0.33146944642066956 lr:5.010043234396356e-06 tokens_per_second_per_gpu:4132.5457101756665 mfu:19.493923199902834 +Step 31100 | loss:0.3122318685054779 lr:5.00815149102643e-06 tokens_per_second_per_gpu:3906.799880934987 mfu:18.429041607164418 +Step 31101 | loss:0.35714611411094666 lr:5.006260086045955e-06 tokens_per_second_per_gpu:3975.6024700221333 mfu:18.753595158820662 +Step 31102 | loss:0.25976458191871643 lr:5.004369019469135e-06 tokens_per_second_per_gpu:3412.6452723652937 mfu:16.098030007071028 +Step 31103 | loss:0.32546377182006836 lr:5.002478291310186e-06 tokens_per_second_per_gpu:3702.198246071315 mfu:17.463900786873904 +Step 31104 | loss:0.3597102761268616 lr:5.000587901583342e-06 tokens_per_second_per_gpu:3769.4390327849123 mfu:17.78108705026316 +Step 31105 | loss:0.2948583662509918 lr:4.998697850302819e-06 tokens_per_second_per_gpu:5344.03881038587 mfu:25.208742855631705 +Step 31106 | loss:0.26372697949409485 lr:4.996808137482839e-06 tokens_per_second_per_gpu:4636.4609488101 mfu:21.870977357346813 +Step 31107 | loss:0.3223278820514679 lr:4.994918763137596e-06 tokens_per_second_per_gpu:4507.483407656494 mfu:21.262568289887568 +Step 31108 | loss:0.34174618124961853 lr:4.993029727281323e-06 tokens_per_second_per_gpu:3823.549752202034 mfu:18.036336545994452 +Step 31109 | loss:0.2546384930610657 lr:4.99114102992822e-06 tokens_per_second_per_gpu:4145.857553175934 mfu:19.55671743456972 +Step 31110 | loss:0.3564835488796234 lr:4.9892526710924754e-06 tokens_per_second_per_gpu:4560.250481460548 mfu:21.51147958001244 +Step 31111 | loss:0.3690784275531769 lr:4.987364650788307e-06 tokens_per_second_per_gpu:4782.803234400058 mfu:22.561298887043247 +Step 31112 | loss:0.23728246986865997 lr:4.985476969029912e-06 tokens_per_second_per_gpu:5327.045482488005 mfu:25.12858242857676 +Step 31113 | loss:0.2733539342880249 lr:4.9835896258314975e-06 tokens_per_second_per_gpu:5284.699310310049 mfu:24.928828309411355 +Step 31114 | loss:0.21513134241104126 lr:4.98170262120724e-06 tokens_per_second_per_gpu:4905.530618177665 mfu:23.140224895773102 +Step 31115 | loss:0.32222986221313477 lr:4.979815955171341e-06 tokens_per_second_per_gpu:5320.809366676738 mfu:25.099165606303945 +Step 31116 | loss:0.35399848222732544 lr:4.977929627737998e-06 tokens_per_second_per_gpu:5201.232155753326 mfu:24.535099500402236 +Step 31117 | loss:0.2786300480365753 lr:4.976043638921396e-06 tokens_per_second_per_gpu:4442.894607448302 mfu:20.957891899319865 +Step 31118 | loss:0.2602439820766449 lr:4.974157988735706e-06 tokens_per_second_per_gpu:3480.0987466312754 mfu:16.416219553933118 +Step 31119 | loss:0.39689990878105164 lr:4.972272677195116e-06 tokens_per_second_per_gpu:5018.454056447827 mfu:23.67290402081874 +Step 31120 | loss:0.2865148186683655 lr:4.9703877043138166e-06 tokens_per_second_per_gpu:3896.9836155885487 mfu:18.382736608697698 +Step 31121 | loss:0.35753875970840454 lr:4.9685030701059675e-06 tokens_per_second_per_gpu:4096.056312975629 mfu:19.32179648757712 +Step 31122 | loss:0.2957187592983246 lr:4.966618774585757e-06 tokens_per_second_per_gpu:5140.589928482242 mfu:24.24903976773365 +Step 31123 | loss:0.2591383457183838 lr:4.964734817767363e-06 tokens_per_second_per_gpu:4835.278432564011 mfu:22.80883334997465 +Step 31124 | loss:0.3053305447101593 lr:4.962851199664942e-06 tokens_per_second_per_gpu:3698.497071293932 mfu:17.446441713963456 +Step 31125 | loss:0.3577812612056732 lr:4.9609679202926525e-06 tokens_per_second_per_gpu:5366.480419109733 mfu:25.31460375291544 +Step 31126 | loss:0.22512409090995789 lr:4.959084979664675e-06 tokens_per_second_per_gpu:3768.616261692756 mfu:17.777205898642233 +Step 31127 | loss:0.3851190209388733 lr:4.957202377795173e-06 tokens_per_second_per_gpu:5355.7493082898545 mfu:25.26398327226177 +Step 31128 | loss:0.20464663207530975 lr:4.955320114698292e-06 tokens_per_second_per_gpu:4293.293923397847 mfu:20.252199948145165 +Step 31129 | loss:0.34436196088790894 lr:4.953438190388193e-06 tokens_per_second_per_gpu:5064.880157894878 mfu:23.891903862454193 +Step 31130 | loss:0.3285878598690033 lr:4.951556604879048e-06 tokens_per_second_per_gpu:5368.355424212158 mfu:25.323448471892437 +Step 31131 | loss:0.25593215227127075 lr:4.949675358184991e-06 tokens_per_second_per_gpu:5223.7080136281365 mfu:24.641121956774505 +Step 31132 | loss:0.2576028108596802 lr:4.947794450320164e-06 tokens_per_second_per_gpu:4411.640527279731 mfu:20.81046107066877 +Step 31133 | loss:0.37219247221946716 lr:4.945913881298725e-06 tokens_per_second_per_gpu:5331.273680245065 mfu:25.14852755880174 +Step 31134 | loss:0.2582645118236542 lr:4.944033651134822e-06 tokens_per_second_per_gpu:5307.657605360494 mfu:25.03712650425335 +Step 31135 | loss:0.3009606897830963 lr:4.942153759842588e-06 tokens_per_second_per_gpu:4977.750998465418 mfu:23.48090075165792 +Step 31136 | loss:0.23516523838043213 lr:4.940274207436158e-06 tokens_per_second_per_gpu:4105.587783731103 mfu:19.366758061367292 +Step 31137 | loss:0.3881896436214447 lr:4.938394993929691e-06 tokens_per_second_per_gpu:4607.123718378194 mfu:21.732588635951863 +Step 31138 | loss:0.317727267742157 lr:4.9365161193373036e-06 tokens_per_second_per_gpu:4349.9577364980905 mfu:20.519492822382446 +Step 31139 | loss:0.35959696769714355 lr:4.934637583673118e-06 tokens_per_second_per_gpu:4312.083579331005 mfu:20.34083396102775 +Step 31140 | loss:0.24774108827114105 lr:4.932759386951274e-06 tokens_per_second_per_gpu:3956.042097068339 mfu:18.661325542253856 +Step 31141 | loss:0.35098859667778015 lr:4.930881529185904e-06 tokens_per_second_per_gpu:4635.31180598753 mfu:21.865556654588772 +Step 31142 | loss:0.2916964292526245 lr:4.92900401039112e-06 tokens_per_second_per_gpu:4860.379431162924 mfu:22.927239043037694 +Step 31143 | loss:0.2767510712146759 lr:4.927126830581047e-06 tokens_per_second_per_gpu:5323.684230414924 mfu:25.112726829059298 +Step 31144 | loss:0.3269784152507782 lr:4.925249989769809e-06 tokens_per_second_per_gpu:5155.482791234372 mfu:24.31929194231985 +Step 31145 | loss:0.3064883351325989 lr:4.9233734879715225e-06 tokens_per_second_per_gpu:5255.796369191713 mfu:24.792488204805903 +Step 31146 | loss:0.3338957130908966 lr:4.921497325200286e-06 tokens_per_second_per_gpu:4715.6320030167935 mfu:22.244440728014535 +Step 31147 | loss:0.24508075416088104 lr:4.919621501470217e-06 tokens_per_second_per_gpu:4489.897926412905 mfu:21.179614574469067 +Step 31148 | loss:0.29503902792930603 lr:4.917746016795433e-06 tokens_per_second_per_gpu:5328.0117377740535 mfu:25.133140419621203 +Step 31149 | loss:0.3588934540748596 lr:4.915870871190037e-06 tokens_per_second_per_gpu:4670.620984081322 mfu:22.03211607202418 +Step 31150 | loss:0.3538433015346527 lr:4.913996064668119e-06 tokens_per_second_per_gpu:5198.156327596708 mfu:24.520590294197074 +Step 31151 | loss:0.20925429463386536 lr:4.912121597243796e-06 tokens_per_second_per_gpu:3242.4064302422553 mfu:15.294984343035088 +Step 31152 | loss:0.2704731523990631 lr:4.910247468931167e-06 tokens_per_second_per_gpu:4742.287772104819 mfu:22.37018053874593 +Step 31153 | loss:0.35172557830810547 lr:4.908373679744316e-06 tokens_per_second_per_gpu:3611.3037499886323 mfu:17.035135940651703 +Step 31154 | loss:0.31154167652130127 lr:4.906500229697336e-06 tokens_per_second_per_gpu:4329.745649167741 mfu:20.424148957907487 +Step 31155 | loss:0.23593878746032715 lr:4.904627118804317e-06 tokens_per_second_per_gpu:4852.564134304051 mfu:22.890372954327248 +Step 31156 | loss:0.17695270478725433 lr:4.902754347079363e-06 tokens_per_second_per_gpu:5336.848283954385 mfu:25.174823915624067 +Step 31157 | loss:0.2931292951107025 lr:4.900881914536542e-06 tokens_per_second_per_gpu:4827.639261713822 mfu:22.772798077697356 +Step 31158 | loss:0.4423677921295166 lr:4.899009821189937e-06 tokens_per_second_per_gpu:4139.039186638388 mfu:19.524554036278822 +Step 31159 | loss:0.27720409631729126 lr:4.897138067053647e-06 tokens_per_second_per_gpu:3234.2143261813258 mfu:15.256340790462602 +Step 31160 | loss:0.2978106141090393 lr:4.895266652141739e-06 tokens_per_second_per_gpu:4600.694682624313 mfu:21.702261777394195 +Step 31161 | loss:0.23294921219348907 lr:4.893395576468273e-06 tokens_per_second_per_gpu:4196.676792995536 mfu:19.796440459455184 +Step 31162 | loss:0.19364489614963531 lr:4.891524840047335e-06 tokens_per_second_per_gpu:3055.742332594954 mfu:14.41445794625363 +Step 31163 | loss:0.21988415718078613 lr:4.8896544428930045e-06 tokens_per_second_per_gpu:5397.533517112456 mfu:25.461086514399135 +Step 31164 | loss:0.3207792341709137 lr:4.887784385019328e-06 tokens_per_second_per_gpu:4376.886537436662 mfu:20.6465205709369 +Step 31165 | loss:0.21800965070724487 lr:4.885914666440383e-06 tokens_per_second_per_gpu:4018.503723983985 mfu:18.95596769346743 +Step 31166 | loss:0.18874959647655487 lr:4.8840452871702416e-06 tokens_per_second_per_gpu:4878.400925545609 mfu:23.012249506824826 +Step 31167 | loss:0.2684355676174164 lr:4.882176247222947e-06 tokens_per_second_per_gpu:4661.86259516386 mfu:21.990801257165273 +Step 31168 | loss:0.35358715057373047 lr:4.880307546612561e-06 tokens_per_second_per_gpu:5065.492526353066 mfu:23.894792508952296 +Step 31169 | loss:0.2910088896751404 lr:4.878439185353134e-06 tokens_per_second_per_gpu:5350.164401637909 mfu:25.237638314701204 +Step 31170 | loss:0.34713631868362427 lr:4.876571163458732e-06 tokens_per_second_per_gpu:4177.460007673538 mfu:19.70579160436965 +Step 31171 | loss:0.273805171251297 lr:4.874703480943388e-06 tokens_per_second_per_gpu:5382.4970939415525 mfu:25.390157140823646 +Step 31172 | loss:0.3348439633846283 lr:4.872836137821157e-06 tokens_per_second_per_gpu:5322.297657905692 mfu:25.10618612995959 +Step 31173 | loss:0.3230000138282776 lr:4.8709691341060905e-06 tokens_per_second_per_gpu:4201.8103176998 mfu:19.820656171354827 +Step 31174 | loss:0.3021899163722992 lr:4.869102469812226e-06 tokens_per_second_per_gpu:3871.5457703601805 mfu:18.262741952611545 +Step 31175 | loss:0.33855387568473816 lr:4.867236144953591e-06 tokens_per_second_per_gpu:3698.050982771318 mfu:17.444337438291736 +Step 31176 | loss:0.2802644670009613 lr:4.8653701595442355e-06 tokens_per_second_per_gpu:5353.890735658644 mfu:25.25521606805516 +Step 31177 | loss:0.23388388752937317 lr:4.863504513598194e-06 tokens_per_second_per_gpu:4891.93507767186 mfu:23.076092411568816 +Step 31178 | loss:0.3585272431373596 lr:4.8616392071294866e-06 tokens_per_second_per_gpu:4793.623455802717 mfu:22.612339717519728 +Step 31179 | loss:0.40220382809638977 lr:4.8597742401521536e-06 tokens_per_second_per_gpu:4871.254428389878 mfu:22.978538260423093 +Step 31180 | loss:0.33061277866363525 lr:4.857909612680217e-06 tokens_per_second_per_gpu:3945.3902329662874 mfu:18.61107888188906 +Step 31181 | loss:0.2928324043750763 lr:4.85604532472772e-06 tokens_per_second_per_gpu:4741.358463187344 mfu:22.365796830025747 +Step 31182 | loss:0.25973179936408997 lr:4.854181376308647e-06 tokens_per_second_per_gpu:5353.467042251661 mfu:25.25321743396487 +Step 31183 | loss:0.35364097356796265 lr:4.852317767437037e-06 tokens_per_second_per_gpu:5089.745559752517 mfu:24.0091982054917 +Step 31184 | loss:0.2765345275402069 lr:4.850454498126905e-06 tokens_per_second_per_gpu:4630.0330687145715 mfu:21.84065595022678 +Step 31185 | loss:0.27508533000946045 lr:4.84859156839228e-06 tokens_per_second_per_gpu:4691.739132339469 mfu:22.131733980485496 +Step 31186 | loss:0.32493335008621216 lr:4.846728978247145e-06 tokens_per_second_per_gpu:4461.97887110857 mfu:21.04791562711649 +Step 31187 | loss:0.3365030288696289 lr:4.844866727705522e-06 tokens_per_second_per_gpu:5308.027981589114 mfu:25.038873632116452 +Step 31188 | loss:0.31406259536743164 lr:4.8430048167814325e-06 tokens_per_second_per_gpu:4795.027396304515 mfu:22.618962344403656 +Step 31189 | loss:0.24854740500450134 lr:4.841143245488866e-06 tokens_per_second_per_gpu:5338.231409930102 mfu:25.18134835683715 +Step 31190 | loss:0.2585325539112091 lr:4.839282013841806e-06 tokens_per_second_per_gpu:3851.4263996204018 mfu:18.16783544811339 +Step 31191 | loss:0.4083079397678375 lr:4.8374211218542746e-06 tokens_per_second_per_gpu:4909.602707437707 mfu:23.15943363558279 +Step 31192 | loss:0.3404386341571808 lr:4.835560569540271e-06 tokens_per_second_per_gpu:4000.55298019975 mfu:18.871290972324484 +Step 31193 | loss:0.3806176781654358 lr:4.833700356913773e-06 tokens_per_second_per_gpu:3113.0323159767804 mfu:14.684704572544803 +Step 31194 | loss:0.26417291164398193 lr:4.83184048398877e-06 tokens_per_second_per_gpu:5315.981322726802 mfu:25.076390899242856 +Step 31195 | loss:0.4220547676086426 lr:4.829980950779273e-06 tokens_per_second_per_gpu:4936.297373415941 mfu:23.28535692958179 +Step 31196 | loss:0.2602509558200836 lr:4.828121757299253e-06 tokens_per_second_per_gpu:4713.892887187028 mfu:22.23623702192177 +Step 31197 | loss:0.23890209197998047 lr:4.826262903562684e-06 tokens_per_second_per_gpu:4477.834620645786 mfu:21.122709902953368 +Step 31198 | loss:0.38746339082717896 lr:4.824404389583554e-06 tokens_per_second_per_gpu:3694.383306835842 mfu:17.427036385134052 +Step 31199 | loss:0.34630751609802246 lr:4.82254621537585e-06 tokens_per_second_per_gpu:5336.592466513345 mfu:25.17361718110773 +Step 31200 | loss:0.312391459941864 lr:4.820688380953536e-06 tokens_per_second_per_gpu:4375.112886869543 mfu:20.63815395859581 +Step 31201 | loss:0.2519160509109497 lr:4.8188308863305884e-06 tokens_per_second_per_gpu:5103.318221019127 mfu:24.073222764421104 +Step 31202 | loss:0.2756986916065216 lr:4.816973731520991e-06 tokens_per_second_per_gpu:5357.813528421061 mfu:25.273720550812943 +Step 31203 | loss:0.31726986169815063 lr:4.8151169165386924e-06 tokens_per_second_per_gpu:4267.64209744481 mfu:20.131195908471877 +Step 31204 | loss:0.3255874812602997 lr:4.81326044139766e-06 tokens_per_second_per_gpu:4597.331982879899 mfu:21.686399348964056 +Step 31205 | loss:0.22732864320278168 lr:4.81140430611186e-06 tokens_per_second_per_gpu:3104.236147937226 mfu:14.6432115471212 +Step 31206 | loss:0.4026264548301697 lr:4.809548510695266e-06 tokens_per_second_per_gpu:5417.195049121618 mfu:25.553833315489626 +Step 31207 | loss:0.3494126796722412 lr:4.807693055161816e-06 tokens_per_second_per_gpu:5149.910902278129 mfu:24.293008391450968 +Step 31208 | loss:0.25192737579345703 lr:4.805837939525471e-06 tokens_per_second_per_gpu:4939.494747280379 mfu:23.300439487628573 +Step 31209 | loss:0.188583642244339 lr:4.803983163800196e-06 tokens_per_second_per_gpu:3299.9685923745415 mfu:15.566514882930663 +Step 31210 | loss:0.27692919969558716 lr:4.802128727999927e-06 tokens_per_second_per_gpu:5345.470286618986 mfu:25.215495373258342 +Step 31211 | loss:0.29473012685775757 lr:4.8002746321386075e-06 tokens_per_second_per_gpu:4068.3550261400974 mfu:19.19112479124702 +Step 31212 | loss:0.2832907438278198 lr:4.798420876230192e-06 tokens_per_second_per_gpu:4938.621525364654 mfu:23.29632035086508 +Step 31213 | loss:0.25062480568885803 lr:4.796567460288632e-06 tokens_per_second_per_gpu:5365.698465039325 mfu:25.310915142150204 +Step 31214 | loss:0.2681746482849121 lr:4.794714384327842e-06 tokens_per_second_per_gpu:4147.584652895634 mfu:19.564864458621138 +Step 31215 | loss:0.22923338413238525 lr:4.792861648361774e-06 tokens_per_second_per_gpu:5335.905454125101 mfu:25.17037642645598 +Step 31216 | loss:0.25259459018707275 lr:4.7910092524043605e-06 tokens_per_second_per_gpu:3525.8300850825913 mfu:16.63194207997889 +Step 31217 | loss:0.33068573474884033 lr:4.789157196469557e-06 tokens_per_second_per_gpu:4611.948780494822 mfu:21.75534927719648 +Step 31218 | loss:0.36853012442588806 lr:4.787305480571253e-06 tokens_per_second_per_gpu:4711.301739591593 mfu:22.22401413661874 +Step 31219 | loss:0.28017744421958923 lr:4.785454104723391e-06 tokens_per_second_per_gpu:3995.0140769259774 mfu:18.845163020547755 +Step 31220 | loss:0.30287331342697144 lr:4.783603068939902e-06 tokens_per_second_per_gpu:4354.277653771612 mfu:20.539870609215534 +Step 31221 | loss:0.345586359500885 lr:4.78175237323471e-06 tokens_per_second_per_gpu:4847.557901860059 mfu:22.866757701737605 +Step 31222 | loss:0.27276045083999634 lr:4.779902017621718e-06 tokens_per_second_per_gpu:4110.021074943195 mfu:19.38767065241221 +Step 31223 | loss:0.36538660526275635 lr:4.778052002114858e-06 tokens_per_second_per_gpu:4856.472838734276 mfu:22.908810980018362 +Step 31224 | loss:0.30676358938217163 lr:4.776202326728046e-06 tokens_per_second_per_gpu:4300.877319198958 mfu:20.28797211999975 +Step 31225 | loss:0.34250640869140625 lr:4.774352991475184e-06 tokens_per_second_per_gpu:4866.524824337361 mfu:22.956227911153707 +Step 31226 | loss:0.27637454867362976 lr:4.7725039963701756e-06 tokens_per_second_per_gpu:4061.7784887004027 mfu:19.16010214207099 +Step 31227 | loss:0.24473638832569122 lr:4.770655341426938e-06 tokens_per_second_per_gpu:3497.3802582223584 mfu:16.497739392638177 +Step 31228 | loss:0.3702259659767151 lr:4.768807026659383e-06 tokens_per_second_per_gpu:5362.11079622546 mfu:25.293991496235783 +Step 31229 | loss:0.27632710337638855 lr:4.7669590520813935e-06 tokens_per_second_per_gpu:5355.2306509753125 mfu:25.261536677217013 +Step 31230 | loss:0.36731380224227905 lr:4.765111417706869e-06 tokens_per_second_per_gpu:3254.334238004238 mfu:15.351249847342219 +Step 31231 | loss:0.31491684913635254 lr:4.763264123549732e-06 tokens_per_second_per_gpu:5192.207131350089 mfu:24.492526920464226 +Step 31232 | loss:0.33330631256103516 lr:4.761417169623855e-06 tokens_per_second_per_gpu:5338.618100414309 mfu:25.183172441827356 +Step 31233 | loss:0.2439546138048172 lr:4.759570555943121e-06 tokens_per_second_per_gpu:3758.1957416261516 mfu:17.728050527564005 +Step 31234 | loss:0.30577361583709717 lr:4.757724282521425e-06 tokens_per_second_per_gpu:4691.698432048741 mfu:22.131541990269948 +Step 31235 | loss:0.33223286271095276 lr:4.755878349372672e-06 tokens_per_second_per_gpu:5339.695815648386 mfu:25.18825620846368 +Step 31236 | loss:0.2902720868587494 lr:4.754032756510723e-06 tokens_per_second_per_gpu:3611.5923963880377 mfu:17.036497534965854 +Step 31237 | loss:0.25206881761550903 lr:4.752187503949468e-06 tokens_per_second_per_gpu:5346.023196821536 mfu:25.218103545019975 +Step 31238 | loss:0.34175485372543335 lr:4.750342591702789e-06 tokens_per_second_per_gpu:4726.63821620162 mfu:22.29635891346097 +Step 31239 | loss:0.43444523215293884 lr:4.748498019784564e-06 tokens_per_second_per_gpu:5305.77136959006 mfu:25.028228808299207 +Step 31240 | loss:0.32927843928337097 lr:4.746653788208644e-06 tokens_per_second_per_gpu:4012.000891214703 mfu:18.925292721797046 +Step 31241 | loss:0.31576210260391235 lr:4.744809896988916e-06 tokens_per_second_per_gpu:2812.5043722015453 mfu:13.267062986402463 +Step 31242 | loss:0.3532874286174774 lr:4.7429663461392605e-06 tokens_per_second_per_gpu:3930.3923412353515 mfu:18.54033126769004 +Step 31243 | loss:0.2618310749530792 lr:4.741123135673514e-06 tokens_per_second_per_gpu:5398.784832932626 mfu:25.466989184619372 +Step 31244 | loss:0.305721640586853 lr:4.739280265605561e-06 tokens_per_second_per_gpu:4920.707713176696 mfu:23.211817842362958 +Step 31245 | loss:0.3077884912490845 lr:4.737437735949263e-06 tokens_per_second_per_gpu:4479.138689198953 mfu:21.128861416815695 +Step 31246 | loss:0.2442638874053955 lr:4.735595546718474e-06 tokens_per_second_per_gpu:5087.7808893940355 mfu:23.999930520203417 +Step 31247 | loss:0.346312552690506 lr:4.733753697927035e-06 tokens_per_second_per_gpu:4391.128013684448 mfu:20.713700044244113 +Step 31248 | loss:0.20245252549648285 lr:4.731912189588811e-06 tokens_per_second_per_gpu:3845.50663556539 mfu:18.139910911050006 +Step 31249 | loss:0.3052631616592407 lr:4.7300710217176635e-06 tokens_per_second_per_gpu:4631.371546038409 mfu:21.846969776131097 +Step 31250 | loss:0.38627398014068604 lr:4.7282301943274154e-06 tokens_per_second_per_gpu:3678.213702757782 mfu:17.350761603878905 +Step 31251 | loss:0.2818496823310852 lr:4.726389707431928e-06 tokens_per_second_per_gpu:4234.715695928852 mfu:19.975876454697715 +Step 31252 | loss:0.3517821729183197 lr:4.724549561045038e-06 tokens_per_second_per_gpu:5346.026671806545 mfu:25.218119937117155 +Step 31253 | loss:0.3763292133808136 lr:4.72270975518061e-06 tokens_per_second_per_gpu:3753.4756024551543 mfu:17.705784825755625 +Step 31254 | loss:0.3558255136013031 lr:4.720870289852436e-06 tokens_per_second_per_gpu:4926.984717699502 mfu:23.241427543664436 +Step 31255 | loss:0.24445100128650665 lr:4.719031165074378e-06 tokens_per_second_per_gpu:3193.7089754629037 mfu:15.065270140198342 +Step 31256 | loss:0.2463199496269226 lr:4.717192380860264e-06 tokens_per_second_per_gpu:5365.937253474191 mfu:25.31204154793897 +Step 31257 | loss:0.26827895641326904 lr:4.715353937223932e-06 tokens_per_second_per_gpu:4883.214606612741 mfu:23.034956461717165 +Step 31258 | loss:0.3658962845802307 lr:4.7135158341791884e-06 tokens_per_second_per_gpu:3801.3750686511444 mfu:17.931734780293763 +Step 31259 | loss:0.36396360397338867 lr:4.711678071739878e-06 tokens_per_second_per_gpu:4972.086666432614 mfu:23.454181131024484 +Step 31260 | loss:0.3345378339290619 lr:4.7098406499198225e-06 tokens_per_second_per_gpu:5304.798851254755 mfu:25.023641273382836 +Step 31261 | loss:0.3290565609931946 lr:4.7080035687328325e-06 tokens_per_second_per_gpu:5331.51390953145 mfu:25.14966076133296 +Step 31262 | loss:0.2257862240076065 lr:4.706166828192721e-06 tokens_per_second_per_gpu:5107.260085162594 mfu:24.091817210137215 +Step 31263 | loss:0.3428517282009125 lr:4.704330428313303e-06 tokens_per_second_per_gpu:4554.94079195792 mfu:21.4864328687021 +Step 31264 | loss:0.2987610399723053 lr:4.702494369108406e-06 tokens_per_second_per_gpu:5321.702382214761 mfu:25.103378112960957 +Step 31265 | loss:0.25937265157699585 lr:4.700658650591827e-06 tokens_per_second_per_gpu:5050.3171266293475 mfu:23.823207559265335 +Step 31266 | loss:0.3785269558429718 lr:4.698823272777364e-06 tokens_per_second_per_gpu:3637.6713288899314 mfu:17.15951630910199 +Step 31267 | loss:0.3177073001861572 lr:4.696988235678851e-06 tokens_per_second_per_gpu:4335.1322993619715 mfu:20.449558705930965 +Step 31268 | loss:0.35364261269569397 lr:4.695153539310065e-06 tokens_per_second_per_gpu:5369.023398262023 mfu:25.326599419453775 +Step 31269 | loss:0.33498814702033997 lr:4.693319183684797e-06 tokens_per_second_per_gpu:4792.861464298457 mfu:22.608745273584976 +Step 31270 | loss:0.2532498836517334 lr:4.691485168816862e-06 tokens_per_second_per_gpu:5076.3150640890835 mfu:23.945844265968436 +Step 31271 | loss:0.3704582452774048 lr:4.68965149472006e-06 tokens_per_second_per_gpu:5296.488081557717 mfu:24.98443795475838 +Step 31272 | loss:0.28670281171798706 lr:4.6878181614081584e-06 tokens_per_second_per_gpu:5379.19029368131 mfu:25.374558399797785 +Step 31273 | loss:0.2861866056919098 lr:4.685985168894957e-06 tokens_per_second_per_gpu:5359.130198123413 mfu:25.279931506445948 +Step 31274 | loss:0.295244425535202 lr:4.684152517194251e-06 tokens_per_second_per_gpu:4279.1012917759235 mfu:20.185250883271898 +Step 31275 | loss:0.4008632302284241 lr:4.682320206319818e-06 tokens_per_second_per_gpu:3454.3014974921166 mfu:16.29452952827914 +Step 31276 | loss:0.28167200088500977 lr:4.680488236285429e-06 tokens_per_second_per_gpu:5245.664206326115 mfu:24.744693063843627 +Step 31277 | loss:0.3798139989376068 lr:4.67865660710487e-06 tokens_per_second_per_gpu:4506.599870527225 mfu:21.258400494501828 +Step 31278 | loss:0.3401118218898773 lr:4.676825318791928e-06 tokens_per_second_per_gpu:4679.268099214615 mfu:22.072905989457904 +Step 31279 | loss:0.23544958233833313 lr:4.6749943713603585e-06 tokens_per_second_per_gpu:5355.250676227037 mfu:25.261631139746054 +Step 31280 | loss:0.4310157001018524 lr:4.673163764823934e-06 tokens_per_second_per_gpu:3855.340757742073 mfu:18.18630014323168 +Step 31281 | loss:0.32435598969459534 lr:4.671333499196445e-06 tokens_per_second_per_gpu:4110.543046227317 mfu:19.390132880016896 +Step 31282 | loss:0.34080275893211365 lr:4.669503574491635e-06 tokens_per_second_per_gpu:5361.345187709844 mfu:25.290379990241338 +Step 31283 | loss:0.36358094215393066 lr:4.667673990723265e-06 tokens_per_second_per_gpu:3967.510876575652 mfu:18.71542573196541 +Step 31284 | loss:0.306625097990036 lr:4.665844747905107e-06 tokens_per_second_per_gpu:4997.381248525262 mfu:23.57349999045623 +Step 31285 | loss:0.2648612856864929 lr:4.6640158460509234e-06 tokens_per_second_per_gpu:4401.627036050027 mfu:20.763225724060476 +Step 31286 | loss:0.24937160313129425 lr:4.662187285174446e-06 tokens_per_second_per_gpu:4402.975547993782 mfu:20.769586884979034 +Step 31287 | loss:0.3637349605560303 lr:4.660359065289449e-06 tokens_per_second_per_gpu:5382.605165794873 mfu:25.39066693419472 +Step 31288 | loss:0.27665969729423523 lr:4.658531186409676e-06 tokens_per_second_per_gpu:5329.260311415898 mfu:25.139030154518416 +Step 31289 | loss:0.3188404142856598 lr:4.656703648548894e-06 tokens_per_second_per_gpu:5364.75221412993 mfu:25.30645151516353 +Step 31290 | loss:0.25925901532173157 lr:4.6548764517208135e-06 tokens_per_second_per_gpu:5351.678552780465 mfu:25.24478082398147 +Step 31291 | loss:0.33613982796669006 lr:4.653049595939191e-06 tokens_per_second_per_gpu:4576.776167047848 mfu:21.58943405850044 +Step 31292 | loss:0.2759058177471161 lr:4.651223081217765e-06 tokens_per_second_per_gpu:4509.722216143175 mfu:21.273129131499356 +Step 31293 | loss:0.2803754210472107 lr:4.649396907570292e-06 tokens_per_second_per_gpu:4752.564982332958 mfu:22.418659892864937 +Step 31294 | loss:0.35470008850097656 lr:4.6475710750104765e-06 tokens_per_second_per_gpu:4350.0990698772575 mfu:20.520159516045936 +Step 31295 | loss:0.28185930848121643 lr:4.645745583552069e-06 tokens_per_second_per_gpu:5380.58093801726 mfu:25.381118306398847 +Step 31296 | loss:0.3539530038833618 lr:4.643920433208815e-06 tokens_per_second_per_gpu:5349.535487238478 mfu:25.234671618175227 +Step 31297 | loss:0.3184920847415924 lr:4.642095623994408e-06 tokens_per_second_per_gpu:4783.043764873877 mfu:22.562433510326986 +Step 31298 | loss:0.3503924012184143 lr:4.640271155922582e-06 tokens_per_second_per_gpu:4548.669573847756 mfu:21.456850462896313 +Step 31299 | loss:0.30741336941719055 lr:4.638447029007065e-06 tokens_per_second_per_gpu:3751.1323304287484 mfu:17.69473121180363 +Step 31300 | loss:0.3594811260700226 lr:4.6366232432615895e-06 tokens_per_second_per_gpu:4752.410701264852 mfu:22.417932122743615 +Step 31301 | loss:0.3968488276004791 lr:4.634799798699846e-06 tokens_per_second_per_gpu:4887.880654555707 mfu:23.056967005974006 +Step 31302 | loss:0.2480924129486084 lr:4.632976695335562e-06 tokens_per_second_per_gpu:4281.865997160422 mfu:20.19829247027796 +Step 31303 | loss:0.2866925597190857 lr:4.631153933182458e-06 tokens_per_second_per_gpu:5253.444172609311 mfu:24.781392492200588 +Step 31304 | loss:0.3522179126739502 lr:4.629331512254237e-06 tokens_per_second_per_gpu:3681.3555375391807 mfu:17.365582174595055 +Step 31305 | loss:0.3049269914627075 lr:4.627509432564592e-06 tokens_per_second_per_gpu:3106.765998105434 mfu:14.655145282001586 +Step 31306 | loss:0.28427350521087646 lr:4.62568769412724e-06 tokens_per_second_per_gpu:4549.637995085912 mfu:21.461418671106394 +Step 31307 | loss:0.32751229405403137 lr:4.6238662969558865e-06 tokens_per_second_per_gpu:4584.448335848884 mfu:21.625624987741904 +Step 31308 | loss:0.35144609212875366 lr:4.622045241064215e-06 tokens_per_second_per_gpu:4994.079928882638 mfu:23.55792710243876 +Step 31309 | loss:0.32062968611717224 lr:4.620224526465938e-06 tokens_per_second_per_gpu:4059.5666801905386 mfu:19.149668663956515 +Step 31310 | loss:0.36413076519966125 lr:4.618404153174749e-06 tokens_per_second_per_gpu:4571.181294851133 mfu:21.563042091764906 +Step 31311 | loss:0.2720927298069 lr:4.616584121204332e-06 tokens_per_second_per_gpu:4992.674973065318 mfu:23.551299686138282 +Step 31312 | loss:0.2753652036190033 lr:4.6147644305683635e-06 tokens_per_second_per_gpu:4622.3142226639 mfu:21.804244836434822 +Step 31313 | loss:0.325512170791626 lr:4.612945081280551e-06 tokens_per_second_per_gpu:5326.202796831932 mfu:25.124607336560064 +Step 31314 | loss:0.39384207129478455 lr:4.611126073354571e-06 tokens_per_second_per_gpu:4809.995395947634 mfu:22.689568952524333 +Step 31315 | loss:0.3479066789150238 lr:4.609307406804098e-06 tokens_per_second_per_gpu:3463.624830772929 mfu:16.33850928208942 +Step 31316 | loss:0.3167014420032501 lr:4.607489081642818e-06 tokens_per_second_per_gpu:4662.161449863729 mfu:21.99221100577443 +Step 31317 | loss:0.28938519954681396 lr:4.605671097884412e-06 tokens_per_second_per_gpu:4143.73566724187 mfu:19.546708136587664 +Step 31318 | loss:0.23382355272769928 lr:4.603853455542545e-06 tokens_per_second_per_gpu:5151.099388171196 mfu:24.298614682185168 +Step 31319 | loss:0.32148343324661255 lr:4.602036154630884e-06 tokens_per_second_per_gpu:5212.074299401405 mfu:24.586243741850687 +Step 31320 | loss:0.2937110960483551 lr:4.600219195163097e-06 tokens_per_second_per_gpu:5285.482364764691 mfu:24.93252211087676 +Step 31321 | loss:0.2849627137184143 lr:4.598402577152866e-06 tokens_per_second_per_gpu:3462.135446464073 mfu:16.33148360218938 +Step 31322 | loss:0.33361533284187317 lr:4.596586300613831e-06 tokens_per_second_per_gpu:4830.258411264134 mfu:22.785153052999174 +Step 31323 | loss:0.4852455258369446 lr:4.59477036555967e-06 tokens_per_second_per_gpu:4023.6555859787754 mfu:18.98026990549515 +Step 31324 | loss:0.37291309237480164 lr:4.5929547720040265e-06 tokens_per_second_per_gpu:3679.0498141435237 mfu:17.35470568394076 +Step 31325 | loss:0.3522704541683197 lr:4.591139519960585e-06 tokens_per_second_per_gpu:5245.295918375904 mfu:24.74295578674608 +Step 31326 | loss:0.34122931957244873 lr:4.5893246094429636e-06 tokens_per_second_per_gpu:4389.180618270259 mfu:20.704513847815466 +Step 31327 | loss:0.2755017578601837 lr:4.587510040464821e-06 tokens_per_second_per_gpu:4645.848244968778 mfu:21.91525883496518 +Step 31328 | loss:0.3838708698749542 lr:4.585695813039814e-06 tokens_per_second_per_gpu:4337.964184020156 mfu:20.462917189032968 +Step 31329 | loss:0.3037290573120117 lr:4.583881927181588e-06 tokens_per_second_per_gpu:3043.0289814238695 mfu:14.35448689965824 +Step 31330 | loss:0.2740708887577057 lr:4.582068382903776e-06 tokens_per_second_per_gpu:4860.69561291823 mfu:22.928730526323644 +Step 31331 | loss:0.34919556975364685 lr:4.580255180220022e-06 tokens_per_second_per_gpu:5024.862659456817 mfu:23.703134494632575 +Step 31332 | loss:0.3018409311771393 lr:4.578442319143978e-06 tokens_per_second_per_gpu:5169.188858472453 mfu:24.383945799978086 +Step 31333 | loss:0.303024023771286 lr:4.576629799689258e-06 tokens_per_second_per_gpu:5062.168252672381 mfu:23.87911134282093 +Step 31334 | loss:0.30370596051216125 lr:4.574817621869493e-06 tokens_per_second_per_gpu:3943.7018992075964 mfu:18.603114723489867 +Step 31335 | loss:0.37513267993927 lr:4.5730057856983255e-06 tokens_per_second_per_gpu:5311.846805001597 mfu:25.056887673711596 +Step 31336 | loss:0.34337297081947327 lr:4.571194291189385e-06 tokens_per_second_per_gpu:3836.575885840614 mfu:18.09778304085521 +Step 31337 | loss:0.35252511501312256 lr:4.569383138356276e-06 tokens_per_second_per_gpu:4478.507757161554 mfu:21.125885202747174 +Step 31338 | loss:0.39901864528656006 lr:4.567572327212638e-06 tokens_per_second_per_gpu:5319.214753619313 mfu:25.091643544443027 +Step 31339 | loss:0.3157525062561035 lr:4.565761857772094e-06 tokens_per_second_per_gpu:5174.625616092622 mfu:24.409591913278927 +Step 31340 | loss:0.31351420283317566 lr:4.5639517300482545e-06 tokens_per_second_per_gpu:4362.906566852504 mfu:20.58057466446249 +Step 31341 | loss:0.2674020230770111 lr:4.5621419440547205e-06 tokens_per_second_per_gpu:5286.97482609936 mfu:24.93956230563256 +Step 31342 | loss:0.2651953101158142 lr:4.560332499805115e-06 tokens_per_second_per_gpu:4376.870766590279 mfu:20.646446177163728 +Step 31343 | loss:0.3293394446372986 lr:4.558523397313053e-06 tokens_per_second_per_gpu:4632.976987543668 mfu:21.85454291762797 +Step 31344 | loss:0.30471372604370117 lr:4.5567146365921295e-06 tokens_per_second_per_gpu:3852.6489068788683 mfu:18.173602223432827 +Step 31345 | loss:0.3643015921115875 lr:4.554906217655952e-06 tokens_per_second_per_gpu:3574.011694580739 mfu:16.85922295261192 +Step 31346 | loss:0.2855398654937744 lr:4.553098140518136e-06 tokens_per_second_per_gpu:5360.236141991702 mfu:25.285148432366153 +Step 31347 | loss:0.2827737331390381 lr:4.551290405192265e-06 tokens_per_second_per_gpu:4910.236415957072 mfu:23.162422946790453 +Step 31348 | loss:0.36349838972091675 lr:4.54948301169193e-06 tokens_per_second_per_gpu:4683.983675406579 mfu:22.095150166915662 +Step 31349 | loss:0.2950623631477356 lr:4.547675960030728e-06 tokens_per_second_per_gpu:5378.9585475707745 mfu:25.373465213854423 +Step 31350 | loss:0.27906790375709534 lr:4.545869250222273e-06 tokens_per_second_per_gpu:5044.326089436579 mfu:23.794946814650697 +Step 31351 | loss:0.30676713585853577 lr:4.54406288228012e-06 tokens_per_second_per_gpu:5351.138219674963 mfu:25.24223198053248 +Step 31352 | loss:0.2897285521030426 lr:4.5422568562178695e-06 tokens_per_second_per_gpu:5349.384901289814 mfu:25.233961278562766 +Step 31353 | loss:0.40127578377723694 lr:4.540451172049121e-06 tokens_per_second_per_gpu:4973.997464740762 mfu:23.463194692659826 +Step 31354 | loss:0.31350305676460266 lr:4.538645829787436e-06 tokens_per_second_per_gpu:3434.4677303535245 mfu:16.20097026469678 +Step 31355 | loss:0.2791903018951416 lr:4.536840829446387e-06 tokens_per_second_per_gpu:5364.250060944213 mfu:25.304082772910338 +Step 31356 | loss:0.2418794482946396 lr:4.535036171039558e-06 tokens_per_second_per_gpu:5089.07730701326 mfu:24.00604594330495 +Step 31357 | loss:0.33356326818466187 lr:4.533231854580533e-06 tokens_per_second_per_gpu:4001.3908504650108 mfu:18.875243349321128 +Step 31358 | loss:0.1950029581785202 lr:4.531427880082867e-06 tokens_per_second_per_gpu:5370.7642683194135 mfu:25.334811400537546 +Step 31359 | loss:0.32822075486183167 lr:4.529624247560132e-06 tokens_per_second_per_gpu:5115.238293957886 mfu:24.12945178224739 +Step 31360 | loss:0.31901541352272034 lr:4.527820957025891e-06 tokens_per_second_per_gpu:5330.175647423028 mfu:25.143347950636475 +Step 31361 | loss:0.3304039239883423 lr:4.526018008493738e-06 tokens_per_second_per_gpu:5187.93204440404 mfu:24.472360605934664 +Step 31362 | loss:0.2912508249282837 lr:4.524215401977178e-06 tokens_per_second_per_gpu:3879.996881869472 mfu:18.302607287509186 +Step 31363 | loss:0.3366973102092743 lr:4.522413137489806e-06 tokens_per_second_per_gpu:5079.103607767118 mfu:23.958998302272335 +Step 31364 | loss:0.2899974286556244 lr:4.520611215045167e-06 tokens_per_second_per_gpu:5339.295903810681 mfu:25.186369756093228 +Step 31365 | loss:0.2701669931411743 lr:4.518809634656812e-06 tokens_per_second_per_gpu:5350.783300315986 mfu:25.2405577653607 +Step 31366 | loss:0.25637590885162354 lr:4.51700839633829e-06 tokens_per_second_per_gpu:4708.640896450342 mfu:22.211462485534806 +Step 31367 | loss:0.3175084590911865 lr:4.515207500103158e-06 tokens_per_second_per_gpu:4229.442151091124 mfu:19.95100024393862 +Step 31368 | loss:0.2828102111816406 lr:4.513406945964954e-06 tokens_per_second_per_gpu:3322.268869614288 mfu:15.671709095490186 +Step 31369 | loss:0.33515068888664246 lr:4.511606733937229e-06 tokens_per_second_per_gpu:4915.868977341152 mfu:23.18899269985323 +Step 31370 | loss:0.42879828810691833 lr:4.5098068640335e-06 tokens_per_second_per_gpu:4682.58762127273 mfu:22.088564741374896 +Step 31371 | loss:0.2710907459259033 lr:4.508007336267317e-06 tokens_per_second_per_gpu:3579.3823057271757 mfu:16.884557041710412 +Step 31372 | loss:0.32425984740257263 lr:4.50620815065223e-06 tokens_per_second_per_gpu:4807.360338728914 mfu:22.677138938036506 +Step 31373 | loss:0.37872204184532166 lr:4.504409307201746e-06 tokens_per_second_per_gpu:5290.60463669649 mfu:24.95668474152877 +Step 31374 | loss:0.4648534953594208 lr:4.502610805929403e-06 tokens_per_second_per_gpu:4082.2167921963483 mfu:19.256513106795524 +Step 31375 | loss:0.3185850977897644 lr:4.500812646848734e-06 tokens_per_second_per_gpu:3838.9504197593187 mfu:18.108984122487 +Step 31376 | loss:0.35785040259361267 lr:4.499014829973264e-06 tokens_per_second_per_gpu:4473.149570508039 mfu:21.100609722101634 +Step 31377 | loss:0.3645758032798767 lr:4.497217355316502e-06 tokens_per_second_per_gpu:5392.557780841282 mfu:25.43761511745986 +Step 31378 | loss:0.3386644423007965 lr:4.4954202228919715e-06 tokens_per_second_per_gpu:4613.439337223287 mfu:21.762380487599856 +Step 31379 | loss:0.29700568318367004 lr:4.493623432713201e-06 tokens_per_second_per_gpu:3260.7406842333035 mfu:15.38147014725716 +Step 31380 | loss:0.36508381366729736 lr:4.49182698479369e-06 tokens_per_second_per_gpu:5351.925549747017 mfu:25.245945950814722 +Step 31381 | loss:0.37965431809425354 lr:4.49003087914695e-06 tokens_per_second_per_gpu:5320.562497835746 mfu:25.09800108386082 +Step 31382 | loss:0.2390187680721283 lr:4.48823511578651e-06 tokens_per_second_per_gpu:5320.4917382168815 mfu:25.097667298665797 +Step 31383 | loss:0.22126071155071259 lr:4.486439694725858e-06 tokens_per_second_per_gpu:3861.1673172218684 mfu:18.21378501841143 +Step 31384 | loss:0.27171453833580017 lr:4.484644615978489e-06 tokens_per_second_per_gpu:5349.097752034003 mfu:25.232606746530394 +Step 31385 | loss:0.2733496129512787 lr:4.482849879557915e-06 tokens_per_second_per_gpu:4982.838529955914 mfu:23.50489950572097 +Step 31386 | loss:0.2484322190284729 lr:4.481055485477647e-06 tokens_per_second_per_gpu:4693.62242263458 mfu:22.140617782129997 +Step 31387 | loss:0.3152301013469696 lr:4.479261433751164e-06 tokens_per_second_per_gpu:4258.043822753685 mfu:20.08591920911931 +Step 31388 | loss:0.29883772134780884 lr:4.4774677243919596e-06 tokens_per_second_per_gpu:5364.412644239284 mfu:25.30484970605276 +Step 31389 | loss:0.315180242061615 lr:4.4756743574135344e-06 tokens_per_second_per_gpu:4117.123502166847 mfu:19.42117401342522 +Step 31390 | loss:0.42553403973579407 lr:4.4738813328293775e-06 tokens_per_second_per_gpu:3837.3559689043063 mfu:18.10146282576254 +Step 31391 | loss:0.264809787273407 lr:4.472088650652956e-06 tokens_per_second_per_gpu:4448.65765563196 mfu:20.985077180879593 +Step 31392 | loss:0.3809802532196045 lr:4.47029631089777e-06 tokens_per_second_per_gpu:3828.2296309201042 mfu:18.058412332377234 +Step 31393 | loss:0.29761114716529846 lr:4.468504313577298e-06 tokens_per_second_per_gpu:4099.469932250715 mfu:19.337899112121207 +Step 31394 | loss:0.37829408049583435 lr:4.466712658705008e-06 tokens_per_second_per_gpu:5331.480479325551 mfu:25.14950306534795 +Step 31395 | loss:0.2494858205318451 lr:4.464921346294382e-06 tokens_per_second_per_gpu:5080.559435445704 mfu:23.965865689822042 +Step 31396 | loss:0.41231900453567505 lr:4.463130376358893e-06 tokens_per_second_per_gpu:5339.296187397631 mfu:25.186371093821258 +Step 31397 | loss:0.3007175326347351 lr:4.461339748912019e-06 tokens_per_second_per_gpu:3284.1199642539236 mfu:15.491754200030918 +Step 31398 | loss:0.36883389949798584 lr:4.459549463967222e-06 tokens_per_second_per_gpu:5331.528112244607 mfu:25.149727757954153 +Step 31399 | loss:0.3179593086242676 lr:4.457759521537957e-06 tokens_per_second_per_gpu:3622.909795772626 mfu:17.089883638810402 +Step 31400 | loss:0.3697047233581543 lr:4.455969921637698e-06 tokens_per_second_per_gpu:4555.599721817008 mfu:21.48954115327244 +Step 31401 | loss:0.2535701394081116 lr:4.454180664279894e-06 tokens_per_second_per_gpu:5011.398854089335 mfu:23.639623427552092 +Step 31402 | loss:0.3238736093044281 lr:4.452391749478007e-06 tokens_per_second_per_gpu:5348.966121630834 mfu:25.231985823460185 +Step 31403 | loss:0.3109802007675171 lr:4.450603177245499e-06 tokens_per_second_per_gpu:5333.287055941038 mfu:25.158024995476957 +Step 31404 | loss:0.25486987829208374 lr:4.44881494759582e-06 tokens_per_second_per_gpu:4337.91178005604 mfu:20.46266999059319 +Step 31405 | loss:0.2423425316810608 lr:4.447027060542419e-06 tokens_per_second_per_gpu:5286.98761555476 mfu:24.939622635675786 +Step 31406 | loss:0.36168375611305237 lr:4.445239516098732e-06 tokens_per_second_per_gpu:5311.585964922887 mfu:25.05565724655641 +Step 31407 | loss:0.3569268584251404 lr:4.443452314278207e-06 tokens_per_second_per_gpu:4207.392464142848 mfu:19.84698810853908 +Step 31408 | loss:0.3104437291622162 lr:4.441665455094302e-06 tokens_per_second_per_gpu:5384.4396450703925 mfu:25.399320485931785 +Step 31409 | loss:0.2942686378955841 lr:4.439878938560438e-06 tokens_per_second_per_gpu:4597.567221847486 mfu:21.687509011310066 +Step 31410 | loss:0.34800249338150024 lr:4.438092764690055e-06 tokens_per_second_per_gpu:4148.434018610279 mfu:19.568871061613855 +Step 31411 | loss:0.32670533657073975 lr:4.4363069334965975e-06 tokens_per_second_per_gpu:4609.820399769344 mfu:21.745309342175204 +Step 31412 | loss:0.2676960825920105 lr:4.434521444993489e-06 tokens_per_second_per_gpu:5359.239887090211 mfu:25.2804489280172 +Step 31413 | loss:0.29047060012817383 lr:4.4327362991941565e-06 tokens_per_second_per_gpu:4399.5438932293855 mfu:20.7533991839547 +Step 31414 | loss:0.24384982883930206 lr:4.430951496112023e-06 tokens_per_second_per_gpu:4931.761102312273 mfu:23.26395856481813 +Step 31415 | loss:0.31534311175346375 lr:4.429167035760529e-06 tokens_per_second_per_gpu:4636.095729555864 mfu:21.869254555811754 +Step 31416 | loss:0.3956008553504944 lr:4.427382918153073e-06 tokens_per_second_per_gpu:5303.881727956448 mfu:25.01931504630735 +Step 31417 | loss:0.24329937994480133 lr:4.425599143303088e-06 tokens_per_second_per_gpu:4410.808153342155 mfu:20.806534620787584 +Step 31418 | loss:0.29025304317474365 lr:4.423815711223994e-06 tokens_per_second_per_gpu:4676.270703032277 mfu:22.058766760257296 +Step 31419 | loss:0.389560729265213 lr:4.422032621929201e-06 tokens_per_second_per_gpu:5341.082698993213 mfu:25.194798373827926 +Step 31420 | loss:0.25840285420417786 lr:4.420249875432103e-06 tokens_per_second_per_gpu:3698.4021645005114 mfu:17.44599402242612 +Step 31421 | loss:0.27105993032455444 lr:4.418467471746124e-06 tokens_per_second_per_gpu:3322.459665513493 mfu:15.672609112299872 +Step 31422 | loss:0.30055221915245056 lr:4.41668541088468e-06 tokens_per_second_per_gpu:4510.651785266288 mfu:21.27751406765393 +Step 31423 | loss:0.30644491314888 lr:4.41490369286115e-06 tokens_per_second_per_gpu:4807.782781803102 mfu:22.67913167409367 +Step 31424 | loss:0.24354593455791473 lr:4.413122317688945e-06 tokens_per_second_per_gpu:4104.35639020158 mfu:19.360949367991136 +Step 31425 | loss:0.3378479778766632 lr:4.411341285381471e-06 tokens_per_second_per_gpu:4155.621812608092 mfu:19.602777112265652 +Step 31426 | loss:0.2546943724155426 lr:4.409560595952106e-06 tokens_per_second_per_gpu:5165.461624134746 mfu:24.366363799675373 +Step 31427 | loss:0.32259246706962585 lr:4.407780249414267e-06 tokens_per_second_per_gpu:4142.449036114562 mfu:19.540638878038244 +Step 31428 | loss:0.4076095223426819 lr:4.4060002457813224e-06 tokens_per_second_per_gpu:5110.813322009782 mfu:24.108578434590115 +Step 31429 | loss:0.3133927285671234 lr:4.4042205850666705e-06 tokens_per_second_per_gpu:5283.060954927162 mfu:24.921099907538018 +Step 31430 | loss:0.29117220640182495 lr:4.402441267283691e-06 tokens_per_second_per_gpu:5367.087164900486 mfu:25.317465876331926 +Step 31431 | loss:0.3500327467918396 lr:4.400662292445767e-06 tokens_per_second_per_gpu:3525.3809151691116 mfu:16.62982326886072 +Step 31432 | loss:0.30153951048851013 lr:4.398883660566289e-06 tokens_per_second_per_gpu:5352.645630848391 mfu:25.249342696228407 +Step 31433 | loss:0.3946189880371094 lr:4.397105371658616e-06 tokens_per_second_per_gpu:4601.509229426607 mfu:21.706104133636632 +Step 31434 | loss:0.35573840141296387 lr:4.39532742573614e-06 tokens_per_second_per_gpu:4796.24308834674 mfu:22.624696971185283 +Step 31435 | loss:0.29041942954063416 lr:4.393549822812221e-06 tokens_per_second_per_gpu:4655.449341423032 mfu:21.96054884505587 +Step 31436 | loss:0.2529667615890503 lr:4.391772562900237e-06 tokens_per_second_per_gpu:3772.9582920854655 mfu:17.79768799683142 +Step 31437 | loss:0.3263430595397949 lr:4.389995646013545e-06 tokens_per_second_per_gpu:4496.050943409404 mfu:21.2086394054545 +Step 31438 | loss:0.3481996953487396 lr:4.388219072165517e-06 tokens_per_second_per_gpu:5368.431573005132 mfu:25.32380767874122 +Step 31439 | loss:0.3189548850059509 lr:4.386442841369509e-06 tokens_per_second_per_gpu:4640.9827950913805 mfu:21.892307677762147 +Step 31440 | loss:0.31894397735595703 lr:4.3846669536389e-06 tokens_per_second_per_gpu:5377.007252736454 mfu:25.3642606231957 +Step 31441 | loss:0.28589022159576416 lr:4.382891408987027e-06 tokens_per_second_per_gpu:5068.92258374452 mfu:23.910972674895923 +Step 31442 | loss:0.3984958231449127 lr:4.381116207427239e-06 tokens_per_second_per_gpu:5360.512858540801 mfu:25.28645375154096 +Step 31443 | loss:0.2535669207572937 lr:4.379341348972899e-06 tokens_per_second_per_gpu:5325.664309820042 mfu:25.12206720144962 +Step 31444 | loss:0.35435718297958374 lr:4.377566833637364e-06 tokens_per_second_per_gpu:4672.1060849157675 mfu:22.039121546044594 +Step 31445 | loss:0.3033973276615143 lr:4.375792661433959e-06 tokens_per_second_per_gpu:5327.451500367757 mfu:25.130497684189343 +Step 31446 | loss:0.29022783041000366 lr:4.3740188323760425e-06 tokens_per_second_per_gpu:5309.739230932151 mfu:25.046945887236806 +Step 31447 | loss:0.40817755460739136 lr:4.372245346476955e-06 tokens_per_second_per_gpu:3327.3091629193555 mfu:15.695485018972905 +Step 31448 | loss:0.32571929693222046 lr:4.370472203750037e-06 tokens_per_second_per_gpu:3928.512153193346 mfu:18.53146209990274 +Step 31449 | loss:0.2179616093635559 lr:4.368699404208609e-06 tokens_per_second_per_gpu:5173.370902773971 mfu:24.40367321647868 +Step 31450 | loss:0.3958342969417572 lr:4.366926947866018e-06 tokens_per_second_per_gpu:5367.784834891327 mfu:25.320756904713498 +Step 31451 | loss:0.3106914162635803 lr:4.365154834735602e-06 tokens_per_second_per_gpu:4724.830882542045 mfu:22.287833412225837 +Step 31452 | loss:0.2936966121196747 lr:4.363383064830667e-06 tokens_per_second_per_gpu:5351.734977356763 mfu:25.24504698833489 +Step 31453 | loss:0.33031001687049866 lr:4.361611638164553e-06 tokens_per_second_per_gpu:3678.1220010843554 mfu:17.350329031439433 +Step 31454 | loss:0.20398081839084625 lr:4.359840554750589e-06 tokens_per_second_per_gpu:5331.524596298557 mfu:25.149711172636756 +Step 31455 | loss:0.34907832741737366 lr:4.3580698146020895e-06 tokens_per_second_per_gpu:4502.794574667652 mfu:21.2404502646818 +Step 31456 | loss:0.2987317740917206 lr:4.3562994177323625e-06 tokens_per_second_per_gpu:4579.055653730851 mfu:21.6001867861047 +Step 31457 | loss:0.2315782606601715 lr:4.35452936415473e-06 tokens_per_second_per_gpu:5198.046503694909 mfu:24.520072236114398 +Step 31458 | loss:0.2557530701160431 lr:4.352759653882515e-06 tokens_per_second_per_gpu:4151.376649947535 mfu:19.582751955696367 +Step 31459 | loss:0.21786390244960785 lr:4.350990286929013e-06 tokens_per_second_per_gpu:4308.621455121641 mfu:20.324502530432877 +Step 31460 | loss:0.3433413803577423 lr:4.349221263307534e-06 tokens_per_second_per_gpu:4470.575515548764 mfu:21.088467465679674 +Step 31461 | loss:0.2576009929180145 lr:4.347452583031397e-06 tokens_per_second_per_gpu:5124.249926333403 mfu:24.171961189705733 +Step 31462 | loss:0.3609667718410492 lr:4.345684246113885e-06 tokens_per_second_per_gpu:4314.555978367542 mfu:20.352496689117867 +Step 31463 | loss:0.27989596128463745 lr:4.343916252568319e-06 tokens_per_second_per_gpu:5354.830244216373 mfu:25.25964789021866 +Step 31464 | loss:0.2921014428138733 lr:4.342148602407975e-06 tokens_per_second_per_gpu:5332.903800899391 mfu:25.15621711605546 +Step 31465 | loss:0.25462520122528076 lr:4.340381295646168e-06 tokens_per_second_per_gpu:4549.50444761026 mfu:21.460788704878198 +Step 31466 | loss:0.31570956110954285 lr:4.338614332296164e-06 tokens_per_second_per_gpu:3700.5405850572174 mfu:17.456081317044273 +Step 31467 | loss:0.32639166712760925 lr:4.336847712371272e-06 tokens_per_second_per_gpu:5065.53024272811 mfu:23.894970423528456 +Step 31468 | loss:0.24720481038093567 lr:4.335081435884786e-06 tokens_per_second_per_gpu:5332.4170391806465 mfu:25.153920977978025 +Step 31469 | loss:0.2896011173725128 lr:4.3333155028499715e-06 tokens_per_second_per_gpu:4734.766314194375 mfu:22.334700538489315 +Step 31470 | loss:0.2782750427722931 lr:4.331549913280125e-06 tokens_per_second_per_gpu:3243.887409960912 mfu:15.301970377049187 +Step 31471 | loss:0.36555609107017517 lr:4.329784667188508e-06 tokens_per_second_per_gpu:3981.053667248464 mfu:18.779309386207423 +Step 31472 | loss:0.32470014691352844 lr:4.328019764588414e-06 tokens_per_second_per_gpu:5359.938139343757 mfu:25.2837427030313 +Step 31473 | loss:0.2674294114112854 lr:4.3262552054931074e-06 tokens_per_second_per_gpu:4975.741515182877 mfu:23.471421676160976 +Step 31474 | loss:0.35374224185943604 lr:4.324490989915864e-06 tokens_per_second_per_gpu:5260.290144983669 mfu:24.813686111933766 +Step 31475 | loss:0.41317814588546753 lr:4.322727117869951e-06 tokens_per_second_per_gpu:4819.441890287093 mfu:22.73412967806169 +Step 31476 | loss:0.3111582100391388 lr:4.320963589368643e-06 tokens_per_second_per_gpu:4337.629949412067 mfu:20.461340547360308 +Step 31477 | loss:0.294892817735672 lr:4.3192004044252e-06 tokens_per_second_per_gpu:3636.360511977596 mfu:17.15333296207808 +Step 31478 | loss:0.35577160120010376 lr:4.317437563052867e-06 tokens_per_second_per_gpu:5370.111679829425 mfu:25.33173302928719 +Step 31479 | loss:0.27050620317459106 lr:4.315675065264918e-06 tokens_per_second_per_gpu:4216.185562221983 mfu:19.888466652435593 +Step 31480 | loss:0.285228967666626 lr:4.313912911074614e-06 tokens_per_second_per_gpu:4091.172258312107 mfu:19.298757568423632 +Step 31481 | loss:0.31018492579460144 lr:4.312151100495193e-06 tokens_per_second_per_gpu:5082.25860266193 mfu:23.973880951488802 +Step 31482 | loss:0.40766069293022156 lr:4.310389633539913e-06 tokens_per_second_per_gpu:4824.6149510143705 mfu:22.75853188812743 +Step 31483 | loss:0.2383735626935959 lr:4.308628510222029e-06 tokens_per_second_per_gpu:4453.022834734605 mfu:21.005668475483123 +Step 31484 | loss:0.35947662591934204 lr:4.306867730554781e-06 tokens_per_second_per_gpu:3974.135182106069 mfu:18.746673711375458 +Step 31485 | loss:0.34129711985588074 lr:4.3051072945514024e-06 tokens_per_second_per_gpu:5324.6195944398005 mfu:25.1171391007542 +Step 31486 | loss:0.25688812136650085 lr:4.303347202225145e-06 tokens_per_second_per_gpu:4919.816148588481 mfu:23.207612180083743 +Step 31487 | loss:0.4400317668914795 lr:4.301587453589251e-06 tokens_per_second_per_gpu:5158.928190063499 mfu:24.3355444764425 +Step 31488 | loss:0.24363794922828674 lr:4.29982804865694e-06 tokens_per_second_per_gpu:5317.896105013696 mfu:25.085423253985677 +Step 31489 | loss:0.18257582187652588 lr:4.298068987441456e-06 tokens_per_second_per_gpu:5352.591019216666 mfu:25.249085083843855 +Step 31490 | loss:0.290355920791626 lr:4.296310269956033e-06 tokens_per_second_per_gpu:4587.396962646282 mfu:21.63953416343284 +Step 31491 | loss:0.2472735047340393 lr:4.294551896213883e-06 tokens_per_second_per_gpu:2590.7155120098637 mfu:12.220847091796704 +Step 31492 | loss:0.3620791435241699 lr:4.292793866228251e-06 tokens_per_second_per_gpu:3618.0452101375554 mfu:17.06693655838588 +Step 31493 | loss:0.28251826763153076 lr:4.291036180012337e-06 tokens_per_second_per_gpu:3846.709099089092 mfu:18.14558313665274 +Step 31494 | loss:0.2944400906562805 lr:4.289278837579386e-06 tokens_per_second_per_gpu:4349.699474630211 mfu:20.51827455708764 +Step 31495 | loss:0.36419153213500977 lr:4.287521838942587e-06 tokens_per_second_per_gpu:5382.317324150889 mfu:25.389309136049203 +Step 31496 | loss:0.23199737071990967 lr:4.285765184115176e-06 tokens_per_second_per_gpu:5219.948938587205 mfu:24.62338975844209 +Step 31497 | loss:0.23476921021938324 lr:4.284008873110362e-06 tokens_per_second_per_gpu:3713.239776207646 mfu:17.51598556840628 +Step 31498 | loss:0.2830130159854889 lr:4.282252905941342e-06 tokens_per_second_per_gpu:5340.777388855585 mfu:25.19335817381745 +Step 31499 | loss:0.2758795917034149 lr:4.280497282621343e-06 tokens_per_second_per_gpu:3577.432551499711 mfu:16.875359718357085 +Step 31500 | loss:0.3793609142303467 lr:4.278742003163544e-06 tokens_per_second_per_gpu:4635.043426549319 mfu:21.864290662988477 +Step 31501 | loss:0.43237313628196716 lr:4.276987067581173e-06 tokens_per_second_per_gpu:5333.260360241024 mfu:25.157899067305234 +Step 31502 | loss:0.41937094926834106 lr:4.275232475887409e-06 tokens_per_second_per_gpu:5315.312295298776 mfu:25.073234982717004 +Step 31503 | loss:0.3703089654445648 lr:4.2734782280954564e-06 tokens_per_second_per_gpu:5101.695677084582 mfu:24.06556894784731 +Step 31504 | loss:0.3482421338558197 lr:4.271724324218512e-06 tokens_per_second_per_gpu:4545.725535629563 mfu:21.44296293231638 +Step 31505 | loss:0.25792762637138367 lr:4.26997076426976e-06 tokens_per_second_per_gpu:4916.610225675884 mfu:23.19248929471817 +Step 31506 | loss:0.304739773273468 lr:4.268217548262405e-06 tokens_per_second_per_gpu:4897.937070966593 mfu:23.104404837985722 +Step 31507 | loss:0.3655347526073456 lr:4.2664646762096024e-06 tokens_per_second_per_gpu:5386.619438946063 mfu:25.409602945554507 +Step 31508 | loss:0.26336362957954407 lr:4.264712148124572e-06 tokens_per_second_per_gpu:4718.195107626754 mfu:22.256531329770464 +Step 31509 | loss:0.36351293325424194 lr:4.262959964020463e-06 tokens_per_second_per_gpu:5345.874829410713 mfu:25.217403670629032 +Step 31510 | loss:0.31008073687553406 lr:4.2612081239104705e-06 tokens_per_second_per_gpu:5130.801762200137 mfu:24.202867317347586 +Step 31511 | loss:0.3006559908390045 lr:4.259456627807773e-06 tokens_per_second_per_gpu:5322.88667932924 mfu:25.10896464451188 +Step 31512 | loss:0.3634803891181946 lr:4.257705475725543e-06 tokens_per_second_per_gpu:4962.890634355707 mfu:23.410801878712853 +Step 31513 | loss:0.3105684220790863 lr:4.255954667676942e-06 tokens_per_second_per_gpu:4045.6576546695314 mfu:19.08405741744946 +Step 31514 | loss:0.3560222387313843 lr:4.254204203675144e-06 tokens_per_second_per_gpu:4657.1294499867245 mfu:21.96847420380689 +Step 31515 | loss:0.3595675528049469 lr:4.2524540837333035e-06 tokens_per_second_per_gpu:4435.660521856635 mfu:20.923767483321694 +Step 31516 | loss:0.26927992701530457 lr:4.250704307864612e-06 tokens_per_second_per_gpu:4669.967028249908 mfu:22.029031250791036 +Step 31517 | loss:0.2192421853542328 lr:4.248954876082195e-06 tokens_per_second_per_gpu:4521.00910186708 mfu:21.3263712972003 +Step 31518 | loss:0.3804214596748352 lr:4.247205788399228e-06 tokens_per_second_per_gpu:4445.916130858244 mfu:20.972144940769617 +Step 31519 | loss:0.3686937391757965 lr:4.245457044828871e-06 tokens_per_second_per_gpu:4758.039704315315 mfu:22.444485090539608 +Step 31520 | loss:0.3213406801223755 lr:4.243708645384265e-06 tokens_per_second_per_gpu:5350.508194320506 mfu:25.239260043442098 +Step 31521 | loss:0.3115687966346741 lr:4.241960590078576e-06 tokens_per_second_per_gpu:4277.941916383148 mfu:20.179781911734338 +Step 31522 | loss:0.3532721996307373 lr:4.240212878924926e-06 tokens_per_second_per_gpu:5165.147335532126 mfu:24.36488124671365 +Step 31523 | loss:0.2565983533859253 lr:4.238465511936485e-06 tokens_per_second_per_gpu:5240.302832384723 mfu:24.719402548217364 +Step 31524 | loss:0.32412776350975037 lr:4.236718489126373e-06 tokens_per_second_per_gpu:5073.7337330902765 mfu:23.93366768722658 +Step 31525 | loss:0.22893354296684265 lr:4.234971810507743e-06 tokens_per_second_per_gpu:4921.082584757884 mfu:23.21358617557111 +Step 31526 | loss:0.2934420704841614 lr:4.233225476093739e-06 tokens_per_second_per_gpu:3079.326990234099 mfu:14.525710800295007 +Step 31527 | loss:0.3069591820240021 lr:4.231479485897471e-06 tokens_per_second_per_gpu:5381.6178685620325 mfu:25.386009684697132 +Step 31528 | loss:0.255946546792984 lr:4.229733839932104e-06 tokens_per_second_per_gpu:3871.0911635934453 mfu:18.260597494928945 +Step 31529 | loss:0.3457871675491333 lr:4.2279885382107364e-06 tokens_per_second_per_gpu:2924.0499084659023 mfu:13.793242312592403 +Step 31530 | loss:0.34555768966674805 lr:4.226243580746514e-06 tokens_per_second_per_gpu:5363.353652284573 mfu:25.29985426031973 +Step 31531 | loss:0.2768798768520355 lr:4.2244989675525485e-06 tokens_per_second_per_gpu:3486.2960674877995 mfu:16.445453373785316 +Step 31532 | loss:0.16024532914161682 lr:4.222754698641967e-06 tokens_per_second_per_gpu:4046.1076341825274 mfu:19.086180047586687 +Step 31533 | loss:0.33604156970977783 lr:4.221010774027895e-06 tokens_per_second_per_gpu:5329.914935310596 mfu:25.142118127120057 +Step 31534 | loss:0.25940176844596863 lr:4.219267193723436e-06 tokens_per_second_per_gpu:5360.551042833061 mfu:25.28663387336246 +Step 31535 | loss:0.26201194524765015 lr:4.21752395774172e-06 tokens_per_second_per_gpu:5364.114396207531 mfu:25.30344281920062 +Step 31536 | loss:0.3731078803539276 lr:4.215781066095836e-06 tokens_per_second_per_gpu:4501.633456912892 mfu:21.23497307412548 +Step 31537 | loss:0.3818313479423523 lr:4.214038518798913e-06 tokens_per_second_per_gpu:3453.4245029308945 mfu:16.290392595303153 +Step 31538 | loss:0.2957117557525635 lr:4.212296315864045e-06 tokens_per_second_per_gpu:3170.953800966323 mfu:14.957930099664857 +Step 31539 | loss:0.40838733315467834 lr:4.210554457304333e-06 tokens_per_second_per_gpu:5389.568698265038 mfu:25.423515104957637 +Step 31540 | loss:0.2893029749393463 lr:4.208812943132895e-06 tokens_per_second_per_gpu:4524.537176627502 mfu:21.343013827796835 +Step 31541 | loss:0.24983595311641693 lr:4.207071773362814e-06 tokens_per_second_per_gpu:4591.247094171008 mfu:21.65769589073515 +Step 31542 | loss:0.38503497838974 lr:4.2053309480071915e-06 tokens_per_second_per_gpu:4666.221561630216 mfu:22.011363245703446 +Step 31543 | loss:0.3554656207561493 lr:4.2035904670791105e-06 tokens_per_second_per_gpu:5363.6867528155735 mfu:25.30142555235727 +Step 31544 | loss:0.30277132987976074 lr:4.2018503305916775e-06 tokens_per_second_per_gpu:4208.468993522296 mfu:19.852066281296786 +Step 31545 | loss:0.26070818305015564 lr:4.2001105385579666e-06 tokens_per_second_per_gpu:3739.292621419132 mfu:17.638881284342613 +Step 31546 | loss:0.28726986050605774 lr:4.198371090991071e-06 tokens_per_second_per_gpu:4376.929686533085 mfu:20.646724112586725 +Step 31547 | loss:0.2884443402290344 lr:4.196631987904065e-06 tokens_per_second_per_gpu:4911.479290465603 mfu:23.168285797903458 +Step 31548 | loss:0.47577306628227234 lr:4.194893229310043e-06 tokens_per_second_per_gpu:5381.495109286592 mfu:25.38543060824999 +Step 31549 | loss:0.35527303814888 lr:4.193154815222067e-06 tokens_per_second_per_gpu:5353.456989031619 mfu:25.253170011210674 +Step 31550 | loss:0.3508402705192566 lr:4.191416745653226e-06 tokens_per_second_per_gpu:3062.3372212960817 mfu:14.445567161460888 +Step 31551 | loss:0.26651671528816223 lr:4.189679020616577e-06 tokens_per_second_per_gpu:5327.247018913618 mfu:25.12953311028204 +Step 31552 | loss:0.3373304605484009 lr:4.187941640125209e-06 tokens_per_second_per_gpu:4402.8874777573865 mfu:20.769171442647625 +Step 31553 | loss:0.40077346563339233 lr:4.186204604192167e-06 tokens_per_second_per_gpu:4509.346349644368 mfu:21.271356105095048 +Step 31554 | loss:0.3294784128665924 lr:4.184467912830525e-06 tokens_per_second_per_gpu:4048.785638463867 mfu:19.09881265069641 +Step 31555 | loss:0.31979018449783325 lr:4.182731566053355e-06 tokens_per_second_per_gpu:4960.389200652326 mfu:23.399002189145403 +Step 31556 | loss:0.27377060055732727 lr:4.180995563873696e-06 tokens_per_second_per_gpu:5361.994631739458 mfu:25.293443528535605 +Step 31557 | loss:0.2911243736743927 lr:4.179259906304628e-06 tokens_per_second_per_gpu:4682.951240671889 mfu:22.090279996120593 +Step 31558 | loss:0.30946195125579834 lr:4.1775245933591835e-06 tokens_per_second_per_gpu:3690.2838089502907 mfu:17.407698354161443 +Step 31559 | loss:0.3965696394443512 lr:4.175789625050436e-06 tokens_per_second_per_gpu:3507.8071551431544 mfu:16.546924844425632 +Step 31560 | loss:0.32275864481925964 lr:4.174055001391408e-06 tokens_per_second_per_gpu:5060.707453575768 mfu:23.872220504244183 +Step 31561 | loss:0.30732545256614685 lr:4.172320722395162e-06 tokens_per_second_per_gpu:4448.978816111927 mfu:20.986592149659224 +Step 31562 | loss:0.3833734095096588 lr:4.170586788074748e-06 tokens_per_second_per_gpu:5101.987490812819 mfu:24.066945482991965 +Step 31563 | loss:0.30235788226127625 lr:4.16885319844319e-06 tokens_per_second_per_gpu:3496.5455479382617 mfu:16.493801921811833 +Step 31564 | loss:0.39716875553131104 lr:4.167119953513543e-06 tokens_per_second_per_gpu:4117.176352948628 mfu:19.421423319580153 +Step 31565 | loss:0.29912254214286804 lr:4.165387053298825e-06 tokens_per_second_per_gpu:3674.362223709044 mfu:17.332593520075925 +Step 31566 | loss:0.320969820022583 lr:4.163654497812092e-06 tokens_per_second_per_gpu:5389.786456616802 mfu:25.424542308256317 +Step 31567 | loss:0.3051399886608124 lr:4.16192228706635e-06 tokens_per_second_per_gpu:4928.099406487118 mfu:23.246685720861144 +Step 31568 | loss:0.31469210982322693 lr:4.160190421074639e-06 tokens_per_second_per_gpu:4511.0048122992475 mfu:21.279179356401023 +Step 31569 | loss:0.24504457414150238 lr:4.158458899849999e-06 tokens_per_second_per_gpu:5270.705077043426 mfu:24.862815123430284 +Step 31570 | loss:0.2627362608909607 lr:4.156727723405423e-06 tokens_per_second_per_gpu:4266.270081868046 mfu:20.124723876906284 +Step 31571 | loss:0.3348325192928314 lr:4.154996891753965e-06 tokens_per_second_per_gpu:3783.9126463605876 mfu:17.849361554952736 +Step 31572 | loss:0.3696969449520111 lr:4.1532664049086125e-06 tokens_per_second_per_gpu:4202.842189894324 mfu:19.825523688551947 +Step 31573 | loss:0.2223500907421112 lr:4.1515362628823996e-06 tokens_per_second_per_gpu:4149.208985836454 mfu:19.572526714242766 +Step 31574 | loss:0.37160882353782654 lr:4.1498064656883275e-06 tokens_per_second_per_gpu:5180.122981461375 mfu:24.435523923670587 +Step 31575 | loss:0.35530030727386475 lr:4.1480770133394075e-06 tokens_per_second_per_gpu:5118.450779383932 mfu:24.144605624108582 +Step 31576 | loss:0.3613617718219757 lr:4.1463479058486635e-06 tokens_per_second_per_gpu:3426.0237082432654 mfu:16.161138371703956 +Step 31577 | loss:0.31668609380722046 lr:4.144619143229078e-06 tokens_per_second_per_gpu:4785.03486880837 mfu:22.57182588730335 +Step 31578 | loss:0.3715779185295105 lr:4.14289072549367e-06 tokens_per_second_per_gpu:4390.501981680365 mfu:20.71074694446878 +Step 31579 | loss:0.2697213292121887 lr:4.141162652655428e-06 tokens_per_second_per_gpu:4455.543461932438 mfu:21.01755870403865 +Step 31580 | loss:0.2773684859275818 lr:4.139434924727359e-06 tokens_per_second_per_gpu:5091.737437179672 mfu:24.018594231164926 +Step 31581 | loss:0.27281299233436584 lr:4.137707541722441e-06 tokens_per_second_per_gpu:5338.550733254509 mfu:25.18285465944056 +Step 31582 | loss:0.3095792233943939 lr:4.13598050365368e-06 tokens_per_second_per_gpu:5327.939808484916 mfu:25.13280111689194 +Step 31583 | loss:0.3493003249168396 lr:4.134253810534061e-06 tokens_per_second_per_gpu:4636.888448952035 mfu:21.872993948455374 +Step 31584 | loss:0.27768856287002563 lr:4.132527462376579e-06 tokens_per_second_per_gpu:4559.78399277659 mfu:21.509279073299034 +Step 31585 | loss:0.31312230229377747 lr:4.1308014591942e-06 tokens_per_second_per_gpu:4280.51015498024 mfu:20.191896731383512 +Step 31586 | loss:0.43914514780044556 lr:4.129075800999932e-06 tokens_per_second_per_gpu:5356.380669724698 mfu:25.266961511871298 +Step 31587 | loss:0.33706608414649963 lr:4.12735048780673e-06 tokens_per_second_per_gpu:3925.435142041048 mfu:18.516947313305113 +Step 31588 | loss:0.29981109499931335 lr:4.125625519627574e-06 tokens_per_second_per_gpu:5384.057077281434 mfu:25.397515848398474 +Step 31589 | loss:0.3923047184944153 lr:4.123900896475447e-06 tokens_per_second_per_gpu:4336.677562104264 mfu:20.456847973935375 +Step 31590 | loss:0.2836771607398987 lr:4.122176618363305e-06 tokens_per_second_per_gpu:5346.265699558176 mfu:25.21924747180394 +Step 31591 | loss:0.310971200466156 lr:4.120452685304143e-06 tokens_per_second_per_gpu:4785.230976926731 mfu:22.57275096275715 +Step 31592 | loss:0.24418151378631592 lr:4.1187290973108976e-06 tokens_per_second_per_gpu:4917.717696709344 mfu:23.19771342453704 +Step 31593 | loss:0.4190119206905365 lr:4.117005854396555e-06 tokens_per_second_per_gpu:4409.615007724343 mfu:20.80090634933673 +Step 31594 | loss:0.22676074504852295 lr:4.115282956574063e-06 tokens_per_second_per_gpu:3918.916416245982 mfu:18.486197371520475 +Step 31595 | loss:0.32176241278648376 lr:4.113560403856381e-06 tokens_per_second_per_gpu:3700.0255211504855 mfu:17.453651672717253 +Step 31596 | loss:0.29148444533348083 lr:4.111838196256467e-06 tokens_per_second_per_gpu:5355.386313592589 mfu:25.26227096434149 +Step 31597 | loss:0.3094344735145569 lr:4.110116333787267e-06 tokens_per_second_per_gpu:5072.794947072877 mfu:23.929239273409095 +Step 31598 | loss:0.34546351432800293 lr:4.108394816461747e-06 tokens_per_second_per_gpu:5021.322752112762 mfu:23.686436147720194 +Step 31599 | loss:0.3549058437347412 lr:4.106673644292841e-06 tokens_per_second_per_gpu:3264.0305077384787 mfu:15.396988806032839 +Step 31600 | loss:0.2963700592517853 lr:4.104952817293506e-06 tokens_per_second_per_gpu:4741.943085470765 mfu:22.368554593083495 +Step 31601 | loss:0.2629677653312683 lr:4.1032323354766644e-06 tokens_per_second_per_gpu:4992.968583051838 mfu:23.552684694539725 +Step 31602 | loss:0.2682148218154907 lr:4.101512198855284e-06 tokens_per_second_per_gpu:4497.059927847259 mfu:21.213398957198365 +Step 31603 | loss:0.21047137677669525 lr:4.099792407442271e-06 tokens_per_second_per_gpu:5070.040049021962 mfu:23.916243949268853 +Step 31604 | loss:0.2675847113132477 lr:4.098072961250582e-06 tokens_per_second_per_gpu:5094.405223002128 mfu:24.031178632060648 +Step 31605 | loss:0.3795836865901947 lr:4.096353860293151e-06 tokens_per_second_per_gpu:4797.247823756799 mfu:22.62943647954007 +Step 31606 | loss:0.32238584756851196 lr:4.094635104582889e-06 tokens_per_second_per_gpu:5339.769986581915 mfu:25.188606085412 +Step 31607 | loss:0.37052783370018005 lr:4.092916694132743e-06 tokens_per_second_per_gpu:4359.8410703877325 mfu:20.56611419461053 +Step 31608 | loss:0.36003196239471436 lr:4.091198628955622e-06 tokens_per_second_per_gpu:4535.2916962193995 mfu:21.393744731622068 +Step 31609 | loss:0.3574654459953308 lr:4.089480909064458e-06 tokens_per_second_per_gpu:4947.407749117911 mfu:23.33776646739355 +Step 31610 | loss:0.39801737666130066 lr:4.087763534472161e-06 tokens_per_second_per_gpu:5066.412793296851 mfu:23.899133565139863 +Step 31611 | loss:0.36706244945526123 lr:4.086046505191654e-06 tokens_per_second_per_gpu:4651.854220990898 mfu:21.943590048586696 +Step 31612 | loss:0.3516765534877777 lr:4.084329821235855e-06 tokens_per_second_per_gpu:5334.539096760619 mfu:25.163931085643114 +Step 31613 | loss:0.34832319617271423 lr:4.082613482617664e-06 tokens_per_second_per_gpu:5316.311802028217 mfu:25.077949826493214 +Step 31614 | loss:0.28825387358665466 lr:4.08089748935e-06 tokens_per_second_per_gpu:5339.4054419168615 mfu:25.186886466029208 +Step 31615 | loss:0.289920449256897 lr:4.0791818414457675e-06 tokens_per_second_per_gpu:5349.16487492524 mfu:25.232923376660704 +Step 31616 | loss:0.25602248311042786 lr:4.077466538917868e-06 tokens_per_second_per_gpu:4263.862547256857 mfu:20.113367125377543 +Step 31617 | loss:0.3270072340965271 lr:4.075751581779197e-06 tokens_per_second_per_gpu:4005.52940940813 mfu:18.894765637966792 +Step 31618 | loss:0.3885461390018463 lr:4.07403697004266e-06 tokens_per_second_per_gpu:4732.199107682147 mfu:22.322590587360537 +Step 31619 | loss:0.3455178141593933 lr:4.0723227037211484e-06 tokens_per_second_per_gpu:5350.0201028455795 mfu:25.23695763267788 +Step 31620 | loss:0.2740612328052521 lr:4.070608782827568e-06 tokens_per_second_per_gpu:4680.574161528154 mfu:22.07906690822808 +Step 31621 | loss:0.2865440249443054 lr:4.06889520737479e-06 tokens_per_second_per_gpu:5355.138552069506 mfu:25.261102230964628 +Step 31622 | loss:0.21053554117679596 lr:4.067181977375717e-06 tokens_per_second_per_gpu:5402.3540836986185 mfu:25.483825949459536 +Step 31623 | loss:0.32473936676979065 lr:4.065469092843233e-06 tokens_per_second_per_gpu:5374.125371770788 mfu:25.35066629898106 +Step 31624 | loss:0.3599106967449188 lr:4.06375655379021e-06 tokens_per_second_per_gpu:4209.607874568792 mfu:19.857438577506247 +Step 31625 | loss:0.32241421937942505 lr:4.0620443602295334e-06 tokens_per_second_per_gpu:4530.095942111411 mfu:21.369235473007624 +Step 31626 | loss:0.3868092894554138 lr:4.0603325121740865e-06 tokens_per_second_per_gpu:4717.106034399448 mfu:22.251393985542204 +Step 31627 | loss:0.29644420742988586 lr:4.0586210096367425e-06 tokens_per_second_per_gpu:4553.481652959717 mfu:21.479549861092792 +Step 31628 | loss:0.2869252562522888 lr:4.056909852630369e-06 tokens_per_second_per_gpu:5349.795011815937 mfu:25.235895839886563 +Step 31629 | loss:0.3096443712711334 lr:4.0551990411678455e-06 tokens_per_second_per_gpu:4733.638518328301 mfu:22.329380532967228 +Step 31630 | loss:0.37354186177253723 lr:4.053488575262026e-06 tokens_per_second_per_gpu:5304.3684563923225 mfu:25.021611027383805 +Step 31631 | loss:0.30813512206077576 lr:4.0517784549257866e-06 tokens_per_second_per_gpu:3186.776753210501 mfu:15.03256966507468 +Step 31632 | loss:0.3647614121437073 lr:4.050068680171981e-06 tokens_per_second_per_gpu:5330.095709447703 mfu:25.142970869568035 +Step 31633 | loss:0.24846290051937103 lr:4.048359251013467e-06 tokens_per_second_per_gpu:3226.1486442247065 mfu:15.218293592525983 +Step 31634 | loss:0.3288586139678955 lr:4.046650167463118e-06 tokens_per_second_per_gpu:5367.076320354731 mfu:25.317414720759334 +Step 31635 | loss:0.1982976645231247 lr:4.0449414295337666e-06 tokens_per_second_per_gpu:5353.739356982033 mfu:25.254501989009718 +Step 31636 | loss:0.3683314621448517 lr:4.043233037238281e-06 tokens_per_second_per_gpu:4431.920050534366 mfu:20.90612304190434 +Step 31637 | loss:0.44492170214653015 lr:4.0415249905895015e-06 tokens_per_second_per_gpu:4142.4907602590265 mfu:19.54083569794656 +Step 31638 | loss:0.2916615903377533 lr:4.039817289600278e-06 tokens_per_second_per_gpu:5327.588475955277 mfu:25.131143821406123 +Step 31639 | loss:0.2668928802013397 lr:4.038109934283451e-06 tokens_per_second_per_gpu:4793.592239956668 mfu:22.612192467047944 +Step 31640 | loss:0.268553227186203 lr:4.0364029246518605e-06 tokens_per_second_per_gpu:4201.621162462265 mfu:19.81976389382616 +Step 31641 | loss:0.24546736478805542 lr:4.034696260718363e-06 tokens_per_second_per_gpu:4519.222523056692 mfu:21.31794370012992 +Step 31642 | loss:0.3060682415962219 lr:4.032989942495763e-06 tokens_per_second_per_gpu:4029.6432821226044 mfu:19.0085148898119 +Step 31643 | loss:0.24128910899162292 lr:4.031283969996918e-06 tokens_per_second_per_gpu:4799.232291521401 mfu:22.63879755257081 +Step 31644 | loss:0.45393723249435425 lr:4.029578343234658e-06 tokens_per_second_per_gpu:5000.513137386071 mfu:23.58827364456801 +Step 31645 | loss:0.33652862906455994 lr:4.027873062221804e-06 tokens_per_second_per_gpu:5352.9507846684855 mfu:25.250782158862407 +Step 31646 | loss:0.27873483300209045 lr:4.026168126971175e-06 tokens_per_second_per_gpu:4185.3314721834695 mfu:19.74292264547333 +Step 31647 | loss:0.33263370394706726 lr:4.0244635374955985e-06 tokens_per_second_per_gpu:5311.358364880041 mfu:25.05458361832174 +Step 31648 | loss:0.33236342668533325 lr:4.02275929380791e-06 tokens_per_second_per_gpu:5375.001457445319 mfu:25.354798944583454 +Step 31649 | loss:0.3445170819759369 lr:4.021055395920908e-06 tokens_per_second_per_gpu:5172.524305324235 mfu:24.399679671863858 +Step 31650 | loss:0.25522926449775696 lr:4.019351843847413e-06 tokens_per_second_per_gpu:4540.985888632851 mfu:21.42060520876573 +Step 31651 | loss:0.24753008782863617 lr:4.017648637600252e-06 tokens_per_second_per_gpu:5309.8590178614895 mfu:25.04751094262025 +Step 31652 | loss:0.26792624592781067 lr:4.01594577719222e-06 tokens_per_second_per_gpu:3406.1703910730666 mfu:16.06748688728693 +Step 31653 | loss:0.3035781681537628 lr:4.014243262636119e-06 tokens_per_second_per_gpu:5346.887079721367 mfu:25.222178628052102 +Step 31654 | loss:0.3850035071372986 lr:4.012541093944766e-06 tokens_per_second_per_gpu:5226.833155896294 mfu:24.655863786057655 +Step 31655 | loss:0.28623664379119873 lr:4.010839271130962e-06 tokens_per_second_per_gpu:3440.511955757773 mfu:16.229481907179814 +Step 31656 | loss:0.3346834182739258 lr:4.009137794207502e-06 tokens_per_second_per_gpu:3567.9055971072676 mfu:16.830419449021896 +Step 31657 | loss:0.2147921919822693 lr:4.00743666318718e-06 tokens_per_second_per_gpu:4054.9064676930943 mfu:19.12768564649183 +Step 31658 | loss:0.3727070987224579 lr:4.0057358780828055e-06 tokens_per_second_per_gpu:5190.816315868056 mfu:24.485966206537885 +Step 31659 | loss:0.30258241295814514 lr:4.004035438907161e-06 tokens_per_second_per_gpu:4258.693629432887 mfu:20.088984458093556 +Step 31660 | loss:0.36316365003585815 lr:4.002335345673025e-06 tokens_per_second_per_gpu:5184.66787732495 mfu:24.45696297289826 +Step 31661 | loss:0.23047371208667755 lr:4.000635598393199e-06 tokens_per_second_per_gpu:4405.068139471033 mfu:20.779457996010148 +Step 31662 | loss:0.29920870065689087 lr:3.998936197080455e-06 tokens_per_second_per_gpu:5210.989385373391 mfu:24.58112601727512 +Step 31663 | loss:0.3001997768878937 lr:3.997237141747595e-06 tokens_per_second_per_gpu:4386.808006709421 mfu:20.693321834273714 +Step 31664 | loss:0.33643797039985657 lr:3.995538432407375e-06 tokens_per_second_per_gpu:4094.0431007804086 mfu:19.312299822162846 +Step 31665 | loss:0.45074424147605896 lr:3.993840069072591e-06 tokens_per_second_per_gpu:4489.203962197968 mfu:21.17634102686471 +Step 31666 | loss:0.2906872034072876 lr:3.992142051755993e-06 tokens_per_second_per_gpu:5025.0271623241 mfu:23.70391048272463 +Step 31667 | loss:0.31644532084465027 lr:3.9904443804703765e-06 tokens_per_second_per_gpu:3571.7946491663693 mfu:16.848764771125385 +Step 31668 | loss:0.34129950404167175 lr:3.988747055228492e-06 tokens_per_second_per_gpu:3465.524801405295 mfu:16.347471767730706 +Step 31669 | loss:0.2637862265110016 lr:3.987050076043108e-06 tokens_per_second_per_gpu:3748.1801237206582 mfu:17.68080514906317 +Step 31670 | loss:0.3102515935897827 lr:3.985353442927003e-06 tokens_per_second_per_gpu:5387.951016003022 mfu:25.415884221722656 +Step 31671 | loss:0.42090538144111633 lr:3.983657155892917e-06 tokens_per_second_per_gpu:3302.767338775215 mfu:15.579717047218018 +Step 31672 | loss:0.2952388823032379 lr:3.981961214953622e-06 tokens_per_second_per_gpu:3762.8790440948956 mfu:17.750142464362867 +Step 31673 | loss:0.33367085456848145 lr:3.980265620121864e-06 tokens_per_second_per_gpu:5035.081628540217 mfu:23.751339115334886 +Step 31674 | loss:0.22328607738018036 lr:3.978570371410412e-06 tokens_per_second_per_gpu:5350.751193638166 mfu:25.240406312683458 +Step 31675 | loss:0.2661646008491516 lr:3.976875468831992e-06 tokens_per_second_per_gpu:5183.452261032709 mfu:24.45122870344224 +Step 31676 | loss:0.32608598470687866 lr:3.975180912399362e-06 tokens_per_second_per_gpu:5346.130496175044 mfu:25.21860969437763 +Step 31677 | loss:0.40465864539146423 lr:3.973486702125284e-06 tokens_per_second_per_gpu:4914.156932708513 mfu:23.180916693216137 +Step 31678 | loss:0.36913350224494934 lr:3.9717928380224755e-06 tokens_per_second_per_gpu:4157.38161045313 mfu:19.61107837895317 +Step 31679 | loss:0.2748767137527466 lr:3.9700993201036815e-06 tokens_per_second_per_gpu:5386.0776490405215 mfu:25.40704723013118 +Step 31680 | loss:0.24144712090492249 lr:3.968406148381659e-06 tokens_per_second_per_gpu:5339.013325195669 mfu:25.18503678455333 +Step 31681 | loss:0.29245561361312866 lr:3.96671332286912e-06 tokens_per_second_per_gpu:4508.44355241642 mfu:21.267097456537726 +Step 31682 | loss:0.375053346157074 lr:3.9650208435788035e-06 tokens_per_second_per_gpu:5240.452067018105 mfu:24.720106513445074 +Step 31683 | loss:0.35742905735969543 lr:3.963328710523434e-06 tokens_per_second_per_gpu:5343.122610114519 mfu:25.204420982631856 +Step 31684 | loss:0.3532525300979614 lr:3.96163692371575e-06 tokens_per_second_per_gpu:3514.3587436680787 mfu:16.577829805313968 +Step 31685 | loss:0.29100164771080017 lr:3.959945483168465e-06 tokens_per_second_per_gpu:4498.429183121273 mfu:21.21985796794493 +Step 31686 | loss:0.3104988932609558 lr:3.958254388894306e-06 tokens_per_second_per_gpu:4186.423771306087 mfu:19.748075206800163 +Step 31687 | loss:0.2877586781978607 lr:3.956563640905997e-06 tokens_per_second_per_gpu:5339.54868734683 mfu:25.187562179162562 +Step 31688 | loss:0.318300724029541 lr:3.95487323921625e-06 tokens_per_second_per_gpu:4130.220724566386 mfu:19.48295584610041 +Step 31689 | loss:0.2494231015443802 lr:3.953183183837767e-06 tokens_per_second_per_gpu:4227.851964773593 mfu:19.943499063765064 +Step 31690 | loss:0.27335986495018005 lr:3.951493474783263e-06 tokens_per_second_per_gpu:4955.467942349709 mfu:23.37578777407871 +Step 31691 | loss:0.3668765425682068 lr:3.949804112065469e-06 tokens_per_second_per_gpu:5316.725880858149 mfu:25.079903106983135 +Step 31692 | loss:0.43610358238220215 lr:3.948115095697064e-06 tokens_per_second_per_gpu:3943.2484623129044 mfu:18.600975784293293 +Step 31693 | loss:0.39043259620666504 lr:3.9464264256907575e-06 tokens_per_second_per_gpu:5313.825049004279 mfu:25.066219388195478 +Step 31694 | loss:0.31362438201904297 lr:3.94473810205927e-06 tokens_per_second_per_gpu:5299.360612851729 mfu:24.997988174976612 +Step 31695 | loss:0.3103828430175781 lr:3.943050124815279e-06 tokens_per_second_per_gpu:4681.560434934874 mfu:22.08371932816317 +Step 31696 | loss:0.28089308738708496 lr:3.9413624939714785e-06 tokens_per_second_per_gpu:5360.710838980352 mfu:25.28738765905327 +Step 31697 | loss:0.22411610186100006 lr:3.939675209540566e-06 tokens_per_second_per_gpu:5117.245953828237 mfu:24.138922256397933 +Step 31698 | loss:0.37269845604896545 lr:3.937988271535231e-06 tokens_per_second_per_gpu:4513.175518750893 mfu:21.28941894909415 +Step 31699 | loss:0.27979904413223267 lr:3.93630167996818e-06 tokens_per_second_per_gpu:5369.648377491475 mfu:25.329547553111553 +Step 31700 | loss:0.39279547333717346 lr:3.934615434852068e-06 tokens_per_second_per_gpu:4449.302583693136 mfu:20.98811941658023 +Step 31701 | loss:0.3386402726173401 lr:3.932929536199598e-06 tokens_per_second_per_gpu:5325.076296694088 mfu:25.119293443209113 +Step 31702 | loss:0.3144642412662506 lr:3.931243984023442e-06 tokens_per_second_per_gpu:5351.425449681667 mfu:25.243586893480593 +Step 31703 | loss:0.24448491632938385 lr:3.929558778336279e-06 tokens_per_second_per_gpu:3731.4939245901787 mfu:17.60209350080555 +Step 31704 | loss:0.30331385135650635 lr:3.927873919150776e-06 tokens_per_second_per_gpu:4048.5437722435463 mfu:19.097671726468526 +Step 31705 | loss:0.35514819622039795 lr:3.9261894064796135e-06 tokens_per_second_per_gpu:5339.352213426465 mfu:25.186635378158293 +Step 31706 | loss:0.17390413582324982 lr:3.924505240335469e-06 tokens_per_second_per_gpu:3328.8526938938608 mfu:15.702766117933185 +Step 31707 | loss:0.3197803795337677 lr:3.922821420730993e-06 tokens_per_second_per_gpu:5057.211825939779 mfu:23.855731032269645 +Step 31708 | loss:0.4146774411201477 lr:3.921137947678855e-06 tokens_per_second_per_gpu:5372.605590399396 mfu:25.343497230950764 +Step 31709 | loss:0.286981999874115 lr:3.919454821191726e-06 tokens_per_second_per_gpu:4970.87374200728 mfu:23.44845955956394 +Step 31710 | loss:0.2933768332004547 lr:3.917772041282258e-06 tokens_per_second_per_gpu:5339.820751657722 mfu:25.188845552936 +Step 31711 | loss:0.44838619232177734 lr:3.916089607963097e-06 tokens_per_second_per_gpu:4568.133799584038 mfu:21.548666536633707 +Step 31712 | loss:0.2703877091407776 lr:3.914407521246905e-06 tokens_per_second_per_gpu:4207.297627812791 mfu:19.84654074938945 +Step 31713 | loss:0.2561057507991791 lr:3.912725781146343e-06 tokens_per_second_per_gpu:4954.6482052495785 mfu:23.37192093431582 +Step 31714 | loss:0.3363412320613861 lr:3.911044387674046e-06 tokens_per_second_per_gpu:3943.89162775477 mfu:18.604009705417916 +Step 31715 | loss:0.20875152945518494 lr:3.909363340842665e-06 tokens_per_second_per_gpu:3400.3612413498536 mfu:16.040084136899388 +Step 31716 | loss:0.25492820143699646 lr:3.90768264066485e-06 tokens_per_second_per_gpu:5061.24809268746 mfu:23.874770791176573 +Step 31717 | loss:0.37934035062789917 lr:3.9060022871532375e-06 tokens_per_second_per_gpu:5313.134970234476 mfu:25.062964168898738 +Step 31718 | loss:0.3531554639339447 lr:3.904322280320449e-06 tokens_per_second_per_gpu:4564.238108265135 mfu:21.530289896008966 +Step 31719 | loss:0.23372703790664673 lr:3.902642620179142e-06 tokens_per_second_per_gpu:4955.062805401617 mfu:23.37387667397107 +Step 31720 | loss:0.3624323010444641 lr:3.900963306741945e-06 tokens_per_second_per_gpu:4761.368203620202 mfu:22.460186189661297 +Step 31721 | loss:0.3399377167224884 lr:3.89928434002148e-06 tokens_per_second_per_gpu:4910.715984965656 mfu:23.16468515562686 +Step 31722 | loss:0.29905518889427185 lr:3.897605720030378e-06 tokens_per_second_per_gpu:4502.42887376261 mfu:21.238725191117705 +Step 31723 | loss:0.3214398920536041 lr:3.895927446781278e-06 tokens_per_second_per_gpu:4902.151065559066 mfu:23.124282969460037 +Step 31724 | loss:0.29832619428634644 lr:3.894249520286786e-06 tokens_per_second_per_gpu:3860.6740787691674 mfu:18.211458328474883 +Step 31725 | loss:0.2529524862766266 lr:3.8925719405595195e-06 tokens_per_second_per_gpu:3521.1962681580726 mfu:16.610083574935242 +Step 31726 | loss:0.2858615815639496 lr:3.890894707612108e-06 tokens_per_second_per_gpu:5369.631556302703 mfu:25.329468204694123 +Step 31727 | loss:0.34339281916618347 lr:3.889217821457164e-06 tokens_per_second_per_gpu:4046.873604248593 mfu:19.089793259076288 +Step 31728 | loss:0.328986257314682 lr:3.887541282107288e-06 tokens_per_second_per_gpu:4017.253981074589 mfu:18.95007244293568 +Step 31729 | loss:0.27758291363716125 lr:3.885865089575097e-06 tokens_per_second_per_gpu:4153.293559963959 mfu:19.591794347302415 +Step 31730 | loss:0.2774902880191803 lr:3.88418924387321e-06 tokens_per_second_per_gpu:4995.053770732418 mfu:23.562520880598726 +Step 31731 | loss:0.35212981700897217 lr:3.8825137450142216e-06 tokens_per_second_per_gpu:4609.175772325114 mfu:21.742268524535987 +Step 31732 | loss:0.37654775381088257 lr:3.880838593010722e-06 tokens_per_second_per_gpu:5074.615546737531 mfu:23.937827352653603 +Step 31733 | loss:0.2995605766773224 lr:3.879163787875317e-06 tokens_per_second_per_gpu:4027.940752960102 mfu:19.00048376927095 +Step 31734 | loss:0.26155731081962585 lr:3.8774893296206095e-06 tokens_per_second_per_gpu:4690.793475708485 mfu:22.12727315681149 +Step 31735 | loss:0.24064084887504578 lr:3.875815218259199e-06 tokens_per_second_per_gpu:5067.295356368959 mfu:23.90329676573178 +Step 31736 | loss:0.36733558773994446 lr:3.874141453803659e-06 tokens_per_second_per_gpu:5353.471126433998 mfu:25.253236699749785 +Step 31737 | loss:0.2526426911354065 lr:3.872468036266585e-06 tokens_per_second_per_gpu:3506.304632380477 mfu:16.539837188197264 +Step 31738 | loss:0.2992781400680542 lr:3.870794965660574e-06 tokens_per_second_per_gpu:5334.167758030488 mfu:25.162179417496166 +Step 31739 | loss:0.24407079815864563 lr:3.869122241998197e-06 tokens_per_second_per_gpu:4224.764267350532 mfu:19.92893386820498 +Step 31740 | loss:0.34468480944633484 lr:3.867449865292034e-06 tokens_per_second_per_gpu:4135.687209167083 mfu:19.508742186640347 +Step 31741 | loss:0.3323550522327423 lr:3.865777835554668e-06 tokens_per_second_per_gpu:4120.244883856047 mfu:19.435898103415806 +Step 31742 | loss:0.2703123688697815 lr:3.86410615279868e-06 tokens_per_second_per_gpu:4215.263088498037 mfu:19.884115186489492 +Step 31743 | loss:0.3381597697734833 lr:3.862434817036626e-06 tokens_per_second_per_gpu:3889.6171378526456 mfu:18.34798767636674 +Step 31744 | loss:0.25839269161224365 lr:3.860763828281089e-06 tokens_per_second_per_gpu:3564.2718977010923 mfu:16.81327866894992 +Step 31745 | loss:0.3759075701236725 lr:3.8590931865446435e-06 tokens_per_second_per_gpu:4800.5173218607715 mfu:22.64485925993041 +Step 31746 | loss:0.29701727628707886 lr:3.857422891839846e-06 tokens_per_second_per_gpu:5354.6419308364075 mfu:25.258759583875783 +Step 31747 | loss:0.24095581471920013 lr:3.855752944179247e-06 tokens_per_second_per_gpu:5319.360494852817 mfu:25.09233103070748 +Step 31748 | loss:0.2657138407230377 lr:3.854083343575421e-06 tokens_per_second_per_gpu:3388.614846754874 mfu:15.984674389453291 +Step 31749 | loss:0.22530195116996765 lr:3.852414090040934e-06 tokens_per_second_per_gpu:5337.593871611864 mfu:25.178340979814774 +Step 31750 | loss:0.2484906017780304 lr:3.850745183588316e-06 tokens_per_second_per_gpu:4613.561089742577 mfu:21.762954815006463 +Step 31751 | loss:0.33241763710975647 lr:3.8490766242301355e-06 tokens_per_second_per_gpu:4243.2699593776315 mfu:20.016228374893448 +Step 31752 | loss:0.2900432348251343 lr:3.847408411978942e-06 tokens_per_second_per_gpu:3910.473906539133 mfu:18.446372612792583 +Step 31753 | loss:0.3492709696292877 lr:3.845740546847282e-06 tokens_per_second_per_gpu:3601.893344800747 mfu:16.99074545379976 +Step 31754 | loss:0.2901950180530548 lr:3.844073028847689e-06 tokens_per_second_per_gpu:4859.60420607611 mfu:22.923582174036152 +Step 31755 | loss:0.2846565544605255 lr:3.842405857992714e-06 tokens_per_second_per_gpu:5309.970158161361 mfu:25.048035210377233 +Step 31756 | loss:0.305711567401886 lr:3.840739034294899e-06 tokens_per_second_per_gpu:5231.650159238266 mfu:24.678586412687253 +Step 31757 | loss:0.24574512243270874 lr:3.839072557766771e-06 tokens_per_second_per_gpu:4387.635104140499 mfu:20.697223393973484 +Step 31758 | loss:0.3103410303592682 lr:3.837406428420864e-06 tokens_per_second_per_gpu:3937.4514255226563 mfu:18.57363017267723 +Step 31759 | loss:0.32894688844680786 lr:3.835740646269726e-06 tokens_per_second_per_gpu:4347.030485991091 mfu:20.505684482300495 +Step 31760 | loss:0.4196329712867737 lr:3.834075211325872e-06 tokens_per_second_per_gpu:5332.708845260403 mfu:25.15529747704349 +Step 31761 | loss:0.2336197942495346 lr:3.832410123601815e-06 tokens_per_second_per_gpu:4689.052778056068 mfu:22.119061988991394 +Step 31762 | loss:0.26322853565216064 lr:3.830745383110101e-06 tokens_per_second_per_gpu:4961.910125051819 mfu:23.406176649034272 +Step 31763 | loss:0.3358864188194275 lr:3.829080989863243e-06 tokens_per_second_per_gpu:5374.804147575298 mfu:25.35386820026203 +Step 31764 | loss:0.31547167897224426 lr:3.827416943873757e-06 tokens_per_second_per_gpu:5312.108824193736 mfu:25.058123662946 +Step 31765 | loss:0.35628557205200195 lr:3.825753245154157e-06 tokens_per_second_per_gpu:3555.977740595659 mfu:16.774153714754465 +Step 31766 | loss:0.3046579957008362 lr:3.824089893716959e-06 tokens_per_second_per_gpu:5335.609178020581 mfu:25.168978841521042 +Step 31767 | loss:0.3123566508293152 lr:3.822426889574687e-06 tokens_per_second_per_gpu:4893.286404955548 mfu:23.082466852926185 +Step 31768 | loss:0.2832832932472229 lr:3.82076423273982e-06 tokens_per_second_per_gpu:4746.8617209494605 mfu:22.39175663584225 +Step 31769 | loss:0.1953238844871521 lr:3.819101923224877e-06 tokens_per_second_per_gpu:4651.252134956176 mfu:21.940749906034444 +Step 31770 | loss:0.31494656205177307 lr:3.817439961042363e-06 tokens_per_second_per_gpu:5141.092205581644 mfu:24.251409094508837 +Step 31771 | loss:0.27736371755599976 lr:3.815778346204785e-06 tokens_per_second_per_gpu:4261.750250766811 mfu:20.103403062439057 +Step 31772 | loss:0.3166866600513458 lr:3.814117078724622e-06 tokens_per_second_per_gpu:5310.883841368754 mfu:25.052345209956933 +Step 31773 | loss:0.26403796672821045 lr:3.812456158614375e-06 tokens_per_second_per_gpu:4309.028295860627 mfu:20.326421667612866 +Step 31774 | loss:0.30291256308555603 lr:3.810795585886551e-06 tokens_per_second_per_gpu:5058.264363140937 mfu:23.860696029828873 +Step 31775 | loss:0.3148079812526703 lr:3.809135360553623e-06 tokens_per_second_per_gpu:4027.3991734739766 mfu:18.997929046431906 +Step 31776 | loss:0.30054330825805664 lr:3.807475482628081e-06 tokens_per_second_per_gpu:5347.390145360133 mfu:25.224551674502006 +Step 31777 | loss:0.24481932818889618 lr:3.8058159521224046e-06 tokens_per_second_per_gpu:5313.342043087836 mfu:25.06394096687831 +Step 31778 | loss:0.3164041042327881 lr:3.8041567690490887e-06 tokens_per_second_per_gpu:3235.4682290351348 mfu:15.262255664161907 +Step 31779 | loss:0.3671680986881256 lr:3.8024979334205955e-06 tokens_per_second_per_gpu:3539.1610419385925 mfu:16.694826477964426 +Step 31780 | loss:0.3113507032394409 lr:3.80083944524941e-06 tokens_per_second_per_gpu:5347.633157386705 mfu:25.225698003693473 +Step 31781 | loss:0.39091652631759644 lr:3.7991813045480163e-06 tokens_per_second_per_gpu:4527.068686670324 mfu:21.354955392588973 +Step 31782 | loss:0.27629268169403076 lr:3.7975235113288765e-06 tokens_per_second_per_gpu:5331.62052395732 mfu:25.15016367977009 +Step 31783 | loss:0.24057801067829132 lr:3.7958660656044478e-06 tokens_per_second_per_gpu:3580.346206817784 mfu:16.889103927613295 +Step 31784 | loss:0.2564498782157898 lr:3.7942089673872085e-06 tokens_per_second_per_gpu:4689.401886693601 mfu:22.120708793998713 +Step 31785 | loss:0.22833271324634552 lr:3.7925522166896265e-06 tokens_per_second_per_gpu:4592.25885954321 mfu:21.6624685606208 +Step 31786 | loss:0.28592097759246826 lr:3.7908958135241422e-06 tokens_per_second_per_gpu:5139.223885298792 mfu:24.24259590896632 +Step 31787 | loss:0.35238897800445557 lr:3.7892397579032346e-06 tokens_per_second_per_gpu:3576.7068366619546 mfu:16.87193639764767 +Step 31788 | loss:0.3061293959617615 lr:3.7875840498393546e-06 tokens_per_second_per_gpu:4639.424353060207 mfu:21.884956240803476 +Step 31789 | loss:0.31709522008895874 lr:3.785928689344953e-06 tokens_per_second_per_gpu:5315.586601031003 mfu:25.0745289296573 +Step 31790 | loss:0.32032477855682373 lr:3.7842736764324706e-06 tokens_per_second_per_gpu:5348.783408232308 mfu:25.231123933185216 +Step 31791 | loss:0.23390701413154602 lr:3.7826190111143635e-06 tokens_per_second_per_gpu:4372.139683644291 mfu:20.624128851701517 +Step 31792 | loss:0.29106664657592773 lr:3.780964693403083e-06 tokens_per_second_per_gpu:4457.289651218251 mfu:21.02579577683044 +Step 31793 | loss:0.27037379145622253 lr:3.779310723311058e-06 tokens_per_second_per_gpu:3500.7675744588328 mfu:16.513717941261373 +Step 31794 | loss:0.30032864212989807 lr:3.7776571008507346e-06 tokens_per_second_per_gpu:4551.638938810104 mfu:21.470857464049615 +Step 31795 | loss:0.3095228672027588 lr:3.7760038260345577e-06 tokens_per_second_per_gpu:4580.280402639541 mfu:21.605964135670696 +Step 31796 | loss:0.26493674516677856 lr:3.774350898874951e-06 tokens_per_second_per_gpu:5218.981712866284 mfu:24.618827189690784 +Step 31797 | loss:0.2875499427318573 lr:3.772698319384349e-06 tokens_per_second_per_gpu:4472.7863297766935 mfu:21.098896253596617 +Step 31798 | loss:0.21750354766845703 lr:3.771046087575175e-06 tokens_per_second_per_gpu:5317.158023507844 mfu:25.08194159759288 +Step 31799 | loss:0.2704668939113617 lr:3.769394203459875e-06 tokens_per_second_per_gpu:4834.752737822576 mfu:22.806353558186522 +Step 31800 | loss:0.4099142849445343 lr:3.7677426670508497e-06 tokens_per_second_per_gpu:5353.970060988472 mfu:25.25559025917025 +Step 31801 | loss:0.38871586322784424 lr:3.766091478360528e-06 tokens_per_second_per_gpu:4897.083507105393 mfu:23.100378431619433 +Step 31802 | loss:0.33110156655311584 lr:3.764440637401334e-06 tokens_per_second_per_gpu:4650.727818696644 mfu:21.938276616780655 +Step 31803 | loss:0.3196045756340027 lr:3.762790144185702e-06 tokens_per_second_per_gpu:3354.0862775436863 mfu:15.82179723730273 +Step 31804 | loss:0.3198530375957489 lr:3.761139998726004e-06 tokens_per_second_per_gpu:4306.970452031885 mfu:20.316714467168325 +Step 31805 | loss:0.31564095616340637 lr:3.7594902010346768e-06 tokens_per_second_per_gpu:4901.9618203771615 mfu:23.123390267648432 +Step 31806 | loss:0.2328423112630844 lr:3.75784075112412e-06 tokens_per_second_per_gpu:5146.7713292065255 mfu:24.278198489605813 +Step 31807 | loss:0.3331131339073181 lr:3.7561916490067515e-06 tokens_per_second_per_gpu:4525.025864615356 mfu:21.34531905241431 +Step 31808 | loss:0.25910648703575134 lr:3.7545428946949623e-06 tokens_per_second_per_gpu:5352.099429526387 mfu:25.246766171400964 +Step 31809 | loss:0.3598801791667938 lr:3.752894488201153e-06 tokens_per_second_per_gpu:4735.3418676366555 mfu:22.337415522276096 +Step 31810 | loss:0.27022188901901245 lr:3.751246429537736e-06 tokens_per_second_per_gpu:5236.3334327110415 mfu:24.70067821270719 +Step 31811 | loss:0.4430927038192749 lr:3.7495987187170954e-06 tokens_per_second_per_gpu:5227.709984200324 mfu:24.659999934770095 +Step 31812 | loss:0.28474441170692444 lr:3.7479513557516166e-06 tokens_per_second_per_gpu:3870.543594070137 mfu:18.258014516062502 +Step 31813 | loss:0.3102629780769348 lr:3.746304340653695e-06 tokens_per_second_per_gpu:4002.001583475504 mfu:18.878124281133424 +Step 31814 | loss:0.30466946959495544 lr:3.7446576734357307e-06 tokens_per_second_per_gpu:5359.872803403084 mfu:25.283434502251684 +Step 31815 | loss:0.3302489221096039 lr:3.7430113541100875e-06 tokens_per_second_per_gpu:2987.804704675778 mfu:14.09398456400429 +Step 31816 | loss:0.2978737950325012 lr:3.7413653826891603e-06 tokens_per_second_per_gpu:3536.4504738901005 mfu:16.682040266009093 +Step 31817 | loss:0.3252454400062561 lr:3.7397197591853394e-06 tokens_per_second_per_gpu:4114.496951838684 mfu:19.40878412739158 +Step 31818 | loss:0.2791493535041809 lr:3.7380744836109815e-06 tokens_per_second_per_gpu:5235.045786004715 mfu:24.694604163498266 +Step 31819 | loss:0.3515511155128479 lr:3.736429555978466e-06 tokens_per_second_per_gpu:5311.7079352698665 mfu:25.056232601493402 +Step 31820 | loss:0.34470948576927185 lr:3.734784976300165e-06 tokens_per_second_per_gpu:4631.920966210665 mfu:21.849561484824378 +Step 31821 | loss:0.23576821386814117 lr:3.7331407445884536e-06 tokens_per_second_per_gpu:3857.721077887488 mfu:18.197528519482095 +Step 31822 | loss:0.3164116442203522 lr:3.7314968608556932e-06 tokens_per_second_per_gpu:3928.293513962381 mfu:18.530430741346578 +Step 31823 | loss:0.23333412408828735 lr:3.7298533251142466e-06 tokens_per_second_per_gpu:5032.81025829045 mfu:23.74062467433101 +Step 31824 | loss:0.22638800740242004 lr:3.728210137376481e-06 tokens_per_second_per_gpu:5051.435265286548 mfu:23.828482010085477 +Step 31825 | loss:0.3053469657897949 lr:3.7265672976547593e-06 tokens_per_second_per_gpu:3979.6157279649897 mfu:18.772526381269532 +Step 31826 | loss:0.2437548190355301 lr:3.724924805961416e-06 tokens_per_second_per_gpu:5325.973325073859 mfu:25.123524879125313 +Step 31827 | loss:0.34210866689682007 lr:3.723282662308819e-06 tokens_per_second_per_gpu:4236.249480268872 mfu:19.983111577120177 +Step 31828 | loss:0.3087567389011383 lr:3.721640866709325e-06 tokens_per_second_per_gpu:4691.658438935147 mfu:22.131353335929603 +Step 31829 | loss:0.421123743057251 lr:3.7199994191752686e-06 tokens_per_second_per_gpu:5344.47847491298 mfu:25.210816828220587 +Step 31830 | loss:0.25042200088500977 lr:3.718358319719001e-06 tokens_per_second_per_gpu:5354.3604910955455 mfu:25.257431984599563 +Step 31831 | loss:0.48951655626296997 lr:3.716717568352873e-06 tokens_per_second_per_gpu:5349.82670029592 mfu:25.236045319853254 +Step 31832 | loss:0.4805370271205902 lr:3.7150771650892204e-06 tokens_per_second_per_gpu:4456.703417198161 mfu:21.023030410935842 +Step 31833 | loss:0.23214854300022125 lr:3.713437109940371e-06 tokens_per_second_per_gpu:3669.8119885791825 mfu:17.311129284618197 +Step 31834 | loss:0.321420818567276 lr:3.711797402918665e-06 tokens_per_second_per_gpu:5296.320982079485 mfu:24.98364971800945 +Step 31835 | loss:0.2720497250556946 lr:3.7101580440364435e-06 tokens_per_second_per_gpu:3541.931070979153 mfu:16.70789317756444 +Step 31836 | loss:0.38425660133361816 lr:3.7085190333060237e-06 tokens_per_second_per_gpu:5364.7978781538795 mfu:25.30666692015559 +Step 31837 | loss:0.29800277948379517 lr:3.7068803707397403e-06 tokens_per_second_per_gpu:3750.418314549303 mfu:17.691363077076616 +Step 31838 | loss:0.251700222492218 lr:3.7052420563499167e-06 tokens_per_second_per_gpu:4530.303710759658 mfu:21.370215553170393 +Step 31839 | loss:0.28934159874916077 lr:3.7036040901488934e-06 tokens_per_second_per_gpu:5354.85210669026 mfu:25.259751019238102 +Step 31840 | loss:0.28126293420791626 lr:3.701966472148949e-06 tokens_per_second_per_gpu:4501.557780899225 mfu:21.23461609745735 +Step 31841 | loss:0.30895161628723145 lr:3.7003292023624236e-06 tokens_per_second_per_gpu:4080.9203521220684 mfu:19.250397577769657 +Step 31842 | loss:0.356107622385025 lr:3.6986922808016356e-06 tokens_per_second_per_gpu:3850.525350105176 mfu:18.16358504381517 +Step 31843 | loss:0.2302067130804062 lr:3.6970557074788913e-06 tokens_per_second_per_gpu:4140.802786691218 mfu:19.532873238629218 +Step 31844 | loss:0.30307212471961975 lr:3.695419482406498e-06 tokens_per_second_per_gpu:5356.329837185246 mfu:25.266721726109772 +Step 31845 | loss:0.2978225648403168 lr:3.6937836055967625e-06 tokens_per_second_per_gpu:5095.02133819663 mfu:24.0340849525532 +Step 31846 | loss:0.30636587738990784 lr:3.692148077062002e-06 tokens_per_second_per_gpu:4044.8422343114726 mfu:19.080210940495437 +Step 31847 | loss:0.2693970799446106 lr:3.690512896814491e-06 tokens_per_second_per_gpu:4438.440244167145 mfu:20.9368799077299 +Step 31848 | loss:0.22077606618404388 lr:3.688878064866541e-06 tokens_per_second_per_gpu:4227.899611747148 mfu:19.943723822668737 +Step 31849 | loss:0.33390864729881287 lr:3.6872435812304485e-06 tokens_per_second_per_gpu:4003.65370608841 mfu:18.885917625379683 +Step 31850 | loss:0.34721794724464417 lr:3.6856094459185142e-06 tokens_per_second_per_gpu:4047.2902009077416 mfu:19.09175841659621 +Step 31851 | loss:0.26693323254585266 lr:3.6839756589430117e-06 tokens_per_second_per_gpu:3778.2702324233173 mfu:17.822745325715545 +Step 31852 | loss:0.29774022102355957 lr:3.6823422203162374e-06 tokens_per_second_per_gpu:5367.177775797718 mfu:25.317893303393983 +Step 31853 | loss:0.33565545082092285 lr:3.6807091300504913e-06 tokens_per_second_per_gpu:4854.07694877882 mfu:22.8975091583175 +Step 31854 | loss:0.33740806579589844 lr:3.679076388158037e-06 tokens_per_second_per_gpu:3501.218545725285 mfu:16.51584524966886 +Step 31855 | loss:0.32106027007102966 lr:3.677443994651153e-06 tokens_per_second_per_gpu:4909.650538442065 mfu:23.15965926259116 +Step 31856 | loss:0.323693186044693 lr:3.675811949542124e-06 tokens_per_second_per_gpu:3771.546760095034 mfu:17.791029559600364 +Step 31857 | loss:0.30926692485809326 lr:3.6741802528432347e-06 tokens_per_second_per_gpu:5367.818488158758 mfu:25.320915652917808 +Step 31858 | loss:0.3000624179840088 lr:3.672548904566736e-06 tokens_per_second_per_gpu:5336.915096533899 mfu:25.17513908196104 +Step 31859 | loss:0.2774079144001007 lr:3.6709179047249078e-06 tokens_per_second_per_gpu:5181.784665503714 mfu:24.44336237081083 +Step 31860 | loss:0.33971014618873596 lr:3.669287253330028e-06 tokens_per_second_per_gpu:5371.069415539484 mfu:25.336250832038104 +Step 31861 | loss:0.33512696623802185 lr:3.6676569503943493e-06 tokens_per_second_per_gpu:5321.135994701241 mfu:25.100706366424006 +Step 31862 | loss:0.4399254322052002 lr:3.6660269959301276e-06 tokens_per_second_per_gpu:4292.348147294 mfu:20.24773856089728 +Step 31863 | loss:0.21717648208141327 lr:3.6643973899496253e-06 tokens_per_second_per_gpu:3885.629337700015 mfu:18.329176542657912 +Step 31864 | loss:0.22880935668945312 lr:3.6627681324651163e-06 tokens_per_second_per_gpu:5333.435390014927 mfu:25.1587247126119 +Step 31865 | loss:0.3690699338912964 lr:3.6611392234888296e-06 tokens_per_second_per_gpu:5351.632374339056 mfu:25.24456299239449 +Step 31866 | loss:0.35939154028892517 lr:3.6595106630330276e-06 tokens_per_second_per_gpu:4970.31791437393 mfu:23.44583762578345 +Step 31867 | loss:0.256686806678772 lr:3.6578824511099675e-06 tokens_per_second_per_gpu:3883.297248586385 mfu:18.31817568041349 +Step 31868 | loss:0.2772074043750763 lr:3.6562545877318887e-06 tokens_per_second_per_gpu:4616.4951534016545 mfu:21.776795293889215 +Step 31869 | loss:0.342989981174469 lr:3.654627072911021e-06 tokens_per_second_per_gpu:3462.0948116802297 mfu:16.33129192098098 +Step 31870 | loss:0.3149731159210205 lr:3.652999906659621e-06 tokens_per_second_per_gpu:5129.477264954498 mfu:24.196619437857375 +Step 31871 | loss:0.2595363259315491 lr:3.6513730889899233e-06 tokens_per_second_per_gpu:5055.136178975294 mfu:23.845939851396277 +Step 31872 | loss:0.2871026396751404 lr:3.6497466199141626e-06 tokens_per_second_per_gpu:4849.648444215661 mfu:22.876619146712283 +Step 31873 | loss:0.3155078887939453 lr:3.6481204994445684e-06 tokens_per_second_per_gpu:3298.250188328725 mfu:15.558408877856584 +Step 31874 | loss:0.29180189967155457 lr:3.6464947275933747e-06 tokens_per_second_per_gpu:5158.415247723076 mfu:24.33312483990574 +Step 31875 | loss:0.27701976895332336 lr:3.6448693043728276e-06 tokens_per_second_per_gpu:4367.8129799063345 mfu:20.603719051957796 +Step 31876 | loss:0.22870127856731415 lr:3.643244229795112e-06 tokens_per_second_per_gpu:3818.63029296023 mfu:18.013130617404563 +Step 31877 | loss:0.2548869252204895 lr:3.6416195038724788e-06 tokens_per_second_per_gpu:5365.729145860516 mfu:25.311059868818447 +Step 31878 | loss:0.3972630798816681 lr:3.6399951266171408e-06 tokens_per_second_per_gpu:4915.857076372907 mfu:23.188936560955504 +Step 31879 | loss:0.21097534894943237 lr:3.63837109804131e-06 tokens_per_second_per_gpu:3473.2173793495995 mfu:16.383758970383127 +Step 31880 | loss:0.38665440678596497 lr:3.6367474181572103e-06 tokens_per_second_per_gpu:4840.545428498478 mfu:22.833678668439696 +Step 31881 | loss:0.334861695766449 lr:3.6351240869770432e-06 tokens_per_second_per_gpu:5350.638706127974 mfu:25.23987568990553 +Step 31882 | loss:0.22481416165828705 lr:3.633501104513043e-06 tokens_per_second_per_gpu:4370.09896107834 mfu:20.61450241517475 +Step 31883 | loss:0.2717467248439789 lr:3.6318784707773835e-06 tokens_per_second_per_gpu:5171.468774250642 mfu:24.39470055169755 +Step 31884 | loss:0.24668844044208527 lr:3.630256185782277e-06 tokens_per_second_per_gpu:5268.9203684321665 mfu:24.854396348408287 +Step 31885 | loss:0.2989608943462372 lr:3.628634249539936e-06 tokens_per_second_per_gpu:3712.6673837148637 mfu:17.513285495357593 +Step 31886 | loss:0.3023397922515869 lr:3.627012662062562e-06 tokens_per_second_per_gpu:4280.900978581144 mfu:20.193740312990357 +Step 31887 | loss:0.2888854146003723 lr:3.625391423362329e-06 tokens_per_second_per_gpu:5346.32000834338 mfu:25.219503655609785 +Step 31888 | loss:0.3080923855304718 lr:3.6237705334514483e-06 tokens_per_second_per_gpu:4125.4005725970865 mfu:19.460218366858495 +Step 31889 | loss:0.23297423124313354 lr:3.6221499923421167e-06 tokens_per_second_per_gpu:2898.1888928120434 mfu:13.671251489408892 +Step 31890 | loss:0.3102573752403259 lr:3.6205298000465127e-06 tokens_per_second_per_gpu:5356.3988329408485 mfu:25.267047190860836 +Step 31891 | loss:0.27195748686790466 lr:3.6189099565768103e-06 tokens_per_second_per_gpu:3827.040041872973 mfu:18.052800837877097 +Step 31892 | loss:0.2354302853345871 lr:3.6172904619452054e-06 tokens_per_second_per_gpu:5407.139805121341 mfu:25.506401013938792 +Step 31893 | loss:0.3080003261566162 lr:3.615671316163888e-06 tokens_per_second_per_gpu:3811.10931886873 mfu:17.97765289940446 +Step 31894 | loss:0.35551348328590393 lr:3.614052519245015e-06 tokens_per_second_per_gpu:5321.836616859851 mfu:25.104011320684638 +Step 31895 | loss:0.20938624441623688 lr:3.612434071200771e-06 tokens_per_second_per_gpu:3414.276903912944 mfu:16.105726691466344 +Step 31896 | loss:0.400600403547287 lr:3.610815972043341e-06 tokens_per_second_per_gpu:3456.586031722014 mfu:16.3053060660233 +Step 31897 | loss:0.38702279329299927 lr:3.609198221784882e-06 tokens_per_second_per_gpu:4241.707373501499 mfu:20.008857390711093 +Step 31898 | loss:0.3937152326107025 lr:3.607580820437556e-06 tokens_per_second_per_gpu:5339.444724659437 mfu:25.187071769428268 +Step 31899 | loss:0.33180180191993713 lr:3.6059637680135315e-06 tokens_per_second_per_gpu:5085.390463561921 mfu:23.988654473704628 +Step 31900 | loss:0.26863691210746765 lr:3.604347064524982e-06 tokens_per_second_per_gpu:4827.433884156149 mfu:22.771829276716794 +Step 31901 | loss:0.30954810976982117 lr:3.6027307099840534e-06 tokens_per_second_per_gpu:5322.030007018891 mfu:25.104923575060596 +Step 31902 | loss:0.270658940076828 lr:3.6011147044029023e-06 tokens_per_second_per_gpu:4876.687021581946 mfu:23.004164729406053 +Step 31903 | loss:0.41287800669670105 lr:3.5994990477936974e-06 tokens_per_second_per_gpu:3885.404517325462 mfu:18.328116026592838 +Step 31904 | loss:0.28739041090011597 lr:3.597883740168584e-06 tokens_per_second_per_gpu:5078.119017219614 mfu:23.954353820671287 +Step 31905 | loss:0.4595845937728882 lr:3.5962687815396967e-06 tokens_per_second_per_gpu:4009.3211585372396 mfu:18.912651965547916 +Step 31906 | loss:0.29582154750823975 lr:3.5946541719191984e-06 tokens_per_second_per_gpu:3741.3743067737673 mfu:17.64870094933257 +Step 31907 | loss:0.39927637577056885 lr:3.593039911319229e-06 tokens_per_second_per_gpu:4805.154927930263 mfu:22.666735639017386 +Step 31908 | loss:0.33563125133514404 lr:3.591425999751924e-06 tokens_per_second_per_gpu:5392.374599583829 mfu:25.43675102021451 +Step 31909 | loss:0.3299010097980499 lr:3.5898124372294227e-06 tokens_per_second_per_gpu:4705.439361459809 mfu:22.19636029874627 +Step 31910 | loss:0.3482714295387268 lr:3.5881992237638662e-06 tokens_per_second_per_gpu:4862.944590455233 mfu:22.93933933708082 +Step 31911 | loss:0.3016829490661621 lr:3.5865863593674e-06 tokens_per_second_per_gpu:3917.8791356608986 mfu:18.481304342022323 +Step 31912 | loss:0.34304994344711304 lr:3.584973844052125e-06 tokens_per_second_per_gpu:4207.051745871664 mfu:19.845380882321372 +Step 31913 | loss:0.30709290504455566 lr:3.583361677830188e-06 tokens_per_second_per_gpu:5362.91652204982 mfu:25.297792242420286 +Step 31914 | loss:0.2223237007856369 lr:3.5817498607137125e-06 tokens_per_second_per_gpu:4244.751299022103 mfu:20.023216106739028 +Step 31915 | loss:0.3682088553905487 lr:3.580138392714816e-06 tokens_per_second_per_gpu:4998.347923368757 mfu:23.57805996062473 +Step 31916 | loss:0.3037738800048828 lr:3.5785272738456166e-06 tokens_per_second_per_gpu:4421.9582682948985 mfu:20.859131615425454 +Step 31917 | loss:0.28060320019721985 lr:3.576916504118244e-06 tokens_per_second_per_gpu:5336.372587956661 mfu:25.1725799764407 +Step 31918 | loss:0.3647652268409729 lr:3.575306083544816e-06 tokens_per_second_per_gpu:5188.596713702651 mfu:24.47549596442049 +Step 31919 | loss:0.3303891122341156 lr:3.573696012137423e-06 tokens_per_second_per_gpu:4001.2248432586457 mfu:18.874460265004956 +Step 31920 | loss:0.3909126818180084 lr:3.572086289908183e-06 tokens_per_second_per_gpu:5331.156504343925 mfu:25.14797482008379 +Step 31921 | loss:0.24953511357307434 lr:3.5704769168692088e-06 tokens_per_second_per_gpu:3979.4546135574474 mfu:18.7717663771202 +Step 31922 | loss:0.2909495532512665 lr:3.568867893032607e-06 tokens_per_second_per_gpu:4098.704166717967 mfu:19.334286865450164 +Step 31923 | loss:0.36280563473701477 lr:3.567259218410468e-06 tokens_per_second_per_gpu:5316.053578480862 mfu:25.076731741962856 +Step 31924 | loss:0.2863762676715851 lr:3.565650893014899e-06 tokens_per_second_per_gpu:4152.164953802736 mfu:19.586470519479857 +Step 31925 | loss:0.27361324429512024 lr:3.5640429168580015e-06 tokens_per_second_per_gpu:4144.54545824074 mfu:19.550528058893924 +Step 31926 | loss:0.28997230529785156 lr:3.56243528995186e-06 tokens_per_second_per_gpu:3929.8777850422807 mfu:18.537904018334984 +Step 31927 | loss:0.36817827820777893 lr:3.5608280123085646e-06 tokens_per_second_per_gpu:4385.223687656993 mfu:20.68584833099071 +Step 31928 | loss:0.2802758514881134 lr:3.5592210839402062e-06 tokens_per_second_per_gpu:5388.842150355105 mfu:25.42008785450463 +Step 31929 | loss:0.19377903640270233 lr:3.557614504858875e-06 tokens_per_second_per_gpu:3763.896047183166 mfu:17.754939841448856 +Step 31930 | loss:0.2892819941043854 lr:3.55600827507665e-06 tokens_per_second_per_gpu:4335.726350372779 mfu:20.4523609459045 +Step 31931 | loss:0.34433507919311523 lr:3.5544023946056106e-06 tokens_per_second_per_gpu:4078.266191841516 mfu:19.237877450879854 +Step 31932 | loss:0.28335219621658325 lr:3.552796863457841e-06 tokens_per_second_per_gpu:4345.152856283231 mfu:20.496827382611855 +Step 31933 | loss:0.3247629404067993 lr:3.5511916816454215e-06 tokens_per_second_per_gpu:5398.276038646492 mfu:25.464589115162422 +Step 31934 | loss:0.4097937345504761 lr:3.549586849180403e-06 tokens_per_second_per_gpu:3199.0088563212844 mfu:15.09027058246024 +Step 31935 | loss:0.41345712542533875 lr:3.54798236607487e-06 tokens_per_second_per_gpu:3392.8017186364727 mfu:16.004424578472797 +Step 31936 | loss:0.35938140749931335 lr:3.5463782323408967e-06 tokens_per_second_per_gpu:5337.606187662428 mfu:25.178399076726617 +Step 31937 | loss:0.3054497539997101 lr:3.544774447990529e-06 tokens_per_second_per_gpu:4760.133558868005 mfu:22.454362159713067 +Step 31938 | loss:0.2836267948150635 lr:3.543171013035845e-06 tokens_per_second_per_gpu:5369.657568869628 mfu:25.329590910410513 +Step 31939 | loss:0.3410959541797638 lr:3.5415679274889035e-06 tokens_per_second_per_gpu:5353.398859971687 mfu:25.252895806517138 +Step 31940 | loss:0.33458411693573 lr:3.5399651913617606e-06 tokens_per_second_per_gpu:4575.784606329114 mfu:21.584756697412466 +Step 31941 | loss:0.31088972091674805 lr:3.5383628046664617e-06 tokens_per_second_per_gpu:3734.436864907483 mfu:17.615975852399675 +Step 31942 | loss:0.3685794770717621 lr:3.536760767415065e-06 tokens_per_second_per_gpu:4596.014808552961 mfu:21.680186012930776 +Step 31943 | loss:0.32785797119140625 lr:3.535159079619621e-06 tokens_per_second_per_gpu:5376.4203733206305 mfu:25.361492212859307 +Step 31944 | loss:0.3905965983867645 lr:3.5335577412921707e-06 tokens_per_second_per_gpu:4334.432827963801 mfu:20.44625918000367 +Step 31945 | loss:0.32025963068008423 lr:3.5319567524447648e-06 tokens_per_second_per_gpu:4581.224335978958 mfu:21.610416830284365 +Step 31946 | loss:0.25059807300567627 lr:3.5303561130894446e-06 tokens_per_second_per_gpu:5146.459641585255 mfu:24.276728205916886 +Step 31947 | loss:0.36762163043022156 lr:3.5287558232382445e-06 tokens_per_second_per_gpu:5211.2129017726975 mfu:24.58218038226648 +Step 31948 | loss:0.33233729004859924 lr:3.5271558829031994e-06 tokens_per_second_per_gpu:3935.1471337706607 mfu:18.562760435330354 +Step 31949 | loss:0.31954261660575867 lr:3.5255562920963437e-06 tokens_per_second_per_gpu:4742.629595839094 mfu:22.371792979621645 +Step 31950 | loss:0.3570877015590668 lr:3.5239570508297127e-06 tokens_per_second_per_gpu:5310.696546088005 mfu:25.05146170616177 +Step 31951 | loss:0.2846367657184601 lr:3.5223581591153243e-06 tokens_per_second_per_gpu:4309.553732028847 mfu:20.328900239667014 +Step 31952 | loss:0.3178759515285492 lr:3.5207596169652134e-06 tokens_per_second_per_gpu:5356.128052822516 mfu:25.265769874842814 +Step 31953 | loss:0.3784441351890564 lr:3.519161424391398e-06 tokens_per_second_per_gpu:4637.1226723907 mfu:21.874098820378197 +Step 31954 | loss:0.24327921867370605 lr:3.5175635814059138e-06 tokens_per_second_per_gpu:5377.783379438458 mfu:25.367921745269925 +Step 31955 | loss:0.2818208336830139 lr:3.5159660880207613e-06 tokens_per_second_per_gpu:4789.044954227665 mfu:22.590742144417288 +Step 31956 | loss:0.3310997486114502 lr:3.514368944247948e-06 tokens_per_second_per_gpu:4117.189555064846 mfu:19.42148559621487 +Step 31957 | loss:0.2509709894657135 lr:3.5127721500995027e-06 tokens_per_second_per_gpu:5329.696480193391 mfu:25.141087637059915 +Step 31958 | loss:0.3370858430862427 lr:3.511175705587433e-06 tokens_per_second_per_gpu:5336.848015198508 mfu:25.174822647856743 +Step 31959 | loss:0.30164724588394165 lr:3.50957961072374e-06 tokens_per_second_per_gpu:4408.176229375101 mfu:20.79411938638246 +Step 31960 | loss:0.29424986243247986 lr:3.5079838655204254e-06 tokens_per_second_per_gpu:5040.285562054685 mfu:23.775886957605245 +Step 31961 | loss:0.3113541901111603 lr:3.5063884699895077e-06 tokens_per_second_per_gpu:4413.991397058072 mfu:20.821550524512947 +Step 31962 | loss:0.20616301894187927 lr:3.504793424142977e-06 tokens_per_second_per_gpu:5323.134410123234 mfu:25.11013323293438 +Step 31963 | loss:0.3097669184207916 lr:3.503198727992818e-06 tokens_per_second_per_gpu:2990.7347393682035 mfu:14.107806037563929 +Step 31964 | loss:0.3718417286872864 lr:3.5016043815510324e-06 tokens_per_second_per_gpu:4073.043407443936 mfu:19.21324068577756 +Step 31965 | loss:0.3476632833480835 lr:3.5000103848296217e-06 tokens_per_second_per_gpu:4137.903927064603 mfu:19.51919882317402 +Step 31966 | loss:0.3595801293849945 lr:3.49841673784056e-06 tokens_per_second_per_gpu:4052.4037853719337 mfu:19.115880066981276 +Step 31967 | loss:0.35205137729644775 lr:3.4968234405958424e-06 tokens_per_second_per_gpu:4817.56486080422 mfu:22.725275409736728 +Step 31968 | loss:0.27534809708595276 lr:3.495230493107454e-06 tokens_per_second_per_gpu:5327.58564605657 mfu:25.13113047229109 +Step 31969 | loss:0.3089330494403839 lr:3.4936378953873696e-06 tokens_per_second_per_gpu:4386.303736807656 mfu:20.69094310710977 +Step 31970 | loss:0.30420252680778503 lr:3.4920456474475615e-06 tokens_per_second_per_gpu:3572.9091892714505 mfu:16.854022246961204 +Step 31971 | loss:0.2224767506122589 lr:3.4904537493000154e-06 tokens_per_second_per_gpu:5356.449020943274 mfu:25.26728393623124 +Step 31972 | loss:0.3819501996040344 lr:3.488862200956711e-06 tokens_per_second_per_gpu:5340.977791356766 mfu:25.194303506607717 +Step 31973 | loss:0.216729536652565 lr:3.4872710024295928e-06 tokens_per_second_per_gpu:3844.262217367384 mfu:18.134040778090373 +Step 31974 | loss:0.3038245439529419 lr:3.4856801537306526e-06 tokens_per_second_per_gpu:4046.4697927875095 mfu:19.087888411517014 +Step 31975 | loss:0.2801317572593689 lr:3.484089654871853e-06 tokens_per_second_per_gpu:4452.996567841387 mfu:21.00554457006622 +Step 31976 | loss:0.34472325444221497 lr:3.4824995058651445e-06 tokens_per_second_per_gpu:5194.196633896638 mfu:24.50191174342043 +Step 31977 | loss:0.3102836608886719 lr:3.4809097067224906e-06 tokens_per_second_per_gpu:5327.588070992393 mfu:25.131141911127102 +Step 31978 | loss:0.3941289782524109 lr:3.4793202574558536e-06 tokens_per_second_per_gpu:5335.813797002617 mfu:25.16994406417091 +Step 31979 | loss:0.26225748658180237 lr:3.4777311580771854e-06 tokens_per_second_per_gpu:3609.1373302255947 mfu:17.02491656900013 +Step 31980 | loss:0.3404485881328583 lr:3.4761424085984374e-06 tokens_per_second_per_gpu:4828.350523342401 mfu:22.776153220153112 +Step 31981 | loss:0.26032695174217224 lr:3.4745540090315554e-06 tokens_per_second_per_gpu:4585.216030051577 mfu:21.629246332278147 +Step 31982 | loss:0.27339300513267517 lr:3.4729659593884966e-06 tokens_per_second_per_gpu:4786.825358980074 mfu:22.5802719349324 +Step 31983 | loss:0.3323149085044861 lr:3.471378259681185e-06 tokens_per_second_per_gpu:3975.2224116442026 mfu:18.75180235860732 +Step 31984 | loss:0.41626325249671936 lr:3.4697909099215885e-06 tokens_per_second_per_gpu:4972.519836756991 mfu:23.45622446935105 +Step 31985 | loss:0.28975674510002136 lr:3.4682039101216256e-06 tokens_per_second_per_gpu:4911.040327905719 mfu:23.166215136613747 +Step 31986 | loss:0.3424176573753357 lr:3.4666172602932423e-06 tokens_per_second_per_gpu:4644.059742788432 mfu:21.90682216503055 +Step 31987 | loss:0.31194671988487244 lr:3.465030960448362e-06 tokens_per_second_per_gpu:4337.370803775235 mfu:20.460118113176566 +Step 31988 | loss:0.277627170085907 lr:3.4634450105989202e-06 tokens_per_second_per_gpu:5308.193919294928 mfu:25.039656388586668 +Step 31989 | loss:0.3028416633605957 lr:3.4618594107568512e-06 tokens_per_second_per_gpu:3939.4963274150805 mfu:18.583276323800032 +Step 31990 | loss:0.42803752422332764 lr:3.4602741609340796e-06 tokens_per_second_per_gpu:4684.477475656058 mfu:22.09747950694346 +Step 31991 | loss:0.39001771807670593 lr:3.4586892611425225e-06 tokens_per_second_per_gpu:4780.796124132162 mfu:22.551831005461256 +Step 31992 | loss:0.2748536765575409 lr:3.457104711394099e-06 tokens_per_second_per_gpu:5324.493339824489 mfu:25.116543536192516 +Step 31993 | loss:0.31400930881500244 lr:3.4555205117007272e-06 tokens_per_second_per_gpu:3692.1488900420577 mfu:17.416496259886955 +Step 31994 | loss:0.27194827795028687 lr:3.4539366620743363e-06 tokens_per_second_per_gpu:5336.833791134802 mfu:25.17475555052137 +Step 31995 | loss:0.30685365200042725 lr:3.4523531625268167e-06 tokens_per_second_per_gpu:4680.064459643067 mfu:22.07666255747201 +Step 31996 | loss:0.224201500415802 lr:3.450770013070087e-06 tokens_per_second_per_gpu:4533.797208230393 mfu:21.38669497679198 +Step 31997 | loss:0.3303019404411316 lr:3.449187213716065e-06 tokens_per_second_per_gpu:5354.670249201064 mfu:25.258893166433477 +Step 31998 | loss:0.36907529830932617 lr:3.447604764476647e-06 tokens_per_second_per_gpu:4439.292392846781 mfu:20.940899638443305 +Step 31999 | loss:0.35309019684791565 lr:3.446022665363724e-06 tokens_per_second_per_gpu:5363.507521070684 mfu:25.300580085636533 +Step 32000 | loss:0.3633328974246979 lr:3.4444409163892076e-06 tokens_per_second_per_gpu:5207.780584141609 mfu:24.565989554386864 +Step 32001 | loss:0.30725494027137756 lr:3.442859517564995e-06 tokens_per_second_per_gpu:5366.164598660404 mfu:25.31311397397143 +Step 32002 | loss:0.3703565001487732 lr:3.44127846890297e-06 tokens_per_second_per_gpu:3783.2311922208655 mfu:17.8461470195076 +Step 32003 | loss:0.4325401186943054 lr:3.4396977704150347e-06 tokens_per_second_per_gpu:4164.78423251009 mfu:19.645997810213412 +Step 32004 | loss:0.2584281861782074 lr:3.43811742211308e-06 tokens_per_second_per_gpu:4036.8492451585394 mfu:19.04250664691648 +Step 32005 | loss:0.3606179654598236 lr:3.436537424008979e-06 tokens_per_second_per_gpu:3223.0130999571743 mfu:15.203502695237017 +Step 32006 | loss:0.2930299639701843 lr:3.4349577761146225e-06 tokens_per_second_per_gpu:5361.153050224645 mfu:25.289473644941623 +Step 32007 | loss:0.3200033903121948 lr:3.4333784784418845e-06 tokens_per_second_per_gpu:5345.008277055499 mfu:25.213315995320414 +Step 32008 | loss:0.2694042921066284 lr:3.4317995310026605e-06 tokens_per_second_per_gpu:2995.4724633212613 mfu:14.130154689789563 +Step 32009 | loss:0.3076825439929962 lr:3.430220933808803e-06 tokens_per_second_per_gpu:5370.188307952343 mfu:25.332094497217103 +Step 32010 | loss:0.2760913372039795 lr:3.4286426868721967e-06 tokens_per_second_per_gpu:4506.714367355999 mfu:21.258940595578327 +Step 32011 | loss:0.24110747873783112 lr:3.4270647902047204e-06 tokens_per_second_per_gpu:5019.557933004921 mfu:23.67811119487932 +Step 32012 | loss:0.27159932255744934 lr:3.4254872438182373e-06 tokens_per_second_per_gpu:3862.6676810229224 mfu:18.220862490449463 +Step 32013 | loss:0.3046427071094513 lr:3.423910047724593e-06 tokens_per_second_per_gpu:4549.700711686313 mfu:21.461714516011046 +Step 32014 | loss:0.3637310862541199 lr:3.422333201935668e-06 tokens_per_second_per_gpu:5344.613279746014 mfu:25.211452725618077 +Step 32015 | loss:0.4036303758621216 lr:3.420756706463324e-06 tokens_per_second_per_gpu:5337.384512305915 mfu:25.17735339624823 +Step 32016 | loss:0.2735273838043213 lr:3.4191805613194016e-06 tokens_per_second_per_gpu:4411.886155271825 mfu:20.811619739815935 +Step 32017 | loss:0.37010329961776733 lr:3.4176047665157696e-06 tokens_per_second_per_gpu:4047.501179242841 mfu:19.092753637399486 +Step 32018 | loss:0.3390924334526062 lr:3.4160293220642846e-06 tokens_per_second_per_gpu:4116.311850712577 mfu:19.417345315033412 +Step 32019 | loss:0.27030935883522034 lr:3.414454227976782e-06 tokens_per_second_per_gpu:5359.459200122389 mfu:25.28148346500865 +Step 32020 | loss:0.250427782535553 lr:3.4128794842651134e-06 tokens_per_second_per_gpu:5369.3182233375765 mfu:25.327990159637505 +Step 32021 | loss:0.3040207624435425 lr:3.4113050909411246e-06 tokens_per_second_per_gpu:4903.7218029743335 mfu:23.13169240584329 +Step 32022 | loss:0.3443315029144287 lr:3.4097310480166566e-06 tokens_per_second_per_gpu:4198.841379688785 mfu:19.806651184203815 +Step 32023 | loss:0.2835744321346283 lr:3.4081573555035385e-06 tokens_per_second_per_gpu:4449.1362442185555 mfu:20.987334764898197 +Step 32024 | loss:0.27835899591445923 lr:3.4065840134136164e-06 tokens_per_second_per_gpu:5326.770440723098 mfu:25.127285009644726 +Step 32025 | loss:0.3406510055065155 lr:3.4050110217587192e-06 tokens_per_second_per_gpu:5312.150890371588 mfu:25.058322096284158 +Step 32026 | loss:0.3540867269039154 lr:3.4034383805506885e-06 tokens_per_second_per_gpu:5106.872615651547 mfu:24.089989450344486 +Step 32027 | loss:0.3486296832561493 lr:3.401866089801342e-06 tokens_per_second_per_gpu:4535.395157334259 mfu:21.39423277535315 +Step 32028 | loss:0.2198607176542282 lr:3.400294149522504e-06 tokens_per_second_per_gpu:4837.645607064853 mfu:22.81999971597673 +Step 32029 | loss:0.26037725806236267 lr:3.398722559725992e-06 tokens_per_second_per_gpu:2833.754817648265 mfu:13.367304963274503 +Step 32030 | loss:0.3553233742713928 lr:3.397151320423647e-06 tokens_per_second_per_gpu:5347.804781550215 mfu:25.226507584154856 +Step 32031 | loss:0.37700772285461426 lr:3.3955804316272656e-06 tokens_per_second_per_gpu:5343.783841967183 mfu:25.207540126098657 +Step 32032 | loss:0.2935400605201721 lr:3.3940098933486708e-06 tokens_per_second_per_gpu:5012.172524904984 mfu:23.64327296479136 +Step 32033 | loss:0.3056824207305908 lr:3.392439705599676e-06 tokens_per_second_per_gpu:5349.093006051181 mfu:25.232584358919677 +Step 32034 | loss:0.3060832917690277 lr:3.390869868392099e-06 tokens_per_second_per_gpu:3654.7135976719915 mfu:17.239907598657858 +Step 32035 | loss:0.35330307483673096 lr:3.38930038173772e-06 tokens_per_second_per_gpu:3927.2046211937504 mfu:18.525294248372685 +Step 32036 | loss:0.3137672543525696 lr:3.3877312456483676e-06 tokens_per_second_per_gpu:2790.1896544899446 mfu:13.161800655673407 +Step 32037 | loss:0.2653888165950775 lr:3.3861624601358387e-06 tokens_per_second_per_gpu:3699.1245092675595 mfu:17.449401445937013 +Step 32038 | loss:0.33809545636177063 lr:3.384594025211929e-06 tokens_per_second_per_gpu:5107.356554961052 mfu:24.092272274635985 +Step 32039 | loss:0.32015055418014526 lr:3.383025940888429e-06 tokens_per_second_per_gpu:5224.128217837963 mfu:24.64310413172649 +Step 32040 | loss:0.3599223494529724 lr:3.3814582071771517e-06 tokens_per_second_per_gpu:3603.475377700319 mfu:16.99820817290969 +Step 32041 | loss:0.3006458580493927 lr:3.379890824089871e-06 tokens_per_second_per_gpu:4191.161791482604 mfu:19.770425256362394 +Step 32042 | loss:0.3305412828922272 lr:3.378323791638377e-06 tokens_per_second_per_gpu:4688.890204558916 mfu:22.118295102067428 +Step 32043 | loss:0.32461345195770264 lr:3.376757109834455e-06 tokens_per_second_per_gpu:5374.726675007727 mfu:25.353502748942613 +Step 32044 | loss:0.3548113703727722 lr:3.3751907786899017e-06 tokens_per_second_per_gpu:3579.7157462504692 mfu:16.886129937549605 +Step 32045 | loss:0.27298852801322937 lr:3.373624798216485e-06 tokens_per_second_per_gpu:5384.940630624622 mfu:25.401683720268032 +Step 32046 | loss:0.4296765923500061 lr:3.3720591684259784e-06 tokens_per_second_per_gpu:3808.729200191962 mfu:17.966425473516026 +Step 32047 | loss:0.2887834310531616 lr:3.370493889330173e-06 tokens_per_second_per_gpu:5333.237963933546 mfu:25.157793420101658 +Step 32048 | loss:0.31515100598335266 lr:3.3689289609408313e-06 tokens_per_second_per_gpu:4746.463222434244 mfu:22.389876850355115 +Step 32049 | loss:0.2738797664642334 lr:3.3673643832697277e-06 tokens_per_second_per_gpu:5322.649568386412 mfu:25.107846151739263 +Step 32050 | loss:0.2528066337108612 lr:3.365800156328619e-06 tokens_per_second_per_gpu:4986.505239215867 mfu:23.52219599890528 +Step 32051 | loss:0.32674211263656616 lr:3.364236280129285e-06 tokens_per_second_per_gpu:5350.540259026288 mfu:25.239411298130076 +Step 32052 | loss:0.2716120183467865 lr:3.362672754683477e-06 tokens_per_second_per_gpu:4652.505489576545 mfu:21.946662193623073 +Step 32053 | loss:0.2831639051437378 lr:3.3611095800029522e-06 tokens_per_second_per_gpu:4508.182274677764 mfu:21.26586496486591 +Step 32054 | loss:0.4405965805053711 lr:3.3595467560994853e-06 tokens_per_second_per_gpu:5370.103814689858 mfu:25.331695928081924 +Step 32055 | loss:0.32715675234794617 lr:3.3579842829848107e-06 tokens_per_second_per_gpu:5276.8717517980995 mfu:24.89190438039129 +Step 32056 | loss:0.3698742687702179 lr:3.356422160670697e-06 tokens_per_second_per_gpu:4250.061776037923 mfu:20.048266532883456 +Step 32057 | loss:0.2150239646434784 lr:3.3548603891688734e-06 tokens_per_second_per_gpu:4743.792016740307 mfu:22.377276317342126 +Step 32058 | loss:0.28708019852638245 lr:3.3532989684911086e-06 tokens_per_second_per_gpu:5355.690853890066 mfu:25.26370753288462 +Step 32059 | loss:0.3334459662437439 lr:3.3517378986491212e-06 tokens_per_second_per_gpu:5231.239341741386 mfu:24.67664851646178 +Step 32060 | loss:0.28332066535949707 lr:3.3501771796546676e-06 tokens_per_second_per_gpu:3675.703653480384 mfu:17.338921273178855 +Step 32061 | loss:0.25474321842193604 lr:3.3486168115194837e-06 tokens_per_second_per_gpu:5387.538450148886 mfu:25.413938078197702 +Step 32062 | loss:0.27060699462890625 lr:3.3470567942553155e-06 tokens_per_second_per_gpu:4887.385432343564 mfu:23.054630958305708 +Step 32063 | loss:0.2907867133617401 lr:3.3454971278738867e-06 tokens_per_second_per_gpu:4147.660169636584 mfu:19.565220683972687 +Step 32064 | loss:0.28514769673347473 lr:3.343937812386916e-06 tokens_per_second_per_gpu:4273.9325445301865 mfu:20.160869020634543 +Step 32065 | loss:0.24894869327545166 lr:3.3423788478061437e-06 tokens_per_second_per_gpu:5347.4508796352875 mfu:25.224838168440247 +Step 32066 | loss:0.34077733755111694 lr:3.340820234143299e-06 tokens_per_second_per_gpu:5329.484782092826 mfu:25.14008902100013 +Step 32067 | loss:0.2949160933494568 lr:3.339261971410096e-06 tokens_per_second_per_gpu:3127.223567074701 mfu:14.751647125250805 +Step 32068 | loss:0.21762779355049133 lr:3.3377040596182517e-06 tokens_per_second_per_gpu:5362.270910936305 mfu:25.294746784645227 +Step 32069 | loss:0.3809650242328644 lr:3.336146498779502e-06 tokens_per_second_per_gpu:5374.586186037245 mfu:25.35284003849896 +Step 32070 | loss:0.2810171842575073 lr:3.3345892889055484e-06 tokens_per_second_per_gpu:5257.679420357681 mfu:24.801370878437428 +Step 32071 | loss:0.30738121271133423 lr:3.3330324300080928e-06 tokens_per_second_per_gpu:5214.041837464469 mfu:24.595524954590765 +Step 32072 | loss:0.2638661563396454 lr:3.331475922098859e-06 tokens_per_second_per_gpu:4074.492196076237 mfu:19.22007486894496 +Step 32073 | loss:0.3112574815750122 lr:3.3299197651895542e-06 tokens_per_second_per_gpu:3696.9475643867804 mfu:17.439132425509282 +Step 32074 | loss:0.3002574145793915 lr:3.3283639592918693e-06 tokens_per_second_per_gpu:5029.418242389084 mfu:23.724623956586676 +Step 32075 | loss:0.33820629119873047 lr:3.326808504417517e-06 tokens_per_second_per_gpu:4444.968051582888 mfu:20.967672689068756 +Step 32076 | loss:0.2581150531768799 lr:3.3252534005782043e-06 tokens_per_second_per_gpu:4398.999717040748 mfu:20.750832212026864 +Step 32077 | loss:0.33491018414497375 lr:3.3236986477856e-06 tokens_per_second_per_gpu:3874.007350343789 mfu:18.274353645383083 +Step 32078 | loss:0.3339572846889496 lr:3.3221442460514286e-06 tokens_per_second_per_gpu:5356.977362463681 mfu:25.269776213327358 +Step 32079 | loss:0.33030325174331665 lr:3.3205901953873576e-06 tokens_per_second_per_gpu:4686.726610434896 mfu:22.108089059437205 +Step 32080 | loss:0.26039060950279236 lr:3.3190364958050945e-06 tokens_per_second_per_gpu:4210.897245757031 mfu:19.863520761390582 +Step 32081 | loss:0.29934629797935486 lr:3.317483147316303e-06 tokens_per_second_per_gpu:3587.7201838454653 mfu:16.92388823538308 +Step 32082 | loss:0.2826226055622101 lr:3.315930149932678e-06 tokens_per_second_per_gpu:5356.117009540891 mfu:25.265717781799207 +Step 32083 | loss:0.39088472723960876 lr:3.3143775036659053e-06 tokens_per_second_per_gpu:3995.375563525807 mfu:18.846868214515126 +Step 32084 | loss:0.3692399859428406 lr:3.312825208527648e-06 tokens_per_second_per_gpu:5353.373710527504 mfu:25.252777172297986 +Step 32085 | loss:0.27693232893943787 lr:3.311273264529602e-06 tokens_per_second_per_gpu:5290.518352336977 mfu:24.956277723483318 +Step 32086 | loss:0.30920204520225525 lr:3.309721671683419e-06 tokens_per_second_per_gpu:5320.259535536686 mfu:25.09657195900545 +Step 32087 | loss:0.2234175056219101 lr:3.308170430000779e-06 tokens_per_second_per_gpu:5380.064023582115 mfu:25.37867993281357 +Step 32088 | loss:0.3063022792339325 lr:3.3066195394933442e-06 tokens_per_second_per_gpu:4395.934835373661 mfu:20.73637464228048 +Step 32089 | loss:0.37790530920028687 lr:3.3050690001727834e-06 tokens_per_second_per_gpu:4675.670467367508 mfu:22.05593534621581 +Step 32090 | loss:0.43733057379722595 lr:3.30351881205076e-06 tokens_per_second_per_gpu:4466.121658311571 mfu:21.067457861185428 +Step 32091 | loss:0.26156172156333923 lr:3.301968975138925e-06 tokens_per_second_per_gpu:5394.752769729454 mfu:25.447969254549086 +Step 32092 | loss:0.2910110652446747 lr:3.300419489448947e-06 tokens_per_second_per_gpu:5337.310846455075 mfu:25.17700590186113 +Step 32093 | loss:0.21342717111110687 lr:3.298870354992467e-06 tokens_per_second_per_gpu:4994.523949326962 mfu:23.560021622632334 +Step 32094 | loss:0.40955623984336853 lr:3.297321571781148e-06 tokens_per_second_per_gpu:4284.1998660111685 mfu:20.209301727846018 +Step 32095 | loss:0.2090005874633789 lr:3.29577313982663e-06 tokens_per_second_per_gpu:4727.308986765076 mfu:22.299523052654056 +Step 32096 | loss:0.36340567469596863 lr:3.2942250591405542e-06 tokens_per_second_per_gpu:5385.225269654138 mfu:25.40302641113469 +Step 32097 | loss:0.22864601016044617 lr:3.292677329734578e-06 tokens_per_second_per_gpu:3412.5572225633014 mfu:16.097614661132187 +Step 32098 | loss:0.3122554123401642 lr:3.2911299516203365e-06 tokens_per_second_per_gpu:5324.870224600899 mfu:25.118321365986013 +Step 32099 | loss:0.28560498356819153 lr:3.28958292480947e-06 tokens_per_second_per_gpu:4259.481987252867 mfu:20.09270327643812 +Step 32100 | loss:0.33960631489753723 lr:3.288036249313603e-06 tokens_per_second_per_gpu:3223.9652174002713 mfu:15.207993995664093 +Step 32101 | loss:0.251973420381546 lr:3.286489925144376e-06 tokens_per_second_per_gpu:4491.09412269023 mfu:21.185257236401338 +Step 32102 | loss:0.2616049647331238 lr:3.2849439523134296e-06 tokens_per_second_per_gpu:5369.515289600088 mfu:25.328919754820745 +Step 32103 | loss:0.3528500497341156 lr:3.283398330832371e-06 tokens_per_second_per_gpu:5381.20840853986 mfu:25.384078191911463 +Step 32104 | loss:0.3402201533317566 lr:3.2818530607128306e-06 tokens_per_second_per_gpu:4923.2364373344435 mfu:23.223746265659884 +Step 32105 | loss:0.29778358340263367 lr:3.2803081419664484e-06 tokens_per_second_per_gpu:5353.349932970024 mfu:25.252665009502486 +Step 32106 | loss:0.17655351758003235 lr:3.278763574604826e-06 tokens_per_second_per_gpu:3374.3751408165836 mfu:15.917503266998015 +Step 32107 | loss:0.266920804977417 lr:3.277219358639577e-06 tokens_per_second_per_gpu:5349.989127951737 mfu:25.236811518818758 +Step 32108 | loss:0.2359822392463684 lr:3.2756754940823253e-06 tokens_per_second_per_gpu:5331.617379402603 mfu:25.150148846368968 +Step 32109 | loss:0.2729419469833374 lr:3.2741319809446837e-06 tokens_per_second_per_gpu:5349.896958694416 mfu:25.23637674070628 +Step 32110 | loss:0.3710734248161316 lr:3.272588819238254e-06 tokens_per_second_per_gpu:4940.437494795867 mfu:23.304886588508445 +Step 32111 | loss:0.32296326756477356 lr:3.2710460089746433e-06 tokens_per_second_per_gpu:5368.882959868787 mfu:25.325936947591618 +Step 32112 | loss:0.2863428294658661 lr:3.26950355016547e-06 tokens_per_second_per_gpu:4830.216026816974 mfu:22.784953118330385 +Step 32113 | loss:0.2643943428993225 lr:3.267961442822315e-06 tokens_per_second_per_gpu:5322.521130770783 mfu:25.107240289592625 +Step 32114 | loss:0.3693998157978058 lr:3.2664196869567896e-06 tokens_per_second_per_gpu:5328.564486280141 mfu:25.135747828633036 +Step 32115 | loss:0.3296643793582916 lr:3.2648782825804802e-06 tokens_per_second_per_gpu:3710.179237535786 mfu:17.50154848530937 +Step 32116 | loss:0.3152684271335602 lr:3.263337229704988e-06 tokens_per_second_per_gpu:4272.476019607019 mfu:20.153998344063183 +Step 32117 | loss:0.2928239405155182 lr:3.2617965283418984e-06 tokens_per_second_per_gpu:5365.831918611673 mfu:25.3115446654205 +Step 32118 | loss:0.23808708786964417 lr:3.2602561785027962e-06 tokens_per_second_per_gpu:4458.209671825364 mfu:21.030135671005898 +Step 32119 | loss:0.24091273546218872 lr:3.258716180199278e-06 tokens_per_second_per_gpu:5332.359532611324 mfu:25.1536497096789 +Step 32120 | loss:0.28554341197013855 lr:3.2571765334429184e-06 tokens_per_second_per_gpu:2869.5129355652075 mfu:13.53598210645275 +Step 32121 | loss:0.3697897493839264 lr:3.2556372382453016e-06 tokens_per_second_per_gpu:5345.428365136823 mfu:25.215297622474434 +Step 32122 | loss:0.4107758402824402 lr:3.2540982946179964e-06 tokens_per_second_per_gpu:5318.978746446795 mfu:25.090530258343286 +Step 32123 | loss:0.32404646277427673 lr:3.2525597025725885e-06 tokens_per_second_per_gpu:5364.872887155848 mfu:25.307020750415727 +Step 32124 | loss:0.3313448131084442 lr:3.251021462120635e-06 tokens_per_second_per_gpu:3726.174355904567 mfu:17.577000187703515 +Step 32125 | loss:0.2538149952888489 lr:3.249483573273715e-06 tokens_per_second_per_gpu:3936.091150817749 mfu:18.567213524807386 +Step 32126 | loss:0.33639973402023315 lr:3.2479460360434032e-06 tokens_per_second_per_gpu:4916.873926183288 mfu:23.19373321500354 +Step 32127 | loss:0.2587940990924835 lr:3.2464088504412517e-06 tokens_per_second_per_gpu:3816.188381198732 mfu:18.00161170299247 +Step 32128 | loss:0.2610710859298706 lr:3.2448720164788283e-06 tokens_per_second_per_gpu:5351.11493562615 mfu:25.242122145701565 +Step 32129 | loss:0.34144681692123413 lr:3.2433355341676798e-06 tokens_per_second_per_gpu:4168.198228785747 mfu:19.662102213133515 +Step 32130 | loss:0.3214971721172333 lr:3.24179940351938e-06 tokens_per_second_per_gpu:5370.353020328686 mfu:25.33287147359914 +Step 32131 | loss:0.361482709646225 lr:3.240263624545464e-06 tokens_per_second_per_gpu:5032.120879090934 mfu:23.737372755026673 +Step 32132 | loss:0.27769532799720764 lr:3.23872819725749e-06 tokens_per_second_per_gpu:5042.898564707553 mfu:23.78821293694305 +Step 32133 | loss:0.2577684819698334 lr:3.2371931216670148e-06 tokens_per_second_per_gpu:4791.052421065572 mfu:22.600211708000856 +Step 32134 | loss:0.24304410815238953 lr:3.2356583977855794e-06 tokens_per_second_per_gpu:5376.795656252048 mfu:25.363262486476675 +Step 32135 | loss:0.3010030686855316 lr:3.234124025624724e-06 tokens_per_second_per_gpu:4976.675571236058 mfu:23.475827777930224 +Step 32136 | loss:0.3133932054042816 lr:3.2325900051959846e-06 tokens_per_second_per_gpu:5092.765184474704 mfu:24.02344229050736 +Step 32137 | loss:0.3363282084465027 lr:3.2310563365109124e-06 tokens_per_second_per_gpu:3744.520695573575 mfu:17.66354300213055 +Step 32138 | loss:0.2636577785015106 lr:3.229523019581021e-06 tokens_per_second_per_gpu:3662.385929576174 mfu:17.27609929728469 +Step 32139 | loss:0.2312823086977005 lr:3.2279900544178566e-06 tokens_per_second_per_gpu:4361.582447742596 mfu:20.574328568703038 +Step 32140 | loss:0.23774337768554688 lr:3.2264574410329483e-06 tokens_per_second_per_gpu:4891.861123644762 mfu:23.075743557805136 +Step 32141 | loss:0.3412042558193207 lr:3.2249251794378265e-06 tokens_per_second_per_gpu:4513.532363787195 mfu:21.291102248014422 +Step 32142 | loss:0.21740730106830597 lr:3.2233932696440093e-06 tokens_per_second_per_gpu:5339.929189650141 mfu:25.189357073447667 +Step 32143 | loss:0.24974694848060608 lr:3.221861711663021e-06 tokens_per_second_per_gpu:3888.477015402283 mfu:18.34260952424343 +Step 32144 | loss:0.29440397024154663 lr:3.220330505506375e-06 tokens_per_second_per_gpu:4909.993688932561 mfu:23.161277962002337 +Step 32145 | loss:0.24401676654815674 lr:3.2187996511856e-06 tokens_per_second_per_gpu:4404.705465437821 mfu:20.777747200716917 +Step 32146 | loss:0.36929798126220703 lr:3.217269148712199e-06 tokens_per_second_per_gpu:4137.351032603505 mfu:19.5165907256675 +Step 32147 | loss:0.26937684416770935 lr:3.2157389980976783e-06 tokens_per_second_per_gpu:3403.607737473553 mfu:16.055398413024232 +Step 32148 | loss:0.21299530565738678 lr:3.214209199353568e-06 tokens_per_second_per_gpu:4188.779094194752 mfu:19.75918566672552 +Step 32149 | loss:0.335482656955719 lr:3.212679752491349e-06 tokens_per_second_per_gpu:5219.530117662695 mfu:24.62141410868296 +Step 32150 | loss:0.28891828656196594 lr:3.2111506575225434e-06 tokens_per_second_per_gpu:5315.871699494828 mfu:25.07587378775398 +Step 32151 | loss:0.2765446901321411 lr:3.209621914458638e-06 tokens_per_second_per_gpu:4571.528081389663 mfu:21.564677943035036 +Step 32152 | loss:0.2311084121465683 lr:3.2080935233111454e-06 tokens_per_second_per_gpu:5366.9759379316065 mfu:25.31694119974244 +Step 32153 | loss:0.2834160327911377 lr:3.2065654840915394e-06 tokens_per_second_per_gpu:3764.2357718444946 mfu:17.75654238063935 +Step 32154 | loss:0.2747802436351776 lr:3.205037796811333e-06 tokens_per_second_per_gpu:5258.053167325834 mfu:24.803133906654015 +Step 32155 | loss:0.34426021575927734 lr:3.2035104614820067e-06 tokens_per_second_per_gpu:5328.1709456780645 mfu:25.133891430468093 +Step 32156 | loss:0.23831552267074585 lr:3.2019834781150504e-06 tokens_per_second_per_gpu:4160.649083733287 mfu:19.626491607904097 +Step 32157 | loss:0.37984034419059753 lr:3.20045684672195e-06 tokens_per_second_per_gpu:4443.406487888484 mfu:20.960306526692012 +Step 32158 | loss:0.3628310561180115 lr:3.198930567314179e-06 tokens_per_second_per_gpu:3847.0064823309895 mfu:18.14698594414358 +Step 32159 | loss:0.24442395567893982 lr:3.197404639903234e-06 tokens_per_second_per_gpu:5314.425905260768 mfu:25.069053729673982 +Step 32160 | loss:0.2843814492225647 lr:3.1958790645005677e-06 tokens_per_second_per_gpu:5035.811757431499 mfu:23.754783257887897 +Step 32161 | loss:0.3141205310821533 lr:3.19435384111767e-06 tokens_per_second_per_gpu:4000.471807010598 mfu:18.870908064541684 +Step 32162 | loss:0.35130593180656433 lr:3.1928289697660207e-06 tokens_per_second_per_gpu:5362.460948014818 mfu:25.295643221968238 +Step 32163 | loss:0.21355409920215607 lr:3.191304450457061e-06 tokens_per_second_per_gpu:4847.634553281235 mfu:22.867119279569373 +Step 32164 | loss:0.3086933195590973 lr:3.1897802832022872e-06 tokens_per_second_per_gpu:4968.036611466397 mfu:23.435076330738205 +Step 32165 | loss:0.3207922577857971 lr:3.18825646801314e-06 tokens_per_second_per_gpu:3510.8565556704075 mfu:16.561309386994065 +Step 32166 | loss:0.2817896902561188 lr:3.186733004901099e-06 tokens_per_second_per_gpu:5314.5392048756985 mfu:25.069588183288523 +Step 32167 | loss:0.33735302090644836 lr:3.1852098938775996e-06 tokens_per_second_per_gpu:4117.933962915155 mfu:19.424997095053847 +Step 32168 | loss:0.32690441608428955 lr:3.1836871349541164e-06 tokens_per_second_per_gpu:4322.678095415935 mfu:20.390810101011205 +Step 32169 | loss:0.2875705659389496 lr:3.182164728142101e-06 tokens_per_second_per_gpu:5381.308856582009 mfu:25.384552022464483 +Step 32170 | loss:0.31259867548942566 lr:3.1806426734529883e-06 tokens_per_second_per_gpu:5107.656366034463 mfu:24.093686534622304 +Step 32171 | loss:0.38895702362060547 lr:3.179120970898242e-06 tokens_per_second_per_gpu:4978.806534951833 mfu:23.485879897357435 +Step 32172 | loss:0.3672153949737549 lr:3.1775996204893023e-06 tokens_per_second_per_gpu:4531.368401327419 mfu:21.37523788023346 +Step 32173 | loss:0.2440711259841919 lr:3.176078622237616e-06 tokens_per_second_per_gpu:4353.92940254923 mfu:20.538227848781776 +Step 32174 | loss:0.2687681019306183 lr:3.1745579761546073e-06 tokens_per_second_per_gpu:5304.195091720983 mfu:25.020793236650036 +Step 32175 | loss:0.38714179396629333 lr:3.1730376822517226e-06 tokens_per_second_per_gpu:4053.798840885733 mfu:19.122460781860887 +Step 32176 | loss:0.30711647868156433 lr:3.1715177405403973e-06 tokens_per_second_per_gpu:4634.167191344399 mfu:21.860157312025727 +Step 32177 | loss:0.344683974981308 lr:3.1699981510320664e-06 tokens_per_second_per_gpu:5322.762796976275 mfu:25.108380270316438 +Step 32178 | loss:0.35115793347358704 lr:3.1684789137381544e-06 tokens_per_second_per_gpu:5337.813854064265 mfu:25.179378674575624 +Step 32179 | loss:0.35173851251602173 lr:3.1669600286700905e-06 tokens_per_second_per_gpu:4060.1825246871 mfu:19.1525737075207 +Step 32180 | loss:0.2864997982978821 lr:3.1654414958392886e-06 tokens_per_second_per_gpu:5361.925866514731 mfu:25.29311915123801 +Step 32181 | loss:0.24589216709136963 lr:3.163923315257189e-06 tokens_per_second_per_gpu:4832.85363365226 mfu:22.79739515979456 +Step 32182 | loss:0.45474037528038025 lr:3.162405486935188e-06 tokens_per_second_per_gpu:5347.330470585091 mfu:25.224270178406567 +Step 32183 | loss:0.22343862056732178 lr:3.1608880108847107e-06 tokens_per_second_per_gpu:3792.824576158007 mfu:17.89140064833918 +Step 32184 | loss:0.3144221305847168 lr:3.15937088711718e-06 tokens_per_second_per_gpu:4328.884008666591 mfu:20.420084452652535 +Step 32185 | loss:0.21348460018634796 lr:3.157854115643993e-06 tokens_per_second_per_gpu:4360.175321860053 mfu:20.567690915834653 +Step 32186 | loss:0.28094738721847534 lr:3.1563376964765633e-06 tokens_per_second_per_gpu:5360.198030780291 mfu:25.284968655279485 +Step 32187 | loss:0.3568577468395233 lr:3.154821629626292e-06 tokens_per_second_per_gpu:4371.177045333482 mfu:20.619587922546398 +Step 32188 | loss:0.31401684880256653 lr:3.153305915104593e-06 tokens_per_second_per_gpu:4038.2018618835154 mfu:19.0488871707884 +Step 32189 | loss:0.35090234875679016 lr:3.151790552922851e-06 tokens_per_second_per_gpu:3599.1842800517866 mfu:16.977966333164808 +Step 32190 | loss:0.2833586037158966 lr:3.150275543092468e-06 tokens_per_second_per_gpu:5348.341721354712 mfu:25.229040420824045 +Step 32191 | loss:0.3409133851528168 lr:3.148760885624846e-06 tokens_per_second_per_gpu:4856.8192151555895 mfu:22.910444896695612 +Step 32192 | loss:0.3597066104412079 lr:3.147246580531371e-06 tokens_per_second_per_gpu:5032.135343786463 mfu:23.737440987463412 +Step 32193 | loss:0.35479065775871277 lr:3.1457326278234334e-06 tokens_per_second_per_gpu:4944.867279940375 mfu:23.32578264083437 +Step 32194 | loss:0.21998777985572815 lr:3.144219027512413e-06 tokens_per_second_per_gpu:4824.223929049661 mfu:22.756687370804666 +Step 32195 | loss:0.28441405296325684 lr:3.1427057796097116e-06 tokens_per_second_per_gpu:5334.631182453641 mfu:25.164365469568537 +Step 32196 | loss:0.30399543046951294 lr:3.141192884126687e-06 tokens_per_second_per_gpu:5315.990493540603 mfu:25.076434159536255 +Step 32197 | loss:0.3469694256782532 lr:3.1396803410747353e-06 tokens_per_second_per_gpu:3153.781176653095 mfu:14.876923900827524 +Step 32198 | loss:0.2298918515443802 lr:3.1381681504652317e-06 tokens_per_second_per_gpu:3428.443646096687 mfu:16.17255363144252 +Step 32199 | loss:0.3619895577430725 lr:3.1366563123095383e-06 tokens_per_second_per_gpu:5336.864925824371 mfu:25.17490241816419 +Step 32200 | loss:0.3521029055118561 lr:3.13514482661903e-06 tokens_per_second_per_gpu:5332.691328914769 mfu:25.155214849452534 +Step 32201 | loss:0.3495228588581085 lr:3.133633693405086e-06 tokens_per_second_per_gpu:5286.274526956306 mfu:24.93625887509142 +Step 32202 | loss:0.2943861484527588 lr:3.132122912679064e-06 tokens_per_second_per_gpu:4589.335857207629 mfu:21.648680259888234 +Step 32203 | loss:0.4707711637020111 lr:3.1306124844523166e-06 tokens_per_second_per_gpu:5375.344169382544 mfu:25.356415575264034 +Step 32204 | loss:0.3783426284790039 lr:3.1291024087362174e-06 tokens_per_second_per_gpu:5122.005635174904 mfu:24.161374485395534 +Step 32205 | loss:0.29554039239883423 lr:3.1275926855421243e-06 tokens_per_second_per_gpu:4824.157615673868 mfu:22.756374559296827 +Step 32206 | loss:0.34826165437698364 lr:3.1260833148813783e-06 tokens_per_second_per_gpu:4846.884186427032 mfu:22.863579671092307 +Step 32207 | loss:0.35239550471305847 lr:3.1245742967653426e-06 tokens_per_second_per_gpu:3762.2056814501607 mfu:17.746966097880136 +Step 32208 | loss:0.4111495018005371 lr:3.123065631205374e-06 tokens_per_second_per_gpu:4395.806806564213 mfu:20.735770708542347 +Step 32209 | loss:0.20186659693717957 lr:3.121557318212809e-06 tokens_per_second_per_gpu:4422.278783032644 mfu:20.86064353812886 +Step 32210 | loss:0.30257096886634827 lr:3.1200493577989875e-06 tokens_per_second_per_gpu:4306.443248210958 mfu:20.314227556794766 +Step 32211 | loss:0.4467467665672302 lr:3.118541749975257e-06 tokens_per_second_per_gpu:4908.778554504342 mfu:23.155545965581116 +Step 32212 | loss:0.2019387036561966 lr:3.1170344947529517e-06 tokens_per_second_per_gpu:5357.936758250715 mfu:25.274301846944596 +Step 32213 | loss:0.42114320397377014 lr:3.1155275921434245e-06 tokens_per_second_per_gpu:4121.750391031689 mfu:19.442999837629323 +Step 32214 | loss:0.3462579846382141 lr:3.114021042157983e-06 tokens_per_second_per_gpu:5245.421102445398 mfu:24.743546301360485 +Step 32215 | loss:0.3283945918083191 lr:3.112514844807984e-06 tokens_per_second_per_gpu:4741.166983055864 mfu:22.364893585575675 +Step 32216 | loss:0.3591717779636383 lr:3.1110090001047365e-06 tokens_per_second_per_gpu:4265.59215309933 mfu:20.121525971237677 +Step 32217 | loss:0.2646867334842682 lr:3.1095035080595804e-06 tokens_per_second_per_gpu:4226.799597366505 mfu:19.938534867153436 +Step 32218 | loss:0.3280591368675232 lr:3.1079983686838234e-06 tokens_per_second_per_gpu:5369.187886787305 mfu:25.327375339892164 +Step 32219 | loss:0.2697507441043854 lr:3.106493581988795e-06 tokens_per_second_per_gpu:5343.76398029499 mfu:25.207446435202172 +Step 32220 | loss:0.28170594573020935 lr:3.1049891479858206e-06 tokens_per_second_per_gpu:4984.917711761826 mfu:23.514707361044454 +Step 32221 | loss:0.23879548907279968 lr:3.103485066686196e-06 tokens_per_second_per_gpu:5356.091529608996 mfu:25.265597588613137 +Step 32222 | loss:0.33580702543258667 lr:3.1019813381012564e-06 tokens_per_second_per_gpu:4908.400497480576 mfu:23.153762606096834 +Step 32223 | loss:0.27980443835258484 lr:3.1004779622422875e-06 tokens_per_second_per_gpu:5336.169954220769 mfu:25.171624118535615 +Step 32224 | loss:0.3287118673324585 lr:3.098974939120619e-06 tokens_per_second_per_gpu:5345.198864390845 mfu:25.214215028298792 +Step 32225 | loss:0.3898313045501709 lr:3.097472268747531e-06 tokens_per_second_per_gpu:5383.666377203811 mfu:25.395672849472017 +Step 32226 | loss:0.3258628845214844 lr:3.0959699511343475e-06 tokens_per_second_per_gpu:5319.269253974265 mfu:25.09190063191605 +Step 32227 | loss:0.31067708134651184 lr:3.0944679862923597e-06 tokens_per_second_per_gpu:4353.114840548362 mfu:20.534385420845073 +Step 32228 | loss:0.3233540952205658 lr:3.0929663742328583e-06 tokens_per_second_per_gpu:5086.118782770096 mfu:23.992090079673662 +Step 32229 | loss:0.3203314244747162 lr:3.0914651149671514e-06 tokens_per_second_per_gpu:5356.807969310312 mfu:25.26897715692146 +Step 32230 | loss:0.3018398582935333 lr:3.089964208506513e-06 tokens_per_second_per_gpu:4014.3187766677743 mfu:18.936226582951225 +Step 32231 | loss:0.33710819482803345 lr:3.0884636548622446e-06 tokens_per_second_per_gpu:3872.5073683338824 mfu:18.26727797431903 +Step 32232 | loss:0.29762041568756104 lr:3.0869634540456217e-06 tokens_per_second_per_gpu:4112.2676864099585 mfu:19.398268302012376 +Step 32233 | loss:0.3079560101032257 lr:3.085463606067929e-06 tokens_per_second_per_gpu:4416.513544763534 mfu:20.833447925562385 +Step 32234 | loss:0.2822454273700714 lr:3.083964110940463e-06 tokens_per_second_per_gpu:2991.6609997496134 mfu:14.112175365819358 +Step 32235 | loss:0.27081298828125 lr:3.082464968674481e-06 tokens_per_second_per_gpu:3752.5063979550127 mfu:17.701212922765116 +Step 32236 | loss:0.22561047971248627 lr:3.0809661792812637e-06 tokens_per_second_per_gpu:4317.160743991027 mfu:20.36478381298269 +Step 32237 | loss:0.3271295130252838 lr:3.0794677427720964e-06 tokens_per_second_per_gpu:4906.008247632788 mfu:23.142477955404907 +Step 32238 | loss:0.3074582815170288 lr:3.077969659158242e-06 tokens_per_second_per_gpu:5063.177378783065 mfu:23.883871562859696 +Step 32239 | loss:0.2526545822620392 lr:3.0764719284509526e-06 tokens_per_second_per_gpu:4714.327102880028 mfu:22.238285291430497 +Step 32240 | loss:0.27699801325798035 lr:3.074974550661508e-06 tokens_per_second_per_gpu:4273.664881092301 mfu:20.159606406529942 +Step 32241 | loss:0.36005833745002747 lr:3.0734775258011716e-06 tokens_per_second_per_gpu:4683.287427042684 mfu:22.091865844590945 +Step 32242 | loss:0.26487722992897034 lr:3.0719808538811958e-06 tokens_per_second_per_gpu:3480.6383619986664 mfu:16.418765011688976 +Step 32243 | loss:0.23708397150039673 lr:3.0704845349128373e-06 tokens_per_second_per_gpu:5374.348669683628 mfu:25.35171963333448 +Step 32244 | loss:0.2071571946144104 lr:3.0689885689073605e-06 tokens_per_second_per_gpu:3830.104589061685 mfu:18.06725682983192 +Step 32245 | loss:0.34845080971717834 lr:3.0674929558760115e-06 tokens_per_second_per_gpu:4766.381279810587 mfu:22.483833725370342 +Step 32246 | loss:0.396581768989563 lr:3.065997695830025e-06 tokens_per_second_per_gpu:4455.616788416024 mfu:21.017904597572514 +Step 32247 | loss:0.22435805201530457 lr:3.0645027887806654e-06 tokens_per_second_per_gpu:5364.258108499605 mfu:25.30412073460218 +Step 32248 | loss:0.2901807427406311 lr:3.063008234739162e-06 tokens_per_second_per_gpu:5337.4934701816555 mfu:25.17786736913823 +Step 32249 | loss:0.349104642868042 lr:3.0615140337167726e-06 tokens_per_second_per_gpu:4877.31144099787 mfu:23.007110222327906 +Step 32250 | loss:0.3343580663204193 lr:3.060020185724721e-06 tokens_per_second_per_gpu:4309.50716329736 mfu:20.328680567015148 +Step 32251 | loss:0.31980887055397034 lr:3.058526690774249e-06 tokens_per_second_per_gpu:4404.215944933716 mfu:20.775438048978245 +Step 32252 | loss:0.3631584048271179 lr:3.0570335488765866e-06 tokens_per_second_per_gpu:5167.954955392469 mfu:24.37812526862423 +Step 32253 | loss:0.3428267538547516 lr:3.0555407600429686e-06 tokens_per_second_per_gpu:5367.2526990764045 mfu:25.31824672928278 +Step 32254 | loss:0.2530462443828583 lr:3.054048324284614e-06 tokens_per_second_per_gpu:5387.10607903985 mfu:25.411898509906262 +Step 32255 | loss:0.3299449682235718 lr:3.0525562416127472e-06 tokens_per_second_per_gpu:4328.685597703854 mfu:20.419148514288903 +Step 32256 | loss:0.26086631417274475 lr:3.0510645120386094e-06 tokens_per_second_per_gpu:2649.0456632219334 mfu:12.496000367214068 +Step 32257 | loss:0.3700195550918579 lr:3.0495731355733914e-06 tokens_per_second_per_gpu:4034.7567891896865 mfu:19.032636175101597 +Step 32258 | loss:0.3390894830226898 lr:3.0480821122283398e-06 tokens_per_second_per_gpu:3621.476927366053 mfu:17.08312455406404 +Step 32259 | loss:0.3302031457424164 lr:3.0465914420146404e-06 tokens_per_second_per_gpu:5367.3073220931265 mfu:25.318504395372305 +Step 32260 | loss:0.3205382525920868 lr:3.045101124943528e-06 tokens_per_second_per_gpu:2819.0879763039056 mfu:13.298118970231986 +Step 32261 | loss:0.3656240403652191 lr:3.043611161026194e-06 tokens_per_second_per_gpu:4481.507706004196 mfu:21.14003647328197 +Step 32262 | loss:0.30137506127357483 lr:3.042121550273852e-06 tokens_per_second_per_gpu:3743.396381932274 mfu:17.658239422856845 +Step 32263 | loss:0.3569628894329071 lr:3.0406322926977095e-06 tokens_per_second_per_gpu:3811.5916681135377 mfu:17.979928223089487 +Step 32264 | loss:0.2963256239891052 lr:3.0391433883089626e-06 tokens_per_second_per_gpu:4718.821206264387 mfu:22.25948474386762 +Step 32265 | loss:0.2808307707309723 lr:3.0376548371188084e-06 tokens_per_second_per_gpu:4989.6173087443585 mfu:23.536876161844987 +Step 32266 | loss:0.2803879976272583 lr:3.0361666391384493e-06 tokens_per_second_per_gpu:4264.714017440002 mfu:20.117383655507233 +Step 32267 | loss:0.30397680401802063 lr:3.0346787943790755e-06 tokens_per_second_per_gpu:5352.570710937785 mfu:25.248989286227427 +Step 32268 | loss:0.23365378379821777 lr:3.033191302851868e-06 tokens_per_second_per_gpu:2942.7332022910673 mfu:13.881374597264339 +Step 32269 | loss:0.27373695373535156 lr:3.0317041645680224e-06 tokens_per_second_per_gpu:4220.137231491774 mfu:19.907107350605497 +Step 32270 | loss:0.2141844481229782 lr:3.03021737953873e-06 tokens_per_second_per_gpu:5377.690963473498 mfu:25.36748580339542 +Step 32271 | loss:0.34309089183807373 lr:3.0287309477751546e-06 tokens_per_second_per_gpu:4551.706552008786 mfu:21.4711764070438 +Step 32272 | loss:0.2995474934577942 lr:3.0272448692884927e-06 tokens_per_second_per_gpu:4430.834238280708 mfu:20.901001080245162 +Step 32273 | loss:0.390342116355896 lr:3.0257591440899235e-06 tokens_per_second_per_gpu:3994.980929765205 mfu:18.845006659734903 +Step 32274 | loss:0.30962073802948 lr:3.024273772190611e-06 tokens_per_second_per_gpu:4219.078760528895 mfu:19.90211436247983 +Step 32275 | loss:0.27366217970848083 lr:3.0227887536017185e-06 tokens_per_second_per_gpu:5372.462103178771 mfu:25.342820377249634 +Step 32276 | loss:0.3057175874710083 lr:3.0213040883344313e-06 tokens_per_second_per_gpu:3317.217320295291 mfu:15.647880075469288 +Step 32277 | loss:0.34006795287132263 lr:3.019819776399918e-06 tokens_per_second_per_gpu:4500.185590272887 mfu:21.228143240153052 +Step 32278 | loss:0.32678067684173584 lr:3.0183358178093258e-06 tokens_per_second_per_gpu:4608.1073010533555 mfu:21.737228363247105 +Step 32279 | loss:0.21190650761127472 lr:3.0168522125738286e-06 tokens_per_second_per_gpu:4860.293820412511 mfu:22.9268352025213 +Step 32280 | loss:0.30230143666267395 lr:3.0153689607045845e-06 tokens_per_second_per_gpu:3680.802858873153 mfu:17.3629750950307 +Step 32281 | loss:0.2333102971315384 lr:3.013886062212745e-06 tokens_per_second_per_gpu:4471.757242844875 mfu:21.09404187495889 +Step 32282 | loss:0.3770197927951813 lr:3.012403517109458e-06 tokens_per_second_per_gpu:4772.557752063258 mfu:22.512969198801024 +Step 32283 | loss:0.3125559091567993 lr:3.01092132540588e-06 tokens_per_second_per_gpu:3469.037553375742 mfu:16.364042018112592 +Step 32284 | loss:0.23565985262393951 lr:3.0094394871131582e-06 tokens_per_second_per_gpu:4195.899189765649 mfu:19.792772372347013 +Step 32285 | loss:0.25427788496017456 lr:3.0079580022424447e-06 tokens_per_second_per_gpu:5384.046345304914 mfu:25.397465223834164 +Step 32286 | loss:0.3489796221256256 lr:3.0064768708048694e-06 tokens_per_second_per_gpu:4748.986594285011 mfu:22.401780025906817 +Step 32287 | loss:0.30771446228027344 lr:3.004996092811585e-06 tokens_per_second_per_gpu:4880.173278606109 mfu:23.020609998605817 +Step 32288 | loss:0.34655848145484924 lr:3.0035156682737155e-06 tokens_per_second_per_gpu:4756.059775681575 mfu:22.435145429363594 +Step 32289 | loss:0.3074283003807068 lr:3.002035597202407e-06 tokens_per_second_per_gpu:5311.719719973808 mfu:25.056288191952575 +Step 32290 | loss:0.2869666814804077 lr:3.0005558796087798e-06 tokens_per_second_per_gpu:3724.743309223608 mfu:17.57024969634769 +Step 32291 | loss:0.30629634857177734 lr:2.999076515503968e-06 tokens_per_second_per_gpu:3996.608756593977 mfu:18.852685396622114 +Step 32292 | loss:0.3451024889945984 lr:2.997597504899108e-06 tokens_per_second_per_gpu:4622.652379313927 mfu:21.805839979048475 +Step 32293 | loss:0.2984468936920166 lr:2.9961188478053016e-06 tokens_per_second_per_gpu:4382.080636547808 mfu:20.67102202264887 +Step 32294 | loss:0.3706966042518616 lr:2.99464054423369e-06 tokens_per_second_per_gpu:4832.518597788202 mfu:22.795814738460784 +Step 32295 | loss:0.33201929926872253 lr:2.993162594195387e-06 tokens_per_second_per_gpu:3853.492714744237 mfu:18.177582609102238 +Step 32296 | loss:0.27286297082901 lr:2.9916849977015103e-06 tokens_per_second_per_gpu:5322.399292142903 mfu:25.106665555997097 +Step 32297 | loss:0.3122905492782593 lr:2.9902077547631578e-06 tokens_per_second_per_gpu:4226.527288677364 mfu:19.937250340606102 +Step 32298 | loss:0.411112517118454 lr:2.9887308653914536e-06 tokens_per_second_per_gpu:4751.098926647649 mfu:22.41174425806244 +Step 32299 | loss:0.3271263837814331 lr:2.987254329597511e-06 tokens_per_second_per_gpu:4155.397846139244 mfu:19.601720624219478 +Step 32300 | loss:0.2938200831413269 lr:2.9857781473924205e-06 tokens_per_second_per_gpu:5308.072095668158 mfu:25.03908172575394 +Step 32301 | loss:0.28620681166648865 lr:2.984302318787291e-06 tokens_per_second_per_gpu:3833.513080133798 mfu:18.08333526376762 +Step 32302 | loss:0.3078466057777405 lr:2.9828268437932295e-06 tokens_per_second_per_gpu:3944.7994351544935 mfu:18.608291987810325 +Step 32303 | loss:0.4065417945384979 lr:2.9813517224213273e-06 tokens_per_second_per_gpu:4962.253823278086 mfu:23.40779793220854 +Step 32304 | loss:0.3017864227294922 lr:2.97987695468267e-06 tokens_per_second_per_gpu:4195.415912133962 mfu:19.790492669302676 +Step 32305 | loss:0.30013343691825867 lr:2.9784025405883597e-06 tokens_per_second_per_gpu:4311.958011736469 mfu:20.340241637260288 +Step 32306 | loss:0.24947190284729004 lr:2.9769284801494935e-06 tokens_per_second_per_gpu:4418.688884249235 mfu:20.84370937306763 +Step 32307 | loss:0.30639347434043884 lr:2.975454773377134e-06 tokens_per_second_per_gpu:4582.26776016833 mfu:21.615338840212157 +Step 32308 | loss:0.30005788803100586 lr:2.9739814202823844e-06 tokens_per_second_per_gpu:3279.661792398973 mfu:15.470724242748561 +Step 32309 | loss:0.22625885903835297 lr:2.9725084208763297e-06 tokens_per_second_per_gpu:4324.9012443166075 mfu:20.401297073683992 +Step 32310 | loss:0.24239659309387207 lr:2.971035775170039e-06 tokens_per_second_per_gpu:4984.481856876893 mfu:23.512651359187714 +Step 32311 | loss:0.26279929280281067 lr:2.9695634831745812e-06 tokens_per_second_per_gpu:5290.570383047179 mfu:24.956523161220694 +Step 32312 | loss:0.29509419202804565 lr:2.968091544901036e-06 tokens_per_second_per_gpu:5305.975564373358 mfu:25.029192029931245 +Step 32313 | loss:0.39362236857414246 lr:2.9666199603604838e-06 tokens_per_second_per_gpu:3892.0583285408884 mfu:18.359503189353262 +Step 32314 | loss:0.3818725347518921 lr:2.965148729563977e-06 tokens_per_second_per_gpu:5337.6945167064605 mfu:25.178815739897807 +Step 32315 | loss:0.2657831609249115 lr:2.9636778525225895e-06 tokens_per_second_per_gpu:4753.718720698018 mfu:22.424102273580992 +Step 32316 | loss:0.22672627866268158 lr:2.9622073292473853e-06 tokens_per_second_per_gpu:4269.975674476493 mfu:20.142203789479623 +Step 32317 | loss:0.2759299874305725 lr:2.9607371597494273e-06 tokens_per_second_per_gpu:3757.2601553236163 mfu:17.723637207347462 +Step 32318 | loss:0.2545091211795807 lr:2.9592673440397567e-06 tokens_per_second_per_gpu:5341.9991519544565 mfu:25.19912143880911 +Step 32319 | loss:0.3147686719894409 lr:2.9577978821294428e-06 tokens_per_second_per_gpu:4800.643054167771 mfu:22.645452360674668 +Step 32320 | loss:0.3112074136734009 lr:2.956328774029532e-06 tokens_per_second_per_gpu:3791.854487718647 mfu:17.886824575656565 +Step 32321 | loss:0.28137001395225525 lr:2.954860019751082e-06 tokens_per_second_per_gpu:4062.1907332524765 mfu:19.162046769959296 +Step 32322 | loss:0.26594358682632446 lr:2.9533916193051235e-06 tokens_per_second_per_gpu:4869.372559184782 mfu:22.96966116230266 +Step 32323 | loss:0.38387972116470337 lr:2.9519235727027193e-06 tokens_per_second_per_gpu:4565.885282299139 mfu:21.538059896964235 +Step 32324 | loss:0.28382688760757446 lr:2.9504558799548942e-06 tokens_per_second_per_gpu:5367.43291650908 mfu:25.319096845660976 +Step 32325 | loss:0.2610171139240265 lr:2.9489885410727003e-06 tokens_per_second_per_gpu:3127.4996623725146 mfu:14.752949513877308 +Step 32326 | loss:0.35495525598526 lr:2.947521556067162e-06 tokens_per_second_per_gpu:5342.50996016767 mfu:25.201531008303707 +Step 32327 | loss:0.3081000745296478 lr:2.946054924949315e-06 tokens_per_second_per_gpu:5339.03272866108 mfu:25.18512831401025 +Step 32328 | loss:0.22118449211120605 lr:2.944588647730201e-06 tokens_per_second_per_gpu:5056.027633775301 mfu:23.850144995785783 +Step 32329 | loss:0.35548996925354004 lr:2.9431227244208327e-06 tokens_per_second_per_gpu:4482.252671143664 mfu:21.143550600949002 +Step 32330 | loss:0.4120674133300781 lr:2.941657155032246e-06 tokens_per_second_per_gpu:4562.099173460604 mfu:21.520200175596084 +Step 32331 | loss:0.27679139375686646 lr:2.940191939575465e-06 tokens_per_second_per_gpu:4902.738870701078 mfu:23.127055746604803 +Step 32332 | loss:0.28068241477012634 lr:2.9387270780615095e-06 tokens_per_second_per_gpu:3930.512677611474 mfu:18.540898914907714 +Step 32333 | loss:0.42786136269569397 lr:2.937262570501381e-06 tokens_per_second_per_gpu:5296.163080436002 mfu:24.98290486901634 +Step 32334 | loss:0.35583174228668213 lr:2.9357984169061047e-06 tokens_per_second_per_gpu:3938.378578413888 mfu:18.578003711053764 +Step 32335 | loss:0.38842344284057617 lr:2.9343346172867047e-06 tokens_per_second_per_gpu:4115.127220048152 mfu:19.41175721007048 +Step 32336 | loss:0.2881014347076416 lr:2.9328711716541725e-06 tokens_per_second_per_gpu:5075.336250241176 mfu:23.941227034044992 +Step 32337 | loss:0.3029603064060211 lr:2.931408080019521e-06 tokens_per_second_per_gpu:4460.3866475831865 mfu:21.04040483708591 +Step 32338 | loss:0.2866900563240051 lr:2.9299453423937586e-06 tokens_per_second_per_gpu:4686.366683306476 mfu:22.106391221762546 +Step 32339 | loss:0.22167691588401794 lr:2.928482958787887e-06 tokens_per_second_per_gpu:3835.537198335896 mfu:18.092883374677093 +Step 32340 | loss:0.3609124720096588 lr:2.927020929212898e-06 tokens_per_second_per_gpu:3549.7749256503657 mfu:16.74489397834816 +Step 32341 | loss:0.26012977957725525 lr:2.9255592536797828e-06 tokens_per_second_per_gpu:5181.775019681647 mfu:24.443316869822393 +Step 32342 | loss:0.2569579482078552 lr:2.92409793219956e-06 tokens_per_second_per_gpu:5336.825273044696 mfu:25.174715369237 +Step 32343 | loss:0.3615957498550415 lr:2.9226369647831874e-06 tokens_per_second_per_gpu:5326.1622574096655 mfu:25.124416105188494 +Step 32344 | loss:0.3229750990867615 lr:2.9211763514416734e-06 tokens_per_second_per_gpu:5219.755664390767 mfu:24.622478051081515 +Step 32345 | loss:0.41360825300216675 lr:2.919716092186009e-06 tokens_per_second_per_gpu:5330.071849886761 mfu:25.142858319948214 +Step 32346 | loss:0.3403061628341675 lr:2.918256187027163e-06 tokens_per_second_per_gpu:5346.023836907262 mfu:25.218106564413556 +Step 32347 | loss:0.2713067829608917 lr:2.9167966359761156e-06 tokens_per_second_per_gpu:5358.9001266049 mfu:25.278846219838876 +Step 32348 | loss:0.2639107406139374 lr:2.915337439043847e-06 tokens_per_second_per_gpu:4262.057915495194 mfu:20.10485436945585 +Step 32349 | loss:0.39253541827201843 lr:2.913878596241343e-06 tokens_per_second_per_gpu:5120.5411972732145 mfu:24.154466481954397 +Step 32350 | loss:0.3196013867855072 lr:2.9124201075795552e-06 tokens_per_second_per_gpu:4392.392352415546 mfu:20.719664145757136 +Step 32351 | loss:0.3269067406654358 lr:2.910961973069465e-06 tokens_per_second_per_gpu:4399.591873715918 mfu:20.753625516095987 +Step 32352 | loss:0.4163283705711365 lr:2.909504192722046e-06 tokens_per_second_per_gpu:3747.1459738211547 mfu:17.675926887543042 +Step 32353 | loss:0.27092286944389343 lr:2.908046766548256e-06 tokens_per_second_per_gpu:4891.910569837548 mfu:23.0759768039329 +Step 32354 | loss:0.2772519886493683 lr:2.9065896945590476e-06 tokens_per_second_per_gpu:5329.557293991647 mfu:25.140431071998677 +Step 32355 | loss:0.3311622440814972 lr:2.90513297676539e-06 tokens_per_second_per_gpu:5352.072952229929 mfu:25.246641273476072 +Step 32356 | loss:0.3559764623641968 lr:2.90367661317823e-06 tokens_per_second_per_gpu:4479.631423454325 mfu:21.131185728367416 +Step 32357 | loss:0.2885976731777191 lr:2.902220603808542e-06 tokens_per_second_per_gpu:4076.5872794688894 mfu:19.229957734766185 +Step 32358 | loss:0.29178449511528015 lr:2.9007649486672508e-06 tokens_per_second_per_gpu:4617.314007481296 mfu:21.7806579682938 +Step 32359 | loss:0.25618433952331543 lr:2.899309647765319e-06 tokens_per_second_per_gpu:5181.073430933743 mfu:24.440007355995146 +Step 32360 | loss:0.3168551027774811 lr:2.8978547011137003e-06 tokens_per_second_per_gpu:5331.266825132048 mfu:25.148495222064035 +Step 32361 | loss:0.420272558927536 lr:2.896400108723318e-06 tokens_per_second_per_gpu:4465.273096934514 mfu:21.06345505239035 +Step 32362 | loss:0.3275412917137146 lr:2.8949458706051146e-06 tokens_per_second_per_gpu:4281.940703479406 mfu:20.19864487273082 +Step 32363 | loss:0.3558990955352783 lr:2.8934919867700362e-06 tokens_per_second_per_gpu:4478.313529709507 mfu:21.12496899871725 +Step 32364 | loss:0.3546281158924103 lr:2.892038457229024e-06 tokens_per_second_per_gpu:5195.926096708482 mfu:24.510069914580683 +Step 32365 | loss:0.3290041387081146 lr:2.890585281992997e-06 tokens_per_second_per_gpu:5350.988459822078 mfu:25.241525537754757 +Step 32366 | loss:0.3468143045902252 lr:2.8891324610728854e-06 tokens_per_second_per_gpu:3844.328720451625 mfu:18.134354484485332 +Step 32367 | loss:0.2930341362953186 lr:2.887679994479625e-06 tokens_per_second_per_gpu:5304.3408786868795 mfu:25.02148093864199 +Step 32368 | loss:0.3318128287792206 lr:2.8862278822241407e-06 tokens_per_second_per_gpu:5350.089015868141 mfu:25.237282707163075 +Step 32369 | loss:0.3628027141094208 lr:2.8847761243173342e-06 tokens_per_second_per_gpu:5201.785502480481 mfu:24.537709731324917 +Step 32370 | loss:0.2415461391210556 lr:2.883324720770142e-06 tokens_per_second_per_gpu:4904.945396899088 mfu:23.137464307153007 +Step 32371 | loss:0.26816263794898987 lr:2.881873671593488e-06 tokens_per_second_per_gpu:4635.1452234527615 mfu:21.864770856350358 +Step 32372 | loss:0.3231385052204132 lr:2.8804229767982637e-06 tokens_per_second_per_gpu:4518.705854766439 mfu:21.315506487652097 +Step 32373 | loss:0.28822362422943115 lr:2.8789726363953883e-06 tokens_per_second_per_gpu:4843.172084194863 mfu:22.84606905151322 +Step 32374 | loss:0.31499984860420227 lr:2.877522650395781e-06 tokens_per_second_per_gpu:3454.983537788906 mfu:16.297746828727167 +Step 32375 | loss:0.3022369146347046 lr:2.876073018810338e-06 tokens_per_second_per_gpu:5327.879590821515 mfu:25.132517059899488 +Step 32376 | loss:0.2642122507095337 lr:2.8746237416499568e-06 tokens_per_second_per_gpu:5309.913030543615 mfu:25.04776572965735 +Step 32377 | loss:0.29723912477493286 lr:2.8731748189255446e-06 tokens_per_second_per_gpu:5355.495531436794 mfu:25.262786163547087 +Step 32378 | loss:0.3400290012359619 lr:2.871726250648005e-06 tokens_per_second_per_gpu:5203.937481081995 mfu:24.547860981553672 +Step 32379 | loss:0.3029059171676636 lr:2.8702780368282168e-06 tokens_per_second_per_gpu:3267.885096064504 mfu:15.415171556827284 +Step 32380 | loss:0.37800663709640503 lr:2.8688301774770833e-06 tokens_per_second_per_gpu:4942.96112189394 mfu:23.316790968105238 +Step 32381 | loss:0.32502204179763794 lr:2.8673826726055008e-06 tokens_per_second_per_gpu:5345.314876702776 mfu:25.21476227816791 +Step 32382 | loss:0.3035595417022705 lr:2.8659355222243446e-06 tokens_per_second_per_gpu:4736.532112792492 mfu:22.343030111752938 +Step 32383 | loss:0.2173376828432083 lr:2.864488726344494e-06 tokens_per_second_per_gpu:4544.04951374482 mfu:21.43505685112723 +Step 32384 | loss:0.290004700422287 lr:2.863042284976841e-06 tokens_per_second_per_gpu:4220.006829568167 mfu:19.906492222482978 +Step 32385 | loss:0.28208327293395996 lr:2.8615961981322657e-06 tokens_per_second_per_gpu:5322.032632984318 mfu:25.10493596218755 +Step 32386 | loss:0.23533186316490173 lr:2.8601504658216316e-06 tokens_per_second_per_gpu:4247.058720667892 mfu:20.03410060832894 +Step 32387 | loss:0.30804961919784546 lr:2.8587050880558242e-06 tokens_per_second_per_gpu:3990.6600862477217 mfu:18.82462450365101 +Step 32388 | loss:0.41738831996917725 lr:2.8572600648457126e-06 tokens_per_second_per_gpu:5252.636271745826 mfu:24.77758148598543 +Step 32389 | loss:0.3580111563205719 lr:2.8558153962021772e-06 tokens_per_second_per_gpu:5352.488655334674 mfu:25.24860221594742 +Step 32390 | loss:0.291429340839386 lr:2.8543710821360535e-06 tokens_per_second_per_gpu:4321.163503730439 mfu:20.38366551361473 +Step 32391 | loss:0.3750138282775879 lr:2.8529271226582223e-06 tokens_per_second_per_gpu:3811.6326673661065 mfu:17.980121623559047 +Step 32392 | loss:0.32291579246520996 lr:2.851483517779541e-06 tokens_per_second_per_gpu:4569.130717537449 mfu:21.55336916871169 +Step 32393 | loss:0.37776100635528564 lr:2.850040267510873e-06 tokens_per_second_per_gpu:4566.949329116573 mfu:21.543079187347175 +Step 32394 | loss:0.29752758145332336 lr:2.8485973718630656e-06 tokens_per_second_per_gpu:3520.9087614353675 mfu:16.60872735667054 +Step 32395 | loss:0.37982577085494995 lr:2.847154830846971e-06 tokens_per_second_per_gpu:5026.5625076511815 mfu:23.7111529685894 +Step 32396 | loss:0.4311743676662445 lr:2.8457126444734582e-06 tokens_per_second_per_gpu:4656.144054118046 mfu:21.96382592336846 +Step 32397 | loss:0.33773282170295715 lr:2.844270812753347e-06 tokens_per_second_per_gpu:4032.751996237725 mfu:19.023179224694246 +Step 32398 | loss:0.4554494023323059 lr:2.8428293356974833e-06 tokens_per_second_per_gpu:4269.182359886306 mfu:20.138461589204283 +Step 32399 | loss:0.21639752388000488 lr:2.841388213316726e-06 tokens_per_second_per_gpu:3503.845493828025 mfu:16.52823701207305 +Step 32400 | loss:0.28761205077171326 lr:2.8399474456219045e-06 tokens_per_second_per_gpu:3607.4839893505014 mfu:17.01711747800327 +Step 32401 | loss:0.33194389939308167 lr:2.838507032623855e-06 tokens_per_second_per_gpu:4242.907009287809 mfu:20.014516277394087 +Step 32402 | loss:0.4151662290096283 lr:2.8370669743334076e-06 tokens_per_second_per_gpu:4139.056671225458 mfu:19.524636514059207 +Step 32403 | loss:0.31406983733177185 lr:2.8356272707614094e-06 tokens_per_second_per_gpu:3601.733014677001 mfu:16.98998914925091 +Step 32404 | loss:0.2741950750350952 lr:2.834187921918674e-06 tokens_per_second_per_gpu:3758.4492173550434 mfu:17.729246215824606 +Step 32405 | loss:0.37198513746261597 lr:2.8327489278160202e-06 tokens_per_second_per_gpu:4517.8989202891 mfu:21.31170004004527 +Step 32406 | loss:0.33404844999313354 lr:2.831310288464284e-06 tokens_per_second_per_gpu:3590.466264076742 mfu:16.936841964363616 +Step 32407 | loss:0.3742216229438782 lr:2.82987200387429e-06 tokens_per_second_per_gpu:4302.067556992621 mfu:20.293586674747964 +Step 32408 | loss:0.2983814477920532 lr:2.828434074056835e-06 tokens_per_second_per_gpu:4761.685869148893 mfu:22.46168467215946 +Step 32409 | loss:0.23721355199813843 lr:2.8269964990227495e-06 tokens_per_second_per_gpu:5305.85310600032 mfu:25.028614372892072 +Step 32410 | loss:0.3248881995677948 lr:2.8255592787828524e-06 tokens_per_second_per_gpu:5349.139384650726 mfu:25.232803134686733 +Step 32411 | loss:0.2974326014518738 lr:2.8241224133479405e-06 tokens_per_second_per_gpu:2801.3357607096154 mfu:13.214378740433414 +Step 32412 | loss:0.26894572377204895 lr:2.8226859027288168e-06 tokens_per_second_per_gpu:5317.138966747521 mfu:25.08185170360302 +Step 32413 | loss:0.2858249843120575 lr:2.8212497469362886e-06 tokens_per_second_per_gpu:5069.92408488317 mfu:23.915696926640646 +Step 32414 | loss:0.3117724359035492 lr:2.8198139459811756e-06 tokens_per_second_per_gpu:4576.782048133469 mfu:21.589461800584733 +Step 32415 | loss:0.28400832414627075 lr:2.8183784998742524e-06 tokens_per_second_per_gpu:4455.246293183279 mfu:21.016156908346648 +Step 32416 | loss:0.3881377577781677 lr:2.816943408626321e-06 tokens_per_second_per_gpu:3704.042957730356 mfu:17.472602606509863 +Step 32417 | loss:0.285312682390213 lr:2.8155086722481905e-06 tokens_per_second_per_gpu:3758.5274289361273 mfu:17.72961515319685 +Step 32418 | loss:0.31957271695137024 lr:2.81407429075064e-06 tokens_per_second_per_gpu:4269.473935697682 mfu:20.139837002054506 +Step 32419 | loss:0.24179260432720184 lr:2.812640264144445e-06 tokens_per_second_per_gpu:4837.583045360737 mfu:22.81970460174429 +Step 32420 | loss:0.34866848587989807 lr:2.8112065924404072e-06 tokens_per_second_per_gpu:3654.7187156268274 mfu:17.239931740924032 +Step 32421 | loss:0.2921126186847687 lr:2.8097732756493133e-06 tokens_per_second_per_gpu:4801.671837769033 mfu:22.650305308450406 +Step 32422 | loss:0.34973788261413574 lr:2.808340313781932e-06 tokens_per_second_per_gpu:5357.650586161874 mfu:25.27295192437554 +Step 32423 | loss:0.3099222183227539 lr:2.8069077068490433e-06 tokens_per_second_per_gpu:5311.075948869581 mfu:25.05325141381526 +Step 32424 | loss:0.3068179488182068 lr:2.8054754548614225e-06 tokens_per_second_per_gpu:4790.10397801926 mfu:22.59573774033033 +Step 32425 | loss:0.2990192770957947 lr:2.8040435578298552e-06 tokens_per_second_per_gpu:5365.0655466111675 mfu:25.307929557937566 +Step 32426 | loss:0.36320871114730835 lr:2.802612015765088e-06 tokens_per_second_per_gpu:4098.6414066973375 mfu:19.33399081572489 +Step 32427 | loss:0.2767774164676666 lr:2.801180828677896e-06 tokens_per_second_per_gpu:4915.088197244534 mfu:23.18530962692266 +Step 32428 | loss:0.2780193090438843 lr:2.7997499965790543e-06 tokens_per_second_per_gpu:5359.558870493371 mfu:25.281953627154092 +Step 32429 | loss:0.3110784590244293 lr:2.798319519479309e-06 tokens_per_second_per_gpu:4474.061087781229 mfu:21.104909504600155 +Step 32430 | loss:0.29091766476631165 lr:2.7968893973894186e-06 tokens_per_second_per_gpu:5312.101658974481 mfu:25.058089863384264 +Step 32431 | loss:0.3502251207828522 lr:2.795459630320152e-06 tokens_per_second_per_gpu:5080.375047633713 mfu:23.964995901053076 +Step 32432 | loss:0.3941151797771454 lr:2.794030218282273e-06 tokens_per_second_per_gpu:4975.117642379756 mfu:23.468478761704976 +Step 32433 | loss:0.29621562361717224 lr:2.792601161286501e-06 tokens_per_second_per_gpu:3599.6399683271907 mfu:16.980115892508323 +Step 32434 | loss:0.2896142303943634 lr:2.7911724593435995e-06 tokens_per_second_per_gpu:5351.539531465751 mfu:25.244125036720483 +Step 32435 | loss:0.29060307145118713 lr:2.7897441124643097e-06 tokens_per_second_per_gpu:4357.748729063912 mfu:20.55624426359583 +Step 32436 | loss:0.37690818309783936 lr:2.78831612065939e-06 tokens_per_second_per_gpu:5354.860572680904 mfu:25.259790954760415 +Step 32437 | loss:0.31834399700164795 lr:2.7868884839395592e-06 tokens_per_second_per_gpu:3534.222969253848 mfu:16.67153274658841 +Step 32438 | loss:0.32342082262039185 lr:2.7854612023155646e-06 tokens_per_second_per_gpu:4104.06807451346 mfu:19.35958933369854 +Step 32439 | loss:0.37125852704048157 lr:2.7840342757981476e-06 tokens_per_second_per_gpu:5334.685921614012 mfu:25.16462368352747 +Step 32440 | loss:0.3479357659816742 lr:2.782607704398038e-06 tokens_per_second_per_gpu:4404.450608821552 mfu:20.776544998574956 +Step 32441 | loss:0.36036136746406555 lr:2.7811814881259503e-06 tokens_per_second_per_gpu:3487.6969417342575 mfu:16.45206153661965 +Step 32442 | loss:0.34610092639923096 lr:2.7797556269926195e-06 tokens_per_second_per_gpu:5023.048233139685 mfu:23.694575536121615 +Step 32443 | loss:0.42001691460609436 lr:2.778330121008782e-06 tokens_per_second_per_gpu:5374.062641218311 mfu:25.350370388262093 +Step 32444 | loss:0.4078846871852875 lr:2.776904970185146e-06 tokens_per_second_per_gpu:4335.011275012417 mfu:20.448987813425234 +Step 32445 | loss:0.3558572828769684 lr:2.7754801745324244e-06 tokens_per_second_per_gpu:5219.903493916966 mfu:24.62317538817891 +Step 32446 | loss:0.32357555627822876 lr:2.7740557340613595e-06 tokens_per_second_per_gpu:3952.8779408714136 mfu:18.646399677612244 +Step 32447 | loss:0.23071345686912537 lr:2.772631648782642e-06 tokens_per_second_per_gpu:5338.603132487549 mfu:25.18310183556304 +Step 32448 | loss:0.33139026165008545 lr:2.7712079187069807e-06 tokens_per_second_per_gpu:5323.8109033947185 mfu:25.113324367117627 +Step 32449 | loss:0.32760199904441833 lr:2.7697845438450887e-06 tokens_per_second_per_gpu:4839.091689205469 mfu:22.826821132161076 +Step 32450 | loss:0.3502703607082367 lr:2.7683615242076855e-06 tokens_per_second_per_gpu:5071.357587173814 mfu:23.92245900152633 +Step 32451 | loss:0.3173513114452362 lr:2.7669388598054514e-06 tokens_per_second_per_gpu:4356.123265301294 mfu:20.548576673694843 +Step 32452 | loss:0.25188884139060974 lr:2.7655165506490997e-06 tokens_per_second_per_gpu:4997.480325895407 mfu:23.573967355315546 +Step 32453 | loss:0.4032919108867645 lr:2.764094596749334e-06 tokens_per_second_per_gpu:5331.00409355269 mfu:25.147255872377574 +Step 32454 | loss:0.25665411353111267 lr:2.762672998116839e-06 tokens_per_second_per_gpu:4959.034837325045 mfu:23.39261342625241 +Step 32455 | loss:0.23188266158103943 lr:2.761251754762295e-06 tokens_per_second_per_gpu:4582.923862241907 mfu:21.618433785635776 +Step 32456 | loss:0.33092546463012695 lr:2.759830866696411e-06 tokens_per_second_per_gpu:3953.532844111234 mfu:18.649488967932488 +Step 32457 | loss:0.25323399901390076 lr:2.758410333929873e-06 tokens_per_second_per_gpu:5368.174659956897 mfu:25.322595776058808 +Step 32458 | loss:0.2772151529788971 lr:2.756990156473349e-06 tokens_per_second_per_gpu:5145.5507660014655 mfu:24.27244088471799 +Step 32459 | loss:0.2488236129283905 lr:2.7555703343375317e-06 tokens_per_second_per_gpu:4433.14188830292 mfu:20.91188665912563 +Step 32460 | loss:0.2503066956996918 lr:2.7541508675331005e-06 tokens_per_second_per_gpu:3065.1253756025694 mfu:14.458719361032768 +Step 32461 | loss:0.3953078091144562 lr:2.752731756070742e-06 tokens_per_second_per_gpu:4801.016022255662 mfu:22.64721171478017 +Step 32462 | loss:0.297676146030426 lr:2.751312999961103e-06 tokens_per_second_per_gpu:4512.728679589462 mfu:21.287311132531055 +Step 32463 | loss:0.25475630164146423 lr:2.749894599214864e-06 tokens_per_second_per_gpu:4582.829751514235 mfu:21.6179898492764 +Step 32464 | loss:0.2222667932510376 lr:2.7484765538427105e-06 tokens_per_second_per_gpu:5340.335752196621 mfu:25.191274898346137 +Step 32465 | loss:0.3289022147655487 lr:2.747058863855284e-06 tokens_per_second_per_gpu:3393.2777862292182 mfu:16.006670270533284 +Step 32466 | loss:0.3536727726459503 lr:2.7456415292632542e-06 tokens_per_second_per_gpu:4421.329867365705 mfu:20.856167341026197 +Step 32467 | loss:0.26011204719543457 lr:2.7442245500772844e-06 tokens_per_second_per_gpu:5391.652809186982 mfu:25.433346211762444 +Step 32468 | loss:0.3516172468662262 lr:2.7428079263080498e-06 tokens_per_second_per_gpu:4952.073634682112 mfu:23.35977624568205 +Step 32469 | loss:0.25705572962760925 lr:2.7413916579661747e-06 tokens_per_second_per_gpu:5354.140025914214 mfu:25.256392012723985 +Step 32470 | loss:0.26344189047813416 lr:2.7399757450623175e-06 tokens_per_second_per_gpu:4261.3089048142765 mfu:20.101321158279564 +Step 32471 | loss:0.3898940086364746 lr:2.7385601876071363e-06 tokens_per_second_per_gpu:3553.4535811300725 mfu:16.76224682388928 +Step 32472 | loss:0.2903960347175598 lr:2.7371449856112787e-06 tokens_per_second_per_gpu:4930.75004449206 mfu:23.259189232575928 +Step 32473 | loss:0.33918997645378113 lr:2.7357301390853797e-06 tokens_per_second_per_gpu:5140.20402245891 mfu:24.247219383179235 +Step 32474 | loss:0.2513994574546814 lr:2.7343156480400865e-06 tokens_per_second_per_gpu:4260.611113072957 mfu:20.09802955558021 +Step 32475 | loss:0.2714128792285919 lr:2.732901512486041e-06 tokens_per_second_per_gpu:5216.274173532077 mfu:24.60605526469748 +Step 32476 | loss:0.40373504161834717 lr:2.731487732433874e-06 tokens_per_second_per_gpu:5229.185624583578 mfu:24.666960782228116 +Step 32477 | loss:0.28976523876190186 lr:2.730074307894215e-06 tokens_per_second_per_gpu:4399.170064213566 mfu:20.75163576870527 +Step 32478 | loss:0.3091266453266144 lr:2.7286612388776943e-06 tokens_per_second_per_gpu:3887.4209737208544 mfu:18.33762799545284 +Step 32479 | loss:0.28047576546669006 lr:2.7272485253949597e-06 tokens_per_second_per_gpu:4076.011461292672 mfu:19.227241502184256 +Step 32480 | loss:0.39103296399116516 lr:2.725836167456608e-06 tokens_per_second_per_gpu:3647.9538497428853 mfu:17.208020714344336 +Step 32481 | loss:0.2806060314178467 lr:2.7244241650732747e-06 tokens_per_second_per_gpu:4596.1197619634995 mfu:21.680681096075183 +Step 32482 | loss:0.30309706926345825 lr:2.7230125182555853e-06 tokens_per_second_per_gpu:3570.2678664140026 mfu:16.84156267638647 +Step 32483 | loss:0.3074772357940674 lr:2.7216012270141533e-06 tokens_per_second_per_gpu:5362.737513425594 mfu:25.296947828197794 +Step 32484 | loss:0.3339826464653015 lr:2.7201902913595865e-06 tokens_per_second_per_gpu:4584.8655474866355 mfu:21.627593046220824 +Step 32485 | loss:0.3004191815853119 lr:2.7187797113024995e-06 tokens_per_second_per_gpu:4776.433245573637 mfu:22.531250562916853 +Step 32486 | loss:0.30832579731941223 lr:2.7173694868535103e-06 tokens_per_second_per_gpu:5348.852380471848 mfu:25.231449287007006 +Step 32487 | loss:0.3417412340641022 lr:2.715959618023212e-06 tokens_per_second_per_gpu:5347.692845420174 mfu:25.22597956233118 +Step 32488 | loss:0.3038037419319153 lr:2.714550104822211e-06 tokens_per_second_per_gpu:5345.097489497366 mfu:25.21373682562991 +Step 32489 | loss:0.3169093728065491 lr:2.7131409472611224e-06 tokens_per_second_per_gpu:5304.963912399955 mfu:25.024419894967078 +Step 32490 | loss:0.23061618208885193 lr:2.7117321453505372e-06 tokens_per_second_per_gpu:5361.394827943438 mfu:25.2906141516927 +Step 32491 | loss:0.395975261926651 lr:2.710323699101036e-06 tokens_per_second_per_gpu:5057.362235449317 mfu:23.856440539588647 +Step 32492 | loss:0.2519785463809967 lr:2.7089156085232325e-06 tokens_per_second_per_gpu:4090.0825287300086 mfu:19.293617128057836 +Step 32493 | loss:0.34953898191452026 lr:2.707507873627707e-06 tokens_per_second_per_gpu:5115.657324584686 mfu:24.131428421208 +Step 32494 | loss:0.2984139919281006 lr:2.7061004944250455e-06 tokens_per_second_per_gpu:3958.9799574816657 mfu:18.67518393107419 +Step 32495 | loss:0.22945742309093475 lr:2.70469347092584e-06 tokens_per_second_per_gpu:4398.109942622389 mfu:20.746634994284296 +Step 32496 | loss:0.23050355911254883 lr:2.7032868031406753e-06 tokens_per_second_per_gpu:5375.689063771778 mfu:25.3580425009431 +Step 32497 | loss:0.2725647985935211 lr:2.7018804910801276e-06 tokens_per_second_per_gpu:4269.151731521684 mfu:20.138317109982626 +Step 32498 | loss:0.3609355688095093 lr:2.70047453475476e-06 tokens_per_second_per_gpu:5328.691873428457 mfu:25.136348735546136 +Step 32499 | loss:0.33999618887901306 lr:2.699068934175164e-06 tokens_per_second_per_gpu:3955.8392780932213 mfu:18.66036881054374 +Step 32500 | loss:0.3439401090145111 lr:2.697663689351909e-06 tokens_per_second_per_gpu:5356.287469590374 mfu:25.266521870936106 +Step 32501 | loss:0.36878788471221924 lr:2.6962588002955535e-06 tokens_per_second_per_gpu:5204.177538727093 mfu:24.548993374423652 +Step 32502 | loss:0.42224493622779846 lr:2.6948542670166777e-06 tokens_per_second_per_gpu:4763.134013967849 mfu:22.46851581834946 +Step 32503 | loss:0.3880327343940735 lr:2.693450089525834e-06 tokens_per_second_per_gpu:5333.906897266602 mfu:25.16094889276548 +Step 32504 | loss:0.3223876655101776 lr:2.6920462678336033e-06 tokens_per_second_per_gpu:4969.2433556792475 mfu:23.440768749082014 +Step 32505 | loss:0.241871178150177 lr:2.690642801950516e-06 tokens_per_second_per_gpu:5207.844841433107 mfu:24.566292666994283 +Step 32506 | loss:0.3890772759914398 lr:2.689239691887141e-06 tokens_per_second_per_gpu:5348.4331524651325 mfu:25.229471716971652 +Step 32507 | loss:0.2776649296283722 lr:2.6878369376540314e-06 tokens_per_second_per_gpu:5292.51603983289 mfu:24.965701156242172 +Step 32508 | loss:0.3162955939769745 lr:2.6864345392617453e-06 tokens_per_second_per_gpu:5314.835872034779 mfu:25.07098761289454 +Step 32509 | loss:0.3482484221458435 lr:2.685032496720813e-06 tokens_per_second_per_gpu:5329.221152554828 mfu:25.138845435489557 +Step 32510 | loss:0.27032724022865295 lr:2.6836308100417873e-06 tokens_per_second_per_gpu:4285.201027372399 mfu:20.214024376802687 +Step 32511 | loss:0.24088047444820404 lr:2.682229479235221e-06 tokens_per_second_per_gpu:2996.829779471661 mfu:14.136557381653013 +Step 32512 | loss:0.2777763605117798 lr:2.6808285043116445e-06 tokens_per_second_per_gpu:5366.580362262466 mfu:25.31507520181894 +Step 32513 | loss:0.3084869682788849 lr:2.679427885281588e-06 tokens_per_second_per_gpu:5333.739977715311 mfu:25.16016150476263 +Step 32514 | loss:0.33998098969459534 lr:2.6780276221555932e-06 tokens_per_second_per_gpu:5338.933311493224 mfu:25.184659346267154 +Step 32515 | loss:0.26968762278556824 lr:2.6766277149441964e-06 tokens_per_second_per_gpu:4276.50429616242 mfu:20.173000411869864 +Step 32516 | loss:0.31284981966018677 lr:2.6752281636579168e-06 tokens_per_second_per_gpu:5342.066513709887 mfu:25.199439195702578 +Step 32517 | loss:0.43763431906700134 lr:2.673828968307279e-06 tokens_per_second_per_gpu:5338.832096347789 mfu:25.184181897157057 +Step 32518 | loss:0.3347495496273041 lr:2.6724301289028254e-06 tokens_per_second_per_gpu:4103.015747428282 mfu:19.354625327292247 +Step 32519 | loss:0.3426859378814697 lr:2.671031645455063e-06 tokens_per_second_per_gpu:4764.3110498218975 mfu:22.474068097296033 +Step 32520 | loss:0.24369587004184723 lr:2.6696335179745014e-06 tokens_per_second_per_gpu:4617.066724499835 mfu:21.77949149227942 +Step 32521 | loss:0.2650008201599121 lr:2.6682357464716644e-06 tokens_per_second_per_gpu:4350.843842737646 mfu:20.523672736699744 +Step 32522 | loss:0.2667763829231262 lr:2.6668383309570777e-06 tokens_per_second_per_gpu:5337.6061892819 mfu:25.178399084365942 +Step 32523 | loss:0.27109554409980774 lr:2.665441271441227e-06 tokens_per_second_per_gpu:3746.609586633414 mfu:17.673396657661378 +Step 32524 | loss:0.3162294626235962 lr:2.664044567934637e-06 tokens_per_second_per_gpu:5330.313317736279 mfu:25.143997364993055 +Step 32525 | loss:0.3723719120025635 lr:2.662648220447811e-06 tokens_per_second_per_gpu:5255.764447800844 mfu:24.79233762615882 +Step 32526 | loss:0.309306800365448 lr:2.661252228991251e-06 tokens_per_second_per_gpu:3953.473044945921 mfu:18.64920688506692 +Step 32527 | loss:0.3964146375656128 lr:2.6598565935754493e-06 tokens_per_second_per_gpu:4007.56405882693 mfu:18.904363426422258 +Step 32528 | loss:0.32944604754447937 lr:2.658461314210903e-06 tokens_per_second_per_gpu:5100.133257529427 mfu:24.058198748229383 +Step 32529 | loss:0.27037313580513 lr:2.657066390908114e-06 tokens_per_second_per_gpu:3897.9216068002615 mfu:18.387161273281148 +Step 32530 | loss:0.2645465135574341 lr:2.655671823677569e-06 tokens_per_second_per_gpu:4167.291421785989 mfu:19.657824649798428 +Step 32531 | loss:0.26332321763038635 lr:2.654277612529754e-06 tokens_per_second_per_gpu:5350.8590539428415 mfu:25.24091510814348 +Step 32532 | loss:0.3399290144443512 lr:2.6528837574751663e-06 tokens_per_second_per_gpu:5117.155656004937 mfu:24.138496306159027 +Step 32533 | loss:0.3166445195674896 lr:2.6514902585242807e-06 tokens_per_second_per_gpu:3706.8064251708647 mfu:17.48563835392291 +Step 32534 | loss:0.30270376801490784 lr:2.650097115687572e-06 tokens_per_second_per_gpu:5392.451839590248 mfu:25.43711537450329 +Step 32535 | loss:0.37217575311660767 lr:2.6487043289755263e-06 tokens_per_second_per_gpu:4654.088817897834 mfu:21.95413102345826 +Step 32536 | loss:0.31031790375709534 lr:2.647311898398619e-06 tokens_per_second_per_gpu:4908.502957812388 mfu:23.154245928963523 +Step 32537 | loss:0.2393612563610077 lr:2.6459198239673188e-06 tokens_per_second_per_gpu:4535.826214802582 mfu:21.39626614697722 +Step 32538 | loss:0.29187092185020447 lr:2.644528105692096e-06 tokens_per_second_per_gpu:5339.117988916417 mfu:25.185530501181574 +Step 32539 | loss:0.3721475601196289 lr:2.643136743583424e-06 tokens_per_second_per_gpu:4204.249712868956 mfu:19.832163214571466 +Step 32540 | loss:0.302742063999176 lr:2.6417457376517684e-06 tokens_per_second_per_gpu:5310.410354826998 mfu:25.05011169315434 +Step 32541 | loss:0.34694692492485046 lr:2.640355087907581e-06 tokens_per_second_per_gpu:5323.710674850943 mfu:25.112851571975888 +Step 32542 | loss:0.3211705982685089 lr:2.638964794361326e-06 tokens_per_second_per_gpu:5320.184237173671 mfu:25.096216763779882 +Step 32543 | loss:0.23640038073062897 lr:2.637574857023456e-06 tokens_per_second_per_gpu:3662.3674480396207 mfu:17.276012116723493 +Step 32544 | loss:0.2116381973028183 lr:2.6361852759044346e-06 tokens_per_second_per_gpu:3941.8413696193843 mfu:18.59433828798319 +Step 32545 | loss:0.249215766787529 lr:2.634796051014704e-06 tokens_per_second_per_gpu:3771.601465364285 mfu:17.791287613689104 +Step 32546 | loss:0.3536492586135864 lr:2.6334071823647168e-06 tokens_per_second_per_gpu:4746.0410707710735 mfu:22.387885488933435 +Step 32547 | loss:0.3976290225982666 lr:2.6320186699649253e-06 tokens_per_second_per_gpu:4660.743366755912 mfu:21.985521665804928 +Step 32548 | loss:0.3536754548549652 lr:2.6306305138257605e-06 tokens_per_second_per_gpu:3813.3179768322116 mfu:17.98807151585411 +Step 32549 | loss:0.2511061429977417 lr:2.6292427139576635e-06 tokens_per_second_per_gpu:4154.631059724749 mfu:19.598103561875053 +Step 32550 | loss:0.43850356340408325 lr:2.6278552703710767e-06 tokens_per_second_per_gpu:4931.732473721006 mfu:23.263823518868577 +Step 32551 | loss:0.34273722767829895 lr:2.626468183076447e-06 tokens_per_second_per_gpu:4322.86677798604 mfu:20.391700148886958 +Step 32552 | loss:0.30178171396255493 lr:2.6250814520841825e-06 tokens_per_second_per_gpu:4430.104272998284 mfu:20.897557709463044 +Step 32553 | loss:0.3187159597873688 lr:2.6236950774047253e-06 tokens_per_second_per_gpu:5040.152638661113 mfu:23.775259935278697 +Step 32554 | loss:0.2333332747220993 lr:2.622309059048511e-06 tokens_per_second_per_gpu:5367.632549953242 mfu:25.320038550678063 +Step 32555 | loss:0.19982390105724335 lr:2.6209233970259593e-06 tokens_per_second_per_gpu:3826.409092054384 mfu:18.049824539931674 +Step 32556 | loss:0.23523353040218353 lr:2.619538091347473e-06 tokens_per_second_per_gpu:4611.815114042693 mfu:21.75471874973594 +Step 32557 | loss:0.25780147314071655 lr:2.618153142023494e-06 tokens_per_second_per_gpu:4892.444636172914 mfu:23.078496085957934 +Step 32558 | loss:0.2929752767086029 lr:2.6167685490644357e-06 tokens_per_second_per_gpu:4152.185820700968 mfu:19.586568952199077 +Step 32559 | loss:0.27593544125556946 lr:2.6153843124807008e-06 tokens_per_second_per_gpu:4529.2707846013345 mfu:21.365343064246204 +Step 32560 | loss:0.3096684515476227 lr:2.6140004322827038e-06 tokens_per_second_per_gpu:5375.973210207508 mfu:25.359382868160814 +Step 32561 | loss:0.35681772232055664 lr:2.612616908480869e-06 tokens_per_second_per_gpu:3825.6215696259424 mfu:18.04610966227153 +Step 32562 | loss:0.35618144273757935 lr:2.6112337410855838e-06 tokens_per_second_per_gpu:4090.350761868672 mfu:19.294882429538205 +Step 32563 | loss:0.24086523056030273 lr:2.6098509301072495e-06 tokens_per_second_per_gpu:5342.843667379527 mfu:25.20310516215836 +Step 32564 | loss:0.23687197268009186 lr:2.608468475556275e-06 tokens_per_second_per_gpu:3195.5578789194915 mfu:15.073991733258524 +Step 32565 | loss:0.29295796155929565 lr:2.607086377443063e-06 tokens_per_second_per_gpu:4312.135774947508 mfu:20.341080176656288 +Step 32566 | loss:0.33741942048072815 lr:2.6057046357779945e-06 tokens_per_second_per_gpu:5335.743341594268 mfu:25.169611713988793 +Step 32567 | loss:0.31186649203300476 lr:2.604323250571472e-06 tokens_per_second_per_gpu:5362.720976272273 mfu:25.296869819624053 +Step 32568 | loss:0.3362899124622345 lr:2.6029422218338873e-06 tokens_per_second_per_gpu:4339.075405865984 mfu:20.468159012073734 +Step 32569 | loss:0.30067646503448486 lr:2.6015615495756207e-06 tokens_per_second_per_gpu:3858.9169230921393 mfu:18.203169525344666 +Step 32570 | loss:0.28864821791648865 lr:2.600181233807053e-06 tokens_per_second_per_gpu:4166.648375031935 mfu:19.654791288544466 +Step 32571 | loss:0.24221614003181458 lr:2.598801274538565e-06 tokens_per_second_per_gpu:4411.635895582655 mfu:20.810439222163414 +Step 32572 | loss:0.3864997625350952 lr:2.5974216717805534e-06 tokens_per_second_per_gpu:5208.619381225226 mfu:24.569946303343823 +Step 32573 | loss:0.2803657054901123 lr:2.596042425543377e-06 tokens_per_second_per_gpu:4963.5305937549665 mfu:23.41382067638749 +Step 32574 | loss:0.3713424503803253 lr:2.5946635358374105e-06 tokens_per_second_per_gpu:5047.042440106724 mfu:23.807760304218984 +Step 32575 | loss:0.2710939049720764 lr:2.5932850026730293e-06 tokens_per_second_per_gpu:5354.93165774115 mfu:25.260126275117962 +Step 32576 | loss:0.25265148282051086 lr:2.5919068260606028e-06 tokens_per_second_per_gpu:4485.664572592596 mfu:21.15964512221403 +Step 32577 | loss:0.2648369073867798 lr:2.5905290060105002e-06 tokens_per_second_per_gpu:5378.993523174451 mfu:25.373630199744415 +Step 32578 | loss:0.33101099729537964 lr:2.589151542533069e-06 tokens_per_second_per_gpu:3582.63548919772 mfu:16.8999028631913 +Step 32579 | loss:0.25663965940475464 lr:2.587774435638679e-06 tokens_per_second_per_gpu:4742.207730392353 mfu:22.369802968331705 +Step 32580 | loss:0.38902747631073 lr:2.5863976853376992e-06 tokens_per_second_per_gpu:5336.805597441467 mfu:25.174622556059568 +Step 32581 | loss:0.2586718499660492 lr:2.5850212916404603e-06 tokens_per_second_per_gpu:4394.870341665488 mfu:20.731353243837674 +Step 32582 | loss:0.3075793981552124 lr:2.5836452545573263e-06 tokens_per_second_per_gpu:5351.8320535956045 mfu:25.24550491351642 +Step 32583 | loss:0.26505646109580994 lr:2.582269574098656e-06 tokens_per_second_per_gpu:5056.234072168972 mfu:23.85111880091075 +Step 32584 | loss:0.3306320309638977 lr:2.5808942502747845e-06 tokens_per_second_per_gpu:4288.229432562526 mfu:20.22830988078372 +Step 32585 | loss:0.43123939633369446 lr:2.5795192830960546e-06 tokens_per_second_per_gpu:4754.47146080038 mfu:22.427653077074535 +Step 32586 | loss:0.28491270542144775 lr:2.5781446725728075e-06 tokens_per_second_per_gpu:4177.688019054721 mfu:19.706867172957445 +Step 32587 | loss:0.34070491790771484 lr:2.576770418715396e-06 tokens_per_second_per_gpu:3462.72144873562 mfu:16.334247874886564 +Step 32588 | loss:0.20949840545654297 lr:2.5753965215341347e-06 tokens_per_second_per_gpu:4817.152727144188 mfu:22.7233313049451 +Step 32589 | loss:0.40323391556739807 lr:2.57402298103937e-06 tokens_per_second_per_gpu:4741.750314382658 mfu:22.36764525897075 +Step 32590 | loss:0.3316153585910797 lr:2.572649797241433e-06 tokens_per_second_per_gpu:5128.473706905536 mfu:24.19188548331576 +Step 32591 | loss:0.3061782717704773 lr:2.5712769701506545e-06 tokens_per_second_per_gpu:3994.7439687452043 mfu:18.84388887417384 +Step 32592 | loss:0.3401256203651428 lr:2.5699044997773425e-06 tokens_per_second_per_gpu:4857.1484538280265 mfu:22.911997971687406 +Step 32593 | loss:0.31480830907821655 lr:2.5685323861318335e-06 tokens_per_second_per_gpu:4789.683612705863 mfu:22.593754805425323 +Step 32594 | loss:0.24462871253490448 lr:2.5671606292244523e-06 tokens_per_second_per_gpu:5342.724058915167 mfu:25.202540948624662 +Step 32595 | loss:0.24718241393566132 lr:2.5657892290654963e-06 tokens_per_second_per_gpu:3503.732916397985 mfu:16.527705965127858 +Step 32596 | loss:0.3016248643398285 lr:2.564418185665296e-06 tokens_per_second_per_gpu:5295.844519247227 mfu:24.98140216153675 +Step 32597 | loss:0.28704994916915894 lr:2.563047499034166e-06 tokens_per_second_per_gpu:3799.126490290638 mfu:17.921127852520222 +Step 32598 | loss:0.36071905493736267 lr:2.561677169182408e-06 tokens_per_second_per_gpu:4819.856078377658 mfu:22.736083473952664 +Step 32599 | loss:0.33362382650375366 lr:2.5603071961203262e-06 tokens_per_second_per_gpu:4052.31350434777 mfu:19.115454195986768 +Step 32600 | loss:0.246517613530159 lr:2.5589375798582225e-06 tokens_per_second_per_gpu:5336.092752980414 mfu:25.171259947112972 +Step 32601 | loss:0.27480432391166687 lr:2.5575683204064116e-06 tokens_per_second_per_gpu:4816.461538232593 mfu:22.720070848920116 +Step 32602 | loss:0.3302473723888397 lr:2.5561994177751737e-06 tokens_per_second_per_gpu:5301.657533422115 mfu:25.008823141202186 +Step 32603 | loss:0.36744293570518494 lr:2.554830871974817e-06 tokens_per_second_per_gpu:4041.657565647893 mfu:19.06518831504883 +Step 32604 | loss:0.29535818099975586 lr:2.5534626830156392e-06 tokens_per_second_per_gpu:3481.961330560247 mfu:16.425005680115667 +Step 32605 | loss:0.4262385964393616 lr:2.5520948509079103e-06 tokens_per_second_per_gpu:4137.88881615409 mfu:19.519127542430944 +Step 32606 | loss:0.3406609296798706 lr:2.550727375661943e-06 tokens_per_second_per_gpu:5291.119347232656 mfu:24.959112718946535 +Step 32607 | loss:0.3004414141178131 lr:2.5493602572880025e-06 tokens_per_second_per_gpu:4301.152678639004 mfu:20.289271037459613 +Step 32608 | loss:0.3602352440357208 lr:2.54799349579638e-06 tokens_per_second_per_gpu:5317.411648361085 mfu:25.083137989298482 +Step 32609 | loss:0.3179481029510498 lr:2.546627091197351e-06 tokens_per_second_per_gpu:4276.362978860468 mfu:20.172333794045336 +Step 32610 | loss:0.30522990226745605 lr:2.545261043501196e-06 tokens_per_second_per_gpu:4485.174531411437 mfu:21.157333514352853 +Step 32611 | loss:0.31772997975349426 lr:2.543895352718184e-06 tokens_per_second_per_gpu:4144.65812920948 mfu:19.55105954707712 +Step 32612 | loss:0.2834755480289459 lr:2.5425300188585967e-06 tokens_per_second_per_gpu:4259.908879154259 mfu:20.094716998372256 +Step 32613 | loss:0.34430602192878723 lr:2.5411650419327028e-06 tokens_per_second_per_gpu:4520.729436810099 mfu:21.32505206941074 +Step 32614 | loss:0.4003138542175293 lr:2.53980042195075e-06 tokens_per_second_per_gpu:5384.331466580993 mfu:25.398810189540132 +Step 32615 | loss:0.23823124170303345 lr:2.5384361589230133e-06 tokens_per_second_per_gpu:4319.17506556968 mfu:20.374285711547895 +Step 32616 | loss:0.34968993067741394 lr:2.5370722528597625e-06 tokens_per_second_per_gpu:3911.321777204476 mfu:18.450372163382657 +Step 32617 | loss:0.2662883400917053 lr:2.5357087037712447e-06 tokens_per_second_per_gpu:5168.4517157323835 mfu:24.38046857190366 +Step 32618 | loss:0.2868850827217102 lr:2.5343455116677127e-06 tokens_per_second_per_gpu:4393.321449678779 mfu:20.724046856068142 +Step 32619 | loss:0.2946256101131439 lr:2.532982676559437e-06 tokens_per_second_per_gpu:4206.632540478571 mfu:19.84340341895834 +Step 32620 | loss:0.2831365168094635 lr:2.5316201984566522e-06 tokens_per_second_per_gpu:4164.043936003746 mfu:19.64250570528538 +Step 32621 | loss:0.32675325870513916 lr:2.5302580773696016e-06 tokens_per_second_per_gpu:5283.031879697787 mfu:24.92096275472031 +Step 32622 | loss:0.29358845949172974 lr:2.5288963133085374e-06 tokens_per_second_per_gpu:5213.40873786776 mfu:24.592538515775157 +Step 32623 | loss:0.2931162416934967 lr:2.527534906283713e-06 tokens_per_second_per_gpu:5305.125753414957 mfu:25.02518332664777 +Step 32624 | loss:0.2888752222061157 lr:2.5261738563053415e-06 tokens_per_second_per_gpu:3149.981499038843 mfu:14.85900017322925 +Step 32625 | loss:0.31772297620773315 lr:2.5248131633836826e-06 tokens_per_second_per_gpu:4042.9016189259105 mfu:19.07105672661861 +Step 32626 | loss:0.3745805323123932 lr:2.5234528275289604e-06 tokens_per_second_per_gpu:5373.001528082639 mfu:25.345364936556557 +Step 32627 | loss:0.2949904203414917 lr:2.5220928487514117e-06 tokens_per_second_per_gpu:4635.826480516696 mfu:21.86798446215538 +Step 32628 | loss:0.3841862976551056 lr:2.520733227061256e-06 tokens_per_second_per_gpu:5315.398510051119 mfu:25.07364167241356 +Step 32629 | loss:0.20672088861465454 lr:2.5193739624687186e-06 tokens_per_second_per_gpu:3366.8734703333735 mfu:15.882116607411747 +Step 32630 | loss:0.24025844037532806 lr:2.518015054984041e-06 tokens_per_second_per_gpu:4026.7181954267726 mfu:18.994716756796407 +Step 32631 | loss:0.33542191982269287 lr:2.5166565046174207e-06 tokens_per_second_per_gpu:3716.4843215038645 mfu:17.531290642145052 +Step 32632 | loss:0.3499082922935486 lr:2.515298311379083e-06 tokens_per_second_per_gpu:3963.157011206305 mfu:18.69488780617217 +Step 32633 | loss:0.32864153385162354 lr:2.5139404752792585e-06 tokens_per_second_per_gpu:5344.688051753873 mfu:25.21180543793657 +Step 32634 | loss:0.3185463547706604 lr:2.512582996328139e-06 tokens_per_second_per_gpu:4272.989071464885 mfu:20.156418497212407 +Step 32635 | loss:0.3617058992385864 lr:2.5112258745359443e-06 tokens_per_second_per_gpu:4441.620452664954 mfu:20.951881493813595 +Step 32636 | loss:0.33996573090553284 lr:2.509869109912877e-06 tokens_per_second_per_gpu:4426.009613674124 mfu:20.878242502810064 +Step 32637 | loss:0.28856542706489563 lr:2.508512702469151e-06 tokens_per_second_per_gpu:4949.220721757645 mfu:23.3463185686602 +Step 32638 | loss:0.28411221504211426 lr:2.507156652214948e-06 tokens_per_second_per_gpu:5135.353405460575 mfu:24.224338195197493 +Step 32639 | loss:0.25308758020401 lr:2.505800959160487e-06 tokens_per_second_per_gpu:4749.437922926446 mfu:22.40390902011307 +Step 32640 | loss:0.18593202531337738 lr:2.5044456233159595e-06 tokens_per_second_per_gpu:5333.15342328668 mfu:25.15739462744627 +Step 32641 | loss:0.302089124917984 lr:2.5030906446915526e-06 tokens_per_second_per_gpu:4369.266114860767 mfu:20.610573737468243 +Step 32642 | loss:0.3616589903831482 lr:2.5017360232974687e-06 tokens_per_second_per_gpu:5384.495005766734 mfu:25.399581631781935 +Step 32643 | loss:0.20265229046344757 lr:2.5003817591438828e-06 tokens_per_second_per_gpu:3816.672353882561 mfu:18.003894684716347 +Step 32644 | loss:0.2927875816822052 lr:2.4990278522409927e-06 tokens_per_second_per_gpu:4644.675834827821 mfu:21.90972837629649 +Step 32645 | loss:0.40676987171173096 lr:2.497674302598968e-06 tokens_per_second_per_gpu:5315.4408258908525 mfu:25.073841283449408 +Step 32646 | loss:0.28285694122314453 lr:2.496321110227995e-06 tokens_per_second_per_gpu:4905.628981316635 mfu:23.140688891482103 +Step 32647 | loss:0.26455938816070557 lr:2.494968275138254e-06 tokens_per_second_per_gpu:3663.498510643114 mfu:17.281347532003352 +Step 32648 | loss:0.3466416299343109 lr:2.4936157973399268e-06 tokens_per_second_per_gpu:3711.651091213938 mfu:17.50849147023335 +Step 32649 | loss:0.24651244282722473 lr:2.4922636768431707e-06 tokens_per_second_per_gpu:3096.07950302813 mfu:14.60473526141787 +Step 32650 | loss:0.2000228464603424 lr:2.4909119136581616e-06 tokens_per_second_per_gpu:5365.911352685242 mfu:25.31191936949837 +Step 32651 | loss:0.2695733904838562 lr:2.489560507795069e-06 tokens_per_second_per_gpu:3802.177020158096 mfu:17.93551772237909 +Step 32652 | loss:0.28183844685554504 lr:2.4882094592640514e-06 tokens_per_second_per_gpu:4113.039657522257 mfu:19.40190982146094 +Step 32653 | loss:0.3004189133644104 lr:2.4868587680752676e-06 tokens_per_second_per_gpu:4825.706730961026 mfu:22.763682000412672 +Step 32654 | loss:0.6119699478149414 lr:2.4855084342388866e-06 tokens_per_second_per_gpu:5075.18646140355 mfu:23.940520454541304 +Step 32655 | loss:0.3191818594932556 lr:2.4841584577650624e-06 tokens_per_second_per_gpu:5337.860574294646 mfu:25.17959906187362 +Step 32656 | loss:0.3283311724662781 lr:2.482808838663947e-06 tokens_per_second_per_gpu:3995.5225227880283 mfu:18.84756144642886 +Step 32657 | loss:0.29714736342430115 lr:2.481459576945683e-06 tokens_per_second_per_gpu:4176.185642760808 mfu:19.699780207647226 +Step 32658 | loss:0.22193914651870728 lr:2.4801106726204283e-06 tokens_per_second_per_gpu:5045.460546913986 mfu:23.80029824413003 +Step 32659 | loss:0.20031119883060455 lr:2.478762125698325e-06 tokens_per_second_per_gpu:4619.001604364309 mfu:21.788618650724686 +Step 32660 | loss:0.3161774277687073 lr:2.4774139361895155e-06 tokens_per_second_per_gpu:4028.642793224068 mfu:19.003795412976498 +Step 32661 | loss:0.2755914330482483 lr:2.4760661041041357e-06 tokens_per_second_per_gpu:4425.722592646328 mfu:20.876888575651115 +Step 32662 | loss:0.3368818163871765 lr:2.474718629452333e-06 tokens_per_second_per_gpu:5323.218774841028 mfu:25.11053119570232 +Step 32663 | loss:0.3225177526473999 lr:2.473371512244238e-06 tokens_per_second_per_gpu:4657.177329698438 mfu:21.968700060576165 +Step 32664 | loss:0.3349573612213135 lr:2.472024752489971e-06 tokens_per_second_per_gpu:3973.9728410223215 mfu:18.74590792078517 +Step 32665 | loss:0.3916100263595581 lr:2.470678350199673e-06 tokens_per_second_per_gpu:4267.095289642479 mfu:20.12861652277325 +Step 32666 | loss:0.3282772898674011 lr:2.4693323053834704e-06 tokens_per_second_per_gpu:4374.2905782526705 mfu:20.634274988549702 +Step 32667 | loss:0.3050280809402466 lr:2.467986618051482e-06 tokens_per_second_per_gpu:5330.867231429674 mfu:25.14661027039222 +Step 32668 | loss:0.3889124095439911 lr:2.466641288213828e-06 tokens_per_second_per_gpu:4408.113560909418 mfu:20.79382376853749 +Step 32669 | loss:0.3359898328781128 lr:2.4652963158806443e-06 tokens_per_second_per_gpu:5382.702365495576 mfu:25.391125441766913 +Step 32670 | loss:0.31388166546821594 lr:2.4639517010620173e-06 tokens_per_second_per_gpu:3805.09917254874 mfu:17.949302013776364 +Step 32671 | loss:0.28716421127319336 lr:2.4626074437680835e-06 tokens_per_second_per_gpu:4910.013273183016 mfu:23.161370344253154 +Step 32672 | loss:0.394011914730072 lr:2.4612635440089404e-06 tokens_per_second_per_gpu:4731.562700435705 mfu:22.319588545796798 +Step 32673 | loss:0.3333951532840729 lr:2.459920001794708e-06 tokens_per_second_per_gpu:4290.188279696286 mfu:20.23755009692745 +Step 32674 | loss:0.26827967166900635 lr:2.458576817135477e-06 tokens_per_second_per_gpu:5318.185290198405 mfu:25.08678738984203 +Step 32675 | loss:0.3062906861305237 lr:2.457233990041352e-06 tokens_per_second_per_gpu:4492.860497948995 mfu:21.193589529871627 +Step 32676 | loss:0.36963900923728943 lr:2.4558915205224463e-06 tokens_per_second_per_gpu:3832.4259409489355 mfu:18.07820704274755 +Step 32677 | loss:0.3330085873603821 lr:2.454549408588841e-06 tokens_per_second_per_gpu:3702.0419016506025 mfu:17.463163283566296 +Step 32678 | loss:0.32067975401878357 lr:2.453207654250639e-06 tokens_per_second_per_gpu:5320.138985170028 mfu:25.09600330235791 +Step 32679 | loss:0.3150794506072998 lr:2.4518662575179264e-06 tokens_per_second_per_gpu:5318.149676493638 mfu:25.08661939392032 +Step 32680 | loss:0.29690980911254883 lr:2.450525218400801e-06 tokens_per_second_per_gpu:4984.499892148039 mfu:23.512736434638843 +Step 32681 | loss:0.3434178829193115 lr:2.4491845369093327e-06 tokens_per_second_per_gpu:4583.643537744711 mfu:21.6218286177718 +Step 32682 | loss:0.2308274805545807 lr:2.4478442130536183e-06 tokens_per_second_per_gpu:5166.2303617256 mfu:24.36999006605159 +Step 32683 | loss:0.4130929112434387 lr:2.4465042468437394e-06 tokens_per_second_per_gpu:4449.047247820885 mfu:20.986914953706034 +Step 32684 | loss:0.3124243915081024 lr:2.445164638289771e-06 tokens_per_second_per_gpu:5332.971696403891 mfu:25.15653739073435 +Step 32685 | loss:0.37294802069664 lr:2.443825387401788e-06 tokens_per_second_per_gpu:3637.446176412595 mfu:17.15845422645142 +Step 32686 | loss:0.37770137190818787 lr:2.442486494189855e-06 tokens_per_second_per_gpu:4881.709240535392 mfu:23.027855393088373 +Step 32687 | loss:0.26887860894203186 lr:2.4411479586640585e-06 tokens_per_second_per_gpu:5356.8491821158605 mfu:25.26917156475686 +Step 32688 | loss:0.3573978841304779 lr:2.439809780834451e-06 tokens_per_second_per_gpu:4652.308539738501 mfu:21.945733147633913 +Step 32689 | loss:0.28470656275749207 lr:2.4384719607110972e-06 tokens_per_second_per_gpu:4578.670859175916 mfu:21.598371644536346 +Step 32690 | loss:0.3188991844654083 lr:2.4371344983040665e-06 tokens_per_second_per_gpu:4829.338486414685 mfu:22.780813610528956 +Step 32691 | loss:0.22731691598892212 lr:2.435797393623429e-06 tokens_per_second_per_gpu:4812.814053496918 mfu:22.70286504109687 +Step 32692 | loss:0.39565861225128174 lr:2.434460646679221e-06 tokens_per_second_per_gpu:5058.707963489125 mfu:23.8627885683573 +Step 32693 | loss:0.2385118156671524 lr:2.4331242574815004e-06 tokens_per_second_per_gpu:5383.21507917164 mfu:25.393543999654884 +Step 32694 | loss:0.30877187848091125 lr:2.431788226040327e-06 tokens_per_second_per_gpu:5350.182493432338 mfu:25.237723656782304 +Step 32695 | loss:0.3266759216785431 lr:2.4304525523657417e-06 tokens_per_second_per_gpu:4313.367254111346 mfu:20.34688927398412 +Step 32696 | loss:0.3287549316883087 lr:2.4291172364677926e-06 tokens_per_second_per_gpu:4749.651510189821 mfu:22.40491654767616 +Step 32697 | loss:0.36958223581314087 lr:2.4277822783565163e-06 tokens_per_second_per_gpu:3732.9605169006863 mfu:17.60901166696067 +Step 32698 | loss:0.285629540681839 lr:2.426447678041971e-06 tokens_per_second_per_gpu:4762.181460183403 mfu:22.464022459625795 +Step 32699 | loss:0.3284291923046112 lr:2.4251134355341767e-06 tokens_per_second_per_gpu:3964.564963646423 mfu:18.701529358053662 +Step 32700 | loss:0.28860220313072205 lr:2.4237795508431803e-06 tokens_per_second_per_gpu:4514.119598505191 mfu:21.29387233437178 +Step 32701 | loss:0.2957182228565216 lr:2.4224460239790024e-06 tokens_per_second_per_gpu:4416.805922290169 mfu:20.834827120240274 +Step 32702 | loss:0.36345192790031433 lr:2.421112854951685e-06 tokens_per_second_per_gpu:3631.4005962689544 mfu:17.129936193431533 +Step 32703 | loss:0.3415806293487549 lr:2.419780043771247e-06 tokens_per_second_per_gpu:4119.20823633799 mfu:19.431008060202824 +Step 32704 | loss:0.26753106713294983 lr:2.418447590447709e-06 tokens_per_second_per_gpu:4756.7196778824955 mfu:22.438258300636264 +Step 32705 | loss:0.34182655811309814 lr:2.4171154949911067e-06 tokens_per_second_per_gpu:5353.7558648565555 mfu:25.254579859470375 +Step 32706 | loss:0.4002194106578827 lr:2.4157837574114438e-06 tokens_per_second_per_gpu:5341.905155572947 mfu:25.198678041840093 +Step 32707 | loss:0.36596059799194336 lr:2.4144523777187454e-06 tokens_per_second_per_gpu:4043.472418601578 mfu:19.073749286077206 +Step 32708 | loss:0.3210909962654114 lr:2.41312135592302e-06 tokens_per_second_per_gpu:5263.7788744720765 mfu:24.830143044169212 +Step 32709 | loss:0.32069507241249084 lr:2.411790692034288e-06 tokens_per_second_per_gpu:5309.780326248456 mfu:25.047139740855396 +Step 32710 | loss:0.26135289669036865 lr:2.4104603860625463e-06 tokens_per_second_per_gpu:5314.5305576446435 mfu:25.069547392824262 +Step 32711 | loss:0.24752019345760345 lr:2.4091304380177983e-06 tokens_per_second_per_gpu:4586.675505544878 mfu:21.636130927192248 +Step 32712 | loss:0.24526099860668182 lr:2.4078008479100635e-06 tokens_per_second_per_gpu:3828.009688395698 mfu:18.057374825963546 +Step 32713 | loss:0.22490772604942322 lr:2.406471615749323e-06 tokens_per_second_per_gpu:5326.604701023563 mfu:25.126503187203916 +Step 32714 | loss:0.3445374667644501 lr:2.405142741545591e-06 tokens_per_second_per_gpu:4770.8574407013175 mfu:22.504948540003717 +Step 32715 | loss:0.3513627350330353 lr:2.403814225308848e-06 tokens_per_second_per_gpu:3883.4429560550006 mfu:18.31886300740321 +Step 32716 | loss:0.2849975526332855 lr:2.402486067049092e-06 tokens_per_second_per_gpu:5336.834360492744 mfu:25.174758236279924 +Step 32717 | loss:0.31882399320602417 lr:2.40115826677631e-06 tokens_per_second_per_gpu:4371.385074302775 mfu:20.620569230687956 +Step 32718 | loss:0.26612767577171326 lr:2.3998308245004875e-06 tokens_per_second_per_gpu:3937.028642424575 mfu:18.571635832669347 +Step 32719 | loss:0.34850889444351196 lr:2.3985037402316224e-06 tokens_per_second_per_gpu:4847.400126813859 mfu:22.86601344992548 +Step 32720 | loss:0.3477374315261841 lr:2.3971770139796733e-06 tokens_per_second_per_gpu:4353.662175532124 mfu:20.536967293348457 +Step 32721 | loss:0.37567993998527527 lr:2.395850645754638e-06 tokens_per_second_per_gpu:5343.98166325525 mfu:25.20847328286626 +Step 32722 | loss:0.3240740895271301 lr:2.3945246355664806e-06 tokens_per_second_per_gpu:4170.122262156534 mfu:19.67117820681746 +Step 32723 | loss:0.2978793978691101 lr:2.393198983425182e-06 tokens_per_second_per_gpu:5275.91665264851 mfu:24.887399014745952 +Step 32724 | loss:0.2900111675262451 lr:2.3918736893407013e-06 tokens_per_second_per_gpu:4600.535341432461 mfu:21.701510137805816 +Step 32725 | loss:0.24497391283512115 lr:2.3905487533230185e-06 tokens_per_second_per_gpu:4001.261075604833 mfu:18.8746311791638 +Step 32726 | loss:0.3345428705215454 lr:2.3892241753820877e-06 tokens_per_second_per_gpu:4905.099827172155 mfu:23.138192781099583 +Step 32727 | loss:0.2756820023059845 lr:2.3878999555278836e-06 tokens_per_second_per_gpu:5342.66777713251 mfu:25.202275457853208 +Step 32728 | loss:0.3403402268886566 lr:2.386576093770354e-06 tokens_per_second_per_gpu:5325.907663305491 mfu:25.12321514136087 +Step 32729 | loss:0.3523330092430115 lr:2.385252590119469e-06 tokens_per_second_per_gpu:4403.463324331815 mfu:20.77188780918876 +Step 32730 | loss:0.32766425609588623 lr:2.38392944458517e-06 tokens_per_second_per_gpu:4311.042907843798 mfu:20.33592493606573 +Step 32731 | loss:0.26497119665145874 lr:2.3826066571774163e-06 tokens_per_second_per_gpu:5367.604990155885 mfu:25.319908546411746 +Step 32732 | loss:0.4299851953983307 lr:2.38128422790615e-06 tokens_per_second_per_gpu:4211.466234250127 mfu:19.866204777191964 +Step 32733 | loss:0.3248133957386017 lr:2.379962156781318e-06 tokens_per_second_per_gpu:4207.037526328928 mfu:19.84531380631218 +Step 32734 | loss:0.26006200909614563 lr:2.378640443812874e-06 tokens_per_second_per_gpu:4637.725977176255 mfu:21.876944711986066 +Step 32735 | loss:0.2951321005821228 lr:2.377319089010749e-06 tokens_per_second_per_gpu:5321.256078413917 mfu:25.101272821785095 +Step 32736 | loss:0.3247396945953369 lr:2.37599809238489e-06 tokens_per_second_per_gpu:4642.782044834881 mfu:21.900795045785905 +Step 32737 | loss:0.24822397530078888 lr:2.3746774539452178e-06 tokens_per_second_per_gpu:5356.429888623709 mfu:25.26719368581554 +Step 32738 | loss:0.36102941632270813 lr:2.373357173701679e-06 tokens_per_second_per_gpu:5042.85679605971 mfu:23.788015907104707 +Step 32739 | loss:0.30379921197891235 lr:2.3720372516641886e-06 tokens_per_second_per_gpu:4435.965559693903 mfu:20.92520639884491 +Step 32740 | loss:0.3593544065952301 lr:2.3707176878426884e-06 tokens_per_second_per_gpu:4325.878498126428 mfu:20.40590694664161 +Step 32741 | loss:0.19420179724693298 lr:2.369398482247104e-06 tokens_per_second_per_gpu:4642.922778309602 mfu:21.901458909597768 +Step 32742 | loss:0.34685927629470825 lr:2.3680796348873435e-06 tokens_per_second_per_gpu:5142.143208254622 mfu:24.256366853436823 +Step 32743 | loss:0.4278853237628937 lr:2.366761145773344e-06 tokens_per_second_per_gpu:5299.186619914207 mfu:24.997167420604065 +Step 32744 | loss:0.3917039930820465 lr:2.365443014915003e-06 tokens_per_second_per_gpu:4019.8891149894066 mfu:18.962502819212084 +Step 32745 | loss:0.36641573905944824 lr:2.364125242322246e-06 tokens_per_second_per_gpu:5334.189589588588 mfu:25.162282400680574 +Step 32746 | loss:0.27824267745018005 lr:2.3628078280049815e-06 tokens_per_second_per_gpu:5352.127064802135 mfu:25.246896531711727 +Step 32747 | loss:0.41026923060417175 lr:2.361490771973113e-06 tokens_per_second_per_gpu:4923.205669864642 mfu:23.223601130255357 +Step 32748 | loss:0.31048887968063354 lr:2.36017407423656e-06 tokens_per_second_per_gpu:4555.115870104598 mfu:21.48725874219139 +Step 32749 | loss:0.29480940103530884 lr:2.3588577348052088e-06 tokens_per_second_per_gpu:4704.9046274289 mfu:22.19383786708744 +Step 32750 | loss:0.2525193393230438 lr:2.3575417536889745e-06 tokens_per_second_per_gpu:5353.640849513904 mfu:25.254037312475447 +Step 32751 | loss:0.33212757110595703 lr:2.356226130897743e-06 tokens_per_second_per_gpu:4891.026874703948 mfu:23.07180826321377 +Step 32752 | loss:0.25079014897346497 lr:2.3549108664414176e-06 tokens_per_second_per_gpu:4212.086253121135 mfu:19.86912951199264 +Step 32753 | loss:0.22732339799404144 lr:2.353595960329885e-06 tokens_per_second_per_gpu:5171.915931752687 mfu:24.396809869927637 +Step 32754 | loss:0.34428870677948 lr:2.352281412573032e-06 tokens_per_second_per_gpu:5259.524061222552 mfu:24.810072364125208 +Step 32755 | loss:0.22975529730319977 lr:2.3509672231807613e-06 tokens_per_second_per_gpu:3271.2406021474135 mfu:15.431000051528741 +Step 32756 | loss:0.26648589968681335 lr:2.349653392162937e-06 tokens_per_second_per_gpu:3739.4676843031834 mfu:17.63970708583549 +Step 32757 | loss:0.2688122093677521 lr:2.348339919529452e-06 tokens_per_second_per_gpu:5353.775890485287 mfu:25.254674323777827 +Step 32758 | loss:0.4236406981945038 lr:2.3470268052901866e-06 tokens_per_second_per_gpu:4976.41849828209 mfu:23.474615120945185 +Step 32759 | loss:0.4167729914188385 lr:2.345714049455011e-06 tokens_per_second_per_gpu:4209.615588525098 mfu:19.857474965554285 +Step 32760 | loss:0.3401877284049988 lr:2.344401652033801e-06 tokens_per_second_per_gpu:4988.172661459467 mfu:23.53006151411991 +Step 32761 | loss:0.3202780485153198 lr:2.34308961303642e-06 tokens_per_second_per_gpu:5318.532560073879 mfu:25.08842552109499 +Step 32762 | loss:0.28565680980682373 lr:2.3417779324727496e-06 tokens_per_second_per_gpu:4812.483918725564 mfu:22.701307739053593 +Step 32763 | loss:0.27855899930000305 lr:2.340466610352654e-06 tokens_per_second_per_gpu:4968.791106382515 mfu:23.43863541198764 +Step 32764 | loss:0.23062758147716522 lr:2.339155646685981e-06 tokens_per_second_per_gpu:3957.119120756021 mfu:18.66640604674784 +Step 32765 | loss:0.29476237297058105 lr:2.337845041482606e-06 tokens_per_second_per_gpu:4167.846423664409 mfu:19.66044268835202 +Step 32766 | loss:0.28435632586479187 lr:2.3365347947523764e-06 tokens_per_second_per_gpu:5067.3280172065615 mfu:23.90345083247527 +Step 32767 | loss:0.3611299693584442 lr:2.335224906505157e-06 tokens_per_second_per_gpu:4405.617272153877 mfu:20.782048348566825 +Step 32768 | loss:0.25420859456062317 lr:2.333915376750784e-06 tokens_per_second_per_gpu:3735.4254620207985 mfu:17.620639233655243 +Step 32769 | loss:0.2797780930995941 lr:2.332606205499116e-06 tokens_per_second_per_gpu:4201.798500115794 mfu:19.820600425794783 +Step 32770 | loss:0.25439316034317017 lr:2.331297392760007e-06 tokens_per_second_per_gpu:5371.977568665434 mfu:25.34053474528011 +Step 32771 | loss:0.24835677444934845 lr:2.329988938543287e-06 tokens_per_second_per_gpu:4407.869033815979 mfu:20.792670292517112 +Step 32772 | loss:0.37193071842193604 lr:2.32868084285881e-06 tokens_per_second_per_gpu:5181.795475201236 mfu:24.443413361998367 +Step 32773 | loss:0.31485459208488464 lr:2.327373105716396e-06 tokens_per_second_per_gpu:4083.265933144425 mfu:19.261462083649505 +Step 32774 | loss:0.3336895704269409 lr:2.3260657271259033e-06 tokens_per_second_per_gpu:4554.98565779385 mfu:21.48664450850508 +Step 32775 | loss:0.26041972637176514 lr:2.3247587070971466e-06 tokens_per_second_per_gpu:4050.224031815961 mfu:19.105597797553447 +Step 32776 | loss:0.34021154046058655 lr:2.323452045639962e-06 tokens_per_second_per_gpu:3434.9653090599495 mfu:16.20331742835067 +Step 32777 | loss:0.25591161847114563 lr:2.322145742764181e-06 tokens_per_second_per_gpu:3625.760665638952 mfu:17.10333167837842 +Step 32778 | loss:0.22746102511882782 lr:2.3208397984796235e-06 tokens_per_second_per_gpu:4067.541174128122 mfu:19.18728571244906 +Step 32779 | loss:0.2010938674211502 lr:2.3195342127961208e-06 tokens_per_second_per_gpu:4775.640679809276 mfu:22.527511894980698 +Step 32780 | loss:0.3919644057750702 lr:2.318228985723475e-06 tokens_per_second_per_gpu:5342.255614564347 mfu:25.200331216696995 +Step 32781 | loss:0.2300666868686676 lr:2.316924117271524e-06 tokens_per_second_per_gpu:5281.765237942103 mfu:24.914987789447384 +Step 32782 | loss:0.3143574893474579 lr:2.3156196074500646e-06 tokens_per_second_per_gpu:4497.50673775044 mfu:21.215506635745303 +Step 32783 | loss:0.27078571915626526 lr:2.314315456268912e-06 tokens_per_second_per_gpu:4806.474827246117 mfu:22.67296182928831 +Step 32784 | loss:0.24825331568717957 lr:2.3130116637378853e-06 tokens_per_second_per_gpu:3360.2477301050717 mfu:15.850861860286996 +Step 32785 | loss:0.29795941710472107 lr:2.3117082298667768e-06 tokens_per_second_per_gpu:5194.082564495671 mfu:24.501373658593824 +Step 32786 | loss:0.2816064953804016 lr:2.3104051546654013e-06 tokens_per_second_per_gpu:4165.240417013336 mfu:19.648149710348324 +Step 32787 | loss:0.3124437630176544 lr:2.3091024381435456e-06 tokens_per_second_per_gpu:5079.37641725568 mfu:23.960285191175984 +Step 32788 | loss:0.33918243646621704 lr:2.3078000803110233e-06 tokens_per_second_per_gpu:3887.7161327333883 mfu:18.339020310874282 +Step 32789 | loss:0.2966804504394531 lr:2.306498081177616e-06 tokens_per_second_per_gpu:4097.540912286699 mfu:19.328799595826197 +Step 32790 | loss:0.3244975805282593 lr:2.305196440753127e-06 tokens_per_second_per_gpu:5323.242275824147 mfu:25.110642053849375 +Step 32791 | loss:0.3607611358165741 lr:2.3038951590473424e-06 tokens_per_second_per_gpu:5367.537153120157 mfu:25.319588547540025 +Step 32792 | loss:0.26262322068214417 lr:2.3025942360700437e-06 tokens_per_second_per_gpu:4258.817011753985 mfu:20.089566473553372 +Step 32793 | loss:0.3033086061477661 lr:2.3012936718310176e-06 tokens_per_second_per_gpu:5342.202960292158 mfu:25.200082837511342 +Step 32794 | loss:0.30006861686706543 lr:2.299993466340061e-06 tokens_per_second_per_gpu:4441.123934967243 mfu:20.949539335118835 +Step 32795 | loss:0.31013223528862 lr:2.298693619606934e-06 tokens_per_second_per_gpu:3911.8533876336355 mfu:18.452879860479094 +Step 32796 | loss:0.25495949387550354 lr:2.2973941316414117e-06 tokens_per_second_per_gpu:3676.6249600130363 mfu:17.34326723328448 +Step 32797 | loss:0.2751748561859131 lr:2.29609500245328e-06 tokens_per_second_per_gpu:4045.1886935114994 mfu:19.081845247653998 +Step 32798 | loss:0.19345642626285553 lr:2.294796232052304e-06 tokens_per_second_per_gpu:5391.179377298681 mfu:25.431112952769517 +Step 32799 | loss:0.3703376054763794 lr:2.2934978204482594e-06 tokens_per_second_per_gpu:5186.1655167697045 mfu:24.46402759368281 +Step 32800 | loss:0.3135961592197418 lr:2.292199767650893e-06 tokens_per_second_per_gpu:5353.503695852554 mfu:25.253390335254036 +Step 32801 | loss:0.2884661555290222 lr:2.2909020736699926e-06 tokens_per_second_per_gpu:4922.206322352881 mfu:23.218887037536113 +Step 32802 | loss:0.3229706585407257 lr:2.2896047385152943e-06 tokens_per_second_per_gpu:4759.149148853296 mfu:22.449718529716016 +Step 32803 | loss:0.29830268025398254 lr:2.2883077621965787e-06 tokens_per_second_per_gpu:4412.020527407714 mfu:20.812253596107176 +Step 32804 | loss:0.27784937620162964 lr:2.2870111447235775e-06 tokens_per_second_per_gpu:4681.806702892351 mfu:22.08488101613623 +Step 32805 | loss:0.3197299540042877 lr:2.2857148861060553e-06 tokens_per_second_per_gpu:5312.511177378283 mfu:25.060021631566038 +Step 32806 | loss:0.2743969261646271 lr:2.2844189863537647e-06 tokens_per_second_per_gpu:4387.382720495047 mfu:20.696032857256796 +Step 32807 | loss:0.36528515815734863 lr:2.283123445476443e-06 tokens_per_second_per_gpu:5354.440221474825 mfu:25.257808086402033 +Step 32808 | loss:0.25335192680358887 lr:2.2818282634838426e-06 tokens_per_second_per_gpu:4595.979025090781 mfu:21.680017216234383 +Step 32809 | loss:0.24092061817646027 lr:2.280533440385696e-06 tokens_per_second_per_gpu:4197.737780154873 mfu:19.80144531690885 +Step 32810 | loss:0.3586789667606354 lr:2.2792389761917556e-06 tokens_per_second_per_gpu:5364.849388537816 mfu:25.306909903425193 +Step 32811 | loss:0.2995724081993103 lr:2.277944870911736e-06 tokens_per_second_per_gpu:4305.237107246464 mfu:20.30853798407638 +Step 32812 | loss:0.3369768261909485 lr:2.2766511245553846e-06 tokens_per_second_per_gpu:4140.331168585414 mfu:19.530648535557795 +Step 32813 | loss:0.27073419094085693 lr:2.2753577371324384e-06 tokens_per_second_per_gpu:5362.711386862548 mfu:25.296824584742257 +Step 32814 | loss:0.3517034351825714 lr:2.274064708652607e-06 tokens_per_second_per_gpu:4767.213639381706 mfu:22.4877601074825 +Step 32815 | loss:0.3796987235546112 lr:2.2727720391256315e-06 tokens_per_second_per_gpu:4174.7696694298575 mfu:19.69310081985517 +Step 32816 | loss:0.39017438888549805 lr:2.2714797285612266e-06 tokens_per_second_per_gpu:4314.221213819265 mfu:20.350917547626928 +Step 32817 | loss:0.2234116643667221 lr:2.270187776969113e-06 tokens_per_second_per_gpu:5331.283705253539 mfu:25.148574848477153 +Step 32818 | loss:0.30189448595046997 lr:2.268896184359004e-06 tokens_per_second_per_gpu:5366.738635303255 mfu:25.31582180275651 +Step 32819 | loss:0.3936513662338257 lr:2.2676049507406148e-06 tokens_per_second_per_gpu:4240.565827927056 mfu:20.003472525468684 +Step 32820 | loss:0.30852776765823364 lr:2.2663140761236656e-06 tokens_per_second_per_gpu:4217.522187578992 mfu:19.894771742296406 +Step 32821 | loss:0.3112555146217346 lr:2.2650235605178537e-06 tokens_per_second_per_gpu:2944.5325194852044 mfu:13.88986228346425 +Step 32822 | loss:0.3005214333534241 lr:2.263733403932894e-06 tokens_per_second_per_gpu:4153.520708440282 mfu:19.592865845420757 +Step 32823 | loss:0.34494614601135254 lr:2.262443606378489e-06 tokens_per_second_per_gpu:4316.610904149075 mfu:20.362190124635855 +Step 32824 | loss:0.3400161862373352 lr:2.2611541678643377e-06 tokens_per_second_per_gpu:4997.784104628141 mfu:23.57540033142785 +Step 32825 | loss:0.30215802788734436 lr:2.2598650884001317e-06 tokens_per_second_per_gpu:3945.6522106970497 mfu:18.612314675543075 +Step 32826 | loss:0.34332340955734253 lr:2.258576367995574e-06 tokens_per_second_per_gpu:3600.028582458182 mfu:16.981949051668618 +Step 32827 | loss:0.28814682364463806 lr:2.2572880066603573e-06 tokens_per_second_per_gpu:4314.720102003478 mfu:20.35327088831096 +Step 32828 | loss:0.2637791335582733 lr:2.2560000044041627e-06 tokens_per_second_per_gpu:4871.405351541423 mfu:22.97925019067908 +Step 32829 | loss:0.31392523646354675 lr:2.254712361236688e-06 tokens_per_second_per_gpu:4828.624315632762 mfu:22.777444745100524 +Step 32830 | loss:0.320748895406723 lr:2.2534250771676203e-06 tokens_per_second_per_gpu:4173.42836475122 mfu:19.686773656835868 +Step 32831 | loss:0.31691011786460876 lr:2.2521381522066344e-06 tokens_per_second_per_gpu:4156.880442251166 mfu:19.60871428303639 +Step 32832 | loss:0.27828460931777954 lr:2.250851586363406e-06 tokens_per_second_per_gpu:5149.752477232474 mfu:24.292261073480514 +Step 32833 | loss:0.3024919927120209 lr:2.2495653796476112e-06 tokens_per_second_per_gpu:4552.311883768734 mfu:21.47403186023572 +Step 32834 | loss:0.4902474284172058 lr:2.248279532068931e-06 tokens_per_second_per_gpu:4932.010708961643 mfu:23.265136001970493 +Step 32835 | loss:0.3079698383808136 lr:2.246994043637035e-06 tokens_per_second_per_gpu:5344.850872282469 mfu:25.212573490150007 +Step 32836 | loss:0.3424301743507385 lr:2.2457089143615885e-06 tokens_per_second_per_gpu:5190.176226116792 mfu:24.4829467939706 +Step 32837 | loss:0.35765016078948975 lr:2.2444241442522663e-06 tokens_per_second_per_gpu:5317.149695597245 mfu:25.081902313417704 +Step 32838 | loss:0.23232969641685486 lr:2.2431397333187167e-06 tokens_per_second_per_gpu:5365.04752091089 mfu:25.30784452763385 +Step 32839 | loss:0.2607105076313019 lr:2.2418556815706094e-06 tokens_per_second_per_gpu:4976.856100351486 mfu:23.47667936456914 +Step 32840 | loss:0.24472220242023468 lr:2.240571989017598e-06 tokens_per_second_per_gpu:4946.866958279112 mfu:23.3352154647385 +Step 32841 | loss:0.35918375849723816 lr:2.239288655669336e-06 tokens_per_second_per_gpu:4566.5054472923985 mfu:21.5409853210485 +Step 32842 | loss:0.29704156517982483 lr:2.238005681535482e-06 tokens_per_second_per_gpu:4600.411045111355 mfu:21.70092381085202 +Step 32843 | loss:0.2320457398891449 lr:2.236723066625679e-06 tokens_per_second_per_gpu:5358.649101776986 mfu:25.277662092896353 +Step 32844 | loss:0.42288461327552795 lr:2.235440810949585e-06 tokens_per_second_per_gpu:5366.682599528411 mfu:25.315557472445878 +Step 32845 | loss:0.25353938341140747 lr:2.2341589145168264e-06 tokens_per_second_per_gpu:5348.123775723611 mfu:25.228012334096785 +Step 32846 | loss:0.4144396483898163 lr:2.2328773773370613e-06 tokens_per_second_per_gpu:3012.8040810007433 mfu:14.211910887462581 +Step 32847 | loss:0.2679409384727478 lr:2.231596199419911e-06 tokens_per_second_per_gpu:4420.576999959176 mfu:20.852615937017173 +Step 32848 | loss:0.3253074288368225 lr:2.230315380775022e-06 tokens_per_second_per_gpu:3559.3500522982044 mfu:16.79006148443108 +Step 32849 | loss:0.34563779830932617 lr:2.2290349214120376e-06 tokens_per_second_per_gpu:4315.100250799863 mfu:20.35506411504358 +Step 32850 | loss:0.3644372522830963 lr:2.2277548213405665e-06 tokens_per_second_per_gpu:4086.261636291208 mfu:19.275593326512542 +Step 32851 | loss:0.38716909289360046 lr:2.226475080570245e-06 tokens_per_second_per_gpu:5326.071452310633 mfu:25.12398776204341 +Step 32852 | loss:0.34050366282463074 lr:2.2251956991107102e-06 tokens_per_second_per_gpu:5264.853094192297 mfu:24.83521032187028 +Step 32853 | loss:0.3883429765701294 lr:2.223916676971577e-06 tokens_per_second_per_gpu:4256.233803198573 mfu:20.077381038056668 +Step 32854 | loss:0.321045458316803 lr:2.222638014162448e-06 tokens_per_second_per_gpu:4163.967299177087 mfu:19.642144196298435 +Step 32855 | loss:0.2851937711238861 lr:2.221359710692961e-06 tokens_per_second_per_gpu:5213.322570679729 mfu:24.592132050447585 +Step 32856 | loss:0.3550474941730499 lr:2.2200817665727292e-06 tokens_per_second_per_gpu:3907.6913343738615 mfu:18.43324674513323 +Step 32857 | loss:0.31838247179985046 lr:2.218804181811357e-06 tokens_per_second_per_gpu:4521.0592987460395 mfu:21.326608084442825 +Step 32858 | loss:0.34160205721855164 lr:2.2175269564184532e-06 tokens_per_second_per_gpu:4945.546753490487 mfu:23.328987833500392 +Step 32859 | loss:0.3983091413974762 lr:2.2162500904036267e-06 tokens_per_second_per_gpu:5358.460191648078 mfu:25.276770971586902 +Step 32860 | loss:0.24616138637065887 lr:2.214973583776486e-06 tokens_per_second_per_gpu:5320.982163872886 mfu:25.099980720160413 +Step 32861 | loss:0.2539733648300171 lr:2.213697436546619e-06 tokens_per_second_per_gpu:5342.606995696697 mfu:25.201988741450094 +Step 32862 | loss:0.2961385250091553 lr:2.2124216487236338e-06 tokens_per_second_per_gpu:3722.709055292743 mfu:17.56065377347629 +Step 32863 | loss:0.31165698170661926 lr:2.2111462203171285e-06 tokens_per_second_per_gpu:4961.937783358739 mfu:23.406307117987005 +Step 32864 | loss:0.15355119109153748 lr:2.209871151336679e-06 tokens_per_second_per_gpu:4413.587654757382 mfu:20.819646003195054 +Step 32865 | loss:0.22764776647090912 lr:2.208596441791888e-06 tokens_per_second_per_gpu:5367.790459225734 mfu:25.32078343565851 +Step 32866 | loss:0.35921230912208557 lr:2.2073220916923544e-06 tokens_per_second_per_gpu:5019.859737656467 mfu:23.679534858913843 +Step 32867 | loss:0.3293055295944214 lr:2.206048101047642e-06 tokens_per_second_per_gpu:4320.060362694136 mfu:20.37846180913122 +Step 32868 | loss:0.23803363740444183 lr:2.2047744698673377e-06 tokens_per_second_per_gpu:4432.864498501857 mfu:20.910578164106436 +Step 32869 | loss:0.26348233222961426 lr:2.203501198161023e-06 tokens_per_second_per_gpu:5180.449942023493 mfu:24.437066252427325 +Step 32870 | loss:0.2979739010334015 lr:2.2022282859382735e-06 tokens_per_second_per_gpu:5309.454593506802 mfu:25.04560320393728 +Step 32871 | loss:0.22076080739498138 lr:2.2009557332086704e-06 tokens_per_second_per_gpu:5358.499629591386 mfu:25.276957007094293 +Step 32872 | loss:0.32181647419929504 lr:2.1996835399817785e-06 tokens_per_second_per_gpu:4905.337198329389 mfu:23.139312501347955 +Step 32873 | loss:0.27812936902046204 lr:2.1984117062671617e-06 tokens_per_second_per_gpu:4945.33514711605 mfu:23.327989650128405 +Step 32874 | loss:0.33156076073646545 lr:2.197140232074396e-06 tokens_per_second_per_gpu:5325.179850512788 mfu:25.119781924240023 +Step 32875 | loss:0.33839961886405945 lr:2.1958691174130354e-06 tokens_per_second_per_gpu:4394.117407746522 mfu:20.727801526078135 +Step 32876 | loss:0.3693690598011017 lr:2.194598362292638e-06 tokens_per_second_per_gpu:3876.8469242266992 mfu:18.28774839987003 +Step 32877 | loss:0.3875519335269928 lr:2.1933279667227637e-06 tokens_per_second_per_gpu:5297.979831369049 mfu:24.991474793137197 +Step 32878 | loss:0.2918010950088501 lr:2.1920579307129817e-06 tokens_per_second_per_gpu:3703.838193062097 mfu:17.471636696633233 +Step 32879 | loss:0.28483960032463074 lr:2.190788254272824e-06 tokens_per_second_per_gpu:4695.325149295972 mfu:22.14864983430691 +Step 32880 | loss:0.2611834704875946 lr:2.1895189374118553e-06 tokens_per_second_per_gpu:4553.993335128429 mfu:21.481963553184585 +Step 32881 | loss:0.33888161182403564 lr:2.188249980139606e-06 tokens_per_second_per_gpu:4285.069878112535 mfu:20.21340572336753 +Step 32882 | loss:0.2046985924243927 lr:2.1869813824656415e-06 tokens_per_second_per_gpu:5334.840568480943 mfu:25.16535317918426 +Step 32883 | loss:0.3754451274871826 lr:2.1857131443994814e-06 tokens_per_second_per_gpu:5380.8810183644455 mfu:25.382533836595506 +Step 32884 | loss:0.30614033341407776 lr:2.184445265950674e-06 tokens_per_second_per_gpu:5346.587292012804 mfu:25.22076447828175 +Step 32885 | loss:0.34981364011764526 lr:2.1831777471287617e-06 tokens_per_second_per_gpu:4008.8111716730646 mfu:18.91024627049687 +Step 32886 | loss:0.3560459613800049 lr:2.1819105879432646e-06 tokens_per_second_per_gpu:5276.091642775356 mfu:24.888224473030686 +Step 32887 | loss:0.27822935581207275 lr:2.180643788403719e-06 tokens_per_second_per_gpu:4825.130985418344 mfu:22.76096611045552 +Step 32888 | loss:0.25689807534217834 lr:2.1793773485196567e-06 tokens_per_second_per_gpu:4470.217076076412 mfu:21.086776645532506 +Step 32889 | loss:0.28114983439445496 lr:2.1781112683006033e-06 tokens_per_second_per_gpu:5268.098888884919 mfu:24.850521289225057 +Step 32890 | loss:0.39182519912719727 lr:2.1768455477560678e-06 tokens_per_second_per_gpu:5349.329311901394 mfu:25.23369905393307 +Step 32891 | loss:0.36273646354675293 lr:2.175580186895582e-06 tokens_per_second_per_gpu:4698.036643781787 mfu:22.16144041642516 +Step 32892 | loss:0.3106154203414917 lr:2.1743151857286648e-06 tokens_per_second_per_gpu:5347.040299562151 mfu:25.22290139218308 +Step 32893 | loss:0.30280935764312744 lr:2.1730505442648208e-06 tokens_per_second_per_gpu:4767.910808288145 mfu:22.49104877216345 +Step 32894 | loss:0.24705183506011963 lr:2.1717862625135697e-06 tokens_per_second_per_gpu:5345.204826402142 mfu:25.214243152123306 +Step 32895 | loss:0.3084661066532135 lr:2.1705223404844155e-06 tokens_per_second_per_gpu:4169.499092002056 mfu:19.668238606874997 +Step 32896 | loss:0.3542836010456085 lr:2.1692587781868723e-06 tokens_per_second_per_gpu:4662.281334307218 mfu:21.99277652114876 +Step 32897 | loss:0.26256564259529114 lr:2.167995575630427e-06 tokens_per_second_per_gpu:4828.492484615272 mfu:22.776822875698333 +Step 32898 | loss:0.35667386651039124 lr:2.1667327328245947e-06 tokens_per_second_per_gpu:5092.69849654374 mfu:24.023127712158868 +Step 32899 | loss:0.3351544141769409 lr:2.1654702497788724e-06 tokens_per_second_per_gpu:4452.668180042891 mfu:21.00399550879198 +Step 32900 | loss:0.27215346693992615 lr:2.1642081265027424e-06 tokens_per_second_per_gpu:4623.519870573119 mfu:21.809932083327354 +Step 32901 | loss:0.3334692418575287 lr:2.1629463630057135e-06 tokens_per_second_per_gpu:3835.8278327961375 mfu:18.094254347013404 +Step 32902 | loss:0.26493093371391296 lr:2.1616849592972723e-06 tokens_per_second_per_gpu:5374.199121566506 mfu:25.351014189350476 +Step 32903 | loss:0.28608161211013794 lr:2.1604239153869e-06 tokens_per_second_per_gpu:5246.169761721598 mfu:24.747077854900766 +Step 32904 | loss:0.3048548698425293 lr:2.1591632312840836e-06 tokens_per_second_per_gpu:5212.0796993390495 mfu:24.586269214277877 +Step 32905 | loss:0.2695840299129486 lr:2.1579029069982993e-06 tokens_per_second_per_gpu:3828.5304851783176 mfu:18.059831513244255 +Step 32906 | loss:0.20798373222351074 lr:2.1566429425390335e-06 tokens_per_second_per_gpu:5334.72472404097 mfu:25.16480672119581 +Step 32907 | loss:0.2770775258541107 lr:2.1553833379157673e-06 tokens_per_second_per_gpu:4172.231059022287 mfu:19.681125761431066 +Step 32908 | loss:0.3231959342956543 lr:2.154124093137966e-06 tokens_per_second_per_gpu:4925.904261416365 mfu:23.236330847032033 +Step 32909 | loss:0.28266507387161255 lr:2.1528652082151046e-06 tokens_per_second_per_gpu:4336.975008357473 mfu:20.45825107866132 +Step 32910 | loss:0.26271265745162964 lr:2.1516066831566482e-06 tokens_per_second_per_gpu:4788.805094053721 mfu:22.589610683052427 +Step 32911 | loss:0.2847021222114563 lr:2.150348517972062e-06 tokens_per_second_per_gpu:5089.728135534378 mfu:24.009116012481865 +Step 32912 | loss:0.3182111978530884 lr:2.1490907126708036e-06 tokens_per_second_per_gpu:5225.778268054697 mfu:24.650887700126418 +Step 32913 | loss:0.3105711340904236 lr:2.147833267262345e-06 tokens_per_second_per_gpu:4823.829968463812 mfu:22.754828991504613 +Step 32914 | loss:0.3702875077724457 lr:2.1465761817561436e-06 tokens_per_second_per_gpu:4663.03072618835 mfu:21.9963115305115 +Step 32915 | loss:0.40256792306900024 lr:2.145319456161643e-06 tokens_per_second_per_gpu:5365.5633210742235 mfu:25.31027764500888 +Step 32916 | loss:0.28164881467819214 lr:2.1440630904882964e-06 tokens_per_second_per_gpu:3040.5836443974245 mfu:14.342951827686981 +Step 32917 | loss:0.19221828877925873 lr:2.1428070847455685e-06 tokens_per_second_per_gpu:5147.4186861863545 mfu:24.281252182930157 +Step 32918 | loss:0.2313574254512787 lr:2.141551438942896e-06 tokens_per_second_per_gpu:3962.93944054239 mfu:18.6938614882291 +Step 32919 | loss:0.28964027762413025 lr:2.140296153089716e-06 tokens_per_second_per_gpu:5315.851150059482 mfu:25.075776852561358 +Step 32920 | loss:0.3206617534160614 lr:2.1390412271954706e-06 tokens_per_second_per_gpu:5344.158338038385 mfu:25.209306687962137 +Step 32921 | loss:0.37125030159950256 lr:2.137786661269614e-06 tokens_per_second_per_gpu:3741.180943000696 mfu:17.647788819423734 +Step 32922 | loss:0.2607628405094147 lr:2.1365324553215603e-06 tokens_per_second_per_gpu:3595.261375924399 mfu:16.95946132507946 +Step 32923 | loss:0.32116803526878357 lr:2.135278609360758e-06 tokens_per_second_per_gpu:4490.996427913857 mfu:21.18479639347274 +Step 32924 | loss:0.290227472782135 lr:2.134025123396638e-06 tokens_per_second_per_gpu:5337.495986537489 mfu:25.17787923921803 +Step 32925 | loss:0.3509005904197693 lr:2.132771997438621e-06 tokens_per_second_per_gpu:4711.712905878319 mfu:22.22595367814543 +Step 32926 | loss:0.34217995405197144 lr:2.131519231496132e-06 tokens_per_second_per_gpu:4422.5602056088555 mfu:20.861971056436488 +Step 32927 | loss:0.3515249490737915 lr:2.1302668255785917e-06 tokens_per_second_per_gpu:4240.310866933352 mfu:20.002269830960554 +Step 32928 | loss:0.2597774267196655 lr:2.129014779695426e-06 tokens_per_second_per_gpu:5352.576088368129 mfu:25.249014652483844 +Step 32929 | loss:0.4136483371257782 lr:2.1277630938560446e-06 tokens_per_second_per_gpu:4584.745095902914 mfu:21.62702485554925 +Step 32930 | loss:0.30286866426467896 lr:2.1265117680698665e-06 tokens_per_second_per_gpu:5132.748018322938 mfu:24.212048139544375 +Step 32931 | loss:0.3425881564617157 lr:2.125260802346313e-06 tokens_per_second_per_gpu:3779.483450886762 mfu:17.82846828420403 +Step 32932 | loss:0.2681867182254791 lr:2.1240101966947754e-06 tokens_per_second_per_gpu:3769.921298760693 mfu:17.783361981153977 +Step 32933 | loss:0.3692227900028229 lr:2.1227599511246586e-06 tokens_per_second_per_gpu:4414.253517355845 mfu:20.82278699067988 +Step 32934 | loss:0.32260751724243164 lr:2.1215100656453768e-06 tokens_per_second_per_gpu:4869.968793065346 mfu:22.97247369924527 +Step 32935 | loss:0.4071342349052429 lr:2.1202605402663333e-06 tokens_per_second_per_gpu:4124.039837212247 mfu:19.45379954588281 +Step 32936 | loss:0.264764666557312 lr:2.1190113749969153e-06 tokens_per_second_per_gpu:3680.1232696923953 mfu:17.359769356915194 +Step 32937 | loss:0.3078998327255249 lr:2.117762569846521e-06 tokens_per_second_per_gpu:3316.5567460424686 mfu:15.644764034013136 +Step 32938 | loss:0.3341141641139984 lr:2.116514124824548e-06 tokens_per_second_per_gpu:4722.892487536265 mfu:22.278689672259283 +Step 32939 | loss:0.20904472470283508 lr:2.1152660399403832e-06 tokens_per_second_per_gpu:4361.864969078278 mfu:20.575661270046556 +Step 32940 | loss:0.34749504923820496 lr:2.1140183152034087e-06 tokens_per_second_per_gpu:5371.267553348775 mfu:25.337185481890327 +Step 32941 | loss:0.37008094787597656 lr:2.1127709506230107e-06 tokens_per_second_per_gpu:3371.047156662064 mfu:15.901804597928498 +Step 32942 | loss:0.36580657958984375 lr:2.1115239462085766e-06 tokens_per_second_per_gpu:3857.223191850704 mfu:18.19517990609847 +Step 32943 | loss:0.2016412615776062 lr:2.1102773019694822e-06 tokens_per_second_per_gpu:4055.2256833612887 mfu:19.129191441262513 +Step 32944 | loss:0.26107287406921387 lr:2.1090310179150974e-06 tokens_per_second_per_gpu:3600.399578297091 mfu:16.98369910233917 +Step 32945 | loss:0.36370840668678284 lr:2.1077850940548038e-06 tokens_per_second_per_gpu:4795.44731973303 mfu:22.62094319486229 +Step 32946 | loss:0.23769837617874146 lr:2.1065395303979828e-06 tokens_per_second_per_gpu:5245.440794157098 mfu:24.743639190524327 +Step 32947 | loss:0.3568832576274872 lr:2.1052943269539716e-06 tokens_per_second_per_gpu:5343.897823810055 mfu:25.208077797898795 +Step 32948 | loss:0.29592353105545044 lr:2.1040494837321568e-06 tokens_per_second_per_gpu:5320.687825381996 mfu:25.098592275279557 +Step 32949 | loss:0.2940639555454254 lr:2.1028050007418975e-06 tokens_per_second_per_gpu:5336.603623693485 mfu:25.173669811430063 +Step 32950 | loss:0.2925640344619751 lr:2.101560877992559e-06 tokens_per_second_per_gpu:4293.877190559299 mfu:20.254951318861377 +Step 32951 | loss:0.4142332375049591 lr:2.1003171154934886e-06 tokens_per_second_per_gpu:4291.22304858125 mfu:20.2424312783076 +Step 32952 | loss:0.2826789319515228 lr:2.099073713254046e-06 tokens_per_second_per_gpu:3992.866369711739 mfu:18.835031919181557 +Step 32953 | loss:0.31239229440689087 lr:2.097830671283585e-06 tokens_per_second_per_gpu:4566.2496036173025 mfu:21.539778462781637 +Step 32954 | loss:0.1868281066417694 lr:2.096587989591453e-06 tokens_per_second_per_gpu:5347.607717470208 mfu:25.22557799926686 +Step 32955 | loss:0.38473132252693176 lr:2.0953456681869878e-06 tokens_per_second_per_gpu:5327.297184794683 mfu:25.1297697513022 +Step 32956 | loss:0.31697580218315125 lr:2.0941037070795423e-06 tokens_per_second_per_gpu:5344.269823676954 mfu:25.209832584740674 +Step 32957 | loss:0.169288769364357 lr:2.0928621062784594e-06 tokens_per_second_per_gpu:4313.594524213636 mfu:20.347961345832903 +Step 32958 | loss:0.32563528418540955 lr:2.0916208657930704e-06 tokens_per_second_per_gpu:4417.723471232301 mfu:20.8391553551503 +Step 32959 | loss:0.26531022787094116 lr:2.090379985632712e-06 tokens_per_second_per_gpu:4987.11040722654 mfu:23.52505067966393 +Step 32960 | loss:0.287009060382843 lr:2.0891394658067277e-06 tokens_per_second_per_gpu:3795.4083977403093 mfu:17.90358897558832 +Step 32961 | loss:0.3549824655056 lr:2.0878993063244367e-06 tokens_per_second_per_gpu:3574.2660661564873 mfu:16.86042286673499 +Step 32962 | loss:0.3332673907279968 lr:2.0866595071951657e-06 tokens_per_second_per_gpu:3838.535698494227 mfu:18.10702781152082 +Step 32963 | loss:0.340326189994812 lr:2.08542006842824e-06 tokens_per_second_per_gpu:3991.070865589437 mfu:18.826562219892065 +Step 32964 | loss:0.3060026168823242 lr:2.0841809900329913e-06 tokens_per_second_per_gpu:4686.92173967945 mfu:22.109009517376247 +Step 32965 | loss:0.2452530562877655 lr:2.082942272018723e-06 tokens_per_second_per_gpu:4716.779256775948 mfu:22.249852519738145 +Step 32966 | loss:0.35446637868881226 lr:2.0817039143947613e-06 tokens_per_second_per_gpu:4508.481083245886 mfu:21.267274495863738 +Step 32967 | loss:0.20894630253314972 lr:2.0804659171704266e-06 tokens_per_second_per_gpu:4308.629086436639 mfu:20.32453852864776 +Step 32968 | loss:0.2541874945163727 lr:2.0792282803550165e-06 tokens_per_second_per_gpu:4559.336055909718 mfu:21.50716607867209 +Step 32969 | loss:0.3243039846420288 lr:2.0779910039578464e-06 tokens_per_second_per_gpu:5322.58922063851 mfu:25.107561481116015 +Step 32970 | loss:0.334473192691803 lr:2.076754087988214e-06 tokens_per_second_per_gpu:4573.504027331373 mfu:21.573998817173226 +Step 32971 | loss:0.3012167811393738 lr:2.07551753245544e-06 tokens_per_second_per_gpu:5360.093638071076 mfu:25.284476217058884 +Step 32972 | loss:0.4506601095199585 lr:2.074281337368805e-06 tokens_per_second_per_gpu:4967.999669703097 mfu:23.43490207013781 +Step 32973 | loss:0.34081488847732544 lr:2.0730455027376196e-06 tokens_per_second_per_gpu:5369.954686299135 mfu:25.33099246401895 +Step 32974 | loss:0.2270221710205078 lr:2.0718100285711693e-06 tokens_per_second_per_gpu:5354.035155333295 mfu:25.255897320301024 +Step 32975 | loss:0.2793999910354614 lr:2.07057491487877e-06 tokens_per_second_per_gpu:5308.105679812215 mfu:25.03924014789151 +Step 32976 | loss:0.23443862795829773 lr:2.069340161669675e-06 tokens_per_second_per_gpu:3102.351964841638 mfu:14.634323533984528 +Step 32977 | loss:0.2344108521938324 lr:2.0681057689531936e-06 tokens_per_second_per_gpu:4803.917253273733 mfu:22.660897316493465 +Step 32978 | loss:0.36962568759918213 lr:2.066871736738607e-06 tokens_per_second_per_gpu:5365.922195797108 mfu:25.311970518307067 +Step 32979 | loss:0.2575874924659729 lr:2.065638065035186e-06 tokens_per_second_per_gpu:4814.050885833633 mfu:22.708699390254516 +Step 32980 | loss:0.4623062014579773 lr:2.0644047538522226e-06 tokens_per_second_per_gpu:4963.540576803274 mfu:23.41386776812964 +Step 32981 | loss:0.340721994638443 lr:2.0631718031989823e-06 tokens_per_second_per_gpu:4522.381856285867 mfu:21.33284681400977 +Step 32982 | loss:0.35676488280296326 lr:2.0619392130847625e-06 tokens_per_second_per_gpu:5392.334638294555 mfu:25.436562515994904 +Step 32983 | loss:0.32924753427505493 lr:2.060706983518801e-06 tokens_per_second_per_gpu:4356.578848300414 mfu:20.55072573643217 +Step 32984 | loss:0.29772692918777466 lr:2.059475114510373e-06 tokens_per_second_per_gpu:5354.718761904274 mfu:25.25912200913164 +Step 32985 | loss:0.2671554386615753 lr:2.058243606068755e-06 tokens_per_second_per_gpu:5185.292703536207 mfu:24.45991038474311 +Step 32986 | loss:0.3129265010356903 lr:2.0570124582032113e-06 tokens_per_second_per_gpu:4425.798049893039 mfu:20.877244520358182 +Step 32987 | loss:0.3707612454891205 lr:2.0557816709229848e-06 tokens_per_second_per_gpu:5364.658519588364 mfu:25.306009542025027 +Step 32988 | loss:0.26305684447288513 lr:2.054551244237346e-06 tokens_per_second_per_gpu:4166.2101858548995 mfu:19.652724275433606 +Step 32989 | loss:0.2172408252954483 lr:2.0533211781555486e-06 tokens_per_second_per_gpu:5366.349868346424 mfu:25.313987922690654 +Step 32990 | loss:0.2777819335460663 lr:2.0520914726868457e-06 tokens_per_second_per_gpu:4128.587012343255 mfu:19.475249346803853 +Step 32991 | loss:0.3778098225593567 lr:2.0508621278404694e-06 tokens_per_second_per_gpu:5323.210716136955 mfu:25.11049318142025 +Step 32992 | loss:0.3799487054347992 lr:2.049633143625679e-06 tokens_per_second_per_gpu:4464.9539713420245 mfu:21.061949682522137 +Step 32993 | loss:0.19172066450119019 lr:2.048404520051722e-06 tokens_per_second_per_gpu:5169.596703691746 mfu:24.385869675463066 +Step 32994 | loss:0.3032888174057007 lr:2.047176257127825e-06 tokens_per_second_per_gpu:5358.6166003013905 mfu:25.277508777890645 +Step 32995 | loss:0.2646094262599945 lr:2.0459483548632363e-06 tokens_per_second_per_gpu:5350.538371324883 mfu:25.23940239352047 +Step 32996 | loss:0.328269362449646 lr:2.044720813267198e-06 tokens_per_second_per_gpu:4173.325496382313 mfu:19.686288409188574 +Step 32997 | loss:0.31136810779571533 lr:2.0434936323489308e-06 tokens_per_second_per_gpu:4449.491729565444 mfu:20.989011649931523 +Step 32998 | loss:0.3044402301311493 lr:2.0422668121176603e-06 tokens_per_second_per_gpu:5343.689797378368 mfu:25.207096501727555 +Step 32999 | loss:0.2875825762748718 lr:2.041040352582624e-06 tokens_per_second_per_gpu:4072.0923154185416 mfu:19.208754222420744 +Step 33000 | loss:0.2658846974372864 lr:2.039814253753042e-06 tokens_per_second_per_gpu:5332.279679428906 mfu:25.153273028593077 +Step 33001 | loss:0.2650129795074463 lr:2.0385885156381345e-06 tokens_per_second_per_gpu:4098.500864260908 mfu:19.333327853073243 +Step 33002 | loss:0.273444801568985 lr:2.0373631382471283e-06 tokens_per_second_per_gpu:5357.176618579624 mfu:25.27071613842286 +Step 33003 | loss:0.2238776534795761 lr:2.0361381215892317e-06 tokens_per_second_per_gpu:4388.163843198395 mfu:20.699717546321764 +Step 33004 | loss:0.35801973938941956 lr:2.034913465673666e-06 tokens_per_second_per_gpu:5366.75371066048 mfu:25.315892915788655 +Step 33005 | loss:0.22054798901081085 lr:2.0336891705096284e-06 tokens_per_second_per_gpu:3600.8654097891767 mfu:16.985896508966388 +Step 33006 | loss:0.2342866063117981 lr:2.0324652361063345e-06 tokens_per_second_per_gpu:4035.7889111088552 mfu:19.037504870292473 +Step 33007 | loss:0.27386054396629333 lr:2.0312416624729936e-06 tokens_per_second_per_gpu:4374.602505862524 mfu:20.63574640430572 +Step 33008 | loss:0.3155590295791626 lr:2.0300184496188037e-06 tokens_per_second_per_gpu:4864.6473324349945 mfu:22.947371461517097 +Step 33009 | loss:0.34005168080329895 lr:2.0287955975529626e-06 tokens_per_second_per_gpu:5355.668545975706 mfu:25.263602302646458 +Step 33010 | loss:0.18362365663051605 lr:2.0275731062846805e-06 tokens_per_second_per_gpu:2995.405529151944 mfu:14.129838949892747 +Step 33011 | loss:0.2590550482273102 lr:2.0263509758231436e-06 tokens_per_second_per_gpu:4115.918039229371 mfu:19.41548763907596 +Step 33012 | loss:0.35128092765808105 lr:2.025129206177534e-06 tokens_per_second_per_gpu:4357.431363180017 mfu:20.554747194573295 +Step 33013 | loss:0.4809642732143402 lr:2.02390779735705e-06 tokens_per_second_per_gpu:3277.266011701164 mfu:15.459422936434661 +Step 33014 | loss:0.379483699798584 lr:2.022686749370878e-06 tokens_per_second_per_gpu:4091.059889197802 mfu:19.29822750413931 +Step 33015 | loss:0.36463823914527893 lr:2.0214660622282e-06 tokens_per_second_per_gpu:4544.542920314132 mfu:21.43738433409851 +Step 33016 | loss:0.35142046213150024 lr:2.0202457359381976e-06 tokens_per_second_per_gpu:4839.750636712962 mfu:22.829929499981958 +Step 33017 | loss:0.3511399030685425 lr:2.019025770510047e-06 tokens_per_second_per_gpu:5335.143299085816 mfu:25.166781211098613 +Step 33018 | loss:0.31868237257003784 lr:2.0178061659529402e-06 tokens_per_second_per_gpu:4029.4214767507156 mfu:19.0074685960288 +Step 33019 | loss:0.32980436086654663 lr:2.0165869222760257e-06 tokens_per_second_per_gpu:3474.1334167206996 mfu:16.3880800749547 +Step 33020 | loss:0.34110531210899353 lr:2.01536803948848e-06 tokens_per_second_per_gpu:5049.096031120352 mfu:23.817447443409023 +Step 33021 | loss:0.33319154381752014 lr:2.014149517599473e-06 tokens_per_second_per_gpu:5300.512527123268 mfu:25.003421951132303 +Step 33022 | loss:0.1809839904308319 lr:2.012931356618181e-06 tokens_per_second_per_gpu:4305.828559575081 mfu:20.31132796562264 +Step 33023 | loss:0.23270820081233978 lr:2.011713556553746e-06 tokens_per_second_per_gpu:5276.32817110054 mfu:24.889340217496468 +Step 33024 | loss:0.40193143486976624 lr:2.010496117415339e-06 tokens_per_second_per_gpu:4634.788625577262 mfu:21.863088723329756 +Step 33025 | loss:0.4575125277042389 lr:2.0092790392121198e-06 tokens_per_second_per_gpu:4898.587003633161 mfu:23.10747068126387 +Step 33026 | loss:0.40116071701049805 lr:2.008062321953236e-06 tokens_per_second_per_gpu:4244.209433784624 mfu:20.020660035962095 +Step 33027 | loss:0.40967699885368347 lr:2.0068459656478355e-06 tokens_per_second_per_gpu:5023.130583974911 mfu:23.694963999061873 +Step 33028 | loss:0.3318043351173401 lr:2.0056299703050674e-06 tokens_per_second_per_gpu:4986.131086593068 mfu:23.520431057146517 +Step 33029 | loss:0.46794697642326355 lr:2.004414335934085e-06 tokens_per_second_per_gpu:3265.903851184826 mfu:15.40582568669461 +Step 33030 | loss:0.4291875958442688 lr:2.0031990625440255e-06 tokens_per_second_per_gpu:5396.275638718225 mfu:25.455152887397368 +Step 33031 | loss:0.38880541920661926 lr:2.001984150144026e-06 tokens_per_second_per_gpu:4692.872764377282 mfu:22.137081516225432 +Step 33032 | loss:0.3523387312889099 lr:2.000769598743235e-06 tokens_per_second_per_gpu:5379.710952323728 mfu:25.37701443544759 +Step 33033 | loss:0.31746238470077515 lr:1.9995554083507838e-06 tokens_per_second_per_gpu:5009.745225619818 mfu:23.631822979923268 +Step 33034 | loss:0.33666640520095825 lr:1.9983415789757932e-06 tokens_per_second_per_gpu:4467.961162060272 mfu:21.076135114219753 +Step 33035 | loss:0.4120447337627411 lr:1.9971281106274e-06 tokens_per_second_per_gpu:4093.4700880223045 mfu:19.309596823216808 +Step 33036 | loss:0.28989508748054504 lr:1.9959150033147356e-06 tokens_per_second_per_gpu:4247.403490908903 mfu:20.03572694837966 +Step 33037 | loss:0.3390768766403198 lr:1.9947022570469207e-06 tokens_per_second_per_gpu:4216.52904087411 mfu:19.890086899841176 +Step 33038 | loss:0.27803459763526917 lr:1.9934898718330707e-06 tokens_per_second_per_gpu:5361.1792694370415 mfu:25.289597325439953 +Step 33039 | loss:0.32149067521095276 lr:1.9922778476823166e-06 tokens_per_second_per_gpu:3572.181417302782 mfu:16.850589222414293 +Step 33040 | loss:0.21485869586467743 lr:1.9910661846037626e-06 tokens_per_second_per_gpu:4768.891780409757 mfu:22.4956761850322 +Step 33041 | loss:0.2892708480358124 lr:1.9898548826065233e-06 tokens_per_second_per_gpu:3697.2007812861266 mfu:17.440326892826754 +Step 33042 | loss:0.336433470249176 lr:1.9886439416997085e-06 tokens_per_second_per_gpu:5034.763225203472 mfu:23.74983715247025 +Step 33043 | loss:0.28004175424575806 lr:1.9874333618924377e-06 tokens_per_second_per_gpu:4896.935868566362 mfu:23.099681995441088 +Step 33044 | loss:0.26929476857185364 lr:1.986223143193805e-06 tokens_per_second_per_gpu:4945.845744623085 mfu:23.330398225686007 +Step 33045 | loss:0.22132237255573273 lr:1.985013285612908e-06 tokens_per_second_per_gpu:2955.8772967035884 mfu:13.943377533221955 +Step 33046 | loss:0.3025699257850647 lr:1.983803789158861e-06 tokens_per_second_per_gpu:3751.2471480421987 mfu:17.69527282607618 +Step 33047 | loss:0.32802313566207886 lr:1.9825946538407525e-06 tokens_per_second_per_gpu:5271.245525911854 mfu:24.86536451295275 +Step 33048 | loss:0.32353314757347107 lr:1.9813858796676745e-06 tokens_per_second_per_gpu:3546.1543439527763 mfu:16.727815076745692 +Step 33049 | loss:0.2668015956878662 lr:1.9801774666487136e-06 tokens_per_second_per_gpu:5365.739042288144 mfu:25.311106551956076 +Step 33050 | loss:0.27687591314315796 lr:1.978969414792975e-06 tokens_per_second_per_gpu:5325.039036809352 mfu:25.11911768197562 +Step 33051 | loss:0.280638724565506 lr:1.977761724109528e-06 tokens_per_second_per_gpu:3355.256572920501 mfu:15.827317720268148 +Step 33052 | loss:0.35526803135871887 lr:1.976554394607466e-06 tokens_per_second_per_gpu:4509.273079344451 mfu:21.271010476589137 +Step 33053 | loss:0.3035174012184143 lr:1.975347426295865e-06 tokens_per_second_per_gpu:5355.425667174337 mfu:25.262456601901004 +Step 33054 | loss:0.25385794043540955 lr:1.974140819183812e-06 tokens_per_second_per_gpu:4207.636819291247 mfu:19.848140773465364 +Step 33055 | loss:0.39880070090293884 lr:1.972934573280366e-06 tokens_per_second_per_gpu:5357.359042805789 mfu:25.271576664622458 +Step 33056 | loss:0.349668025970459 lr:1.971728688594604e-06 tokens_per_second_per_gpu:4266.699515031238 mfu:20.126749586405914 +Step 33057 | loss:0.23520249128341675 lr:1.9705231651356016e-06 tokens_per_second_per_gpu:3570.8956098694566 mfu:16.844523849369924 +Step 33058 | loss:0.24341163039207458 lr:1.969318002912429e-06 tokens_per_second_per_gpu:4498.224743469983 mfu:21.21889359122879 +Step 33059 | loss:0.25733649730682373 lr:1.9681132019341407e-06 tokens_per_second_per_gpu:4580.251887475525 mfu:21.60582962477685 +Step 33060 | loss:0.2357068657875061 lr:1.966908762209796e-06 tokens_per_second_per_gpu:5325.9544114549335 mfu:25.12343566035785 +Step 33061 | loss:0.29365840554237366 lr:1.9657046837484703e-06 tokens_per_second_per_gpu:4480.689872829437 mfu:21.136178614659908 +Step 33062 | loss:0.2773928642272949 lr:1.9645009665592072e-06 tokens_per_second_per_gpu:5216.777956227869 mfu:24.608431693627725 +Step 33063 | loss:0.32151857018470764 lr:1.963297610651055e-06 tokens_per_second_per_gpu:5356.517892553912 mfu:25.267608815369325 +Step 33064 | loss:0.41576629877090454 lr:1.962094616033072e-06 tokens_per_second_per_gpu:5307.369893460543 mfu:25.03576931813257 +Step 33065 | loss:0.24133896827697754 lr:1.9608919827143135e-06 tokens_per_second_per_gpu:5333.837235186598 mfu:25.160620284848527 +Step 33066 | loss:0.3335556089878082 lr:1.959689710703805e-06 tokens_per_second_per_gpu:5154.361586957423 mfu:24.314003030448312 +Step 33067 | loss:0.33018895983695984 lr:1.9584878000106055e-06 tokens_per_second_per_gpu:5009.217943868693 mfu:23.629335701939667 +Step 33068 | loss:0.33405032753944397 lr:1.957286250643753e-06 tokens_per_second_per_gpu:4835.514696775943 mfu:22.80994784857316 +Step 33069 | loss:0.29808154702186584 lr:1.956085062612278e-06 tokens_per_second_per_gpu:4488.120585618723 mfu:21.171230554697303 +Step 33070 | loss:0.3191158175468445 lr:1.954884235925214e-06 tokens_per_second_per_gpu:4093.2184407855043 mfu:19.30840976026538 +Step 33071 | loss:0.2585851848125458 lr:1.953683770591597e-06 tokens_per_second_per_gpu:5333.024417831065 mfu:25.156786086701533 +Step 33072 | loss:0.36129385232925415 lr:1.9524836666204637e-06 tokens_per_second_per_gpu:4899.741340019315 mfu:23.11291588294746 +Step 33073 | loss:0.34970611333847046 lr:1.951283924020819e-06 tokens_per_second_per_gpu:4679.72523054773 mfu:22.075062355950852 +Step 33074 | loss:0.27768176794052124 lr:1.9500845428017055e-06 tokens_per_second_per_gpu:4184.232943769313 mfu:19.73774069951548 +Step 33075 | loss:0.28924503922462463 lr:1.9488855229721383e-06 tokens_per_second_per_gpu:5362.165250890634 mfu:25.29424836818233 +Step 33076 | loss:0.2688603699207306 lr:1.9476868645411374e-06 tokens_per_second_per_gpu:4339.3880854917925 mfu:20.46963397521732 +Step 33077 | loss:0.2844076454639435 lr:1.946488567517707e-06 tokens_per_second_per_gpu:4201.08637741076 mfu:19.817241221494474 +Step 33078 | loss:0.34850770235061646 lr:1.9452906319108677e-06 tokens_per_second_per_gpu:4012.6850923860598 mfu:18.928520215459095 +Step 33079 | loss:0.2962885797023773 lr:1.94409305772964e-06 tokens_per_second_per_gpu:5327.169990144679 mfu:25.12916975243649 +Step 33080 | loss:0.3538869023323059 lr:1.9428958449830116e-06 tokens_per_second_per_gpu:4449.365548844074 mfu:20.98841643394017 +Step 33081 | loss:0.42573341727256775 lr:1.9416989936799913e-06 tokens_per_second_per_gpu:5319.877547081111 mfu:25.09477005428655 +Step 33082 | loss:0.4268980920314789 lr:1.9405025038296e-06 tokens_per_second_per_gpu:5345.66659542833 mfu:25.216421395405753 +Step 33083 | loss:0.24875253438949585 lr:1.9393063754408135e-06 tokens_per_second_per_gpu:5282.371574291891 mfu:24.917847981460984 +Step 33084 | loss:0.3018769323825836 lr:1.9381106085226363e-06 tokens_per_second_per_gpu:4063.8772111985013 mfu:19.1700021741735 +Step 33085 | loss:0.1877809762954712 lr:1.9369152030840556e-06 tokens_per_second_per_gpu:3271.6079586683236 mfu:15.43273293491536 +Step 33086 | loss:0.35540369153022766 lr:1.93572015913408e-06 tokens_per_second_per_gpu:5114.279858993902 mfu:24.124930680996407 +Step 33087 | loss:0.2620098888874054 lr:1.934525476681681e-06 tokens_per_second_per_gpu:4643.169805416826 mfu:21.902624178609713 +Step 33088 | loss:0.3299902379512787 lr:1.9333311557358457e-06 tokens_per_second_per_gpu:5148.887572160017 mfu:24.28818116868571 +Step 33089 | loss:0.27155011892318726 lr:1.932137196305561e-06 tokens_per_second_per_gpu:5346.443070745656 mfu:25.220084161957928 +Step 33090 | loss:0.21956658363342285 lr:1.9309435983998148e-06 tokens_per_second_per_gpu:5307.734290267236 mfu:25.037488240042247 +Step 33091 | loss:0.3425782024860382 lr:1.9297503620275657e-06 tokens_per_second_per_gpu:5348.583780776308 mfu:25.230182256415308 +Step 33092 | loss:0.24210043251514435 lr:1.9285574871978018e-06 tokens_per_second_per_gpu:5339.592657570035 mfu:25.187769594207516 +Step 33093 | loss:0.2936762273311615 lr:1.9273649739194876e-06 tokens_per_second_per_gpu:4732.5803789188685 mfu:22.32438910883574 +Step 33094 | loss:0.2185996174812317 lr:1.9261728222015995e-06 tokens_per_second_per_gpu:5355.584081828368 mfu:25.263203870852394 +Step 33095 | loss:0.2552803158760071 lr:1.924981032053097e-06 tokens_per_second_per_gpu:5347.188026296629 mfu:25.223598244394392 +Step 33096 | loss:0.36253494024276733 lr:1.923789603482945e-06 tokens_per_second_per_gpu:3976.126041773785 mfu:18.75606493610376 +Step 33097 | loss:0.3747970759868622 lr:1.9225985365001143e-06 tokens_per_second_per_gpu:3965.4154273281233 mfu:18.705541140344312 +Step 33098 | loss:0.2878294885158539 lr:1.921407831113553e-06 tokens_per_second_per_gpu:5298.524085231451 mfu:24.99404213146543 +Step 33099 | loss:0.2816692590713501 lr:1.9202174873322155e-06 tokens_per_second_per_gpu:4479.646131955346 mfu:21.131255110876264 +Step 33100 | loss:0.3503548204898834 lr:1.9190275051650553e-06 tokens_per_second_per_gpu:5385.718640515434 mfu:25.405353725665172 +Step 33101 | loss:0.2696244418621063 lr:1.9178378846210266e-06 tokens_per_second_per_gpu:5344.218642781428 mfu:25.209591155724116 +Step 33102 | loss:0.37572452425956726 lr:1.916648625709072e-06 tokens_per_second_per_gpu:4050.4023717155846 mfu:19.1064390572879 +Step 33103 | loss:0.3125919997692108 lr:1.9154597284381347e-06 tokens_per_second_per_gpu:4780.759013600322 mfu:22.55165594875086 +Step 33104 | loss:0.3411812484264374 lr:1.9142711928171743e-06 tokens_per_second_per_gpu:4375.386431713007 mfu:20.639444316293993 +Step 33105 | loss:0.3250110447406769 lr:1.9130830188551106e-06 tokens_per_second_per_gpu:4262.49407147388 mfu:20.106911791623194 +Step 33106 | loss:0.29345086216926575 lr:1.9118952065608765e-06 tokens_per_second_per_gpu:3960.01494094102 mfu:18.680066124638355 +Step 33107 | loss:0.34880512952804565 lr:1.9107077559434194e-06 tokens_per_second_per_gpu:3249.584040127119 mfu:15.328842353488517 +Step 33108 | loss:0.2930814325809479 lr:1.9095206670116715e-06 tokens_per_second_per_gpu:5161.645389962181 mfu:24.348361971966693 +Step 33109 | loss:0.288276344537735 lr:1.908333939774548e-06 tokens_per_second_per_gpu:4337.202577296262 mfu:20.45932456017285 +Step 33110 | loss:0.46756860613822937 lr:1.9071475742409806e-06 tokens_per_second_per_gpu:4607.298822663322 mfu:21.733414632740683 +Step 33111 | loss:0.2761126458644867 lr:1.9059615704199007e-06 tokens_per_second_per_gpu:4362.322351474371 mfu:20.577818820846414 +Step 33112 | loss:0.27669215202331543 lr:1.9047759283202181e-06 tokens_per_second_per_gpu:3133.488183769122 mfu:14.781198391052214 +Step 33113 | loss:0.3703587055206299 lr:1.903590647950848e-06 tokens_per_second_per_gpu:3855.740493890083 mfu:18.188185766843862 +Step 33114 | loss:0.3415292799472809 lr:1.902405729320711e-06 tokens_per_second_per_gpu:5336.016482996564 mfu:25.17090016858615 +Step 33115 | loss:0.2850518226623535 lr:1.9012211724387218e-06 tokens_per_second_per_gpu:4870.049479597084 mfu:22.972854311382022 +Step 33116 | loss:0.20774242281913757 lr:1.9000369773137848e-06 tokens_per_second_per_gpu:4502.436633699015 mfu:21.23876179606171 +Step 33117 | loss:0.3853508532047272 lr:1.898853143954804e-06 tokens_per_second_per_gpu:4697.249535176989 mfu:22.157727490841754 +Step 33118 | loss:0.31103357672691345 lr:1.8976696723706889e-06 tokens_per_second_per_gpu:4762.8206989285345 mfu:22.46703785785196 +Step 33119 | loss:0.23897692561149597 lr:1.8964865625703432e-06 tokens_per_second_per_gpu:3957.999632230234 mfu:18.670559569602457 +Step 33120 | loss:0.26318734884262085 lr:1.895303814562649e-06 tokens_per_second_per_gpu:5331.529380419922 mfu:25.14973374015349 +Step 33121 | loss:0.27137869596481323 lr:1.8941214283565155e-06 tokens_per_second_per_gpu:4310.9023929208 mfu:20.33526210319963 +Step 33122 | loss:0.3566526770591736 lr:1.8929394039608418e-06 tokens_per_second_per_gpu:4281.005638785788 mfu:20.194234013032535 +Step 33123 | loss:0.2065412551164627 lr:1.8917577413844979e-06 tokens_per_second_per_gpu:4890.114833489346 mfu:23.067506009194215 +Step 33124 | loss:0.3520909249782562 lr:1.890576440636388e-06 tokens_per_second_per_gpu:5348.089398765168 mfu:25.22785017211849 +Step 33125 | loss:0.3003672659397125 lr:1.8893955017253884e-06 tokens_per_second_per_gpu:4681.841921027845 mfu:22.085047146090197 +Step 33126 | loss:0.3561122715473175 lr:1.8882149246603976e-06 tokens_per_second_per_gpu:4975.867141032052 mfu:23.472014274725513 +Step 33127 | loss:0.2891601324081421 lr:1.8870347094502695e-06 tokens_per_second_per_gpu:3604.9514708623506 mfu:17.005171156202454 +Step 33128 | loss:0.36300721764564514 lr:1.8858548561038913e-06 tokens_per_second_per_gpu:5109.168424626183 mfu:24.100819172983247 +Step 33129 | loss:0.28926095366477966 lr:1.8846753646301397e-06 tokens_per_second_per_gpu:5328.320751156084 mfu:25.13459808846736 +Step 33130 | loss:0.36152344942092896 lr:1.8834962350378904e-06 tokens_per_second_per_gpu:3806.4288229381114 mfu:17.95557420152478 +Step 33131 | loss:0.30092930793762207 lr:1.8823174673359978e-06 tokens_per_second_per_gpu:5376.201754929185 mfu:25.36046095260672 +Step 33132 | loss:0.3740499019622803 lr:1.8811390615333324e-06 tokens_per_second_per_gpu:5341.385478405358 mfu:25.19622663597451 +Step 33133 | loss:0.29921773076057434 lr:1.8799610176387706e-06 tokens_per_second_per_gpu:3753.551756406267 mfu:17.706144056936225 +Step 33134 | loss:0.2842240333557129 lr:1.878783335661166e-06 tokens_per_second_per_gpu:3824.0859593887685 mfu:18.03886592678066 +Step 33135 | loss:0.31722012162208557 lr:1.8776060156093567e-06 tokens_per_second_per_gpu:3803.6752269693925 mfu:17.942585019528416 +Step 33136 | loss:0.32563555240631104 lr:1.8764290574922183e-06 tokens_per_second_per_gpu:4299.918683981654 mfu:20.283450074119692 +Step 33137 | loss:0.35838112235069275 lr:1.8752524613186052e-06 tokens_per_second_per_gpu:5339.593226427692 mfu:25.187772277606136 +Step 33138 | loss:0.32498374581336975 lr:1.874076227097349e-06 tokens_per_second_per_gpu:3917.754087865507 mfu:18.480714470236105 +Step 33139 | loss:0.22755736112594604 lr:1.8729003548373091e-06 tokens_per_second_per_gpu:3955.131193106322 mfu:18.65702865284878 +Step 33140 | loss:0.18738162517547607 lr:1.8717248445473346e-06 tokens_per_second_per_gpu:4998.636356175266 mfu:23.579420547384743 +Step 33141 | loss:0.33620887994766235 lr:1.8705496962362567e-06 tokens_per_second_per_gpu:4622.258755449727 mfu:21.803983188121883 +Step 33142 | loss:0.3213585615158081 lr:1.8693749099129077e-06 tokens_per_second_per_gpu:3423.7737313635594 mfu:16.150524846876742 +Step 33143 | loss:0.2386258989572525 lr:1.8682004855861357e-06 tokens_per_second_per_gpu:5381.464465201603 mfu:25.385286054872676 +Step 33144 | loss:0.3609479069709778 lr:1.8670264232647727e-06 tokens_per_second_per_gpu:5316.083231538507 mfu:25.0768716204951 +Step 33145 | loss:0.32193195819854736 lr:1.8658527229576394e-06 tokens_per_second_per_gpu:4057.5602739682 mfu:19.14020410347754 +Step 33146 | loss:0.34232503175735474 lr:1.8646793846735677e-06 tokens_per_second_per_gpu:5285.504053846695 mfu:24.932624421977113 +Step 33147 | loss:0.3520451486110687 lr:1.8635064084213948e-06 tokens_per_second_per_gpu:5349.55619381511 mfu:25.234769294630112 +Step 33148 | loss:0.29388725757598877 lr:1.8623337942099307e-06 tokens_per_second_per_gpu:3960.0497882026916 mfu:18.680230505117212 +Step 33149 | loss:0.2881373465061188 lr:1.86116154204799e-06 tokens_per_second_per_gpu:5021.217158333361 mfu:23.685938043846605 +Step 33150 | loss:0.2806226313114166 lr:1.8599896519443994e-06 tokens_per_second_per_gpu:4845.962942972013 mfu:22.859234008534642 +Step 33151 | loss:0.37092965841293335 lr:1.8588181239079739e-06 tokens_per_second_per_gpu:5349.631901312786 mfu:25.235126419813678 +Step 33152 | loss:0.3330194652080536 lr:1.857646957947512e-06 tokens_per_second_per_gpu:4670.499347605441 mfu:22.03154229201413 +Step 33153 | loss:0.2431437075138092 lr:1.8564761540718289e-06 tokens_per_second_per_gpu:5272.291918269834 mfu:24.870300524238285 +Step 33154 | loss:0.4230249226093292 lr:1.85530571228974e-06 tokens_per_second_per_gpu:4940.952322709246 mfu:23.307315119614557 +Step 33155 | loss:0.2713129222393036 lr:1.8541356326100433e-06 tokens_per_second_per_gpu:4539.851807239661 mfu:21.41525555333981 +Step 33156 | loss:0.34100499749183655 lr:1.8529659150415269e-06 tokens_per_second_per_gpu:4592.323604405684 mfu:21.6627739731837 +Step 33157 | loss:0.37579166889190674 lr:1.8517965595929943e-06 tokens_per_second_per_gpu:5359.81740733741 mfu:25.283173189558358 +Step 33158 | loss:0.37154144048690796 lr:1.8506275662732553e-06 tokens_per_second_per_gpu:4140.643095150964 mfu:19.532119946387652 +Step 33159 | loss:0.3223899304866791 lr:1.8494589350910807e-06 tokens_per_second_per_gpu:4401.616301142903 mfu:20.763175085672007 +Step 33160 | loss:0.2535536289215088 lr:1.848290666055269e-06 tokens_per_second_per_gpu:5366.595841096754 mfu:25.315148218121237 +Step 33161 | loss:0.33371612429618835 lr:1.8471227591746077e-06 tokens_per_second_per_gpu:4589.277277895321 mfu:21.648403931277418 +Step 33162 | loss:0.30616873502731323 lr:1.8459552144578896e-06 tokens_per_second_per_gpu:5315.7014458006 mfu:25.07507067203003 +Step 33163 | loss:0.3481728434562683 lr:1.8447880319138854e-06 tokens_per_second_per_gpu:4258.5924126452055 mfu:20.0885070012367 +Step 33164 | loss:0.367464542388916 lr:1.8436212115513663e-06 tokens_per_second_per_gpu:4845.03007187428 mfu:22.854833496402417 +Step 33165 | loss:0.33587563037872314 lr:1.8424547533791248e-06 tokens_per_second_per_gpu:3282.13162815647 mfu:15.48237487941411 +Step 33166 | loss:0.2742936909198761 lr:1.8412886574059207e-06 tokens_per_second_per_gpu:3108.2844729829585 mfu:14.6623081870774 +Step 33167 | loss:0.2963336110115051 lr:1.8401229236405303e-06 tokens_per_second_per_gpu:4544.383483765033 mfu:21.43663224469373 +Step 33168 | loss:0.30988451838493347 lr:1.8389575520917191e-06 tokens_per_second_per_gpu:4055.9660492413705 mfu:19.1326838734383 +Step 33169 | loss:0.24619658291339874 lr:1.837792542768263e-06 tokens_per_second_per_gpu:2822.228445604778 mfu:13.312933099743514 +Step 33170 | loss:0.26269274950027466 lr:1.8366278956789107e-06 tokens_per_second_per_gpu:4424.273828958669 mfu:20.87005451015185 +Step 33171 | loss:0.25467950105667114 lr:1.8354636108324218e-06 tokens_per_second_per_gpu:4569.3834843586665 mfu:21.55456151293384 +Step 33172 | loss:0.3308727443218231 lr:1.8342996882375562e-06 tokens_per_second_per_gpu:3755.3557635616116 mfu:17.714653866483346 +Step 33173 | loss:0.3079761266708374 lr:1.8331361279030734e-06 tokens_per_second_per_gpu:4912.8922429160375 mfu:23.174950935686702 +Step 33174 | loss:0.35263311862945557 lr:1.8319729298377164e-06 tokens_per_second_per_gpu:3741.726634219212 mfu:17.650362938005998 +Step 33175 | loss:0.35683193802833557 lr:1.8308100940502338e-06 tokens_per_second_per_gpu:4720.127663723757 mfu:22.265647526608095 +Step 33176 | loss:0.3068157136440277 lr:1.8296476205493796e-06 tokens_per_second_per_gpu:5007.821690738841 mfu:23.62274933770081 +Step 33177 | loss:0.30687665939331055 lr:1.8284855093438969e-06 tokens_per_second_per_gpu:3899.303366142561 mfu:18.393679267851233 +Step 33178 | loss:0.4077463150024414 lr:1.8273237604425176e-06 tokens_per_second_per_gpu:3554.247432096922 mfu:16.76599155434485 +Step 33179 | loss:0.3534458577632904 lr:1.826162373853979e-06 tokens_per_second_per_gpu:4431.230759231315 mfu:20.90287153722262 +Step 33180 | loss:0.33392155170440674 lr:1.8250013495870243e-06 tokens_per_second_per_gpu:4307.224078694208 mfu:20.317910867407676 +Step 33181 | loss:0.34021857380867004 lr:1.82384068765038e-06 tokens_per_second_per_gpu:5202.215760277678 mfu:24.539739330763844 +Step 33182 | loss:0.378341406583786 lr:1.822680388052772e-06 tokens_per_second_per_gpu:4779.305640189305 mfu:22.54480013840002 +Step 33183 | loss:0.36360809206962585 lr:1.8215204508029437e-06 tokens_per_second_per_gpu:5374.232777704435 mfu:25.351172951095414 +Step 33184 | loss:0.27518215775489807 lr:1.8203608759096048e-06 tokens_per_second_per_gpu:4735.938359399371 mfu:22.340229275692813 +Step 33185 | loss:0.3309237062931061 lr:1.8192016633814758e-06 tokens_per_second_per_gpu:4774.544513198464 mfu:22.522341090052194 +Step 33186 | loss:0.3039959967136383 lr:1.8180428132272776e-06 tokens_per_second_per_gpu:3866.1103972963397 mfu:18.237102370499137 +Step 33187 | loss:0.27289190888404846 lr:1.8168843254557312e-06 tokens_per_second_per_gpu:4467.829456601015 mfu:21.075513837097173 +Step 33188 | loss:0.2900541126728058 lr:1.8157262000755403e-06 tokens_per_second_per_gpu:4369.64241692696 mfu:20.612348818518228 +Step 33189 | loss:0.23675379157066345 lr:1.8145684370954264e-06 tokens_per_second_per_gpu:4465.926247481564 mfu:21.06653607495976 +Step 33190 | loss:0.3731776177883148 lr:1.813411036524093e-06 tokens_per_second_per_gpu:4373.395801079523 mfu:20.63005417195924 +Step 33191 | loss:0.39402300119400024 lr:1.8122539983702391e-06 tokens_per_second_per_gpu:4331.764028725616 mfu:20.43366999865305 +Step 33192 | loss:0.3187789022922516 lr:1.8110973226425799e-06 tokens_per_second_per_gpu:5290.0511941353125 mfu:24.954074058540655 +Step 33193 | loss:0.32487618923187256 lr:1.809941009349797e-06 tokens_per_second_per_gpu:5355.01172324249 mfu:25.260503957748362 +Step 33194 | loss:0.37301480770111084 lr:1.8087850585006004e-06 tokens_per_second_per_gpu:3157.619086499734 mfu:14.895027976389065 +Step 33195 | loss:0.3705638349056244 lr:1.8076294701036834e-06 tokens_per_second_per_gpu:5171.19777868261 mfu:24.39342221936653 +Step 33196 | loss:0.30335575342178345 lr:1.8064742441677273e-06 tokens_per_second_per_gpu:5340.205319420919 mfu:25.190659624689026 +Step 33197 | loss:0.30394089221954346 lr:1.8053193807014313e-06 tokens_per_second_per_gpu:5324.044925927188 mfu:25.11442829133165 +Step 33198 | loss:0.35688430070877075 lr:1.8041648797134825e-06 tokens_per_second_per_gpu:4287.912805922179 mfu:20.226816298899127 +Step 33199 | loss:0.33486539125442505 lr:1.8030107412125574e-06 tokens_per_second_per_gpu:4523.565022966872 mfu:21.338428013113123 +Step 33200 | loss:0.48515698313713074 lr:1.8018569652073381e-06 tokens_per_second_per_gpu:5388.424960430382 mfu:25.41811989844977 +Step 33201 | loss:0.406642347574234 lr:1.8007035517065062e-06 tokens_per_second_per_gpu:4463.781785094924 mfu:21.056420279998967 +Step 33202 | loss:0.33692029118537903 lr:1.7995505007187275e-06 tokens_per_second_per_gpu:4151.520171579182 mfu:19.583428971720164 +Step 33203 | loss:0.19239157438278198 lr:1.7983978122526779e-06 tokens_per_second_per_gpu:5379.290110430561 mfu:25.37502925243449 +Step 33204 | loss:0.24241241812705994 lr:1.7972454863170285e-06 tokens_per_second_per_gpu:3495.334211888033 mfu:16.488087842987664 +Step 33205 | loss:0.23945583403110504 lr:1.7960935229204557e-06 tokens_per_second_per_gpu:5374.165038999835 mfu:25.3508534160684 +Step 33206 | loss:0.3564521372318268 lr:1.794941922071608e-06 tokens_per_second_per_gpu:5108.779182245276 mfu:24.098983050260912 +Step 33207 | loss:0.34837326407432556 lr:1.7937906837791562e-06 tokens_per_second_per_gpu:5058.899351807415 mfu:23.863691379708886 +Step 33208 | loss:0.2562595009803772 lr:1.7926398080517492e-06 tokens_per_second_per_gpu:4686.523669853881 mfu:22.107131754091323 +Step 33209 | loss:0.3311172425746918 lr:1.7914892948980576e-06 tokens_per_second_per_gpu:4202.012675083103 mfu:19.821610725657596 +Step 33210 | loss:0.2992520332336426 lr:1.7903391443267192e-06 tokens_per_second_per_gpu:5207.843817283481 mfu:24.566287835905758 +Step 33211 | loss:0.29938995838165283 lr:1.7891893563463936e-06 tokens_per_second_per_gpu:5348.393873940504 mfu:25.229286433469372 +Step 33212 | loss:0.234490305185318 lr:1.7880399309657347e-06 tokens_per_second_per_gpu:3629.1131481324824 mfu:17.119145910292897 +Step 33213 | loss:0.3398192524909973 lr:1.786890868193375e-06 tokens_per_second_per_gpu:5348.467486686964 mfu:25.229633677354055 +Step 33214 | loss:0.3227830231189728 lr:1.7857421680379572e-06 tokens_per_second_per_gpu:4042.9879972483873 mfu:19.0714641879034 +Step 33215 | loss:0.23167650401592255 lr:1.7845938305081244e-06 tokens_per_second_per_gpu:4436.037034103122 mfu:20.925543555827993 +Step 33216 | loss:0.2975231111049652 lr:1.78344585561252e-06 tokens_per_second_per_gpu:5325.5543352390505 mfu:25.121548432587485 +Step 33217 | loss:0.4407448470592499 lr:1.7822982433597645e-06 tokens_per_second_per_gpu:5324.040968693901 mfu:25.114409624387058 +Step 33218 | loss:0.39313805103302 lr:1.7811509937584958e-06 tokens_per_second_per_gpu:3692.468374874365 mfu:17.418003324350682 +Step 33219 | loss:0.4150024354457855 lr:1.7800041068173512e-06 tokens_per_second_per_gpu:5301.935847743529 mfu:25.010135997341614 +Step 33220 | loss:0.2527613043785095 lr:1.7788575825449517e-06 tokens_per_second_per_gpu:4441.492573046576 mfu:20.95127826383519 +Step 33221 | loss:0.270447701215744 lr:1.7777114209499069e-06 tokens_per_second_per_gpu:3958.79043348665 mfu:18.674289914053574 +Step 33222 | loss:0.31890302896499634 lr:1.7765656220408489e-06 tokens_per_second_per_gpu:3365.446965187671 mfu:15.875387539252594 +Step 33223 | loss:0.40044233202934265 lr:1.7754201858263986e-06 tokens_per_second_per_gpu:4446.704521423225 mfu:20.97586391357793 +Step 33224 | loss:0.3803538382053375 lr:1.7742751123151602e-06 tokens_per_second_per_gpu:4483.695655158185 mfu:21.15035740274396 +Step 33225 | loss:0.36118340492248535 lr:1.7731304015157546e-06 tokens_per_second_per_gpu:5033.216365165955 mfu:23.742540349751746 +Step 33226 | loss:0.40233662724494934 lr:1.7719860534367972e-06 tokens_per_second_per_gpu:5353.098816999368 mfu:25.251480452624097 +Step 33227 | loss:0.3102162778377533 lr:1.7708420680868753e-06 tokens_per_second_per_gpu:4717.293416995712 mfu:22.252277901219546 +Step 33228 | loss:0.2706482708454132 lr:1.7696984454746158e-06 tokens_per_second_per_gpu:5318.356729743409 mfu:25.087596099426417 +Step 33229 | loss:0.3863259255886078 lr:1.7685551856086002e-06 tokens_per_second_per_gpu:5352.323898681457 mfu:25.24782503070381 +Step 33230 | loss:0.2639419734477997 lr:1.7674122884974386e-06 tokens_per_second_per_gpu:4539.36290942597 mfu:21.412949338938063 +Step 33231 | loss:0.30084657669067383 lr:1.7662697541497187e-06 tokens_per_second_per_gpu:4530.417947811285 mfu:21.370754428833763 +Step 33232 | loss:0.25398197770118713 lr:1.7651275825740443e-06 tokens_per_second_per_gpu:4958.145630147589 mfu:23.388418884281958 +Step 33233 | loss:0.3804765045642853 lr:1.7639857737789977e-06 tokens_per_second_per_gpu:5338.09295442175 mfu:25.18069523858131 +Step 33234 | loss:0.2197979837656021 lr:1.7628443277731666e-06 tokens_per_second_per_gpu:4379.745690113765 mfu:20.66000768193654 +Step 33235 | loss:0.247431680560112 lr:1.7617032445651494e-06 tokens_per_second_per_gpu:4920.013574993745 mfu:23.208543474121793 +Step 33236 | loss:0.2896007001399994 lr:1.7605625241635059e-06 tokens_per_second_per_gpu:3420.628615439975 mfu:16.135688798453085 +Step 33237 | loss:0.3790356516838074 lr:1.7594221665768351e-06 tokens_per_second_per_gpu:5169.761898785035 mfu:24.386648928903295 +Step 33238 | loss:0.18574172258377075 lr:1.7582821718136967e-06 tokens_per_second_per_gpu:3759.586207987942 mfu:17.734609594630594 +Step 33239 | loss:0.2864290177822113 lr:1.757142539882678e-06 tokens_per_second_per_gpu:5059.780735137274 mfu:23.867849015255334 +Step 33240 | loss:0.3202829957008362 lr:1.7560032707923447e-06 tokens_per_second_per_gpu:3792.9113743643384 mfu:17.89181009028775 +Step 33241 | loss:0.3202275037765503 lr:1.7548643645512675e-06 tokens_per_second_per_gpu:5330.271881720962 mfu:25.14380190423909 +Step 33242 | loss:0.22836260497570038 lr:1.7537258211680175e-06 tokens_per_second_per_gpu:5366.795546530181 mfu:25.316090262723975 +Step 33243 | loss:0.3742017447948456 lr:1.7525876406511433e-06 tokens_per_second_per_gpu:4319.590457903076 mfu:20.376245188057766 +Step 33244 | loss:0.25579121708869934 lr:1.7514498230092158e-06 tokens_per_second_per_gpu:3354.886478693624 mfu:15.825571922655383 +Step 33245 | loss:0.2698822021484375 lr:1.750312368250795e-06 tokens_per_second_per_gpu:3632.1192638171888 mfu:17.13332627087388 +Step 33246 | loss:0.2861559987068176 lr:1.7491752763844293e-06 tokens_per_second_per_gpu:4309.874576293449 mfu:20.330413716804816 +Step 33247 | loss:0.38173845410346985 lr:1.7480385474186678e-06 tokens_per_second_per_gpu:5313.740811311859 mfu:25.065822024627902 +Step 33248 | loss:0.31163081526756287 lr:1.7469021813620756e-06 tokens_per_second_per_gpu:5011.778602021673 mfu:23.641414763340606 +Step 33249 | loss:0.295825332403183 lr:1.745766178223185e-06 tokens_per_second_per_gpu:5316.659992186019 mfu:25.07959229887493 +Step 33250 | loss:0.34156787395477295 lr:1.7446305380105443e-06 tokens_per_second_per_gpu:3536.550364300489 mfu:16.68251146611796 +Step 33251 | loss:0.3359931409358978 lr:1.7434952607326916e-06 tokens_per_second_per_gpu:5001.338976011982 mfu:23.59216926626903 +Step 33252 | loss:0.33756354451179504 lr:1.7423603463981696e-06 tokens_per_second_per_gpu:5297.54954644813 mfu:24.989445065751045 +Step 33253 | loss:0.4084206819534302 lr:1.7412257950155164e-06 tokens_per_second_per_gpu:3643.7800248333924 mfu:17.188332070118296 +Step 33254 | loss:0.40583011507987976 lr:1.7400916065932582e-06 tokens_per_second_per_gpu:5319.876364664003 mfu:25.09476447662331 +Step 33255 | loss:0.40100789070129395 lr:1.7389577811399326e-06 tokens_per_second_per_gpu:3347.319016530101 mfu:15.78987491248786 +Step 33256 | loss:0.29861024022102356 lr:1.7378243186640607e-06 tokens_per_second_per_gpu:5375.981682329454 mfu:25.359422832605528 +Step 33257 | loss:0.38842687010765076 lr:1.7366912191741746e-06 tokens_per_second_per_gpu:4685.807506240198 mfu:22.103753488135467 +Step 33258 | loss:0.25861579179763794 lr:1.7355584826787842e-06 tokens_per_second_per_gpu:4144.193318697275 mfu:19.5488669565862 +Step 33259 | loss:0.3253449499607086 lr:1.7344261091864267e-06 tokens_per_second_per_gpu:4806.266939686589 mfu:22.671981188200743 +Step 33260 | loss:0.26800239086151123 lr:1.7332940987056013e-06 tokens_per_second_per_gpu:4853.205248074201 mfu:22.893397197366333 +Step 33261 | loss:0.3481982946395874 lr:1.7321624512448287e-06 tokens_per_second_per_gpu:4781.190570082339 mfu:22.553691674307665 +Step 33262 | loss:0.3585704267024994 lr:1.73103116681263e-06 tokens_per_second_per_gpu:3992.7136908921966 mfu:18.83431170714008 +Step 33263 | loss:0.4427936375141144 lr:1.7299002454174984e-06 tokens_per_second_per_gpu:5335.677657175936 mfu:25.169301869380604 +Step 33264 | loss:0.20214904844760895 lr:1.7287696870679493e-06 tokens_per_second_per_gpu:4946.225979629197 mfu:23.33219185908462 +Step 33265 | loss:0.25843214988708496 lr:1.7276394917724813e-06 tokens_per_second_per_gpu:5325.523583345002 mfu:25.12140337065645 +Step 33266 | loss:0.27914783358573914 lr:1.726509659539599e-06 tokens_per_second_per_gpu:4624.299577070315 mfu:21.81361009190627 +Step 33267 | loss:0.3287600874900818 lr:1.7253801903777955e-06 tokens_per_second_per_gpu:3852.4228168761565 mfu:18.172535718315103 +Step 33268 | loss:0.32116302847862244 lr:1.7242510842955638e-06 tokens_per_second_per_gpu:5182.010595387882 mfu:24.444428120622067 +Step 33269 | loss:0.22416052222251892 lr:1.7231223413014085e-06 tokens_per_second_per_gpu:5328.090488501876 mfu:25.133511900240602 +Step 33270 | loss:0.39315688610076904 lr:1.7219939614038005e-06 tokens_per_second_per_gpu:4294.682430711913 mfu:20.25874977404939 +Step 33271 | loss:0.2252868264913559 lr:1.720865944611244e-06 tokens_per_second_per_gpu:4560.7368364395215 mfu:21.513773799428975 +Step 33272 | loss:0.3845350444316864 lr:1.7197382909322102e-06 tokens_per_second_per_gpu:4764.910847000122 mfu:22.476897442921672 +Step 33273 | loss:0.3131254315376282 lr:1.718611000375192e-06 tokens_per_second_per_gpu:5360.482459483131 mfu:25.286310353999315 +Step 33274 | loss:0.32010430097579956 lr:1.7174840729486552e-06 tokens_per_second_per_gpu:4238.71488660734 mfu:19.994741319460434 +Step 33275 | loss:0.3319569230079651 lr:1.7163575086610762e-06 tokens_per_second_per_gpu:5357.251187848563 mfu:25.271067894387272 +Step 33276 | loss:0.2999359965324402 lr:1.715231307520937e-06 tokens_per_second_per_gpu:4814.949830669131 mfu:22.71293986641936 +Step 33277 | loss:0.3831854462623596 lr:1.7141054695367142e-06 tokens_per_second_per_gpu:3993.0319784062153 mfu:18.835813123649157 +Step 33278 | loss:0.3536856472492218 lr:1.7129799947168624e-06 tokens_per_second_per_gpu:5355.454090018165 mfu:25.262590677304775 +Step 33279 | loss:0.24458791315555573 lr:1.711854883069841e-06 tokens_per_second_per_gpu:5300.28850956592 mfu:25.00236522209296 +Step 33280 | loss:0.2593514323234558 lr:1.7107301346041215e-06 tokens_per_second_per_gpu:4541.847410920622 mfu:21.424669156388 +Step 33281 | loss:0.409039169549942 lr:1.7096057493281693e-06 tokens_per_second_per_gpu:3792.0233595654113 mfu:17.88762117297026 +Step 33282 | loss:0.2794981300830841 lr:1.7084817272504271e-06 tokens_per_second_per_gpu:4211.059838290467 mfu:19.864287738111948 +Step 33283 | loss:0.39075762033462524 lr:1.7073580683793612e-06 tokens_per_second_per_gpu:4803.966511542683 mfu:22.661129676152427 +Step 33284 | loss:0.3338484764099121 lr:1.7062347727234195e-06 tokens_per_second_per_gpu:4720.555095820976 mfu:22.267663796738113 +Step 33285 | loss:0.2457389086484909 lr:1.7051118402910405e-06 tokens_per_second_per_gpu:4107.020351770132 mfu:19.373515729228703 +Step 33286 | loss:0.33723485469818115 lr:1.7039892710906836e-06 tokens_per_second_per_gpu:4802.343242951624 mfu:22.653472441250223 +Step 33287 | loss:0.3662795126438141 lr:1.7028670651307864e-06 tokens_per_second_per_gpu:4832.0517563125995 mfu:22.793612567569 +Step 33288 | loss:0.20701536536216736 lr:1.701745222419787e-06 tokens_per_second_per_gpu:5338.448045470167 mfu:25.182370263641477 +Step 33289 | loss:0.34469717741012573 lr:1.7006237429661233e-06 tokens_per_second_per_gpu:5344.084510463567 mfu:25.20895843069544 +Step 33290 | loss:0.31284594535827637 lr:1.699502626778232e-06 tokens_per_second_per_gpu:4478.937463813828 mfu:21.127912202341466 +Step 33291 | loss:0.2955401837825775 lr:1.6983818738645463e-06 tokens_per_second_per_gpu:5322.870311344112 mfu:25.10888743393337 +Step 33292 | loss:0.34574711322784424 lr:1.6972614842334921e-06 tokens_per_second_per_gpu:3936.747570354618 mfu:18.57030996775952 +Step 33293 | loss:0.24036283791065216 lr:1.696141457893502e-06 tokens_per_second_per_gpu:4566.485267084975 mfu:21.540890127566982 +Step 33294 | loss:0.23926295340061188 lr:1.6950217948529911e-06 tokens_per_second_per_gpu:4284.496657267088 mfu:20.210701742838 +Step 33295 | loss:0.2811484932899475 lr:1.6939024951203863e-06 tokens_per_second_per_gpu:5337.630719525923 mfu:25.178514797712293 +Step 33296 | loss:0.3705688416957855 lr:1.692783558704103e-06 tokens_per_second_per_gpu:5346.0886666882125 mfu:25.218412377551402 +Step 33297 | loss:0.3532314896583557 lr:1.6916649856125566e-06 tokens_per_second_per_gpu:3857.0599290042073 mfu:18.194409767396028 +Step 33298 | loss:0.35864800214767456 lr:1.6905467758541683e-06 tokens_per_second_per_gpu:4657.490135388875 mfu:21.970175618388108 +Step 33299 | loss:0.21592698991298676 lr:1.6894289294373368e-06 tokens_per_second_per_gpu:4938.815774298677 mfu:23.297236656229153 +Step 33300 | loss:0.2885720729827881 lr:1.688311446370483e-06 tokens_per_second_per_gpu:3913.6161754358122 mfu:18.461195231304576 +Step 33301 | loss:0.24461787939071655 lr:1.6871943266619949e-06 tokens_per_second_per_gpu:4553.370973874942 mfu:21.47902776896611 +Step 33302 | loss:0.34657642245292664 lr:1.686077570320288e-06 tokens_per_second_per_gpu:4827.7022161986515 mfu:22.773095044743442 +Step 33303 | loss:0.3139786422252655 lr:1.6849611773537555e-06 tokens_per_second_per_gpu:5336.363413051687 mfu:25.172536696848535 +Step 33304 | loss:0.28098249435424805 lr:1.6838451477707905e-06 tokens_per_second_per_gpu:4470.297452227027 mfu:21.087155793548543 +Step 33305 | loss:0.2564563751220703 lr:1.6827294815797978e-06 tokens_per_second_per_gpu:3897.296576989065 mfu:18.384212901020657 +Step 33306 | loss:0.2713233232498169 lr:1.6816141787891593e-06 tokens_per_second_per_gpu:4654.780085861064 mfu:21.957391852383296 +Step 33307 | loss:0.2296929508447647 lr:1.6804992394072682e-06 tokens_per_second_per_gpu:4131.968179608106 mfu:19.491198889678778 +Step 33308 | loss:0.29393520951271057 lr:1.6793846634425014e-06 tokens_per_second_per_gpu:5335.973334753616 mfu:25.170696630962315 +Step 33309 | loss:0.39900916814804077 lr:1.6782704509032576e-06 tokens_per_second_per_gpu:5318.316222069159 mfu:25.08740501781566 +Step 33310 | loss:0.31841424107551575 lr:1.6771566017978968e-06 tokens_per_second_per_gpu:5182.281712109862 mfu:24.44570702445693 +Step 33311 | loss:0.36407041549682617 lr:1.6760431161348123e-06 tokens_per_second_per_gpu:5355.520078023476 mfu:25.262901953984546 +Step 33312 | loss:0.33358123898506165 lr:1.6749299939223694e-06 tokens_per_second_per_gpu:4343.982568086891 mfu:20.49130693351796 +Step 33313 | loss:0.3564688265323639 lr:1.6738172351689563e-06 tokens_per_second_per_gpu:4533.644640003802 mfu:21.385975286436327 +Step 33314 | loss:0.2573387026786804 lr:1.672704839882927e-06 tokens_per_second_per_gpu:4014.70434178124 mfu:18.938045359375636 +Step 33315 | loss:0.39362069964408875 lr:1.6715928080726417e-06 tokens_per_second_per_gpu:4393.275693025633 mfu:20.723831014128365 +Step 33316 | loss:0.1989385038614273 lr:1.670481139746477e-06 tokens_per_second_per_gpu:4852.008497897353 mfu:22.887751922595506 +Step 33317 | loss:0.3193715214729309 lr:1.6693698349127984e-06 tokens_per_second_per_gpu:3468.6590761484117 mfu:16.362256676456404 +Step 33318 | loss:0.3855084478855133 lr:1.6682588935799436e-06 tokens_per_second_per_gpu:5355.427471919761 mfu:25.262465115193088 +Step 33319 | loss:0.29719290137290955 lr:1.6671483157562896e-06 tokens_per_second_per_gpu:4445.375368277476 mfu:20.969594071412523 +Step 33320 | loss:0.3945058584213257 lr:1.6660381014501792e-06 tokens_per_second_per_gpu:4988.0895002784355 mfu:23.529669228640223 +Step 33321 | loss:0.3327074348926544 lr:1.6649282506699615e-06 tokens_per_second_per_gpu:3884.588219713288 mfu:18.324265411481477 +Step 33322 | loss:0.38029348850250244 lr:1.663818763423991e-06 tokens_per_second_per_gpu:3931.0838782285464 mfu:18.543593365675317 +Step 33323 | loss:0.35638439655303955 lr:1.6627096397206056e-06 tokens_per_second_per_gpu:5315.226632567565 mfu:25.07283089699831 +Step 33324 | loss:0.31742188334465027 lr:1.6616008795681482e-06 tokens_per_second_per_gpu:5330.848652956775 mfu:25.146522632565745 +Step 33325 | loss:0.28257301449775696 lr:1.6604924829749568e-06 tokens_per_second_per_gpu:3678.3941176773774 mfu:17.3516126518366 +Step 33326 | loss:0.25075647234916687 lr:1.6593844499493638e-06 tokens_per_second_per_gpu:4513.9772103494115 mfu:21.29320066514705 +Step 33327 | loss:0.20975539088249207 lr:1.6582767804997235e-06 tokens_per_second_per_gpu:4813.958966662873 mfu:22.708265791843008 +Step 33328 | loss:0.2632189095020294 lr:1.6571694746343402e-06 tokens_per_second_per_gpu:4170.628173425103 mfu:19.67356467658867 +Step 33329 | loss:0.3138158619403839 lr:1.6560625323615574e-06 tokens_per_second_per_gpu:4564.4098399204395 mfu:21.53109998352748 +Step 33330 | loss:0.32656624913215637 lr:1.6549559536896964e-06 tokens_per_second_per_gpu:4957.110690981413 mfu:23.383536899656278 +Step 33331 | loss:0.2479550838470459 lr:1.6538497386270834e-06 tokens_per_second_per_gpu:3490.172451704432 mfu:16.463738939514517 +Step 33332 | loss:0.27923309803009033 lr:1.6527438871820345e-06 tokens_per_second_per_gpu:5240.490278081513 mfu:24.720286761550653 +Step 33333 | loss:0.3612986207008362 lr:1.6516383993628647e-06 tokens_per_second_per_gpu:5311.538358721232 mfu:25.05543267998075 +Step 33334 | loss:0.28047245740890503 lr:1.6505332751778958e-06 tokens_per_second_per_gpu:5345.111943317078 mfu:25.213805006763568 +Step 33335 | loss:0.4434192478656769 lr:1.6494285146354315e-06 tokens_per_second_per_gpu:3638.150251981839 mfu:17.161775471039224 +Step 33336 | loss:0.32327085733413696 lr:1.648324117743788e-06 tokens_per_second_per_gpu:3617.7310838255144 mfu:17.06545477097651 +Step 33337 | loss:0.3421982526779175 lr:1.6472200845112696e-06 tokens_per_second_per_gpu:4405.8198682769225 mfu:20.783004029045692 +Step 33338 | loss:0.34953799843788147 lr:1.6461164149461805e-06 tokens_per_second_per_gpu:4615.719791832091 mfu:21.77313778107526 +Step 33339 | loss:0.4027402102947235 lr:1.6450131090568145e-06 tokens_per_second_per_gpu:5031.156564924758 mfu:23.732823920575854 +Step 33340 | loss:0.29739025235176086 lr:1.64391016685147e-06 tokens_per_second_per_gpu:4891.85301538218 mfu:23.075705309747125 +Step 33341 | loss:0.3514207601547241 lr:1.6428075883384574e-06 tokens_per_second_per_gpu:3341.6276695721085 mfu:15.763027857843088 +Step 33342 | loss:0.3138771653175354 lr:1.6417053735260534e-06 tokens_per_second_per_gpu:4019.932848430014 mfu:18.962709117313125 +Step 33343 | loss:0.3294420540332794 lr:1.6406035224225624e-06 tokens_per_second_per_gpu:5390.300623193911 mfu:25.426967719726402 +Step 33344 | loss:0.28926876187324524 lr:1.6395020350362556e-06 tokens_per_second_per_gpu:5323.410433909249 mfu:25.111435284227525 +Step 33345 | loss:0.34563058614730835 lr:1.6384009113754263e-06 tokens_per_second_per_gpu:5102.970942580046 mfu:24.071584592772144 +Step 33346 | loss:0.31889864802360535 lr:1.6373001514483566e-06 tokens_per_second_per_gpu:4709.7399671738885 mfu:22.216646989658578 +Step 33347 | loss:0.30626025795936584 lr:1.636199755263318e-06 tokens_per_second_per_gpu:5311.251966082818 mfu:25.054081717041733 +Step 33348 | loss:0.34739962220191956 lr:1.6350997228285925e-06 tokens_per_second_per_gpu:4132.21434382729 mfu:19.492360088300615 +Step 33349 | loss:0.30449679493904114 lr:1.6340000541524625e-06 tokens_per_second_per_gpu:5328.942352803748 mfu:25.13753028949557 +Step 33350 | loss:0.36241453886032104 lr:1.6329007492431825e-06 tokens_per_second_per_gpu:5186.24314386613 mfu:24.464393773941197 +Step 33351 | loss:0.21851451694965363 lr:1.6318018081090292e-06 tokens_per_second_per_gpu:4601.980003977886 mfu:21.708324857517432 +Step 33352 | loss:0.3546574115753174 lr:1.630703230758268e-06 tokens_per_second_per_gpu:3695.381823537882 mfu:17.431746558782095 +Step 33353 | loss:0.30108073353767395 lr:1.6296050171991594e-06 tokens_per_second_per_gpu:4588.771204329919 mfu:21.646016695924363 +Step 33354 | loss:0.3155710697174072 lr:1.6285071674399576e-06 tokens_per_second_per_gpu:4255.023390001132 mfu:20.071671312486547 +Step 33355 | loss:0.3463083803653717 lr:1.6274096814889283e-06 tokens_per_second_per_gpu:5336.598177844072 mfu:25.17364412242922 +Step 33356 | loss:0.2833074629306793 lr:1.6263125593543316e-06 tokens_per_second_per_gpu:4951.6743079624 mfu:23.357892553413002 +Step 33357 | loss:0.3267718255519867 lr:1.6252158010443997e-06 tokens_per_second_per_gpu:4622.170981299099 mfu:21.80356914247756 +Step 33358 | loss:0.3622470498085022 lr:1.6241194065674038e-06 tokens_per_second_per_gpu:4031.5038180681267 mfu:19.017291355307094 +Step 33359 | loss:0.3829439580440521 lr:1.6230233759315649e-06 tokens_per_second_per_gpu:5352.908398846033 mfu:25.250582217706132 +Step 33360 | loss:0.31680163741111755 lr:1.6219277091451545e-06 tokens_per_second_per_gpu:5329.174514934234 mfu:25.138625437876016 +Step 33361 | loss:0.3144027292728424 lr:1.6208324062163882e-06 tokens_per_second_per_gpu:5322.395835738912 mfu:25.10664925154975 +Step 33362 | loss:0.297666072845459 lr:1.6197374671535148e-06 tokens_per_second_per_gpu:4783.840100347215 mfu:22.566189960623564 +Step 33363 | loss:0.32098326086997986 lr:1.6186428919647778e-06 tokens_per_second_per_gpu:4665.180594980346 mfu:22.006452828408786 +Step 33364 | loss:0.3875437080860138 lr:1.6175486806583927e-06 tokens_per_second_per_gpu:5361.726462861175 mfu:25.292178530182415 +Step 33365 | loss:0.21142616868019104 lr:1.6164548332426032e-06 tokens_per_second_per_gpu:5361.543674092303 mfu:25.291316284372666 +Step 33366 | loss:0.28270381689071655 lr:1.6153613497256249e-06 tokens_per_second_per_gpu:4484.614108075172 mfu:21.154689901857612 +Step 33367 | loss:0.3824869692325592 lr:1.6142682301156953e-06 tokens_per_second_per_gpu:5313.867835051666 mfu:25.066421217281057 +Step 33368 | loss:0.24018222093582153 lr:1.6131754744210193e-06 tokens_per_second_per_gpu:5334.337009848471 mfu:25.162977807198928 +Step 33369 | loss:0.2288844883441925 lr:1.612083082649829e-06 tokens_per_second_per_gpu:4997.343290005208 mfu:23.57332093364074 +Step 33370 | loss:0.2879476547241211 lr:1.6109910548103402e-06 tokens_per_second_per_gpu:5228.909977956291 mfu:24.665660510057045 +Step 33371 | loss:0.2821238934993744 lr:1.6098993909107573e-06 tokens_per_second_per_gpu:5316.082195386758 mfu:25.076866732790517 +Step 33372 | loss:0.38397833704948425 lr:1.6088080909593017e-06 tokens_per_second_per_gpu:5313.000870930015 mfu:25.062331599599894 +Step 33373 | loss:0.4263708293437958 lr:1.6077171549641666e-06 tokens_per_second_per_gpu:5052.519747868903 mfu:23.833597699458256 +Step 33374 | loss:0.32441219687461853 lr:1.6066265829335735e-06 tokens_per_second_per_gpu:4564.051029619841 mfu:21.529407414120207 +Step 33375 | loss:0.3347015678882599 lr:1.6055363748757102e-06 tokens_per_second_per_gpu:4024.246562352973 mfu:18.983057641883207 +Step 33376 | loss:0.3570481836795807 lr:1.6044465307987811e-06 tokens_per_second_per_gpu:4691.057930749542 mfu:22.128520636360836 +Step 33377 | loss:0.40458163619041443 lr:1.6033570507109963e-06 tokens_per_second_per_gpu:4510.045240753714 mfu:21.274652893700953 +Step 33378 | loss:0.2648872137069702 lr:1.602267934620527e-06 tokens_per_second_per_gpu:5345.907701222493 mfu:25.21755873257371 +Step 33379 | loss:0.37025800347328186 lr:1.6011791825355782e-06 tokens_per_second_per_gpu:4651.760335847986 mfu:21.94314717634913 +Step 33380 | loss:0.2514772415161133 lr:1.6000907944643428e-06 tokens_per_second_per_gpu:3862.535778607964 mfu:18.22024028425323 +Step 33381 | loss:0.347873330116272 lr:1.5990027704150035e-06 tokens_per_second_per_gpu:4657.519793044839 mfu:21.97031551861141 +Step 33382 | loss:0.2785680294036865 lr:1.5979151103957312e-06 tokens_per_second_per_gpu:4436.615475402033 mfu:20.928272162127215 +Step 33383 | loss:0.32856443524360657 lr:1.596827814414714e-06 tokens_per_second_per_gpu:3033.730450168208 mfu:14.310624141232259 +Step 33384 | loss:0.26612451672554016 lr:1.5957408824801346e-06 tokens_per_second_per_gpu:3422.9343196585605 mfu:16.146565198645856 +Step 33385 | loss:0.3285045921802521 lr:1.594654314600169e-06 tokens_per_second_per_gpu:5075.538966068923 mfu:23.942183279191365 +Step 33386 | loss:0.3092348873615265 lr:1.593568110782978e-06 tokens_per_second_per_gpu:4554.425008811479 mfu:21.483999831598656 +Step 33387 | loss:0.3250584602355957 lr:1.5924822710367438e-06 tokens_per_second_per_gpu:5372.722345853609 mfu:25.34404798634882 +Step 33388 | loss:0.38316401839256287 lr:1.5913967953696207e-06 tokens_per_second_per_gpu:5220.588721250522 mfu:24.62640772242314 +Step 33389 | loss:0.3211555778980255 lr:1.5903116837897858e-06 tokens_per_second_per_gpu:5335.097429829619 mfu:25.16656483799103 +Step 33390 | loss:0.27304938435554504 lr:1.589226936305388e-06 tokens_per_second_per_gpu:5209.155753049889 mfu:24.572476460755233 +Step 33391 | loss:0.3923864960670471 lr:1.588142552924593e-06 tokens_per_second_per_gpu:3895.1284479135015 mfu:18.373985466250257 +Step 33392 | loss:0.3833006024360657 lr:1.587058533655561e-06 tokens_per_second_per_gpu:5182.065974393794 mfu:24.444689352841586 +Step 33393 | loss:0.2633740305900574 lr:1.5859748785064354e-06 tokens_per_second_per_gpu:5311.167994348668 mfu:25.053685608044688 +Step 33394 | loss:0.26458457112312317 lr:1.584891587485371e-06 tokens_per_second_per_gpu:5297.298484277539 mfu:24.98826076265674 +Step 33395 | loss:0.24957585334777832 lr:1.5838086606005108e-06 tokens_per_second_per_gpu:4586.9266913884385 mfu:21.637315813673045 +Step 33396 | loss:0.37527215480804443 lr:1.58272609786001e-06 tokens_per_second_per_gpu:4590.548572760894 mfu:21.6544008460636 +Step 33397 | loss:0.2557106018066406 lr:1.5816438992720006e-06 tokens_per_second_per_gpu:3931.9623415984893 mfu:18.547737227272645 +Step 33398 | loss:0.3249709904193878 lr:1.5805620648446263e-06 tokens_per_second_per_gpu:5319.595545030748 mfu:25.09343980250048 +Step 33399 | loss:0.3005264401435852 lr:1.5794805945860246e-06 tokens_per_second_per_gpu:4317.1784044150745 mfu:20.3648671202161 +Step 33400 | loss:0.32298120856285095 lr:1.578399488504323e-06 tokens_per_second_per_gpu:5354.607523815216 mfu:25.258597280086377 +Step 33401 | loss:0.24278955161571503 lr:1.5773187466076643e-06 tokens_per_second_per_gpu:3939.268376534392 mfu:18.582201040603394 +Step 33402 | loss:0.26272791624069214 lr:1.5762383689041648e-06 tokens_per_second_per_gpu:3438.791841271548 mfu:16.221367833666495 +Step 33403 | loss:0.28703364729881287 lr:1.5751583554019622e-06 tokens_per_second_per_gpu:5382.381410172773 mfu:25.389611440748627 +Step 33404 | loss:0.2977867126464844 lr:1.574078706109161e-06 tokens_per_second_per_gpu:3892.069727680969 mfu:18.359556961041914 +Step 33405 | loss:0.30699196457862854 lr:1.5729994210338995e-06 tokens_per_second_per_gpu:4178.809579701719 mfu:19.71215776588731 +Step 33406 | loss:0.42652520537376404 lr:1.571920500184293e-06 tokens_per_second_per_gpu:5349.973001234396 mfu:25.236735446341562 +Step 33407 | loss:0.38954004645347595 lr:1.5708419435684462e-06 tokens_per_second_per_gpu:5307.1607676949625 mfu:25.034782836215793 +Step 33408 | loss:0.3208448588848114 lr:1.5697637511944753e-06 tokens_per_second_per_gpu:4815.632658387451 mfu:22.716160881269097 +Step 33409 | loss:0.28047797083854675 lr:1.5686859230704954e-06 tokens_per_second_per_gpu:4429.825098626102 mfu:20.896240796317404 +Step 33410 | loss:0.24580253660678864 lr:1.5676084592046115e-06 tokens_per_second_per_gpu:3206.9662409293455 mfu:15.127806923326279 +Step 33411 | loss:0.37871918082237244 lr:1.566531359604917e-06 tokens_per_second_per_gpu:4151.110958979212 mfu:19.581498646067235 +Step 33412 | loss:0.2334720492362976 lr:1.565454624279522e-06 tokens_per_second_per_gpu:5345.191559280452 mfu:25.214180568846686 +Step 33413 | loss:0.3469604551792145 lr:1.564378253236526e-06 tokens_per_second_per_gpu:4527.843845803492 mfu:21.358611950475915 +Step 33414 | loss:0.29558125138282776 lr:1.5633022464840162e-06 tokens_per_second_per_gpu:4932.754530802242 mfu:23.268644736501688 +Step 33415 | loss:0.2959662079811096 lr:1.5622266040300925e-06 tokens_per_second_per_gpu:4111.807755693164 mfu:19.39609873034898 +Step 33416 | loss:0.326766699552536 lr:1.5611513258828482e-06 tokens_per_second_per_gpu:5319.481649039297 mfu:25.092902535676142 +Step 33417 | loss:0.31192585825920105 lr:1.5600764120503653e-06 tokens_per_second_per_gpu:5219.995083436595 mfu:24.623607431569788 +Step 33418 | loss:0.24917110800743103 lr:1.5590018625407266e-06 tokens_per_second_per_gpu:4421.4208605815 mfu:20.85659657155015 +Step 33419 | loss:0.34985291957855225 lr:1.5579276773620143e-06 tokens_per_second_per_gpu:4816.207616445532 mfu:22.71887305652703 +Step 33420 | loss:0.28434452414512634 lr:1.5568538565223057e-06 tokens_per_second_per_gpu:3332.0213246230505 mfu:15.717713089706924 +Step 33421 | loss:0.38134923577308655 lr:1.5557804000296883e-06 tokens_per_second_per_gpu:5097.834050408308 mfu:24.04735299595388 +Step 33422 | loss:0.3843693733215332 lr:1.5547073078922224e-06 tokens_per_second_per_gpu:3967.643344703776 mfu:18.716050606727165 +Step 33423 | loss:0.3463153541088104 lr:1.5536345801179908e-06 tokens_per_second_per_gpu:3885.282162178865 mfu:18.3275388564904 +Step 33424 | loss:0.28971973061561584 lr:1.5525622167150533e-06 tokens_per_second_per_gpu:5342.907436034512 mfu:25.203405969783976 +Step 33425 | loss:0.2996508777141571 lr:1.5514902176914758e-06 tokens_per_second_per_gpu:5383.198000375089 mfu:25.393463436057605 +Step 33426 | loss:0.2597060203552246 lr:1.5504185830553186e-06 tokens_per_second_per_gpu:4293.432530314243 mfu:20.252853780618583 +Step 33427 | loss:0.3528187870979309 lr:1.5493473128146473e-06 tokens_per_second_per_gpu:5373.407228932564 mfu:25.347278696685354 +Step 33428 | loss:0.27473145723342896 lr:1.5482764069775224e-06 tokens_per_second_per_gpu:5361.816940232765 mfu:25.29260532738124 +Step 33429 | loss:0.3162483274936676 lr:1.5472058655519872e-06 tokens_per_second_per_gpu:4171.738329226007 mfu:19.67880147091364 +Step 33430 | loss:0.27189260721206665 lr:1.5461356885461075e-06 tokens_per_second_per_gpu:4928.30226069013 mfu:23.247642618747246 +Step 33431 | loss:0.38777053356170654 lr:1.545065875967916e-06 tokens_per_second_per_gpu:5365.282291539412 mfu:25.30895198074459 +Step 33432 | loss:0.2774873673915863 lr:1.5439964278254726e-06 tokens_per_second_per_gpu:5349.578055540125 mfu:25.234872420117004 +Step 33433 | loss:0.33148348331451416 lr:1.54292734412681e-06 tokens_per_second_per_gpu:5041.8223919693955 mfu:23.783136446523155 +Step 33434 | loss:0.29845699667930603 lr:1.5418586248799716e-06 tokens_per_second_per_gpu:4932.665686221212 mfu:23.268225641454524 +Step 33435 | loss:0.2948727309703827 lr:1.5407902700930065e-06 tokens_per_second_per_gpu:4818.50854756365 mfu:22.72972694119052 +Step 33436 | loss:0.23233051598072052 lr:1.5397222797739307e-06 tokens_per_second_per_gpu:5320.780916107307 mfu:25.099031400113766 +Step 33437 | loss:0.40571221709251404 lr:1.538654653930799e-06 tokens_per_second_per_gpu:5267.445853883015 mfu:24.847440811701816 +Step 33438 | loss:0.31913068890571594 lr:1.5375873925716211e-06 tokens_per_second_per_gpu:4267.246383538621 mfu:20.129329258460626 +Step 33439 | loss:0.34793782234191895 lr:1.5365204957044355e-06 tokens_per_second_per_gpu:4658.422125982295 mfu:21.974571977031264 +Step 33440 | loss:0.26609647274017334 lr:1.5354539633372632e-06 tokens_per_second_per_gpu:3645.0826684808017 mfu:17.19447686245705 +Step 33441 | loss:0.3205772042274475 lr:1.5343877954781204e-06 tokens_per_second_per_gpu:4604.294765305645 mfu:21.719243981639483 +Step 33442 | loss:0.34628427028656006 lr:1.5333219921350394e-06 tokens_per_second_per_gpu:5190.359073854462 mfu:24.48380931794623 +Step 33443 | loss:0.34707778692245483 lr:1.532256553316025e-06 tokens_per_second_per_gpu:4691.548993607951 mfu:22.13083706364792 +Step 33444 | loss:0.2534933388233185 lr:1.531191479029087e-06 tokens_per_second_per_gpu:3885.650905198967 mfu:18.329278280230266 +Step 33445 | loss:0.3654419481754303 lr:1.530126769282253e-06 tokens_per_second_per_gpu:4464.543718233016 mfu:21.060014448610545 +Step 33446 | loss:0.24286691844463348 lr:1.5290624240835216e-06 tokens_per_second_per_gpu:4771.142969048768 mfu:22.5062954259348 +Step 33447 | loss:0.2395523488521576 lr:1.5279984434408868e-06 tokens_per_second_per_gpu:5325.900069005582 mfu:25.123179317752477 +Step 33448 | loss:0.3646666705608368 lr:1.526934827362364e-06 tokens_per_second_per_gpu:4721.252861826297 mfu:22.270955278036368 +Step 33449 | loss:0.3623950779438019 lr:1.5258715758559528e-06 tokens_per_second_per_gpu:4014.020246175285 mfu:18.9348183636836 +Step 33450 | loss:0.1431276649236679 lr:1.5248086889296409e-06 tokens_per_second_per_gpu:5355.2589217706645 mfu:25.26167003538218 +Step 33451 | loss:0.276542067527771 lr:1.5237461665914332e-06 tokens_per_second_per_gpu:3326.362123838536 mfu:15.69101767404715 +Step 33452 | loss:0.30781394243240356 lr:1.5226840088493177e-06 tokens_per_second_per_gpu:3147.7885456459476 mfu:14.84865563791853 +Step 33453 | loss:0.28692111372947693 lr:1.5216222157112826e-06 tokens_per_second_per_gpu:5327.491741590402 mfu:25.130687508902422 +Step 33454 | loss:0.32702502608299255 lr:1.5205607871853044e-06 tokens_per_second_per_gpu:3864.6324907832036 mfu:18.230130833319823 +Step 33455 | loss:0.34124085307121277 lr:1.5194997232793773e-06 tokens_per_second_per_gpu:5038.494320495674 mfu:23.767437365554066 +Step 33456 | loss:0.32341963052749634 lr:1.518439024001478e-06 tokens_per_second_per_gpu:4379.360338955943 mfu:20.658189914777413 +Step 33457 | loss:0.5532448887825012 lr:1.5173786893595887e-06 tokens_per_second_per_gpu:4027.252076594434 mfu:18.99723516535379 +Step 33458 | loss:0.32834959030151367 lr:1.5163187193616812e-06 tokens_per_second_per_gpu:4120.0882565838365 mfu:19.43515926584942 +Step 33459 | loss:0.3070943355560303 lr:1.515259114015727e-06 tokens_per_second_per_gpu:4514.222024452038 mfu:21.29435549503873 +Step 33460 | loss:0.3901893198490143 lr:1.5141998733296969e-06 tokens_per_second_per_gpu:5151.6612879101385 mfu:24.301265259123742 +Step 33461 | loss:0.29094037413597107 lr:1.5131409973115518e-06 tokens_per_second_per_gpu:4197.868382881538 mfu:19.802061392253655 +Step 33462 | loss:0.28008803725242615 lr:1.5120824859692628e-06 tokens_per_second_per_gpu:5247.216874558082 mfu:24.752017264806334 +Step 33463 | loss:0.3841371536254883 lr:1.5110243393107848e-06 tokens_per_second_per_gpu:4429.196260224336 mfu:20.89327445828538 +Step 33464 | loss:0.23214773833751678 lr:1.5099665573440835e-06 tokens_per_second_per_gpu:3866.7026278810054 mfu:18.239896023211962 +Step 33465 | loss:0.3091205060482025 lr:1.5089091400771084e-06 tokens_per_second_per_gpu:4584.657743326486 mfu:21.626612798542354 +Step 33466 | loss:0.3687650263309479 lr:1.5078520875178248e-06 tokens_per_second_per_gpu:4832.67553495444 mfu:22.796555037850172 +Step 33467 | loss:0.3360249996185303 lr:1.5067953996741601e-06 tokens_per_second_per_gpu:4034.727930070024 mfu:19.03250004171027 +Step 33468 | loss:0.3411245346069336 lr:1.5057390765540857e-06 tokens_per_second_per_gpu:5048.422078099155 mfu:23.814268292019797 +Step 33469 | loss:0.1966256946325302 lr:1.5046831181655286e-06 tokens_per_second_per_gpu:4244.046015177799 mfu:20.01988916251161 +Step 33470 | loss:0.37590599060058594 lr:1.5036275245164377e-06 tokens_per_second_per_gpu:4660.210728680178 mfu:21.983009121124265 +Step 33471 | loss:0.33214542269706726 lr:1.502572295614757e-06 tokens_per_second_per_gpu:4699.339299981061 mfu:22.167585267973205 +Step 33472 | loss:0.23911651968955994 lr:1.5015174314684132e-06 tokens_per_second_per_gpu:4287.6828064271385 mfu:20.225731352039062 +Step 33473 | loss:0.2878839671611786 lr:1.5004629320853503e-06 tokens_per_second_per_gpu:5319.947284880502 mfu:25.095099019384772 +Step 33474 | loss:0.2083371877670288 lr:1.4994087974734948e-06 tokens_per_second_per_gpu:4686.282622231308 mfu:22.10599469132868 +Step 33475 | loss:0.25133660435676575 lr:1.4983550276407742e-06 tokens_per_second_per_gpu:3990.852123239193 mfu:18.82553037490467 +Step 33476 | loss:0.255543977022171 lr:1.4973016225951097e-06 tokens_per_second_per_gpu:4685.666603268838 mfu:22.103088824779036 +Step 33477 | loss:0.23904669284820557 lr:1.4962485823444284e-06 tokens_per_second_per_gpu:5317.038513579307 mfu:25.081377848869476 +Step 33478 | loss:0.28954657912254333 lr:1.495195906896657e-06 tokens_per_second_per_gpu:3382.2173010813126 mfu:15.954496074977254 +Step 33479 | loss:0.3144069314002991 lr:1.4941435962597005e-06 tokens_per_second_per_gpu:5338.381102166887 mfu:25.182054480658195 +Step 33480 | loss:0.32274022698402405 lr:1.493091650441475e-06 tokens_per_second_per_gpu:4457.055033022379 mfu:21.024689042770518 +Step 33481 | loss:0.34781599044799805 lr:1.4920400694499071e-06 tokens_per_second_per_gpu:4097.758220255064 mfu:19.329824674589542 +Step 33482 | loss:0.29593560099601746 lr:1.4909888532928963e-06 tokens_per_second_per_gpu:5372.368992271928 mfu:25.342381157215694 +Step 33483 | loss:0.2440924346446991 lr:1.4899380019783415e-06 tokens_per_second_per_gpu:5309.803591731751 mfu:25.047249488109603 +Step 33484 | loss:0.3743968605995178 lr:1.4888875155141534e-06 tokens_per_second_per_gpu:5265.928813207691 mfu:24.84028467200248 +Step 33485 | loss:0.3276063799858093 lr:1.4878373939082368e-06 tokens_per_second_per_gpu:5083.775056607106 mfu:23.98103432348192 +Step 33486 | loss:0.23692165315151215 lr:1.4867876371684798e-06 tokens_per_second_per_gpu:3820.838203970762 mfu:18.02354571034971 +Step 33487 | loss:0.25209105014801025 lr:1.4857382453027812e-06 tokens_per_second_per_gpu:4691.039411905115 mfu:22.128433279812036 +Step 33488 | loss:0.22507219016551971 lr:1.4846892183190464e-06 tokens_per_second_per_gpu:5332.19117717493 mfu:25.152855548361032 +Step 33489 | loss:0.39127281308174133 lr:1.4836405562251522e-06 tokens_per_second_per_gpu:3837.856226723305 mfu:18.103822627247542 +Step 33490 | loss:0.2947087585926056 lr:1.482592259028981e-06 tokens_per_second_per_gpu:3378.2008658819027 mfu:15.935549864867129 +Step 33491 | loss:0.2424621880054474 lr:1.4815443267384266e-06 tokens_per_second_per_gpu:5097.870583754235 mfu:24.047525329979845 +Step 33492 | loss:0.38700246810913086 lr:1.4804967593613716e-06 tokens_per_second_per_gpu:5324.1835310898905 mfu:25.115082115532594 +Step 33493 | loss:0.2956412434577942 lr:1.4794495569056876e-06 tokens_per_second_per_gpu:5333.6896807183975 mfu:25.159924245246632 +Step 33494 | loss:0.38215774297714233 lr:1.4784027193792515e-06 tokens_per_second_per_gpu:4481.611854085984 mfu:21.14052775755326 +Step 33495 | loss:0.39603468775749207 lr:1.4773562467899515e-06 tokens_per_second_per_gpu:5321.582310553418 mfu:25.102811714448066 +Step 33496 | loss:0.4013904929161072 lr:1.4763101391456424e-06 tokens_per_second_per_gpu:5350.68031065781 mfu:25.240071945571234 +Step 33497 | loss:0.2583504319190979 lr:1.4752643964541902e-06 tokens_per_second_per_gpu:5331.1022860427165 mfu:25.147719063106393 +Step 33498 | loss:0.4127425253391266 lr:1.4742190187234718e-06 tokens_per_second_per_gpu:4589.270609118349 mfu:21.648372473518634 +Step 33499 | loss:0.3057727515697479 lr:1.4731740059613364e-06 tokens_per_second_per_gpu:4233.754063416445 mfu:19.971340270065266 +Step 33500 | loss:0.31927788257598877 lr:1.4721293581756611e-06 tokens_per_second_per_gpu:5342.610748558617 mfu:25.20200644434008 +Step 33501 | loss:0.2768633961677551 lr:1.47108507537429e-06 tokens_per_second_per_gpu:4495.191036379238 mfu:21.204583077277942 +Step 33502 | loss:0.3818866014480591 lr:1.4700411575650775e-06 tokens_per_second_per_gpu:5059.973366800787 mfu:23.868757691678475 +Step 33503 | loss:0.33633220195770264 lr:1.468997604755884e-06 tokens_per_second_per_gpu:4434.173008274598 mfu:20.91675062795924 +Step 33504 | loss:0.3315897285938263 lr:1.4679544169545479e-06 tokens_per_second_per_gpu:5234.607991672257 mfu:24.692539012936642 +Step 33505 | loss:0.22529146075248718 lr:1.4669115941689183e-06 tokens_per_second_per_gpu:5242.526231872439 mfu:24.729890702951987 +Step 33506 | loss:0.4036411643028259 lr:1.465869136406839e-06 tokens_per_second_per_gpu:5065.078824938116 mfu:23.892841008793237 +Step 33507 | loss:0.3090493977069855 lr:1.4648270436761592e-06 tokens_per_second_per_gpu:3749.7550280819096 mfu:17.68823424164158 +Step 33508 | loss:0.3155156970024109 lr:1.463785315984695e-06 tokens_per_second_per_gpu:4412.312041416928 mfu:20.813628717426056 +Step 33509 | loss:0.274117112159729 lr:1.4627439533403009e-06 tokens_per_second_per_gpu:5358.398186967256 mfu:25.276478484928376 +Step 33510 | loss:0.3470701575279236 lr:1.46170295575081e-06 tokens_per_second_per_gpu:4428.628874640886 mfu:20.890598003682285 +Step 33511 | loss:0.20086407661437988 lr:1.460662323224038e-06 tokens_per_second_per_gpu:5372.561452163986 mfu:25.343289023363578 +Step 33512 | loss:0.35342028737068176 lr:1.4596220557678176e-06 tokens_per_second_per_gpu:5337.690020909217 mfu:25.178794532455147 +Step 33513 | loss:0.3377056419849396 lr:1.4585821533899757e-06 tokens_per_second_per_gpu:3317.1617303170137 mfu:15.647617848057127 +Step 33514 | loss:0.34558698534965515 lr:1.4575426160983285e-06 tokens_per_second_per_gpu:4787.910218749066 mfu:22.5853894035589 +Step 33515 | loss:0.24190089106559753 lr:1.4565034439006974e-06 tokens_per_second_per_gpu:5349.818569085305 mfu:25.236006963545456 +Step 33516 | loss:0.29360637068748474 lr:1.4554646368048985e-06 tokens_per_second_per_gpu:4729.241444544171 mfu:22.308638785709338 +Step 33517 | loss:0.31842511892318726 lr:1.4544261948187477e-06 tokens_per_second_per_gpu:4073.9224671235434 mfu:19.217387360269317 +Step 33518 | loss:0.22218206524848938 lr:1.4533881179500551e-06 tokens_per_second_per_gpu:5320.4853033766485 mfu:25.09763694442652 +Step 33519 | loss:0.3747468590736389 lr:1.4523504062066207e-06 tokens_per_second_per_gpu:5279.479890395814 mfu:24.904207415151042 +Step 33520 | loss:0.29444822669029236 lr:1.4513130595962487e-06 tokens_per_second_per_gpu:5366.192834425166 mfu:25.313247166891333 +Step 33521 | loss:0.2769823670387268 lr:1.4502760781267555e-06 tokens_per_second_per_gpu:4415.43915959045 mfu:20.828379867391934 +Step 33522 | loss:0.2354440987110138 lr:1.4492394618059236e-06 tokens_per_second_per_gpu:5334.234814056645 mfu:25.162495732212484 +Step 33523 | loss:0.3184266984462738 lr:1.4482032106415522e-06 tokens_per_second_per_gpu:4866.337077976054 mfu:22.95534227953459 +Step 33524 | loss:0.25399044156074524 lr:1.4471673246414518e-06 tokens_per_second_per_gpu:5340.37089199907 mfu:25.19144065878976 +Step 33525 | loss:0.28982090950012207 lr:1.4461318038133997e-06 tokens_per_second_per_gpu:5350.701270212086 mfu:25.24017081536482 +Step 33526 | loss:0.2269280105829239 lr:1.4450966481651785e-06 tokens_per_second_per_gpu:5340.563914640183 mfu:25.192351179520948 +Step 33527 | loss:0.2596167325973511 lr:1.4440618577045818e-06 tokens_per_second_per_gpu:4666.845776754245 mfu:22.014307774945575 +Step 33528 | loss:0.30721431970596313 lr:1.4430274324393977e-06 tokens_per_second_per_gpu:5363.331174206636 mfu:25.299748227390943 +Step 33529 | loss:0.24976015090942383 lr:1.441993372377398e-06 tokens_per_second_per_gpu:4441.714502457884 mfu:20.95232514273339 +Step 33530 | loss:0.3680259883403778 lr:1.4409596775263545e-06 tokens_per_second_per_gpu:3782.212580220574 mfu:17.8413420529091 +Step 33531 | loss:0.31648504734039307 lr:1.4399263478940606e-06 tokens_per_second_per_gpu:4943.859026014924 mfu:23.32102653504216 +Step 33532 | loss:0.31463348865509033 lr:1.4388933834882767e-06 tokens_per_second_per_gpu:4601.756204160311 mfu:21.707269155593746 +Step 33533 | loss:0.2705141007900238 lr:1.4378607843167635e-06 tokens_per_second_per_gpu:5323.309440812722 mfu:25.11095888256001 +Step 33534 | loss:0.3289608359336853 lr:1.436828550387298e-06 tokens_per_second_per_gpu:3920.190423523686 mfu:18.492207081217465 +Step 33535 | loss:0.2759833335876465 lr:1.4357966817076463e-06 tokens_per_second_per_gpu:5355.785308330427 mfu:25.26415309059539 +Step 33536 | loss:0.3232310712337494 lr:1.434765178285563e-06 tokens_per_second_per_gpu:5320.360385587105 mfu:25.09704768590007 +Step 33537 | loss:0.2446596771478653 lr:1.4337340401288036e-06 tokens_per_second_per_gpu:4481.168530034092 mfu:21.138436522361605 +Step 33538 | loss:0.23912127315998077 lr:1.432703267245128e-06 tokens_per_second_per_gpu:5372.216412668506 mfu:25.341661413193584 +Step 33539 | loss:0.2669920027256012 lr:1.4316728596422967e-06 tokens_per_second_per_gpu:3566.5837981455365 mfu:16.824184297797252 +Step 33540 | loss:0.3425636291503906 lr:1.4306428173280483e-06 tokens_per_second_per_gpu:3493.4116440113175 mfu:16.479018762289293 +Step 33541 | loss:0.3425348997116089 lr:1.4296131403101266e-06 tokens_per_second_per_gpu:3500.1173695452726 mfu:16.51065081374771 +Step 33542 | loss:0.3888935446739197 lr:1.428583828596286e-06 tokens_per_second_per_gpu:3852.1011046619205 mfu:18.171018146910786 +Step 33543 | loss:0.3282485604286194 lr:1.4275548821942652e-06 tokens_per_second_per_gpu:4245.6207418969225 mfu:20.027317417121804 +Step 33544 | loss:0.34615668654441833 lr:1.426526301111797e-06 tokens_per_second_per_gpu:5279.435479564404 mfu:24.903997921682006 +Step 33545 | loss:0.2542230486869812 lr:1.4254980853566247e-06 tokens_per_second_per_gpu:5323.785088690083 mfu:25.11320259475142 +Step 33546 | loss:0.28401827812194824 lr:1.424470234936487e-06 tokens_per_second_per_gpu:4707.604930644861 mfu:22.20657565807636 +Step 33547 | loss:0.22024144232273102 lr:1.4234427498591052e-06 tokens_per_second_per_gpu:3073.7303389077097 mfu:14.499310441101247 +Step 33548 | loss:0.2944810390472412 lr:1.4224156301322012e-06 tokens_per_second_per_gpu:5380.449669046419 mfu:25.380499088266596 +Step 33549 | loss:0.25802960991859436 lr:1.4213888757635075e-06 tokens_per_second_per_gpu:4976.454844068414 mfu:23.47478657022051 +Step 33550 | loss:0.30947253108024597 lr:1.4203624867607513e-06 tokens_per_second_per_gpu:4343.768870891322 mfu:20.49029888738596 +Step 33551 | loss:0.23172366619110107 lr:1.4193364631316485e-06 tokens_per_second_per_gpu:5244.0216123820155 mfu:24.736944667953885 +Step 33552 | loss:0.23871174454689026 lr:1.41831080488391e-06 tokens_per_second_per_gpu:5221.094448818326 mfu:24.628793325646487 +Step 33553 | loss:0.29137352108955383 lr:1.4172855120252627e-06 tokens_per_second_per_gpu:3735.7676282460475 mfu:17.62225329011932 +Step 33554 | loss:0.29153183102607727 lr:1.416260584563406e-06 tokens_per_second_per_gpu:5337.131964399512 mfu:25.17616208468461 +Step 33555 | loss:0.3457387685775757 lr:1.4152360225060501e-06 tokens_per_second_per_gpu:4452.651978488973 mfu:21.003919083297877 +Step 33556 | loss:0.2567562460899353 lr:1.414211825860906e-06 tokens_per_second_per_gpu:5361.368446498171 mfu:25.290489705914233 +Step 33557 | loss:0.2229907065629959 lr:1.4131879946356785e-06 tokens_per_second_per_gpu:4042.4911987462046 mfu:19.06912070460604 +Step 33558 | loss:0.3388465940952301 lr:1.4121645288380557e-06 tokens_per_second_per_gpu:4097.820296219681 mfu:19.33011749750589 +Step 33559 | loss:0.3756035268306732 lr:1.4111414284757484e-06 tokens_per_second_per_gpu:5358.422820558225 mfu:25.276594685779898 +Step 33560 | loss:0.31144458055496216 lr:1.4101186935564447e-06 tokens_per_second_per_gpu:3477.0748988271 mfu:16.401955547919187 +Step 33561 | loss:0.31369662284851074 lr:1.409096324087844e-06 tokens_per_second_per_gpu:4646.5267716937915 mfu:21.91845956129567 +Step 33562 | loss:0.4120028018951416 lr:1.4080743200776237e-06 tokens_per_second_per_gpu:5373.6100580158245 mfu:25.348235476077306 +Step 33563 | loss:0.2672872245311737 lr:1.4070526815334773e-06 tokens_per_second_per_gpu:5170.368140355884 mfu:24.38950867382847 +Step 33564 | loss:0.24398574233055115 lr:1.4060314084630877e-06 tokens_per_second_per_gpu:4950.370428283103 mfu:23.351741930500946 +Step 33565 | loss:0.23462598025798798 lr:1.4050105008741377e-06 tokens_per_second_per_gpu:4698.294585855682 mfu:22.162657173197182 +Step 33566 | loss:0.30828049778938293 lr:1.4039899587743044e-06 tokens_per_second_per_gpu:5388.362887506955 mfu:25.417827089879236 +Step 33567 | loss:0.2962580621242523 lr:1.4029697821712595e-06 tokens_per_second_per_gpu:4320.4464033524655 mfu:20.38028282878189 +Step 33568 | loss:0.29632508754730225 lr:1.4019499710726913e-06 tokens_per_second_per_gpu:4815.194974755243 mfu:22.714096252899445 +Step 33569 | loss:0.3289749324321747 lr:1.4009305254862493e-06 tokens_per_second_per_gpu:4097.180744837754 mfu:19.32710062451862 +Step 33570 | loss:0.3710537850856781 lr:1.3999114454196104e-06 tokens_per_second_per_gpu:3864.9307562862527 mfu:18.231537802587045 +Step 33571 | loss:0.3774159252643585 lr:1.3988927308804411e-06 tokens_per_second_per_gpu:5356.289119988006 mfu:25.26652965614332 +Step 33572 | loss:0.32369232177734375 lr:1.3978743818764018e-06 tokens_per_second_per_gpu:5054.872012179791 mfu:23.84469373155039 +Step 33573 | loss:0.3675220012664795 lr:1.3968563984151473e-06 tokens_per_second_per_gpu:4021.971654850872 mfu:18.972326515056555 +Step 33574 | loss:0.33334892988204956 lr:1.3958387805043384e-06 tokens_per_second_per_gpu:5127.664573958876 mfu:24.188068665154052 +Step 33575 | loss:0.2537150979042053 lr:1.3948215281516353e-06 tokens_per_second_per_gpu:4647.361556956551 mfu:21.922397385809155 +Step 33576 | loss:0.3203871250152588 lr:1.3938046413646766e-06 tokens_per_second_per_gpu:5029.622408142865 mfu:23.725587041281532 +Step 33577 | loss:0.28556838631629944 lr:1.3927881201511172e-06 tokens_per_second_per_gpu:4306.568326376834 mfu:20.31481757184374 +Step 33578 | loss:0.2541765570640564 lr:1.3917719645185956e-06 tokens_per_second_per_gpu:4282.1461194034255 mfu:20.19961385469199 +Step 33579 | loss:0.28587329387664795 lr:1.390756174474761e-06 tokens_per_second_per_gpu:4455.356955395331 mfu:21.01667892088182 +Step 33580 | loss:0.2958938777446747 lr:1.3897407500272518e-06 tokens_per_second_per_gpu:3683.322618794556 mfu:17.37486123249533 +Step 33581 | loss:0.3039904534816742 lr:1.3887256911837065e-06 tokens_per_second_per_gpu:3968.9085345128724 mfu:18.722018722920314 +Step 33582 | loss:0.2448737770318985 lr:1.3877109979517633e-06 tokens_per_second_per_gpu:5220.317968266045 mfu:24.625130534400135 +Step 33583 | loss:0.25157999992370605 lr:1.3866966703390438e-06 tokens_per_second_per_gpu:4231.04388577651 mfu:19.958555899733707 +Step 33584 | loss:0.2551710307598114 lr:1.3856827083531809e-06 tokens_per_second_per_gpu:3758.210391229538 mfu:17.728119632242656 +Step 33585 | loss:0.29013776779174805 lr:1.3846691120017964e-06 tokens_per_second_per_gpu:5102.645382343064 mfu:24.070048869588 +Step 33586 | loss:0.40144190192222595 lr:1.3836558812925283e-06 tokens_per_second_per_gpu:4234.996031631172 mfu:19.97719884603576 +Step 33587 | loss:0.288717120885849 lr:1.3826430162329818e-06 tokens_per_second_per_gpu:5181.72788631699 mfu:24.443094533699526 +Step 33588 | loss:0.35403206944465637 lr:1.381630516830784e-06 tokens_per_second_per_gpu:4050.6575705274813 mfu:19.107642873625025 +Step 33589 | loss:0.32766062021255493 lr:1.380618383093546e-06 tokens_per_second_per_gpu:5360.507314782802 mfu:25.28642760068862 +Step 33590 | loss:0.42460864782333374 lr:1.3796066150288832e-06 tokens_per_second_per_gpu:5141.477356035385 mfu:24.253225914912584 +Step 33591 | loss:0.2978866398334503 lr:1.3785952126444013e-06 tokens_per_second_per_gpu:4742.0067431733805 mfu:22.368854877327855 +Step 33592 | loss:0.3719688951969147 lr:1.3775841759477048e-06 tokens_per_second_per_gpu:5059.186665083411 mfu:23.86504668545246 +Step 33593 | loss:0.2818593978881836 lr:1.3765735049464101e-06 tokens_per_second_per_gpu:4701.298571345401 mfu:22.17682748528496 +Step 33594 | loss:0.30798643827438354 lr:1.3755631996481e-06 tokens_per_second_per_gpu:5310.817177584609 mfu:25.05203074551311 +Step 33595 | loss:0.3059755861759186 lr:1.3745532600603905e-06 tokens_per_second_per_gpu:4892.392573487007 mfu:23.0782504973857 +Step 33596 | loss:0.2107168436050415 lr:1.3735436861908701e-06 tokens_per_second_per_gpu:5325.007015904355 mfu:25.118966633903423 +Step 33597 | loss:0.23749518394470215 lr:1.3725344780471384e-06 tokens_per_second_per_gpu:3322.527858292221 mfu:15.672930789272021 +Step 33598 | loss:0.30324026942253113 lr:1.3715256356367668e-06 tokens_per_second_per_gpu:5347.568907840242 mfu:25.22539492762071 +Step 33599 | loss:0.2602923810482025 lr:1.3705171589673604e-06 tokens_per_second_per_gpu:4770.985708628334 mfu:22.505553601700218 +Step 33600 | loss:0.2640370726585388 lr:1.3695090480465078e-06 tokens_per_second_per_gpu:5394.148562012961 mfu:25.44511910366192 +Step 33601 | loss:0.2753731310367584 lr:1.368501302881775e-06 tokens_per_second_per_gpu:3662.894026379225 mfu:17.278496076594937 +Step 33602 | loss:0.31478050351142883 lr:1.3674939234807449e-06 tokens_per_second_per_gpu:4736.765748965891 mfu:22.344132213445306 +Step 33603 | loss:0.31139999628067017 lr:1.3664869098510058e-06 tokens_per_second_per_gpu:4525.432081685912 mfu:21.3472352476439 +Step 33604 | loss:0.28893202543258667 lr:1.3654802620001295e-06 tokens_per_second_per_gpu:5345.691287532088 mfu:25.216537872272085 +Step 33605 | loss:0.33999955654144287 lr:1.3644739799356764e-06 tokens_per_second_per_gpu:5314.973535651671 mfu:25.071636995663255 +Step 33606 | loss:0.4090971350669861 lr:1.3634680636652131e-06 tokens_per_second_per_gpu:3520.8675845069665 mfu:16.608533118073762 +Step 33607 | loss:0.3198040723800659 lr:1.3624625131963166e-06 tokens_per_second_per_gpu:5061.150003171663 mfu:23.874308086194766 +Step 33608 | loss:0.31261298060417175 lr:1.3614573285365528e-06 tokens_per_second_per_gpu:5353.984693187203 mfu:25.255659281748184 +Step 33609 | loss:0.3082073926925659 lr:1.3604525096934717e-06 tokens_per_second_per_gpu:4281.116998480242 mfu:20.194759315711195 +Step 33610 | loss:0.338615357875824 lr:1.3594480566746282e-06 tokens_per_second_per_gpu:4495.765950338879 mfu:21.207295044516005 +Step 33611 | loss:0.34452444314956665 lr:1.3584439694875883e-06 tokens_per_second_per_gpu:5053.817634709409 mfu:23.839720053127248 +Step 33612 | loss:0.32831713557243347 lr:1.3574402481399017e-06 tokens_per_second_per_gpu:5319.727410051909 mfu:25.094061832303787 +Step 33613 | loss:0.2845157980918884 lr:1.3564368926391014e-06 tokens_per_second_per_gpu:5349.348780662107 mfu:25.23379089139906 +Step 33614 | loss:0.24399623274803162 lr:1.3554339029927532e-06 tokens_per_second_per_gpu:5378.924176005007 mfu:25.3733030773143 +Step 33615 | loss:0.3617783188819885 lr:1.3544312792083957e-06 tokens_per_second_per_gpu:4299.22243693589 mfu:20.280165758012828 +Step 33616 | loss:0.2688971161842346 lr:1.3534290212935674e-06 tokens_per_second_per_gpu:5318.11917082163 mfu:25.086475493460654 +Step 33617 | loss:0.3282797038555145 lr:1.352427129255801e-06 tokens_per_second_per_gpu:4461.136744024651 mfu:21.043943169979947 +Step 33618 | loss:0.40986374020576477 lr:1.3514256031026461e-06 tokens_per_second_per_gpu:4494.354357134513 mfu:21.200636318528375 +Step 33619 | loss:0.20172244310379028 lr:1.3504244428416246e-06 tokens_per_second_per_gpu:3418.072478336171 mfu:16.12363106361248 +Step 33620 | loss:0.4266747832298279 lr:1.3494236484802636e-06 tokens_per_second_per_gpu:5382.475268640861 mfu:25.390054187156505 +Step 33621 | loss:0.33905524015426636 lr:1.348423220026096e-06 tokens_per_second_per_gpu:5096.533072642594 mfu:24.04121606186331 +Step 33622 | loss:0.29311344027519226 lr:1.347423157486649e-06 tokens_per_second_per_gpu:4449.191067163582 mfu:20.987593374083648 +Step 33623 | loss:0.28332191705703735 lr:1.3464234608694392e-06 tokens_per_second_per_gpu:5353.437364916032 mfu:25.25307744090923 +Step 33624 | loss:0.3055731952190399 lr:1.3454241301819826e-06 tokens_per_second_per_gpu:4192.783347610305 mfu:19.778074413282766 +Step 33625 | loss:0.26676666736602783 lr:1.3444251654318063e-06 tokens_per_second_per_gpu:5318.782746160446 mfu:25.089605691551473 +Step 33626 | loss:0.3689448833465576 lr:1.3434265666264157e-06 tokens_per_second_per_gpu:5342.059585392394 mfu:25.19940651364687 +Step 33627 | loss:0.22609032690525055 lr:1.3424283337733212e-06 tokens_per_second_per_gpu:5352.853439817559 mfu:25.25032296659184 +Step 33628 | loss:0.3222527503967285 lr:1.3414304668800282e-06 tokens_per_second_per_gpu:4627.538044606609 mfu:21.82888649581451 +Step 33629 | loss:0.39764460921287537 lr:1.3404329659540528e-06 tokens_per_second_per_gpu:5327.922633266197 mfu:25.132720098454847 +Step 33630 | loss:0.366108775138855 lr:1.3394358310028832e-06 tokens_per_second_per_gpu:4782.812944973604 mfu:22.561344693475462 +Step 33631 | loss:0.25116074085235596 lr:1.3384390620340249e-06 tokens_per_second_per_gpu:2986.4740311394326 mfu:14.087707549897033 +Step 33632 | loss:0.28366902470588684 lr:1.3374426590549828e-06 tokens_per_second_per_gpu:5310.1041477943945 mfu:25.04866726233724 +Step 33633 | loss:0.2308967560529709 lr:1.3364466220732452e-06 tokens_per_second_per_gpu:4167.649852287924 mfu:19.659515427630605 +Step 33634 | loss:0.25525593757629395 lr:1.3354509510962954e-06 tokens_per_second_per_gpu:3833.4584517671624 mfu:18.08307757244164 +Step 33635 | loss:0.3022153377532959 lr:1.3344556461316272e-06 tokens_per_second_per_gpu:4673.03215266507 mfu:22.043489965621504 +Step 33636 | loss:0.4348645508289337 lr:1.3334607071867344e-06 tokens_per_second_per_gpu:5325.185450396671 mfu:25.119808339847737 +Step 33637 | loss:0.31418877840042114 lr:1.3324661342690892e-06 tokens_per_second_per_gpu:4969.294926600881 mfu:23.441012017918265 +Step 33638 | loss:0.3024449348449707 lr:1.3314719273861742e-06 tokens_per_second_per_gpu:5023.762797342067 mfu:23.69794625738147 +Step 33639 | loss:0.2601704001426697 lr:1.3304780865454724e-06 tokens_per_second_per_gpu:5133.644933108519 mfu:24.2162790396174 +Step 33640 | loss:0.24857565760612488 lr:1.3294846117544613e-06 tokens_per_second_per_gpu:3756.0730245901996 mfu:17.718037309132544 +Step 33641 | loss:0.394660621881485 lr:1.3284915030205959e-06 tokens_per_second_per_gpu:4572.726282720867 mfu:21.570330063148553 +Step 33642 | loss:0.1651478409767151 lr:1.327498760351359e-06 tokens_per_second_per_gpu:4227.300576775185 mfu:19.940898072500076 +Step 33643 | loss:0.31895744800567627 lr:1.3265063837542169e-06 tokens_per_second_per_gpu:5348.004936142977 mfu:25.22745174751881 +Step 33644 | loss:0.3564845025539398 lr:1.3255143732366305e-06 tokens_per_second_per_gpu:4142.420524811041 mfu:19.54050438535501 +Step 33645 | loss:0.31878912448883057 lr:1.3245227288060601e-06 tokens_per_second_per_gpu:4200.245681735765 mfu:19.81327551655805 +Step 33646 | loss:0.2541131377220154 lr:1.323531450469967e-06 tokens_per_second_per_gpu:5383.863759960135 mfu:25.396603937610568 +Step 33647 | loss:0.30576515197753906 lr:1.3225405382358114e-06 tokens_per_second_per_gpu:5357.0438643490115 mfu:25.27008991406685 +Step 33648 | loss:0.3481363356113434 lr:1.3215499921110319e-06 tokens_per_second_per_gpu:5326.612129601193 mfu:25.126538229072043 +Step 33649 | loss:0.3396800756454468 lr:1.3205598121030837e-06 tokens_per_second_per_gpu:3600.4183335468515 mfu:16.983787574052265 +Step 33650 | loss:0.292453795671463 lr:1.319569998219422e-06 tokens_per_second_per_gpu:4609.295337482756 mfu:21.742832533784505 +Step 33651 | loss:0.3211475908756256 lr:1.3185805504674852e-06 tokens_per_second_per_gpu:5338.147759789284 mfu:25.180953764850727 +Step 33652 | loss:0.39869821071624756 lr:1.317591468854712e-06 tokens_per_second_per_gpu:4710.632159628895 mfu:22.220855613692574 +Step 33653 | loss:0.30452391505241394 lr:1.316602753388546e-06 tokens_per_second_per_gpu:4975.196797240877 mfu:23.46885214869009 +Step 33654 | loss:0.4343164265155792 lr:1.3156144040764317e-06 tokens_per_second_per_gpu:4833.971104188992 mfu:22.80266645897646 +Step 33655 | loss:0.3516347110271454 lr:1.3146264209257907e-06 tokens_per_second_per_gpu:5303.343328905141 mfu:25.016775326121987 +Step 33656 | loss:0.21407660841941833 lr:1.3136388039440507e-06 tokens_per_second_per_gpu:5357.53781164247 mfu:25.272419947726192 +Step 33657 | loss:0.30160874128341675 lr:1.3126515531386441e-06 tokens_per_second_per_gpu:5365.30603285316 mfu:25.309063972572226 +Step 33658 | loss:0.28144121170043945 lr:1.3116646685170042e-06 tokens_per_second_per_gpu:5321.77820541261 mfu:25.10373578392198 +Step 33659 | loss:0.3298863172531128 lr:1.3106781500865417e-06 tokens_per_second_per_gpu:4044.2646291519713 mfu:19.077486278408468 +Step 33660 | loss:0.2876414358615875 lr:1.309691997854684e-06 tokens_per_second_per_gpu:5341.67010853615 mfu:25.197569284864365 +Step 33661 | loss:0.3568081855773926 lr:1.3087062118288474e-06 tokens_per_second_per_gpu:5130.1207244593625 mfu:24.199654746127333 +Step 33662 | loss:0.38889530301094055 lr:1.3077207920164481e-06 tokens_per_second_per_gpu:3983.837268368572 mfu:18.792440107622294 +Step 33663 | loss:0.37943366169929504 lr:1.3067357384248857e-06 tokens_per_second_per_gpu:5384.440213635702 mfu:25.39932316795136 +Step 33664 | loss:0.2881016433238983 lr:1.3057510510615823e-06 tokens_per_second_per_gpu:4445.784321693612 mfu:20.971523174451082 +Step 33665 | loss:0.26545482873916626 lr:1.3047667299339371e-06 tokens_per_second_per_gpu:3220.2391419699898 mfu:15.190417462125431 +Step 33666 | loss:0.2901272475719452 lr:1.3037827750493503e-06 tokens_per_second_per_gpu:4056.7072965264324 mfu:19.136180463351767 +Step 33667 | loss:0.3066953122615814 lr:1.3027991864152323e-06 tokens_per_second_per_gpu:5170.7294380650455 mfu:24.39121297676633 +Step 33668 | loss:0.36895379424095154 lr:1.3018159640389771e-06 tokens_per_second_per_gpu:5292.68905523046 mfu:24.96651729939198 +Step 33669 | loss:0.3548851013183594 lr:1.300833107927979e-06 tokens_per_second_per_gpu:5319.595960039407 mfu:25.09344176016713 +Step 33670 | loss:0.3083725571632385 lr:1.2998506180896208e-06 tokens_per_second_per_gpu:5378.498637509727 mfu:25.371295739627712 +Step 33671 | loss:0.28789079189300537 lr:1.2988684945313023e-06 tokens_per_second_per_gpu:5330.946100516121 mfu:25.146982309329445 +Step 33672 | loss:0.28404179215431213 lr:1.2978867372604175e-06 tokens_per_second_per_gpu:4986.835944488486 mfu:23.52375599210036 +Step 33673 | loss:0.35000312328338623 lr:1.2969053462843273e-06 tokens_per_second_per_gpu:5341.243014979139 mfu:25.195554611686266 +Step 33674 | loss:0.4292306900024414 lr:1.295924321610431e-06 tokens_per_second_per_gpu:5121.077329953527 mfu:24.156995511281096 +Step 33675 | loss:0.27456408739089966 lr:1.2949436632461065e-06 tokens_per_second_per_gpu:3888.5555076677583 mfu:18.34297978565184 +Step 33676 | loss:0.34444576501846313 lr:1.293963371198731e-06 tokens_per_second_per_gpu:4526.918140194539 mfu:21.35424523917682 +Step 33677 | loss:0.27218085527420044 lr:1.292983445475665e-06 tokens_per_second_per_gpu:4293.181954582962 mfu:20.251671772141506 +Step 33678 | loss:0.30584797263145447 lr:1.2920038860842808e-06 tokens_per_second_per_gpu:4308.306731633856 mfu:20.323017926043633 +Step 33679 | loss:0.3109537959098816 lr:1.291024693031956e-06 tokens_per_second_per_gpu:4676.935038476594 mfu:22.06190054389545 +Step 33680 | loss:0.30553334951400757 lr:1.2900458663260506e-06 tokens_per_second_per_gpu:4902.1124202626825 mfu:23.124100673003426 +Step 33681 | loss:0.32585370540618896 lr:1.2890674059739205e-06 tokens_per_second_per_gpu:4302.6437313416245 mfu:20.296304587457694 +Step 33682 | loss:0.30509036779403687 lr:1.2880893119829318e-06 tokens_per_second_per_gpu:3297.316738608992 mfu:15.55400563626566 +Step 33683 | loss:0.25671353936195374 lr:1.2871115843604509e-06 tokens_per_second_per_gpu:5358.222691631246 mfu:25.275650643485825 +Step 33684 | loss:0.4702214002609253 lr:1.2861342231138052e-06 tokens_per_second_per_gpu:4116.228761183827 mfu:19.41695336754976 +Step 33685 | loss:0.5115562677383423 lr:1.285157228250361e-06 tokens_per_second_per_gpu:3869.368333020703 mfu:18.252470609172114 +Step 33686 | loss:0.2608317732810974 lr:1.284180599777468e-06 tokens_per_second_per_gpu:5355.583754118208 mfu:25.263202324987656 +Step 33687 | loss:0.24229605495929718 lr:1.2832043377024705e-06 tokens_per_second_per_gpu:3638.806109702638 mfu:17.1648692638091 +Step 33688 | loss:0.31868088245391846 lr:1.282228442032707e-06 tokens_per_second_per_gpu:3742.971606929461 mfu:17.656235686694487 +Step 33689 | loss:0.296810120344162 lr:1.2812529127755213e-06 tokens_per_second_per_gpu:4823.05258489128 mfu:22.751161940557665 +Step 33690 | loss:0.30280160903930664 lr:1.2802777499382524e-06 tokens_per_second_per_gpu:3910.2321246869897 mfu:18.445232086543783 +Step 33691 | loss:0.38004618883132935 lr:1.2793029535282275e-06 tokens_per_second_per_gpu:3285.1751051295887 mfu:15.496731479567027 +Step 33692 | loss:0.304029256105423 lr:1.2783285235527798e-06 tokens_per_second_per_gpu:4543.774899897769 mfu:21.433761450756734 +Step 33693 | loss:0.32827863097190857 lr:1.2773544600192423e-06 tokens_per_second_per_gpu:4362.293056730777 mfu:20.57768063254268 +Step 33694 | loss:0.4200383722782135 lr:1.2763807629349423e-06 tokens_per_second_per_gpu:5341.704810130567 mfu:25.197732978205934 +Step 33695 | loss:0.2999032437801361 lr:1.2754074323071964e-06 tokens_per_second_per_gpu:4513.809779170211 mfu:21.292410863708056 +Step 33696 | loss:0.21591289341449738 lr:1.2744344681433261e-06 tokens_per_second_per_gpu:4517.69664220755 mfu:21.3107458598224 +Step 33697 | loss:0.32588422298431396 lr:1.273461870450654e-06 tokens_per_second_per_gpu:5366.112401744088 mfu:25.312867752211453 +Step 33698 | loss:0.2986685037612915 lr:1.272489639236496e-06 tokens_per_second_per_gpu:4523.012433585365 mfu:21.335821354719087 +Step 33699 | loss:0.2571331262588501 lr:1.2715177745081574e-06 tokens_per_second_per_gpu:5355.93270174039 mfu:25.264848370458093 +Step 33700 | loss:0.29472583532333374 lr:1.2705462762729493e-06 tokens_per_second_per_gpu:5397.6797278979875 mfu:25.46177621562062 +Step 33701 | loss:0.28429731726646423 lr:1.2695751445381875e-06 tokens_per_second_per_gpu:5307.734067673305 mfu:25.037487190028695 +Step 33702 | loss:0.270313024520874 lr:1.2686043793111613e-06 tokens_per_second_per_gpu:5374.470938939823 mfu:25.352296398277236 +Step 33703 | loss:0.3031320869922638 lr:1.2676339805991811e-06 tokens_per_second_per_gpu:5372.530017178037 mfu:25.343140739171496 +Step 33704 | loss:0.40437057614326477 lr:1.2666639484095523e-06 tokens_per_second_per_gpu:4542.313339118789 mfu:21.426867019194304 +Step 33705 | loss:0.27770689129829407 lr:1.2656942827495577e-06 tokens_per_second_per_gpu:4529.024008347998 mfu:21.364178978554843 +Step 33706 | loss:0.3505319058895111 lr:1.2647249836264919e-06 tokens_per_second_per_gpu:4434.1055359833035 mfu:20.916432349649245 +Step 33707 | loss:0.33971092104911804 lr:1.2637560510476488e-06 tokens_per_second_per_gpu:4433.824446468856 mfu:20.91510640245078 +Step 33708 | loss:0.33629700541496277 lr:1.262787485020317e-06 tokens_per_second_per_gpu:5332.44118878602 mfu:25.15403489578692 +Step 33709 | loss:0.24278013408184052 lr:1.2618192855517796e-06 tokens_per_second_per_gpu:5348.596910258788 mfu:25.230244190424287 +Step 33710 | loss:0.28730013966560364 lr:1.2608514526493142e-06 tokens_per_second_per_gpu:5358.119273417081 mfu:25.275162802124644 +Step 33711 | loss:0.2727887034416199 lr:1.2598839863202038e-06 tokens_per_second_per_gpu:4618.194235611129 mfu:21.784810154564628 +Step 33712 | loss:0.2791943848133087 lr:1.258916886571737e-06 tokens_per_second_per_gpu:5327.833822177006 mfu:25.1323011613944 +Step 33713 | loss:0.2996995449066162 lr:1.2579501534111693e-06 tokens_per_second_per_gpu:4485.636638063088 mfu:21.159513350272217 +Step 33714 | loss:0.25220781564712524 lr:1.2569837868457724e-06 tokens_per_second_per_gpu:4868.170455084012 mfu:22.96399063215922 +Step 33715 | loss:0.2872343957424164 lr:1.2560177868828293e-06 tokens_per_second_per_gpu:4871.6718013661775 mfu:22.98050707996347 +Step 33716 | loss:0.2873707413673401 lr:1.2550521535295845e-06 tokens_per_second_per_gpu:5159.894037729224 mfu:24.34010054467233 +Step 33717 | loss:0.31113201379776 lr:1.254086886793321e-06 tokens_per_second_per_gpu:4528.2839037852955 mfu:21.36068777905772 +Step 33718 | loss:0.34657785296440125 lr:1.2531219866812826e-06 tokens_per_second_per_gpu:3469.227715633123 mfu:16.364939045925684 +Step 33719 | loss:0.32479798793792725 lr:1.252157453200742e-06 tokens_per_second_per_gpu:5023.064443001669 mfu:23.694652000806276 +Step 33720 | loss:0.3634324371814728 lr:1.2511932863589482e-06 tokens_per_second_per_gpu:4930.570546636461 mfu:23.258342510565576 +Step 33721 | loss:0.2871895134449005 lr:1.2502294861631404e-06 tokens_per_second_per_gpu:4715.070871246743 mfu:22.24179377371626 +Step 33722 | loss:0.371203750371933 lr:1.249266052620579e-06 tokens_per_second_per_gpu:5320.03038897201 mfu:25.095491035563438 +Step 33723 | loss:0.27461546659469604 lr:1.2483029857385142e-06 tokens_per_second_per_gpu:5351.345169326508 mfu:25.243208197348068 +Step 33724 | loss:0.48082786798477173 lr:1.247340285524179e-06 tokens_per_second_per_gpu:3975.017713799555 mfu:18.750836763949064 +Step 33725 | loss:0.3917037844657898 lr:1.2463779519848173e-06 tokens_per_second_per_gpu:4122.216626375111 mfu:19.445199149292197 +Step 33726 | loss:0.3772551715373993 lr:1.245415985127668e-06 tokens_per_second_per_gpu:3983.649360031528 mfu:18.791553711935986 +Step 33727 | loss:0.3773181438446045 lr:1.2444543849599755e-06 tokens_per_second_per_gpu:5358.32376855578 mfu:25.27612744058435 +Step 33728 | loss:0.22234295308589935 lr:1.2434931514889503e-06 tokens_per_second_per_gpu:4890.190801147138 mfu:23.067864361597437 +Step 33729 | loss:0.31727802753448486 lr:1.2425322847218368e-06 tokens_per_second_per_gpu:4473.765292114675 mfu:21.103514185973072 +Step 33730 | loss:0.2488631010055542 lr:1.241571784665868e-06 tokens_per_second_per_gpu:5313.653084083766 mfu:25.065408200325177 +Step 33731 | loss:0.23575904965400696 lr:1.2406116513282495e-06 tokens_per_second_per_gpu:5093.651885337958 mfu:24.027625009766098 +Step 33732 | loss:0.33794504404067993 lr:1.2396518847162141e-06 tokens_per_second_per_gpu:5139.202612369587 mfu:24.24249556092972 +Step 33733 | loss:0.4118226170539856 lr:1.2386924848369841e-06 tokens_per_second_per_gpu:4351.447250904865 mfu:20.526519115976562 +Step 33734 | loss:0.2673606276512146 lr:1.2377334516977702e-06 tokens_per_second_per_gpu:3359.1659279698497 mfu:15.845758815040675 +Step 33735 | loss:0.3058246970176697 lr:1.2367747853057832e-06 tokens_per_second_per_gpu:5094.66277472848 mfu:24.032393547496707 +Step 33736 | loss:0.372112900018692 lr:1.2358164856682286e-06 tokens_per_second_per_gpu:4655.2862486702725 mfu:21.95977950871461 +Step 33737 | loss:0.2254069596529007 lr:1.2348585527923284e-06 tokens_per_second_per_gpu:5355.0555342154985 mfu:25.26071062158297 +Step 33738 | loss:0.3213452696800232 lr:1.233900986685277e-06 tokens_per_second_per_gpu:3910.0145742441878 mfu:18.444205863987154 +Step 33739 | loss:0.27259767055511475 lr:1.2329437873542736e-06 tokens_per_second_per_gpu:4864.752118969006 mfu:22.947865757476453 +Step 33740 | loss:0.25200727581977844 lr:1.2319869548065354e-06 tokens_per_second_per_gpu:5081.2555299699125 mfu:23.969149286458705 +Step 33741 | loss:0.22331520915031433 lr:1.2310304890492396e-06 tokens_per_second_per_gpu:4212.57582830182 mfu:19.87143892165007 +Step 33742 | loss:0.25370723009109497 lr:1.2300743900895862e-06 tokens_per_second_per_gpu:4855.551986194884 mfu:22.904467161476408 +Step 33743 | loss:0.33092862367630005 lr:1.2291186579347636e-06 tokens_per_second_per_gpu:4030.2033998203574 mfu:19.011157060558137 +Step 33744 | loss:0.34532439708709717 lr:1.2281632925919661e-06 tokens_per_second_per_gpu:4256.828155672167 mfu:20.0801847000816 +Step 33745 | loss:0.22979171574115753 lr:1.2272082940683715e-06 tokens_per_second_per_gpu:5053.703445685027 mfu:23.839181404016582 +Step 33746 | loss:0.3122524321079254 lr:1.2262536623711684e-06 tokens_per_second_per_gpu:4319.45372623189 mfu:20.375600201435955 +Step 33747 | loss:0.35763776302337646 lr:1.2252993975075345e-06 tokens_per_second_per_gpu:5311.510284645711 mfu:25.05530024977659 +Step 33748 | loss:0.34606391191482544 lr:1.2243454994846581e-06 tokens_per_second_per_gpu:4326.548831557298 mfu:20.409069023805703 +Step 33749 | loss:0.40448132157325745 lr:1.223391968309695e-06 tokens_per_second_per_gpu:5337.783491407943 mfu:25.179235448745352 +Step 33750 | loss:0.3032659590244293 lr:1.222438803989817e-06 tokens_per_second_per_gpu:5317.967900012688 mfu:25.08576192324541 +Step 33751 | loss:0.2836228907108307 lr:1.2214860065322132e-06 tokens_per_second_per_gpu:3589.462247280675 mfu:16.93210584583364 +Step 33752 | loss:0.3536040782928467 lr:1.2205335759440328e-06 tokens_per_second_per_gpu:4336.813241300615 mfu:20.457487995853818 +Step 33753 | loss:0.3602825701236725 lr:1.2195815122324373e-06 tokens_per_second_per_gpu:4230.687485757581 mfu:19.956874700036302 +Step 33754 | loss:0.342104971408844 lr:1.2186298154045983e-06 tokens_per_second_per_gpu:4094.1997445726483 mfu:19.313038737657028 +Step 33755 | loss:0.31395870447158813 lr:1.217678485467677e-06 tokens_per_second_per_gpu:4392.689726787293 mfu:20.721066911405963 +Step 33756 | loss:0.25576943159103394 lr:1.2167275224288122e-06 tokens_per_second_per_gpu:4770.787183346953 mfu:22.504617124076038 +Step 33757 | loss:0.3117363452911377 lr:1.2157769262951645e-06 tokens_per_second_per_gpu:5366.099683718011 mfu:25.31280775911253 +Step 33758 | loss:0.40736496448516846 lr:1.214826697073884e-06 tokens_per_second_per_gpu:4985.203163289826 mfu:23.516053884605085 +Step 33759 | loss:0.4431544840335846 lr:1.2138768347721207e-06 tokens_per_second_per_gpu:4065.396570125363 mfu:19.17716924946087 +Step 33760 | loss:0.35192885994911194 lr:1.212927339397013e-06 tokens_per_second_per_gpu:4513.451134395967 mfu:21.290719075116833 +Step 33761 | loss:0.2502160370349884 lr:1.2119782109556998e-06 tokens_per_second_per_gpu:3030.6697606623634 mfu:14.296186346625714 +Step 33762 | loss:0.22855375707149506 lr:1.211029449455331e-06 tokens_per_second_per_gpu:5323.381407827705 mfu:25.111298363248522 +Step 33763 | loss:0.32163459062576294 lr:1.2100810549030338e-06 tokens_per_second_per_gpu:4270.592169001267 mfu:20.145111899337795 +Step 33764 | loss:0.21771404147148132 lr:1.2091330273059365e-06 tokens_per_second_per_gpu:4010.084774304959 mfu:18.91625407141053 +Step 33765 | loss:0.3107280135154724 lr:1.2081853666711773e-06 tokens_per_second_per_gpu:3569.326131909039 mfu:16.837120354050082 +Step 33766 | loss:0.20012980699539185 lr:1.2072380730058896e-06 tokens_per_second_per_gpu:4096.722777484155 mfu:19.324940314374405 +Step 33767 | loss:0.24079832434654236 lr:1.206291146317179e-06 tokens_per_second_per_gpu:4775.809661812994 mfu:22.528309011923543 +Step 33768 | loss:0.24039994180202484 lr:1.2053445866121838e-06 tokens_per_second_per_gpu:5343.465850573274 mfu:25.20604010643866 +Step 33769 | loss:0.32926124334335327 lr:1.2043983938980208e-06 tokens_per_second_per_gpu:5348.326344389677 mfu:25.22896788505639 +Step 33770 | loss:0.2818189561367035 lr:1.2034525681818066e-06 tokens_per_second_per_gpu:5361.412010498416 mfu:25.290695204736174 +Step 33771 | loss:0.40862804651260376 lr:1.2025071094706464e-06 tokens_per_second_per_gpu:4497.370666999907 mfu:21.21486476679897 +Step 33772 | loss:0.3153110146522522 lr:1.2015620177716513e-06 tokens_per_second_per_gpu:5332.54789351263 mfu:25.154538240188046 +Step 33773 | loss:0.2240505963563919 lr:1.200617293091949e-06 tokens_per_second_per_gpu:5345.171295742818 mfu:25.21408498228235 +Step 33774 | loss:0.363129585981369 lr:1.199672935438617e-06 tokens_per_second_per_gpu:4693.852485828711 mfu:22.14170302946954 +Step 33775 | loss:0.32450491189956665 lr:1.1987289448187776e-06 tokens_per_second_per_gpu:4476.218971648884 mfu:21.115088610976837 +Step 33776 | loss:0.22150066494941711 lr:1.1977853212395308e-06 tokens_per_second_per_gpu:5364.974282216536 mfu:25.307499048216776 +Step 33777 | loss:0.4046210050582886 lr:1.1968420647079648e-06 tokens_per_second_per_gpu:4335.077233333375 mfu:20.449298950078845 +Step 33778 | loss:0.2946544885635376 lr:1.1958991752311688e-06 tokens_per_second_per_gpu:3550.4471878255017 mfu:16.74806515372803 +Step 33779 | loss:0.3865436315536499 lr:1.194956652816248e-06 tokens_per_second_per_gpu:3855.828487716634 mfu:18.188600848736996 +Step 33780 | loss:0.27271905541419983 lr:1.1940144974702916e-06 tokens_per_second_per_gpu:4250.902526117941 mfu:20.052232494457662 +Step 33781 | loss:0.3527795672416687 lr:1.1930727092003712e-06 tokens_per_second_per_gpu:4877.996390915414 mfu:23.01034124795347 +Step 33782 | loss:0.27489036321640015 lr:1.1921312880135815e-06 tokens_per_second_per_gpu:4369.722722102417 mfu:20.61272763173233 +Step 33783 | loss:0.3801804184913635 lr:1.191190233917e-06 tokens_per_second_per_gpu:5319.125111624932 mfu:25.091220687861032 +Step 33784 | loss:0.39233651757240295 lr:1.1902495469177045e-06 tokens_per_second_per_gpu:4145.011797574424 mfu:19.55272786109661 +Step 33785 | loss:0.3548416793346405 lr:1.1893092270227724e-06 tokens_per_second_per_gpu:3885.7108520071038 mfu:18.329561059552226 +Step 33786 | loss:0.33591151237487793 lr:1.1883692742392706e-06 tokens_per_second_per_gpu:4743.283325102685 mfu:22.37487673209572 +Step 33787 | loss:0.2628982961177826 lr:1.1874296885742764e-06 tokens_per_second_per_gpu:3562.921341996734 mfu:16.8069078672639 +Step 33788 | loss:0.29004350304603577 lr:1.1864904700348511e-06 tokens_per_second_per_gpu:4833.719925289696 mfu:22.801481605252942 +Step 33789 | loss:0.28358903527259827 lr:1.1855516186280558e-06 tokens_per_second_per_gpu:3290.1414632599294 mfu:15.520158638215918 +Step 33790 | loss:0.4645904302597046 lr:1.1846131343609568e-06 tokens_per_second_per_gpu:5366.669651564136 mfu:25.315496394689244 +Step 33791 | loss:0.28251945972442627 lr:1.1836750172406153e-06 tokens_per_second_per_gpu:4794.850280295513 mfu:22.618126857970086 +Step 33792 | loss:0.28318917751312256 lr:1.1827372672740811e-06 tokens_per_second_per_gpu:5393.435518368006 mfu:25.441755555130882 +Step 33793 | loss:0.30298054218292236 lr:1.1817998844684097e-06 tokens_per_second_per_gpu:4149.670537841678 mfu:19.5747039338003 +Step 33794 | loss:0.3137582242488861 lr:1.1808628688306456e-06 tokens_per_second_per_gpu:4858.802970438502 mfu:22.919802608828657 +Step 33795 | loss:0.3076578676700592 lr:1.1799262203678441e-06 tokens_per_second_per_gpu:4971.81324602076 mfu:23.452891360291435 +Step 33796 | loss:0.28406065702438354 lr:1.1789899390870497e-06 tokens_per_second_per_gpu:3997.621307254405 mfu:18.857461770846314 +Step 33797 | loss:0.30642983317375183 lr:1.1780540249952953e-06 tokens_per_second_per_gpu:5362.463368841683 mfu:25.2956546414216 +Step 33798 | loss:0.36381417512893677 lr:1.1771184780996314e-06 tokens_per_second_per_gpu:4169.783028445601 mfu:19.66957798352375 +Step 33799 | loss:0.33909299969673157 lr:1.1761832984070909e-06 tokens_per_second_per_gpu:5344.862102994012 mfu:25.212626467332512 +Step 33800 | loss:0.3110392689704895 lr:1.1752484859247014e-06 tokens_per_second_per_gpu:4635.8176734130775 mfu:21.867942917544962 +Step 33801 | loss:0.31463706493377686 lr:1.1743140406595022e-06 tokens_per_second_per_gpu:5295.5713050982495 mfu:24.980113363780877 +Step 33802 | loss:0.3126467764377594 lr:1.173379962618515e-06 tokens_per_second_per_gpu:4156.018209696781 mfu:19.604646984965314 +Step 33803 | loss:0.26447492837905884 lr:1.1724462518087675e-06 tokens_per_second_per_gpu:3529.0094306615597 mfu:16.646939595527176 +Step 33804 | loss:0.2701808512210846 lr:1.1715129082372767e-06 tokens_per_second_per_gpu:4410.2141838594525 mfu:20.80373276539603 +Step 33805 | loss:0.28911083936691284 lr:1.1705799319110754e-06 tokens_per_second_per_gpu:4674.098162169139 mfu:22.048518514332173 +Step 33806 | loss:0.33616432547569275 lr:1.169647322837175e-06 tokens_per_second_per_gpu:5342.465547636593 mfu:25.20132150681706 +Step 33807 | loss:0.3472728133201599 lr:1.1687150810225812e-06 tokens_per_second_per_gpu:5338.264418708937 mfu:25.181504064878798 +Step 33808 | loss:0.3095993399620056 lr:1.1677832064743155e-06 tokens_per_second_per_gpu:4612.8364970789535 mfu:21.759536787783897 +Step 33809 | loss:0.30299919843673706 lr:1.166851699199384e-06 tokens_per_second_per_gpu:5306.213074199304 mfu:25.030312404303587 +Step 33810 | loss:0.3337444067001343 lr:1.1659205592047917e-06 tokens_per_second_per_gpu:4255.505899089638 mfu:20.073947390181573 +Step 33811 | loss:0.35544323921203613 lr:1.164989786497539e-06 tokens_per_second_per_gpu:5318.933402376664 mfu:25.09031636262798 +Step 33812 | loss:0.3329692780971527 lr:1.1640593810846312e-06 tokens_per_second_per_gpu:3561.584014369261 mfu:16.800599464673486 +Step 33813 | loss:0.2225911021232605 lr:1.1631293429730628e-06 tokens_per_second_per_gpu:4774.790244428743 mfu:22.52350024619149 +Step 33814 | loss:0.2878076434135437 lr:1.1621996721698392e-06 tokens_per_second_per_gpu:5354.646360705286 mfu:25.258780480323093 +Step 33815 | loss:0.36976945400238037 lr:1.1612703686819325e-06 tokens_per_second_per_gpu:4929.879214595866 mfu:23.255081379376684 +Step 33816 | loss:0.38996249437332153 lr:1.1603414325163486e-06 tokens_per_second_per_gpu:5365.571164736818 mfu:25.310314644903602 +Step 33817 | loss:0.33400052785873413 lr:1.159412863680065e-06 tokens_per_second_per_gpu:5349.543786767817 mfu:25.23471076847117 +Step 33818 | loss:0.2354964166879654 lr:1.1584846621800704e-06 tokens_per_second_per_gpu:4621.672388174774 mfu:21.80121719364136 +Step 33819 | loss:0.34267038106918335 lr:1.1575568280233485e-06 tokens_per_second_per_gpu:3900.1504643498206 mfu:18.397675174624734 +Step 33820 | loss:0.3040226101875305 lr:1.156629361216871e-06 tokens_per_second_per_gpu:5332.814577827591 mfu:25.15579623560275 +Step 33821 | loss:0.3393450379371643 lr:1.1557022617676215e-06 tokens_per_second_per_gpu:5336.17591777196 mfu:25.171652249624067 +Step 33822 | loss:0.2910985052585602 lr:1.1547755296825614e-06 tokens_per_second_per_gpu:4432.091393034625 mfu:20.906931293712105 +Step 33823 | loss:0.2920527756214142 lr:1.1538491649686733e-06 tokens_per_second_per_gpu:5278.932666751576 mfu:24.901626067855187 +Step 33824 | loss:0.23889294266700745 lr:1.1529231676329133e-06 tokens_per_second_per_gpu:5331.041950661434 mfu:25.147434450818608 +Step 33825 | loss:0.27382802963256836 lr:1.1519975376822478e-06 tokens_per_second_per_gpu:5340.863204248263 mfu:25.193762979666335 +Step 33826 | loss:0.39718544483184814 lr:1.1510722751236435e-06 tokens_per_second_per_gpu:4874.088213469285 mfu:22.991905708136247 +Step 33827 | loss:0.2587863802909851 lr:1.150147379964067e-06 tokens_per_second_per_gpu:5319.334827835527 mfu:25.092209955007682 +Step 33828 | loss:0.28054553270339966 lr:1.1492228522104577e-06 tokens_per_second_per_gpu:3891.420434391691 mfu:18.35649413381553 +Step 33829 | loss:0.24056346714496613 lr:1.1482986918697814e-06 tokens_per_second_per_gpu:4852.4756732677015 mfu:22.88995566852578 +Step 33830 | loss:0.31453368067741394 lr:1.1473748989489774e-06 tokens_per_second_per_gpu:4106.735559881518 mfu:19.372172317299608 +Step 33831 | loss:0.38444432616233826 lr:1.1464514734550068e-06 tokens_per_second_per_gpu:5334.225615770215 mfu:25.162452342326006 +Step 33832 | loss:0.27705118060112 lr:1.1455284153948033e-06 tokens_per_second_per_gpu:5307.700870960033 mfu:25.03733059546825 +Step 33833 | loss:0.28808483481407166 lr:1.1446057247753162e-06 tokens_per_second_per_gpu:3949.921397818863 mfu:18.6324531595951 +Step 33834 | loss:0.21139483153820038 lr:1.1436834016034847e-06 tokens_per_second_per_gpu:5346.208035253909 mfu:25.218975459442166 +Step 33835 | loss:0.28971266746520996 lr:1.142761445886248e-06 tokens_per_second_per_gpu:4558.377459607597 mfu:21.50264421636159 +Step 33836 | loss:0.3179122507572174 lr:1.1418398576305278e-06 tokens_per_second_per_gpu:4838.047733409758 mfu:22.821896614555477 +Step 33837 | loss:0.33983901143074036 lr:1.140918636843269e-06 tokens_per_second_per_gpu:4740.856885198441 mfu:22.363430801074063 +Step 33838 | loss:0.35471516847610474 lr:1.139997783531399e-06 tokens_per_second_per_gpu:4668.318212883019 mfu:22.021253507388508 +Step 33839 | loss:0.33333197236061096 lr:1.139077297701835e-06 tokens_per_second_per_gpu:3453.773651711413 mfu:16.292039589671784 +Step 33840 | loss:0.3999413251876831 lr:1.1381571793615043e-06 tokens_per_second_per_gpu:4907.231142846184 mfu:23.148246560773934 +Step 33841 | loss:0.3283345699310303 lr:1.1372374285173348e-06 tokens_per_second_per_gpu:5342.866968081497 mfu:25.203215075545064 +Step 33842 | loss:0.2817646265029907 lr:1.1363180451762324e-06 tokens_per_second_per_gpu:4405.017176237233 mfu:20.77921759374109 +Step 33843 | loss:0.30512648820877075 lr:1.1353990293451189e-06 tokens_per_second_per_gpu:4564.567530608129 mfu:21.531843837406118 +Step 33844 | loss:0.23573161661624908 lr:1.1344803810309002e-06 tokens_per_second_per_gpu:5369.154359353658 mfu:25.327217185267283 +Step 33845 | loss:0.291461318731308 lr:1.1335621002404983e-06 tokens_per_second_per_gpu:3925.595606878968 mfu:18.517704253323704 +Step 33846 | loss:0.2818087041378021 lr:1.1326441869808025e-06 tokens_per_second_per_gpu:5358.589443133321 mfu:25.27738067289537 +Step 33847 | loss:0.29390567541122437 lr:1.131726641258729e-06 tokens_per_second_per_gpu:5355.730746747809 mfu:25.26389571430101 +Step 33848 | loss:0.42948251962661743 lr:1.1308094630811783e-06 tokens_per_second_per_gpu:4803.225741353139 mfu:22.65763533677992 +Step 33849 | loss:0.2221696972846985 lr:1.129892652455039e-06 tokens_per_second_per_gpu:4047.083337285021 mfu:19.09078260559326 +Step 33850 | loss:0.31319642066955566 lr:1.1289762093872225e-06 tokens_per_second_per_gpu:4365.725511031999 mfu:20.593872105118642 +Step 33851 | loss:0.35323143005371094 lr:1.128060133884601e-06 tokens_per_second_per_gpu:5343.326047553087 mfu:25.205380631739583 +Step 33852 | loss:0.3874414265155792 lr:1.1271444259540853e-06 tokens_per_second_per_gpu:3992.9703737957934 mfu:18.835522524190964 +Step 33853 | loss:0.2661953866481781 lr:1.1262290856025482e-06 tokens_per_second_per_gpu:4708.3518454281375 mfu:22.210098982545876 +Step 33854 | loss:0.3042926490306854 lr:1.1253141128368728e-06 tokens_per_second_per_gpu:3604.9594918023417 mfu:17.005208992344905 +Step 33855 | loss:0.3275603652000427 lr:1.1243995076639535e-06 tokens_per_second_per_gpu:4477.245086936967 mfu:21.119928971864045 +Step 33856 | loss:0.2908051013946533 lr:1.1234852700906574e-06 tokens_per_second_per_gpu:5327.234634473093 mfu:25.12947469076308 +Step 33857 | loss:0.2789798378944397 lr:1.122571400123873e-06 tokens_per_second_per_gpu:4124.655805629958 mfu:19.456705174004203 +Step 33858 | loss:0.3240254819393158 lr:1.1216578977704561e-06 tokens_per_second_per_gpu:3798.902029749605 mfu:17.920069033851252 +Step 33859 | loss:0.4138615131378174 lr:1.120744763037296e-06 tokens_per_second_per_gpu:4770.247430537077 mfu:22.502071017980526 +Step 33860 | loss:0.33411964774131775 lr:1.1198319959312475e-06 tokens_per_second_per_gpu:5312.365974844319 mfu:25.059336686439224 +Step 33861 | loss:0.34026050567626953 lr:1.1189195964591782e-06 tokens_per_second_per_gpu:4518.168281498997 mfu:21.312970662830118 +Step 33862 | loss:0.3256857097148895 lr:1.1180075646279486e-06 tokens_per_second_per_gpu:5158.9038950608365 mfu:24.335429872769787 +Step 33863 | loss:0.28481850028038025 lr:1.1170959004444314e-06 tokens_per_second_per_gpu:5346.439925061016 mfu:25.22006932322677 +Step 33864 | loss:0.22178424894809723 lr:1.116184603915471e-06 tokens_per_second_per_gpu:4679.474189835975 mfu:22.073878154081545 +Step 33865 | loss:0.30484771728515625 lr:1.1152736750479176e-06 tokens_per_second_per_gpu:5176.677154372208 mfu:24.41926936937226 +Step 33866 | loss:0.3432316482067108 lr:1.1143631138486321e-06 tokens_per_second_per_gpu:4352.654248897087 mfu:20.53221273144097 +Step 33867 | loss:0.3707071840763092 lr:1.1134529203244593e-06 tokens_per_second_per_gpu:4181.723317974775 mfu:19.725902366456328 +Step 33868 | loss:0.2826026678085327 lr:1.1125430944822436e-06 tokens_per_second_per_gpu:4577.716153549889 mfu:21.593868135208805 +Step 33869 | loss:0.2711572051048279 lr:1.1116336363288293e-06 tokens_per_second_per_gpu:5362.896519534191 mfu:25.297697887141286 +Step 33870 | loss:0.3250569701194763 lr:1.1107245458710614e-06 tokens_per_second_per_gpu:5332.260577916673 mfu:25.15318292350082 +Step 33871 | loss:0.3107324540615082 lr:1.109815823115773e-06 tokens_per_second_per_gpu:3811.0495953006416 mfu:17.977371173143894 +Step 33872 | loss:0.2159416675567627 lr:1.108907468069792e-06 tokens_per_second_per_gpu:4841.816353209132 mfu:22.83967383714197 +Step 33873 | loss:0.26174473762512207 lr:1.1079994807399574e-06 tokens_per_second_per_gpu:4901.01805953892 mfu:23.118938386752248 +Step 33874 | loss:0.35379984974861145 lr:1.1070918611330972e-06 tokens_per_second_per_gpu:5165.250241013672 mfu:24.365366669427587 +Step 33875 | loss:0.36133256554603577 lr:1.106184609256039e-06 tokens_per_second_per_gpu:5306.854282137321 mfu:25.03333709154857 +Step 33876 | loss:0.331010103225708 lr:1.1052777251156054e-06 tokens_per_second_per_gpu:5350.282810007183 mfu:25.23819686718162 +Step 33877 | loss:0.269239217042923 lr:1.1043712087186186e-06 tokens_per_second_per_gpu:5318.27289403193 mfu:25.08720063207152 +Step 33878 | loss:0.3413771092891693 lr:1.1034650600718955e-06 tokens_per_second_per_gpu:5322.545968020374 mfu:25.107357451136615 +Step 33879 | loss:0.3427714705467224 lr:1.1025592791822525e-06 tokens_per_second_per_gpu:5308.033965070648 mfu:25.038901857219738 +Step 33880 | loss:0.2832089960575104 lr:1.10165386605649e-06 tokens_per_second_per_gpu:4822.129596642982 mfu:22.74680804751241 +Step 33881 | loss:0.272482693195343 lr:1.1007488207014415e-06 tokens_per_second_per_gpu:4218.1323026562595 mfu:19.897649759212243 +Step 33882 | loss:0.22093574702739716 lr:1.0998441431238904e-06 tokens_per_second_per_gpu:2631.087270725997 mfu:12.411287565793023 +Step 33883 | loss:0.2489885538816452 lr:1.0989398333306532e-06 tokens_per_second_per_gpu:5332.825982717691 mfu:25.155850034415227 +Step 33884 | loss:0.37884819507598877 lr:1.098035891328536e-06 tokens_per_second_per_gpu:4905.084984979199 mfu:23.138122767943074 +Step 33885 | loss:0.24832426011562347 lr:1.0971323171243276e-06 tokens_per_second_per_gpu:3890.2087335415854 mfu:18.3507783341681 +Step 33886 | loss:0.2539878785610199 lr:1.096229110724828e-06 tokens_per_second_per_gpu:5338.16031812759 mfu:25.18101300467519 +Step 33887 | loss:0.23769426345825195 lr:1.0953262721368262e-06 tokens_per_second_per_gpu:4701.768261895819 mfu:22.179043095748753 +Step 33888 | loss:0.408896267414093 lr:1.094423801367117e-06 tokens_per_second_per_gpu:4690.0581771925845 mfu:22.123804628257126 +Step 33889 | loss:0.22512860596179962 lr:1.0935216984224839e-06 tokens_per_second_per_gpu:4512.694060180897 mfu:21.287147826874172 +Step 33890 | loss:0.3083357810974121 lr:1.0926199633097157e-06 tokens_per_second_per_gpu:4848.891655663208 mfu:22.873049245988927 +Step 33891 | loss:0.2961626648902893 lr:1.0917185960356014e-06 tokens_per_second_per_gpu:5324.987651390092 mfu:25.118875288185716 +Step 33892 | loss:0.32382988929748535 lr:1.0908175966069024e-06 tokens_per_second_per_gpu:5104.804874968484 mfu:24.08023556474954 +Step 33893 | loss:0.2938345968723297 lr:1.089916965030413e-06 tokens_per_second_per_gpu:3680.5567477207474 mfu:17.361814146733906 +Step 33894 | loss:0.3406526446342468 lr:1.0890167013128893e-06 tokens_per_second_per_gpu:3266.9160358487916 mfu:15.410600334451047 +Step 33895 | loss:0.28988170623779297 lr:1.0881168054611257e-06 tokens_per_second_per_gpu:4656.560697251591 mfu:21.96579130011604 +Step 33896 | loss:0.22280655801296234 lr:1.0872172774818667e-06 tokens_per_second_per_gpu:5345.564548401837 mfu:25.215940022170766 +Step 33897 | loss:0.40715327858924866 lr:1.0863181173818849e-06 tokens_per_second_per_gpu:5351.961727976387 mfu:25.246116609695303 +Step 33898 | loss:0.23976746201515198 lr:1.0854193251679524e-06 tokens_per_second_per_gpu:5355.605953718997 mfu:25.26330704429222 +Step 33899 | loss:0.3062204122543335 lr:1.0845209008468192e-06 tokens_per_second_per_gpu:5323.436718580627 mfu:25.111559273507044 +Step 33900 | loss:0.3586095869541168 lr:1.0836228444252528e-06 tokens_per_second_per_gpu:4965.22597662079 mfu:23.421818086627724 +Step 33901 | loss:0.32889869809150696 lr:1.0827251559099915e-06 tokens_per_second_per_gpu:5160.073095857941 mfu:24.340945192415955 +Step 33902 | loss:0.23889398574829102 lr:1.0818278353078027e-06 tokens_per_second_per_gpu:5253.47792272067 mfu:24.78155169723423 +Step 33903 | loss:0.41448599100112915 lr:1.0809308826254306e-06 tokens_per_second_per_gpu:5348.751987430434 mfu:25.230975715901835 +Step 33904 | loss:0.37567049264907837 lr:1.0800342978696087e-06 tokens_per_second_per_gpu:5216.673960501434 mfu:24.60794112804263 +Step 33905 | loss:0.320631742477417 lr:1.0791380810470986e-06 tokens_per_second_per_gpu:3833.1407542619854 mfu:18.081578939104933 +Step 33906 | loss:0.3582807183265686 lr:1.0782422321646336e-06 tokens_per_second_per_gpu:5324.085326779844 mfu:25.114618869046723 +Step 33907 | loss:0.35052692890167236 lr:1.0773467512289525e-06 tokens_per_second_per_gpu:5367.9694417254195 mfu:25.321627726647087 +Step 33908 | loss:0.25188615918159485 lr:1.0764516382467893e-06 tokens_per_second_per_gpu:3935.8614876475513 mfu:18.566130164448982 +Step 33909 | loss:0.3014925420284271 lr:1.0755568932248717e-06 tokens_per_second_per_gpu:3770.6777145211254 mfu:17.786930123353773 +Step 33910 | loss:0.24017374217510223 lr:1.0746625161699386e-06 tokens_per_second_per_gpu:4295.903568250547 mfu:20.264510088166677 +Step 33911 | loss:0.3584694564342499 lr:1.0737685070887015e-06 tokens_per_second_per_gpu:5332.691855351804 mfu:25.155217332745853 +Step 33912 | loss:0.2916386127471924 lr:1.0728748659878995e-06 tokens_per_second_per_gpu:3567.209557882494 mfu:16.827136113242105 +Step 33913 | loss:0.322273313999176 lr:1.0719815928742549e-06 tokens_per_second_per_gpu:5346.919171193207 mfu:25.222330009000142 +Step 33914 | loss:0.32305359840393066 lr:1.0710886877544734e-06 tokens_per_second_per_gpu:4651.362168047313 mfu:21.94126895089892 +Step 33915 | loss:0.28622323274612427 lr:1.0701961506352831e-06 tokens_per_second_per_gpu:5348.161541461511 mfu:25.228190481525953 +Step 33916 | loss:0.3827681243419647 lr:1.069303981523384e-06 tokens_per_second_per_gpu:3440.577486972514 mfu:16.229791029100525 +Step 33917 | loss:0.2851168215274811 lr:1.0684121804255042e-06 tokens_per_second_per_gpu:5347.898138666714 mfu:25.22694796560175 +Step 33918 | loss:0.3340221345424652 lr:1.0675207473483274e-06 tokens_per_second_per_gpu:3721.30138641625 mfu:17.554013559212798 +Step 33919 | loss:0.2391425520181656 lr:1.0666296822985755e-06 tokens_per_second_per_gpu:4741.376835791834 mfu:22.36588349673582 +Step 33920 | loss:0.2859288454055786 lr:1.0657389852829491e-06 tokens_per_second_per_gpu:5117.36925205947 mfu:24.139503875191508 +Step 33921 | loss:0.3308614194393158 lr:1.0648486563081373e-06 tokens_per_second_per_gpu:4543.433658005072 mfu:21.432151754527013 +Step 33922 | loss:0.3259485065937042 lr:1.063958695380851e-06 tokens_per_second_per_gpu:4550.998309416094 mfu:21.467835505895422 +Step 33923 | loss:0.3465208411216736 lr:1.063069102507769e-06 tokens_per_second_per_gpu:5022.47339324786 mfu:23.691863918274073 +Step 33924 | loss:0.3506910502910614 lr:1.062179877695596e-06 tokens_per_second_per_gpu:3822.0307690020827 mfu:18.029171243075027 +Step 33925 | loss:0.3650592565536499 lr:1.0612910209510051e-06 tokens_per_second_per_gpu:4866.227256295676 mfu:22.95482423193074 +Step 33926 | loss:0.28106069564819336 lr:1.060402532280691e-06 tokens_per_second_per_gpu:4483.238837345103 mfu:21.148202515178497 +Step 33927 | loss:0.36992016434669495 lr:1.059514411691337e-06 tokens_per_second_per_gpu:4052.02117158898 mfu:19.114075212486444 +Step 33928 | loss:0.2946937382221222 lr:1.0586266591896155e-06 tokens_per_second_per_gpu:5373.717029704938 mfu:25.348740079786104 +Step 33929 | loss:0.28760865330696106 lr:1.0577392747822102e-06 tokens_per_second_per_gpu:5336.903711385023 mfu:25.175085376271284 +Step 33930 | loss:0.33604443073272705 lr:1.0568522584757935e-06 tokens_per_second_per_gpu:4395.168232402038 mfu:20.732758445267375 +Step 33931 | loss:0.2888636887073517 lr:1.0559656102770377e-06 tokens_per_second_per_gpu:4726.306307597493 mfu:22.294793243945858 +Step 33932 | loss:0.3681834638118744 lr:1.0550793301926043e-06 tokens_per_second_per_gpu:4079.580976447802 mfu:19.244079514193093 +Step 33933 | loss:0.3754178285598755 lr:1.0541934182291657e-06 tokens_per_second_per_gpu:4617.52168445796 mfu:21.781637616026067 +Step 33934 | loss:0.38340941071510315 lr:1.053307874393389e-06 tokens_per_second_per_gpu:4788.852460626229 mfu:22.58983411925652 +Step 33935 | loss:0.3567560613155365 lr:1.0524226986919294e-06 tokens_per_second_per_gpu:5189.403706598935 mfu:24.479302687599446 +Step 33936 | loss:0.36362797021865845 lr:1.0515378911314378e-06 tokens_per_second_per_gpu:5316.309107802842 mfu:25.077937117372432 +Step 33937 | loss:0.256687194108963 lr:1.0506534517185806e-06 tokens_per_second_per_gpu:3581.959246946285 mfu:16.89671291311246 +Step 33938 | loss:0.22559307515621185 lr:1.0497693804600083e-06 tokens_per_second_per_gpu:5385.591296946134 mfu:25.40475302432173 +Step 33939 | loss:0.281379759311676 lr:1.0488856773623601e-06 tokens_per_second_per_gpu:3633.250567646399 mfu:17.138662824056603 +Step 33940 | loss:0.26395756006240845 lr:1.048002342432286e-06 tokens_per_second_per_gpu:4509.107299198363 mfu:21.27022846335503 +Step 33941 | loss:0.2715868651866913 lr:1.0471193756764365e-06 tokens_per_second_per_gpu:5354.081792398634 mfu:25.25611731529533 +Step 33942 | loss:0.28738337755203247 lr:1.0462367771014558e-06 tokens_per_second_per_gpu:5009.051090809677 mfu:23.628548627592224 +Step 33943 | loss:0.3568108379840851 lr:1.0453545467139669e-06 tokens_per_second_per_gpu:4073.316684064128 mfu:19.214529778220808 +Step 33944 | loss:0.4034034013748169 lr:1.0444726845206143e-06 tokens_per_second_per_gpu:4130.276762908203 mfu:19.483220188519887 +Step 33945 | loss:0.2840427756309509 lr:1.043591190528026e-06 tokens_per_second_per_gpu:5105.661727949573 mfu:24.08427748645553 +Step 33946 | loss:0.32554328441619873 lr:1.0427100647428467e-06 tokens_per_second_per_gpu:4524.94524238497 mfu:21.344938743598007 +Step 33947 | loss:0.36890628933906555 lr:1.0418293071716822e-06 tokens_per_second_per_gpu:5339.568514455245 mfu:25.18765570701581 +Step 33948 | loss:0.36370280385017395 lr:1.0409489178211662e-06 tokens_per_second_per_gpu:3633.375461069293 mfu:17.13925196764141 +Step 33949 | loss:0.346035897731781 lr:1.0400688966979266e-06 tokens_per_second_per_gpu:3620.8840134750067 mfu:17.080327678078433 +Step 33950 | loss:0.263446182012558 lr:1.0391892438085694e-06 tokens_per_second_per_gpu:4423.977503322004 mfu:20.868656691565334 +Step 33951 | loss:0.32127875089645386 lr:1.0383099591597224e-06 tokens_per_second_per_gpu:5334.152842543612 mfu:25.162109058599725 +Step 33952 | loss:0.3118823766708374 lr:1.0374310427579913e-06 tokens_per_second_per_gpu:3766.9178568591037 mfu:17.769194233264148 +Step 33953 | loss:0.32904377579689026 lr:1.0365524946099935e-06 tokens_per_second_per_gpu:4596.269828869937 mfu:21.6813889872769 +Step 33954 | loss:0.3025706708431244 lr:1.0356743147223235e-06 tokens_per_second_per_gpu:4080.8888193810103 mfu:19.250248832450048 +Step 33955 | loss:0.3112921714782715 lr:1.0347965031015982e-06 tokens_per_second_per_gpu:4371.926583702569 mfu:20.623123622917813 +Step 33956 | loss:0.3422726094722748 lr:1.0339190597544234e-06 tokens_per_second_per_gpu:4639.371921881799 mfu:21.884708913989243 +Step 33957 | loss:0.35590216517448425 lr:1.0330419846873828e-06 tokens_per_second_per_gpu:4591.50849863004 mfu:21.658928980168604 +Step 33958 | loss:0.30257198214530945 lr:1.0321652779070878e-06 tokens_per_second_per_gpu:5340.886665086453 mfu:25.193873648442903 +Step 33959 | loss:0.26782000064849854 lr:1.031288939420122e-06 tokens_per_second_per_gpu:5253.675517571091 mfu:24.78248378585683 +Step 33960 | loss:0.39642244577407837 lr:1.03041296923308e-06 tokens_per_second_per_gpu:4310.33173569242 mfu:20.332570215689202 +Step 33961 | loss:0.2666931748390198 lr:1.029537367352551e-06 tokens_per_second_per_gpu:2985.344950765633 mfu:14.082381485132844 +Step 33962 | loss:0.34935107827186584 lr:1.0286621337851188e-06 tokens_per_second_per_gpu:5233.535210732131 mfu:24.687478522206767 +Step 33963 | loss:0.2809562087059021 lr:1.0277872685373724e-06 tokens_per_second_per_gpu:5331.202215386623 mfu:25.148190446871304 +Step 33964 | loss:0.25474604964256287 lr:1.0269127716158843e-06 tokens_per_second_per_gpu:5322.666051804298 mfu:25.107923906833793 +Step 33965 | loss:0.3728976547718048 lr:1.0260386430272384e-06 tokens_per_second_per_gpu:5333.367742897669 mfu:25.158405609618 +Step 33966 | loss:0.2610311210155487 lr:1.0251648827780013e-06 tokens_per_second_per_gpu:5371.09035966947 mfu:25.336349629072682 +Step 33967 | loss:0.2912108898162842 lr:1.0242914908747459e-06 tokens_per_second_per_gpu:3693.6395533224286 mfu:17.423527972913845 +Step 33968 | loss:0.33010736107826233 lr:1.0234184673240443e-06 tokens_per_second_per_gpu:4096.932392043083 mfu:19.325929102012804 +Step 33969 | loss:0.3501286804676056 lr:1.0225458121324638e-06 tokens_per_second_per_gpu:3304.5032188207483 mfu:15.587905489564474 +Step 33970 | loss:0.31399214267730713 lr:1.0216735253065657e-06 tokens_per_second_per_gpu:5365.745978155825 mfu:25.311139269627304 +Step 33971 | loss:0.37484854459762573 lr:1.0208016068529113e-06 tokens_per_second_per_gpu:4924.432714632948 mfu:23.22938930978288 +Step 33972 | loss:0.2780658006668091 lr:1.0199300567780567e-06 tokens_per_second_per_gpu:5359.278432023709 mfu:25.28063075104496 +Step 33973 | loss:0.3258908689022064 lr:1.0190588750885576e-06 tokens_per_second_per_gpu:3701.5962894661093 mfu:17.461061254862763 +Step 33974 | loss:0.3206068277359009 lr:1.01818806179097e-06 tokens_per_second_per_gpu:5349.676995947948 mfu:25.235339138901676 +Step 33975 | loss:0.26466524600982666 lr:1.0173176168918386e-06 tokens_per_second_per_gpu:5281.736287640669 mfu:24.914851225936086 +Step 33976 | loss:0.2870521545410156 lr:1.016447540397708e-06 tokens_per_second_per_gpu:4513.579567880954 mfu:21.29132491777853 +Step 33977 | loss:0.37590935826301575 lr:1.0155778323151233e-06 tokens_per_second_per_gpu:4643.602194298339 mfu:21.90466383073696 +Step 33978 | loss:0.3732944428920746 lr:1.0147084926506345e-06 tokens_per_second_per_gpu:5362.5489311917745 mfu:25.296058253625418 +Step 33979 | loss:0.3574782907962799 lr:1.0138395214107754e-06 tokens_per_second_per_gpu:4813.63052686308 mfu:22.70671648526978 +Step 33980 | loss:0.2905206084251404 lr:1.0129709186020742e-06 tokens_per_second_per_gpu:4419.929962961508 mfu:20.84956375310328 +Step 33981 | loss:0.30431848764419556 lr:1.0121026842310755e-06 tokens_per_second_per_gpu:5335.547492293131 mfu:25.168687859419993 +Step 33982 | loss:0.2893047630786896 lr:1.011234818304302e-06 tokens_per_second_per_gpu:4639.272317727504 mfu:21.88423906419921 +Step 33983 | loss:0.3106943964958191 lr:1.010367320828276e-06 tokens_per_second_per_gpu:4538.727859558505 mfu:21.409953700362546 +Step 33984 | loss:0.35921192169189453 lr:1.0095001918095314e-06 tokens_per_second_per_gpu:4408.947305187414 mfu:20.79775668254824 +Step 33985 | loss:0.3542187809944153 lr:1.0086334312545908e-06 tokens_per_second_per_gpu:4134.214221856458 mfu:19.501793854178576 +Step 33986 | loss:0.3173328638076782 lr:1.0077670391699657e-06 tokens_per_second_per_gpu:4615.927098774634 mfu:21.77411568329345 +Step 33987 | loss:0.2567719519138336 lr:1.0069010155621838e-06 tokens_per_second_per_gpu:3883.4453928803546 mfu:18.318874502324167 +Step 33988 | loss:0.3920132517814636 lr:1.0060353604377404e-06 tokens_per_second_per_gpu:4318.119237376113 mfu:20.369305189815975 +Step 33989 | loss:0.26168692111968994 lr:1.0051700738031633e-06 tokens_per_second_per_gpu:5025.598410569974 mfu:23.70660515816542 +Step 33990 | loss:0.3594953715801239 lr:1.004305155664953e-06 tokens_per_second_per_gpu:3896.313627261819 mfu:18.379576159448717 +Step 33991 | loss:0.28099337220191956 lr:1.0034406060296099e-06 tokens_per_second_per_gpu:5307.378226304232 mfu:25.03580862557798 +Step 33992 | loss:0.3774091303348541 lr:1.0025764249036507e-06 tokens_per_second_per_gpu:4065.1505636175093 mfu:19.176008794790288 +Step 33993 | loss:0.363968163728714 lr:1.0017126122935595e-06 tokens_per_second_per_gpu:4948.999326018544 mfu:23.345274207184854 +Step 33994 | loss:0.3560394048690796 lr:1.0008491682058418e-06 tokens_per_second_per_gpu:5290.261105736682 mfu:24.955064247378928 +Step 33995 | loss:0.250209778547287 lr:9.999860926469929e-07 tokens_per_second_per_gpu:5090.022189061355 mfu:24.010503113139315 +Step 33996 | loss:0.2973858416080475 lr:9.991233856235017e-07 tokens_per_second_per_gpu:5378.479112206613 mfu:25.37120363544156 +Step 33997 | loss:0.2685491740703583 lr:9.982610471418518e-07 tokens_per_second_per_gpu:4388.876637370146 mfu:20.703079918045145 +Step 33998 | loss:0.33407989144325256 lr:9.973990772085273e-07 tokens_per_second_per_gpu:5360.0356750890505 mfu:25.284202795783163 +Step 33999 | loss:0.34435245394706726 lr:9.965374758300283e-07 tokens_per_second_per_gpu:4061.5860393113785 mfu:19.15919432546771 +Step 34000 | loss:0.2452484667301178 lr:9.95676243012822e-07 tokens_per_second_per_gpu:3591.1683035249034 mfu:16.940153604220875 +Step 34001 | loss:0.355020135641098 lr:9.948153787633807e-07 tokens_per_second_per_gpu:4734.539100410147 mfu:22.333628732302444 +Step 34002 | loss:0.310929536819458 lr:9.93954883088194e-07 tokens_per_second_per_gpu:5014.069785824239 mfu:23.652222668253582 +Step 34003 | loss:0.38201943039894104 lr:9.930947559937232e-07 tokens_per_second_per_gpu:3901.5013232208275 mfu:18.40404740639956 +Step 34004 | loss:0.40907225012779236 lr:9.922349974864409e-07 tokens_per_second_per_gpu:4412.196000773657 mfu:20.813081333913168 +Step 34005 | loss:0.31251466274261475 lr:9.913756075728087e-07 tokens_per_second_per_gpu:3829.3620146725843 mfu:18.063753979742273 +Step 34006 | loss:0.21047842502593994 lr:9.905165862592991e-07 tokens_per_second_per_gpu:4301.081130529914 mfu:20.288933532821666 +Step 34007 | loss:0.409017950296402 lr:9.896579335523627e-07 tokens_per_second_per_gpu:4755.493700778348 mfu:22.432475157462736 +Step 34008 | loss:0.2926878333091736 lr:9.887996494584662e-07 tokens_per_second_per_gpu:3782.9543901543834 mfu:17.844841296932703 +Step 34009 | loss:0.2779198884963989 lr:9.879417339840602e-07 tokens_per_second_per_gpu:3282.9172438988394 mfu:15.486080762910735 +Step 34010 | loss:0.35387513041496277 lr:9.870841871356006e-07 tokens_per_second_per_gpu:5319.375915260838 mfu:25.09240377140311 +Step 34011 | loss:0.2899640202522278 lr:9.862270089195325e-07 tokens_per_second_per_gpu:3820.200457760512 mfu:18.020537352665237 +Step 34012 | loss:0.32007282972335815 lr:9.853701993423058e-07 tokens_per_second_per_gpu:5338.287681773665 mfu:25.181613800724207 +Step 34013 | loss:0.18901629745960236 lr:9.845137584103604e-07 tokens_per_second_per_gpu:4995.4960613247995 mfu:23.564607240785545 +Step 34014 | loss:0.22749589383602142 lr:9.836576861301517e-07 tokens_per_second_per_gpu:3809.064531952564 mfu:17.96800729064428 +Step 34015 | loss:0.320866197347641 lr:9.828019825080969e-07 tokens_per_second_per_gpu:5348.111788286244 mfu:25.22795578730945 +Step 34016 | loss:0.32312241196632385 lr:9.819466475506523e-07 tokens_per_second_per_gpu:4640.794377797496 mfu:21.89141888123959 +Step 34017 | loss:0.28927937150001526 lr:9.810916812642345e-07 tokens_per_second_per_gpu:4013.560589856595 mfu:18.932650086402884 +Step 34018 | loss:0.22927936911582947 lr:9.802370836552832e-07 tokens_per_second_per_gpu:4546.480211359178 mfu:21.44652286649393 +Step 34019 | loss:0.24512597918510437 lr:9.793828547302264e-07 tokens_per_second_per_gpu:4111.63039872879 mfu:19.395262107287813 +Step 34020 | loss:0.2752517759799957 lr:9.785289944954812e-07 tokens_per_second_per_gpu:5366.358850362862 mfu:25.314030292394676 +Step 34021 | loss:0.24336716532707214 lr:9.776755029574814e-07 tokens_per_second_per_gpu:4979.862674241941 mfu:23.490861886584817 +Step 34022 | loss:0.21912185847759247 lr:9.768223801226328e-07 tokens_per_second_per_gpu:5349.195532864725 mfu:25.233067995392037 +Step 34023 | loss:0.29135584831237793 lr:9.759696259973638e-07 tokens_per_second_per_gpu:4062.475028167511 mfu:19.16338783757923 +Step 34024 | loss:0.3159145414829254 lr:9.751172405880804e-07 tokens_per_second_per_gpu:4890.231920550015 mfu:23.068058328836507 +Step 34025 | loss:0.3819567561149597 lr:9.74265223901194e-07 tokens_per_second_per_gpu:4900.927446651873 mfu:23.11851095030389 +Step 34026 | loss:0.3759903013706207 lr:9.734135759431162e-07 tokens_per_second_per_gpu:5301.589172826242 mfu:25.00850067260793 +Step 34027 | loss:0.29553112387657166 lr:9.725622967202474e-07 tokens_per_second_per_gpu:5330.064280118418 mfu:25.142822612059405 +Step 34028 | loss:0.3262866139411926 lr:9.717113862389992e-07 tokens_per_second_per_gpu:5372.733510222891 mfu:25.344100650583563 +Step 34029 | loss:0.41824445128440857 lr:9.708608445057665e-07 tokens_per_second_per_gpu:4395.72387960745 mfu:20.735379527938495 +Step 34030 | loss:0.22960612177848816 lr:9.700106715269387e-07 tokens_per_second_per_gpu:3356.323840865753 mfu:15.83235220526059 +Step 34031 | loss:0.41475075483322144 lr:9.691608673089214e-07 tokens_per_second_per_gpu:5348.839747381373 mfu:25.231389694563806 +Step 34032 | loss:0.39695873856544495 lr:9.683114318581043e-07 tokens_per_second_per_gpu:4502.984693483247 mfu:21.241347087572574 +Step 34033 | loss:0.381529837846756 lr:9.674623651808712e-07 tokens_per_second_per_gpu:4661.713409714744 mfu:21.990097523948133 +Step 34034 | loss:0.3266613483428955 lr:9.666136672836056e-07 tokens_per_second_per_gpu:4339.058454203095 mfu:20.468079048187636 +Step 34035 | loss:0.20154714584350586 lr:9.657653381727028e-07 tokens_per_second_per_gpu:5356.430549143929 mfu:25.26719680160211 +Step 34036 | loss:0.34480440616607666 lr:9.649173778545295e-07 tokens_per_second_per_gpu:5189.743078268947 mfu:24.48090356166967 +Step 34037 | loss:0.2939929962158203 lr:9.640697863354698e-07 tokens_per_second_per_gpu:5354.131205338089 mfu:25.256350404561456 +Step 34038 | loss:0.4490760862827301 lr:9.632225636219017e-07 tokens_per_second_per_gpu:4205.94906037252 mfu:19.84017932668436 +Step 34039 | loss:0.41253772377967834 lr:9.623757097201925e-07 tokens_per_second_per_gpu:5332.732945780439 mfu:25.15541116330847 +Step 34040 | loss:0.3782903850078583 lr:9.615292246367091e-07 tokens_per_second_per_gpu:4562.848267037909 mfu:21.523733777808776 +Step 34041 | loss:0.27663981914520264 lr:9.606831083778244e-07 tokens_per_second_per_gpu:5379.75017889552 mfu:25.377199473879475 +Step 34042 | loss:0.2792488932609558 lr:9.598373609499056e-07 tokens_per_second_per_gpu:3681.6426023939975 mfu:17.366936308493543 +Step 34043 | loss:0.25474995374679565 lr:9.589919823592974e-07 tokens_per_second_per_gpu:5336.098942334345 mfu:25.17128914335148 +Step 34044 | loss:0.35770753026008606 lr:9.581469726123725e-07 tokens_per_second_per_gpu:5342.776643952194 mfu:25.20278900121625 +Step 34045 | loss:0.28904643654823303 lr:9.573023317154873e-07 tokens_per_second_per_gpu:4278.01025485399 mfu:20.180104275961043 +Step 34046 | loss:0.304002046585083 lr:9.564580596749862e-07 tokens_per_second_per_gpu:5362.732597715057 mfu:25.296924639952483 +Step 34047 | loss:0.2569532096385956 lr:9.556141564972199e-07 tokens_per_second_per_gpu:5303.248905405115 mfu:25.01632991436205 +Step 34048 | loss:0.27774742245674133 lr:9.547706221885333e-07 tokens_per_second_per_gpu:4306.935036218326 mfu:20.31654740473321 +Step 34049 | loss:0.33739542961120605 lr:9.53927456755277e-07 tokens_per_second_per_gpu:4336.691692245546 mfu:20.456914628222645 +Step 34050 | loss:0.3300996422767639 lr:9.530846602037957e-07 tokens_per_second_per_gpu:5126.755486435567 mfu:24.183780344200276 +Step 34051 | loss:0.3288598954677582 lr:9.522422325404235e-07 tokens_per_second_per_gpu:4474.566357169363 mfu:21.10729294651183 +Step 34052 | loss:0.19896200299263 lr:9.514001737714939e-07 tokens_per_second_per_gpu:5084.530306255849 mfu:23.984596964933917 +Step 34053 | loss:0.2631862759590149 lr:9.505584839033411e-07 tokens_per_second_per_gpu:4297.070834010928 mfu:20.27001627991447 +Step 34054 | loss:0.16253110766410828 lr:9.497171629422985e-07 tokens_per_second_per_gpu:4843.637671648412 mfu:22.84826530696902 +Step 34055 | loss:0.3205645680427551 lr:9.488762108946892e-07 tokens_per_second_per_gpu:4440.038351602766 mfu:20.944418453168975 +Step 34056 | loss:0.2618090510368347 lr:9.480356277668412e-07 tokens_per_second_per_gpu:4767.432567836608 mfu:22.488792830357315 +Step 34057 | loss:0.3119991421699524 lr:9.471954135650774e-07 tokens_per_second_per_gpu:5321.8045056654555 mfu:25.10385984670194 +Step 34058 | loss:0.31992217898368835 lr:9.463555682957148e-07 tokens_per_second_per_gpu:3399.14048071155 mfu:16.034325600685268 +Step 34059 | loss:0.3670028746128082 lr:9.455160919650707e-07 tokens_per_second_per_gpu:4417.096597182347 mfu:20.836198283300927 +Step 34060 | loss:0.19962602853775024 lr:9.446769845794623e-07 tokens_per_second_per_gpu:4461.586729897591 mfu:21.04606583011811 +Step 34061 | loss:0.22017143666744232 lr:9.438382461452011e-07 tokens_per_second_per_gpu:3492.9522322723383 mfu:16.4768516387328 +Step 34062 | loss:0.4217771887779236 lr:9.429998766685877e-07 tokens_per_second_per_gpu:5372.73794710446 mfu:25.344121580110933 +Step 34063 | loss:0.3645803928375244 lr:9.421618761559281e-07 tokens_per_second_per_gpu:5162.692116038702 mfu:24.353299557459525 +Step 34064 | loss:0.32458993792533875 lr:9.41324244613534e-07 tokens_per_second_per_gpu:5297.604680407921 mfu:24.9897051420462 +Step 34065 | loss:0.2800651788711548 lr:9.404869820477003e-07 tokens_per_second_per_gpu:5330.874780132339 mfu:25.146645878910633 +Step 34066 | loss:0.29369214177131653 lr:9.39650088464722e-07 tokens_per_second_per_gpu:5323.432863736989 mfu:25.1115410895519 +Step 34067 | loss:0.3435239791870117 lr:9.388135638708995e-07 tokens_per_second_per_gpu:5179.366891730816 mfu:24.43195731940933 +Step 34068 | loss:0.32333874702453613 lr:9.379774082725168e-07 tokens_per_second_per_gpu:4161.6667413071145 mfu:19.63129207231013 +Step 34069 | loss:0.3037988245487213 lr:9.371416216758632e-07 tokens_per_second_per_gpu:4731.829703876041 mfu:22.32084804658035 +Step 34070 | loss:0.21821755170822144 lr:9.363062040872339e-07 tokens_per_second_per_gpu:3613.1590264298816 mfu:17.043887596167878 +Step 34071 | loss:0.27959102392196655 lr:9.35471155512907e-07 tokens_per_second_per_gpu:4353.276927221595 mfu:20.53515001133755 +Step 34072 | loss:0.3208577632904053 lr:9.346364759591553e-07 tokens_per_second_per_gpu:5320.242853859691 mfu:25.096493268688878 +Step 34073 | loss:0.2727636396884918 lr:9.338021654322626e-07 tokens_per_second_per_gpu:3220.0702429106127 mfu:15.1896207364452 +Step 34074 | loss:0.3185330033302307 lr:9.329682239385129e-07 tokens_per_second_per_gpu:5204.362138294124 mfu:24.54986416207727 +Step 34075 | loss:0.33658653497695923 lr:9.321346514841678e-07 tokens_per_second_per_gpu:3566.2940064643435 mfu:16.822817300993506 +Step 34076 | loss:0.2772025465965271 lr:9.313014480755e-07 tokens_per_second_per_gpu:3850.476951570888 mfu:18.163356739671293 +Step 34077 | loss:0.3580121397972107 lr:9.304686137187713e-07 tokens_per_second_per_gpu:5298.238962660119 mfu:24.992697159649527 +Step 34078 | loss:0.3304695188999176 lr:9.296361484202542e-07 tokens_per_second_per_gpu:5320.753914176803 mfu:25.098904027400575 +Step 34079 | loss:0.344434529542923 lr:9.288040521861996e-07 tokens_per_second_per_gpu:3229.877073275881 mfu:15.235881228495966 +Step 34080 | loss:0.3299572169780731 lr:9.279723250228689e-07 tokens_per_second_per_gpu:4477.476445754299 mfu:21.121020330879283 +Step 34081 | loss:0.31672152876853943 lr:9.271409669365294e-07 tokens_per_second_per_gpu:4399.32094862988 mfu:20.75234751624056 +Step 34082 | loss:0.26300668716430664 lr:9.263099779334206e-07 tokens_per_second_per_gpu:5346.010702183917 mfu:25.21804460568252 +Step 34083 | loss:0.4055856764316559 lr:9.254793580197929e-07 tokens_per_second_per_gpu:5061.091189221854 mfu:23.87403065075882 +Step 34084 | loss:0.33350351452827454 lr:9.24649107201897e-07 tokens_per_second_per_gpu:4883.249048109204 mfu:23.035118928132302 +Step 34085 | loss:0.4896586239337921 lr:9.238192254859835e-07 tokens_per_second_per_gpu:5031.83817662788 mfu:23.73603919927443 +Step 34086 | loss:0.36615708470344543 lr:9.229897128782861e-07 tokens_per_second_per_gpu:5330.00916745095 mfu:25.14256263620377 +Step 34087 | loss:0.3214908242225647 lr:9.221605693850388e-07 tokens_per_second_per_gpu:5308.368233748608 mfu:25.040478659605792 +Step 34088 | loss:0.23305806517601013 lr:9.213317950124923e-07 tokens_per_second_per_gpu:5364.047860954875 mfu:25.303128961061756 +Step 34089 | loss:0.25604015588760376 lr:9.205033897668635e-07 tokens_per_second_per_gpu:5350.717913962998 mfu:25.240249326777594 +Step 34090 | loss:0.4591851830482483 lr:9.196753536544034e-07 tokens_per_second_per_gpu:5317.398451345479 mfu:25.083075736724226 +Step 34091 | loss:0.3112662732601166 lr:9.188476866813178e-07 tokens_per_second_per_gpu:5317.55039215998 mfu:25.083792467470005 +Step 34092 | loss:0.2798183560371399 lr:9.180203888538408e-07 tokens_per_second_per_gpu:4243.198085460968 mfu:20.015889333365593 +Step 34093 | loss:0.33040788769721985 lr:9.171934601782007e-07 tokens_per_second_per_gpu:3283.3108256710298 mfu:15.487937355282131 +Step 34094 | loss:0.2949756383895874 lr:9.163669006606091e-07 tokens_per_second_per_gpu:4373.967027230018 mfu:20.632748743172357 +Step 34095 | loss:0.22192807495594025 lr:9.155407103072833e-07 tokens_per_second_per_gpu:5380.8865714261 mfu:25.382560031334783 +Step 34096 | loss:0.3420858085155487 lr:9.147148891244406e-07 tokens_per_second_per_gpu:3339.3544983927877 mfu:15.752304921547235 +Step 34097 | loss:0.30417805910110474 lr:9.138894371182983e-07 tokens_per_second_per_gpu:3314.0538641469793 mfu:15.632957513076427 +Step 34098 | loss:0.2552649974822998 lr:9.130643542950457e-07 tokens_per_second_per_gpu:4975.286342437653 mfu:23.46927454866127 +Step 34099 | loss:0.3174133598804474 lr:9.122396406609001e-07 tokens_per_second_per_gpu:4773.23321484724 mfu:22.5161554719989 +Step 34100 | loss:0.36223283410072327 lr:9.114152962220735e-07 tokens_per_second_per_gpu:5067.18370286047 mfu:23.902770077082184 +Step 34101 | loss:0.19968052208423615 lr:9.105913209847439e-07 tokens_per_second_per_gpu:4912.986971141438 mfu:23.175397784888215 +Step 34102 | loss:0.22873492538928986 lr:9.097677149551287e-07 tokens_per_second_per_gpu:5343.81808405749 mfu:25.207701651880935 +Step 34103 | loss:0.38885802030563354 lr:9.089444781394174e-07 tokens_per_second_per_gpu:4167.453192659324 mfu:19.658587750608906 +Step 34104 | loss:0.37689104676246643 lr:9.08121610543794e-07 tokens_per_second_per_gpu:4958.684788110421 mfu:23.39096218438247 +Step 34105 | loss:0.33110326528549194 lr:9.072991121744534e-07 tokens_per_second_per_gpu:5355.905583183297 mfu:25.26472044759738 +Step 34106 | loss:0.2522866129875183 lr:9.064769830375741e-07 tokens_per_second_per_gpu:5349.2369685660615 mfu:25.233263454664907 +Step 34107 | loss:0.3395746946334839 lr:9.056552231393567e-07 tokens_per_second_per_gpu:5225.96510017803 mfu:24.651769019129027 +Step 34108 | loss:0.32920020818710327 lr:9.048338324859628e-07 tokens_per_second_per_gpu:4355.419786285696 mfu:20.54525824315237 +Step 34109 | loss:0.29732266068458557 lr:9.040128110835766e-07 tokens_per_second_per_gpu:5336.3768128264755 mfu:25.172599905872453 +Step 34110 | loss:0.2624604105949402 lr:9.031921589383819e-07 tokens_per_second_per_gpu:4603.190116355091 mfu:21.71403316406717 +Step 34111 | loss:0.29806020855903625 lr:9.023718760565458e-07 tokens_per_second_per_gpu:3938.2972594806592 mfu:18.577620115771285 +Step 34112 | loss:0.21157479286193848 lr:9.015519624442248e-07 tokens_per_second_per_gpu:5327.512160093098 mfu:25.130783826463393 +Step 34113 | loss:0.2617904245853424 lr:9.007324181076027e-07 tokens_per_second_per_gpu:4818.040148449877 mfu:22.727517422653722 +Step 34114 | loss:0.29502561688423157 lr:8.999132430528412e-07 tokens_per_second_per_gpu:5363.280605351739 mfu:25.299509685474455 +Step 34115 | loss:0.28951790928840637 lr:8.99094437286091e-07 tokens_per_second_per_gpu:5011.94038386925 mfu:23.642177915918456 +Step 34116 | loss:0.29718029499053955 lr:8.982760008135139e-07 tokens_per_second_per_gpu:5260.757318959038 mfu:24.815889851283682 +Step 34117 | loss:0.26200053095817566 lr:8.97457933641277e-07 tokens_per_second_per_gpu:5013.7767876348735 mfu:23.650840545803696 +Step 34118 | loss:0.27927902340888977 lr:8.966402357755254e-07 tokens_per_second_per_gpu:4346.679822699524 mfu:20.504030343724846 +Step 34119 | loss:0.25628116726875305 lr:8.958229072223989e-07 tokens_per_second_per_gpu:5360.442324304127 mfu:25.28612102951219 +Step 34120 | loss:0.34301725029945374 lr:8.95005947988059e-07 tokens_per_second_per_gpu:4557.825056144818 mfu:21.50003843497791 +Step 34121 | loss:0.26071637868881226 lr:8.941893580786398e-07 tokens_per_second_per_gpu:5223.142895139635 mfu:24.638456196445084 +Step 34122 | loss:0.3090924322605133 lr:8.933731375002974e-07 tokens_per_second_per_gpu:4724.045677292697 mfu:22.284129465094832 +Step 34123 | loss:0.2043234258890152 lr:8.925572862591547e-07 tokens_per_second_per_gpu:5351.390265176957 mfu:25.243420922168784 +Step 34124 | loss:0.3371773064136505 lr:8.917418043613567e-07 tokens_per_second_per_gpu:4004.625018556691 mfu:18.890499472015563 +Step 34125 | loss:0.28951165080070496 lr:8.909266918130377e-07 tokens_per_second_per_gpu:3623.8400532375413 mfu:17.0942718219907 +Step 34126 | loss:0.26200926303863525 lr:8.901119486203313e-07 tokens_per_second_per_gpu:5317.822856791017 mfu:25.08507772962187 +Step 34127 | loss:0.3433997929096222 lr:8.892975747893495e-07 tokens_per_second_per_gpu:4719.731174965163 mfu:22.26377722148585 +Step 34128 | loss:0.3370329439640045 lr:8.884835703262262e-07 tokens_per_second_per_gpu:4807.749472989039 mfu:22.678974550734665 +Step 34129 | loss:0.30899837613105774 lr:8.876699352370899e-07 tokens_per_second_per_gpu:3388.6468627295417 mfu:15.984825414268284 +Step 34130 | loss:0.31922146677970886 lr:8.868566695280523e-07 tokens_per_second_per_gpu:4719.550548771387 mfu:22.262925176911804 +Step 34131 | loss:0.29200300574302673 lr:8.860437732052307e-07 tokens_per_second_per_gpu:4179.930678685214 mfu:19.71744618107167 +Step 34132 | loss:0.30650609731674194 lr:8.852312462747481e-07 tokens_per_second_per_gpu:5189.521233914487 mfu:24.479857083999136 +Step 34133 | loss:0.38072752952575684 lr:8.844190887427051e-07 tokens_per_second_per_gpu:5321.670132586297 mfu:25.103225985961135 +Step 34134 | loss:0.3086409270763397 lr:8.836073006152079e-07 tokens_per_second_per_gpu:5266.972215892731 mfu:24.845206580490668 +Step 34135 | loss:0.22778986394405365 lr:8.82795881898374e-07 tokens_per_second_per_gpu:3839.1630945745496 mfu:18.10998734587651 +Step 34136 | loss:0.3047936260700226 lr:8.819848325982982e-07 tokens_per_second_per_gpu:5341.443226852451 mfu:25.19649904525235 +Step 34137 | loss:0.3615053594112396 lr:8.811741527210759e-07 tokens_per_second_per_gpu:5316.154744655948 mfu:25.077208960071466 +Step 34138 | loss:0.2692835032939911 lr:8.803638422728133e-07 tokens_per_second_per_gpu:5336.54043050759 mfu:25.173371718390325 +Step 34139 | loss:0.3247053027153015 lr:8.79553901259611e-07 tokens_per_second_per_gpu:5369.606367655816 mfu:25.329349385549133 +Step 34140 | loss:0.32846757769584656 lr:8.787443296875475e-07 tokens_per_second_per_gpu:4750.366157966471 mfu:22.4082876631701 +Step 34141 | loss:0.2752106189727783 lr:8.779351275627068e-07 tokens_per_second_per_gpu:5360.345402579046 mfu:25.28566383319843 +Step 34142 | loss:0.3531345725059509 lr:8.771262948911895e-07 tokens_per_second_per_gpu:3813.4532298741847 mfu:17.988709527529643 +Step 34143 | loss:0.19399486482143402 lr:8.763178316790798e-07 tokens_per_second_per_gpu:5328.989113886408 mfu:25.137750869500735 +Step 34144 | loss:0.26801779866218567 lr:8.755097379324451e-07 tokens_per_second_per_gpu:5251.22284347871 mfu:24.770914103693716 +Step 34145 | loss:0.278329461812973 lr:8.747020136573691e-07 tokens_per_second_per_gpu:5310.37804225606 mfu:25.049959269244134 +Step 34146 | loss:0.25670886039733887 lr:8.738946588599306e-07 tokens_per_second_per_gpu:3794.3875091752634 mfu:17.89877327004545 +Step 34147 | loss:0.3272477388381958 lr:8.730876735462023e-07 tokens_per_second_per_gpu:5346.397699152606 mfu:25.219870136412318 +Step 34148 | loss:0.29137861728668213 lr:8.722810577222406e-07 tokens_per_second_per_gpu:5371.368094735925 mfu:25.33765975276764 +Step 34149 | loss:0.35519638657569885 lr:8.714748113941296e-07 tokens_per_second_per_gpu:5339.874563270523 mfu:25.189099391494786 +Step 34150 | loss:0.3358795940876007 lr:8.706689345679253e-07 tokens_per_second_per_gpu:4380.147203575135 mfu:20.661901689439087 +Step 34151 | loss:0.4052514135837555 lr:8.698634272496842e-07 tokens_per_second_per_gpu:5009.612516083496 mfu:23.631196966397855 +Step 34152 | loss:0.31227830052375793 lr:8.690582894454679e-07 tokens_per_second_per_gpu:5076.394847020494 mfu:23.94622061566827 +Step 34153 | loss:0.2328409105539322 lr:8.682535211613385e-07 tokens_per_second_per_gpu:5329.1944926997385 mfu:25.138719676404456 +Step 34154 | loss:0.3387034237384796 lr:8.674491224033465e-07 tokens_per_second_per_gpu:5349.278961193564 mfu:25.233461541053515 +Step 34155 | loss:0.43450868129730225 lr:8.666450931775316e-07 tokens_per_second_per_gpu:5349.304142122018 mfu:25.23358032378934 +Step 34156 | loss:0.35647693276405334 lr:8.658414334899501e-07 tokens_per_second_per_gpu:4565.380287584011 mfu:21.535677750731992 +Step 34157 | loss:0.3080899119377136 lr:8.65038143346647e-07 tokens_per_second_per_gpu:5345.8659791891705 mfu:25.21736192262401 +Step 34158 | loss:0.29324275255203247 lr:8.642352227536621e-07 tokens_per_second_per_gpu:4884.8784952313845 mfu:23.042805308219975 +Step 34159 | loss:0.35346126556396484 lr:8.634326717170294e-07 tokens_per_second_per_gpu:5174.375774900424 mfu:24.40841336974803 +Step 34160 | loss:0.2796669006347656 lr:8.626304902427939e-07 tokens_per_second_per_gpu:4355.68273090676 mfu:20.54649859779291 +Step 34161 | loss:0.26053494215011597 lr:8.618286783369899e-07 tokens_per_second_per_gpu:5367.845329042404 mfu:25.321042265945497 +Step 34162 | loss:0.3114423155784607 lr:8.610272360056349e-07 tokens_per_second_per_gpu:4476.64502883638 mfu:21.11709839542311 +Step 34163 | loss:0.31050339341163635 lr:8.602261632547626e-07 tokens_per_second_per_gpu:5371.99718458657 mfu:25.340627276926927 +Step 34164 | loss:0.2390989065170288 lr:8.594254600904017e-07 tokens_per_second_per_gpu:5360.266241004033 mfu:25.28529041454274 +Step 34165 | loss:0.19307111203670502 lr:8.586251265185752e-07 tokens_per_second_per_gpu:5104.205271938013 mfu:24.07740713495198 +Step 34166 | loss:0.3280719220638275 lr:8.578251625452949e-07 tokens_per_second_per_gpu:5332.781760144536 mfu:25.155641428992507 +Step 34167 | loss:0.3074216842651367 lr:8.570255681765782e-07 tokens_per_second_per_gpu:4845.154433202358 mfu:22.85542013000519 +Step 34168 | loss:0.2724873423576355 lr:8.562263434184537e-07 tokens_per_second_per_gpu:5236.782827824835 mfu:24.702798086133665 +Step 34169 | loss:0.32740673422813416 lr:8.554274882769219e-07 tokens_per_second_per_gpu:4425.025951101786 mfu:20.873602398626204 +Step 34170 | loss:0.3272005021572113 lr:8.54629002757984e-07 tokens_per_second_per_gpu:5344.923013700357 mfu:25.212913793526777 +Step 34171 | loss:0.20118598639965057 lr:8.538308868676514e-07 tokens_per_second_per_gpu:5333.355677614675 mfu:25.158348695619573 +Step 34172 | loss:0.28319400548934937 lr:8.530331406119308e-07 tokens_per_second_per_gpu:5337.391786376525 mfu:25.17738770928038 +Step 34173 | loss:0.3700868487358093 lr:8.522357639968171e-07 tokens_per_second_per_gpu:5351.310132379229 mfu:25.243042922089835 +Step 34174 | loss:0.3393736481666565 lr:8.514387570283111e-07 tokens_per_second_per_gpu:4145.2270947582 mfu:19.55374345464597 +Step 34175 | loss:0.2889401316642761 lr:8.506421197124081e-07 tokens_per_second_per_gpu:4197.9246584562225 mfu:19.802326853741043 +Step 34176 | loss:0.35879215598106384 lr:8.498458520550978e-07 tokens_per_second_per_gpu:4397.835031755832 mfu:20.74533819282211 +Step 34177 | loss:0.2572261095046997 lr:8.490499540623642e-07 tokens_per_second_per_gpu:4745.018037383201 mfu:22.38305966589535 +Step 34178 | loss:0.35942891240119934 lr:8.482544257401969e-07 tokens_per_second_per_gpu:4643.521843052256 mfu:21.904284800199854 +Step 34179 | loss:0.36819934844970703 lr:8.474592670945858e-07 tokens_per_second_per_gpu:5226.028082183818 mfu:24.652066115996167 +Step 34180 | loss:0.48449382185935974 lr:8.46664478131498e-07 tokens_per_second_per_gpu:5362.292146090824 mfu:25.294846954492183 +Step 34181 | loss:0.2663053274154663 lr:8.458700588569235e-07 tokens_per_second_per_gpu:3947.3716509230967 mfu:18.620425568456092 +Step 34182 | loss:0.3449386954307556 lr:8.45076009276835e-07 tokens_per_second_per_gpu:3838.600330433483 mfu:18.107332691405638 +Step 34183 | loss:0.2711460590362549 lr:8.442823293972002e-07 tokens_per_second_per_gpu:4937.216651729735 mfu:23.289693322233006 +Step 34184 | loss:0.3327191472053528 lr:8.434890192239864e-07 tokens_per_second_per_gpu:4363.554294606469 mfu:20.58363010679169 +Step 34185 | loss:0.3561115264892578 lr:8.426960787631666e-07 tokens_per_second_per_gpu:3458.031384061785 mfu:16.312124039612534 +Step 34186 | loss:0.3166535198688507 lr:8.419035080207027e-07 tokens_per_second_per_gpu:5376.208436922235 mfu:25.360492472708 +Step 34187 | loss:0.33121925592422485 lr:8.411113070025512e-07 tokens_per_second_per_gpu:4625.568627370317 mfu:21.819596418693692 +Step 34188 | loss:0.23298555612564087 lr:8.403194757146737e-07 tokens_per_second_per_gpu:3598.835976455469 mfu:16.976323325674255 +Step 34189 | loss:0.312700092792511 lr:8.395280141630324e-07 tokens_per_second_per_gpu:3180.2739626059683 mfu:15.001894892302614 +Step 34190 | loss:0.34810420870780945 lr:8.387369223535779e-07 tokens_per_second_per_gpu:5319.409592278237 mfu:25.092562631640074 +Step 34191 | loss:0.30218008160591125 lr:8.379462002922445e-07 tokens_per_second_per_gpu:4169.690721000359 mfu:19.66914255355527 +Step 34192 | loss:0.3393840193748474 lr:8.371558479849939e-07 tokens_per_second_per_gpu:3266.043221531476 mfu:15.406483120398791 +Step 34193 | loss:0.2550126612186432 lr:8.36365865437766e-07 tokens_per_second_per_gpu:4680.650915822781 mfu:22.079428971331463 +Step 34194 | loss:0.4088953137397766 lr:8.355762526565114e-07 tokens_per_second_per_gpu:4044.215572094795 mfu:19.077254867899768 +Step 34195 | loss:0.3471129238605499 lr:8.347870096471533e-07 tokens_per_second_per_gpu:4560.58233670545 mfu:21.513044997822984 +Step 34196 | loss:0.32928481698036194 lr:8.339981364156368e-07 tokens_per_second_per_gpu:5352.9681232579715 mfu:25.25086394794728 +Step 34197 | loss:0.29818469285964966 lr:8.332096329679073e-07 tokens_per_second_per_gpu:5337.32912670568 mfu:25.177092132922162 +Step 34198 | loss:0.27820494771003723 lr:8.324214993098655e-07 tokens_per_second_per_gpu:5006.948974880949 mfu:23.618632588098453 +Step 34199 | loss:0.314301073551178 lr:8.316337354474624e-07 tokens_per_second_per_gpu:4689.118533073328 mfu:22.11937216662677 +Step 34200 | loss:0.36206603050231934 lr:8.308463413866097e-07 tokens_per_second_per_gpu:4234.244808994886 mfu:19.973655200688597 +Step 34201 | loss:0.2622779607772827 lr:8.300593171332416e-07 tokens_per_second_per_gpu:3562.4059618234105 mfu:16.804476731053406 +Step 34202 | loss:0.27560701966285706 lr:8.292726626932701e-07 tokens_per_second_per_gpu:3965.4343275716583 mfu:18.70563029601778 +Step 34203 | loss:0.33494070172309875 lr:8.284863780726127e-07 tokens_per_second_per_gpu:5321.597923684005 mfu:25.10288536424892 +Step 34204 | loss:0.2706463038921356 lr:8.277004632771868e-07 tokens_per_second_per_gpu:5318.061234319648 mfu:25.08620219709582 +Step 34205 | loss:0.3437157869338989 lr:8.269149183128988e-07 tokens_per_second_per_gpu:3828.096713406432 mfu:18.057785337787177 +Step 34206 | loss:0.2675192058086395 lr:8.261297431856551e-07 tokens_per_second_per_gpu:4200.161542781748 mfu:19.81287861875637 +Step 34207 | loss:0.405043363571167 lr:8.253449379013623e-07 tokens_per_second_per_gpu:4255.2554865415395 mfu:20.072766151467405 +Step 34208 | loss:0.327303409576416 lr:8.24560502465932e-07 tokens_per_second_per_gpu:3282.985933870335 mfu:15.486404785225997 +Step 34209 | loss:0.3429488241672516 lr:8.237764368852541e-07 tokens_per_second_per_gpu:5378.172065285308 mfu:25.369755242726164 +Step 34210 | loss:0.32685619592666626 lr:8.229927411652239e-07 tokens_per_second_per_gpu:4399.874378410975 mfu:20.754958138942843 +Step 34211 | loss:0.301767498254776 lr:8.222094153117476e-07 tokens_per_second_per_gpu:3769.7472958356802 mfu:17.78254117966873 +Step 34212 | loss:0.16935771703720093 lr:8.214264593307098e-07 tokens_per_second_per_gpu:4253.418633075677 mfu:20.06410140026941 +Step 34213 | loss:0.3355077803134918 lr:8.206438732279942e-07 tokens_per_second_per_gpu:4629.72665215452 mfu:21.839210531032958 +Step 34214 | loss:0.2764175236225128 lr:8.198616570094908e-07 tokens_per_second_per_gpu:4769.866611067442 mfu:22.50027462756776 +Step 34215 | loss:0.3070550858974457 lr:8.190798106810838e-07 tokens_per_second_per_gpu:4461.5164562739055 mfu:21.04573433744527 +Step 34216 | loss:0.23920755088329315 lr:8.182983342486517e-07 tokens_per_second_per_gpu:5309.121898410372 mfu:25.044033824403783 +Step 34217 | loss:0.4328128695487976 lr:8.175172277180787e-07 tokens_per_second_per_gpu:5028.024746610928 mfu:23.718050599246215 +Step 34218 | loss:0.2967497408390045 lr:8.167364910952324e-07 tokens_per_second_per_gpu:5354.874787586011 mfu:25.25985800889312 +Step 34219 | loss:0.2751327157020569 lr:8.159561243859859e-07 tokens_per_second_per_gpu:4440.769999009515 mfu:20.947869758818527 +Step 34220 | loss:0.25877004861831665 lr:8.151761275962123e-07 tokens_per_second_per_gpu:3557.7024980459646 mfu:16.782289689921523 +Step 34221 | loss:0.3343483805656433 lr:8.143965007317677e-07 tokens_per_second_per_gpu:4107.469820888443 mfu:19.375635951747118 +Step 34222 | loss:0.3255603015422821 lr:8.136172437985312e-07 tokens_per_second_per_gpu:3795.665014522886 mfu:17.904799480735072 +Step 34223 | loss:0.3718525171279907 lr:8.128383568023479e-07 tokens_per_second_per_gpu:4285.923958461157 mfu:20.217434566093115 +Step 34224 | loss:0.28846868872642517 lr:8.120598397490853e-07 tokens_per_second_per_gpu:4867.085273215875 mfu:22.958871644136096 +Step 34225 | loss:0.32773545384407043 lr:8.112816926445998e-07 tokens_per_second_per_gpu:5135.678796062322 mfu:24.225873118183298 +Step 34226 | loss:0.305372029542923 lr:8.10503915494748e-07 tokens_per_second_per_gpu:4356.921508438647 mfu:20.55234212276806 +Step 34227 | loss:0.24353311955928802 lr:8.09726508305364e-07 tokens_per_second_per_gpu:5325.675816350792 mfu:25.122121479718388 +Step 34228 | loss:0.31485968828201294 lr:8.089494710823043e-07 tokens_per_second_per_gpu:4675.873231916741 mfu:22.056891821189744 +Step 34229 | loss:0.25087079405784607 lr:8.08172803831414e-07 tokens_per_second_per_gpu:4946.432287668919 mfu:23.33316504930777 +Step 34230 | loss:0.3335602581501007 lr:8.073965065585331e-07 tokens_per_second_per_gpu:2928.040867057452 mfu:13.812068345196494 +Step 34231 | loss:0.24514001607894897 lr:8.066205792694959e-07 tokens_per_second_per_gpu:4133.490517092815 mfu:19.49838001533171 +Step 34232 | loss:0.31459468603134155 lr:8.058450219701474e-07 tokens_per_second_per_gpu:3620.0146568983846 mfu:17.076226774778977 +Step 34233 | loss:0.20015482604503632 lr:8.050698346663165e-07 tokens_per_second_per_gpu:4268.287802514881 mfu:20.13424180945605 +Step 34234 | loss:0.31594935059547424 lr:8.04295017363832e-07 tokens_per_second_per_gpu:3976.074167501057 mfu:18.75582023630864 +Step 34235 | loss:0.24160347878932953 lr:8.035205700685167e-07 tokens_per_second_per_gpu:4982.880366020797 mfu:23.50509685357699 +Step 34236 | loss:0.2731805443763733 lr:8.027464927862049e-07 tokens_per_second_per_gpu:3166.664343564049 mfu:14.937695997241432 +Step 34237 | loss:0.26259732246398926 lr:8.019727855227144e-07 tokens_per_second_per_gpu:5303.423807840875 mfu:25.01715495899297 +Step 34238 | loss:0.31396204233169556 lr:8.011994482838625e-07 tokens_per_second_per_gpu:4172.271104260277 mfu:19.681314661651086 +Step 34239 | loss:0.3480668067932129 lr:8.00426481075467e-07 tokens_per_second_per_gpu:4928.952243231663 mfu:23.25070869729411 +Step 34240 | loss:0.3196098208427429 lr:7.996538839033507e-07 tokens_per_second_per_gpu:3861.5456184407863 mfu:18.215569529805908 +Step 34241 | loss:0.28907379508018494 lr:7.988816567733093e-07 tokens_per_second_per_gpu:4401.041058458627 mfu:20.76046156778376 +Step 34242 | loss:0.33684617280960083 lr:7.981097996911546e-07 tokens_per_second_per_gpu:4064.197502407519 mfu:19.171513042453757 +Step 34243 | loss:0.3481714427471161 lr:7.973383126626932e-07 tokens_per_second_per_gpu:4217.129211347358 mfu:19.892918006363132 +Step 34244 | loss:0.3022700846195221 lr:7.965671956937315e-07 tokens_per_second_per_gpu:4681.940797840684 mfu:22.08551356488648 +Step 34245 | loss:0.3223935663700104 lr:7.957964487900649e-07 tokens_per_second_per_gpu:3694.1287787272904 mfu:17.425835732619042 +Step 34246 | loss:0.22474367916584015 lr:7.950260719574942e-07 tokens_per_second_per_gpu:4861.485906763244 mfu:22.93245847722861 +Step 34247 | loss:0.2294187992811203 lr:7.942560652018093e-07 tokens_per_second_per_gpu:3661.466585441154 mfu:17.271762594143517 +Step 34248 | loss:0.31355977058410645 lr:7.934864285288113e-07 tokens_per_second_per_gpu:4237.08643887826 mfu:19.98705965368053 +Step 34249 | loss:0.31322306394577026 lr:7.927171619442675e-07 tokens_per_second_per_gpu:4518.768379365271 mfu:21.315801426852587 +Step 34250 | loss:0.2982642352581024 lr:7.919482654539845e-07 tokens_per_second_per_gpu:4311.462738961616 mfu:20.337905351079055 +Step 34251 | loss:0.2721741497516632 lr:7.911797390637354e-07 tokens_per_second_per_gpu:3707.008298063067 mfu:17.486590622798495 +Step 34252 | loss:0.3758624792098999 lr:7.904115827792991e-07 tokens_per_second_per_gpu:5395.819865939446 mfu:25.45300292943701 +Step 34253 | loss:0.3303218483924866 lr:7.896437966064596e-07 tokens_per_second_per_gpu:4268.050175781833 mfu:20.133120883612957 +Step 34254 | loss:0.37736672163009644 lr:7.888763805509902e-07 tokens_per_second_per_gpu:5340.815590844056 mfu:25.193538379115 +Step 34255 | loss:0.2721126079559326 lr:7.881093346186641e-07 tokens_per_second_per_gpu:4334.955472459835 mfu:20.448724583263807 +Step 34256 | loss:0.30987223982810974 lr:7.873426588152433e-07 tokens_per_second_per_gpu:4057.460752292492 mfu:19.13973464275308 +Step 34257 | loss:0.28391343355178833 lr:7.865763531464954e-07 tokens_per_second_per_gpu:3953.1756449863237 mfu:18.647803998715833 +Step 34258 | loss:0.21321365237236023 lr:7.85810417618188e-07 tokens_per_second_per_gpu:5009.839325412627 mfu:23.632266864701027 +Step 34259 | loss:0.24905212223529816 lr:7.850448522360776e-07 tokens_per_second_per_gpu:3931.928735370124 mfu:18.54757870095963 +Step 34260 | loss:0.24883155524730682 lr:7.842796570059263e-07 tokens_per_second_per_gpu:4349.01044233943 mfu:20.515024274210994 +Step 34261 | loss:0.39992091059684753 lr:7.835148319334907e-07 tokens_per_second_per_gpu:3710.753437270637 mfu:17.50425708342724 +Step 34262 | loss:0.28726670145988464 lr:7.827503770245214e-07 tokens_per_second_per_gpu:5345.885192069692 mfu:25.217452553059488 +Step 34263 | loss:0.24377767741680145 lr:7.819862922847642e-07 tokens_per_second_per_gpu:4759.358005883296 mfu:22.450703743961213 +Step 34264 | loss:0.38400793075561523 lr:7.812225777199644e-07 tokens_per_second_per_gpu:4525.67062405849 mfu:21.34836049271488 +Step 34265 | loss:0.28647303581237793 lr:7.804592333358784e-07 tokens_per_second_per_gpu:4295.316341775483 mfu:20.26174004069263 +Step 34266 | loss:0.32909485697746277 lr:7.796962591382295e-07 tokens_per_second_per_gpu:4846.821030059547 mfu:22.8632817517313 +Step 34267 | loss:0.3643135726451874 lr:7.789336551327686e-07 tokens_per_second_per_gpu:5372.836712278792 mfu:25.344587472289334 +Step 34268 | loss:0.35423436760902405 lr:7.781714213252301e-07 tokens_per_second_per_gpu:4063.015857435921 mfu:19.165939021513307 +Step 34269 | loss:0.20609627664089203 lr:7.774095577213536e-07 tokens_per_second_per_gpu:5362.838623432914 mfu:25.297424781353467 +Step 34270 | loss:0.3445312976837158 lr:7.766480643268458e-07 tokens_per_second_per_gpu:4317.281065928853 mfu:20.365351392092457 +Step 34271 | loss:0.3332323133945465 lr:7.758869411474579e-07 tokens_per_second_per_gpu:4086.5522593747164 mfu:19.276964245182864 +Step 34272 | loss:0.3145162761211395 lr:7.751261881889016e-07 tokens_per_second_per_gpu:4199.003737177042 mfu:19.807417052176962 +Step 34273 | loss:0.30379873514175415 lr:7.743658054569003e-07 tokens_per_second_per_gpu:5344.961966815922 mfu:25.213097542018986 +Step 34274 | loss:0.3025582432746887 lr:7.736057929571771e-07 tokens_per_second_per_gpu:5002.196952091571 mfu:23.596216485822264 +Step 34275 | loss:0.30741533637046814 lr:7.728461506954499e-07 tokens_per_second_per_gpu:4081.9544413193853 mfu:19.25527555294721 +Step 34276 | loss:0.28519728779792786 lr:7.720868786774249e-07 tokens_per_second_per_gpu:4897.0802242773425 mfu:23.10036294595941 +Step 34277 | loss:0.4255969524383545 lr:7.7132797690882e-07 tokens_per_second_per_gpu:3313.2178972886477 mfu:15.629014114775023 +Step 34278 | loss:0.2694127559661865 lr:7.705694453953305e-07 tokens_per_second_per_gpu:3162.539132070808 mfu:14.918236670794156 +Step 34279 | loss:0.366540789604187 lr:7.698112841426741e-07 tokens_per_second_per_gpu:5354.48410809393 mfu:25.258015107072183 +Step 34280 | loss:0.24331648647785187 lr:7.690534931565518e-07 tokens_per_second_per_gpu:3968.5099023698403 mfu:18.720138307087975 +Step 34281 | loss:0.41059738397598267 lr:7.682960724426591e-07 tokens_per_second_per_gpu:4620.693720047557 mfu:21.796600649107273 +Step 34282 | loss:0.2529659569263458 lr:7.675390220066858e-07 tokens_per_second_per_gpu:5350.683410884351 mfu:25.240086569868783 +Step 34283 | loss:0.33324021100997925 lr:7.66782341854344e-07 tokens_per_second_per_gpu:3902.005865878084 mfu:18.406427420197694 +Step 34284 | loss:0.29150256514549255 lr:7.660260319913127e-07 tokens_per_second_per_gpu:5367.915481690242 mfu:25.321373187954624 +Step 34285 | loss:0.33597782254219055 lr:7.652700924232814e-07 tokens_per_second_per_gpu:4454.1573253687875 mfu:21.011020061368622 +Step 34286 | loss:0.2449992448091507 lr:7.645145231559292e-07 tokens_per_second_per_gpu:4708.7241216513385 mfu:22.21185507300759 +Step 34287 | loss:0.28096118569374084 lr:7.63759324194957e-07 tokens_per_second_per_gpu:5010.254509926986 mfu:23.634225360893048 +Step 34288 | loss:0.2940573990345001 lr:7.630044955460214e-07 tokens_per_second_per_gpu:5352.13292829794 mfu:25.246924190821858 +Step 34289 | loss:0.31465086340904236 lr:7.622500372148178e-07 tokens_per_second_per_gpu:5360.158298357474 mfu:25.284781230662006 +Step 34290 | loss:0.319551944732666 lr:7.614959492070195e-07 tokens_per_second_per_gpu:5238.9347500860495 mfu:24.71294907059561 +Step 34291 | loss:0.36875781416893005 lr:7.607422315282886e-07 tokens_per_second_per_gpu:3551.808125747129 mfu:16.7544849301043 +Step 34292 | loss:0.2555994391441345 lr:7.599888841842983e-07 tokens_per_second_per_gpu:4549.716970918083 mfu:21.46179121358141 +Step 34293 | loss:0.37671300768852234 lr:7.592359071807109e-07 tokens_per_second_per_gpu:3603.13209400809 mfu:16.996588845163206 +Step 34294 | loss:0.37052395939826965 lr:7.584833005231994e-07 tokens_per_second_per_gpu:5338.680913483194 mfu:25.1834687417903 +Step 34295 | loss:0.34232956171035767 lr:7.577310642174151e-07 tokens_per_second_per_gpu:5364.471161588638 mfu:25.30512574237375 +Step 34296 | loss:0.2921743392944336 lr:7.569791982690144e-07 tokens_per_second_per_gpu:3716.712120335563 mfu:17.532365208100664 +Step 34297 | loss:0.36653971672058105 lr:7.56227702683665e-07 tokens_per_second_per_gpu:3628.294188488087 mfu:17.115282737921362 +Step 34298 | loss:0.3194475769996643 lr:7.554765774670069e-07 tokens_per_second_per_gpu:3021.608997763013 mfu:14.253445182103917 +Step 34299 | loss:0.29919177293777466 lr:7.547258226246912e-07 tokens_per_second_per_gpu:4826.997667704504 mfu:22.769771569288178 +Step 34300 | loss:0.22835683822631836 lr:7.539754381623688e-07 tokens_per_second_per_gpu:4104.895605344401 mfu:19.36349293781941 +Step 34301 | loss:0.3594316840171814 lr:7.532254240856851e-07 tokens_per_second_per_gpu:3919.7679489953666 mfu:18.490214196785793 +Step 34302 | loss:0.2789144814014435 lr:7.524757804002691e-07 tokens_per_second_per_gpu:5362.687202608539 mfu:25.296710503489837 +Step 34303 | loss:0.4216260015964508 lr:7.517265071117663e-07 tokens_per_second_per_gpu:5344.157577263932 mfu:25.209303099259234 +Step 34304 | loss:0.32213935256004333 lr:7.509776042258165e-07 tokens_per_second_per_gpu:5019.414272008261 mfu:23.677433521446982 +Step 34305 | loss:0.2644631862640381 lr:7.502290717480542e-07 tokens_per_second_per_gpu:5375.09660739235 mfu:25.355247783117854 +Step 34306 | loss:0.3136676847934723 lr:7.494809096840971e-07 tokens_per_second_per_gpu:4386.488055978728 mfu:20.691812572087812 +Step 34307 | loss:0.37887051701545715 lr:7.48733118039574e-07 tokens_per_second_per_gpu:5205.782304222751 mfu:24.5565633270679 +Step 34308 | loss:0.34813064336776733 lr:7.479856968201193e-07 tokens_per_second_per_gpu:4123.422931869345 mfu:19.45088949812544 +Step 34309 | loss:0.29815003275871277 lr:7.472386460313507e-07 tokens_per_second_per_gpu:5333.995416623103 mfu:25.161366453673054 +Step 34310 | loss:0.2781374156475067 lr:7.464919656788804e-07 tokens_per_second_per_gpu:5347.438958606517 mfu:25.2247819349136 +Step 34311 | loss:0.2768014967441559 lr:7.457456557683263e-07 tokens_per_second_per_gpu:5342.885422716142 mfu:25.203302129205365 +Step 34312 | loss:0.30749937891960144 lr:7.449997163053113e-07 tokens_per_second_per_gpu:5354.426838391108 mfu:25.25774495611275 +Step 34313 | loss:0.31566184759140015 lr:7.442541472954367e-07 tokens_per_second_per_gpu:4770.288232311323 mfu:22.502263486911147 +Step 34314 | loss:0.3195890188217163 lr:7.435089487443093e-07 tokens_per_second_per_gpu:3725.03451271575 mfu:17.571623352904563 +Step 34315 | loss:0.32262954115867615 lr:7.427641206575353e-07 tokens_per_second_per_gpu:4719.8634527674585 mfu:22.264401198448198 +Step 34316 | loss:0.23717933893203735 lr:7.420196630407217e-07 tokens_per_second_per_gpu:4737.069230195321 mfu:22.345563786183764 +Step 34317 | loss:0.3115859627723694 lr:7.412755758994583e-07 tokens_per_second_per_gpu:4608.289436457771 mfu:21.738087527025243 +Step 34318 | loss:0.31024715304374695 lr:7.405318592393517e-07 tokens_per_second_per_gpu:4924.197677982908 mfu:23.228280601803256 +Step 34319 | loss:0.30455702543258667 lr:7.397885130659921e-07 tokens_per_second_per_gpu:4465.195099862502 mfu:21.063087126893993 +Step 34320 | loss:0.2884010672569275 lr:7.390455373849636e-07 tokens_per_second_per_gpu:5361.916459716594 mfu:25.293074777766233 +Step 34321 | loss:0.3565343916416168 lr:7.38302932201862e-07 tokens_per_second_per_gpu:5339.1440773032655 mfu:25.185653564553476 +Step 34322 | loss:0.3101898729801178 lr:7.37560697522266e-07 tokens_per_second_per_gpu:3175.7750597734375 mfu:14.980672800049591 +Step 34323 | loss:0.26902610063552856 lr:7.368188333517656e-07 tokens_per_second_per_gpu:3755.999316456605 mfu:17.717689615290475 +Step 34324 | loss:0.3680598735809326 lr:7.360773396959342e-07 tokens_per_second_per_gpu:5301.00811057988 mfu:25.005759702852398 +Step 34325 | loss:0.31878769397735596 lr:7.353362165603561e-07 tokens_per_second_per_gpu:4844.816028630395 mfu:22.853823817901468 +Step 34326 | loss:0.31789734959602356 lr:7.345954639505992e-07 tokens_per_second_per_gpu:5365.774292982409 mfu:25.311272835495313 +Step 34327 | loss:0.22751405835151672 lr:7.338550818722367e-07 tokens_per_second_per_gpu:4631.691655306753 mfu:21.848479786166152 +Step 34328 | loss:0.2590799629688263 lr:7.331150703308309e-07 tokens_per_second_per_gpu:4530.677638269201 mfu:21.37197943303152 +Step 34329 | loss:0.29831698536872864 lr:7.323754293319552e-07 tokens_per_second_per_gpu:4325.311148109546 mfu:20.40323065981208 +Step 34330 | loss:0.3081797659397125 lr:7.316361588811715e-07 tokens_per_second_per_gpu:5127.1862975663735 mfu:24.185812553808248 +Step 34331 | loss:0.334963858127594 lr:7.308972589840424e-07 tokens_per_second_per_gpu:4532.431599759942 mfu:21.38025316864003 +Step 34332 | loss:0.40130865573883057 lr:7.301587296461133e-07 tokens_per_second_per_gpu:4692.60047951073 mfu:22.135797102053317 +Step 34333 | loss:0.24106572568416595 lr:7.294205708729574e-07 tokens_per_second_per_gpu:5314.959090215155 mfu:25.071568854074563 +Step 34334 | loss:0.30483314394950867 lr:7.286827826701148e-07 tokens_per_second_per_gpu:3906.5840153277722 mfu:18.428023332264715 +Step 34335 | loss:0.3506401479244232 lr:7.279453650431312e-07 tokens_per_second_per_gpu:4054.329640697887 mfu:19.1249646551386 +Step 34336 | loss:0.2978307604789734 lr:7.272083179975575e-07 tokens_per_second_per_gpu:5326.495245608119 mfu:25.125986867334245 +Step 34337 | loss:0.25886377692222595 lr:7.264716415389394e-07 tokens_per_second_per_gpu:4994.289830253704 mfu:23.558917243018943 +Step 34338 | loss:0.3312366306781769 lr:7.257353356728114e-07 tokens_per_second_per_gpu:4063.8426983339164 mfu:19.169839371102753 +Step 34339 | loss:0.42889225482940674 lr:7.249994004047189e-07 tokens_per_second_per_gpu:4566.73613381295 mfu:21.54207350872512 +Step 34340 | loss:0.27996140718460083 lr:7.24263835740191e-07 tokens_per_second_per_gpu:3617.9431701278763 mfu:17.066455218250404 +Step 34341 | loss:0.23049762845039368 lr:7.235286416847675e-07 tokens_per_second_per_gpu:3966.924078077732 mfu:18.712657703332574 +Step 34342 | loss:0.2744864523410797 lr:7.22793818243972e-07 tokens_per_second_per_gpu:3366.1407651500667 mfu:15.878660312049695 +Step 34343 | loss:0.3432217538356781 lr:7.220593654233332e-07 tokens_per_second_per_gpu:4720.299499268224 mfu:22.266458104189514 +Step 34344 | loss:0.38198813796043396 lr:7.21325283228369e-07 tokens_per_second_per_gpu:5298.278391792668 mfu:24.99288315359507 +Step 34345 | loss:0.3010335862636566 lr:7.205915716646084e-07 tokens_per_second_per_gpu:4925.889098577743 mfu:23.236259321335204 +Step 34346 | loss:0.32902196049690247 lr:7.198582307375634e-07 tokens_per_second_per_gpu:3296.2830581281455 mfu:15.549129589073429 +Step 34347 | loss:0.33678096532821655 lr:7.191252604527521e-07 tokens_per_second_per_gpu:5321.983939538023 mfu:25.10470626689342 +Step 34348 | loss:0.2701309323310852 lr:7.183926608156921e-07 tokens_per_second_per_gpu:5244.929945651938 mfu:24.741229430965745 +Step 34349 | loss:0.34938400983810425 lr:7.176604318318903e-07 tokens_per_second_per_gpu:5350.043409908102 mfu:25.23706757606839 +Step 34350 | loss:0.2894977033138275 lr:7.169285735068531e-07 tokens_per_second_per_gpu:4225.6911333724365 mfu:19.933306053369954 +Step 34351 | loss:0.2906643748283386 lr:7.161970858460765e-07 tokens_per_second_per_gpu:4669.88219452089 mfu:22.02863107561713 +Step 34352 | loss:0.351257860660553 lr:7.15465968855078e-07 tokens_per_second_per_gpu:4715.593034897269 mfu:22.24425690874609 +Step 34353 | loss:0.21704967319965363 lr:7.147352225393478e-07 tokens_per_second_per_gpu:3641.8583049739705 mfu:17.179266989662185 +Step 34354 | loss:0.30491653084754944 lr:7.140048469043814e-07 tokens_per_second_per_gpu:4093.7679829439053 mfu:19.31100204438812 +Step 34355 | loss:0.3610427677631378 lr:7.132748419556745e-07 tokens_per_second_per_gpu:4650.315937867044 mfu:21.936333704633313 +Step 34356 | loss:0.21596835553646088 lr:7.125452076987171e-07 tokens_per_second_per_gpu:4567.58787755578 mfu:21.546091329282515 +Step 34357 | loss:0.27143359184265137 lr:7.118159441389938e-07 tokens_per_second_per_gpu:5361.534093794596 mfu:25.291271092473814 +Step 34358 | loss:0.2090309113264084 lr:7.110870512819945e-07 tokens_per_second_per_gpu:4495.318909592407 mfu:21.205186277041886 +Step 34359 | loss:0.24580927193164825 lr:7.103585291331982e-07 tokens_per_second_per_gpu:4476.982374783477 mfu:21.11868971381272 +Step 34360 | loss:0.2799134850502014 lr:7.096303776980839e-07 tokens_per_second_per_gpu:5341.216975807957 mfu:25.19543178047306 +Step 34361 | loss:0.25666743516921997 lr:7.089025969821251e-07 tokens_per_second_per_gpu:5379.49094436829 mfu:25.375976620384563 +Step 34362 | loss:0.2996167838573456 lr:7.08175186990806e-07 tokens_per_second_per_gpu:4847.470746532638 mfu:22.866346575187933 +Step 34363 | loss:0.19797983765602112 lr:7.074481477295891e-07 tokens_per_second_per_gpu:5300.274488191062 mfu:25.002299080875456 +Step 34364 | loss:0.2303522378206253 lr:7.067214792039423e-07 tokens_per_second_per_gpu:5360.453997940841 mfu:25.286176096048298 +Step 34365 | loss:0.24852855503559113 lr:7.059951814193333e-07 tokens_per_second_per_gpu:4761.028224488253 mfu:22.458582450089473 +Step 34366 | loss:0.46869397163391113 lr:7.052692543812301e-07 tokens_per_second_per_gpu:5333.218879039143 mfu:25.15770339339854 +Step 34367 | loss:0.2888568341732025 lr:7.045436980950837e-07 tokens_per_second_per_gpu:4730.405568221108 mfu:22.314130155713432 +Step 34368 | loss:0.29371270537376404 lr:7.03818512566351e-07 tokens_per_second_per_gpu:4332.387125401049 mfu:20.436609251982322 +Step 34369 | loss:0.3326526880264282 lr:7.030936978004943e-07 tokens_per_second_per_gpu:3854.398474223413 mfu:18.181855231104013 +Step 34370 | loss:0.3294984698295593 lr:7.023692538029536e-07 tokens_per_second_per_gpu:4708.292715250387 mfu:22.20982005540878 +Step 34371 | loss:0.3829393684864044 lr:7.016451805791913e-07 tokens_per_second_per_gpu:5333.156165330665 mfu:25.157407562135592 +Step 34372 | loss:0.3836677670478821 lr:7.009214781346418e-07 tokens_per_second_per_gpu:4587.265344193689 mfu:21.638913296735534 +Step 34373 | loss:0.2793000042438507 lr:7.001981464747565e-07 tokens_per_second_per_gpu:5346.280702359458 mfu:25.219318242577327 +Step 34374 | loss:0.32464754581451416 lr:6.994751856049697e-07 tokens_per_second_per_gpu:4620.683042986333 mfu:21.796550283587784 +Step 34375 | loss:0.3572663366794586 lr:6.987525955307162e-07 tokens_per_second_per_gpu:4864.214493614252 mfu:22.945329686949734 +Step 34376 | loss:0.3868153989315033 lr:6.980303762574359e-07 tokens_per_second_per_gpu:4575.5409185972285 mfu:21.58360718080375 +Step 34377 | loss:0.3257651627063751 lr:6.973085277905689e-07 tokens_per_second_per_gpu:5323.9355869643805 mfu:25.113912520788933 +Step 34378 | loss:0.28458845615386963 lr:6.965870501355332e-07 tokens_per_second_per_gpu:4703.192348465259 mfu:22.185760755071026 +Step 34379 | loss:0.4398367404937744 lr:6.958659432977521e-07 tokens_per_second_per_gpu:4788.6413181477055 mfu:22.588838124160702 +Step 34380 | loss:0.255941241979599 lr:6.951452072826548e-07 tokens_per_second_per_gpu:4996.371939482641 mfu:23.568738907496307 +Step 34381 | loss:0.2671201229095459 lr:6.944248420956645e-07 tokens_per_second_per_gpu:4335.808333023753 mfu:20.452747672056827 +Step 34382 | loss:0.2472236305475235 lr:6.937048477421937e-07 tokens_per_second_per_gpu:4336.434047586111 mfu:20.45569927440538 +Step 34383 | loss:0.3438255488872528 lr:6.929852242276602e-07 tokens_per_second_per_gpu:5355.099915062688 mfu:25.260919973611312 +Step 34384 | loss:0.18330110609531403 lr:6.92265971557482e-07 tokens_per_second_per_gpu:5321.952724938036 mfu:25.10455902229952 +Step 34385 | loss:0.2808409333229065 lr:6.915470897370602e-07 tokens_per_second_per_gpu:4466.297665018488 mfu:21.068288114851796 +Step 34386 | loss:0.2983510494232178 lr:6.908285787718016e-07 tokens_per_second_per_gpu:5382.694211448923 mfu:25.39108697773763 +Step 34387 | loss:0.2841900587081909 lr:6.901104386671131e-07 tokens_per_second_per_gpu:5353.076186980743 mfu:25.251373702965164 +Step 34388 | loss:0.2999798655509949 lr:6.893926694284014e-07 tokens_per_second_per_gpu:4181.98748287856 mfu:19.727148477378663 +Step 34389 | loss:0.33029431104660034 lr:6.886752710610566e-07 tokens_per_second_per_gpu:4899.66595801797 mfu:23.112560293185616 +Step 34390 | loss:0.336099237203598 lr:6.879582435704745e-07 tokens_per_second_per_gpu:5332.969347397647 mfu:25.156526310071115 +Step 34391 | loss:0.29150861501693726 lr:6.872415869620564e-07 tokens_per_second_per_gpu:5346.837542287098 mfu:25.221944951522694 +Step 34392 | loss:0.36804330348968506 lr:6.865253012411865e-07 tokens_per_second_per_gpu:3907.819490871274 mfu:18.433851281197278 +Step 34393 | loss:0.23336979746818542 lr:6.858093864132442e-07 tokens_per_second_per_gpu:3950.659868419336 mfu:18.63593665141361 +Step 34394 | loss:0.4243764877319336 lr:6.85093842483625e-07 tokens_per_second_per_gpu:5345.815679037244 mfu:25.217124648225287 +Step 34395 | loss:0.38997918367385864 lr:6.843786694577137e-07 tokens_per_second_per_gpu:5360.2431355027575 mfu:25.285181421951016 +Step 34396 | loss:0.32833364605903625 lr:6.83663867340878e-07 tokens_per_second_per_gpu:4494.569656487831 mfu:21.201651922311836 +Step 34397 | loss:0.3007467985153198 lr:6.829494361384969e-07 tokens_per_second_per_gpu:5294.233794379191 mfu:24.97380409751622 +Step 34398 | loss:0.4298098683357239 lr:6.822353758559552e-07 tokens_per_second_per_gpu:4085.2541207598124 mfu:19.270840703851075 +Step 34399 | loss:0.3526764214038849 lr:6.815216864986096e-07 tokens_per_second_per_gpu:4564.185910877202 mfu:21.530043672024295 +Step 34400 | loss:0.38354530930519104 lr:6.80808368071828e-07 tokens_per_second_per_gpu:4717.92493845786 mfu:22.255256895705173 +Step 34401 | loss:0.3894733190536499 lr:6.80095420580984e-07 tokens_per_second_per_gpu:5249.474506740433 mfu:24.76266690100225 +Step 34402 | loss:0.303254634141922 lr:6.793828440314343e-07 tokens_per_second_per_gpu:5327.11872167608 mfu:25.1289279103228 +Step 34403 | loss:0.28163453936576843 lr:6.786706384285412e-07 tokens_per_second_per_gpu:4717.178326378587 mfu:22.25173499914209 +Step 34404 | loss:0.2829928696155548 lr:6.779588037776563e-07 tokens_per_second_per_gpu:5368.470869530255 mfu:25.323993047154904 +Step 34405 | loss:0.3520981967449188 lr:6.772473400841417e-07 tokens_per_second_per_gpu:5122.452346850614 mfu:24.163481700586733 +Step 34406 | loss:0.30469921231269836 lr:6.765362473533377e-07 tokens_per_second_per_gpu:3954.0195259662923 mfu:18.65178472927929 +Step 34407 | loss:0.26049935817718506 lr:6.758255255906009e-07 tokens_per_second_per_gpu:4523.180041352568 mfu:21.336611989153816 +Step 34408 | loss:0.3012157082557678 lr:6.751151748012774e-07 tokens_per_second_per_gpu:5323.541747095468 mfu:25.112054710931265 +Step 34409 | loss:0.3434408903121948 lr:6.744051949907015e-07 tokens_per_second_per_gpu:3534.255484033012 mfu:16.67168612434932 +Step 34410 | loss:0.3008343279361725 lr:6.736955861642191e-07 tokens_per_second_per_gpu:4124.674951639742 mfu:19.456795488999003 +Step 34411 | loss:0.29349765181541443 lr:6.729863483271704e-07 tokens_per_second_per_gpu:3441.2370858725462 mfu:16.232902469651155 +Step 34412 | loss:0.4037252366542816 lr:6.722774814848787e-07 tokens_per_second_per_gpu:5019.5136246606935 mfu:23.677902184859814 +Step 34413 | loss:0.3024583160877228 lr:6.715689856426898e-07 tokens_per_second_per_gpu:3849.4891526577553 mfu:18.158697123662172 +Step 34414 | loss:0.2378803938627243 lr:6.708608608059219e-07 tokens_per_second_per_gpu:5341.583376796193 mfu:25.197160156449023 +Step 34415 | loss:0.3023240566253662 lr:6.701531069799038e-07 tokens_per_second_per_gpu:5343.953162386231 mfu:25.20833883940435 +Step 34416 | loss:0.29194891452789307 lr:6.694457241699536e-07 tokens_per_second_per_gpu:4015.863378813716 mfu:18.943512734809932 +Step 34417 | loss:0.27342918515205383 lr:6.68738712381406e-07 tokens_per_second_per_gpu:4503.7522107691275 mfu:21.244967597562027 +Step 34418 | loss:0.2894558608531952 lr:6.68032071619562e-07 tokens_per_second_per_gpu:4180.511692229769 mfu:19.72018692109247 +Step 34419 | loss:0.3071720600128174 lr:6.673258018897455e-07 tokens_per_second_per_gpu:4833.018875380279 mfu:22.79817463321877 +Step 34420 | loss:0.2979574203491211 lr:6.666199031972687e-07 tokens_per_second_per_gpu:4291.346443689775 mfu:20.243013354087886 +Step 34421 | loss:0.4180510640144348 lr:6.659143755474385e-07 tokens_per_second_per_gpu:5346.580233344772 mfu:25.220731181340312 +Step 34422 | loss:0.3322273790836334 lr:6.652092189455616e-07 tokens_per_second_per_gpu:4486.327736263362 mfu:21.16277337839642 +Step 34423 | loss:0.39164602756500244 lr:6.645044333969342e-07 tokens_per_second_per_gpu:3918.4510964208275 mfu:18.484002378513477 +Step 34424 | loss:0.28405559062957764 lr:6.638000189068738e-07 tokens_per_second_per_gpu:4031.002462156674 mfu:19.014926373932635 +Step 34425 | loss:0.31059929728507996 lr:6.630959754806654e-07 tokens_per_second_per_gpu:5313.171754013831 mfu:25.063137684261914 +Step 34426 | loss:0.2627188265323639 lr:6.623923031236046e-07 tokens_per_second_per_gpu:4960.0838829712575 mfu:23.397561953551236 +Step 34427 | loss:0.3102016746997833 lr:6.616890018409927e-07 tokens_per_second_per_gpu:5356.702990929159 mfu:25.268481955986402 +Step 34428 | loss:0.3362055718898773 lr:6.609860716381089e-07 tokens_per_second_per_gpu:5345.055175983546 mfu:25.21353722556579 +Step 34429 | loss:0.26144939661026 lr:6.602835125202434e-07 tokens_per_second_per_gpu:3366.541617548444 mfu:15.88055120120531 +Step 34430 | loss:0.249013289809227 lr:6.595813244926808e-07 tokens_per_second_per_gpu:3586.7093718546544 mfu:16.91912006276012 +Step 34431 | loss:0.30447447299957275 lr:6.588795075607113e-07 tokens_per_second_per_gpu:5340.060278393989 mfu:25.189975441418447 +Step 34432 | loss:0.2899034321308136 lr:6.581780617295974e-07 tokens_per_second_per_gpu:5194.49851375165 mfu:24.503335762202536 +Step 34433 | loss:0.21199840307235718 lr:6.574769870046238e-07 tokens_per_second_per_gpu:4269.758437613594 mfu:20.141179046132898 +Step 34434 | loss:0.34549179673194885 lr:6.56776283391064e-07 tokens_per_second_per_gpu:4564.288177380951 mfu:21.530526080570944 +Step 34435 | loss:0.2755686342716217 lr:6.560759508941805e-07 tokens_per_second_per_gpu:5367.030791097302 mfu:25.317199951483946 +Step 34436 | loss:0.2703385651111603 lr:6.553759895192524e-07 tokens_per_second_per_gpu:4436.358038165779 mfu:20.927057786760486 +Step 34437 | loss:0.37171974778175354 lr:6.546763992715365e-07 tokens_per_second_per_gpu:4754.0489823827775 mfu:22.425660174296453 +Step 34438 | loss:0.2651800215244293 lr:6.539771801562955e-07 tokens_per_second_per_gpu:5336.632860518577 mfu:25.173807726522337 +Step 34439 | loss:0.3560698330402374 lr:6.532783321787861e-07 tokens_per_second_per_gpu:5340.804834369634 mfu:25.193487638989907 +Step 34440 | loss:0.3479657769203186 lr:6.525798553442708e-07 tokens_per_second_per_gpu:4073.9561415782105 mfu:19.217546208417435 +Step 34441 | loss:0.24396659433841705 lr:6.518817496579954e-07 tokens_per_second_per_gpu:4239.59517862172 mfu:19.998893807085565 +Step 34442 | loss:0.3081822693347931 lr:6.511840151252169e-07 tokens_per_second_per_gpu:5311.352141828733 mfu:25.054554263126956 +Step 34443 | loss:0.28140750527381897 lr:6.504866517511809e-07 tokens_per_second_per_gpu:4576.950932173936 mfu:21.590258455418205 +Step 34444 | loss:0.24872121214866638 lr:6.497896595411279e-07 tokens_per_second_per_gpu:4764.615996521416 mfu:22.47550658290693 +Step 34445 | loss:0.2916709780693054 lr:6.490930385003091e-07 tokens_per_second_per_gpu:4758.012660690824 mfu:22.444357521148763 +Step 34446 | loss:0.4317188858985901 lr:6.483967886339593e-07 tokens_per_second_per_gpu:3469.9672560583435 mfu:16.36842758429006 +Step 34447 | loss:0.37359747290611267 lr:6.477009099473074e-07 tokens_per_second_per_gpu:4191.611374277851 mfu:19.772546015114308 +Step 34448 | loss:0.25192728638648987 lr:6.470054024455996e-07 tokens_per_second_per_gpu:4551.873475980721 mfu:21.471963815899578 +Step 34449 | loss:0.32642051577568054 lr:6.463102661340592e-07 tokens_per_second_per_gpu:4566.637832278049 mfu:21.54160980361302 +Step 34450 | loss:0.33545413613319397 lr:6.45615501017921e-07 tokens_per_second_per_gpu:4408.445991632441 mfu:20.79539190097652 +Step 34451 | loss:0.340741366147995 lr:6.449211071024031e-07 tokens_per_second_per_gpu:4869.877059154281 mfu:22.972040974735357 +Step 34452 | loss:0.3322810232639313 lr:6.442270843927289e-07 tokens_per_second_per_gpu:3693.248945736339 mfu:17.421685410285736 +Step 34453 | loss:0.3218514919281006 lr:6.435334328941278e-07 tokens_per_second_per_gpu:4283.649801692948 mfu:20.20670698060632 +Step 34454 | loss:0.2341245412826538 lr:6.428401526118067e-07 tokens_per_second_per_gpu:4514.842192960798 mfu:21.29728093570621 +Step 34455 | loss:0.31512951850891113 lr:6.421472435509779e-07 tokens_per_second_per_gpu:4893.588168252832 mfu:23.083890321885594 +Step 34456 | loss:0.3262401819229126 lr:6.414547057168652e-07 tokens_per_second_per_gpu:5349.727986988052 mfu:25.235579672337817 +Step 34457 | loss:0.3236182630062103 lr:6.407625391146644e-07 tokens_per_second_per_gpu:5062.081925123261 mfu:23.87870412104273 +Step 34458 | loss:0.3749638497829437 lr:6.400707437495878e-07 tokens_per_second_per_gpu:5131.929773400588 mfu:24.20818833863867 +Step 34459 | loss:0.3146372437477112 lr:6.393793196268372e-07 tokens_per_second_per_gpu:3598.722215959136 mfu:16.97578669800351 +Step 34460 | loss:0.3055635094642639 lr:6.386882667516192e-07 tokens_per_second_per_gpu:5110.673046285068 mfu:24.10791673006314 +Step 34461 | loss:0.23935596644878387 lr:6.379975851291186e-07 tokens_per_second_per_gpu:4993.45858203672 mfu:23.55499610335386 +Step 34462 | loss:0.26031747460365295 lr:6.37307274764537e-07 tokens_per_second_per_gpu:5341.616148817407 mfu:25.197314747664574 +Step 34463 | loss:0.35339438915252686 lr:6.366173356630701e-07 tokens_per_second_per_gpu:4710.231441573401 mfu:22.218965358255257 +Step 34464 | loss:0.30460888147354126 lr:6.359277678298969e-07 tokens_per_second_per_gpu:4759.130372683278 mfu:22.4496299593185 +Step 34465 | loss:0.30631816387176514 lr:6.352385712702191e-07 tokens_per_second_per_gpu:3567.799460720865 mfu:16.829918785578503 +Step 34466 | loss:0.32512181997299194 lr:6.345497459891991e-07 tokens_per_second_per_gpu:4998.937579393791 mfu:23.580841468700505 +Step 34467 | loss:0.3354119062423706 lr:6.338612919920384e-07 tokens_per_second_per_gpu:5389.800160706013 mfu:25.424606952783286 +Step 34468 | loss:0.32857459783554077 lr:6.331732092839048e-07 tokens_per_second_per_gpu:4496.4658097045485 mfu:21.21059640055219 +Step 34469 | loss:0.3261137008666992 lr:6.324854978699724e-07 tokens_per_second_per_gpu:4530.154520463215 mfu:21.36951179708758 +Step 34470 | loss:0.3181755840778351 lr:6.3179815775542e-07 tokens_per_second_per_gpu:3779.2572097654656 mfu:17.827401066234156 +Step 34471 | loss:0.2922828793525696 lr:6.311111889454046e-07 tokens_per_second_per_gpu:4690.250336720436 mfu:22.124711077535796 +Step 34472 | loss:0.22243483364582062 lr:6.304245914451112e-07 tokens_per_second_per_gpu:4126.796233125121 mfu:19.466801935693475 +Step 34473 | loss:0.24199023842811584 lr:6.297383652596911e-07 tokens_per_second_per_gpu:4321.618999266688 mfu:20.38581416377477 +Step 34474 | loss:0.38794878125190735 lr:6.290525103943068e-07 tokens_per_second_per_gpu:3158.4688988899243 mfu:14.899036686426538 +Step 34475 | loss:0.28798484802246094 lr:6.283670268541153e-07 tokens_per_second_per_gpu:5345.730828497646 mfu:25.216724393753008 +Step 34476 | loss:0.3386366665363312 lr:6.276819146442792e-07 tokens_per_second_per_gpu:5326.81253483343 mfu:25.12748357474516 +Step 34477 | loss:0.37841832637786865 lr:6.269971737699498e-07 tokens_per_second_per_gpu:4282.675437548603 mfu:20.202110738693783 +Step 34478 | loss:0.2944895923137665 lr:6.263128042362676e-07 tokens_per_second_per_gpu:5344.6934037101055 mfu:25.21183068402726 +Step 34479 | loss:0.3132658302783966 lr:6.25628806048395e-07 tokens_per_second_per_gpu:5169.359997194576 mfu:24.38475309052962 +Step 34480 | loss:0.31504908204078674 lr:6.249451792114614e-07 tokens_per_second_per_gpu:4416.56820010783 mfu:20.83370574414659 +Step 34481 | loss:0.3497675061225891 lr:6.242619237306236e-07 tokens_per_second_per_gpu:3979.5576238302665 mfu:18.77225229415245 +Step 34482 | loss:0.2047605812549591 lr:6.235790396110052e-07 tokens_per_second_per_gpu:4550.745215155656 mfu:21.466641617086605 +Step 34483 | loss:0.3478696346282959 lr:6.22896526857747e-07 tokens_per_second_per_gpu:4951.836924522275 mfu:23.358659643471178 +Step 34484 | loss:0.2659448981285095 lr:6.222143854759887e-07 tokens_per_second_per_gpu:4729.752727166142 mfu:22.311050593070988 +Step 34485 | loss:0.3100646138191223 lr:6.2153261547086e-07 tokens_per_second_per_gpu:4300.9537769400795 mfu:20.288332784209654 +Step 34486 | loss:0.3422608971595764 lr:6.208512168474844e-07 tokens_per_second_per_gpu:4425.20016439187 mfu:20.874424192439385 +Step 34487 | loss:0.30371832847595215 lr:6.201701896109802e-07 tokens_per_second_per_gpu:5324.439014294548 mfu:25.116287273398893 +Step 34488 | loss:0.3349932134151459 lr:6.194895337664874e-07 tokens_per_second_per_gpu:4130.069871777934 mfu:19.48224424775914 +Step 34489 | loss:0.28247442841529846 lr:6.188092493191078e-07 tokens_per_second_per_gpu:4220.0809091543115 mfu:19.906841668530102 +Step 34490 | loss:0.3658211827278137 lr:6.181293362739648e-07 tokens_per_second_per_gpu:5289.792208897042 mfu:24.952852380984297 +Step 34491 | loss:0.3339645266532898 lr:6.174497946361713e-07 tokens_per_second_per_gpu:3521.0161035557794 mfu:16.609233707766975 +Step 34492 | loss:0.295936644077301 lr:6.167706244108451e-07 tokens_per_second_per_gpu:5083.838874911367 mfu:23.981335365311644 +Step 34493 | loss:0.38245704770088196 lr:6.160918256030823e-07 tokens_per_second_per_gpu:4523.604470587532 mfu:21.338614094270238 +Step 34494 | loss:0.3678344190120697 lr:6.15413398218001e-07 tokens_per_second_per_gpu:5018.089181917037 mfu:23.671182845402996 +Step 34495 | loss:0.33953404426574707 lr:6.147353422606916e-07 tokens_per_second_per_gpu:4920.084668859585 mfu:23.208878836016822 +Step 34496 | loss:0.22806386649608612 lr:6.140576577362612e-07 tokens_per_second_per_gpu:4562.032266923129 mfu:21.51988456604338 +Step 34497 | loss:0.351681113243103 lr:6.133803446497998e-07 tokens_per_second_per_gpu:5324.8634955443595 mfu:25.118289623878223 +Step 34498 | loss:0.2688012421131134 lr:6.127034030064094e-07 tokens_per_second_per_gpu:4191.458382537987 mfu:19.77182432697429 +Step 34499 | loss:0.35579603910446167 lr:6.120268328111855e-07 tokens_per_second_per_gpu:4419.270507166564 mfu:20.84645298760377 +Step 34500 | loss:0.3115902841091156 lr:6.113506340692021e-07 tokens_per_second_per_gpu:4219.60742453088 mfu:19.90460816077657 +Step 34501 | loss:0.31373095512390137 lr:6.10674806785555e-07 tokens_per_second_per_gpu:3877.5026398769082 mfu:18.290841522468693 +Step 34502 | loss:0.4131220281124115 lr:6.099993509653235e-07 tokens_per_second_per_gpu:4766.501786255518 mfu:22.4844021748316 +Step 34503 | loss:0.2925907075405121 lr:6.093242666135978e-07 tokens_per_second_per_gpu:3773.4739049550117 mfu:17.800120230709698 +Step 34504 | loss:0.23385795950889587 lr:6.086495537354354e-07 tokens_per_second_per_gpu:4503.92269495737 mfu:21.24577180056557 +Step 34505 | loss:0.32617735862731934 lr:6.079752123359261e-07 tokens_per_second_per_gpu:4165.773643788807 mfu:19.65066503202609 +Step 34506 | loss:0.270929753780365 lr:6.073012424201386e-07 tokens_per_second_per_gpu:4069.6178944912226 mfu:19.197081956726834 +Step 34507 | loss:0.3302467167377472 lr:6.066276439931351e-07 tokens_per_second_per_gpu:4355.155676788138 mfu:20.544012393590666 +Step 34508 | loss:0.2586335241794586 lr:6.059544170599952e-07 tokens_per_second_per_gpu:5368.660404256451 mfu:25.32488711479633 +Step 34509 | loss:0.3124571442604065 lr:6.052815616257701e-07 tokens_per_second_per_gpu:4970.911480282202 mfu:23.448637577445556 +Step 34510 | loss:0.44368013739585876 lr:6.046090776955338e-07 tokens_per_second_per_gpu:3774.986588209194 mfu:17.807255815710054 +Step 34511 | loss:0.22310473024845123 lr:6.039369652743265e-07 tokens_per_second_per_gpu:5094.011212502208 mfu:24.029320017307402 +Step 34512 | loss:0.23704488575458527 lr:6.03265224367211e-07 tokens_per_second_per_gpu:5367.850973158098 mfu:25.32106889020221 +Step 34513 | loss:0.293122261762619 lr:6.025938549792498e-07 tokens_per_second_per_gpu:3747.270822991122 mfu:17.67651582237924 +Step 34514 | loss:0.25575095415115356 lr:6.019228571154778e-07 tokens_per_second_per_gpu:4212.231617049409 mfu:19.869815218443946 +Step 34515 | loss:0.30419889092445374 lr:6.012522307809465e-07 tokens_per_second_per_gpu:5191.003315743037 mfu:24.48684831685364 +Step 34516 | loss:0.313968688249588 lr:6.005819759807019e-07 tokens_per_second_per_gpu:5085.7136443928175 mfu:23.990178972824303 +Step 34517 | loss:0.1465531289577484 lr:5.999120927197843e-07 tokens_per_second_per_gpu:3988.5927182541654 mfu:18.81487237609605 +Step 34518 | loss:0.3275538682937622 lr:5.992425810032287e-07 tokens_per_second_per_gpu:3740.887199744552 mfu:17.646403182366736 +Step 34519 | loss:0.3204766809940338 lr:5.985734408360755e-07 tokens_per_second_per_gpu:5388.265719268796 mfu:25.417368730721225 +Step 34520 | loss:0.2659410834312439 lr:5.979046722233539e-07 tokens_per_second_per_gpu:3038.1297923816073 mfu:14.331376589058047 +Step 34521 | loss:0.256072461605072 lr:5.972362751700933e-07 tokens_per_second_per_gpu:5105.03657279772 mfu:24.08132852294177 +Step 34522 | loss:0.23335392773151398 lr:5.965682496813285e-07 tokens_per_second_per_gpu:5162.382411915371 mfu:24.351838630268805 +Step 34523 | loss:0.3002835214138031 lr:5.959005957620723e-07 tokens_per_second_per_gpu:3719.8511403765815 mfu:17.54717250120619 +Step 34524 | loss:0.24198417365550995 lr:5.952333134173537e-07 tokens_per_second_per_gpu:4822.296156329873 mfu:22.74759373797349 +Step 34525 | loss:0.27869704365730286 lr:5.945664026521913e-07 tokens_per_second_per_gpu:4047.353232019697 mfu:19.092055745105515 +Step 34526 | loss:0.32387056946754456 lr:5.93899863471592e-07 tokens_per_second_per_gpu:4342.105968343577 mfu:20.48245469234782 +Step 34527 | loss:0.31331631541252136 lr:5.932336958805851e-07 tokens_per_second_per_gpu:4180.232702714541 mfu:19.718870879948618 +Step 34528 | loss:0.2643948197364807 lr:5.925678998841666e-07 tokens_per_second_per_gpu:4698.654039619657 mfu:22.164352777930986 +Step 34529 | loss:0.2849707305431366 lr:5.919024754873493e-07 tokens_per_second_per_gpu:5259.0648659063545 mfu:24.807906261471906 +Step 34530 | loss:0.2922368049621582 lr:5.9123742269514e-07 tokens_per_second_per_gpu:5320.230907427153 mfu:25.096436915328322 +Step 34531 | loss:0.37246644496917725 lr:5.905727415125406e-07 tokens_per_second_per_gpu:5178.3886143554755 mfu:24.427342618118516 +Step 34532 | loss:0.3493225574493408 lr:5.899084319445469e-07 tokens_per_second_per_gpu:4224.693341648502 mfu:19.92859929956733 +Step 34533 | loss:0.3032825291156769 lr:5.89244493996155e-07 tokens_per_second_per_gpu:4729.085953370444 mfu:22.307905307313337 +Step 34534 | loss:0.35382843017578125 lr:5.885809276723608e-07 tokens_per_second_per_gpu:4125.539690184203 mfu:19.460874608252833 +Step 34535 | loss:0.2503327429294586 lr:5.879177329781549e-07 tokens_per_second_per_gpu:5264.195466286789 mfu:24.832108178837107 +Step 34536 | loss:0.26180869340896606 lr:5.872549099185221e-07 tokens_per_second_per_gpu:5227.313985696642 mfu:24.658131942263918 +Step 34537 | loss:0.27245479822158813 lr:5.865924584984528e-07 tokens_per_second_per_gpu:5377.615769860942 mfu:25.36713110229564 +Step 34538 | loss:0.26721158623695374 lr:5.859303787229265e-07 tokens_per_second_per_gpu:5364.305123389647 mfu:25.304342511860067 +Step 34539 | loss:0.32186800241470337 lr:5.852686705969223e-07 tokens_per_second_per_gpu:4346.404124980837 mfu:20.502729830546915 +Step 34540 | loss:0.30646923184394836 lr:5.846073341254199e-07 tokens_per_second_per_gpu:5144.88399447749 mfu:24.269295609676444 +Step 34541 | loss:0.2841954827308655 lr:5.839463693133873e-07 tokens_per_second_per_gpu:3883.479677560547 mfu:18.319036229010205 +Step 34542 | loss:0.20692697167396545 lr:5.832857761658039e-07 tokens_per_second_per_gpu:4884.960962945539 mfu:23.04319432249809 +Step 34543 | loss:0.27858245372772217 lr:5.826255546876325e-07 tokens_per_second_per_gpu:4624.924392629479 mfu:21.81655745350338 +Step 34544 | loss:0.39436087012290955 lr:5.819657048838412e-07 tokens_per_second_per_gpu:5358.843683063729 mfu:25.278579966024914 +Step 34545 | loss:0.27816110849380493 lr:5.813062267593871e-07 tokens_per_second_per_gpu:5022.7651035962745 mfu:23.693239965758462 +Step 34546 | loss:0.2394823133945465 lr:5.806471203192387e-07 tokens_per_second_per_gpu:3967.054372537102 mfu:18.713272324527953 +Step 34547 | loss:0.4063766896724701 lr:5.799883855683474e-07 tokens_per_second_per_gpu:4481.465977676231 mfu:21.139839633639372 +Step 34548 | loss:0.1928078830242157 lr:5.793300225116704e-07 tokens_per_second_per_gpu:4464.0322786919405 mfu:21.057601901034708 +Step 34549 | loss:0.33001115918159485 lr:5.786720311541594e-07 tokens_per_second_per_gpu:5032.306929493354 mfu:23.73825038651836 +Step 34550 | loss:0.3156372010707855 lr:5.780144115007601e-07 tokens_per_second_per_gpu:5353.834469282835 mfu:25.25495064996043 +Step 34551 | loss:0.3833194375038147 lr:5.77357163556419e-07 tokens_per_second_per_gpu:3366.0097025169634 mfu:15.878042067247755 +Step 34552 | loss:0.387283056974411 lr:5.767002873260819e-07 tokens_per_second_per_gpu:5105.391909019047 mfu:24.08300470452438 +Step 34553 | loss:0.2415357381105423 lr:5.760437828146947e-07 tokens_per_second_per_gpu:3624.77206876193 mfu:17.098668298237488 +Step 34554 | loss:0.4191231429576874 lr:5.753876500271815e-07 tokens_per_second_per_gpu:3935.2451281827734 mfu:18.56322269169186 +Step 34555 | loss:0.3185054361820221 lr:5.747318889684883e-07 tokens_per_second_per_gpu:3696.0696363754046 mfu:17.434991089289657 +Step 34556 | loss:0.24803364276885986 lr:5.740764996435444e-07 tokens_per_second_per_gpu:3019.1141747482247 mfu:14.241676676282461 +Step 34557 | loss:0.2973264753818512 lr:5.734214820572736e-07 tokens_per_second_per_gpu:4911.845515245345 mfu:23.17001334267389 +Step 34558 | loss:0.29685500264167786 lr:5.727668362146055e-07 tokens_per_second_per_gpu:3513.8036933236017 mfu:16.575211538137715 +Step 34559 | loss:0.38183078169822693 lr:5.721125621204693e-07 tokens_per_second_per_gpu:4741.775066093942 mfu:22.367762017015956 +Step 34560 | loss:0.3344310522079468 lr:5.714586597797833e-07 tokens_per_second_per_gpu:4093.486707337605 mfu:19.309675219362628 +Step 34561 | loss:0.22884248197078705 lr:5.708051291974603e-07 tokens_per_second_per_gpu:4194.174885431134 mfu:19.78463853460445 +Step 34562 | loss:0.26402851939201355 lr:5.701519703784186e-07 tokens_per_second_per_gpu:4522.240036180926 mfu:21.33217782437776 +Step 34563 | loss:0.33414968848228455 lr:5.694991833275709e-07 tokens_per_second_per_gpu:5382.453031817296 mfu:25.389949292265754 +Step 34564 | loss:0.2689930200576782 lr:5.6884676804983e-07 tokens_per_second_per_gpu:3700.3847518939033 mfu:17.455346225425043 +Step 34565 | loss:0.3044061064720154 lr:5.681947245500973e-07 tokens_per_second_per_gpu:5365.4688863549945 mfu:25.309832180326055 +Step 34566 | loss:0.2596246898174286 lr:5.675430528332804e-07 tokens_per_second_per_gpu:4474.42478542486 mfu:21.106625128438377 +Step 34567 | loss:0.3272334933280945 lr:5.668917529042806e-07 tokens_per_second_per_gpu:3630.688342380722 mfu:17.126576370317697 +Step 34568 | loss:0.3421454131603241 lr:5.662408247679995e-07 tokens_per_second_per_gpu:4274.67296368432 mfu:20.164361704113123 +Step 34569 | loss:0.2954867482185364 lr:5.655902684293223e-07 tokens_per_second_per_gpu:5381.168887461552 mfu:25.38389176424209 +Step 34570 | loss:0.31835049390792847 lr:5.649400838931506e-07 tokens_per_second_per_gpu:5019.443777188541 mfu:23.677572702416466 +Step 34571 | loss:0.4595126211643219 lr:5.642902711643749e-07 tokens_per_second_per_gpu:3754.154215104806 mfu:17.70898595740729 +Step 34572 | loss:0.35933077335357666 lr:5.636408302478746e-07 tokens_per_second_per_gpu:4061.5351266362004 mfu:19.158954161692296 +Step 34573 | loss:0.3524056077003479 lr:5.629917611485458e-07 tokens_per_second_per_gpu:3532.0967954873354 mfu:16.661503222169117 +Step 34574 | loss:0.42409539222717285 lr:5.623430638712624e-07 tokens_per_second_per_gpu:5364.8342895205105 mfu:25.30683867878441 +Step 34575 | loss:0.5088152885437012 lr:5.616947384209037e-07 tokens_per_second_per_gpu:3739.8488069139557 mfu:17.641504906216728 +Step 34576 | loss:0.3078766167163849 lr:5.610467848023493e-07 tokens_per_second_per_gpu:5345.75051312486 mfu:25.21681724949812 +Step 34577 | loss:0.3606199026107788 lr:5.603992030204674e-07 tokens_per_second_per_gpu:4716.466165427481 mfu:22.248375614428532 +Step 34578 | loss:0.37537577748298645 lr:5.597519930801376e-07 tokens_per_second_per_gpu:4227.515789056676 mfu:19.941913265550877 +Step 34579 | loss:0.27557533979415894 lr:5.591051549862114e-07 tokens_per_second_per_gpu:5373.576119154137 mfu:25.348075380676157 +Step 34580 | loss:0.37834617495536804 lr:5.584586887435739e-07 tokens_per_second_per_gpu:4327.2522004247385 mfu:20.412386934760523 +Step 34581 | loss:0.2806648313999176 lr:5.578125943570711e-07 tokens_per_second_per_gpu:5358.176708629237 mfu:25.27543373381984 +Step 34582 | loss:0.2766300141811371 lr:5.57166871831577e-07 tokens_per_second_per_gpu:5339.077195940533 mfu:25.18533807375433 +Step 34583 | loss:0.3010823428630829 lr:5.565215211719321e-07 tokens_per_second_per_gpu:4980.086555253002 mfu:23.49191797151203 +Step 34584 | loss:0.36877700686454773 lr:5.558765423829993e-07 tokens_per_second_per_gpu:4346.059143379236 mfu:20.501102493472235 +Step 34585 | loss:0.3601246178150177 lr:5.5523193546963e-07 tokens_per_second_per_gpu:4548.19696645076 mfu:21.454621093169546 +Step 34586 | loss:0.30906665325164795 lr:5.545877004366651e-07 tokens_per_second_per_gpu:4464.750484066831 mfu:21.06098979832654 +Step 34587 | loss:0.2567351758480072 lr:5.539438372889616e-07 tokens_per_second_per_gpu:3821.1815232656236 mfu:18.025165206039883 +Step 34588 | loss:0.4602711498737335 lr:5.533003460313546e-07 tokens_per_second_per_gpu:3856.265489505582 mfu:18.190662260735657 +Step 34589 | loss:0.29652807116508484 lr:5.526572266686847e-07 tokens_per_second_per_gpu:3312.7643776219265 mfu:15.626874785128665 +Step 34590 | loss:0.32920241355895996 lr:5.520144792057868e-07 tokens_per_second_per_gpu:4870.048850528242 mfu:22.97285134395696 +Step 34591 | loss:0.3357459604740143 lr:5.513721036474962e-07 tokens_per_second_per_gpu:4182.547108062705 mfu:19.729788324854177 +Step 34592 | loss:0.23528707027435303 lr:5.507300999986531e-07 tokens_per_second_per_gpu:4063.699590737312 mfu:19.169164308152897 +Step 34593 | loss:0.20560818910598755 lr:5.500884682640761e-07 tokens_per_second_per_gpu:3972.797843682962 mfu:18.740365257860198 +Step 34594 | loss:0.3524007499217987 lr:5.494472084485891e-07 tokens_per_second_per_gpu:5312.948007134927 mfu:25.062082232058973 +Step 34595 | loss:0.3844163119792938 lr:5.488063205570216e-07 tokens_per_second_per_gpu:4980.80053045688 mfu:23.495285914365457 +Step 34596 | loss:0.30832499265670776 lr:5.481658045941973e-07 tokens_per_second_per_gpu:5303.860759416215 mfu:25.019216134125468 +Step 34597 | loss:0.22381378710269928 lr:5.475256605649237e-07 tokens_per_second_per_gpu:4877.759429423085 mfu:23.009223460164353 +Step 34598 | loss:0.30848395824432373 lr:5.468858884740191e-07 tokens_per_second_per_gpu:4952.491283678601 mfu:23.361746367256753 +Step 34599 | loss:0.30162927508354187 lr:5.462464883262908e-07 tokens_per_second_per_gpu:5338.159953670917 mfu:25.181011285470884 +Step 34600 | loss:0.33003759384155273 lr:5.456074601265626e-07 tokens_per_second_per_gpu:5130.982982304319 mfu:24.203722163498348 +Step 34601 | loss:0.32970571517944336 lr:5.449688038796253e-07 tokens_per_second_per_gpu:4357.92899355958 mfu:20.55709460197819 +Step 34602 | loss:0.3828750252723694 lr:5.443305195902914e-07 tokens_per_second_per_gpu:3488.4951069272342 mfu:16.455826618015944 +Step 34603 | loss:0.4108133316040039 lr:5.436926072633574e-07 tokens_per_second_per_gpu:4685.995241532502 mfu:22.10463906754021 +Step 34604 | loss:0.3139350116252899 lr:5.430550669036249e-07 tokens_per_second_per_gpu:5321.1584616993505 mfu:25.100812347087356 +Step 34605 | loss:0.3767644762992859 lr:5.42417898515879e-07 tokens_per_second_per_gpu:5268.631354713971 mfu:24.853033021388732 +Step 34606 | loss:0.4040754437446594 lr:5.417811021049212e-07 tokens_per_second_per_gpu:4460.943463226748 mfu:21.04303143147913 +Step 34607 | loss:0.2439783215522766 lr:5.411446776755424e-07 tokens_per_second_per_gpu:4932.008324685285 mfu:23.26512475493211 +Step 34608 | loss:0.32880645990371704 lr:5.40508625232522e-07 tokens_per_second_per_gpu:4714.679370354507 mfu:22.23994699721064 +Step 34609 | loss:0.2725640535354614 lr:5.398729447806506e-07 tokens_per_second_per_gpu:3378.3876383972183 mfu:15.93643090268855 +Step 34610 | loss:0.4056176245212555 lr:5.39237636324702e-07 tokens_per_second_per_gpu:3930.431520583285 mfu:18.540516083358927 +Step 34611 | loss:0.3425634503364563 lr:5.386026998694615e-07 tokens_per_second_per_gpu:5119.724249376718 mfu:24.150612799340937 +Step 34612 | loss:0.3064272999763489 lr:5.379681354196975e-07 tokens_per_second_per_gpu:3721.5529102815653 mfu:17.555200040199885 +Step 34613 | loss:0.37822407484054565 lr:5.373339429801894e-07 tokens_per_second_per_gpu:5386.878343234215 mfu:25.410824241255348 +Step 34614 | loss:0.2949253022670746 lr:5.367001225557056e-07 tokens_per_second_per_gpu:4934.873076763275 mfu:23.278638279261095 +Step 34615 | loss:0.2723928689956665 lr:5.36066674151009e-07 tokens_per_second_per_gpu:4596.979457810247 mfu:21.68473642806353 +Step 34616 | loss:0.39591410756111145 lr:5.35433597770868e-07 tokens_per_second_per_gpu:5362.376117475686 mfu:25.295243061841575 +Step 34617 | loss:0.40066227316856384 lr:5.348008934200399e-07 tokens_per_second_per_gpu:5212.178057283317 mfu:24.586733185482583 +Step 34618 | loss:0.3837575912475586 lr:5.341685611032931e-07 tokens_per_second_per_gpu:3724.1917688665458 mfu:17.56764798638094 +Step 34619 | loss:0.26704925298690796 lr:5.335366008253683e-07 tokens_per_second_per_gpu:5329.060411266028 mfu:25.138087191404992 +Step 34620 | loss:0.45504531264305115 lr:5.329050125910284e-07 tokens_per_second_per_gpu:4731.933733419212 mfu:22.32133877168476 +Step 34621 | loss:0.31464070081710815 lr:5.322737964050251e-07 tokens_per_second_per_gpu:4973.217314175273 mfu:23.459514589335 +Step 34622 | loss:0.331158310174942 lr:5.316429522720989e-07 tokens_per_second_per_gpu:4502.219262091192 mfu:21.23773641710019 +Step 34623 | loss:0.39329612255096436 lr:5.310124801970018e-07 tokens_per_second_per_gpu:4213.633335349399 mfu:19.876427362822806 +Step 34624 | loss:0.2619125247001648 lr:5.303823801844743e-07 tokens_per_second_per_gpu:5335.998863990231 mfu:25.170817056727167 +Step 34625 | loss:0.24674001336097717 lr:5.297526522392515e-07 tokens_per_second_per_gpu:3946.9095462907694 mfu:18.618245742062225 +Step 34626 | loss:0.3672998547554016 lr:5.291232963660686e-07 tokens_per_second_per_gpu:4169.229559051022 mfu:19.666967173957904 +Step 34627 | loss:0.3616483807563782 lr:5.284943125696607e-07 tokens_per_second_per_gpu:4906.372442082408 mfu:23.144195922761742 +Step 34628 | loss:0.3166376054286957 lr:5.278657008547627e-07 tokens_per_second_per_gpu:4525.307434827891 mfu:21.346647267147667 +Step 34629 | loss:0.37656524777412415 lr:5.272374612260989e-07 tokens_per_second_per_gpu:5356.965364666462 mfu:25.26971961767084 +Step 34630 | loss:0.24212093651294708 lr:5.26609593688393e-07 tokens_per_second_per_gpu:4510.675311985933 mfu:21.277625047204566 +Step 34631 | loss:0.44510263204574585 lr:5.259820982463748e-07 tokens_per_second_per_gpu:4202.661349496674 mfu:19.82467063354168 +Step 34632 | loss:0.3558259606361389 lr:5.253549749047515e-07 tokens_per_second_per_gpu:4630.878091046291 mfu:21.844642064741173 +Step 34633 | loss:0.28434833884239197 lr:5.247282236682527e-07 tokens_per_second_per_gpu:5116.723541946757 mfu:24.136457950420333 +Step 34634 | loss:0.325845867395401 lr:5.241018445415802e-07 tokens_per_second_per_gpu:4426.7625119439845 mfu:20.881794052406686 +Step 34635 | loss:0.25762900710105896 lr:5.234758375294524e-07 tokens_per_second_per_gpu:5069.8962229413955 mfu:23.915565497107565 +Step 34636 | loss:0.387401282787323 lr:5.228502026365767e-07 tokens_per_second_per_gpu:4348.4842958191475 mfu:20.51254235130497 +Step 34637 | loss:0.24076442420482635 lr:5.222249398676604e-07 tokens_per_second_per_gpu:5391.6757946369 mfu:25.433454638051426 +Step 34638 | loss:0.29160621762275696 lr:5.216000492273998e-07 tokens_per_second_per_gpu:4509.510870927433 mfu:21.27213218005749 +Step 34639 | loss:0.33822688460350037 lr:5.209755307205022e-07 tokens_per_second_per_gpu:5143.3891372907565 mfu:24.262244113277685 +Step 34640 | loss:0.3093230724334717 lr:5.203513843516583e-07 tokens_per_second_per_gpu:5367.701894975956 mfu:25.320365662981686 +Step 34641 | loss:0.3882802724838257 lr:5.197276101255643e-07 tokens_per_second_per_gpu:5390.218376176083 mfu:25.426579746510523 +Step 34642 | loss:0.32684972882270813 lr:5.191042080469166e-07 tokens_per_second_per_gpu:4240.917286157223 mfu:20.00513041390534 +Step 34643 | loss:0.21950112283229828 lr:5.184811781204002e-07 tokens_per_second_per_gpu:4310.758731313891 mfu:20.334584426888643 +Step 34644 | loss:0.2971699833869934 lr:5.178585203506948e-07 tokens_per_second_per_gpu:3608.982640789085 mfu:17.024186872535374 +Step 34645 | loss:0.33428430557250977 lr:5.172362347424964e-07 tokens_per_second_per_gpu:3210.51519363976 mfu:15.144547938775077 +Step 34646 | loss:0.2559291124343872 lr:5.166143213004737e-07 tokens_per_second_per_gpu:5168.722184526627 mfu:24.38174441935284 +Step 34647 | loss:0.3355163037776947 lr:5.159927800293119e-07 tokens_per_second_per_gpu:3678.1885436425437 mfu:17.35064292403974 +Step 34648 | loss:0.31848859786987305 lr:5.153716109336792e-07 tokens_per_second_per_gpu:5341.241860344011 mfu:25.19554916507537 +Step 34649 | loss:0.22997012734413147 lr:5.147508140182556e-07 tokens_per_second_per_gpu:4880.683337562664 mfu:23.02301603372892 +Step 34650 | loss:0.24727003276348114 lr:5.141303892877036e-07 tokens_per_second_per_gpu:5346.961465469902 mfu:25.222529518319455 +Step 34651 | loss:0.32404714822769165 lr:5.135103367466865e-07 tokens_per_second_per_gpu:4429.288092193928 mfu:20.89370764535409 +Step 34652 | loss:0.28224390745162964 lr:5.128906563998781e-07 tokens_per_second_per_gpu:3570.711351531107 mfu:16.843654671350215 +Step 34653 | loss:0.2521461546421051 lr:5.122713482519359e-07 tokens_per_second_per_gpu:4307.157636761221 mfu:20.31759744947361 +Step 34654 | loss:0.25736552476882935 lr:5.116524123075117e-07 tokens_per_second_per_gpu:4165.021701093926 mfu:19.647117990039686 +Step 34655 | loss:0.36217084527015686 lr:5.110338485712629e-07 tokens_per_second_per_gpu:3823.3316421613413 mfu:18.035307683719093 +Step 34656 | loss:0.33496972918510437 lr:5.104156570478413e-07 tokens_per_second_per_gpu:5358.347819598052 mfu:25.276240893454286 +Step 34657 | loss:0.22293385863304138 lr:5.097978377419044e-07 tokens_per_second_per_gpu:5330.2822178930255 mfu:25.143850661726248 +Step 34658 | loss:0.27054563164711 lr:5.091803906580928e-07 tokens_per_second_per_gpu:4628.858615497246 mfu:21.835115854018273 +Step 34659 | loss:0.22899822890758514 lr:5.085633158010472e-07 tokens_per_second_per_gpu:4419.734258974532 mfu:20.848640584006173 +Step 34660 | loss:0.35302531719207764 lr:5.07946613175414e-07 tokens_per_second_per_gpu:4542.043520363907 mfu:21.425594238091698 +Step 34661 | loss:0.22797270119190216 lr:5.073302827858339e-07 tokens_per_second_per_gpu:3462.2640150363027 mfu:16.332090082080647 +Step 34662 | loss:0.26264822483062744 lr:5.06714324636931e-07 tokens_per_second_per_gpu:5289.061447548146 mfu:24.949405255020814 +Step 34663 | loss:0.27505558729171753 lr:5.06098738733346e-07 tokens_per_second_per_gpu:4045.050630029727 mfu:19.08119397865442 +Step 34664 | loss:0.32450905442237854 lr:5.054835250797086e-07 tokens_per_second_per_gpu:4768.208809234036 mfu:22.492454493469424 +Step 34665 | loss:0.28821977972984314 lr:5.048686836806427e-07 tokens_per_second_per_gpu:3022.412877494661 mfu:14.257237219954126 +Step 34666 | loss:0.24251410365104675 lr:5.04254214540778e-07 tokens_per_second_per_gpu:4113.792104560253 mfu:19.405459242519875 +Step 34667 | loss:0.27538740634918213 lr:5.036401176647332e-07 tokens_per_second_per_gpu:5355.195199604597 mfu:25.26136944705272 +Step 34668 | loss:0.359988272190094 lr:5.030263930571322e-07 tokens_per_second_per_gpu:5342.179768537278 mfu:25.199973438046644 +Step 34669 | loss:0.23516890406608582 lr:5.024130407225769e-07 tokens_per_second_per_gpu:5326.801673397533 mfu:25.127432339498885 +Step 34670 | loss:0.286783367395401 lr:5.018000606656914e-07 tokens_per_second_per_gpu:3954.4244270725235 mfu:18.65369471688063 +Step 34671 | loss:0.22707590460777283 lr:5.011874528910832e-07 tokens_per_second_per_gpu:4725.114196347866 mfu:22.289169851786212 +Step 34672 | loss:0.2510873079299927 lr:5.005752174033651e-07 tokens_per_second_per_gpu:3905.3400663964426 mfu:18.422155412916062 +Step 34673 | loss:0.34220343828201294 lr:4.999633542071281e-07 tokens_per_second_per_gpu:4783.5716024511175 mfu:22.564923410237206 +Step 34674 | loss:0.3579215705394745 lr:4.993518633069905e-07 tokens_per_second_per_gpu:3947.2671543942233 mfu:18.619932640500448 +Step 34675 | loss:0.3118791878223419 lr:4.987407447075432e-07 tokens_per_second_per_gpu:3790.384561746091 mfu:17.879890684048785 +Step 34676 | loss:0.2931003272533417 lr:4.981299984133825e-07 tokens_per_second_per_gpu:5334.262232805007 mfu:25.162625071126634 +Step 34677 | loss:0.3436979353427887 lr:4.975196244290992e-07 tokens_per_second_per_gpu:4359.197782173469 mfu:20.563079694347564 +Step 34678 | loss:0.2776033282279968 lr:4.969096227592896e-07 tokens_per_second_per_gpu:5371.205883289001 mfu:25.336894573696373 +Step 34679 | loss:0.30515992641448975 lr:4.962999934085388e-07 tokens_per_second_per_gpu:3886.153320804935 mfu:18.33164826036456 +Step 34680 | loss:0.24814234673976898 lr:4.956907363814323e-07 tokens_per_second_per_gpu:5327.655469925672 mfu:25.131459843394822 +Step 34681 | loss:0.3023758828639984 lr:4.950818516825551e-07 tokens_per_second_per_gpu:5357.2440863124175 mfu:25.271034395229627 +Step 34682 | loss:0.19813686609268188 lr:4.944733393164813e-07 tokens_per_second_per_gpu:4280.415493697258 mfu:20.191450197960805 +Step 34683 | loss:0.36546874046325684 lr:4.938651992877963e-07 tokens_per_second_per_gpu:4010.228610655704 mfu:18.916932572018005 +Step 34684 | loss:0.3487206697463989 lr:4.93257431601063e-07 tokens_per_second_per_gpu:4537.346897780849 mfu:21.403439468040972 +Step 34685 | loss:0.3551652133464813 lr:4.926500362608555e-07 tokens_per_second_per_gpu:5317.433667641465 mfu:25.083241858000914 +Step 34686 | loss:0.3724272847175598 lr:4.920430132717479e-07 tokens_per_second_per_gpu:4868.918727229136 mfu:22.967520359536163 +Step 34687 | loss:0.31343597173690796 lr:4.914363626382978e-07 tokens_per_second_per_gpu:4704.278592475971 mfu:22.190884753401892 +Step 34688 | loss:0.40967389941215515 lr:4.908300843650792e-07 tokens_per_second_per_gpu:5388.955616033826 mfu:25.4206230914704 +Step 34689 | loss:0.2543330788612366 lr:4.902241784566386e-07 tokens_per_second_per_gpu:5015.165799532609 mfu:23.65739275191518 +Step 34690 | loss:0.3359939754009247 lr:4.896186449175444e-07 tokens_per_second_per_gpu:3928.8070957284735 mfu:18.532853394163322 +Step 34691 | loss:0.27415162324905396 lr:4.89013483752343e-07 tokens_per_second_per_gpu:5376.568654938738 mfu:25.362191682551376 +Step 34692 | loss:0.331956148147583 lr:4.884086949655864e-07 tokens_per_second_per_gpu:5324.49869814705 mfu:25.11656881231426 +Step 34693 | loss:0.27795112133026123 lr:4.878042785618319e-07 tokens_per_second_per_gpu:4496.69663426309 mfu:21.211685239377566 +Step 34694 | loss:0.3306886851787567 lr:4.872002345456151e-07 tokens_per_second_per_gpu:4213.337962698394 mfu:19.875034039632247 +Step 34695 | loss:0.3083544671535492 lr:4.865965629214819e-07 tokens_per_second_per_gpu:5403.39561528322 mfu:25.488739031647714 +Step 34696 | loss:0.3238162398338318 lr:4.85993263693979e-07 tokens_per_second_per_gpu:5314.105646570231 mfu:25.06754301478896 +Step 34697 | loss:0.3483095169067383 lr:4.85390336867636e-07 tokens_per_second_per_gpu:4000.151431019377 mfu:18.869396796329315 +Step 34698 | loss:0.20481397211551666 lr:4.847877824469937e-07 tokens_per_second_per_gpu:5298.1738608892265 mfu:24.992390063488724 +Step 34699 | loss:0.27208447456359863 lr:4.841856004365764e-07 tokens_per_second_per_gpu:5339.987010350635 mfu:25.189629823557123 +Step 34700 | loss:0.3296809494495392 lr:4.83583790840919e-07 tokens_per_second_per_gpu:2878.2627107190083 mfu:13.577256288718809 +Step 34701 | loss:0.3411741852760315 lr:4.829823536645406e-07 tokens_per_second_per_gpu:4022.4889535402267 mfu:18.9747667012345 +Step 34702 | loss:0.29468968510627747 lr:4.82381288911976e-07 tokens_per_second_per_gpu:4782.76046136499 mfu:22.561097119339 +Step 34703 | loss:0.23413820564746857 lr:4.817805965877442e-07 tokens_per_second_per_gpu:5073.755881102327 mfu:23.933772163178265 +Step 34704 | loss:0.29425176978111267 lr:4.811802766963524e-07 tokens_per_second_per_gpu:4881.143340147098 mfu:23.025185944404228 +Step 34705 | loss:0.16466069221496582 lr:4.80580329242325e-07 tokens_per_second_per_gpu:5365.264236565588 mfu:25.30886681235257 +Step 34706 | loss:0.276404470205307 lr:4.799807542301693e-07 tokens_per_second_per_gpu:3741.47582216092 mfu:17.649179814734083 +Step 34707 | loss:0.2965831458568573 lr:4.793815516643985e-07 tokens_per_second_per_gpu:4158.0293125373755 mfu:19.61413370019393 +Step 34708 | loss:0.376581609249115 lr:4.7878272154952e-07 tokens_per_second_per_gpu:5019.3603810157665 mfu:23.677179308440635 +Step 34709 | loss:0.3235691487789154 lr:4.781842638900303e-07 tokens_per_second_per_gpu:4337.396011931052 mfu:20.46023702434851 +Step 34710 | loss:0.30795085430145264 lr:4.775861786904423e-07 tokens_per_second_per_gpu:3893.794953317932 mfu:18.367695144725314 +Step 34711 | loss:0.30252203345298767 lr:4.76988465955247e-07 tokens_per_second_per_gpu:4576.991454129421 mfu:21.590449604395985 +Step 34712 | loss:0.37765851616859436 lr:4.763911256889353e-07 tokens_per_second_per_gpu:3435.6693244196194 mfu:16.20663838891964 +Step 34713 | loss:0.27546706795692444 lr:4.75794157896009e-07 tokens_per_second_per_gpu:4130.070125325157 mfu:19.482245443784656 +Step 34714 | loss:0.34439975023269653 lr:4.751975625809535e-07 tokens_per_second_per_gpu:5354.144062237792 mfu:25.256411052750973 +Step 34715 | loss:0.26684075593948364 lr:4.746013397482596e-07 tokens_per_second_per_gpu:5351.571040377166 mfu:25.2442736696316 +Step 34716 | loss:0.34492096304893494 lr:4.740054894024015e-07 tokens_per_second_per_gpu:4863.074106197532 mfu:22.939950284934913 +Step 34717 | loss:0.2447894960641861 lr:4.734100115478701e-07 tokens_per_second_per_gpu:5336.138841675759 mfu:25.171477355352458 +Step 34718 | loss:0.30991995334625244 lr:4.728149061891451e-07 tokens_per_second_per_gpu:5296.878831687829 mfu:24.986281189790965 +Step 34719 | loss:0.32503747940063477 lr:4.722201733307008e-07 tokens_per_second_per_gpu:5361.934792871071 mfu:25.29316125838386 +Step 34720 | loss:0.30503416061401367 lr:4.7162581297700014e-07 tokens_per_second_per_gpu:3513.2777005765033 mfu:16.572730340606036 +Step 34721 | loss:0.3453159034252167 lr:4.7103182513252853e-07 tokens_per_second_per_gpu:4343.316314646553 mfu:20.48816410236533 +Step 34722 | loss:0.3187309801578522 lr:4.7043820980174346e-07 tokens_per_second_per_gpu:3689.871321008549 mfu:17.405752578162545 +Step 34723 | loss:0.32323354482650757 lr:4.698449669891136e-07 tokens_per_second_per_gpu:5145.980197399996 mfu:24.274466585893407 +Step 34724 | loss:0.20540878176689148 lr:4.6925209669909655e-07 tokens_per_second_per_gpu:3801.7955570881827 mfu:17.933718295993994 +Step 34725 | loss:0.3421582877635956 lr:4.6865959893616086e-07 tokens_per_second_per_gpu:4874.342324025129 mfu:22.99310439098389 +Step 34726 | loss:0.30772119760513306 lr:4.6806747370475304e-07 tokens_per_second_per_gpu:4319.96626874436 mfu:20.378017951916046 +Step 34727 | loss:0.2586973011493683 lr:4.674757210093306e-07 tokens_per_second_per_gpu:4947.444697551818 mfu:23.337940759460334 +Step 34728 | loss:0.3836295008659363 lr:4.6688434085433996e-07 tokens_per_second_per_gpu:5322.40448122613 mfu:25.106690033788027 +Step 34729 | loss:0.3756934702396393 lr:4.662933332442387e-07 tokens_per_second_per_gpu:4375.516673979837 mfu:20.64005869128825 +Step 34730 | loss:0.3005928099155426 lr:4.6570269818346224e-07 tokens_per_second_per_gpu:3533.7597471102576 mfu:16.669347648703244 +Step 34731 | loss:0.3660051226615906 lr:4.6511243567645693e-07 tokens_per_second_per_gpu:4459.03804737926 mfu:21.03404325982907 +Step 34732 | loss:0.39391615986824036 lr:4.645225457276636e-07 tokens_per_second_per_gpu:4482.288211294479 mfu:21.14371824990417 +Step 34733 | loss:0.20412175357341766 lr:4.6393302834151773e-07 tokens_per_second_per_gpu:4775.337758168057 mfu:22.52608296191531 +Step 34734 | loss:0.3277950584888458 lr:4.6334388352244904e-07 tokens_per_second_per_gpu:5299.754090603176 mfu:24.999844276664415 +Step 34735 | loss:0.2811982333660126 lr:4.627551112748929e-07 tokens_per_second_per_gpu:5014.4843275152425 mfu:23.65417813213815 +Step 34736 | loss:0.34794434905052185 lr:4.6216671160327904e-07 tokens_per_second_per_gpu:4034.1131844738807 mfu:19.029600181846853 +Step 34737 | loss:0.26748406887054443 lr:4.6157868451202625e-07 tokens_per_second_per_gpu:4181.132847062688 mfu:19.723117014419287 +Step 34738 | loss:0.35894593596458435 lr:4.6099103000555863e-07 tokens_per_second_per_gpu:4834.600312583894 mfu:22.805634542329237 +Step 34739 | loss:0.38421082496643066 lr:4.604037480883061e-07 tokens_per_second_per_gpu:5212.617746979013 mfu:24.58880727679598 +Step 34740 | loss:0.27958717942237854 lr:4.5981683876467616e-07 tokens_per_second_per_gpu:4498.95058071283 mfu:21.222317489343904 +Step 34741 | loss:0.32441723346710205 lr:4.59230302039082e-07 tokens_per_second_per_gpu:4751.818817596829 mfu:22.415140106495787 +Step 34742 | loss:0.36774662137031555 lr:4.586441379159423e-07 tokens_per_second_per_gpu:5376.8119861647365 mfu:25.363339517461 +Step 34743 | loss:0.3032829463481903 lr:4.5805834639965905e-07 tokens_per_second_per_gpu:5348.947566417202 mfu:25.231898295351638 +Step 34744 | loss:0.32215002179145813 lr:4.574729274946399e-07 tokens_per_second_per_gpu:4534.275579004929 mfu:21.388951533353932 +Step 34745 | loss:0.3231840133666992 lr:4.5688788120529236e-07 tokens_per_second_per_gpu:5271.551376675467 mfu:24.866807263188228 +Step 34746 | loss:0.2781021296977997 lr:4.5630320753600744e-07 tokens_per_second_per_gpu:4104.926941140057 mfu:19.36364075411396 +Step 34747 | loss:0.3179790675640106 lr:4.5571890649119267e-07 tokens_per_second_per_gpu:4536.651643969505 mfu:21.40015983719033 +Step 34748 | loss:0.3250570595264435 lr:4.5513497807523895e-07 tokens_per_second_per_gpu:5311.184866144763 mfu:25.053765195185495 +Step 34749 | loss:0.3362647294998169 lr:4.545514222925318e-07 tokens_per_second_per_gpu:4577.6968871646795 mfu:21.593777252382583 +Step 34750 | loss:0.22558902204036713 lr:4.5396823914746754e-07 tokens_per_second_per_gpu:4507.901939319367 mfu:21.264542575150205 +Step 34751 | loss:0.3449278473854065 lr:4.5338542864443166e-07 tokens_per_second_per_gpu:5367.75467922597 mfu:25.320614655294925 +Step 34752 | loss:0.3696731925010681 lr:4.52802990787804e-07 tokens_per_second_per_gpu:4992.344801383077 mfu:23.54974220998005 +Step 34753 | loss:0.33925580978393555 lr:4.522209255819643e-07 tokens_per_second_per_gpu:5148.893509763511 mfu:24.288209177374455 +Step 34754 | loss:0.24366390705108643 lr:4.51639233031298e-07 tokens_per_second_per_gpu:4027.538547645471 mfu:18.998586498178533 +Step 34755 | loss:0.3126305937767029 lr:4.510579131401793e-07 tokens_per_second_per_gpu:4982.261152796925 mfu:23.502175919151316 +Step 34756 | loss:0.3114413022994995 lr:4.504769659129715e-07 tokens_per_second_per_gpu:3408.704203174603 mfu:16.0794393112828 +Step 34757 | loss:0.3455043137073517 lr:4.498963913540488e-07 tokens_per_second_per_gpu:5307.4419853057925 mfu:25.03610938766675 +Step 34758 | loss:0.3297017216682434 lr:4.4931618946778e-07 tokens_per_second_per_gpu:5304.8404373780695 mfu:25.023837442221833 +Step 34759 | loss:0.3754726052284241 lr:4.487363602585226e-07 tokens_per_second_per_gpu:4287.291015155145 mfu:20.223883205762583 +Step 34760 | loss:0.2756942808628082 lr:4.4815690373064543e-07 tokens_per_second_per_gpu:3864.019452195226 mfu:18.227239025705163 +Step 34761 | loss:0.3633938431739807 lr:4.475778198885006e-07 tokens_per_second_per_gpu:5330.073662692041 mfu:25.142866871260015 +Step 34762 | loss:0.21089249849319458 lr:4.469991087364511e-07 tokens_per_second_per_gpu:4907.6191923570295 mfu:23.150077056524662 +Step 34763 | loss:0.3045123219490051 lr:4.464207702788381e-07 tokens_per_second_per_gpu:5371.142978171123 mfu:25.33659783952262 +Step 34764 | loss:0.13967540860176086 lr:4.4584280452001913e-07 tokens_per_second_per_gpu:4969.5787198209455 mfu:23.442350718957428 +Step 34765 | loss:0.26436275243759155 lr:4.452652114643463e-07 tokens_per_second_per_gpu:4555.102620775262 mfu:21.487196242844345 +Step 34766 | loss:0.34649816155433655 lr:4.4468799111614946e-07 tokens_per_second_per_gpu:4177.502587600708 mfu:19.70599246115099 +Step 34767 | loss:0.36797139048576355 lr:4.441111434797751e-07 tokens_per_second_per_gpu:4918.453596720835 mfu:23.201184790448732 +Step 34768 | loss:0.19018276035785675 lr:4.4353466855957537e-07 tokens_per_second_per_gpu:5307.338278754899 mfu:25.035620186172025 +Step 34769 | loss:0.23605328798294067 lr:4.4295856635986897e-07 tokens_per_second_per_gpu:4724.235423335555 mfu:22.285024529553006 +Step 34770 | loss:0.37983420491218567 lr:4.423828368849914e-07 tokens_per_second_per_gpu:5098.047278109302 mfu:24.048358827399866 +Step 34771 | loss:0.30680665373802185 lr:4.4180748013927797e-07 tokens_per_second_per_gpu:4854.783513081536 mfu:22.900842142685708 +Step 34772 | loss:0.3175402283668518 lr:4.4123249612705864e-07 tokens_per_second_per_gpu:5053.436000943708 mfu:23.837919821541004 +Step 34773 | loss:0.3298441767692566 lr:4.406578848526466e-07 tokens_per_second_per_gpu:4130.727149079993 mfu:19.485344736936028 +Step 34774 | loss:0.2669052481651306 lr:4.400836463203717e-07 tokens_per_second_per_gpu:3533.945661266281 mfu:16.670224637497437 +Step 34775 | loss:0.29025551676750183 lr:4.395097805345527e-07 tokens_per_second_per_gpu:4919.9618813314355 mfu:23.208299626296828 +Step 34776 | loss:0.27813878655433655 lr:4.389362874995084e-07 tokens_per_second_per_gpu:3923.531070831648 mfu:18.507965484542325 +Step 34777 | loss:0.28145429491996765 lr:4.383631672195465e-07 tokens_per_second_per_gpu:3574.4300864079496 mfu:16.861196578245877 +Step 34778 | loss:0.3117212653160095 lr:4.377904196989746e-07 tokens_per_second_per_gpu:4020.863034247201 mfu:18.96709696251886 +Step 34779 | loss:0.3126508593559265 lr:4.372180449421115e-07 tokens_per_second_per_gpu:4825.977253363343 mfu:22.764958100740284 +Step 34780 | loss:0.27992138266563416 lr:4.366460429532482e-07 tokens_per_second_per_gpu:5334.747297846323 mfu:25.164913205687146 +Step 34781 | loss:0.40385669469833374 lr:4.360744137366979e-07 tokens_per_second_per_gpu:5366.8449537428005 mfu:25.31632332497564 +Step 34782 | loss:0.27610281109809875 lr:4.3550315729675717e-07 tokens_per_second_per_gpu:4516.941341139685 mfu:21.307182975817636 +Step 34783 | loss:0.35962003469467163 lr:4.3493227363772817e-07 tokens_per_second_per_gpu:4129.3378876040715 mfu:19.478791353521654 +Step 34784 | loss:0.3874995708465576 lr:4.3436176276389075e-07 tokens_per_second_per_gpu:4160.843803429512 mfu:19.6274101339338 +Step 34785 | loss:0.33850857615470886 lr:4.337916246795415e-07 tokens_per_second_per_gpu:4826.520294463909 mfu:22.767519718264015 +Step 34786 | loss:0.2545803487300873 lr:4.3322185938897143e-07 tokens_per_second_per_gpu:5170.73663239114 mfu:24.39124691363 +Step 34787 | loss:0.341626912355423 lr:4.32652466896466e-07 tokens_per_second_per_gpu:4632.323226966082 mfu:21.85145901744044 +Step 34788 | loss:0.35490527749061584 lr:4.320834472063051e-07 tokens_per_second_per_gpu:4675.137661429769 mfu:22.05342200970143 +Step 34789 | loss:0.2513682544231415 lr:4.315148003227687e-07 tokens_per_second_per_gpu:4115.306972815528 mfu:19.412605134543178 +Step 34790 | loss:0.39461570978164673 lr:4.309465262501422e-07 tokens_per_second_per_gpu:5361.319715197433 mfu:25.290259832054236 +Step 34791 | loss:0.21665355563163757 lr:4.3037862499268334e-07 tokens_per_second_per_gpu:5003.05082177954 mfu:23.6002443348256 +Step 34792 | loss:0.4206385910511017 lr:4.2981109655467753e-07 tokens_per_second_per_gpu:5353.485993757473 mfu:25.253306831451244 +Step 34793 | loss:0.3326776623725891 lr:4.2924394094038254e-07 tokens_per_second_per_gpu:3941.1375435722366 mfu:18.59101822043369 +Step 34794 | loss:0.29924577474594116 lr:4.286771581540727e-07 tokens_per_second_per_gpu:5385.663821829744 mfu:25.405095136571738 +Step 34795 | loss:0.2832271456718445 lr:4.2811074820000576e-07 tokens_per_second_per_gpu:3947.7776109583765 mfu:18.62234055120643 +Step 34796 | loss:0.25078538060188293 lr:4.275447110824393e-07 tokens_per_second_per_gpu:4390.386351596984 mfu:20.710201497636948 +Step 34797 | loss:0.29204222559928894 lr:4.2697904680564227e-07 tokens_per_second_per_gpu:4808.812908012065 mfu:22.683990955179006 +Step 34798 | loss:0.4014621675014496 lr:4.264137553738612e-07 tokens_per_second_per_gpu:5379.440458536527 mfu:25.375738470102366 +Step 34799 | loss:0.26320499181747437 lr:4.258488367913427e-07 tokens_per_second_per_gpu:4653.167719501364 mfu:21.949786045166523 +Step 34800 | loss:0.3630552887916565 lr:4.2528429106233893e-07 tokens_per_second_per_gpu:5362.70461250308 mfu:25.29679262893282 +Step 34801 | loss:0.32658690214157104 lr:4.2472011819110203e-07 tokens_per_second_per_gpu:5342.298305567895 mfu:25.20053259744457 +Step 34802 | loss:0.3456111252307892 lr:4.2415631818187305e-07 tokens_per_second_per_gpu:3426.3904134001773 mfu:16.16286818249546 +Step 34803 | loss:0.32249999046325684 lr:4.23592891038882e-07 tokens_per_second_per_gpu:3600.856108978053 mfu:16.985852635453433 +Step 34804 | loss:0.2902001440525055 lr:4.230298367663865e-07 tokens_per_second_per_gpu:4217.428200201704 mfu:19.894328387801842 +Step 34805 | loss:0.3005903661251068 lr:4.224671553686055e-07 tokens_per_second_per_gpu:5354.536713011281 mfu:25.25826325344265 +Step 34806 | loss:0.33522793650627136 lr:4.2190484684976885e-07 tokens_per_second_per_gpu:5346.367992907548 mfu:25.219730006985973 +Step 34807 | loss:0.4260476529598236 lr:4.213429112141176e-07 tokens_per_second_per_gpu:5356.180924876488 mfu:25.2660192813423 +Step 34808 | loss:0.3039892315864563 lr:4.207813484658707e-07 tokens_per_second_per_gpu:4399.1347013674285 mfu:20.751468956126516 +Step 34809 | loss:0.3385637104511261 lr:4.2022015860925246e-07 tokens_per_second_per_gpu:3376.8870453943578 mfu:15.92935234946649 +Step 34810 | loss:0.26957568526268005 lr:4.196593416484873e-07 tokens_per_second_per_gpu:4310.209994461246 mfu:20.33199594153056 +Step 34811 | loss:0.3424833416938782 lr:4.190988975877941e-07 tokens_per_second_per_gpu:4771.853092749977 mfu:22.509645200593 +Step 34812 | loss:0.27379491925239563 lr:4.1853882643138056e-07 tokens_per_second_per_gpu:3351.175975485864 mfu:15.80806884594725 +Step 34813 | loss:0.22933587431907654 lr:4.1797912818346e-07 tokens_per_second_per_gpu:5351.204172985492 mfu:25.242543093551042 +Step 34814 | loss:0.38318517804145813 lr:4.174198028482512e-07 tokens_per_second_per_gpu:4412.606829362184 mfu:20.815019282459833 +Step 34815 | loss:0.3454926609992981 lr:4.168608504299509e-07 tokens_per_second_per_gpu:4643.244256484282 mfu:21.902975376997563 +Step 34816 | loss:0.27684512734413147 lr:4.1630227093276683e-07 tokens_per_second_per_gpu:5361.971815831428 mfu:25.293335902004632 +Step 34817 | loss:0.2951948344707489 lr:4.1574406436090117e-07 tokens_per_second_per_gpu:5338.102433313062 mfu:25.180739952128892 +Step 34818 | loss:0.4050605595111847 lr:4.151862307185506e-07 tokens_per_second_per_gpu:5353.060191502486 mfu:25.251298249565117 +Step 34819 | loss:0.36834225058555603 lr:4.1462877000991715e-07 tokens_per_second_per_gpu:4196.415371311243 mfu:19.79520728876723 +Step 34820 | loss:0.2740730345249176 lr:4.1407168223918105e-07 tokens_per_second_per_gpu:3740.2709215844166 mfu:17.64349609313748 +Step 34821 | loss:0.27058956027030945 lr:4.135149674105443e-07 tokens_per_second_per_gpu:3737.453863540375 mfu:17.63020754970407 +Step 34822 | loss:0.241117462515831 lr:4.129586255281814e-07 tokens_per_second_per_gpu:4292.485644872168 mfu:20.248387160433005 +Step 34823 | loss:0.36211273074150085 lr:4.1240265659629464e-07 tokens_per_second_per_gpu:5352.624954091998 mfu:25.249245160440683 +Step 34824 | loss:0.24995796382427216 lr:4.118470606190472e-07 tokens_per_second_per_gpu:5388.427644072671 mfu:25.418132557648324 +Step 34825 | loss:0.35591086745262146 lr:4.1129183760062473e-07 tokens_per_second_per_gpu:4603.46254166811 mfu:21.715318240749234 +Step 34826 | loss:0.3678229749202728 lr:4.1073698754520717e-07 tokens_per_second_per_gpu:3790.2079086606545 mfu:17.87905738130466 +Step 34827 | loss:0.34922170639038086 lr:4.101825104569634e-07 tokens_per_second_per_gpu:3618.2854228447154 mfu:17.068069682710068 +Step 34828 | loss:0.4079459011554718 lr:4.096284063400624e-07 tokens_per_second_per_gpu:4455.9555470265805 mfu:21.019502579736685 +Step 34829 | loss:0.3750391900539398 lr:4.090746751986785e-07 tokens_per_second_per_gpu:4977.634310044057 mfu:23.480350312465266 +Step 34830 | loss:0.33024582266807556 lr:4.085213170369695e-07 tokens_per_second_per_gpu:4096.903864024986 mfu:19.32579453048406 +Step 34831 | loss:0.31596750020980835 lr:4.0796833185909877e-07 tokens_per_second_per_gpu:5359.547951849008 mfu:25.281902122045718 +Step 34832 | loss:0.3777658939361572 lr:4.0741571966922407e-07 tokens_per_second_per_gpu:4066.9641710661176 mfu:19.184563890558714 +Step 34833 | loss:0.3214874565601349 lr:4.068634804715088e-07 tokens_per_second_per_gpu:4417.191341566967 mfu:20.83664520872824 +Step 34834 | loss:0.29419809579849243 lr:4.063116142701051e-07 tokens_per_second_per_gpu:4959.862867672855 mfu:23.39651938667957 +Step 34835 | loss:0.4180425703525543 lr:4.0576012106915417e-07 tokens_per_second_per_gpu:4356.341712972701 mfu:20.549607128631887 +Step 34836 | loss:0.35456177592277527 lr:4.052090008728082e-07 tokens_per_second_per_gpu:5064.467543339199 mfu:23.889957489195755 +Step 34837 | loss:0.32418474555015564 lr:4.046582536852139e-07 tokens_per_second_per_gpu:5358.520795714512 mfu:25.27705685130839 +Step 34838 | loss:0.36780402064323425 lr:4.0410787951051244e-07 tokens_per_second_per_gpu:4976.455946071184 mfu:23.474791768555594 +Step 34839 | loss:0.31244510412216187 lr:4.035578783528449e-07 tokens_per_second_per_gpu:3890.9162060544763 mfu:18.354115602718664 +Step 34840 | loss:0.3810928165912628 lr:4.0300825021635234e-07 tokens_per_second_per_gpu:3718.418872146795 mfu:17.54041624759613 +Step 34841 | loss:0.29139968752861023 lr:4.024589951051594e-07 tokens_per_second_per_gpu:5090.433514470897 mfu:24.012443405275476 +Step 34842 | loss:0.3164560794830322 lr:4.019101130234015e-07 tokens_per_second_per_gpu:5097.694887594235 mfu:24.046696541216274 +Step 34843 | loss:0.34549444913864136 lr:4.013616039752033e-07 tokens_per_second_per_gpu:4131.178672329838 mfu:19.48747464914336 +Step 34844 | loss:0.2643318474292755 lr:4.008134679647002e-07 tokens_per_second_per_gpu:5181.3550972163575 mfu:24.44133602390746 +Step 34845 | loss:0.3286861479282379 lr:4.0026570499600016e-07 tokens_per_second_per_gpu:4173.277836337339 mfu:19.686063588624794 +Step 34846 | loss:0.39478176832199097 lr:3.997183150732331e-07 tokens_per_second_per_gpu:3767.5483404237284 mfu:17.77216833181007 +Step 34847 | loss:0.2844349443912506 lr:3.99171298200518e-07 tokens_per_second_per_gpu:4604.164330520106 mfu:21.718628698501625 +Step 34848 | loss:0.3421016335487366 lr:3.9862465438196273e-07 tokens_per_second_per_gpu:4548.281564913627 mfu:21.455020158552877 +Step 34849 | loss:0.31741583347320557 lr:3.980783836216806e-07 tokens_per_second_per_gpu:4639.1030154921655 mfu:21.88344043666918 +Step 34850 | loss:0.33588308095932007 lr:3.975324859237739e-07 tokens_per_second_per_gpu:3678.0980679738855 mfu:17.35021613487394 +Step 34851 | loss:0.2714410424232483 lr:3.9698696129236157e-07 tokens_per_second_per_gpu:4527.7077493789375 mfu:21.35796996042085 +Step 34852 | loss:0.3261103928089142 lr:3.9644180973154034e-07 tokens_per_second_per_gpu:5324.840562080375 mfu:25.118181442815757 +Step 34853 | loss:0.39919787645339966 lr:3.958970312454069e-07 tokens_per_second_per_gpu:4041.2457830251183 mfu:19.063245866159843 +Step 34854 | loss:0.36140039563179016 lr:3.95352625838058e-07 tokens_per_second_per_gpu:3379.775717958582 mfu:15.942978710810296 +Step 34855 | loss:0.3463204801082611 lr:3.9480859351360145e-07 tokens_per_second_per_gpu:4213.937424585317 mfu:19.877861803631603 +Step 34856 | loss:0.23860269784927368 lr:3.9426493427611177e-07 tokens_per_second_per_gpu:4939.228732549125 mfu:23.29918465075446 +Step 34857 | loss:0.2999383807182312 lr:3.937216481296857e-07 tokens_per_second_per_gpu:5353.8702757621295 mfu:25.255119555232483 +Step 34858 | loss:0.30941081047058105 lr:3.9317873507840883e-07 tokens_per_second_per_gpu:5332.2449611891 mfu:25.153109256732368 +Step 34859 | loss:0.29118892550468445 lr:3.9263619512636685e-07 tokens_per_second_per_gpu:4818.956551260777 mfu:22.73184025106718 +Step 34860 | loss:0.4296211302280426 lr:3.9209402827763417e-07 tokens_per_second_per_gpu:4034.6140981231433 mfu:19.031963076995105 +Step 34861 | loss:0.36787480115890503 lr:3.9155223453629096e-07 tokens_per_second_per_gpu:4151.069960349045 mfu:19.58130524853365 +Step 34862 | loss:0.28274211287498474 lr:3.910108139064228e-07 tokens_per_second_per_gpu:2845.6195009442167 mfu:13.423272698705219 +Step 34863 | loss:0.3494664132595062 lr:3.9046976639208755e-07 tokens_per_second_per_gpu:5388.373265353218 mfu:25.4178760439507 +Step 34864 | loss:0.42391490936279297 lr:3.8992909199735973e-07 tokens_per_second_per_gpu:5356.21566849305 mfu:25.266183172909496 +Step 34865 | loss:0.2839374542236328 lr:3.893887907263027e-07 tokens_per_second_per_gpu:4484.59789201198 mfu:21.154613407920788 +Step 34866 | loss:0.3257032036781311 lr:3.8884886258298556e-07 tokens_per_second_per_gpu:4775.537710368874 mfu:22.52702617056193 +Step 34867 | loss:0.3331833481788635 lr:3.883093075714661e-07 tokens_per_second_per_gpu:5320.3470082022595 mfu:25.096984582493334 +Step 34868 | loss:0.3624909222126007 lr:3.877701256957966e-07 tokens_per_second_per_gpu:3620.7633469975467 mfu:17.07975847371643 +Step 34869 | loss:0.3161669671535492 lr:3.872313169600461e-07 tokens_per_second_per_gpu:5357.7924837387645 mfu:25.273621279455956 +Step 34870 | loss:0.2131575047969818 lr:3.8669288136826134e-07 tokens_per_second_per_gpu:5361.229442497016 mfu:25.28983400032426 +Step 34871 | loss:0.2882183790206909 lr:3.861548189244835e-07 tokens_per_second_per_gpu:5316.5369891571345 mfu:25.079012072601202 +Step 34872 | loss:0.34228888154029846 lr:3.856171296327704e-07 tokens_per_second_per_gpu:4179.631107797499 mfu:19.716033054082253 +Step 34873 | loss:0.24538461863994598 lr:3.8507981349716336e-07 tokens_per_second_per_gpu:4055.2301663238004 mfu:19.12921258816155 +Step 34874 | loss:0.4615303575992584 lr:3.8454287052169803e-07 tokens_per_second_per_gpu:5346.535748240779 mfu:25.220521337514835 +Step 34875 | loss:0.30763736367225647 lr:3.8400630071041557e-07 tokens_per_second_per_gpu:4861.42161999923 mfu:22.9321552255943 +Step 34876 | loss:0.3601545989513397 lr:3.8347010406735716e-07 tokens_per_second_per_gpu:5319.678586657775 mfu:25.09383152402353 +Step 34877 | loss:0.36881333589553833 lr:3.829342805965475e-07 tokens_per_second_per_gpu:4461.673375293216 mfu:21.046474551232507 +Step 34878 | loss:0.3936924338340759 lr:3.8239883030202207e-07 tokens_per_second_per_gpu:4490.420627083047 mfu:21.18208024271206 +Step 34879 | loss:0.25918081402778625 lr:3.8186375318780554e-07 tokens_per_second_per_gpu:5224.9355593191785 mfu:24.64691249924004 +Step 34880 | loss:0.22498849034309387 lr:3.81329049257928e-07 tokens_per_second_per_gpu:3779.531144886326 mfu:17.828693264937407 +Step 34881 | loss:0.31968802213668823 lr:3.8079471851639737e-07 tokens_per_second_per_gpu:4350.238246339268 mfu:20.520816035163197 +Step 34882 | loss:0.31911805272102356 lr:3.802607609672437e-07 tokens_per_second_per_gpu:3557.761800840171 mfu:16.782569431319896 +Step 34883 | loss:0.31073737144470215 lr:3.797271766144861e-07 tokens_per_second_per_gpu:3443.7334068221076 mfu:16.24467804148651 +Step 34884 | loss:0.34985607862472534 lr:3.791939654621268e-07 tokens_per_second_per_gpu:5349.566793614225 mfu:25.23481929569105 +Step 34885 | loss:0.22757765650749207 lr:3.7866112751418493e-07 tokens_per_second_per_gpu:5325.387154008591 mfu:25.12075981019935 +Step 34886 | loss:0.2976285517215729 lr:3.781286627746572e-07 tokens_per_second_per_gpu:5171.216076840376 mfu:24.3935085348987 +Step 34887 | loss:0.3674849569797516 lr:3.7759657124756263e-07 tokens_per_second_per_gpu:4977.089634568248 mfu:23.477780985314652 +Step 34888 | loss:0.35224539041519165 lr:3.7706485293689254e-07 tokens_per_second_per_gpu:4896.781331400552 mfu:23.098953017263202 +Step 34889 | loss:0.35668790340423584 lr:3.765335078466492e-07 tokens_per_second_per_gpu:5180.208495808716 mfu:24.435927309437247 +Step 34890 | loss:0.33447012305259705 lr:3.7600253598082944e-07 tokens_per_second_per_gpu:3114.231977164187 mfu:14.690363579049103 +Step 34891 | loss:0.3163595497608185 lr:3.7547193734343565e-07 tokens_per_second_per_gpu:4071.0908105374256 mfu:19.204029953022168 +Step 34892 | loss:0.24241380393505096 lr:3.749417119384424e-07 tokens_per_second_per_gpu:5382.904121375058 mfu:25.392077158673533 +Step 34893 | loss:0.3134585916996002 lr:3.744118597698409e-07 tokens_per_second_per_gpu:3845.727783501848 mfu:18.140954103598006 +Step 34894 | loss:0.33403322100639343 lr:3.7388238084162254e-07 tokens_per_second_per_gpu:4785.878648112593 mfu:22.575806138244964 +Step 34895 | loss:0.3032691478729248 lr:3.733532751577673e-07 tokens_per_second_per_gpu:4363.292754874035 mfu:20.582396379250525 +Step 34896 | loss:0.18316222727298737 lr:3.728245427222554e-07 tokens_per_second_per_gpu:4772.0008006404705 mfu:22.510341963914023 +Step 34897 | loss:0.20870831608772278 lr:3.722961835390615e-07 tokens_per_second_per_gpu:2956.8126103818445 mfu:13.94778956742326 +Step 34898 | loss:0.3635662794113159 lr:3.717681976121656e-07 tokens_per_second_per_gpu:4543.150993847172 mfu:21.430818379466718 +Step 34899 | loss:0.3890664279460907 lr:3.7124058494553137e-07 tokens_per_second_per_gpu:4304.179375175796 mfu:20.303548481431335 +Step 34900 | loss:0.3374179005622864 lr:3.7071334554312777e-07 tokens_per_second_per_gpu:3377.3218467165016 mfu:15.931403381487797 +Step 34901 | loss:0.3615029454231262 lr:3.701864794089183e-07 tokens_per_second_per_gpu:4606.758431452982 mfu:21.730865515201263 +Step 34902 | loss:0.39336392283439636 lr:3.6965998654687194e-07 tokens_per_second_per_gpu:5128.804785181264 mfu:24.193447236030153 +Step 34903 | loss:0.33072251081466675 lr:3.691338669609468e-07 tokens_per_second_per_gpu:4911.023872955841 mfu:23.16613751580767 +Step 34904 | loss:0.29098209738731384 lr:3.6860812065510065e-07 tokens_per_second_per_gpu:4597.944126484282 mfu:21.689286934788395 +Step 34905 | loss:0.3197125196456909 lr:3.6808274763328043e-07 tokens_per_second_per_gpu:5330.066259984043 mfu:25.14283195142336 +Step 34906 | loss:0.411714643239975 lr:3.6755774789944965e-07 tokens_per_second_per_gpu:5293.725993734531 mfu:24.971408715235633 +Step 34907 | loss:0.3041560649871826 lr:3.67033121457544e-07 tokens_per_second_per_gpu:5328.010726012913 mfu:25.13313564697128 +Step 34908 | loss:0.3879540264606476 lr:3.6650886831152145e-07 tokens_per_second_per_gpu:5331.3250077167095 mfu:25.148769679242854 +Step 34909 | loss:0.3297995328903198 lr:3.6598498846531214e-07 tokens_per_second_per_gpu:3390.049268797104 mfu:15.991440802964126 +Step 34910 | loss:0.28773677349090576 lr:3.6546148192286857e-07 tokens_per_second_per_gpu:5352.785944188925 mfu:25.250004578195632 +Step 34911 | loss:0.26997023820877075 lr:3.649383486881208e-07 tokens_per_second_per_gpu:5325.553912851521 mfu:25.121546440113445 +Step 34912 | loss:0.3849107325077057 lr:3.644155887650047e-07 tokens_per_second_per_gpu:5372.319207677244 mfu:25.34214631478844 +Step 34913 | loss:0.3341422975063324 lr:3.6389320215745594e-07 tokens_per_second_per_gpu:4212.905700116875 mfu:19.872994983283615 +Step 34914 | loss:0.28874829411506653 lr:3.633711888693936e-07 tokens_per_second_per_gpu:5369.688022910002 mfu:25.329734567314947 +Step 34915 | loss:0.2924230098724365 lr:3.628495489047534e-07 tokens_per_second_per_gpu:4406.967277672762 mfu:20.788416554934095 +Step 34916 | loss:0.4825555086135864 lr:3.623282822674545e-07 tokens_per_second_per_gpu:5042.664483666587 mfu:23.787108736734467 +Step 34917 | loss:0.2922585606575012 lr:3.618073889614215e-07 tokens_per_second_per_gpu:5382.870412541114 mfu:25.391918148352502 +Step 34918 | loss:0.2670283615589142 lr:3.612868689905624e-07 tokens_per_second_per_gpu:5174.846883690191 mfu:24.41063567028881 +Step 34919 | loss:0.3278936743736267 lr:3.607667223588074e-07 tokens_per_second_per_gpu:5285.252489441242 mfu:24.93143774975528 +Step 34920 | loss:0.23932106792926788 lr:3.6024694907005885e-07 tokens_per_second_per_gpu:4587.408855545932 mfu:21.639590264269625 +Step 34921 | loss:0.2636159360408783 lr:3.5972754912821925e-07 tokens_per_second_per_gpu:3904.772551885494 mfu:18.419478350139954 +Step 34922 | loss:0.341966837644577 lr:3.592085225372077e-07 tokens_per_second_per_gpu:5371.954860766378 mfu:25.340427628245898 +Step 34923 | loss:0.3752627670764923 lr:3.586898693009266e-07 tokens_per_second_per_gpu:3905.963916792078 mfu:18.425098221672126 +Step 34924 | loss:0.3078274726867676 lr:3.581715894232673e-07 tokens_per_second_per_gpu:4385.696943182291 mfu:20.688080758049257 +Step 34925 | loss:0.31024831533432007 lr:3.576536829081323e-07 tokens_per_second_per_gpu:4910.627997982211 mfu:23.16427010601382 +Step 34926 | loss:0.3471332788467407 lr:3.571361497594239e-07 tokens_per_second_per_gpu:5307.554333171219 mfu:25.036639351716488 +Step 34927 | loss:0.376823365688324 lr:3.5661898998102796e-07 tokens_per_second_per_gpu:3628.0531235387093 mfu:17.114145593425224 +Step 34928 | loss:0.24692609906196594 lr:3.5610220357683575e-07 tokens_per_second_per_gpu:5172.149727440269 mfu:24.397912724075617 +Step 34929 | loss:0.31934013962745667 lr:3.5558579055073314e-07 tokens_per_second_per_gpu:5301.753701587455 mfu:25.009276782846072 +Step 34930 | loss:0.34890973567962646 lr:3.5506975090660034e-07 tokens_per_second_per_gpu:3533.751041581428 mfu:16.66930658323844 +Step 34931 | loss:0.3221568763256073 lr:3.545540846483286e-07 tokens_per_second_per_gpu:3925.29821215824 mfu:18.516301391685257 +Step 34932 | loss:0.28426554799079895 lr:3.540387917797872e-07 tokens_per_second_per_gpu:4402.909490665176 mfu:20.76927528128946 +Step 34933 | loss:0.29698339104652405 lr:3.535238723048506e-07 tokens_per_second_per_gpu:4561.911018719478 mfu:21.519312617579686 +Step 34934 | loss:0.22901681065559387 lr:3.530093262274048e-07 tokens_per_second_per_gpu:4615.758282720863 mfu:21.77331934916482 +Step 34935 | loss:0.33068421483039856 lr:3.524951535513077e-07 tokens_per_second_per_gpu:4190.463228665534 mfu:19.767130016367172 +Step 34936 | loss:0.28460586071014404 lr:3.5198135428042846e-07 tokens_per_second_per_gpu:3633.1805278569586 mfu:17.13833243441979 +Step 34937 | loss:0.2296612560749054 lr:3.5146792841863064e-07 tokens_per_second_per_gpu:4482.665778392547 mfu:21.145499298325348 +Step 34938 | loss:0.25128743052482605 lr:3.509548759697834e-07 tokens_per_second_per_gpu:3939.9051476471222 mfu:18.5852047985865 +Step 34939 | loss:0.31039464473724365 lr:3.5044219693773364e-07 tokens_per_second_per_gpu:5345.229878038713 mfu:25.214361324967257 +Step 34940 | loss:0.31165561079978943 lr:3.499298913263449e-07 tokens_per_second_per_gpu:5321.377210592023 mfu:25.101844222936506 +Step 34941 | loss:0.3042067289352417 lr:3.494179591394697e-07 tokens_per_second_per_gpu:3434.248253466679 mfu:16.19993495477491 +Step 34942 | loss:0.232025146484375 lr:3.4890640038096057e-07 tokens_per_second_per_gpu:4217.193084936565 mfu:19.893219308981376 +Step 34943 | loss:0.34240275621414185 lr:3.483952150546588e-07 tokens_per_second_per_gpu:4436.885629091994 mfu:20.929546523175734 +Step 34944 | loss:0.3823999762535095 lr:3.4788440316441127e-07 tokens_per_second_per_gpu:4455.405728081297 mfu:21.016908989962996 +Step 34945 | loss:0.2322763055562973 lr:3.47373964714065e-07 tokens_per_second_per_gpu:4917.451586073874 mfu:23.196458135266376 +Step 34946 | loss:0.29360032081604004 lr:3.4686389970745026e-07 tokens_per_second_per_gpu:4337.80022557089 mfu:20.462143769053583 +Step 34947 | loss:0.3041911721229553 lr:3.4635420814840837e-07 tokens_per_second_per_gpu:4805.957751508702 mfu:22.67052269481271 +Step 34948 | loss:0.35681846737861633 lr:3.4584489004077514e-07 tokens_per_second_per_gpu:4506.740309484752 mfu:21.25906296902584 +Step 34949 | loss:0.2632613480091095 lr:3.453359453883753e-07 tokens_per_second_per_gpu:5310.0648307736665 mfu:25.048481797242292 +Step 34950 | loss:0.3360857665538788 lr:3.448273741950392e-07 tokens_per_second_per_gpu:5373.1435342994355 mfu:25.346034804109944 +Step 34951 | loss:0.3508561849594116 lr:3.4431917646459697e-07 tokens_per_second_per_gpu:4036.5660487124887 mfu:19.04117076096216 +Step 34952 | loss:0.21521461009979248 lr:3.4381135220086236e-07 tokens_per_second_per_gpu:4558.923730654569 mfu:21.505221070093608 +Step 34953 | loss:0.3078506290912628 lr:3.4330390140766e-07 tokens_per_second_per_gpu:4460.256440947818 mfu:21.039790630171332 +Step 34954 | loss:0.2998577654361725 lr:3.4279682408880356e-07 tokens_per_second_per_gpu:4532.232772264167 mfu:21.379315265418988 +Step 34955 | loss:0.28787490725517273 lr:3.4229012024811214e-07 tokens_per_second_per_gpu:5375.7531546750615 mfu:25.358344828668923 +Step 34956 | loss:0.3251354396343231 lr:3.4178378988939387e-07 tokens_per_second_per_gpu:4450.2316877193725 mfu:20.992502158793485 +Step 34957 | loss:0.25125569105148315 lr:3.412778330164512e-07 tokens_per_second_per_gpu:4589.910269600888 mfu:21.6513898611521 +Step 34958 | loss:0.2763175666332245 lr:3.4077224963309784e-07 tokens_per_second_per_gpu:4266.795702586662 mfu:20.127203319516138 +Step 34959 | loss:0.31913137435913086 lr:3.4026703974313067e-07 tokens_per_second_per_gpu:5265.866710301831 mfu:24.839991721999688 +Step 34960 | loss:0.28413155674934387 lr:3.3976220335035227e-07 tokens_per_second_per_gpu:5339.5543223535915 mfu:25.187588760450886 +Step 34961 | loss:0.4031963050365448 lr:3.3925774045855954e-07 tokens_per_second_per_gpu:4822.148977575164 mfu:22.746899470676244 +Step 34962 | loss:0.3207933306694031 lr:3.3875365107154945e-07 tokens_per_second_per_gpu:3822.9315206271767 mfu:18.033420242174667 +Step 34963 | loss:0.269038587808609 lr:3.3824993519310787e-07 tokens_per_second_per_gpu:5271.145869552435 mfu:24.864894416902242 +Step 34964 | loss:0.24103359878063202 lr:3.3774659282702623e-07 tokens_per_second_per_gpu:4363.025649076791 mfu:20.581136395631972 +Step 34965 | loss:0.2757965624332428 lr:3.372436239770849e-07 tokens_per_second_per_gpu:4689.522074706842 mfu:22.121275741363362 +Step 34966 | loss:0.2517131268978119 lr:3.367410286470696e-07 tokens_per_second_per_gpu:4577.936820118524 mfu:21.594909057062655 +Step 34967 | loss:0.2230125069618225 lr:3.362388068407718e-07 tokens_per_second_per_gpu:5079.061641218177 mfu:23.95880033890075 +Step 34968 | loss:0.4380805194377899 lr:3.3573695856194965e-07 tokens_per_second_per_gpu:5345.205039157354 mfu:25.214244155725936 +Step 34969 | loss:0.3017367124557495 lr:3.35235483814389e-07 tokens_per_second_per_gpu:5181.136477681188 mfu:24.440304758259728 +Step 34970 | loss:0.2927623689174652 lr:3.347343826018645e-07 tokens_per_second_per_gpu:5175.347360107086 mfu:24.412996502937382 +Step 34971 | loss:0.2806455194950104 lr:3.3423365492813997e-07 tokens_per_second_per_gpu:4331.639426040225 mfu:20.433082226526643 +Step 34972 | loss:0.42330101132392883 lr:3.3373330079697893e-07 tokens_per_second_per_gpu:5366.371042512013 mfu:25.314087804842423 +Step 34973 | loss:0.29567867517471313 lr:3.3323332021214517e-07 tokens_per_second_per_gpu:3909.944861468938 mfu:18.443877016932078 +Step 34974 | loss:0.15610666573047638 lr:3.327337131774022e-07 tokens_per_second_per_gpu:2966.6640379672995 mfu:13.994260432170718 +Step 34975 | loss:0.2938631772994995 lr:3.322344796965082e-07 tokens_per_second_per_gpu:3674.6504283961885 mfu:17.333953030757975 +Step 34976 | loss:0.3196363151073456 lr:3.3173561977321574e-07 tokens_per_second_per_gpu:5376.688540597366 mfu:25.362757203657726 +Step 34977 | loss:0.34317559003829956 lr:3.312371334112774e-07 tokens_per_second_per_gpu:4268.240021702081 mfu:20.134016419209825 +Step 34978 | loss:0.24409182369709015 lr:3.3073902061444564e-07 tokens_per_second_per_gpu:5320.863486199723 mfu:25.099420897327608 +Step 34979 | loss:0.37764912843704224 lr:3.3024128138645636e-07 tokens_per_second_per_gpu:4843.8333777832895 mfu:22.84918848619815 +Step 34980 | loss:0.24284043908119202 lr:3.297439157310622e-07 tokens_per_second_per_gpu:5369.711618022115 mfu:25.329845869484526 +Step 34981 | loss:0.3403586447238922 lr:3.292469236520046e-07 tokens_per_second_per_gpu:4683.367714556222 mfu:22.09224457449077 +Step 34982 | loss:0.3343542516231537 lr:3.2875030515301944e-07 tokens_per_second_per_gpu:4416.687244408324 mfu:20.834267296423075 +Step 34983 | loss:0.1672932505607605 lr:3.28254060237837e-07 tokens_per_second_per_gpu:5322.852895003688 mfu:25.10880527808405 +Step 34984 | loss:0.20170865952968597 lr:3.2775818891019885e-07 tokens_per_second_per_gpu:3720.465909827376 mfu:17.550072473596074 +Step 34985 | loss:0.4362342059612274 lr:3.272626911738297e-07 tokens_per_second_per_gpu:4147.1912320507745 mfu:19.563008625371417 +Step 34986 | loss:0.29917651414871216 lr:3.2676756703245435e-07 tokens_per_second_per_gpu:2682.979482349869 mfu:12.656071981747099 +Step 34987 | loss:0.29930952191352844 lr:3.2627281648979767e-07 tokens_per_second_per_gpu:4541.299304894625 mfu:21.422083646746838 +Step 34988 | loss:0.4166978597640991 lr:3.2577843954957885e-07 tokens_per_second_per_gpu:4799.164021843117 mfu:22.638475512850338 +Step 34989 | loss:0.22372452914714813 lr:3.252844362155227e-07 tokens_per_second_per_gpu:4448.198340650335 mfu:20.982910513744052 +Step 34990 | loss:0.2577323317527771 lr:3.247908064913374e-07 tokens_per_second_per_gpu:5328.217586008309 mfu:25.134111440863514 +Step 34991 | loss:0.30125755071640015 lr:3.2429755038074773e-07 tokens_per_second_per_gpu:4321.31078134293 mfu:20.384360247241002 +Step 34992 | loss:0.3587365746498108 lr:3.238046678874451e-07 tokens_per_second_per_gpu:4395.87214121423 mfu:20.736078903233718 +Step 34993 | loss:0.29593122005462646 lr:3.233121590151544e-07 tokens_per_second_per_gpu:4257.557854630313 mfu:20.083626814566653 +Step 34994 | loss:0.3293365240097046 lr:3.228200237675616e-07 tokens_per_second_per_gpu:3930.109364764521 mfu:18.53899641939637 +Step 34995 | loss:0.25757765769958496 lr:3.223282621483859e-07 tokens_per_second_per_gpu:5312.945242509017 mfu:25.06206919084686 +Step 34996 | loss:0.3592286705970764 lr:3.218368741613187e-07 tokens_per_second_per_gpu:4335.1473587909295 mfu:20.449629743826762 +Step 34997 | loss:0.2967120110988617 lr:3.213458598100516e-07 tokens_per_second_per_gpu:5333.630450400676 mfu:25.159644845732206 +Step 34998 | loss:0.2334357649087906 lr:3.2085521909828165e-07 tokens_per_second_per_gpu:5362.833774564463 mfu:25.297401908413683 +Step 34999 | loss:0.424694299697876 lr:3.203649520297003e-07 tokens_per_second_per_gpu:3877.4225574287793 mfu:18.290463759897396 +Step 35000 | loss:0.3240860402584076 lr:3.1987505860799347e-07 tokens_per_second_per_gpu:4618.462552081253 mfu:21.78607584913371 +Step 35001 | loss:0.30456891655921936 lr:3.193855388368472e-07 tokens_per_second_per_gpu:4937.647860145681 mfu:23.291727405902765 +Step 35002 | loss:0.3347952961921692 lr:3.1889639271994176e-07 tokens_per_second_per_gpu:3942.1672707700814 mfu:18.595875619315503 +Step 35003 | loss:0.2595647871494293 lr:3.1840762026095206e-07 tokens_per_second_per_gpu:3883.370414436961 mfu:18.318520816214033 +Step 35004 | loss:0.26974087953567505 lr:3.179192214635585e-07 tokens_per_second_per_gpu:5244.146879068051 mfu:24.737535572284614 +Step 35005 | loss:0.398303359746933 lr:3.174311963314303e-07 tokens_per_second_per_gpu:5220.5943982682675 mfu:24.62643450188444 +Step 35006 | loss:0.2624557316303253 lr:3.1694354486824786e-07 tokens_per_second_per_gpu:4069.8787961182757 mfu:19.19831267421605 +Step 35007 | loss:0.3643094599246979 lr:3.1645626707767496e-07 tokens_per_second_per_gpu:3465.5814606849976 mfu:16.347739039220134 +Step 35008 | loss:0.23198474943637848 lr:3.1596936296336975e-07 tokens_per_second_per_gpu:3379.953882831139 mfu:15.943819144912208 +Step 35009 | loss:0.3598199188709259 lr:3.15482832528996e-07 tokens_per_second_per_gpu:3937.0769602446594 mfu:18.571863756070552 +Step 35010 | loss:0.23140257596969604 lr:3.14996675778223e-07 tokens_per_second_per_gpu:4836.275209875931 mfu:22.81353531862272 +Step 35011 | loss:0.3784673810005188 lr:3.1451089271469224e-07 tokens_per_second_per_gpu:5330.217732139644 mfu:25.14354647142514 +Step 35012 | loss:0.26612645387649536 lr:3.140254833420675e-07 tokens_per_second_per_gpu:3934.6626325880666 mfu:18.560474960587634 +Step 35013 | loss:0.35102641582489014 lr:3.135404476640014e-07 tokens_per_second_per_gpu:5206.28260704927 mfu:24.558923340861053 +Step 35014 | loss:0.32100820541381836 lr:3.130557856841354e-07 tokens_per_second_per_gpu:5326.948577505236 mfu:25.128125311239344 +Step 35015 | loss:0.37271955609321594 lr:3.1257149740611114e-07 tokens_per_second_per_gpu:5349.8409568178895 mfu:25.2361125702998 +Step 35016 | loss:0.3928719758987427 lr:3.120875828335812e-07 tokens_per_second_per_gpu:5000.51404594359 mfu:23.588277930388838 +Step 35017 | loss:0.2736862897872925 lr:3.1160404197018154e-07 tokens_per_second_per_gpu:3475.729037369667 mfu:16.395606889796834 +Step 35018 | loss:0.3241003453731537 lr:3.111208748195482e-07 tokens_per_second_per_gpu:4880.0277058513275 mfu:23.019923307084238 +Step 35019 | loss:0.2176782339811325 lr:3.106380813853116e-07 tokens_per_second_per_gpu:4197.556804545769 mfu:19.800591624083086 +Step 35020 | loss:0.34378373622894287 lr:3.101556616711132e-07 tokens_per_second_per_gpu:4032.5819115563513 mfu:19.02237690623277 +Step 35021 | loss:0.2762153446674347 lr:3.0967361568056686e-07 tokens_per_second_per_gpu:5317.57067988043 mfu:25.08388816810881 +Step 35022 | loss:0.3040749728679657 lr:3.0919194341731406e-07 tokens_per_second_per_gpu:3934.3142414638924 mfu:18.558831540213244 +Step 35023 | loss:0.17909425497055054 lr:3.0871064488496305e-07 tokens_per_second_per_gpu:3282.0454833620547 mfu:15.481968519721036 +Step 35024 | loss:0.3024325966835022 lr:3.082297200871442e-07 tokens_per_second_per_gpu:5316.305750066036 mfu:25.077921278355035 +Step 35025 | loss:0.35225436091423035 lr:3.077491690274659e-07 tokens_per_second_per_gpu:5206.802684044675 mfu:24.56137663278266 +Step 35026 | loss:0.3468920886516571 lr:3.0726899170954724e-07 tokens_per_second_per_gpu:3415.962751089822 mfu:16.113679120234867 +Step 35027 | loss:0.331833153963089 lr:3.067891881370022e-07 tokens_per_second_per_gpu:4086.5005916735145 mfu:19.276720519820874 +Step 35028 | loss:0.27849799394607544 lr:3.0630975831343334e-07 tokens_per_second_per_gpu:4848.82011030271 mfu:22.87271175431656 +Step 35029 | loss:0.3509107232093811 lr:3.058307022424545e-07 tokens_per_second_per_gpu:5348.378238139261 mfu:25.229212676727165 +Step 35030 | loss:0.2225477397441864 lr:3.053520199276572e-07 tokens_per_second_per_gpu:4653.9087014976085 mfu:21.953281383667374 +Step 35031 | loss:0.26561182737350464 lr:3.048737113726552e-07 tokens_per_second_per_gpu:3532.135672047671 mfu:16.661686609537245 +Step 35032 | loss:0.2866198420524597 lr:3.0439577658103456e-07 tokens_per_second_per_gpu:5344.358577975993 mfu:25.21025125391229 +Step 35033 | loss:0.2970885932445526 lr:3.0391821555639246e-07 tokens_per_second_per_gpu:4672.4180308512505 mfu:22.040593048245732 +Step 35034 | loss:0.39306655526161194 lr:3.034410283023259e-07 tokens_per_second_per_gpu:5101.399855245288 mfu:24.06417350575895 +Step 35035 | loss:0.2026972472667694 lr:3.0296421482242653e-07 tokens_per_second_per_gpu:5331.353972429498 mfu:25.14890631073497 +Step 35036 | loss:0.34613990783691406 lr:3.024877751202693e-07 tokens_per_second_per_gpu:4842.766800572065 mfu:22.844157259515967 +Step 35037 | loss:0.3092201054096222 lr:3.020117091994401e-07 tokens_per_second_per_gpu:5364.567970830457 mfu:25.305582408084767 +Step 35038 | loss:0.2920359671115875 lr:3.0153601706353064e-07 tokens_per_second_per_gpu:3626.5194467503297 mfu:17.106910978342196 +Step 35039 | loss:0.4291488826274872 lr:3.010606987160991e-07 tokens_per_second_per_gpu:5362.1051919588335 mfu:25.293965059953916 +Step 35040 | loss:0.27196913957595825 lr:3.0058575416073707e-07 tokens_per_second_per_gpu:3990.1778391422013 mfu:18.82234966177443 +Step 35041 | loss:0.32588815689086914 lr:3.0011118340101394e-07 tokens_per_second_per_gpu:5161.49594618052 mfu:24.34765702015086 +Step 35042 | loss:0.30835381150245667 lr:2.9963698644049353e-07 tokens_per_second_per_gpu:3809.5772803608857 mfu:17.970426012370236 +Step 35043 | loss:0.3512812554836273 lr:2.991631632827452e-07 tokens_per_second_per_gpu:4111.280938837805 mfu:19.39361364535803 +Step 35044 | loss:0.2881428599357605 lr:2.9868971393133274e-07 tokens_per_second_per_gpu:4023.836710857635 mfu:18.98112430245182 +Step 35045 | loss:0.2554243803024292 lr:2.9821663838981993e-07 tokens_per_second_per_gpu:4336.399721128726 mfu:20.455537350649198 +Step 35046 | loss:0.28482574224472046 lr:2.977439366617596e-07 tokens_per_second_per_gpu:4576.78013261487 mfu:21.58945276475668 +Step 35047 | loss:0.29287365078926086 lr:2.9727160875070436e-07 tokens_per_second_per_gpu:4929.965544645993 mfu:23.255488612952554 +Step 35048 | loss:0.28499385714530945 lr:2.967996546602181e-07 tokens_per_second_per_gpu:4013.9024750091385 mfu:18.934262816999052 +Step 35049 | loss:0.2736794054508209 lr:2.963280743938424e-07 tokens_per_second_per_gpu:3760.3879428058053 mfu:17.73839151455652 +Step 35050 | loss:0.3890710473060608 lr:2.958568679551243e-07 tokens_per_second_per_gpu:5383.529738489072 mfu:25.395028301341622 +Step 35051 | loss:0.289040744304657 lr:2.953860353476112e-07 tokens_per_second_per_gpu:3374.8756866459526 mfu:15.919864427077071 +Step 35052 | loss:0.3305049240589142 lr:2.9491557657483906e-07 tokens_per_second_per_gpu:3982.85252421219 mfu:18.78779490142196 +Step 35053 | loss:0.25188666582107544 lr:2.944454916403494e-07 tokens_per_second_per_gpu:5390.771327099549 mfu:25.429188110359668 +Step 35054 | loss:0.4992977976799011 lr:2.9397578054767836e-07 tokens_per_second_per_gpu:3797.44391750667 mfu:17.91319086962173 +Step 35055 | loss:0.3377298414707184 lr:2.935064433003565e-07 tokens_per_second_per_gpu:3903.2385680210205 mfu:18.412242286526595 +Step 35056 | loss:0.34621715545654297 lr:2.930374799019198e-07 tokens_per_second_per_gpu:3725.9540475050203 mfu:17.575960955394365 +Step 35057 | loss:0.2431425005197525 lr:2.9256889035588764e-07 tokens_per_second_per_gpu:4184.297976308954 mfu:19.738047469100664 +Step 35058 | loss:0.3459416925907135 lr:2.921006746657906e-07 tokens_per_second_per_gpu:4236.706474448671 mfu:19.985267296637538 +Step 35059 | loss:0.37646135687828064 lr:2.9163283283514254e-07 tokens_per_second_per_gpu:4500.593899278011 mfu:21.230069303394924 +Step 35060 | loss:0.3283398151397705 lr:2.9116536486746836e-07 tokens_per_second_per_gpu:3916.0891265261225 mfu:18.47286056352653 +Step 35061 | loss:0.34478074312210083 lr:2.9069827076628196e-07 tokens_per_second_per_gpu:4488.257863732042 mfu:21.17187811897989 +Step 35062 | loss:0.27042123675346375 lr:2.902315505350972e-07 tokens_per_second_per_gpu:5345.914129426865 mfu:25.217589055510494 +Step 35063 | loss:0.2774309515953064 lr:2.8976520417742794e-07 tokens_per_second_per_gpu:5345.101664856807 mfu:25.21375652151278 +Step 35064 | loss:0.38544172048568726 lr:2.892992316967713e-07 tokens_per_second_per_gpu:4776.245250173493 mfu:22.530363756539057 +Step 35065 | loss:0.2757672667503357 lr:2.8883363309664124e-07 tokens_per_second_per_gpu:5361.750636816438 mfu:25.29229256285389 +Step 35066 | loss:0.29441896080970764 lr:2.8836840838053493e-07 tokens_per_second_per_gpu:5361.295324554153 mfu:25.290144777228413 +Step 35067 | loss:0.368203729391098 lr:2.8790355755195506e-07 tokens_per_second_per_gpu:5029.549624584395 mfu:23.72524370881807 +Step 35068 | loss:0.3896070122718811 lr:2.8743908061439893e-07 tokens_per_second_per_gpu:4079.5349289164114 mfu:19.243862300131003 +Step 35069 | loss:0.31240689754486084 lr:2.8697497757134705e-07 tokens_per_second_per_gpu:5010.355249015538 mfu:23.634700564361605 +Step 35070 | loss:0.2512640058994293 lr:2.8651124842630773e-07 tokens_per_second_per_gpu:5332.907001680594 mfu:25.1562322146865 +Step 35071 | loss:0.34336817264556885 lr:2.8604789318276147e-07 tokens_per_second_per_gpu:5018.190814250965 mfu:23.671662262462394 +Step 35072 | loss:0.334551066160202 lr:2.855849118441889e-07 tokens_per_second_per_gpu:4414.064097467535 mfu:20.821893464748356 +Step 35073 | loss:0.37907686829566956 lr:2.851223044140761e-07 tokens_per_second_per_gpu:4816.1749446400145 mfu:22.718718938046017 +Step 35074 | loss:0.34351643919944763 lr:2.846600708959035e-07 tokens_per_second_per_gpu:5327.555183694199 mfu:25.13098677613037 +Step 35075 | loss:0.29240337014198303 lr:2.841982112931463e-07 tokens_per_second_per_gpu:5335.019798517333 mfu:25.16619863784571 +Step 35076 | loss:0.16534225642681122 lr:2.837367256092793e-07 tokens_per_second_per_gpu:5354.77349880264 mfu:25.259380212420307 +Step 35077 | loss:0.24901622533798218 lr:2.832756138477666e-07 tokens_per_second_per_gpu:5348.829030117883 mfu:25.23133913940337 +Step 35078 | loss:0.29647526144981384 lr:2.8281487601208857e-07 tokens_per_second_per_gpu:4271.097021300447 mfu:20.14749337376971 +Step 35079 | loss:0.24907055497169495 lr:2.823545121057036e-07 tokens_per_second_per_gpu:4822.174298602834 mfu:22.74701891428397 +Step 35080 | loss:0.3596680760383606 lr:2.818945221320701e-07 tokens_per_second_per_gpu:3236.747153276797 mfu:15.268288568016803 +Step 35081 | loss:0.2482319176197052 lr:2.8143490609465197e-07 tokens_per_second_per_gpu:5189.110429452098 mfu:24.477919249259575 +Step 35082 | loss:0.29611095786094666 lr:2.8097566399690744e-07 tokens_per_second_per_gpu:3798.0007717795556 mfu:17.915817646236064 +Step 35083 | loss:0.269534170627594 lr:2.805167958422894e-07 tokens_per_second_per_gpu:4020.229769436202 mfu:18.96410974435926 +Step 35084 | loss:0.34131214022636414 lr:2.800583016342506e-07 tokens_per_second_per_gpu:4034.179267730244 mfu:19.029911907842088 +Step 35085 | loss:0.22039753198623657 lr:2.7960018137623833e-07 tokens_per_second_per_gpu:3908.0330101138916 mfu:18.434858487894843 +Step 35086 | loss:0.48672235012054443 lr:2.7914243507169423e-07 tokens_per_second_per_gpu:3902.4878820430304 mfu:18.408701172688183 +Step 35087 | loss:0.39137643575668335 lr:2.7868506272406557e-07 tokens_per_second_per_gpu:4413.529933005221 mfu:20.819373719841764 +Step 35088 | loss:0.35441604256629944 lr:2.782280643367885e-07 tokens_per_second_per_gpu:3969.24199032355 mfu:18.72359169087271 +Step 35089 | loss:0.37415024638175964 lr:2.7777143991330466e-07 tokens_per_second_per_gpu:3752.184030186952 mfu:17.699692259001562 +Step 35090 | loss:0.3343311548233032 lr:2.773151894570447e-07 tokens_per_second_per_gpu:4721.270412348411 mfu:22.27103806684361 +Step 35091 | loss:0.27378177642822266 lr:2.768593129714447e-07 tokens_per_second_per_gpu:3542.2489771677438 mfu:16.709392795296413 +Step 35092 | loss:0.36676034331321716 lr:2.764038104599298e-07 tokens_per_second_per_gpu:4137.577063064129 mfu:19.517656949915036 +Step 35093 | loss:0.25772514939308167 lr:2.759486819259249e-07 tokens_per_second_per_gpu:5357.586236000243 mfu:25.272648373683854 +Step 35094 | loss:0.2498053014278412 lr:2.754939273728607e-07 tokens_per_second_per_gpu:5347.662398146906 mfu:25.225835937348332 +Step 35095 | loss:0.3274076282978058 lr:2.7503954680414555e-07 tokens_per_second_per_gpu:4380.494524456877 mfu:20.66354006129732 +Step 35096 | loss:0.31373101472854614 lr:2.745855402232045e-07 tokens_per_second_per_gpu:5364.588720329146 mfu:25.30568028701032 +Step 35097 | loss:0.29358938336372375 lr:2.7413190763345143e-07 tokens_per_second_per_gpu:4120.974854498139 mfu:19.439341499480392 +Step 35098 | loss:0.2957123816013336 lr:2.7367864903829475e-07 tokens_per_second_per_gpu:4535.934555426472 mfu:21.39677720818492 +Step 35099 | loss:0.32699689269065857 lr:2.732257644411484e-07 tokens_per_second_per_gpu:5331.335924767141 mfu:25.14882117683238 +Step 35100 | loss:0.338631808757782 lr:2.727732538454153e-07 tokens_per_second_per_gpu:4935.839015407394 mfu:23.28319477665153 +Step 35101 | loss:0.3061440587043762 lr:2.7232111725449816e-07 tokens_per_second_per_gpu:4206.956023948427 mfu:19.844929345677617 +Step 35102 | loss:0.29136714339256287 lr:2.718693546717999e-07 tokens_per_second_per_gpu:4998.15671155023 mfu:23.57715798185264 +Step 35103 | loss:0.3798378109931946 lr:2.7141796610072324e-07 tokens_per_second_per_gpu:3495.7347539281163 mfu:16.489977268130254 +Step 35104 | loss:0.32066482305526733 lr:2.709669515446489e-07 tokens_per_second_per_gpu:4671.441696768291 mfu:22.035987513796645 +Step 35105 | loss:0.3012549579143524 lr:2.7051631100698524e-07 tokens_per_second_per_gpu:4503.1943910425725 mfu:21.242336266738008 +Step 35106 | loss:0.3364403545856476 lr:2.700660444911074e-07 tokens_per_second_per_gpu:3273.0266173472423 mfu:15.439424989951949 +Step 35107 | loss:0.32116201519966125 lr:2.696161520004126e-07 tokens_per_second_per_gpu:5326.531082455695 mfu:25.126155915858657 +Step 35108 | loss:0.2724025547504425 lr:2.6916663353828143e-07 tokens_per_second_per_gpu:5349.759554286784 mfu:25.23572858067187 +Step 35109 | loss:0.2307351529598236 lr:2.68717489108089e-07 tokens_per_second_per_gpu:5066.204792524806 mfu:23.898152390009535 +Step 35110 | loss:0.40584132075309753 lr:2.6826871871322156e-07 tokens_per_second_per_gpu:5101.655985444379 mfu:24.065381715608385 +Step 35111 | loss:0.26365983486175537 lr:2.6782032235704856e-07 tokens_per_second_per_gpu:5031.412146882438 mfu:23.73402954427701 +Step 35112 | loss:0.26346009969711304 lr:2.673723000429451e-07 tokens_per_second_per_gpu:5352.660786738681 mfu:25.24941418914877 +Step 35113 | loss:0.42509445548057556 lr:2.669246517742807e-07 tokens_per_second_per_gpu:4869.300257627953 mfu:22.969320103523184 +Step 35114 | loss:0.28505009412765503 lr:2.6647737755442494e-07 tokens_per_second_per_gpu:4086.1209435342266 mfu:19.274929654773203 +Step 35115 | loss:0.3277797996997833 lr:2.660304773867361e-07 tokens_per_second_per_gpu:5189.598275245294 mfu:24.480220501101154 +Step 35116 | loss:0.33568906784057617 lr:2.6558395127457834e-07 tokens_per_second_per_gpu:4433.47141208902 mfu:20.91344107904722 +Step 35117 | loss:0.3226521909236908 lr:2.6513779922131e-07 tokens_per_second_per_gpu:5349.576521520851 mfu:25.234865183886352 +Step 35118 | loss:0.289443701505661 lr:2.6469202123028946e-07 tokens_per_second_per_gpu:4901.806567177799 mfu:23.122657911818113 +Step 35119 | loss:0.2507813572883606 lr:2.642466173048641e-07 tokens_per_second_per_gpu:3755.1858098884895 mfu:17.713852166010142 +Step 35120 | loss:0.24971282482147217 lr:2.6380158744838124e-07 tokens_per_second_per_gpu:3879.88508871928 mfu:18.30207994014585 +Step 35121 | loss:0.21217501163482666 lr:2.633569316641993e-07 tokens_per_second_per_gpu:4099.092645462627 mfu:19.33611938597003 +Step 35122 | loss:0.3401760160923004 lr:2.6291264995565443e-07 tokens_per_second_per_gpu:5354.766436087329 mfu:25.259346896387164 +Step 35123 | loss:0.18801681697368622 lr:2.6246874232609406e-07 tokens_per_second_per_gpu:4108.925254374303 mfu:19.38250147982248 +Step 35124 | loss:0.29169729351997375 lr:2.6202520877884887e-07 tokens_per_second_per_gpu:4405.904459075887 mfu:20.783403058277084 +Step 35125 | loss:0.25700682401657104 lr:2.615820493172605e-07 tokens_per_second_per_gpu:3874.2254512394834 mfu:18.27538246451992 +Step 35126 | loss:0.311410516500473 lr:2.611392639446542e-07 tokens_per_second_per_gpu:4276.59119014125 mfu:20.173410305593645 +Step 35127 | loss:0.2694949507713318 lr:2.6069685266437163e-07 tokens_per_second_per_gpu:5377.731382018786 mfu:25.367676464569655 +Step 35128 | loss:0.29069194197654724 lr:2.602548154797324e-07 tokens_per_second_per_gpu:5197.395742084263 mfu:24.517002482565598 +Step 35129 | loss:0.32846710085868835 lr:2.5981315239406725e-07 tokens_per_second_per_gpu:3546.8565388289953 mfu:16.731127449783617 +Step 35130 | loss:0.35590583086013794 lr:2.5937186341069007e-07 tokens_per_second_per_gpu:3946.673769984942 mfu:18.617133545000048 +Step 35131 | loss:0.3280342221260071 lr:2.589309485329261e-07 tokens_per_second_per_gpu:5324.112620340685 mfu:25.11474761742997 +Step 35132 | loss:0.37987881898880005 lr:2.584904077640893e-07 tokens_per_second_per_gpu:5211.9334099259595 mfu:24.585579142156785 +Step 35133 | loss:0.17744919657707214 lr:2.5805024110749365e-07 tokens_per_second_per_gpu:4283.4817443525335 mfu:20.205914225458045 +Step 35134 | loss:0.2919967770576477 lr:2.5761044856644764e-07 tokens_per_second_per_gpu:5380.992723678481 mfu:25.383060769620705 +Step 35135 | loss:0.2458871304988861 lr:2.571710301442654e-07 tokens_per_second_per_gpu:4495.021015005313 mfu:21.203781057448502 +Step 35136 | loss:0.2769967019557953 lr:2.567319858442441e-07 tokens_per_second_per_gpu:3643.1617692189066 mfu:17.185415652899483 +Step 35137 | loss:0.3481822907924652 lr:2.5629331566968673e-07 tokens_per_second_per_gpu:3975.490820903541 mfu:18.75306849087896 +Step 35138 | loss:0.25038057565689087 lr:2.5585501962389625e-07 tokens_per_second_per_gpu:3024.0440917528244 mfu:14.264931935923734 +Step 35139 | loss:0.4226810038089752 lr:2.5541709771017e-07 tokens_per_second_per_gpu:5335.293644135872 mfu:25.16749041435124 +Step 35140 | loss:0.2413405478000641 lr:2.549795499317942e-07 tokens_per_second_per_gpu:5348.674228184286 mfu:25.23060891227066 +Step 35141 | loss:0.3711206316947937 lr:2.545423762920662e-07 tokens_per_second_per_gpu:5312.67042779486 mfu:25.06077284295196 +Step 35142 | loss:0.3913308084011078 lr:2.5410557679427236e-07 tokens_per_second_per_gpu:4671.566844670307 mfu:22.036577857802882 +Step 35143 | loss:0.3657394051551819 lr:2.536691514416989e-07 tokens_per_second_per_gpu:5126.709114649378 mfu:24.18356160057298 +Step 35144 | loss:0.2665303349494934 lr:2.532331002376265e-07 tokens_per_second_per_gpu:4334.099472681586 mfu:20.444686686261186 +Step 35145 | loss:0.38122016191482544 lr:2.527974231853414e-07 tokens_per_second_per_gpu:4005.6389780468053 mfu:18.89528249192976 +Step 35146 | loss:0.2560241222381592 lr:2.523621202881077e-07 tokens_per_second_per_gpu:4842.390834930651 mfu:22.842383765439063 +Step 35147 | loss:0.3490472137928009 lr:2.5192719154921164e-07 tokens_per_second_per_gpu:4900.996931410818 mfu:23.11883872176716 +Step 35148 | loss:0.27873051166534424 lr:2.514926369719117e-07 tokens_per_second_per_gpu:4324.7441770930845 mfu:20.400556160792203 +Step 35149 | loss:0.3532751798629761 lr:2.510584565594887e-07 tokens_per_second_per_gpu:4590.611738879776 mfu:21.654698811423614 +Step 35150 | loss:0.2538471817970276 lr:2.5062465031520657e-07 tokens_per_second_per_gpu:4977.51103185853 mfu:23.479768788230707 +Step 35151 | loss:0.2509678602218628 lr:2.501912182423183e-07 tokens_per_second_per_gpu:5338.715660497889 mfu:25.18363264938707 +Step 35152 | loss:0.3089864253997803 lr:2.497581603440935e-07 tokens_per_second_per_gpu:5348.271723302036 mfu:25.22871022806667 +Step 35153 | loss:0.26612162590026855 lr:2.493254766237796e-07 tokens_per_second_per_gpu:4199.805434477228 mfu:19.811198795125605 +Step 35154 | loss:0.2896519601345062 lr:2.488931670846462e-07 tokens_per_second_per_gpu:5380.3443527456775 mfu:25.380002293307182 +Step 35155 | loss:0.3605814576148987 lr:2.4846123172992954e-07 tokens_per_second_per_gpu:3934.271266535961 mfu:18.558628820145913 +Step 35156 | loss:0.39594417810440063 lr:2.480296705628826e-07 tokens_per_second_per_gpu:4219.987396879012 mfu:19.906400555172674 +Step 35157 | loss:0.33199381828308105 lr:2.4759848358675843e-07 tokens_per_second_per_gpu:3722.469191336742 mfu:17.559522294270824 +Step 35158 | loss:0.3542698919773102 lr:2.471676708047932e-07 tokens_per_second_per_gpu:5232.399688377143 mfu:24.682122069518197 +Step 35159 | loss:0.3724985718727112 lr:2.4673723222022885e-07 tokens_per_second_per_gpu:3791.1233470764973 mfu:17.883375660502086 +Step 35160 | loss:0.2832604646682739 lr:2.463071678363016e-07 tokens_per_second_per_gpu:5327.143210465623 mfu:25.1290434281213 +Step 35161 | loss:0.22346965968608856 lr:2.458774776562478e-07 tokens_per_second_per_gpu:4444.3084745172755 mfu:20.96456135151481 +Step 35162 | loss:0.2862827777862549 lr:2.4544816168329265e-07 tokens_per_second_per_gpu:4665.204722030498 mfu:22.00656663982085 +Step 35163 | loss:0.25246769189834595 lr:2.4501921992067244e-07 tokens_per_second_per_gpu:5317.312620107367 mfu:25.082670856129734 +Step 35164 | loss:0.27948611974716187 lr:2.4459065237161795e-07 tokens_per_second_per_gpu:5222.440594025627 mfu:24.63514332226538 +Step 35165 | loss:0.20916076004505157 lr:2.4416245903933765e-07 tokens_per_second_per_gpu:3950.1036783038035 mfu:18.633313007742824 +Step 35166 | loss:0.33543556928634644 lr:2.437346399270679e-07 tokens_per_second_per_gpu:4798.013031193848 mfu:22.633046093578482 +Step 35167 | loss:0.23550738394260406 lr:2.433071950380117e-07 tokens_per_second_per_gpu:4131.970024208054 mfu:19.49120759097146 +Step 35168 | loss:0.27039024233818054 lr:2.4288012437539975e-07 tokens_per_second_per_gpu:5319.656521435111 mfu:25.093727438603505 +Step 35169 | loss:0.3593319058418274 lr:2.4245342794242955e-07 tokens_per_second_per_gpu:3237.555373663081 mfu:15.272101081475522 +Step 35170 | loss:0.28417032957077026 lr:2.4202710574231515e-07 tokens_per_second_per_gpu:5321.903065530664 mfu:25.104324770402176 +Step 35171 | loss:0.21305622160434723 lr:2.416011577782651e-07 tokens_per_second_per_gpu:4429.613339160302 mfu:20.895241890787332 +Step 35172 | loss:0.2516087293624878 lr:2.411755840534824e-07 tokens_per_second_per_gpu:5322.8763251134305 mfu:25.108915801909294 +Step 35173 | loss:0.31489846110343933 lr:2.4075038457117004e-07 tokens_per_second_per_gpu:3596.4244184588238 mfu:16.964947595149855 +Step 35174 | loss:0.3115689754486084 lr:2.4032555933451996e-07 tokens_per_second_per_gpu:3588.117224385512 mfu:16.925761143352585 +Step 35175 | loss:0.2008107602596283 lr:2.399011083467351e-07 tokens_per_second_per_gpu:4354.628367250484 mfu:20.541524984533265 +Step 35176 | loss:0.3265798091888428 lr:2.394770316109962e-07 tokens_per_second_per_gpu:4331.394543336484 mfu:20.431927073031805 +Step 35177 | loss:0.34158116579055786 lr:2.3905332913050636e-07 tokens_per_second_per_gpu:5011.260354090995 mfu:23.638970099429383 +Step 35178 | loss:0.31894367933273315 lr:2.386300009084408e-07 tokens_per_second_per_gpu:4819.920272893908 mfu:22.736386290438528 +Step 35179 | loss:0.3042893409729004 lr:2.3820704694799135e-07 tokens_per_second_per_gpu:4173.888783649311 mfu:19.688945531333722 +Step 35180 | loss:0.2681029736995697 lr:2.3778446725233327e-07 tokens_per_second_per_gpu:5077.2470849468655 mfu:23.950240767373753 +Step 35181 | loss:0.4436715245246887 lr:2.3736226182465293e-07 tokens_per_second_per_gpu:4484.9310007601225 mfu:21.156184738719993 +Step 35182 | loss:0.29222264885902405 lr:2.3694043066812e-07 tokens_per_second_per_gpu:3036.327527583742 mfu:14.322874998508432 +Step 35183 | loss:0.24283239245414734 lr:2.365189737859097e-07 tokens_per_second_per_gpu:5045.314477731474 mfu:23.79960921087415 +Step 35184 | loss:0.26185113191604614 lr:2.360978911811862e-07 tokens_per_second_per_gpu:5034.306636246189 mfu:23.747683344456636 +Step 35185 | loss:0.30477309226989746 lr:2.3567718285711915e-07 tokens_per_second_per_gpu:5358.8593131875305 mfu:25.278653695985657 +Step 35186 | loss:0.33038103580474854 lr:2.3525684881687826e-07 tokens_per_second_per_gpu:4582.001808958626 mfu:21.614084302980025 +Step 35187 | loss:0.3318459689617157 lr:2.3483688906361656e-07 tokens_per_second_per_gpu:4240.364410565717 mfu:20.002522405409902 +Step 35188 | loss:0.28067123889923096 lr:2.3441730360049817e-07 tokens_per_second_per_gpu:5324.050056160205 mfu:25.114452491516097 +Step 35189 | loss:0.308621346950531 lr:2.3399809243067617e-07 tokens_per_second_per_gpu:5347.488711387446 mfu:25.22501662726994 +Step 35190 | loss:0.21316879987716675 lr:2.335792555573091e-07 tokens_per_second_per_gpu:4688.260824171317 mfu:22.115326207395757 +Step 35191 | loss:0.2853260636329651 lr:2.331607929835389e-07 tokens_per_second_per_gpu:5331.3416329304455 mfu:25.148848103212604 +Step 35192 | loss:0.26664605736732483 lr:2.3274270471251858e-07 tokens_per_second_per_gpu:5361.168852301512 mfu:25.289548186034313 +Step 35193 | loss:0.2894515097141266 lr:2.323249907473901e-07 tokens_per_second_per_gpu:4354.401323777553 mfu:20.54045398173384 +Step 35194 | loss:0.24802334606647491 lr:2.3190765109129543e-07 tokens_per_second_per_gpu:4014.7960411624067 mfu:18.938477921002125 +Step 35195 | loss:0.3525039255619049 lr:2.314906857473764e-07 tokens_per_second_per_gpu:3825.316377164169 mfu:18.044670017358143 +Step 35196 | loss:0.5190085172653198 lr:2.310740947187584e-07 tokens_per_second_per_gpu:4190.436547348013 mfu:19.767004156040123 +Step 35197 | loss:0.35557085275650024 lr:2.306578780085944e-07 tokens_per_second_per_gpu:5329.108023552051 mfu:25.13831178668166 +Step 35198 | loss:0.1509183645248413 lr:2.3024203561999302e-07 tokens_per_second_per_gpu:5346.540193955541 mfu:25.220542308709884 +Step 35199 | loss:0.1844717413187027 lr:2.2982656755609622e-07 tokens_per_second_per_gpu:4593.526559492995 mfu:21.66844851757585 +Step 35200 | loss:0.3754383623600006 lr:2.294114738200237e-07 tokens_per_second_per_gpu:3429.042562948138 mfu:16.175378824416565 +Step 35201 | loss:0.2580703794956207 lr:2.2899675441490075e-07 tokens_per_second_per_gpu:5057.212465786653 mfu:23.855734050536523 +Step 35202 | loss:0.3451002240180969 lr:2.2858240934384157e-07 tokens_per_second_per_gpu:4884.51217018388 mfu:23.0410772904694 +Step 35203 | loss:0.3332398533821106 lr:2.2816843860996585e-07 tokens_per_second_per_gpu:5317.263021381138 mfu:25.082436890475698 +Step 35204 | loss:0.2883126139640808 lr:2.2775484221638777e-07 tokens_per_second_per_gpu:4239.515017684373 mfu:19.99851567426723 +Step 35205 | loss:0.34606584906578064 lr:2.273416201662104e-07 tokens_per_second_per_gpu:5361.284130878643 mfu:25.290091974751192 +Step 35206 | loss:0.3486590087413788 lr:2.269287724625535e-07 tokens_per_second_per_gpu:4971.984904506356 mfu:23.45370110265595 +Step 35207 | loss:0.315172016620636 lr:2.2651629910851457e-07 tokens_per_second_per_gpu:3971.071072192931 mfu:18.732219782124126 +Step 35208 | loss:0.2816920876502991 lr:2.2610420010719669e-07 tokens_per_second_per_gpu:5311.78095540749 mfu:25.056577049941097 +Step 35209 | loss:0.361591100692749 lr:2.2569247546169737e-07 tokens_per_second_per_gpu:4916.14957335067 mfu:23.190316319104745 +Step 35210 | loss:0.19836923480033875 lr:2.2528112517512522e-07 tokens_per_second_per_gpu:5259.001729589768 mfu:24.8076084366944 +Step 35211 | loss:0.29790735244750977 lr:2.2487014925056115e-07 tokens_per_second_per_gpu:4505.482820739467 mfu:21.253131180064624 +Step 35212 | loss:0.3714054524898529 lr:2.244595476910971e-07 tokens_per_second_per_gpu:4381.168634483544 mfu:20.66671995330814 +Step 35213 | loss:0.3323732614517212 lr:2.2404932049982508e-07 tokens_per_second_per_gpu:4364.76217392302 mfu:20.589327879613325 +Step 35214 | loss:0.35262641310691833 lr:2.2363946767983146e-07 tokens_per_second_per_gpu:5361.374219153185 mfu:25.290516936512848 +Step 35215 | loss:0.273952454328537 lr:2.2322998923419713e-07 tokens_per_second_per_gpu:3939.6856395388018 mfu:18.584169341387607 +Step 35216 | loss:0.33158475160598755 lr:2.2282088516599743e-07 tokens_per_second_per_gpu:4582.653779134617 mfu:21.61715975753776 +Step 35217 | loss:0.3952232003211975 lr:2.2241215547831874e-07 tokens_per_second_per_gpu:3632.219171620631 mfu:17.13379755302875 +Step 35218 | loss:0.32390639185905457 lr:2.2200380017423084e-07 tokens_per_second_per_gpu:4296.243146285106 mfu:20.266111935693683 +Step 35219 | loss:0.22998248040676117 lr:2.2159581925679794e-07 tokens_per_second_per_gpu:4169.14032354511 mfu:19.66654623485137 +Step 35220 | loss:0.29524070024490356 lr:2.211882127290954e-07 tokens_per_second_per_gpu:4508.84179347676 mfu:21.268976027566286 +Step 35221 | loss:0.2860547602176666 lr:2.207809805941874e-07 tokens_per_second_per_gpu:5324.32846024482 mfu:25.11576577108386 +Step 35222 | loss:0.4121459722518921 lr:2.203741228551437e-07 tokens_per_second_per_gpu:5338.615568222078 mfu:25.183160497044565 +Step 35223 | loss:0.33230650424957275 lr:2.1996763951501188e-07 tokens_per_second_per_gpu:4985.312264880988 mfu:23.516568535425243 +Step 35224 | loss:0.32736825942993164 lr:2.1956153057686168e-07 tokens_per_second_per_gpu:5351.322272522081 mfu:25.243100189214992 +Step 35225 | loss:0.32911956310272217 lr:2.191557960437407e-07 tokens_per_second_per_gpu:3994.2308469943537 mfu:18.841468391327478 +Step 35226 | loss:0.40736934542655945 lr:2.1875043591869647e-07 tokens_per_second_per_gpu:4408.271458561041 mfu:20.794568598700312 +Step 35227 | loss:0.2359190732240677 lr:2.1834545020478213e-07 tokens_per_second_per_gpu:5349.438209629934 mfu:25.23421274309846 +Step 35228 | loss:0.34913185238838196 lr:2.1794083890504523e-07 tokens_per_second_per_gpu:5323.48957866702 mfu:25.111808623553422 +Step 35229 | loss:0.39518681168556213 lr:2.175366020225278e-07 tokens_per_second_per_gpu:4532.816286514972 mfu:21.382067801697836 +Step 35230 | loss:0.433938205242157 lr:2.171327395602718e-07 tokens_per_second_per_gpu:4323.43352550354 mfu:20.394373593624103 +Step 35231 | loss:0.2955373227596283 lr:2.1672925152131373e-07 tokens_per_second_per_gpu:5357.574144105328 mfu:25.272591334152448 +Step 35232 | loss:0.38561299443244934 lr:2.163261379086845e-07 tokens_per_second_per_gpu:4670.413777149136 mfu:22.031138641571914 +Step 35233 | loss:0.35251230001449585 lr:2.1592339872542054e-07 tokens_per_second_per_gpu:4272.391738383625 mfu:20.153600775152576 +Step 35234 | loss:0.364147812128067 lr:2.1552103397454725e-07 tokens_per_second_per_gpu:5366.193111055288 mfu:25.31324847180281 +Step 35235 | loss:0.2903534471988678 lr:2.1511904365908997e-07 tokens_per_second_per_gpu:4046.858247735018 mfu:19.089720819781657 +Step 35236 | loss:0.27545487880706787 lr:2.147174277820796e-07 tokens_per_second_per_gpu:5336.987653713463 mfu:25.175481346556573 +Step 35237 | loss:0.3019886016845703 lr:2.143161863465304e-07 tokens_per_second_per_gpu:5361.873686656883 mfu:25.292873009945744 +Step 35238 | loss:0.36928120255470276 lr:2.1391531935546215e-07 tokens_per_second_per_gpu:4105.59149451645 mfu:19.366775565774905 +Step 35239 | loss:0.32474175095558167 lr:2.1351482681188917e-07 tokens_per_second_per_gpu:4982.420367410328 mfu:23.502926961647557 +Step 35240 | loss:0.24975357949733734 lr:2.1311470871882566e-07 tokens_per_second_per_gpu:4404.902643344588 mfu:20.778677322545157 +Step 35241 | loss:0.3160128891468048 lr:2.1271496507928034e-07 tokens_per_second_per_gpu:5100.298908756064 mfu:24.05898015332844 +Step 35242 | loss:0.2860313057899475 lr:2.1231559589625638e-07 tokens_per_second_per_gpu:5100.8634372700535 mfu:24.06164313064743 +Step 35243 | loss:0.30976179242134094 lr:2.119166011727569e-07 tokens_per_second_per_gpu:5320.056907342895 mfu:25.095616127243403 +Step 35244 | loss:0.2954649031162262 lr:2.1151798091179064e-07 tokens_per_second_per_gpu:4421.660632894756 mfu:20.85772761846115 +Step 35245 | loss:0.35141199827194214 lr:2.111197351163441e-07 tokens_per_second_per_gpu:3739.3578275008085 mfu:17.639188872555184 +Step 35246 | loss:0.22149302065372467 lr:2.1072186378942592e-07 tokens_per_second_per_gpu:3888.8206560126964 mfu:18.344230535633884 +Step 35247 | loss:0.22518934309482574 lr:2.1032436693402268e-07 tokens_per_second_per_gpu:5403.877036565004 mfu:25.49100997797283 +Step 35248 | loss:0.3043949007987976 lr:2.0992724455312086e-07 tokens_per_second_per_gpu:5361.685511603101 mfu:25.291985356111073 +Step 35249 | loss:0.38036447763442993 lr:2.0953049664970693e-07 tokens_per_second_per_gpu:5341.599826010461 mfu:25.197237750199253 +Step 35250 | loss:0.3920270502567291 lr:2.091341232267674e-07 tokens_per_second_per_gpu:3455.4698268902916 mfu:16.30004073738788 +Step 35251 | loss:0.23167695105075836 lr:2.0873812428728324e-07 tokens_per_second_per_gpu:4773.82219973194 mfu:22.51893381419121 +Step 35252 | loss:0.3986200988292694 lr:2.0834249983423538e-07 tokens_per_second_per_gpu:4502.360785413517 mfu:21.238404006757964 +Step 35253 | loss:0.3125222325325012 lr:2.079472498705992e-07 tokens_per_second_per_gpu:5328.869563857452 mfu:25.137186931616814 +Step 35254 | loss:0.39303866028785706 lr:2.075523743993446e-07 tokens_per_second_per_gpu:4219.669509960574 mfu:19.904901028341303 +Step 35255 | loss:0.27817004919052124 lr:2.0715787342343584e-07 tokens_per_second_per_gpu:5297.351801449486 mfu:24.98851226885366 +Step 35256 | loss:0.30866512656211853 lr:2.067637469458539e-07 tokens_per_second_per_gpu:5345.398801223072 mfu:25.21515816444915 +Step 35257 | loss:0.26360172033309937 lr:2.063699949695519e-07 tokens_per_second_per_gpu:2775.0466532464616 mfu:13.090368535145466 +Step 35258 | loss:0.3471173346042633 lr:2.059766174974942e-07 tokens_per_second_per_gpu:4812.380845511452 mfu:22.700821525116528 +Step 35259 | loss:0.29752540588378906 lr:2.0558361453263952e-07 tokens_per_second_per_gpu:5010.582279043064 mfu:23.63577150373676 +Step 35260 | loss:0.31505340337753296 lr:2.0519098607794662e-07 tokens_per_second_per_gpu:4395.396571451117 mfu:20.733835559520546 +Step 35261 | loss:0.3967619836330414 lr:2.047987321363687e-07 tokens_per_second_per_gpu:4618.664354436386 mfu:21.787027785273906 +Step 35262 | loss:0.2919429838657379 lr:2.0440685271084781e-07 tokens_per_second_per_gpu:4685.165321614673 mfu:22.100724193688873 +Step 35263 | loss:0.21395917236804962 lr:2.0401534780433717e-07 tokens_per_second_per_gpu:5345.203634323636 mfu:25.2142375288856 +Step 35264 | loss:0.4164033532142639 lr:2.0362421741978444e-07 tokens_per_second_per_gpu:4864.115650118507 mfu:22.944863425315866 +Step 35265 | loss:0.2756635546684265 lr:2.0323346156012611e-07 tokens_per_second_per_gpu:5027.170918652751 mfu:23.71402294708943 +Step 35266 | loss:0.3545701205730438 lr:2.0284308022830435e-07 tokens_per_second_per_gpu:3902.9160279820285 mfu:18.410720810131668 +Step 35267 | loss:0.33369776606559753 lr:2.0245307342725561e-07 tokens_per_second_per_gpu:5329.903389462286 mfu:25.142063663383656 +Step 35268 | loss:0.32871225476264954 lr:2.0206344115991094e-07 tokens_per_second_per_gpu:4628.111649463897 mfu:21.83159228779342 +Step 35269 | loss:0.38157179951667786 lr:2.0167418342919576e-07 tokens_per_second_per_gpu:4386.129903139093 mfu:20.690123104036076 +Step 35270 | loss:0.3548685610294342 lr:2.012853002380466e-07 tokens_per_second_per_gpu:5333.228791720128 mfu:25.157750153206027 +Step 35271 | loss:0.33209750056266785 lr:2.008967915893889e-07 tokens_per_second_per_gpu:4523.737054385966 mfu:21.339239514668577 +Step 35272 | loss:0.2714323103427887 lr:2.00508657486137e-07 tokens_per_second_per_gpu:5167.852830861197 mfu:24.377643529785978 +Step 35273 | loss:0.2999195456504822 lr:2.0012089793121637e-07 tokens_per_second_per_gpu:5299.300147906536 mfu:24.99770295151375 +Step 35274 | loss:0.3483095169067383 lr:1.9973351292753574e-07 tokens_per_second_per_gpu:5211.376494250101 mfu:24.58295207589415 +Step 35275 | loss:0.3323166072368622 lr:1.9934650247802057e-07 tokens_per_second_per_gpu:3772.478876509826 mfu:17.795426511764397 +Step 35276 | loss:0.33457183837890625 lr:1.989598665855741e-07 tokens_per_second_per_gpu:5336.86013190605 mfu:25.174879804433544 +Step 35277 | loss:0.265099436044693 lr:1.9857360525309954e-07 tokens_per_second_per_gpu:3918.603644708608 mfu:18.48472197481436 +Step 35278 | loss:0.24486221373081207 lr:1.9818771848351125e-07 tokens_per_second_per_gpu:3892.5599730554372 mfu:18.36186953211777 +Step 35279 | loss:0.3392057418823242 lr:1.9780220627970692e-07 tokens_per_second_per_gpu:5354.611474463929 mfu:25.258615915970406 +Step 35280 | loss:0.2903277575969696 lr:1.9741706864458975e-07 tokens_per_second_per_gpu:3836.7488780801314 mfu:18.098599074764532 +Step 35281 | loss:0.2431354522705078 lr:1.9703230558104635e-07 tokens_per_second_per_gpu:4106.644443377103 mfu:19.371742505202203 +Step 35282 | loss:0.3554236590862274 lr:1.966479170919855e-07 tokens_per_second_per_gpu:3738.944791245188 mfu:17.63724051006617 +Step 35283 | loss:0.34158656001091003 lr:1.962639031802882e-07 tokens_per_second_per_gpu:4628.421520009213 mfu:21.83305400002504 +Step 35284 | loss:0.3046658933162689 lr:1.958802638488466e-07 tokens_per_second_per_gpu:3463.251077200734 mfu:16.33674622272044 +Step 35285 | loss:0.25660088658332825 lr:1.9549699910054177e-07 tokens_per_second_per_gpu:3578.895615240367 mfu:16.882261239646173 +Step 35286 | loss:0.3252303898334503 lr:1.951141089382602e-07 tokens_per_second_per_gpu:5393.09749754676 mfu:25.440161053244765 +Step 35287 | loss:0.3205183446407318 lr:1.94731593364883e-07 tokens_per_second_per_gpu:4381.154594554512 mfu:20.666653724567436 +Step 35288 | loss:0.2252882719039917 lr:1.9434945238328007e-07 tokens_per_second_per_gpu:4032.517735018798 mfu:19.02207417455548 +Step 35289 | loss:0.26705753803253174 lr:1.939676859963324e-07 tokens_per_second_per_gpu:4547.030035800975 mfu:21.449116482195613 +Step 35290 | loss:0.3321698009967804 lr:1.9358629420690998e-07 tokens_per_second_per_gpu:5189.250400062623 mfu:24.478579514510873 +Step 35291 | loss:0.3295983672142029 lr:1.932052770178827e-07 tokens_per_second_per_gpu:4765.046942260851 mfu:22.477539427486775 +Step 35292 | loss:0.31495746970176697 lr:1.9282463443210941e-07 tokens_per_second_per_gpu:5318.622670920455 mfu:25.088850589332786 +Step 35293 | loss:0.3158240020275116 lr:1.9244436645246e-07 tokens_per_second_per_gpu:3576.12745394969 mfu:16.869203350541035 +Step 35294 | loss:0.3961363136768341 lr:1.9206447308178777e-07 tokens_per_second_per_gpu:4558.782623916377 mfu:21.504555445534265 +Step 35295 | loss:0.32511669397354126 lr:1.9168495432295707e-07 tokens_per_second_per_gpu:5343.350666691441 mfu:25.205496764415656 +Step 35296 | loss:0.31584087014198303 lr:1.913058101788212e-07 tokens_per_second_per_gpu:4385.05309152503 mfu:20.68504359992954 +Step 35297 | loss:0.45986244082450867 lr:1.9092704065222788e-07 tokens_per_second_per_gpu:4652.705738912562 mfu:21.947606803907128 +Step 35298 | loss:0.3159869909286499 lr:1.9054864574603039e-07 tokens_per_second_per_gpu:3852.7570115059825 mfu:18.174112171403475 +Step 35299 | loss:0.26357758045196533 lr:1.901706254630653e-07 tokens_per_second_per_gpu:3589.3541256443705 mfu:16.931595817627734 +Step 35300 | loss:0.2827930748462677 lr:1.897929798061915e-07 tokens_per_second_per_gpu:5357.787430233019 mfu:25.273597441207137 +Step 35301 | loss:0.33039429783821106 lr:1.8941570877823445e-07 tokens_per_second_per_gpu:3869.8626257850933 mfu:18.254802272476784 +Step 35302 | loss:0.3580021858215332 lr:1.890388123820419e-07 tokens_per_second_per_gpu:4528.942220223863 mfu:21.363793170018806 +Step 35303 | loss:0.204559326171875 lr:1.886622906204394e-07 tokens_per_second_per_gpu:5098.93255273587 mfu:24.05253481885681 +Step 35304 | loss:0.29440543055534363 lr:1.8828614349626905e-07 tokens_per_second_per_gpu:4880.516745277759 mfu:23.022230189497442 +Step 35305 | loss:0.32904359698295593 lr:1.8791037101235643e-07 tokens_per_second_per_gpu:4879.581139295314 mfu:23.01781677644741 +Step 35306 | loss:0.2825404703617096 lr:1.875349731715159e-07 tokens_per_second_per_gpu:3578.1894413094724 mfu:16.878930096728446 +Step 35307 | loss:0.27862727642059326 lr:1.871599499765897e-07 tokens_per_second_per_gpu:3945.9684699871605 mfu:18.613806524573924 +Step 35308 | loss:0.3077617287635803 lr:1.8678530143038665e-07 tokens_per_second_per_gpu:4294.493635475156 mfu:20.25785919470582 +Step 35309 | loss:0.321363240480423 lr:1.864110275357267e-07 tokens_per_second_per_gpu:5302.751484552909 mfu:25.013983495333445 +Step 35310 | loss:0.3737429976463318 lr:1.8603712829542985e-07 tokens_per_second_per_gpu:5325.873979283788 mfu:25.123056248083408 +Step 35311 | loss:0.3212405741214752 lr:1.8566360371229942e-07 tokens_per_second_per_gpu:2878.515877942361 mfu:13.578450521706145 +Step 35312 | loss:0.2263820767402649 lr:1.8529045378914978e-07 tokens_per_second_per_gpu:5370.034814568755 mfu:25.33137044273804 +Step 35313 | loss:0.4394654333591461 lr:1.8491767852878982e-07 tokens_per_second_per_gpu:5362.654143736816 mfu:25.296554559151495 +Step 35314 | loss:0.30427050590515137 lr:1.8454527793401732e-07 tokens_per_second_per_gpu:5321.528549864206 mfu:25.102558116104532 +Step 35315 | loss:0.2589784264564514 lr:1.8417325200764112e-07 tokens_per_second_per_gpu:4742.408376981638 mfu:22.3707494525272 +Step 35316 | loss:0.25568246841430664 lr:1.838016007524479e-07 tokens_per_second_per_gpu:3822.3045412366096 mfu:18.03046267341565 +Step 35317 | loss:0.3321269750595093 lr:1.8343032417124095e-07 tokens_per_second_per_gpu:4436.9390047801 mfu:20.92979830540336 +Step 35318 | loss:0.42515718936920166 lr:1.8305942226681805e-07 tokens_per_second_per_gpu:4482.293960853252 mfu:21.14374537155388 +Step 35319 | loss:0.33633142709732056 lr:1.8268889504195475e-07 tokens_per_second_per_gpu:4416.995475053064 mfu:20.835721272963887 +Step 35320 | loss:0.2821827530860901 lr:1.823187424994488e-07 tokens_per_second_per_gpu:5023.885823303527 mfu:23.698526591832074 +Step 35321 | loss:0.43376967310905457 lr:1.819489646420758e-07 tokens_per_second_per_gpu:5335.419260970342 mfu:25.16808297039188 +Step 35322 | loss:0.3010376989841461 lr:1.8157956147262788e-07 tokens_per_second_per_gpu:5309.101103113917 mfu:25.043935729442364 +Step 35323 | loss:0.32119590044021606 lr:1.812105329938696e-07 tokens_per_second_per_gpu:5180.610720230894 mfu:24.437824670663222 +Step 35324 | loss:0.37385591864585876 lr:1.8084187920858754e-07 tokens_per_second_per_gpu:5330.036340915388 mfu:25.142690818071795 +Step 35325 | loss:0.2641748785972595 lr:1.8047360011955172e-07 tokens_per_second_per_gpu:3609.2934096892304 mfu:17.02565282246 +Step 35326 | loss:0.3004114329814911 lr:1.8010569572953218e-07 tokens_per_second_per_gpu:4149.054668841092 mfu:19.57179877464645 +Step 35327 | loss:0.1980084627866745 lr:1.7973816604129334e-07 tokens_per_second_per_gpu:5346.603347085053 mfu:25.22084021279683 +Step 35328 | loss:0.3548840880393982 lr:1.7937101105759968e-07 tokens_per_second_per_gpu:5368.914525874492 mfu:25.326085849826246 +Step 35329 | loss:0.281055748462677 lr:1.7900423078121565e-07 tokens_per_second_per_gpu:5362.659673954337 mfu:25.296580646131094 +Step 35330 | loss:0.28314483165740967 lr:1.7863782521489458e-07 tokens_per_second_per_gpu:5234.948536948342 mfu:24.69414542310695 +Step 35331 | loss:0.3438618779182434 lr:1.782717943613954e-07 tokens_per_second_per_gpu:5301.117174346877 mfu:25.00627417524952 +Step 35332 | loss:0.3502340316772461 lr:1.7790613822347146e-07 tokens_per_second_per_gpu:3220.380183801551 mfu:15.191082780509124 +Step 35333 | loss:0.3043765127658844 lr:1.7754085680388167e-07 tokens_per_second_per_gpu:4395.878942313593 mfu:20.736110985179803 +Step 35334 | loss:0.3072396516799927 lr:1.7717595010535714e-07 tokens_per_second_per_gpu:4173.1661328434975 mfu:19.685536664185758 +Step 35335 | loss:0.24503128230571747 lr:1.7681141813064571e-07 tokens_per_second_per_gpu:4434.139065871344 mfu:20.916590515851922 +Step 35336 | loss:0.31776779890060425 lr:1.7644726088250074e-07 tokens_per_second_per_gpu:4706.636374704804 mfu:22.202006814454432 +Step 35337 | loss:0.34417107701301575 lr:1.760834783636478e-07 tokens_per_second_per_gpu:5375.460119155286 mfu:25.356962530125124 +Step 35338 | loss:0.3835354745388031 lr:1.757200705768347e-07 tokens_per_second_per_gpu:3760.99415739598 mfu:17.741251132209246 +Step 35339 | loss:0.27640119194984436 lr:1.753570375247815e-07 tokens_per_second_per_gpu:5348.210498865235 mfu:25.228421421952316 +Step 35340 | loss:0.3386889696121216 lr:1.7499437921023044e-07 tokens_per_second_per_gpu:5352.281418569145 mfu:25.247624644766184 +Step 35341 | loss:0.3341504633426666 lr:1.746320956359071e-07 tokens_per_second_per_gpu:4678.986568673747 mfu:22.071577961862562 +Step 35342 | loss:0.27801427245140076 lr:1.74270186804526e-07 tokens_per_second_per_gpu:3628.799840897659 mfu:17.11766798661051 +Step 35343 | loss:0.2607463598251343 lr:1.739086527188183e-07 tokens_per_second_per_gpu:5291.90441204279 mfu:24.96281600360181 +Step 35344 | loss:0.2689287066459656 lr:1.73547493381504e-07 tokens_per_second_per_gpu:5335.770018461672 mfu:25.169737553323884 +Step 35345 | loss:0.3134932219982147 lr:1.7318670879529763e-07 tokens_per_second_per_gpu:5377.70380719071 mfu:25.367546389400875 +Step 35346 | loss:0.23676812648773193 lr:1.7282629896290815e-07 tokens_per_second_per_gpu:4583.719942013601 mfu:21.622189029744074 +Step 35347 | loss:0.21354010701179504 lr:1.7246626388705555e-07 tokens_per_second_per_gpu:4365.172249697106 mfu:20.591262277005832 +Step 35348 | loss:0.2794218361377716 lr:1.7210660357044327e-07 tokens_per_second_per_gpu:5346.722599752302 mfu:25.221402747974842 +Step 35349 | loss:0.31015175580978394 lr:1.7174731801576916e-07 tokens_per_second_per_gpu:5360.441331222506 mfu:25.286116344976747 +Step 35350 | loss:0.3317829966545105 lr:1.713884072257421e-07 tokens_per_second_per_gpu:3578.1352450798013 mfu:16.87867444386626 +Step 35351 | loss:0.2974533140659332 lr:1.7102987120306556e-07 tokens_per_second_per_gpu:5384.586825092252 mfu:25.4000147592064 +Step 35352 | loss:0.3941507935523987 lr:1.7067170995042625e-07 tokens_per_second_per_gpu:4188.595006680867 mfu:19.758317294513958 +Step 35353 | loss:0.26728692650794983 lr:1.70313923470522e-07 tokens_per_second_per_gpu:5359.632905202685 mfu:25.28230286150955 +Step 35354 | loss:0.21181994676589966 lr:1.6995651176605066e-07 tokens_per_second_per_gpu:3889.4411932560965 mfu:18.347157715685572 +Step 35355 | loss:0.33790481090545654 lr:1.6959947483969341e-07 tokens_per_second_per_gpu:4614.656865266999 mfu:21.768123775113747 +Step 35356 | loss:0.28470084071159363 lr:1.69242812694137e-07 tokens_per_second_per_gpu:3869.104755439916 mfu:18.251227268752565 +Step 35357 | loss:0.26905596256256104 lr:1.688865253320626e-07 tokens_per_second_per_gpu:4056.83889961487 mfu:19.136801257573726 +Step 35358 | loss:0.3407249450683594 lr:1.6853061275615146e-07 tokens_per_second_per_gpu:4385.274497169482 mfu:20.686088008130117 +Step 35359 | loss:0.27338334918022156 lr:1.6817507496907913e-07 tokens_per_second_per_gpu:4446.660519773899 mfu:20.975656350290606 +Step 35360 | loss:0.33046984672546387 lr:1.6781991197352133e-07 tokens_per_second_per_gpu:2573.711929241123 mfu:12.14063829076618 +Step 35361 | loss:0.27721089124679565 lr:1.6746512377215361e-07 tokens_per_second_per_gpu:5383.054892291281 mfu:25.39278837080953 +Step 35362 | loss:0.31303703784942627 lr:1.6711071036763504e-07 tokens_per_second_per_gpu:5070.419739271415 mfu:23.918035012957358 +Step 35363 | loss:0.404466837644577 lr:1.6675667176263566e-07 tokens_per_second_per_gpu:4975.1904203173945 mfu:23.46882206765401 +Step 35364 | loss:0.2513376474380493 lr:1.6640300795981444e-07 tokens_per_second_per_gpu:5309.716776794065 mfu:25.04683996723638 +Step 35365 | loss:0.2265467494726181 lr:1.660497189618415e-07 tokens_per_second_per_gpu:5118.744391412328 mfu:24.1459906421419 +Step 35366 | loss:0.37374597787857056 lr:1.6569680477136472e-07 tokens_per_second_per_gpu:4532.136345484698 mfu:21.37886040384813 +Step 35367 | loss:0.2939563989639282 lr:1.6534426539103753e-07 tokens_per_second_per_gpu:3807.6290206454114 mfu:17.96123573888496 +Step 35368 | loss:0.3075469434261322 lr:1.649921008235189e-07 tokens_per_second_per_gpu:5342.882909191622 mfu:25.203290272481354 +Step 35369 | loss:0.26307281851768494 lr:1.6464031107145672e-07 tokens_per_second_per_gpu:4109.890448224351 mfu:19.387054463892206 +Step 35370 | loss:0.2858799993991852 lr:1.6428889613748776e-07 tokens_per_second_per_gpu:4996.388671347565 mfu:23.568817834557905 +Step 35371 | loss:0.2413567751646042 lr:1.6393785602425993e-07 tokens_per_second_per_gpu:4544.784014479793 mfu:21.438521616412842 +Step 35372 | loss:0.3838602304458618 lr:1.6358719073442108e-07 tokens_per_second_per_gpu:4797.029009025731 mfu:22.628404293119985 +Step 35373 | loss:0.29445648193359375 lr:1.632369002705969e-07 tokens_per_second_per_gpu:4296.000513920948 mfu:20.264967397434596 +Step 35374 | loss:0.33816081285476685 lr:1.6288698463542973e-07 tokens_per_second_per_gpu:4975.835911193313 mfu:23.471866958247823 +Step 35375 | loss:0.33856257796287537 lr:1.625374438315508e-07 tokens_per_second_per_gpu:3290.9188281464794 mfu:15.5238256010179 +Step 35376 | loss:0.31725484132766724 lr:1.6218827786158574e-07 tokens_per_second_per_gpu:5348.871392019396 mfu:25.231538967720493 +Step 35377 | loss:0.28155115246772766 lr:1.618394867281603e-07 tokens_per_second_per_gpu:4543.7313851491635 mfu:21.43355618426318 +Step 35378 | loss:0.30071771144866943 lr:1.614910704339001e-07 tokens_per_second_per_gpu:4376.634819944748 mfu:20.645333176580113 +Step 35379 | loss:0.2760539650917053 lr:1.611430289814253e-07 tokens_per_second_per_gpu:4474.52962148911 mfu:21.107119658040325 +Step 35380 | loss:0.2726455628871918 lr:1.6079536237335602e-07 tokens_per_second_per_gpu:3943.011628254077 mfu:18.59985859762975 +Step 35381 | loss:0.2607191801071167 lr:1.604480706122957e-07 tokens_per_second_per_gpu:4760.284091800297 mfu:22.455072249238256 +Step 35382 | loss:0.22432146966457367 lr:1.601011537008701e-07 tokens_per_second_per_gpu:5342.172663101471 mfu:25.19993992049363 +Step 35383 | loss:0.3220025599002838 lr:1.5975461164168813e-07 tokens_per_second_per_gpu:5319.340645435171 mfu:25.092237397617783 +Step 35384 | loss:0.33335551619529724 lr:1.5940844443734782e-07 tokens_per_second_per_gpu:5284.641169362385 mfu:24.928554048641352 +Step 35385 | loss:0.29968440532684326 lr:1.5906265209045257e-07 tokens_per_second_per_gpu:4076.045366382229 mfu:19.2274014382765 +Step 35386 | loss:0.3184913992881775 lr:1.5871723460361143e-07 tokens_per_second_per_gpu:4993.353675046371 mfu:23.55450123918141 +Step 35387 | loss:0.284271240234375 lr:1.5837219197941677e-07 tokens_per_second_per_gpu:5356.271228231566 mfu:25.266445257675546 +Step 35388 | loss:0.26320186257362366 lr:1.5802752422046651e-07 tokens_per_second_per_gpu:4088.729533977787 mfu:19.2872348209669 +Step 35389 | loss:0.25383540987968445 lr:1.5768323132934747e-07 tokens_per_second_per_gpu:5303.053188305118 mfu:25.015406683408585 +Step 35390 | loss:0.3067682683467865 lr:1.5733931330865203e-07 tokens_per_second_per_gpu:4051.6706884673117 mfu:19.11242192380295 +Step 35391 | loss:0.2478751689195633 lr:1.5699577016097255e-07 tokens_per_second_per_gpu:4062.8285492772875 mfu:19.165055456970965 +Step 35392 | loss:0.29789063334465027 lr:1.5665260188888474e-07 tokens_per_second_per_gpu:4975.0465069164675 mfu:23.46814320358768 +Step 35393 | loss:0.32076799869537354 lr:1.563098084949699e-07 tokens_per_second_per_gpu:5382.820916092809 mfu:25.39168466516088 +Step 35394 | loss:0.421130895614624 lr:1.559673899818148e-07 tokens_per_second_per_gpu:5331.303697220226 mfu:25.148669153994998 +Step 35395 | loss:0.20823796093463898 lr:1.5562534635198412e-07 tokens_per_second_per_gpu:4192.454614443866 mfu:19.77652372284901 +Step 35396 | loss:0.3189586102962494 lr:1.5528367760805906e-07 tokens_per_second_per_gpu:4638.124900861827 mfu:21.878826503074308 +Step 35397 | loss:0.2602062225341797 lr:1.5494238375260428e-07 tokens_per_second_per_gpu:5268.675941686427 mfu:24.853243345745128 +Step 35398 | loss:0.3520605266094208 lr:1.5460146478818992e-07 tokens_per_second_per_gpu:5355.1875753471295 mfu:25.261333482129412 +Step 35399 | loss:0.33163103461265564 lr:1.5426092071737508e-07 tokens_per_second_per_gpu:4391.475566277602 mfu:20.71533950912461 +Step 35400 | loss:0.3249361217021942 lr:1.539207515427299e-07 tokens_per_second_per_gpu:4188.102229091827 mfu:19.755992778549928 +Step 35401 | loss:0.3333406150341034 lr:1.5358095726680234e-07 tokens_per_second_per_gpu:3604.759701677377 mfu:17.004266548237737 +Step 35402 | loss:0.2759245038032532 lr:1.5324153789215146e-07 tokens_per_second_per_gpu:4784.112753041147 mfu:22.56747610990018 +Step 35403 | loss:0.2553984820842743 lr:1.5290249342133077e-07 tokens_per_second_per_gpu:5348.394480447531 mfu:25.229289294466504 +Step 35404 | loss:0.30255305767059326 lr:1.5256382385689382e-07 tokens_per_second_per_gpu:5330.754381439005 mfu:25.14607793773204 +Step 35405 | loss:0.23545454442501068 lr:1.5222552920138856e-07 tokens_per_second_per_gpu:3402.7391129650746 mfu:16.051300963015116 +Step 35406 | loss:0.3189619183540344 lr:1.5188760945735181e-07 tokens_per_second_per_gpu:4643.672916611143 mfu:21.904997439952925 +Step 35407 | loss:0.30082932114601135 lr:1.5155006462733158e-07 tokens_per_second_per_gpu:4027.7933578757365 mfu:18.9997884815098 +Step 35408 | loss:0.20173200964927673 lr:1.5121289471385912e-07 tokens_per_second_per_gpu:5350.849502025669 mfu:25.24087005012045 +Step 35409 | loss:0.2344946563243866 lr:1.5087609971947693e-07 tokens_per_second_per_gpu:5329.428946805109 mfu:25.139825636421488 +Step 35410 | loss:0.25644704699516296 lr:1.5053967964671622e-07 tokens_per_second_per_gpu:4061.466617812982 mfu:19.158630993884366 +Step 35411 | loss:0.3710996210575104 lr:1.5020363449810837e-07 tokens_per_second_per_gpu:3576.67841064404 mfu:16.87180230727119 +Step 35412 | loss:0.32669830322265625 lr:1.4986796427617912e-07 tokens_per_second_per_gpu:5006.909451086459 mfu:23.61844614761639 +Step 35413 | loss:0.2973557114601135 lr:1.4953266898345973e-07 tokens_per_second_per_gpu:5214.018994167646 mfu:24.595417198862183 +Step 35414 | loss:0.3782823979854584 lr:1.491977486224594e-07 tokens_per_second_per_gpu:5343.677315268679 mfu:25.20703762148649 +Step 35415 | loss:0.26756101846694946 lr:1.4886320319570378e-07 tokens_per_second_per_gpu:5347.159614737468 mfu:25.22346422222232 +Step 35416 | loss:0.3274536728858948 lr:1.4852903270571318e-07 tokens_per_second_per_gpu:4175.617853832293 mfu:19.697101850396642 +Step 35417 | loss:0.33305326104164124 lr:1.4819523715499107e-07 tokens_per_second_per_gpu:5000.157120935853 mfu:23.58659425422966 +Step 35418 | loss:0.34623587131500244 lr:1.4786181654605768e-07 tokens_per_second_per_gpu:3219.6151017316724 mfu:15.187473757849045 +Step 35419 | loss:0.2727736830711365 lr:1.475287708814166e-07 tokens_per_second_per_gpu:4971.802662305967 mfu:23.45284143510299 +Step 35420 | loss:0.3542173206806183 lr:1.4719610016357134e-07 tokens_per_second_per_gpu:5327.933896221254 mfu:25.13277322773551 +Step 35421 | loss:0.21893735229969025 lr:1.4686380439502545e-07 tokens_per_second_per_gpu:3673.008815512171 mfu:17.326209262696025 +Step 35422 | loss:0.3016957640647888 lr:1.4653188357828253e-07 tokens_per_second_per_gpu:5355.892859683393 mfu:25.26466042867748 +Step 35423 | loss:0.3051680326461792 lr:1.4620033771582943e-07 tokens_per_second_per_gpu:5265.792772684609 mfu:24.839642945644197 +Step 35424 | loss:0.2264525145292282 lr:1.4586916681016416e-07 tokens_per_second_per_gpu:5338.995239186473 mfu:25.184951469762165 +Step 35425 | loss:0.3524828553199768 lr:1.4553837086378475e-07 tokens_per_second_per_gpu:5347.816744851927 mfu:25.22656401709018 +Step 35426 | loss:0.27416282892227173 lr:1.4520794987916698e-07 tokens_per_second_per_gpu:3832.825014812196 mfu:18.080089542249574 +Step 35427 | loss:0.2964843213558197 lr:1.448779038588033e-07 tokens_per_second_per_gpu:5370.650872201314 mfu:25.334276491701097 +Step 35428 | loss:0.29613029956817627 lr:1.4454823280517504e-07 tokens_per_second_per_gpu:4649.111082149345 mfu:21.930650194643345 +Step 35429 | loss:0.21903878450393677 lr:1.4421893672076358e-07 tokens_per_second_per_gpu:5318.0360605168125 mfu:25.086083447972747 +Step 35430 | loss:0.3541033864021301 lr:1.4389001560803916e-07 tokens_per_second_per_gpu:5331.197450993138 mfu:25.148167972414345 +Step 35431 | loss:0.3710261583328247 lr:1.4356146946948313e-07 tokens_per_second_per_gpu:5330.969703480431 mfu:25.14709364853918 +Step 35432 | loss:0.25944310426712036 lr:1.4323329830756017e-07 tokens_per_second_per_gpu:4695.69192777283 mfu:22.15037999096111 +Step 35433 | loss:0.2884911298751831 lr:1.4290550212474052e-07 tokens_per_second_per_gpu:3392.3954151634607 mfu:16.002507975668077 +Step 35434 | loss:0.30528223514556885 lr:1.4257808092349445e-07 tokens_per_second_per_gpu:5346.642080194916 mfu:25.221022923484604 +Step 35435 | loss:0.32674768567085266 lr:1.4225103470627554e-07 tokens_per_second_per_gpu:4149.165045617263 mfu:19.572319440731867 +Step 35436 | loss:0.20681583881378174 lr:1.419243634755485e-07 tokens_per_second_per_gpu:3830.6068508825974 mfu:18.069626084535866 +Step 35437 | loss:0.22674983739852905 lr:1.4159806723377244e-07 tokens_per_second_per_gpu:4678.235057943299 mfu:22.068032957526086 +Step 35438 | loss:0.3124249279499054 lr:1.4127214598340098e-07 tokens_per_second_per_gpu:4641.128084895339 mfu:21.892993034556422 +Step 35439 | loss:0.33271655440330505 lr:1.4094659972688217e-07 tokens_per_second_per_gpu:3337.71068432946 mfu:15.744550770146919 +Step 35440 | loss:0.23362478613853455 lr:1.4062142846666403e-07 tokens_per_second_per_gpu:5312.304774046124 mfu:25.059047991832205 +Step 35441 | loss:0.3308459520339966 lr:1.4029663220520018e-07 tokens_per_second_per_gpu:5327.793661303084 mfu:25.13211171569996 +Step 35442 | loss:0.4116078019142151 lr:1.3997221094492198e-07 tokens_per_second_per_gpu:3751.541340864625 mfu:17.696660583814037 +Step 35443 | loss:0.2728285491466522 lr:1.3964816468827747e-07 tokens_per_second_per_gpu:3226.8008626532064 mfu:15.221370218133156 +Step 35444 | loss:0.292206734418869 lr:1.3932449343769805e-07 tokens_per_second_per_gpu:5372.493376580348 mfu:25.342967899220664 +Step 35445 | loss:0.290942907333374 lr:1.3900119719562622e-07 tokens_per_second_per_gpu:4818.002426340909 mfu:22.727339481029667 +Step 35446 | loss:0.30505070090293884 lr:1.3867827596448223e-07 tokens_per_second_per_gpu:4434.187334732023 mfu:20.916818208303265 +Step 35447 | loss:0.29522621631622314 lr:1.3835572974670864e-07 tokens_per_second_per_gpu:5348.3535732389 mfu:25.2290963281839 +Step 35448 | loss:0.27372586727142334 lr:1.380335585447201e-07 tokens_per_second_per_gpu:4879.020514604164 mfu:23.015172214126824 +Step 35449 | loss:0.2591124176979065 lr:1.3771176236094252e-07 tokens_per_second_per_gpu:3022.2884265608914 mfu:14.256650163666015 +Step 35450 | loss:0.3865250051021576 lr:1.373903411978017e-07 tokens_per_second_per_gpu:4498.52497037045 mfu:21.22030981274229 +Step 35451 | loss:0.2775733172893524 lr:1.3706929505770683e-07 tokens_per_second_per_gpu:5362.302430294764 mfu:25.294895466836834 +Step 35452 | loss:0.3566890358924866 lr:1.3674862394307818e-07 tokens_per_second_per_gpu:3634.469512135295 mfu:17.144412793184102 +Step 35453 | loss:0.23206929862499237 lr:1.3642832785632498e-07 tokens_per_second_per_gpu:5120.566559335684 mfu:24.154586119130283 +Step 35454 | loss:0.29431191086769104 lr:1.3610840679985637e-07 tokens_per_second_per_gpu:3633.380082790294 mfu:17.139273769087918 +Step 35455 | loss:0.25577059388160706 lr:1.35788860776076e-07 tokens_per_second_per_gpu:5345.01002281961 mfu:25.21332423038758 +Step 35456 | loss:0.2671775221824646 lr:1.354696897873986e-07 tokens_per_second_per_gpu:3536.9896449735234 mfu:16.684583628001672 +Step 35457 | loss:0.4039725661277771 lr:1.3515089383621115e-07 tokens_per_second_per_gpu:3671.495517103424 mfu:17.31907077590663 +Step 35458 | loss:0.3487900197505951 lr:1.348324729249173e-07 tokens_per_second_per_gpu:5342.200848211772 mfu:25.200072874467796 +Step 35459 | loss:0.2385050356388092 lr:1.3451442705590954e-07 tokens_per_second_per_gpu:5159.54109462363 mfu:24.338435651824938 +Step 35460 | loss:0.30054527521133423 lr:1.3419675623158046e-07 tokens_per_second_per_gpu:5056.924163194721 mfu:23.85437407802088 +Step 35461 | loss:0.35470712184906006 lr:1.3387946045432254e-07 tokens_per_second_per_gpu:4083.6583055909796 mfu:19.263312971425545 +Step 35462 | loss:0.35990604758262634 lr:1.3356253972652278e-07 tokens_per_second_per_gpu:5326.710322313147 mfu:25.127001420847733 +Step 35463 | loss:0.30717960000038147 lr:1.332459940505626e-07 tokens_per_second_per_gpu:4272.900853130418 mfu:20.156002356277128 +Step 35464 | loss:0.34196802973747253 lr:1.3292982342881787e-07 tokens_per_second_per_gpu:5390.664081707966 mfu:25.428682215549617 +Step 35465 | loss:0.31510940194129944 lr:1.326140278636756e-07 tokens_per_second_per_gpu:5330.26827712563 mfu:25.143784900747907 +Step 35466 | loss:0.3212084472179413 lr:1.3229860735750054e-07 tokens_per_second_per_gpu:4360.765568227831 mfu:20.57047520865787 +Step 35467 | loss:0.26608625054359436 lr:1.3198356191267413e-07 tokens_per_second_per_gpu:5281.345581805209 mfu:24.91300819984893 +Step 35468 | loss:0.37933242321014404 lr:1.3166889153156668e-07 tokens_per_second_per_gpu:3882.0416878121605 mfu:18.312252986020717 +Step 35469 | loss:0.3310195803642273 lr:1.3135459621653745e-07 tokens_per_second_per_gpu:5218.906452684075 mfu:24.618472174570588 +Step 35470 | loss:0.3982885777950287 lr:1.3104067596995117e-07 tokens_per_second_per_gpu:3604.8909101898107 mfu:17.004885481177368 +Step 35471 | loss:0.32106202840805054 lr:1.3072713079417264e-07 tokens_per_second_per_gpu:3849.8202688270844 mfu:18.160259055127142 +Step 35472 | loss:0.3118630647659302 lr:1.3041396069155553e-07 tokens_per_second_per_gpu:5342.037327620141 mfu:25.199301519937585 +Step 35473 | loss:0.3938968777656555 lr:1.3010116566446463e-07 tokens_per_second_per_gpu:4544.149114295141 mfu:21.435526683916695 +Step 35474 | loss:0.3604695796966553 lr:1.2978874571524246e-07 tokens_per_second_per_gpu:5359.116003063065 mfu:25.279864545924365 +Step 35475 | loss:0.36390823125839233 lr:1.2947670084624276e-07 tokens_per_second_per_gpu:3277.3194845607086 mfu:15.459675177036544 +Step 35476 | loss:0.24797113239765167 lr:1.291650310598136e-07 tokens_per_second_per_gpu:4699.911959901808 mfu:22.17028660252522 +Step 35477 | loss:0.35554832220077515 lr:1.2885373635829755e-07 tokens_per_second_per_gpu:5373.634271949462 mfu:25.348349697333596 +Step 35478 | loss:0.4219243824481964 lr:1.2854281674403724e-07 tokens_per_second_per_gpu:4095.0207631246003 mfu:19.316911622246796 +Step 35479 | loss:0.352390855550766 lr:1.282322722193696e-07 tokens_per_second_per_gpu:4909.281683813114 mfu:23.157919312373306 +Step 35480 | loss:0.32178395986557007 lr:1.279221027866262e-07 tokens_per_second_per_gpu:3443.093604956438 mfu:16.241659986924688 +Step 35481 | loss:0.4218502640724182 lr:1.2761230844814954e-07 tokens_per_second_per_gpu:3927.654056195856 mfu:18.527414309959134 +Step 35482 | loss:0.33431074023246765 lr:1.2730288920626555e-07 tokens_per_second_per_gpu:5351.06165245834 mfu:25.241870799907957 +Step 35483 | loss:0.36243927478790283 lr:1.269938450632946e-07 tokens_per_second_per_gpu:5015.643637472268 mfu:23.659646795004114 +Step 35484 | loss:0.29105478525161743 lr:1.266851760215737e-07 tokens_per_second_per_gpu:4037.88676977246 mfu:19.047400827540294 +Step 35485 | loss:0.26602792739868164 lr:1.2637688208341214e-07 tokens_per_second_per_gpu:5323.7367401602105 mfu:25.1129745264869 +Step 35486 | loss:0.26894357800483704 lr:1.260689632511358e-07 tokens_per_second_per_gpu:3767.0727754623827 mfu:17.769925010747656 +Step 35487 | loss:0.27039894461631775 lr:1.2576141952705955e-07 tokens_per_second_per_gpu:4201.302992428845 mfu:19.818263031493117 +Step 35488 | loss:0.2825690805912018 lr:1.2545425091349262e-07 tokens_per_second_per_gpu:4142.758269753189 mfu:19.542097585872348 +Step 35489 | loss:0.29601219296455383 lr:1.251474574127498e-07 tokens_per_second_per_gpu:5107.859673664214 mfu:24.0946455713997 +Step 35490 | loss:0.4165382981300354 lr:1.248410390271404e-07 tokens_per_second_per_gpu:5101.100349607107 mfu:24.062760686562722 +Step 35491 | loss:0.34172680974006653 lr:1.2453499575896254e-07 tokens_per_second_per_gpu:3860.963446817911 mfu:18.212823326932085 +Step 35492 | loss:0.22509263455867767 lr:1.2422932761051998e-07 tokens_per_second_per_gpu:3567.7250267375593 mfu:16.829567667779504 +Step 35493 | loss:0.32680168747901917 lr:1.239240345841164e-07 tokens_per_second_per_gpu:5265.049607403982 mfu:24.836137308218408 +Step 35494 | loss:0.34901612997055054 lr:1.2361911668203886e-07 tokens_per_second_per_gpu:4281.736160366313 mfu:20.197680007967705 +Step 35495 | loss:0.2899213433265686 lr:1.2331457390659107e-07 tokens_per_second_per_gpu:5321.737037607945 mfu:25.103541588363427 +Step 35496 | loss:0.3868504464626312 lr:1.230104062600601e-07 tokens_per_second_per_gpu:3932.137328356525 mfu:18.548562669666115 +Step 35497 | loss:0.33949214220046997 lr:1.2270661374472747e-07 tokens_per_second_per_gpu:4748.148028386442 mfu:22.397824367489168 +Step 35498 | loss:0.2989809215068817 lr:1.2240319636288579e-07 tokens_per_second_per_gpu:4282.519099256904 mfu:20.20137326429777 +Step 35499 | loss:0.2671675980091095 lr:1.2210015411681097e-07 tokens_per_second_per_gpu:5298.518269069204 mfu:24.99401469563577 +Step 35500 | loss:0.2747398912906647 lr:1.2179748700879012e-07 tokens_per_second_per_gpu:3741.091497427492 mfu:17.64736688939394 +Step 35501 | loss:0.34577876329421997 lr:1.214951950410881e-07 tokens_per_second_per_gpu:5352.164628197277 mfu:25.24707372465559 +Step 35502 | loss:0.34137189388275146 lr:1.2119327821599191e-07 tokens_per_second_per_gpu:4321.270276035139 mfu:20.384169176793236 +Step 35503 | loss:0.4187465310096741 lr:1.208917365357609e-07 tokens_per_second_per_gpu:4977.128087134717 mfu:23.477962372631374 +Step 35504 | loss:0.2546270191669464 lr:1.2059057000267104e-07 tokens_per_second_per_gpu:3579.622397103096 mfu:16.88568959369444 +Step 35505 | loss:0.306856244802475 lr:1.202897786189816e-07 tokens_per_second_per_gpu:4398.128450598026 mfu:20.74672229956316 +Step 35506 | loss:0.24206966161727905 lr:1.1998936238696302e-07 tokens_per_second_per_gpu:5360.8861983890265 mfu:25.288214859303388 +Step 35507 | loss:0.3078189194202423 lr:1.1968932130886345e-07 tokens_per_second_per_gpu:2881.099100734331 mfu:13.590636024359124 +Step 35508 | loss:0.4057585895061493 lr:1.193896553869478e-07 tokens_per_second_per_gpu:4226.0919594966645 mfu:19.935196818585926 +Step 35509 | loss:0.3611886203289032 lr:1.1909036462346424e-07 tokens_per_second_per_gpu:5351.188585857327 mfu:25.242469566408047 +Step 35510 | loss:0.2733176648616791 lr:1.1879144902066653e-07 tokens_per_second_per_gpu:5330.009396761727 mfu:25.14256371790183 +Step 35511 | loss:0.33007851243019104 lr:1.1849290858080841e-07 tokens_per_second_per_gpu:3723.71814582577 mfu:17.565413825689177 +Step 35512 | loss:0.20049938559532166 lr:1.1819474330612701e-07 tokens_per_second_per_gpu:3550.3814501372 mfu:16.747755057836304 +Step 35513 | loss:0.3377887010574341 lr:1.1789695319887051e-07 tokens_per_second_per_gpu:4968.811785846854 mfu:23.43873296054921 +Step 35514 | loss:0.3849741816520691 lr:1.1759953826127046e-07 tokens_per_second_per_gpu:4538.068957188181 mfu:21.406845545461394 +Step 35515 | loss:0.33266323804855347 lr:1.1730249849557507e-07 tokens_per_second_per_gpu:4161.64400196316 mfu:19.63118480694496 +Step 35516 | loss:0.35375091433525085 lr:1.1700583390401032e-07 tokens_per_second_per_gpu:3849.918910880247 mfu:18.16072436652198 +Step 35517 | loss:0.34313419461250305 lr:1.1670954448880778e-07 tokens_per_second_per_gpu:5044.857025502179 mfu:23.797451330659168 +Step 35518 | loss:0.27593910694122314 lr:1.1641363025219899e-07 tokens_per_second_per_gpu:4892.695345184029 mfu:23.07967872313873 +Step 35519 | loss:0.3032976984977722 lr:1.1611809119640993e-07 tokens_per_second_per_gpu:3880.2465109917425 mfu:18.303784830670054 +Step 35520 | loss:0.3152838945388794 lr:1.1582292732366107e-07 tokens_per_second_per_gpu:5251.213108531009 mfu:24.770868182284456 +Step 35521 | loss:0.32595863938331604 lr:1.1552813863617284e-07 tokens_per_second_per_gpu:4678.579171155339 mfu:22.069656198259054 +Step 35522 | loss:0.31788235902786255 lr:1.152337251361657e-07 tokens_per_second_per_gpu:5327.031626259569 mfu:25.12851706638311 +Step 35523 | loss:0.2997317612171173 lr:1.1493968682584899e-07 tokens_per_second_per_gpu:3449.96368312341 mfu:16.2740673177945 +Step 35524 | loss:0.3013656735420227 lr:1.1464602370743205e-07 tokens_per_second_per_gpu:4183.399399049248 mfu:19.733808726853542 +Step 35525 | loss:0.2512339651584625 lr:1.1435273578313532e-07 tokens_per_second_per_gpu:5359.281731609411 mfu:25.28064631575368 +Step 35526 | loss:0.309855192899704 lr:1.140598230551515e-07 tokens_per_second_per_gpu:4897.040487848488 mfu:23.100175502444767 +Step 35527 | loss:0.26843059062957764 lr:1.1376728552568993e-07 tokens_per_second_per_gpu:4510.7392639535665 mfu:21.277926719547224 +Step 35528 | loss:0.3394102454185486 lr:1.1347512319695442e-07 tokens_per_second_per_gpu:4423.546717338796 mfu:20.866624600582874 +Step 35529 | loss:0.26903387904167175 lr:1.1318333607113207e-07 tokens_per_second_per_gpu:4486.614479346477 mfu:21.16412599444289 +Step 35530 | loss:0.3834216892719269 lr:1.128919241504267e-07 tokens_per_second_per_gpu:4068.5507356636467 mfu:19.19204798646109 +Step 35531 | loss:0.2957122027873993 lr:1.1260088743703101e-07 tokens_per_second_per_gpu:5221.384434385842 mfu:24.6301612370451 +Step 35532 | loss:0.2854444980621338 lr:1.1231022593312102e-07 tokens_per_second_per_gpu:4533.315631330395 mfu:21.384423296389198 +Step 35533 | loss:0.2853606045246124 lr:1.1201993964089497e-07 tokens_per_second_per_gpu:4703.46077217853 mfu:22.18702695552433 +Step 35534 | loss:0.2921866774559021 lr:1.1173002856253445e-07 tokens_per_second_per_gpu:5350.524379593407 mfu:25.23933639213586 +Step 35535 | loss:0.35816577076911926 lr:1.1144049270022105e-07 tokens_per_second_per_gpu:4908.363917203961 mfu:23.153590050690802 +Step 35536 | loss:0.2616405487060547 lr:1.1115133205612526e-07 tokens_per_second_per_gpu:4550.221142541709 mfu:21.464169477150605 +Step 35537 | loss:0.38862061500549316 lr:1.1086254663242312e-07 tokens_per_second_per_gpu:4761.615840825573 mfu:22.461354336610302 +Step 35538 | loss:0.1983947902917862 lr:1.1057413643129622e-07 tokens_per_second_per_gpu:3273.0684960846065 mfu:15.439622539101347 +Step 35539 | loss:0.41023123264312744 lr:1.1028610145490392e-07 tokens_per_second_per_gpu:4120.3701438437665 mfu:19.436488976149437 +Step 35540 | loss:0.3336217403411865 lr:1.0999844170541673e-07 tokens_per_second_per_gpu:5346.211778554288 mfu:25.218993117228727 +Step 35541 | loss:0.3640854060649872 lr:1.0971115718499403e-07 tokens_per_second_per_gpu:5324.958819252922 mfu:25.118739282075413 +Step 35542 | loss:0.3249247670173645 lr:1.0942424789580629e-07 tokens_per_second_per_gpu:3954.32612218136 mfu:18.653230995936468 +Step 35543 | loss:0.21014437079429626 lr:1.0913771383999627e-07 tokens_per_second_per_gpu:4656.889788481133 mfu:21.967343679594546 +Step 35544 | loss:0.2733423411846161 lr:1.0885155501973443e-07 tokens_per_second_per_gpu:5043.135621183903 mfu:23.789331172787996 +Step 35545 | loss:0.2634674608707428 lr:1.0856577143716351e-07 tokens_per_second_per_gpu:4537.125186427309 mfu:21.402393617758463 +Step 35546 | loss:0.27669820189476013 lr:1.0828036309443734e-07 tokens_per_second_per_gpu:5325.087193003945 mfu:25.11934484296181 +Step 35547 | loss:0.2989571690559387 lr:1.0799532999369866e-07 tokens_per_second_per_gpu:4297.347500469677 mfu:20.27132136280455 +Step 35548 | loss:0.4083617925643921 lr:1.0771067213709018e-07 tokens_per_second_per_gpu:4775.1548036738695 mfu:22.52521993435097 +Step 35549 | loss:0.3080194890499115 lr:1.0742638952676576e-07 tokens_per_second_per_gpu:4521.41938519085 mfu:21.32830667364008 +Step 35550 | loss:0.2614660859107971 lr:1.0714248216484591e-07 tokens_per_second_per_gpu:5334.853641902097 mfu:25.165414848742397 +Step 35551 | loss:0.2643358111381531 lr:1.0685895005347334e-07 tokens_per_second_per_gpu:5248.813101844252 mfu:24.759546941259657 +Step 35552 | loss:0.3650436997413635 lr:1.0657579319478528e-07 tokens_per_second_per_gpu:4407.561455675299 mfu:20.791219393949284 +Step 35553 | loss:0.2387617826461792 lr:1.0629301159090777e-07 tokens_per_second_per_gpu:4216.46445016712 mfu:19.88978221445599 +Step 35554 | loss:0.20786865055561066 lr:1.0601060524396134e-07 tokens_per_second_per_gpu:4996.3484957400915 mfu:23.568628319362787 +Step 35555 | loss:0.27418896555900574 lr:1.0572857415608317e-07 tokens_per_second_per_gpu:3478.851539652457 mfu:16.41033626581858 +Step 35556 | loss:0.3429379165172577 lr:1.054469183293827e-07 tokens_per_second_per_gpu:4866.648620339679 mfu:22.956811878018417 +Step 35557 | loss:0.24246592819690704 lr:1.0516563776598043e-07 tokens_per_second_per_gpu:3981.5264119980025 mfu:18.781539403849564 +Step 35558 | loss:0.25530463457107544 lr:1.0488473246799691e-07 tokens_per_second_per_gpu:4604.871310523702 mfu:21.721963643801814 +Step 35559 | loss:0.32044076919555664 lr:1.0460420243754155e-07 tokens_per_second_per_gpu:5358.418972003741 mfu:25.27657653149177 +Step 35560 | loss:0.26991045475006104 lr:1.0432404767672376e-07 tokens_per_second_per_gpu:5308.593417721932 mfu:25.041540890828312 +Step 35561 | loss:0.25622057914733887 lr:1.0404426818765301e-07 tokens_per_second_per_gpu:5359.264270463527 mfu:25.28056394854937 +Step 35562 | loss:0.3133791983127594 lr:1.0376486397243313e-07 tokens_per_second_per_gpu:4626.812010806795 mfu:21.825461670506645 +Step 35563 | loss:0.34168583154678345 lr:1.0348583503316245e-07 tokens_per_second_per_gpu:4582.059765144597 mfu:21.614357692197597 +Step 35564 | loss:0.31827205419540405 lr:1.032071813719393e-07 tokens_per_second_per_gpu:4855.2748072925415 mfu:22.90315966130246 +Step 35565 | loss:0.32387468218803406 lr:1.0292890299086754e-07 tokens_per_second_per_gpu:5357.199541039025 mfu:25.270824267574824 +Step 35566 | loss:0.3452804982662201 lr:1.0265099989202887e-07 tokens_per_second_per_gpu:4521.682842482037 mfu:21.329549446633052 +Step 35567 | loss:0.4084239602088928 lr:1.0237347207752157e-07 tokens_per_second_per_gpu:3787.464643182926 mfu:17.866116929996856 +Step 35568 | loss:0.3067108690738678 lr:1.0209631954943288e-07 tokens_per_second_per_gpu:5360.071608658642 mfu:25.284372300561824 +Step 35569 | loss:0.20717425644397736 lr:1.0181954230983892e-07 tokens_per_second_per_gpu:5174.584884888241 mfu:24.409399777238402 +Step 35570 | loss:0.23245975375175476 lr:1.0154314036083245e-07 tokens_per_second_per_gpu:4749.579991146708 mfu:22.404579180147387 +Step 35571 | loss:0.22715076804161072 lr:1.0126711370448405e-07 tokens_per_second_per_gpu:4544.798865690782 mfu:21.438591672108956 +Step 35572 | loss:0.3275090157985687 lr:1.0099146234287538e-07 tokens_per_second_per_gpu:4570.747825949319 mfu:21.56099734499763 +Step 35573 | loss:0.2728937566280365 lr:1.0071618627807144e-07 tokens_per_second_per_gpu:4679.60059256429 mfu:22.074474417317536 +Step 35574 | loss:0.32439786195755005 lr:1.0044128551215393e-07 tokens_per_second_per_gpu:4715.182884093517 mfu:22.242322157425853 +Step 35575 | loss:0.2680363655090332 lr:1.0016676004718228e-07 tokens_per_second_per_gpu:5370.233862928431 mfu:25.33230938781179 +Step 35576 | loss:0.3673008978366852 lr:9.989260988522708e-08 tokens_per_second_per_gpu:5337.181868102 mfu:25.176397488963698 +Step 35577 | loss:0.24410787224769592 lr:9.961883502834224e-08 tokens_per_second_per_gpu:5362.603301579775 mfu:25.296314728022164 +Step 35578 | loss:0.3575822114944458 lr:9.934543547859276e-08 tokens_per_second_per_gpu:3623.5280697491366 mfu:17.092800142646023 +Step 35579 | loss:0.27478042244911194 lr:9.907241123803812e-08 tokens_per_second_per_gpu:4133.408189119838 mfu:19.49799166023658 +Step 35580 | loss:0.2816636562347412 lr:9.879976230872112e-08 tokens_per_second_per_gpu:4992.795538656478 mfu:23.551868414602396 +Step 35581 | loss:0.37826117873191833 lr:9.852748869270123e-08 tokens_per_second_per_gpu:4325.229256016179 mfu:20.402844360835438 +Step 35582 | loss:0.3108592629432678 lr:9.82555903920268e-08 tokens_per_second_per_gpu:4088.3263996505098 mfu:19.285333167563323 +Step 35583 | loss:0.29193124175071716 lr:9.798406740873512e-08 tokens_per_second_per_gpu:5359.150430499837 mfu:25.28002694601754 +Step 35584 | loss:0.42727431654930115 lr:9.771291974487451e-08 tokens_per_second_per_gpu:4242.842718099083 mfu:20.014213004887655 +Step 35585 | loss:0.38171032071113586 lr:9.74421474024767e-08 tokens_per_second_per_gpu:5334.009284843182 mfu:25.16143187243333 +Step 35586 | loss:0.34601786732673645 lr:9.717175038359561e-08 tokens_per_second_per_gpu:5324.0200209132445 mfu:25.11431081013164 +Step 35587 | loss:0.30653464794158936 lr:9.690172869024627e-08 tokens_per_second_per_gpu:4717.490393589057 mfu:22.25320707341803 +Step 35588 | loss:0.24733762443065643 lr:9.663208232446597e-08 tokens_per_second_per_gpu:5291.021070695128 mfu:24.95864913175131 +Step 35589 | loss:0.2387210726737976 lr:9.63628112882864e-08 tokens_per_second_per_gpu:4692.958732708229 mfu:22.13748704351022 +Step 35590 | loss:0.29127931594848633 lr:9.609391558372816e-08 tokens_per_second_per_gpu:5110.49856392274 mfu:24.107093666990274 +Step 35591 | loss:0.33678439259529114 lr:9.582539521281187e-08 tokens_per_second_per_gpu:5064.368650594008 mfu:23.889490995243843 +Step 35592 | loss:0.3288169801235199 lr:9.555725017756923e-08 tokens_per_second_per_gpu:4541.029184681874 mfu:21.420809443616033 +Step 35593 | loss:0.378463476896286 lr:9.52894804800042e-08 tokens_per_second_per_gpu:3807.9935461818322 mfu:17.962955268034726 +Step 35594 | loss:0.29001709818840027 lr:9.502208612213181e-08 tokens_per_second_per_gpu:4718.078490553296 mfu:22.25598122713804 +Step 35595 | loss:0.3224789798259735 lr:9.475506710596715e-08 tokens_per_second_per_gpu:2947.258849086408 mfu:13.902722845352924 +Step 35596 | loss:0.22501075267791748 lr:9.448842343351971e-08 tokens_per_second_per_gpu:5388.551464227042 mfu:25.41871663844239 +Step 35597 | loss:0.25137537717819214 lr:9.422215510679344e-08 tokens_per_second_per_gpu:5380.707661296288 mfu:25.381716081727244 +Step 35598 | loss:0.33393558859825134 lr:9.395626212778675e-08 tokens_per_second_per_gpu:3804.633891068857 mfu:17.94710720165055 +Step 35599 | loss:0.2881448566913605 lr:9.369074449850912e-08 tokens_per_second_per_gpu:5187.277811982856 mfu:24.46927447998038 +Step 35600 | loss:0.3758963942527771 lr:9.342560222094787e-08 tokens_per_second_per_gpu:5364.532366281824 mfu:25.305414455354104 +Step 35601 | loss:0.3918241858482361 lr:9.31608352971014e-08 tokens_per_second_per_gpu:5316.9603463278045 mfu:25.081009120607433 +Step 35602 | loss:0.4096861183643341 lr:9.289644372895701e-08 tokens_per_second_per_gpu:5319.655776654599 mfu:25.093723925346758 +Step 35603 | loss:0.3579525649547577 lr:9.263242751850753e-08 tokens_per_second_per_gpu:3150.556762019652 mfu:14.861713786859712 +Step 35604 | loss:0.2606622278690338 lr:9.23687866677403e-08 tokens_per_second_per_gpu:3521.6237164114136 mfu:16.612099921276446 +Step 35605 | loss:0.32244792580604553 lr:9.210552117863703e-08 tokens_per_second_per_gpu:4244.736477494068 mfu:20.02314619106247 +Step 35606 | loss:0.2861942648887634 lr:9.184263105317392e-08 tokens_per_second_per_gpu:3574.015697202953 mfu:16.85924183366382 +Step 35607 | loss:0.4091705083847046 lr:9.158011629333274e-08 tokens_per_second_per_gpu:5293.050479679417 mfu:24.968222200183078 +Step 35608 | loss:0.34118446707725525 lr:9.131797690108412e-08 tokens_per_second_per_gpu:5305.392529082489 mfu:25.026441752988138 +Step 35609 | loss:0.3936414122581482 lr:9.105621287839872e-08 tokens_per_second_per_gpu:5364.015971069126 mfu:25.3029785310297 +Step 35610 | loss:0.30207857489585876 lr:9.079482422725272e-08 tokens_per_second_per_gpu:4184.535573439235 mfu:19.739168255302445 +Step 35611 | loss:0.2739785313606262 lr:9.053381094960567e-08 tokens_per_second_per_gpu:4992.01693973932 mfu:23.548195630667486 +Step 35612 | loss:0.34613391757011414 lr:9.027317304742267e-08 tokens_per_second_per_gpu:5014.454817131692 mfu:23.654038926623947 +Step 35613 | loss:0.2546594738960266 lr:9.001291052266325e-08 tokens_per_second_per_gpu:4611.533958700791 mfu:21.753392492018143 +Step 35614 | loss:0.3583222031593323 lr:8.975302337728698e-08 tokens_per_second_per_gpu:4935.746072250905 mfu:23.282756347924636 +Step 35615 | loss:0.33833110332489014 lr:8.949351161324227e-08 tokens_per_second_per_gpu:4070.032723807128 mfu:19.19903877738714 +Step 35616 | loss:0.3517603874206543 lr:8.923437523248868e-08 tokens_per_second_per_gpu:4395.313437452335 mfu:20.73344340226417 +Step 35617 | loss:0.37124472856521606 lr:8.897561423697465e-08 tokens_per_second_per_gpu:4873.098481499794 mfu:22.98723697357047 +Step 35618 | loss:0.3466796875 lr:8.871722862863751e-08 tokens_per_second_per_gpu:4606.459046167567 mfu:21.72945326372956 +Step 35619 | loss:0.36082035303115845 lr:8.845921840943128e-08 tokens_per_second_per_gpu:4487.439417328067 mfu:21.168017367650986 +Step 35620 | loss:0.31216317415237427 lr:8.820158358128772e-08 tokens_per_second_per_gpu:4375.648577466063 mfu:20.640680902537838 +Step 35621 | loss:0.24718526005744934 lr:8.794432414615528e-08 tokens_per_second_per_gpu:3597.8808505035017 mfu:16.971817833597534 +Step 35622 | loss:0.2813894748687744 lr:8.76874401059602e-08 tokens_per_second_per_gpu:5313.928789672056 mfu:25.066708750625367 +Step 35623 | loss:0.27679261565208435 lr:8.743093146263425e-08 tokens_per_second_per_gpu:5334.132601154765 mfu:25.162013576515 +Step 35624 | loss:0.26518115401268005 lr:8.717479821810926e-08 tokens_per_second_per_gpu:4771.543294885727 mfu:22.508183831210367 +Step 35625 | loss:0.3304065465927124 lr:8.691904037431143e-08 tokens_per_second_per_gpu:4421.14972791334 mfu:20.85531759249443 +Step 35626 | loss:0.20908699929714203 lr:8.6663657933167e-08 tokens_per_second_per_gpu:3724.172776162377 mfu:17.567558394554887 +Step 35627 | loss:0.28495410084724426 lr:8.640865089659112e-08 tokens_per_second_per_gpu:5327.544717732299 mfu:25.130937406402413 +Step 35628 | loss:0.22338652610778809 lr:8.615401926650446e-08 tokens_per_second_per_gpu:4493.031897963313 mfu:21.194398052983026 +Step 35629 | loss:0.3461526930332184 lr:8.589976304482772e-08 tokens_per_second_per_gpu:4210.736221373375 mfu:19.862761181899284 +Step 35630 | loss:0.2717934846878052 lr:8.564588223346492e-08 tokens_per_second_per_gpu:4596.321266997334 mfu:21.681631629700085 +Step 35631 | loss:0.36463430523872375 lr:8.539237683432566e-08 tokens_per_second_per_gpu:5336.432667529244 mfu:25.172863382035207 +Step 35632 | loss:0.3588058352470398 lr:8.513924684932506e-08 tokens_per_second_per_gpu:4265.412597846382 mfu:20.12067897847423 +Step 35633 | loss:0.36770421266555786 lr:8.488649228035605e-08 tokens_per_second_per_gpu:4546.871941687691 mfu:21.448370725289614 +Step 35634 | loss:0.237149178981781 lr:8.463411312932823e-08 tokens_per_second_per_gpu:5011.485423308475 mfu:23.640031789329008 +Step 35635 | loss:0.2586473226547241 lr:8.438210939813452e-08 tokens_per_second_per_gpu:5246.3075028279745 mfu:24.747727603200598 +Step 35636 | loss:0.25280511379241943 lr:8.413048108867339e-08 tokens_per_second_per_gpu:4280.2948251696425 mfu:20.190880983927865 +Step 35637 | loss:0.2710566818714142 lr:8.387922820283777e-08 tokens_per_second_per_gpu:5128.928159025663 mfu:24.194029211503928 +Step 35638 | loss:0.27165132761001587 lr:8.362835074251508e-08 tokens_per_second_per_gpu:3906.619109168955 mfu:18.428188875901203 +Step 35639 | loss:0.345401793718338 lr:8.337784870959264e-08 tokens_per_second_per_gpu:4242.019532616038 mfu:20.010329898514208 +Step 35640 | loss:0.2669176757335663 lr:8.312772210595232e-08 tokens_per_second_per_gpu:5330.27860991841 mfu:25.143833642294403 +Step 35641 | loss:0.33996886014938354 lr:8.287797093348148e-08 tokens_per_second_per_gpu:4174.36783200135 mfu:19.691205284144367 +Step 35642 | loss:0.3836098909378052 lr:8.262859519405086e-08 tokens_per_second_per_gpu:3855.0512902855976 mfu:18.184934675851242 +Step 35643 | loss:0.2538515329360962 lr:8.237959488954227e-08 tokens_per_second_per_gpu:4017.52589143958 mfu:18.951355090520035 +Step 35644 | loss:0.3560669720172882 lr:8.213097002182646e-08 tokens_per_second_per_gpu:4518.408787698975 mfu:21.314105171610002 +Step 35645 | loss:0.3340519964694977 lr:8.188272059277413e-08 tokens_per_second_per_gpu:5322.0413529469615 mfu:25.10497709573912 +Step 35646 | loss:0.37128591537475586 lr:8.163484660425047e-08 tokens_per_second_per_gpu:5245.224282391129 mfu:24.74261786758354 +Step 35647 | loss:0.2641254663467407 lr:8.138734805812065e-08 tokens_per_second_per_gpu:5348.741649431465 mfu:25.230926949796853 +Step 35648 | loss:0.20084652304649353 lr:8.11402249562443e-08 tokens_per_second_per_gpu:5317.611532251814 mfu:25.08408087571476 +Step 35649 | loss:0.21641746163368225 lr:8.089347730048657e-08 tokens_per_second_per_gpu:4873.835096317472 mfu:22.9907117113449 +Step 35650 | loss:0.3214602768421173 lr:8.064710509269602e-08 tokens_per_second_per_gpu:3222.336001651178 mfu:15.200308707002748 +Step 35651 | loss:0.3496979773044586 lr:8.04011083347267e-08 tokens_per_second_per_gpu:5034.107782434115 mfu:23.74674531709712 +Step 35652 | loss:0.24179264903068542 lr:8.015548702843267e-08 tokens_per_second_per_gpu:5006.73448147033 mfu:23.617620786084213 +Step 35653 | loss:0.2899573743343353 lr:7.991024117565693e-08 tokens_per_second_per_gpu:4155.311262297281 mfu:19.60131219346421 +Step 35654 | loss:0.2923004925251007 lr:7.966537077824798e-08 tokens_per_second_per_gpu:4629.566207274807 mfu:21.838453685160633 +Step 35655 | loss:0.3377419412136078 lr:7.94208758380488e-08 tokens_per_second_per_gpu:4695.771016445499 mfu:22.150753065724007 +Step 35656 | loss:0.2935979962348938 lr:7.917675635689125e-08 tokens_per_second_per_gpu:5400.28058452887 mfu:25.47404490010039 +Step 35657 | loss:0.2718028426170349 lr:7.893301233661277e-08 tokens_per_second_per_gpu:3819.026551408262 mfu:18.014999836112597 +Step 35658 | loss:0.2773601710796356 lr:7.86896437790563e-08 tokens_per_second_per_gpu:4349.880554116006 mfu:20.519128739917484 +Step 35659 | loss:0.3567960262298584 lr:7.844665068603707e-08 tokens_per_second_per_gpu:3525.0871672099197 mfu:16.62843760961864 +Step 35660 | loss:0.32101014256477356 lr:7.820403305939806e-08 tokens_per_second_per_gpu:5044.947434475152 mfu:23.79787780521004 +Step 35661 | loss:0.39665094017982483 lr:7.796179090094891e-08 tokens_per_second_per_gpu:3899.539714094464 mfu:18.394794161465477 +Step 35662 | loss:0.3773703873157501 lr:7.771992421252705e-08 tokens_per_second_per_gpu:5200.981251148306 mfu:24.53391594057162 +Step 35663 | loss:0.38494935631752014 lr:7.74784329959366e-08 tokens_per_second_per_gpu:4826.665706476396 mfu:22.768205651536764 +Step 35664 | loss:0.3705992102622986 lr:7.723731725300387e-08 tokens_per_second_per_gpu:4419.461068659306 mfu:20.847351898678177 +Step 35665 | loss:0.36740607023239136 lr:7.699657698554407e-08 tokens_per_second_per_gpu:3470.1952153007187 mfu:16.3695029069302 +Step 35666 | loss:0.2953205704689026 lr:7.675621219535578e-08 tokens_per_second_per_gpu:4296.282029992777 mfu:20.266295356777007 +Step 35667 | loss:0.28400322794914246 lr:7.651622288425975e-08 tokens_per_second_per_gpu:3894.991776010581 mfu:18.373340761565547 +Step 35668 | loss:0.3333088457584381 lr:7.627660905404899e-08 tokens_per_second_per_gpu:5319.507860386106 mfu:25.09302617907116 +Step 35669 | loss:0.3288293480873108 lr:7.603737070653872e-08 tokens_per_second_per_gpu:4947.575773468095 mfu:23.338559066921274 +Step 35670 | loss:0.44688764214515686 lr:7.579850784351639e-08 tokens_per_second_per_gpu:3555.756509995945 mfu:16.773110132269785 +Step 35671 | loss:0.2679617702960968 lr:7.556002046679167e-08 tokens_per_second_per_gpu:4884.358756895975 mfu:23.040353613815345 +Step 35672 | loss:0.2850416600704193 lr:7.532190857814092e-08 tokens_per_second_per_gpu:4573.975087483879 mfu:21.576220888283746 +Step 35673 | loss:0.34482333064079285 lr:7.50841721793738e-08 tokens_per_second_per_gpu:4363.461652546014 mfu:20.58319309838616 +Step 35674 | loss:0.29667752981185913 lr:7.484681127226112e-08 tokens_per_second_per_gpu:3681.280437561578 mfu:17.365227915187326 +Step 35675 | loss:0.26699668169021606 lr:7.460982585860144e-08 tokens_per_second_per_gpu:4144.402148281176 mfu:19.54985204136346 +Step 35676 | loss:0.28602659702301025 lr:7.43732159401711e-08 tokens_per_second_per_gpu:5334.854805101564 mfu:25.16542033575275 +Step 35677 | loss:0.31071773171424866 lr:7.413698151874649e-08 tokens_per_second_per_gpu:4714.723720274691 mfu:22.240156203351024 +Step 35678 | loss:0.43622201681137085 lr:7.390112259610948e-08 tokens_per_second_per_gpu:5344.09666167612 mfu:25.209015750038258 +Step 35679 | loss:0.3332063555717468 lr:7.366563917403647e-08 tokens_per_second_per_gpu:5373.1806301533015 mfu:25.346209791581835 +Step 35680 | loss:0.28765255212783813 lr:7.343053125429267e-08 tokens_per_second_per_gpu:4986.21674330382 mfu:23.520835114465676 +Step 35681 | loss:0.27090558409690857 lr:7.319579883864891e-08 tokens_per_second_per_gpu:4442.075206737148 mfu:20.95402664636117 +Step 35682 | loss:0.29779258370399475 lr:7.296144192887045e-08 tokens_per_second_per_gpu:4431.010922748063 mfu:20.901834531023155 +Step 35683 | loss:0.3354930579662323 lr:7.272746052671697e-08 tokens_per_second_per_gpu:5304.146955204128 mfu:25.020566168487093 +Step 35684 | loss:0.3735755681991577 lr:7.249385463395375e-08 tokens_per_second_per_gpu:5324.566896111107 mfu:25.116890513746487 +Step 35685 | loss:0.3799746036529541 lr:7.226062425233493e-08 tokens_per_second_per_gpu:4166.894347280639 mfu:19.655951581608793 +Step 35686 | loss:0.34433069825172424 lr:7.202776938361467e-08 tokens_per_second_per_gpu:5332.63151783073 mfu:25.154932710364303 +Step 35687 | loss:0.36948177218437195 lr:7.179529002954155e-08 tokens_per_second_per_gpu:5167.59274463472 mfu:24.3764166586804 +Step 35688 | loss:0.3932991325855255 lr:7.156318619186975e-08 tokens_per_second_per_gpu:3728.668367117162 mfu:17.588764864115518 +Step 35689 | loss:0.3840802311897278 lr:7.13314578723423e-08 tokens_per_second_per_gpu:4631.021470016133 mfu:21.845318407804893 +Step 35690 | loss:0.37076637148857117 lr:7.110010507270226e-08 tokens_per_second_per_gpu:3427.6375410047945 mfu:16.16875109618835 +Step 35691 | loss:0.3155769407749176 lr:7.086912779468713e-08 tokens_per_second_per_gpu:5369.512852568666 mfu:25.328908258927726 +Step 35692 | loss:0.29576539993286133 lr:7.06385260400344e-08 tokens_per_second_per_gpu:4236.881096291983 mfu:19.986091017666002 +Step 35693 | loss:0.2844969630241394 lr:7.040829981048713e-08 tokens_per_second_per_gpu:4333.655658035732 mfu:20.442593136854104 +Step 35694 | loss:0.2415032684803009 lr:7.017844910776616e-08 tokens_per_second_per_gpu:5376.700267025286 mfu:25.36281251921896 +Step 35695 | loss:0.2799375653266907 lr:6.994897393360344e-08 tokens_per_second_per_gpu:4459.1360714299335 mfu:21.03450565600072 +Step 35696 | loss:0.36061182618141174 lr:6.971987428972537e-08 tokens_per_second_per_gpu:4732.791068290228 mfu:22.325382966547753 +Step 35697 | loss:0.32676437497138977 lr:6.949115017785835e-08 tokens_per_second_per_gpu:5375.9140112407795 mfu:25.359103616534373 +Step 35698 | loss:0.2770678997039795 lr:6.926280159971765e-08 tokens_per_second_per_gpu:4539.106810266232 mfu:21.411741275506557 +Step 35699 | loss:0.2730117738246918 lr:6.903482855701859e-08 tokens_per_second_per_gpu:5354.02923715578 mfu:25.25586940324794 +Step 35700 | loss:0.3005499243736267 lr:6.880723105148757e-08 tokens_per_second_per_gpu:4296.600024791041 mfu:20.267795392495923 +Step 35701 | loss:0.30703863501548767 lr:6.85800090848232e-08 tokens_per_second_per_gpu:5355.806046756471 mfu:25.264250917289377 +Step 35702 | loss:0.416787326335907 lr:6.835316265874081e-08 tokens_per_second_per_gpu:3881.098624042572 mfu:18.307804393316296 +Step 35703 | loss:0.24729447066783905 lr:6.812669177494457e-08 tokens_per_second_per_gpu:4268.980879667354 mfu:20.137511172635417 +Step 35704 | loss:0.4091513752937317 lr:6.790059643514423e-08 tokens_per_second_per_gpu:3986.9618517517065 mfu:18.807179300549436 +Step 35705 | loss:0.22529983520507812 lr:6.767487664102734e-08 tokens_per_second_per_gpu:4215.933875134547 mfu:19.88727940150191 +Step 35706 | loss:0.32874199748039246 lr:6.744953239430363e-08 tokens_per_second_per_gpu:4682.064157812463 mfu:22.086095474920743 +Step 35707 | loss:0.24434256553649902 lr:6.722456369666619e-08 tokens_per_second_per_gpu:4844.60787695145 mfu:22.852841930917545 +Step 35708 | loss:0.37240204215049744 lr:6.6999970549797e-08 tokens_per_second_per_gpu:5347.594968343038 mfu:25.225517859458776 +Step 35709 | loss:0.34378528594970703 lr:6.677575295540028e-08 tokens_per_second_per_gpu:5302.009564609577 mfu:25.010483732376155 +Step 35710 | loss:0.29014015197753906 lr:6.65519109151469e-08 tokens_per_second_per_gpu:4635.555622171037 mfu:21.866706777125703 +Step 35711 | loss:0.32463136315345764 lr:6.632844443073549e-08 tokens_per_second_per_gpu:3973.2780890896506 mfu:18.742630657381863 +Step 35712 | loss:0.363246887922287 lr:6.610535350383695e-08 tokens_per_second_per_gpu:4090.916540499657 mfu:19.297551303872346 +Step 35713 | loss:0.28355854749679565 lr:6.588263813612772e-08 tokens_per_second_per_gpu:5332.22502842698 mfu:25.15301523049257 +Step 35714 | loss:0.3007660508155823 lr:6.566029832928977e-08 tokens_per_second_per_gpu:4621.504275791998 mfu:21.800424178848875 +Step 35715 | loss:0.30473706126213074 lr:6.5438334084994e-08 tokens_per_second_per_gpu:4238.896080787582 mfu:19.99559604332355 +Step 35716 | loss:0.25394707918167114 lr:6.521674540490574e-08 tokens_per_second_per_gpu:4575.937651592376 mfu:21.58547863803149 +Step 35717 | loss:0.3564211130142212 lr:6.499553229069588e-08 tokens_per_second_per_gpu:4003.0554881484727 mfu:18.883095729290496 +Step 35718 | loss:0.4187043011188507 lr:6.47746947440242e-08 tokens_per_second_per_gpu:5400.338565788272 mfu:25.474318407593614 +Step 35719 | loss:0.35049983859062195 lr:6.455423276655603e-08 tokens_per_second_per_gpu:3992.6636375084677 mfu:18.834075596789045 +Step 35720 | loss:0.28678250312805176 lr:6.433414635994562e-08 tokens_per_second_per_gpu:5347.74332558488 mfu:25.2262176858809 +Step 35721 | loss:0.22458995878696442 lr:6.411443552584718e-08 tokens_per_second_per_gpu:4952.335948446607 mfu:23.36101362446452 +Step 35722 | loss:0.32375630736351013 lr:6.38951002659205e-08 tokens_per_second_per_gpu:5352.245069924174 mfu:25.247453182006137 +Step 35723 | loss:0.37680578231811523 lr:6.36761405818087e-08 tokens_per_second_per_gpu:4434.208223614341 mfu:20.916916744725164 +Step 35724 | loss:0.31888577342033386 lr:6.34575564751605e-08 tokens_per_second_per_gpu:4962.893935397214 mfu:23.410817450288853 +Step 35725 | loss:0.22416481375694275 lr:6.323934794762454e-08 tokens_per_second_per_gpu:5346.851253103994 mfu:25.2220096277853 +Step 35726 | loss:0.2836616039276123 lr:6.302151500083287e-08 tokens_per_second_per_gpu:3690.180936624593 mfu:17.4072130878493 +Step 35727 | loss:0.29876527190208435 lr:6.28040576364286e-08 tokens_per_second_per_gpu:3923.821897521817 mfu:18.50933736366 +Step 35728 | loss:0.2832471430301666 lr:6.258697585604934e-08 tokens_per_second_per_gpu:4771.298114498247 mfu:22.507027273490536 +Step 35729 | loss:0.41450682282447815 lr:6.237026966132153e-08 tokens_per_second_per_gpu:5337.6275906805095 mfu:25.178500038414636 +Step 35730 | loss:0.2779300808906555 lr:6.215393905388278e-08 tokens_per_second_per_gpu:4281.301493748862 mfu:20.195629611371416 +Step 35731 | loss:0.37898334860801697 lr:6.193798403535401e-08 tokens_per_second_per_gpu:5312.385603801907 mfu:25.059429279581213 +Step 35732 | loss:0.2886080741882324 lr:6.172240460736168e-08 tokens_per_second_per_gpu:4236.6188047509995 mfu:19.984853743715632 +Step 35733 | loss:0.2161286622285843 lr:6.150720077152672e-08 tokens_per_second_per_gpu:3195.257060594278 mfu:15.072572721893593 +Step 35734 | loss:0.2531035542488098 lr:6.129237252947007e-08 tokens_per_second_per_gpu:4475.622659697194 mfu:21.112275705759405 +Step 35735 | loss:0.3705863356590271 lr:6.107791988280709e-08 tokens_per_second_per_gpu:4524.726675536494 mfu:21.3439077264824 +Step 35736 | loss:0.3297322690486908 lr:6.08638428331476e-08 tokens_per_second_per_gpu:4539.794361865728 mfu:21.414984573709777 +Step 35737 | loss:0.3453042805194855 lr:6.065014138210146e-08 tokens_per_second_per_gpu:5364.398480069019 mfu:25.304782891244958 +Step 35738 | loss:0.285428911447525 lr:6.043681553128399e-08 tokens_per_second_per_gpu:3502.4169365072853 mfu:16.521498263454106 +Step 35739 | loss:0.2816087603569031 lr:6.02238652822884e-08 tokens_per_second_per_gpu:4748.0761912619055 mfu:22.39748549951605 +Step 35740 | loss:0.4200096130371094 lr:6.00112906367245e-08 tokens_per_second_per_gpu:5304.399602979283 mfu:25.021757951148654 +Step 35741 | loss:0.3079567849636078 lr:5.979909159618547e-08 tokens_per_second_per_gpu:4911.369382032149 mfu:23.16776734107054 +Step 35742 | loss:0.28888413310050964 lr:5.958726816227e-08 tokens_per_second_per_gpu:4744.127689011801 mfu:22.378859740718358 +Step 35743 | loss:0.428584486246109 lr:5.93758203365713e-08 tokens_per_second_per_gpu:5362.082238940072 mfu:25.29385678664823 +Step 35744 | loss:0.3063126504421234 lr:5.916474812067696e-08 tokens_per_second_per_gpu:4007.4344122828365 mfu:18.903751861553786 +Step 35745 | loss:0.3291108310222626 lr:5.8954051516180164e-08 tokens_per_second_per_gpu:5313.739792123237 mfu:25.065817216941277 +Step 35746 | loss:0.37491685152053833 lr:5.8743730524662974e-08 tokens_per_second_per_gpu:5174.021313096198 mfu:24.406741312940078 +Step 35747 | loss:0.27715444564819336 lr:5.8533785147707466e-08 tokens_per_second_per_gpu:3825.9745293393853 mfu:18.04777463346098 +Step 35748 | loss:0.2952348291873932 lr:5.8324215386890146e-08 tokens_per_second_per_gpu:5380.570083305413 mfu:25.38106710287106 +Step 35749 | loss:0.2976243197917938 lr:5.8115021243787535e-08 tokens_per_second_per_gpu:3562.6126847287624 mfu:16.805451878268517 +Step 35750 | loss:0.2706012427806854 lr:5.790620271997615e-08 tokens_per_second_per_gpu:3812.642783699514 mfu:17.984886514647346 +Step 35751 | loss:0.33834320306777954 lr:5.769775981702696e-08 tokens_per_second_per_gpu:4412.240097470232 mfu:20.813289345554587 +Step 35752 | loss:0.4270177185535431 lr:5.7489692536505377e-08 tokens_per_second_per_gpu:5353.517087355104 mfu:25.253453505256388 +Step 35753 | loss:0.35161152482032776 lr:5.728200087997126e-08 tokens_per_second_per_gpu:4589.316583441088 mfu:21.64858934224302 +Step 35754 | loss:0.30427947640419006 lr:5.707468484899559e-08 tokens_per_second_per_gpu:5327.839968183368 mfu:25.132330153155024 +Step 35755 | loss:0.26522788405418396 lr:5.68677444451382e-08 tokens_per_second_per_gpu:4079.753598308179 mfu:19.244893800960813 +Step 35756 | loss:0.312706857919693 lr:5.6661179669942334e-08 tokens_per_second_per_gpu:5378.486673023862 mfu:25.371239301106534 +Step 35757 | loss:0.30936726927757263 lr:5.645499052497893e-08 tokens_per_second_per_gpu:4743.8807960320855 mfu:22.37769510440875 +Step 35758 | loss:0.37988418340682983 lr:5.624917701178567e-08 tokens_per_second_per_gpu:5326.129271611168 mfu:25.124260505549028 +Step 35759 | loss:0.3112773895263672 lr:5.604373913191685e-08 tokens_per_second_per_gpu:4043.106297732328 mfu:19.072022231470292 +Step 35760 | loss:0.2692239284515381 lr:5.583867688691569e-08 tokens_per_second_per_gpu:5338.498292929744 mfu:25.1826072894814 +Step 35761 | loss:0.35156646370887756 lr:5.563399027832539e-08 tokens_per_second_per_gpu:3912.1159467805232 mfu:18.454118396772174 +Step 35762 | loss:0.22779209911823273 lr:5.5429679307683614e-08 tokens_per_second_per_gpu:5344.785775357223 mfu:25.21226641684694 +Step 35763 | loss:0.40764346718788147 lr:5.522574397652802e-08 tokens_per_second_per_gpu:5260.754292011549 mfu:24.815875572655926 +Step 35764 | loss:0.3075505793094635 lr:5.502218428639072e-08 tokens_per_second_per_gpu:5330.848509188557 mfu:25.14652195438653 +Step 35765 | loss:0.4078978896141052 lr:5.4819000238803816e-08 tokens_per_second_per_gpu:5340.563099143228 mfu:25.192347332682672 +Step 35766 | loss:0.34755754470825195 lr:5.4616191835299426e-08 tokens_per_second_per_gpu:3941.9371119857474 mfu:18.59478992106066 +Step 35767 | loss:0.21129195392131805 lr:5.4413759077404094e-08 tokens_per_second_per_gpu:4280.4566741533 mfu:20.191644453198062 +Step 35768 | loss:0.23485243320465088 lr:5.421170196662772e-08 tokens_per_second_per_gpu:4116.682952733908 mfu:19.41909586658456 +Step 35769 | loss:0.33635830879211426 lr:5.401002050450244e-08 tokens_per_second_per_gpu:5330.79752688264 mfu:25.14628146215104 +Step 35770 | loss:0.32894545793533325 lr:5.380871469254367e-08 tokens_per_second_per_gpu:5333.736120872495 mfu:25.16014331137702 +Step 35771 | loss:0.34651538729667664 lr:5.360778453225579e-08 tokens_per_second_per_gpu:5365.391829943008 mfu:25.30946869208353 +Step 35772 | loss:0.24763691425323486 lr:5.3407230025159794e-08 tokens_per_second_per_gpu:5015.553955444771 mfu:23.659223749579077 +Step 35773 | loss:0.35360270738601685 lr:5.3207051172765586e-08 tokens_per_second_per_gpu:5380.763323803297 mfu:25.381978651269808 +Step 35774 | loss:0.24982301890850067 lr:5.3007247976571974e-08 tokens_per_second_per_gpu:3453.5787129156324 mfu:16.2911200301121 +Step 35775 | loss:0.385408490896225 lr:5.28078204380833e-08 tokens_per_second_per_gpu:3790.4331332475294 mfu:17.880119804108222 +Step 35776 | loss:0.3542824387550354 lr:5.260876855879837e-08 tokens_per_second_per_gpu:5351.775789825006 mfu:25.245239507710917 +Step 35777 | loss:0.2826322615146637 lr:5.241009234021599e-08 tokens_per_second_per_gpu:5340.369869054016 mfu:25.191435833383405 +Step 35778 | loss:0.24717192351818085 lr:5.221179178383495e-08 tokens_per_second_per_gpu:5168.5796625999255 mfu:24.38107211910784 +Step 35779 | loss:0.2487405240535736 lr:5.2013866891137406e-08 tokens_per_second_per_gpu:3809.946365340634 mfu:17.972167049192098 +Step 35780 | loss:0.280606210231781 lr:5.181631766362216e-08 tokens_per_second_per_gpu:5373.818570574709 mfu:25.349219065393477 +Step 35781 | loss:0.2770962417125702 lr:5.16191441027658e-08 tokens_per_second_per_gpu:4347.041847237984 mfu:20.50573807524052 +Step 35782 | loss:0.2991478741168976 lr:5.142234621005604e-08 tokens_per_second_per_gpu:5324.019972576121 mfu:25.114310582117184 +Step 35783 | loss:0.4392144978046417 lr:5.122592398697501e-08 tokens_per_second_per_gpu:4745.737431707816 mfu:22.386453171665885 +Step 35784 | loss:0.38019701838493347 lr:5.102987743499377e-08 tokens_per_second_per_gpu:4512.424961816987 mfu:21.285878443978692 +Step 35785 | loss:0.34957557916641235 lr:5.0834206555594457e-08 tokens_per_second_per_gpu:4176.837403419168 mfu:19.702854673874548 +Step 35786 | loss:0.3763604462146759 lr:5.063891135024257e-08 tokens_per_second_per_gpu:2986.9391754676017 mfu:14.089901715054813 +Step 35787 | loss:0.2729729413986206 lr:5.044399182040915e-08 tokens_per_second_per_gpu:3851.111681732935 mfu:18.1663508701418 +Step 35788 | loss:0.2876179814338684 lr:5.02494479675597e-08 tokens_per_second_per_gpu:5308.37455086759 mfu:25.04050845853384 +Step 35789 | loss:0.2962336540222168 lr:5.005527979315416e-08 tokens_per_second_per_gpu:5346.276487788843 mfu:25.219298361728658 +Step 35790 | loss:0.3062056601047516 lr:4.986148729866358e-08 tokens_per_second_per_gpu:3611.085064398593 mfu:17.03410436341188 +Step 35791 | loss:0.3863477110862732 lr:4.966807048553124e-08 tokens_per_second_per_gpu:5339.3571868406 mfu:25.186658838601318 +Step 35792 | loss:0.3052610754966736 lr:4.947502935522263e-08 tokens_per_second_per_gpu:5327.864329678611 mfu:25.132445070484586 +Step 35793 | loss:0.26153796911239624 lr:4.9282363909186614e-08 tokens_per_second_per_gpu:3677.0932213056576 mfu:17.345476101695386 +Step 35794 | loss:0.2267794907093048 lr:4.909007414887201e-08 tokens_per_second_per_gpu:3914.944101121166 mfu:18.467459283329873 +Step 35795 | loss:0.34926044940948486 lr:4.889816007572212e-08 tokens_per_second_per_gpu:4502.933301839661 mfu:21.24110466442146 +Step 35796 | loss:0.29220128059387207 lr:4.870662169118578e-08 tokens_per_second_per_gpu:5049.59025577871 mfu:23.819778785445926 +Step 35797 | loss:0.4467458128929138 lr:4.851545899670073e-08 tokens_per_second_per_gpu:5211.489722971693 mfu:24.583486195092707 +Step 35798 | loss:0.3021981120109558 lr:4.832467199370472e-08 tokens_per_second_per_gpu:4407.367098849543 mfu:20.7903025796415 +Step 35799 | loss:0.2587202489376068 lr:4.813426068362992e-08 tokens_per_second_per_gpu:5364.568464864275 mfu:25.30558473852658 +Step 35800 | loss:0.35679855942726135 lr:4.7944225067914074e-08 tokens_per_second_per_gpu:4273.706009609025 mfu:20.159800416760582 +Step 35801 | loss:0.29543378949165344 lr:4.775456514798382e-08 tokens_per_second_per_gpu:5363.729616950651 mfu:25.301627749795802 +Step 35802 | loss:0.38207879662513733 lr:4.7565280925260244e-08 tokens_per_second_per_gpu:4963.600185307474 mfu:23.414148951614212 +Step 35803 | loss:0.39199021458625793 lr:4.737637240117554e-08 tokens_per_second_per_gpu:3942.5388726824726 mfu:18.597628528938074 +Step 35804 | loss:0.36629170179367065 lr:4.7187839577150784e-08 tokens_per_second_per_gpu:4204.078658833741 mfu:19.831356323502536 +Step 35805 | loss:0.35916242003440857 lr:4.699968245459596e-08 tokens_per_second_per_gpu:3725.4753447581443 mfu:17.573702832862242 +Step 35806 | loss:0.23821377754211426 lr:4.681190103493216e-08 tokens_per_second_per_gpu:4000.1842000623606 mfu:18.86955137349604 +Step 35807 | loss:0.29977065324783325 lr:4.662449531956936e-08 tokens_per_second_per_gpu:5192.456505914255 mfu:24.493703262831083 +Step 35808 | loss:0.27570778131484985 lr:4.643746530991755e-08 tokens_per_second_per_gpu:3951.240857037858 mfu:18.638677273854054 +Step 35809 | loss:0.40734586119651794 lr:4.625081100738671e-08 tokens_per_second_per_gpu:5066.183799035995 mfu:23.89805336014096 +Step 35810 | loss:0.2241889387369156 lr:4.6064532413375715e-08 tokens_per_second_per_gpu:5237.186346750826 mfu:24.70470155375499 +Step 35811 | loss:0.3045911490917206 lr:4.5878629529289006e-08 tokens_per_second_per_gpu:5324.651702547912 mfu:25.117290560178365 +Step 35812 | loss:0.2750479578971863 lr:4.569310235651991e-08 tokens_per_second_per_gpu:5297.5101291008605 mfu:24.989259127398682 +Step 35813 | loss:0.35354384779930115 lr:4.550795089646731e-08 tokens_per_second_per_gpu:5353.977978334181 mfu:25.2556276066408 +Step 35814 | loss:0.41158121824264526 lr:4.532317515052453e-08 tokens_per_second_per_gpu:4143.23168632483 mfu:19.544330772613975 +Step 35815 | loss:0.4353561997413635 lr:4.513877512007936e-08 tokens_per_second_per_gpu:5335.450789083642 mfu:25.168231693881566 +Step 35816 | loss:0.18455931544303894 lr:4.495475080652511e-08 tokens_per_second_per_gpu:5280.133907389662 mfu:24.907292524897827 +Step 35817 | loss:0.37790170311927795 lr:4.477110221123293e-08 tokens_per_second_per_gpu:4216.505725058313 mfu:19.889976915159973 +Step 35818 | loss:0.38867706060409546 lr:4.458782933559613e-08 tokens_per_second_per_gpu:5351.245766776434 mfu:25.242739298559545 +Step 35819 | loss:0.2989823818206787 lr:4.44049321809914e-08 tokens_per_second_per_gpu:4386.604904299898 mfu:20.692363765555154 +Step 35820 | loss:0.2691003084182739 lr:4.422241074878431e-08 tokens_per_second_per_gpu:5331.02399295552 mfu:25.147349741255912 +Step 35821 | loss:0.3313021659851074 lr:4.4040265040362626e-08 tokens_per_second_per_gpu:4339.077460034782 mfu:20.468168701938435 +Step 35822 | loss:0.383870005607605 lr:4.385849505708084e-08 tokens_per_second_per_gpu:4777.380751707246 mfu:22.53572011101031 +Step 35823 | loss:0.21119849383831024 lr:4.367710080032117e-08 tokens_per_second_per_gpu:5101.0387929115 mfu:24.06247031312689 +Step 35824 | loss:0.40028491616249084 lr:4.3496082271432536e-08 tokens_per_second_per_gpu:5330.086703070824 mfu:25.142928384951656 +Step 35825 | loss:0.2689673900604248 lr:4.331543947179162e-08 tokens_per_second_per_gpu:3936.268708943127 mfu:18.56805109677933 +Step 35826 | loss:0.3090454041957855 lr:4.3135172402747336e-08 tokens_per_second_per_gpu:5022.760631154626 mfu:23.69321886848813 +Step 35827 | loss:0.2488275170326233 lr:4.295528106565416e-08 tokens_per_second_per_gpu:5363.977609828268 mfu:25.302797574511473 +Step 35828 | loss:0.3412182629108429 lr:4.277576546187212e-08 tokens_per_second_per_gpu:3777.9121104928695 mfu:17.82105600346772 +Step 35829 | loss:0.3665802478790283 lr:4.2596625592750127e-08 tokens_per_second_per_gpu:4507.965467695531 mfu:21.26484224933963 +Step 35830 | loss:0.30627867579460144 lr:4.241786145963156e-08 tokens_per_second_per_gpu:4865.412840427388 mfu:22.950982493326347 +Step 35831 | loss:0.29400888085365295 lr:4.223947306385978e-08 tokens_per_second_per_gpu:3967.3984728172604 mfu:18.714895504259022 +Step 35832 | loss:0.3672214150428772 lr:4.206146040678372e-08 tokens_per_second_per_gpu:3734.755764676526 mfu:17.61748015702016 +Step 35833 | loss:0.34346938133239746 lr:4.188382348974118e-08 tokens_per_second_per_gpu:4077.297401108955 mfu:19.233307499701617 +Step 35834 | loss:0.2495824247598648 lr:4.1706562314058896e-08 tokens_per_second_per_gpu:4527.928630016824 mfu:21.359011892075884 +Step 35835 | loss:0.37630167603492737 lr:4.152967688108023e-08 tokens_per_second_per_gpu:4704.820720803573 mfu:22.193442065219823 +Step 35836 | loss:0.3291532099246979 lr:4.135316719212634e-08 tokens_per_second_per_gpu:4584.398482110745 mfu:21.62538981915318 +Step 35837 | loss:0.3714221715927124 lr:4.117703324853506e-08 tokens_per_second_per_gpu:4643.6138152297035 mfu:21.904718648652942 +Step 35838 | loss:0.5203781127929688 lr:4.100127505162199e-08 tokens_per_second_per_gpu:4359.770482477201 mfu:20.565781219393013 +Step 35839 | loss:0.2879151701927185 lr:4.0825892602719406e-08 tokens_per_second_per_gpu:4894.770865405509 mfu:23.089469306143712 +Step 35840 | loss:0.27979886531829834 lr:4.065088590313737e-08 tokens_per_second_per_gpu:4802.261822371119 mfu:22.653088366480194 +Step 35841 | loss:0.361633837223053 lr:4.0476254954191495e-08 tokens_per_second_per_gpu:5379.312798142132 mfu:25.375136274240894 +Step 35842 | loss:0.25005456805229187 lr:4.030199975720295e-08 tokens_per_second_per_gpu:3428.4453681472637 mfu:16.172561754648903 +Step 35843 | loss:0.29543206095695496 lr:4.012812031347624e-08 tokens_per_second_per_gpu:4363.660926882075 mfu:20.584133109429068 +Step 35844 | loss:0.2701408267021179 lr:3.995461662431588e-08 tokens_per_second_per_gpu:3924.6025161084744 mfu:18.51301967472053 +Step 35845 | loss:0.4323117733001709 lr:3.9781488691037484e-08 tokens_per_second_per_gpu:5305.9482164034525 mfu:25.02906302489116 +Step 35846 | loss:0.2647378742694855 lr:3.960873651494002e-08 tokens_per_second_per_gpu:3665.503814873898 mfu:17.29080689420014 +Step 35847 | loss:0.2500220239162445 lr:3.9436360097316885e-08 tokens_per_second_per_gpu:5377.164154340388 mfu:25.365000754831637 +Step 35848 | loss:0.26208770275115967 lr:3.926435943946705e-08 tokens_per_second_per_gpu:4014.4094500536858 mfu:18.93665430478304 +Step 35849 | loss:0.2934249937534332 lr:3.909273454268392e-08 tokens_per_second_per_gpu:3434.763756458722 mfu:16.202366670342258 +Step 35850 | loss:0.27670758962631226 lr:3.892148540826091e-08 tokens_per_second_per_gpu:4093.0967035996455 mfu:19.30783550518918 +Step 35851 | loss:0.21856755018234253 lr:3.875061203748587e-08 tokens_per_second_per_gpu:5181.695249863897 mfu:24.442940581981933 +Step 35852 | loss:0.32377973198890686 lr:3.858011443164111e-08 tokens_per_second_per_gpu:4982.637052134817 mfu:23.503949100463196 +Step 35853 | loss:0.35534247756004333 lr:3.8409992592008944e-08 tokens_per_second_per_gpu:5201.744191066389 mfu:24.53751485833619 +Step 35854 | loss:0.2843490242958069 lr:3.824024651987168e-08 tokens_per_second_per_gpu:4214.67101273634 mfu:19.881322264103026 +Step 35855 | loss:0.3246307373046875 lr:3.8070876216506066e-08 tokens_per_second_per_gpu:3691.6417744392684 mfu:17.414104109065754 +Step 35856 | loss:0.2749682068824768 lr:3.7901881683183313e-08 tokens_per_second_per_gpu:5065.657658379408 mfu:23.89557146489498 +Step 35857 | loss:0.3178200423717499 lr:3.773326292117463e-08 tokens_per_second_per_gpu:5345.757210860434 mfu:25.216848843859573 +Step 35858 | loss:0.31770145893096924 lr:3.756501993175121e-08 tokens_per_second_per_gpu:4613.679827714888 mfu:21.763514922280642 +Step 35859 | loss:0.3146442472934723 lr:3.739715271617872e-08 tokens_per_second_per_gpu:4387.118854803071 mfu:20.69478815776762 +Step 35860 | loss:0.23984703421592712 lr:3.7229661275711704e-08 tokens_per_second_per_gpu:4727.033863833896 mfu:22.298225250846862 +Step 35861 | loss:0.3754369616508484 lr:3.706254561162137e-08 tokens_per_second_per_gpu:4398.6574418046575 mfu:20.74921764134005 +Step 35862 | loss:0.282743364572525 lr:3.689580572515672e-08 tokens_per_second_per_gpu:4920.20888384212 mfu:23.209464779282698 +Step 35863 | loss:0.2849864661693573 lr:3.67294416175723e-08 tokens_per_second_per_gpu:4441.579964919693 mfu:20.95169050621127 +Step 35864 | loss:0.27918386459350586 lr:3.656345329012267e-08 tokens_per_second_per_gpu:5006.5853263007 mfu:23.616917195700573 +Step 35865 | loss:0.30425864458084106 lr:3.639784074405683e-08 tokens_per_second_per_gpu:4147.996491281036 mfu:19.56680717055195 +Step 35866 | loss:0.23215624690055847 lr:3.623260398061268e-08 tokens_per_second_per_gpu:4244.687932342443 mfu:20.0229171952996 +Step 35867 | loss:0.27538400888442993 lr:3.6067743001044765e-08 tokens_per_second_per_gpu:5331.81365776863 mfu:25.15107472491006 +Step 35868 | loss:0.2589673399925232 lr:3.5903257806579884e-08 tokens_per_second_per_gpu:5317.984086748428 mfu:25.08583827883964 +Step 35869 | loss:0.3082098364830017 lr:3.573914839846704e-08 tokens_per_second_per_gpu:4890.953749994449 mfu:23.071463321482703 +Step 35870 | loss:0.2899378836154938 lr:3.5575414777933024e-08 tokens_per_second_per_gpu:5384.523622643484 mfu:25.399716622472106 +Step 35871 | loss:0.27430835366249084 lr:3.5412056946215745e-08 tokens_per_second_per_gpu:4475.880636303013 mfu:21.1134926254244 +Step 35872 | loss:0.3567482531070709 lr:3.5249074904536436e-08 tokens_per_second_per_gpu:4246.2264868568345 mfu:20.030174819448455 +Step 35873 | loss:0.3282914459705353 lr:3.508646865412746e-08 tokens_per_second_per_gpu:3781.9757490459297 mfu:17.840224879850737 +Step 35874 | loss:0.27595630288124084 lr:3.4924238196204494e-08 tokens_per_second_per_gpu:5361.104590776361 mfu:25.28924505345603 +Step 35875 | loss:0.28511273860931396 lr:3.4762383531994346e-08 tokens_per_second_per_gpu:4542.71610628553 mfu:21.428766940638983 +Step 35876 | loss:0.2592257559299469 lr:3.460090466271271e-08 tokens_per_second_per_gpu:5371.639837385303 mfu:25.338941609208828 +Step 35877 | loss:0.30792948603630066 lr:3.443980158956972e-08 tokens_per_second_per_gpu:4810.441974141858 mfu:22.691675538060583 +Step 35878 | loss:0.29142701625823975 lr:3.427907431378663e-08 tokens_per_second_per_gpu:4362.640049142362 mfu:20.579317454951138 +Step 35879 | loss:0.3244728744029999 lr:3.4118722836562475e-08 tokens_per_second_per_gpu:3939.2841068642397 mfu:18.582275243253157 +Step 35880 | loss:0.299250066280365 lr:3.3958747159107405e-08 tokens_per_second_per_gpu:5316.792579416809 mfu:25.08021773546428 +Step 35881 | loss:0.2916699945926666 lr:3.379914728262046e-08 tokens_per_second_per_gpu:4182.1269537719 mfu:19.727806385378738 +Step 35882 | loss:0.2914435565471649 lr:3.363992320831177e-08 tokens_per_second_per_gpu:5333.066123965143 mfu:25.156982821651777 +Step 35883 | loss:0.3212532699108124 lr:3.348107493736929e-08 tokens_per_second_per_gpu:3824.699776307918 mfu:18.041761405916255 +Step 35884 | loss:0.33203673362731934 lr:3.332260247099761e-08 tokens_per_second_per_gpu:5341.214552416866 mfu:25.195420348923815 +Step 35885 | loss:0.3054271638393402 lr:3.316450581037356e-08 tokens_per_second_per_gpu:5330.408643826262 mfu:25.144447034423813 +Step 35886 | loss:0.4011470675468445 lr:3.3006784956701734e-08 tokens_per_second_per_gpu:5134.084346728381 mfu:24.218351828633985 +Step 35887 | loss:0.31068849563598633 lr:3.284943991115896e-08 tokens_per_second_per_gpu:5349.966925405474 mfu:25.236706785619894 +Step 35888 | loss:0.19871830940246582 lr:3.2692470674933194e-08 tokens_per_second_per_gpu:5328.982886130639 mfu:25.137721492114213 +Step 35889 | loss:0.25785955786705017 lr:3.253587724920126e-08 tokens_per_second_per_gpu:5328.830186371605 mfu:25.137001181297585 +Step 35890 | loss:0.28155621886253357 lr:3.237965963514e-08 tokens_per_second_per_gpu:5364.096962437868 mfu:25.30336058113461 +Step 35891 | loss:0.34031999111175537 lr:3.2223817833931805e-08 tokens_per_second_per_gpu:5321.358896191138 mfu:25.10175783078277 +Step 35892 | loss:0.3436110019683838 lr:3.20683518467424e-08 tokens_per_second_per_gpu:4787.108835299742 mfu:22.581609141098323 +Step 35893 | loss:0.3146613836288452 lr:3.191326167474307e-08 tokens_per_second_per_gpu:3122.435065478811 mfu:14.729058946220372 +Step 35894 | loss:0.37488141655921936 lr:3.17585473191051e-08 tokens_per_second_per_gpu:5321.831877270835 mfu:25.103988963234595 +Step 35895 | loss:0.36209017038345337 lr:3.160420878097758e-08 tokens_per_second_per_gpu:4499.298200550397 mfu:21.223957271427768 +Step 35896 | loss:0.2562141716480255 lr:3.1450246061537347e-08 tokens_per_second_per_gpu:4781.342884080096 mfu:22.55441016542253 +Step 35897 | loss:0.2509896755218506 lr:3.1296659161933475e-08 tokens_per_second_per_gpu:3645.088750547445 mfu:17.194505552603115 +Step 35898 | loss:0.2949887812137604 lr:3.1143448083326145e-08 tokens_per_second_per_gpu:4849.147529265947 mfu:22.87425624542958 +Step 35899 | loss:0.3392235338687897 lr:3.099061282686444e-08 tokens_per_second_per_gpu:3822.950498984634 mfu:18.033509766324837 +Step 35900 | loss:0.35263070464134216 lr:3.083815339369745e-08 tokens_per_second_per_gpu:4290.94239203627 mfu:20.241107373499982 +Step 35901 | loss:0.3505474030971527 lr:3.0686069784968686e-08 tokens_per_second_per_gpu:4325.5563204840955 mfu:20.404387179733554 +Step 35902 | loss:0.3021547794342041 lr:3.05343620018328e-08 tokens_per_second_per_gpu:5312.577479849148 mfu:25.06033439163348 +Step 35903 | loss:0.30718719959259033 lr:3.03830300454222e-08 tokens_per_second_per_gpu:4101.1489460260855 mfu:19.345819306568917 +Step 35904 | loss:0.2646944224834442 lr:3.023207391687488e-08 tokens_per_second_per_gpu:5351.768347832432 mfu:25.245204402562212 +Step 35905 | loss:0.2966458201408386 lr:3.0081493617328813e-08 tokens_per_second_per_gpu:5303.778313228881 mfu:25.018827221393046 +Step 35906 | loss:0.3590240478515625 lr:2.993128914791643e-08 tokens_per_second_per_gpu:5333.45476589352 mfu:25.158816111937085 +Step 35907 | loss:0.3123590350151062 lr:2.9781460509764602e-08 tokens_per_second_per_gpu:4656.393846542277 mfu:21.965004236852543 +Step 35908 | loss:0.34551993012428284 lr:2.963200770400576e-08 tokens_per_second_per_gpu:5374.555834041253 mfu:25.35269686295528 +Step 35909 | loss:0.26462647318840027 lr:2.9482930731761227e-08 tokens_per_second_per_gpu:5030.4248211064505 mfu:23.729372160136535 +Step 35910 | loss:0.3076641857624054 lr:2.933422959415233e-08 tokens_per_second_per_gpu:4368.820766614975 mfu:20.608472953808594 +Step 35911 | loss:0.25316059589385986 lr:2.918590429229484e-08 tokens_per_second_per_gpu:3161.5221270370776 mfu:14.913439284531098 +Step 35912 | loss:0.25872257351875305 lr:2.9037954827310086e-08 tokens_per_second_per_gpu:3912.0099289979003 mfu:18.453618292803043 +Step 35913 | loss:0.32301291823387146 lr:2.889038120030274e-08 tokens_per_second_per_gpu:4689.296644755524 mfu:22.120212349820676 +Step 35914 | loss:0.3510035574436188 lr:2.8743183412388575e-08 tokens_per_second_per_gpu:4976.427708249714 mfu:23.47465856593385 +Step 35915 | loss:0.25546616315841675 lr:2.8596361464672262e-08 tokens_per_second_per_gpu:5305.901813997617 mfu:25.028844136825764 +Step 35916 | loss:0.3804944157600403 lr:2.8449915358264023e-08 tokens_per_second_per_gpu:5363.877689999167 mfu:25.302326235629522 +Step 35917 | loss:0.27375081181526184 lr:2.830384509425188e-08 tokens_per_second_per_gpu:4662.974559266285 mfu:21.996046581557053 +Step 35918 | loss:0.21568769216537476 lr:2.8158150673751604e-08 tokens_per_second_per_gpu:3818.99634026979 mfu:18.01485732501787 +Step 35919 | loss:0.3258684277534485 lr:2.8012832097845666e-08 tokens_per_second_per_gpu:3883.7796223534096 mfu:18.320451119773995 +Step 35920 | loss:0.28013744950294495 lr:2.7867889367633183e-08 tokens_per_second_per_gpu:5343.237552090939 mfu:25.20496318354575 +Step 35921 | loss:0.29182153940200806 lr:2.7723322484202175e-08 tokens_per_second_per_gpu:5355.413017687475 mfu:25.26239693211325 +Step 35922 | loss:0.2340003401041031 lr:2.757913144864066e-08 tokens_per_second_per_gpu:4066.5017212459 mfu:19.18238243585467 +Step 35923 | loss:0.3359137773513794 lr:2.7435316262031107e-08 tokens_per_second_per_gpu:5321.23510747468 mfu:25.101173898286707 +Step 35924 | loss:0.2794997990131378 lr:2.7291876925461535e-08 tokens_per_second_per_gpu:5350.588801074506 mfu:25.239640279253646 +Step 35925 | loss:0.27299052476882935 lr:2.7148813440003305e-08 tokens_per_second_per_gpu:5364.649854088343 mfu:25.305968665382945 +Step 35926 | loss:0.3654938042163849 lr:2.700612580673334e-08 tokens_per_second_per_gpu:4474.973318098696 mfu:21.10921265065019 +Step 35927 | loss:0.3772038221359253 lr:2.6863814026734102e-08 tokens_per_second_per_gpu:3807.173688366936 mfu:17.95908785883969 +Step 35928 | loss:0.2666812241077423 lr:2.6721878101060305e-08 tokens_per_second_per_gpu:3895.6801675501715 mfu:18.376588021909505 +Step 35929 | loss:0.3030795156955719 lr:2.6580318030794415e-08 tokens_per_second_per_gpu:5342.479258754561 mfu:25.20138618449987 +Step 35930 | loss:0.26861119270324707 lr:2.6439133816991146e-08 tokens_per_second_per_gpu:5352.719932338514 mfu:25.24969318903446 +Step 35931 | loss:0.3382100462913513 lr:2.629832546071631e-08 tokens_per_second_per_gpu:5198.688133000538 mfu:24.5230989110223 +Step 35932 | loss:0.2783474326133728 lr:2.6157892963030172e-08 tokens_per_second_per_gpu:4710.241848596154 mfu:22.219014449957204 +Step 35933 | loss:0.3111962080001831 lr:2.601783632498189e-08 tokens_per_second_per_gpu:4686.828894523451 mfu:22.108571550934382 +Step 35934 | loss:0.3371681272983551 lr:2.5878155547637285e-08 tokens_per_second_per_gpu:3894.158394503952 mfu:18.369409558809235 +Step 35935 | loss:0.2609589695930481 lr:2.5738850632034407e-08 tokens_per_second_per_gpu:4238.258916404295 mfu:19.992590430216126 +Step 35936 | loss:0.31229040026664734 lr:2.559992157922797e-08 tokens_per_second_per_gpu:4523.7965080925505 mfu:21.339519967946387 +Step 35937 | loss:0.3365817070007324 lr:2.5461368390261587e-08 tokens_per_second_per_gpu:4718.450139442949 mfu:22.257734358360366 +Step 35938 | loss:0.3754328787326813 lr:2.5323191066178865e-08 tokens_per_second_per_gpu:5351.678206011833 mfu:25.244779188214665 +Step 35939 | loss:0.29044413566589355 lr:2.518538960801786e-08 tokens_per_second_per_gpu:4841.935825768057 mfu:22.84023740958654 +Step 35940 | loss:0.33279508352279663 lr:2.5047964016811088e-08 tokens_per_second_per_gpu:4171.468657197795 mfu:19.67752938194577 +Step 35941 | loss:0.3726086914539337 lr:2.4910914293596598e-08 tokens_per_second_per_gpu:5097.6672786405115 mfu:24.046566305070986 +Step 35942 | loss:0.35959312319755554 lr:2.4774240439401353e-08 tokens_per_second_per_gpu:5341.69550334415 mfu:25.1976890765065 +Step 35943 | loss:0.34889325499534607 lr:2.4637942455257855e-08 tokens_per_second_per_gpu:4146.209988395905 mfu:19.55837993163397 +Step 35944 | loss:0.28261420130729675 lr:2.4502020342193065e-08 tokens_per_second_per_gpu:3238.468351286828 mfu:15.276407752696782 +Step 35945 | loss:0.31225642561912537 lr:2.4366474101222836e-08 tokens_per_second_per_gpu:5375.506271164851 mfu:25.357180237028587 +Step 35946 | loss:0.2141571342945099 lr:2.423130373336857e-08 tokens_per_second_per_gpu:5314.777596002007 mfu:25.07071271490504 +Step 35947 | loss:0.16936832666397095 lr:2.4096509239646125e-08 tokens_per_second_per_gpu:3715.2969243701814 mfu:17.525689487274626 +Step 35948 | loss:0.32338595390319824 lr:2.3962090621071353e-08 tokens_per_second_per_gpu:4173.42990410293 mfu:19.686780918220517 +Step 35949 | loss:0.43166229128837585 lr:2.382804787866011e-08 tokens_per_second_per_gpu:5335.846378504199 mfu:25.170097756672845 +Step 35950 | loss:0.22383548319339752 lr:2.3694381013406043e-08 tokens_per_second_per_gpu:3995.058811571652 mfu:18.84537404150392 +Step 35951 | loss:0.2563874125480652 lr:2.3561090026330556e-08 tokens_per_second_per_gpu:5122.92827233746 mfu:24.165726722309444 +Step 35952 | loss:0.36659589409828186 lr:2.3428174918427305e-08 tokens_per_second_per_gpu:5119.603872816383 mfu:24.150044962567478 +Step 35953 | loss:0.29367169737815857 lr:2.3295635690695482e-08 tokens_per_second_per_gpu:4315.254865356988 mfu:20.355793458289504 +Step 35954 | loss:0.32552334666252136 lr:2.3163472344139847e-08 tokens_per_second_per_gpu:3258.063704385364 mfu:15.368842376574568 +Step 35955 | loss:0.31089648604393005 lr:2.3031684879742944e-08 tokens_per_second_per_gpu:5367.751896575467 mfu:25.320601529057733 +Step 35956 | loss:0.277750164270401 lr:2.290027329850397e-08 tokens_per_second_per_gpu:4158.453349724888 mfu:19.61613395595563 +Step 35957 | loss:0.23026983439922333 lr:2.276923760141103e-08 tokens_per_second_per_gpu:4849.439112149121 mfu:22.87563169163868 +Step 35958 | loss:0.28379112482070923 lr:2.263857778944667e-08 tokens_per_second_per_gpu:5222.431051126866 mfu:24.635098306783743 +Step 35959 | loss:0.31644493341445923 lr:2.2508293863593434e-08 tokens_per_second_per_gpu:5379.54483304744 mfu:25.376230822478473 +Step 35960 | loss:0.29306909441947937 lr:2.237838582483387e-08 tokens_per_second_per_gpu:4614.422799867023 mfu:21.767019648686386 +Step 35961 | loss:0.34353920817375183 lr:2.2248853674139423e-08 tokens_per_second_per_gpu:4142.2965442673385 mfu:19.5399195479771 +Step 35962 | loss:0.28770947456359863 lr:2.2119697412492646e-08 tokens_per_second_per_gpu:3256.030588426092 mfu:15.359251821709329 +Step 35963 | loss:0.4220413267612457 lr:2.1990917040859428e-08 tokens_per_second_per_gpu:5328.122235053217 mfu:25.133661654139797 +Step 35964 | loss:0.3752492666244507 lr:2.1862512560211212e-08 tokens_per_second_per_gpu:4708.739096142295 mfu:22.211925710236343 +Step 35965 | loss:0.32180237770080566 lr:2.1734483971513898e-08 tokens_per_second_per_gpu:4544.561212104085 mfu:21.437470619592624 +Step 35966 | loss:0.3412732183933258 lr:2.1606831275727825e-08 tokens_per_second_per_gpu:4475.711246369917 mfu:21.112693584209083 +Step 35967 | loss:0.284851610660553 lr:2.1479554473813334e-08 tokens_per_second_per_gpu:5353.787448751361 mfu:25.25472884609094 +Step 35968 | loss:0.3396141827106476 lr:2.1352653566725222e-08 tokens_per_second_per_gpu:5330.708322712192 mfu:25.145860670859236 +Step 35969 | loss:0.3408209979534149 lr:2.122612855542383e-08 tokens_per_second_per_gpu:4462.453074119855 mfu:21.050152523628096 +Step 35970 | loss:0.29710811376571655 lr:2.1099979440858398e-08 tokens_per_second_per_gpu:5303.634556619324 mfu:25.01814909693836 +Step 35971 | loss:0.3134363293647766 lr:2.0974206223978165e-08 tokens_per_second_per_gpu:5235.350689976678 mfu:24.696042447555993 +Step 35972 | loss:0.3581942617893219 lr:2.0848808905726823e-08 tokens_per_second_per_gpu:5116.7143811687865 mfu:24.136414737467657 +Step 35973 | loss:0.3114818334579468 lr:2.0723787487048063e-08 tokens_per_second_per_gpu:3954.9445335779556 mfu:18.656148148006075 +Step 35974 | loss:0.1968446671962738 lr:2.0599141968885572e-08 tokens_per_second_per_gpu:4873.205429924229 mfu:22.987741467535812 +Step 35975 | loss:0.3223757743835449 lr:2.0474872352171936e-08 tokens_per_second_per_gpu:3665.696181510096 mfu:17.291714320444218 +Step 35976 | loss:0.3561837375164032 lr:2.0350978637845298e-08 tokens_per_second_per_gpu:3979.2644281774633 mfu:18.770869240233843 +Step 35977 | loss:0.28448328375816345 lr:2.022746082683824e-08 tokens_per_second_per_gpu:4753.195829183109 mfu:22.42163570509067 +Step 35978 | loss:0.2614036798477173 lr:2.0104318920072252e-08 tokens_per_second_per_gpu:4856.679947752329 mfu:22.909787948592964 +Step 35979 | loss:0.39905980229377747 lr:1.998155291848547e-08 tokens_per_second_per_gpu:3963.061600469453 mfu:18.69443773744769 +Step 35980 | loss:0.2760242223739624 lr:1.9859162822993826e-08 tokens_per_second_per_gpu:4746.389447188364 mfu:22.389528839932943 +Step 35981 | loss:0.3404000401496887 lr:1.973714863451326e-08 tokens_per_second_per_gpu:4281.432673711649 mfu:20.19624840963749 +Step 35982 | loss:0.40480107069015503 lr:1.961551035397635e-08 tokens_per_second_per_gpu:4908.4110432867255 mfu:23.153812352463742 +Step 35983 | loss:0.4343401789665222 lr:1.949424798228239e-08 tokens_per_second_per_gpu:4165.374420091266 mfu:19.648781825730204 +Step 35984 | loss:0.3278507590293884 lr:1.937336152035285e-08 tokens_per_second_per_gpu:4948.497323449417 mfu:23.342906175415813 +Step 35985 | loss:0.4031858444213867 lr:1.9252850969092574e-08 tokens_per_second_per_gpu:5388.7685312890435 mfu:25.41974058080931 +Step 35986 | loss:0.32472577691078186 lr:1.9132716329406387e-08 tokens_per_second_per_gpu:3959.671708490445 mfu:18.678447038607494 +Step 35987 | loss:0.3433966040611267 lr:1.9012957602204674e-08 tokens_per_second_per_gpu:3546.7039765150203 mfu:16.730407787318782 +Step 35988 | loss:0.28461551666259766 lr:1.8893574788381163e-08 tokens_per_second_per_gpu:4677.832615906161 mfu:22.06613456977322 +Step 35989 | loss:0.3009788691997528 lr:1.8774567888835137e-08 tokens_per_second_per_gpu:4429.5102944067 mfu:20.894755811103348 +Step 35990 | loss:0.23602014780044556 lr:1.8655936904465875e-08 tokens_per_second_per_gpu:4787.617384429104 mfu:22.584008054108953 +Step 35991 | loss:0.22268611192703247 lr:1.853768183616711e-08 tokens_per_second_per_gpu:4925.028603966489 mfu:23.232200221438397 +Step 35992 | loss:0.2644764184951782 lr:1.8419802684821465e-08 tokens_per_second_per_gpu:4562.310697052366 mfu:21.521197968468073 +Step 35993 | loss:0.2935107350349426 lr:1.8302299451311565e-08 tokens_per_second_per_gpu:4540.575716390834 mfu:21.418670356317527 +Step 35994 | loss:0.3860234320163727 lr:1.8185172136536697e-08 tokens_per_second_per_gpu:3978.661838274939 mfu:18.7680267208512 +Step 35995 | loss:0.24243751168251038 lr:1.8068420741362836e-08 tokens_per_second_per_gpu:5341.136617124205 mfu:25.1950527148512 +Step 35996 | loss:0.27380192279815674 lr:1.795204526667815e-08 tokens_per_second_per_gpu:3916.7177018505668 mfu:18.475825660577947 +Step 35997 | loss:0.3505680561065674 lr:1.7836045713354177e-08 tokens_per_second_per_gpu:4993.502118356513 mfu:23.555201471602388 +Step 35998 | loss:0.3572491407394409 lr:1.7720422082262435e-08 tokens_per_second_per_gpu:5332.114978278205 mfu:25.152496105164307 +Step 35999 | loss:0.17746031284332275 lr:1.7605174374268896e-08 tokens_per_second_per_gpu:5336.136010779126 mfu:25.171464001530037 +Step 36000 | loss:0.3376941978931427 lr:1.749030259024509e-08 tokens_per_second_per_gpu:5197.049045339277 mfu:24.515367054866935 +Step 36001 | loss:0.31298720836639404 lr:1.7375806731056988e-08 tokens_per_second_per_gpu:4144.708189987107 mfu:19.55129569230644 +Step 36002 | loss:0.32573091983795166 lr:1.7261686797565014e-08 tokens_per_second_per_gpu:5145.653319347716 mfu:24.272924646349658 +Step 36003 | loss:0.3084086775779724 lr:1.7147942790624038e-08 tokens_per_second_per_gpu:5364.6703833691145 mfu:25.306065505502996 +Step 36004 | loss:0.3555644452571869 lr:1.7034574711094486e-08 tokens_per_second_per_gpu:5100.045425079583 mfu:24.057784427577708 +Step 36005 | loss:0.23773100972175598 lr:1.6921582559820125e-08 tokens_per_second_per_gpu:4514.500383354699 mfu:21.295668561475505 +Step 36006 | loss:0.41589590907096863 lr:1.680896633766138e-08 tokens_per_second_per_gpu:5175.723656235803 mfu:24.41477155597928 +Step 36007 | loss:0.2924026548862457 lr:1.669672604545647e-08 tokens_per_second_per_gpu:5065.535517219151 mfu:23.89499530420262 +Step 36008 | loss:0.25325706601142883 lr:1.6584861684054707e-08 tokens_per_second_per_gpu:5078.316871383824 mfu:23.95528713252127 +Step 36009 | loss:0.2195816934108734 lr:1.6473373254294323e-08 tokens_per_second_per_gpu:4769.114702294569 mfu:22.49672774559725 +Step 36010 | loss:0.3215319514274597 lr:1.6362260757019078e-08 tokens_per_second_per_gpu:5331.105142959594 mfu:25.14773253967074 +Step 36011 | loss:0.340040922164917 lr:1.625152419306164e-08 tokens_per_second_per_gpu:5332.472167711239 mfu:25.15418102866273 +Step 36012 | loss:0.2977847754955292 lr:1.6141163563249128e-08 tokens_per_second_per_gpu:4736.47750638009 mfu:22.342772523988824 +Step 36013 | loss:0.33033430576324463 lr:1.6031178868419762e-08 tokens_per_second_per_gpu:5324.24570523212 mfu:25.11537540156949 +Step 36014 | loss:0.26231831312179565 lr:1.5921570109400652e-08 tokens_per_second_per_gpu:5331.50335853187 mfu:25.149610990467753 +Step 36015 | loss:0.32139483094215393 lr:1.5812337287007818e-08 tokens_per_second_per_gpu:5232.039161434457 mfu:24.680421402417306 +Step 36016 | loss:0.4449172616004944 lr:1.5703480402073923e-08 tokens_per_second_per_gpu:5355.955701609741 mfu:25.264956864766003 +Step 36017 | loss:0.281014084815979 lr:1.5594999455409433e-08 tokens_per_second_per_gpu:5295.584636814371 mfu:24.980176251760447 +Step 36018 | loss:0.24836762249469757 lr:1.548689444783591e-08 tokens_per_second_per_gpu:4154.046891155993 mfu:19.595347939066794 +Step 36019 | loss:0.3401232659816742 lr:1.5379165380158268e-08 tokens_per_second_per_gpu:4484.013923943462 mfu:21.151858730906362 +Step 36020 | loss:0.35907015204429626 lr:1.527181225319807e-08 tokens_per_second_per_gpu:5313.685311844086 mfu:25.06556022416923 +Step 36021 | loss:0.2596666216850281 lr:1.516483506775468e-08 tokens_per_second_per_gpu:4051.4272894973924 mfu:19.11127376933369 +Step 36022 | loss:0.2770795226097107 lr:1.5058233824633005e-08 tokens_per_second_per_gpu:3517.5114677995393 mfu:16.592701742952272 +Step 36023 | loss:0.29841598868370056 lr:1.495200852463796e-08 tokens_per_second_per_gpu:4160.4831635143455 mfu:19.625708933922443 +Step 36024 | loss:0.4001609683036804 lr:1.4846159168563355e-08 tokens_per_second_per_gpu:4406.127742471352 mfu:20.784456324149566 +Step 36025 | loss:0.33729907870292664 lr:1.474068575720855e-08 tokens_per_second_per_gpu:3329.585468058216 mfu:15.706222738690615 +Step 36026 | loss:0.32088562846183777 lr:1.4635588291372904e-08 tokens_per_second_per_gpu:3236.335173173525 mfu:15.266345187578617 +Step 36027 | loss:0.38840726017951965 lr:1.4530866771839124e-08 tokens_per_second_per_gpu:5328.867370195772 mfu:25.137176583740395 +Step 36028 | loss:0.27387866377830505 lr:1.442652119939547e-08 tokens_per_second_per_gpu:3612.146486414344 mfu:17.039111272157445 +Step 36029 | loss:0.27301889657974243 lr:1.43225515748302e-08 tokens_per_second_per_gpu:4851.890917468836 mfu:22.887197275652902 +Step 36030 | loss:0.42496630549430847 lr:1.4218957898920471e-08 tokens_per_second_per_gpu:5300.551782092046 mfu:25.003607123517618 +Step 36031 | loss:0.3393140733242035 lr:1.4115740172448988e-08 tokens_per_second_per_gpu:3399.724868986297 mfu:16.03708225988417 +Step 36032 | loss:0.36787617206573486 lr:1.4012898396192908e-08 tokens_per_second_per_gpu:4369.769781912389 mfu:20.6129496208852 +Step 36033 | loss:0.3122923672199249 lr:1.3910432570923836e-08 tokens_per_second_per_gpu:5369.3108816500035 mfu:25.3279555276446 +Step 36034 | loss:0.43526601791381836 lr:1.3808342697413378e-08 tokens_per_second_per_gpu:4925.502343226603 mfu:23.23443493035661 +Step 36035 | loss:0.2609159052371979 lr:1.3706628776427588e-08 tokens_per_second_per_gpu:4759.598009001213 mfu:22.45183587961718 +Step 36036 | loss:0.29574936628341675 lr:1.3605290808732518e-08 tokens_per_second_per_gpu:4933.974618562919 mfu:23.27440009863724 +Step 36037 | loss:0.262613981962204 lr:1.3504328795088672e-08 tokens_per_second_per_gpu:5364.564455340786 mfu:25.305565824920187 +Step 36038 | loss:0.30299925804138184 lr:1.3403742736256553e-08 tokens_per_second_per_gpu:4055.366356716363 mfu:19.129855021479745 +Step 36039 | loss:0.3022192120552063 lr:1.3303532632996663e-08 tokens_per_second_per_gpu:4988.583140517312 mfu:23.531997813870714 +Step 36040 | loss:0.30264055728912354 lr:1.3203698486058402e-08 tokens_per_second_per_gpu:4481.999510116576 mfu:21.142356397190504 +Step 36041 | loss:0.2078733742237091 lr:1.3104240296191172e-08 tokens_per_second_per_gpu:5308.73339022514 mfu:25.042201165007715 +Step 36042 | loss:0.2868395745754242 lr:1.3005158064149925e-08 tokens_per_second_per_gpu:4768.893303900439 mfu:22.49568337159768 +Step 36043 | loss:0.3338632881641388 lr:1.2906451790667406e-08 tokens_per_second_per_gpu:5359.5800482510285 mfu:25.282053526250262 +Step 36044 | loss:0.2776199281215668 lr:1.280812147649857e-08 tokens_per_second_per_gpu:5367.317766372413 mfu:25.31855366281967 +Step 36045 | loss:0.28185543417930603 lr:1.271016712238171e-08 tokens_per_second_per_gpu:4155.261141760869 mfu:19.601075766342507 +Step 36046 | loss:0.3095186650753021 lr:1.2612588729044028e-08 tokens_per_second_per_gpu:4433.146512066733 mfu:20.911908470208427 +Step 36047 | loss:0.24445997178554535 lr:1.251538629722937e-08 tokens_per_second_per_gpu:5328.5326822924235 mfu:25.135597803796657 +Step 36048 | loss:0.2705533802509308 lr:1.2418559827664934e-08 tokens_per_second_per_gpu:4277.966262453442 mfu:20.179896756301783 +Step 36049 | loss:0.2704126238822937 lr:1.2322109321077913e-08 tokens_per_second_per_gpu:3924.2926922066263 mfu:18.511558182514104 +Step 36050 | loss:0.2823868691921234 lr:1.2226034778195505e-08 tokens_per_second_per_gpu:4982.11769705642 mfu:23.501499213947284 +Step 36051 | loss:0.3384980261325836 lr:1.2130336199739357e-08 tokens_per_second_per_gpu:4269.640903279799 mfu:20.140624616626965 +Step 36052 | loss:0.3797965347766876 lr:1.2035013586425558e-08 tokens_per_second_per_gpu:4282.490752918727 mfu:20.201239549784205 +Step 36053 | loss:0.358704537153244 lr:1.194006693898131e-08 tokens_per_second_per_gpu:5360.493125493332 mfu:25.286360667389246 +Step 36054 | loss:0.22572334110736847 lr:1.1845496258111599e-08 tokens_per_second_per_gpu:3659.5070031892624 mfu:17.262518910322136 +Step 36055 | loss:0.3286273777484894 lr:1.1751301544526971e-08 tokens_per_second_per_gpu:4486.390483973267 mfu:21.163069370049904 +Step 36056 | loss:0.2764221727848053 lr:1.165748279894352e-08 tokens_per_second_per_gpu:5337.410624238808 mfu:25.17747657069084 +Step 36057 | loss:0.36297839879989624 lr:1.1564040022060685e-08 tokens_per_second_per_gpu:3652.089487215643 mfu:17.22752922191585 +Step 36058 | loss:0.4547156095504761 lr:1.147097321458901e-08 tokens_per_second_per_gpu:5308.684515499769 mfu:25.04197061458925 +Step 36059 | loss:0.3418709933757782 lr:1.137828237721683e-08 tokens_per_second_per_gpu:5333.006016704271 mfu:25.15669928544689 +Step 36060 | loss:0.3902652859687805 lr:1.1285967510649142e-08 tokens_per_second_per_gpu:4364.48064978019 mfu:20.587999882198623 +Step 36061 | loss:0.22869698703289032 lr:1.119402861557428e-08 tokens_per_second_per_gpu:5374.621142509882 mfu:25.35300493414458 +Step 36062 | loss:0.3125670850276947 lr:1.1102465692691688e-08 tokens_per_second_per_gpu:5055.857230498026 mfu:23.849341174453084 +Step 36063 | loss:0.1944604218006134 lr:1.101127874268415e-08 tokens_per_second_per_gpu:4327.646044277945 mfu:20.41424476341279 +Step 36064 | loss:0.28537943959236145 lr:1.0920467766240006e-08 tokens_per_second_per_gpu:3763.6662129217125 mfu:17.75385567402381 +Step 36065 | loss:0.2656049132347107 lr:1.0830032764042042e-08 tokens_per_second_per_gpu:5363.09842345353 mfu:25.29865030237717 +Step 36066 | loss:0.2294611781835556 lr:1.0739973736773046e-08 tokens_per_second_per_gpu:4500.67280201651 mfu:21.230441501074623 +Step 36067 | loss:0.4101186692714691 lr:1.0650290685104702e-08 tokens_per_second_per_gpu:4270.158028895782 mfu:20.143063986388285 +Step 36068 | loss:0.34208613634109497 lr:1.0560983609714248e-08 tokens_per_second_per_gpu:4692.991343942644 mfu:22.137640876267007 +Step 36069 | loss:0.2926948070526123 lr:1.0472052511273366e-08 tokens_per_second_per_gpu:4848.272665136295 mfu:22.870129362062798 +Step 36070 | loss:0.32129359245300293 lr:1.0383497390448194e-08 tokens_per_second_per_gpu:5325.529037246097 mfu:25.121429097638444 +Step 36071 | loss:0.33830776810646057 lr:1.0295318247910413e-08 tokens_per_second_per_gpu:5304.364131713182 mfu:25.021590627134433 +Step 36072 | loss:0.26186075806617737 lr:1.0207515084320607e-08 tokens_per_second_per_gpu:5338.823105788493 mfu:25.18413948715491 +Step 36073 | loss:0.3058801293373108 lr:1.012008790033936e-08 tokens_per_second_per_gpu:5335.451339838988 mfu:25.168234291888503 +Step 36074 | loss:0.25333476066589355 lr:1.00330366966217e-08 tokens_per_second_per_gpu:4460.372111661815 mfu:21.040336268664717 +Step 36075 | loss:0.30333107709884644 lr:9.946361473822663e-09 tokens_per_second_per_gpu:5345.524746546107 mfu:25.21575227002639 +Step 36076 | loss:0.2871432602405548 lr:9.86006223260283e-09 tokens_per_second_per_gpu:4399.201519419693 mfu:20.75178414827938 +Step 36077 | loss:0.2996600270271301 lr:9.774138973595027e-09 tokens_per_second_per_gpu:4857.8812339665565 mfu:22.915454620626033 +Step 36078 | loss:0.2385093718767166 lr:9.688591697459836e-09 tokens_per_second_per_gpu:4177.085012822052 mfu:19.70402268967447 +Step 36079 | loss:0.2536219656467438 lr:9.603420404835638e-09 tokens_per_second_per_gpu:5058.753715609889 mfu:23.86300438891706 +Step 36080 | loss:0.3720705807209015 lr:9.518625096366363e-09 tokens_per_second_per_gpu:4121.564033579321 mfu:19.442120757730645 +Step 36081 | loss:0.28588441014289856 lr:9.434205772679283e-09 tokens_per_second_per_gpu:5352.223673273249 mfu:25.247352250353075 +Step 36082 | loss:0.27415984869003296 lr:9.35016243441833e-09 tokens_per_second_per_gpu:5200.54066483862 mfu:24.53183761977705 +Step 36083 | loss:0.30823948979377747 lr:9.26649508221633e-09 tokens_per_second_per_gpu:5081.19883492603 mfu:23.96888184626354 +Step 36084 | loss:0.32805588841438293 lr:9.183203716700562e-09 tokens_per_second_per_gpu:4466.714812185105 mfu:21.070255869209376 +Step 36085 | loss:0.29786285758018494 lr:9.100288338492747e-09 tokens_per_second_per_gpu:4853.9434153379 mfu:22.896879258293524 +Step 36086 | loss:0.3472054600715637 lr:9.017748948220162e-09 tokens_per_second_per_gpu:5290.603444523614 mfu:24.956679117845905 +Step 36087 | loss:0.2751806676387787 lr:8.935585546510083e-09 tokens_per_second_per_gpu:3642.1108182977237 mfu:17.180458138093545 +Step 36088 | loss:0.26275157928466797 lr:8.853798133967584e-09 tokens_per_second_per_gpu:5391.627723708613 mfu:25.43322787928095 +Step 36089 | loss:0.3511504530906677 lr:8.772386711219937e-09 tokens_per_second_per_gpu:4525.733734866864 mfu:21.348658197165797 +Step 36090 | loss:0.36477240920066833 lr:8.691351278866666e-09 tokens_per_second_per_gpu:4651.027500785474 mfu:21.939690268324703 +Step 36091 | loss:0.3629648685455322 lr:8.610691837529495e-09 tokens_per_second_per_gpu:4146.787483633896 mfu:19.561104075202422 +Step 36092 | loss:0.3545987904071808 lr:8.530408387813493e-09 tokens_per_second_per_gpu:4634.199446661426 mfu:21.86030946585947 +Step 36093 | loss:0.32282567024230957 lr:8.450500930318183e-09 tokens_per_second_per_gpu:4051.8471838697983 mfu:19.113254482729698 +Step 36094 | loss:0.30438166856765747 lr:8.370969465643086e-09 tokens_per_second_per_gpu:5341.018522431153 mfu:25.194495642035037 +Step 36095 | loss:0.22687795758247375 lr:8.291813994387721e-09 tokens_per_second_per_gpu:4351.635766817947 mfu:20.527408377702706 +Step 36096 | loss:0.2974499762058258 lr:8.213034517151608e-09 tokens_per_second_per_gpu:5386.447117090487 mfu:25.408790073960613 +Step 36097 | loss:0.2510848939418793 lr:8.134631034523167e-09 tokens_per_second_per_gpu:3591.722490259694 mfu:16.942767797602887 +Step 36098 | loss:0.2689374089241028 lr:8.056603547090813e-09 tokens_per_second_per_gpu:2995.7388280635028 mfu:14.131411177725228 +Step 36099 | loss:0.3427286744117737 lr:7.978952055442968e-09 tokens_per_second_per_gpu:4453.3374583301775 mfu:21.00715260866376 +Step 36100 | loss:0.29050177335739136 lr:7.901676560168047e-09 tokens_per_second_per_gpu:5178.702993174728 mfu:24.428825596647492 +Step 36101 | loss:0.3049415647983551 lr:7.824777061843368e-09 tokens_per_second_per_gpu:4575.8649747034315 mfu:21.585135808746383 +Step 36102 | loss:0.3090854585170746 lr:7.748253561046248e-09 tokens_per_second_per_gpu:4003.257010388421 mfu:18.88404634407957 +Step 36103 | loss:0.2616516947746277 lr:7.672106058354e-09 tokens_per_second_per_gpu:5209.0368200020675 mfu:24.571915433276583 +Step 36104 | loss:0.27628082036972046 lr:7.596334554343943e-09 tokens_per_second_per_gpu:4941.275855913265 mfu:23.30884128093837 +Step 36105 | loss:0.3261978328227997 lr:7.520939049571186e-09 tokens_per_second_per_gpu:4209.110642378958 mfu:19.855093048430245 +Step 36106 | loss:0.28548818826675415 lr:7.4459195446185955e-09 tokens_per_second_per_gpu:5371.531911820323 mfu:25.338432505904827 +Step 36107 | loss:0.3151583671569824 lr:7.371276040046837e-09 tokens_per_second_per_gpu:5082.924700608769 mfu:23.97702304915207 +Step 36108 | loss:0.3343050181865692 lr:7.297008536411021e-09 tokens_per_second_per_gpu:4098.617573977669 mfu:19.33387839271993 +Step 36109 | loss:0.28884848952293396 lr:7.223117034271809e-09 tokens_per_second_per_gpu:4494.310205544042 mfu:21.200428047943088 +Step 36110 | loss:0.31245821714401245 lr:7.149601534189864e-09 tokens_per_second_per_gpu:4600.130069644034 mfu:21.69959840163331 +Step 36111 | loss:0.3007371723651886 lr:7.076462036714748e-09 tokens_per_second_per_gpu:5345.686039161755 mfu:25.21651311481376 +Step 36112 | loss:0.3167736232280731 lr:7.00369854239602e-09 tokens_per_second_per_gpu:3554.892715203148 mfu:16.76903546485357 +Step 36113 | loss:0.2428552210330963 lr:6.931311051777689e-09 tokens_per_second_per_gpu:4545.6901476930525 mfu:21.442796001382053 +Step 36114 | loss:0.489846795797348 lr:6.859299565414867e-09 tokens_per_second_per_gpu:4855.853110064794 mfu:22.9058876141478 +Step 36115 | loss:0.31347447633743286 lr:6.7876640838404615e-09 tokens_per_second_per_gpu:3381.8678696893994 mfu:15.952847747482455 +Step 36116 | loss:0.24354463815689087 lr:6.71640460759293e-09 tokens_per_second_per_gpu:4625.706514241459 mfu:21.820246854590795 +Step 36117 | loss:0.2562309205532074 lr:6.6455211372107306e-09 tokens_per_second_per_gpu:3614.9952071764383 mfu:17.052549174034127 +Step 36118 | loss:0.34797751903533936 lr:6.575013673226771e-09 tokens_per_second_per_gpu:4498.472080257033 mfu:21.220060321053314 +Step 36119 | loss:0.3963611423969269 lr:6.5048822161739575e-09 tokens_per_second_per_gpu:4860.974810127175 mfu:22.930047547194192 +Step 36120 | loss:0.32629576325416565 lr:6.435126766574096e-09 tokens_per_second_per_gpu:4967.973204486949 mfu:23.434777229197802 +Step 36121 | loss:0.34300053119659424 lr:6.3657473249545405e-09 tokens_per_second_per_gpu:4565.204374607626 mfu:21.534847939208532 +Step 36122 | loss:0.27762311697006226 lr:6.2967438918426495e-09 tokens_per_second_per_gpu:5343.120265140003 mfu:25.204409920986965 +Step 36123 | loss:0.3147185742855072 lr:6.228116467743572e-09 tokens_per_second_per_gpu:5291.984170589493 mfu:24.963192238274818 +Step 36124 | loss:0.3425733149051666 lr:6.159865053190217e-09 tokens_per_second_per_gpu:3913.111158982345 mfu:18.458812982528876 +Step 36125 | loss:0.30893558263778687 lr:6.091989648682184e-09 tokens_per_second_per_gpu:4003.61790997435 mfu:18.885748769002216 +Step 36126 | loss:0.2913815379142761 lr:6.0244902547412775e-09 tokens_per_second_per_gpu:5117.601122684548 mfu:24.1405976484912 +Step 36127 | loss:0.3192790746688843 lr:5.957366871867098e-09 tokens_per_second_per_gpu:4790.892744930345 mfu:22.59945848842743 +Step 36128 | loss:0.30830007791519165 lr:5.890619500564798e-09 tokens_per_second_per_gpu:3895.2519316350085 mfu:18.3745679600331 +Step 36129 | loss:0.3271560072898865 lr:5.824248141339528e-09 tokens_per_second_per_gpu:4839.401539479143 mfu:22.828282748767904 +Step 36130 | loss:0.4141002595424652 lr:5.758252794690888e-09 tokens_per_second_per_gpu:5334.261134561615 mfu:25.16261989052517 +Step 36131 | loss:0.3596481382846832 lr:5.692633461107377e-09 tokens_per_second_per_gpu:4106.464102847204 mfu:19.370891808153402 +Step 36132 | loss:0.34502723813056946 lr:5.627390141094147e-09 tokens_per_second_per_gpu:5381.381271110453 mfu:25.384893614150347 +Step 36133 | loss:0.2613793909549713 lr:5.562522835139694e-09 tokens_per_second_per_gpu:5351.237094092932 mfu:25.24269838803176 +Step 36134 | loss:0.26869454979896545 lr:5.498031543721416e-09 tokens_per_second_per_gpu:3936.742324105215 mfu:18.570285220305983 +Step 36135 | loss:0.21692140400409698 lr:5.433916267338912e-09 tokens_per_second_per_gpu:5264.369391732495 mfu:24.832928614839243 +Step 36136 | loss:0.3858005106449127 lr:5.370177006464028e-09 tokens_per_second_per_gpu:4488.04547442854 mfu:21.170876242397917 +Step 36137 | loss:0.244112029671669 lr:5.30681376157971e-09 tokens_per_second_per_gpu:4847.319822265207 mfu:22.865634639668123 +Step 36138 | loss:0.25924763083457947 lr:5.243826533163354e-09 tokens_per_second_per_gpu:4165.631683967952 mfu:19.64999538333028 +Step 36139 | loss:0.3053441643714905 lr:5.1812153216868056e-09 tokens_per_second_per_gpu:5374.570211173359 mfu:25.352764682340418 +Step 36140 | loss:0.2948873043060303 lr:5.1189801276163575e-09 tokens_per_second_per_gpu:4267.352164612287 mfu:20.129828245833224 +Step 36141 | loss:0.23093008995056152 lr:5.057120951429406e-09 tokens_per_second_per_gpu:5055.902651374864 mfu:23.849555432478727 +Step 36142 | loss:0.28020158410072327 lr:4.995637793592245e-09 tokens_per_second_per_gpu:5362.081529628894 mfu:25.293853440706382 +Step 36143 | loss:0.23643407225608826 lr:4.9345306545545145e-09 tokens_per_second_per_gpu:5342.374597866814 mfu:25.20089248123538 +Step 36144 | loss:0.28051310777664185 lr:4.873799534788059e-09 tokens_per_second_per_gpu:4349.460926719227 mfu:20.51714928589106 +Step 36145 | loss:0.2603292167186737 lr:4.813444434748071e-09 tokens_per_second_per_gpu:3782.0887984806736 mfu:17.840758153322646 +Step 36146 | loss:0.2914648950099945 lr:4.753465354878639e-09 tokens_per_second_per_gpu:5017.711448236661 mfu:23.66940101118465 +Step 36147 | loss:0.2932962477207184 lr:4.6938622956405055e-09 tokens_per_second_per_gpu:5329.858690545583 mfu:25.141852810967148 +Step 36148 | loss:0.3867855370044708 lr:4.6346352574833104e-09 tokens_per_second_per_gpu:4092.5436655995622 mfu:19.305226730584774 +Step 36149 | loss:0.45658212900161743 lr:4.5757842408400416e-09 tokens_per_second_per_gpu:4714.265917345418 mfu:22.237996668824408 +Step 36150 | loss:0.2741585373878479 lr:4.517309246171442e-09 tokens_per_second_per_gpu:4358.858050634674 mfu:20.56147712271489 +Step 36151 | loss:0.3583083152770996 lr:4.459210273899395e-09 tokens_per_second_per_gpu:4217.842480165798 mfu:19.89628261707592 +Step 36152 | loss:0.3249067962169647 lr:4.401487324473541e-09 tokens_per_second_per_gpu:4612.073387510467 mfu:21.75593706974943 +Step 36153 | loss:0.3101559579372406 lr:4.344140398321317e-09 tokens_per_second_per_gpu:3465.932560110061 mfu:16.349395235111896 +Step 36154 | loss:0.3135572671890259 lr:4.28716949588126e-09 tokens_per_second_per_gpu:4913.710954028563 mfu:23.178812935691006 +Step 36155 | loss:0.3289376199245453 lr:4.230574617569705e-09 tokens_per_second_per_gpu:4760.2699088048685 mfu:22.45500534562895 +Step 36156 | loss:0.24150916934013367 lr:4.174355763825188e-09 tokens_per_second_per_gpu:3871.826025674507 mfu:18.26406396474571 +Step 36157 | loss:0.27633050084114075 lr:4.118512935064045e-09 tokens_per_second_per_gpu:4613.010142747512 mfu:21.760355904029492 +Step 36158 | loss:0.25738343596458435 lr:4.063046131702608e-09 tokens_per_second_per_gpu:4819.797688277735 mfu:22.735808037888948 +Step 36159 | loss:0.32867923378944397 lr:4.007955354168314e-09 tokens_per_second_per_gpu:5312.012444229846 mfu:25.05766902221221 +Step 36160 | loss:0.2960529327392578 lr:3.953240602866392e-09 tokens_per_second_per_gpu:5277.988453389358 mfu:24.89717205232619 +Step 36161 | loss:0.44724494218826294 lr:3.898901878213179e-09 tokens_per_second_per_gpu:4977.069428155382 mfu:23.477685668217596 +Step 36162 | loss:0.432684063911438 lr:3.844939180613904e-09 tokens_per_second_per_gpu:5389.6124757305215 mfu:25.42372161073153 +Step 36163 | loss:0.46119874715805054 lr:3.7913525104793515e-09 tokens_per_second_per_gpu:5344.39233710757 mfu:25.21041050149584 +Step 36164 | loss:0.41932326555252075 lr:3.738141868203649e-09 tokens_per_second_per_gpu:3713.800891169334 mfu:17.518632443416646 +Step 36165 | loss:0.28734806180000305 lr:3.685307254197579e-09 tokens_per_second_per_gpu:4274.9830700550565 mfu:20.16582452877349 +Step 36166 | loss:0.24921166896820068 lr:3.632848668855271e-09 tokens_per_second_per_gpu:4229.410736875193 mfu:19.95085205772226 +Step 36167 | loss:0.29905951023101807 lr:3.5807661125653035e-09 tokens_per_second_per_gpu:5354.528679930049 mfu:25.258225360027893 +Step 36168 | loss:0.22032175958156586 lr:3.5290595857273568e-09 tokens_per_second_per_gpu:4305.549814821489 mfu:20.31001307906119 +Step 36169 | loss:0.3215939402580261 lr:3.4777290887244573e-09 tokens_per_second_per_gpu:3588.075758571299 mfu:16.925565542032142 +Step 36170 | loss:0.2695617973804474 lr:3.4267746219507347e-09 tokens_per_second_per_gpu:4227.702368729741 mfu:19.942793393702495 +Step 36171 | loss:0.3500972092151642 lr:3.3761961857781133e-09 tokens_per_second_per_gpu:4201.419577674429 mfu:19.81881298398803 +Step 36172 | loss:0.3176138699054718 lr:3.3259937805951712e-09 tokens_per_second_per_gpu:4339.782609478942 mfu:20.471495012177435 +Step 36173 | loss:0.3956488072872162 lr:3.276167406779385e-09 tokens_per_second_per_gpu:4038.6022070123913 mfu:19.05077566706723 +Step 36174 | loss:0.29208534955978394 lr:3.226717064697127e-09 tokens_per_second_per_gpu:5381.648502869975 mfu:25.38615419195445 +Step 36175 | loss:0.29683998227119446 lr:3.1776427547314246e-09 tokens_per_second_per_gpu:5000.454199498773 mfu:23.587995624497783 +Step 36176 | loss:0.3856799602508545 lr:3.1289444772431008e-09 tokens_per_second_per_gpu:5333.838721259308 mfu:25.160627294907048 +Step 36177 | loss:0.340693861246109 lr:3.0806222326040802e-09 tokens_per_second_per_gpu:4021.2804074072646 mfu:18.969065782926084 +Step 36178 | loss:0.2070419043302536 lr:3.032676021175185e-09 tokens_per_second_per_gpu:4748.150739307998 mfu:22.39783715536868 +Step 36179 | loss:0.3380664587020874 lr:2.985105843311686e-09 tokens_per_second_per_gpu:5354.208196113315 mfu:25.256713583184183 +Step 36180 | loss:0.2603742480278015 lr:2.9379116993855094e-09 tokens_per_second_per_gpu:5322.098670990175 mfu:25.105247474728426 +Step 36181 | loss:0.3076496422290802 lr:2.891093589735272e-09 tokens_per_second_per_gpu:3983.250836178432 mfu:18.789673806925755 +Step 36182 | loss:0.28568795323371887 lr:2.844651514727348e-09 tokens_per_second_per_gpu:4713.464062377484 mfu:22.2342141821301 +Step 36183 | loss:0.3227802515029907 lr:2.798585474700355e-09 tokens_per_second_per_gpu:4992.061216386815 mfu:23.548404491168096 +Step 36184 | loss:0.3616383671760559 lr:2.7528954700040132e-09 tokens_per_second_per_gpu:4855.489812337452 mfu:22.90417387678295 +Step 36185 | loss:0.3288637697696686 lr:2.7075815009880436e-09 tokens_per_second_per_gpu:4344.860623842928 mfu:20.495448872330954 +Step 36186 | loss:0.2803793251514435 lr:2.6626435679855123e-09 tokens_per_second_per_gpu:4933.954179637182 mfu:23.274303684737298 +Step 36187 | loss:0.2712729871273041 lr:2.6180816713350376e-09 tokens_per_second_per_gpu:5165.6316474741 mfu:24.367165828776077 +Step 36188 | loss:0.36838939785957336 lr:2.5738958113752375e-09 tokens_per_second_per_gpu:4690.02235671444 mfu:22.123635656950114 +Step 36189 | loss:0.30796608328819275 lr:2.530085988433628e-09 tokens_per_second_per_gpu:4502.785260226398 mfu:21.2404063268732 +Step 36190 | loss:0.316864013671875 lr:2.486652202848827e-09 tokens_per_second_per_gpu:4427.891689847366 mfu:20.887120577234555 +Step 36191 | loss:0.3175096809864044 lr:2.443594454937248e-09 tokens_per_second_per_gpu:3879.7450260247724 mfu:18.301419240518698 +Step 36192 | loss:0.2346460223197937 lr:2.400912745031958e-09 tokens_per_second_per_gpu:3993.805972294124 mfu:18.839464184875396 +Step 36193 | loss:0.22940172255039215 lr:2.358607073443819e-09 tokens_per_second_per_gpu:4791.520017439703 mfu:22.6024174398778 +Step 36194 | loss:0.3109995424747467 lr:2.3166774405003476e-09 tokens_per_second_per_gpu:4865.461956733802 mfu:22.95121418332381 +Step 36195 | loss:0.24806049466133118 lr:2.2751238465068546e-09 tokens_per_second_per_gpu:4817.893011137956 mfu:22.726823350849177 +Step 36196 | loss:0.27876925468444824 lr:2.233946291785305e-09 tokens_per_second_per_gpu:4263.538262271484 mfu:20.11183741777282 +Step 36197 | loss:0.24492013454437256 lr:2.1931447766465605e-09 tokens_per_second_per_gpu:4509.219936907682 mfu:21.270759794647862 +Step 36198 | loss:0.2506580352783203 lr:2.1527193013848312e-09 tokens_per_second_per_gpu:5336.516413733717 mfu:25.173258427170282 +Step 36199 | loss:0.38409149646759033 lr:2.11266986631653e-09 tokens_per_second_per_gpu:3376.0890432016367 mfu:15.925588036970211 +Step 36200 | loss:0.3731852173805237 lr:2.0729964717414174e-09 tokens_per_second_per_gpu:4920.141029633102 mfu:23.20914469940164 +Step 36201 | loss:0.3001425862312317 lr:2.0336991179537023e-09 tokens_per_second_per_gpu:5365.569912825853 mfu:25.310308739425984 +Step 36202 | loss:0.2755465805530548 lr:1.9947778052475942e-09 tokens_per_second_per_gpu:3719.4491772582082 mfu:17.545276372594845 +Step 36203 | loss:0.30880802869796753 lr:1.9562325339173014e-09 tokens_per_second_per_gpu:4686.372710196225 mfu:22.10641965162979 +Step 36204 | loss:0.3230251669883728 lr:1.918063304257034e-09 tokens_per_second_per_gpu:5385.940041999466 mfu:25.40639811424034 +Step 36205 | loss:0.3015047013759613 lr:1.880270116549898e-09 tokens_per_second_per_gpu:3467.505845913682 mfu:16.356816692677818 +Step 36206 | loss:0.316773921251297 lr:1.8428529710845522e-09 tokens_per_second_per_gpu:5385.0635867078345 mfu:25.402263725091018 +Step 36207 | loss:0.2479773908853531 lr:1.8058118681330005e-09 tokens_per_second_per_gpu:4804.615780249424 mfu:22.664192387418787 +Step 36208 | loss:0.28947967290878296 lr:1.7691468079839012e-09 tokens_per_second_per_gpu:5387.9914998141685 mfu:25.416075190767085 +Step 36209 | loss:0.38902923464775085 lr:1.732857790903708e-09 tokens_per_second_per_gpu:5330.162813323598 mfu:25.143287409999736 +Step 36210 | loss:0.22558538615703583 lr:1.6969448171755276e-09 tokens_per_second_per_gpu:5211.11516943866 mfu:24.58171936217197 +Step 36211 | loss:0.23366115987300873 lr:1.6614078870658134e-09 tokens_per_second_per_gpu:4670.898431241784 mfu:22.033424837617453 +Step 36212 | loss:0.3438844084739685 lr:1.626247000835468e-09 tokens_per_second_per_gpu:5180.130695659771 mfu:24.4355603128606 +Step 36213 | loss:0.31910720467567444 lr:1.591462158756496e-09 tokens_per_second_per_gpu:5093.525972956374 mfu:24.0270310595793 +Step 36214 | loss:0.2148495465517044 lr:1.5570533610897997e-09 tokens_per_second_per_gpu:3474.2479622901496 mfu:16.388620405949283 +Step 36215 | loss:0.3950088322162628 lr:1.5230206080907305e-09 tokens_per_second_per_gpu:5314.582847652251 mfu:25.069794053711746 +Step 36216 | loss:0.37948495149612427 lr:1.4893639000146398e-09 tokens_per_second_per_gpu:3633.9748240498193 mfu:17.14207926508242 +Step 36217 | loss:0.3375547230243683 lr:1.4560832371168787e-09 tokens_per_second_per_gpu:4253.42472970847 mfu:20.064130159126492 +Step 36218 | loss:0.339223712682724 lr:1.4231786196527985e-09 tokens_per_second_per_gpu:5342.361451866617 mfu:25.200830469309512 +Step 36219 | loss:0.2776361107826233 lr:1.3906500478610972e-09 tokens_per_second_per_gpu:3892.657568910486 mfu:18.362329908417603 +Step 36220 | loss:0.33293506503105164 lr:1.3584975219915752e-09 tokens_per_second_per_gpu:5074.910459512561 mfu:23.939218506530725 +Step 36221 | loss:0.3661806285381317 lr:1.3267210422829302e-09 tokens_per_second_per_gpu:5352.036482670169 mfu:25.24646924034034 +Step 36222 | loss:0.35212647914886475 lr:1.2953206089738602e-09 tokens_per_second_per_gpu:4880.1943549554435 mfu:23.02070941934162 +Step 36223 | loss:0.2236933708190918 lr:1.2642962223086141e-09 tokens_per_second_per_gpu:5010.751660988072 mfu:23.636570507270903 +Step 36224 | loss:0.3182535469532013 lr:1.2336478825092367e-09 tokens_per_second_per_gpu:4832.700795240061 mfu:22.796674194927295 +Step 36225 | loss:0.2190942019224167 lr:1.2033755898144262e-09 tokens_per_second_per_gpu:4483.677043103086 mfu:21.150269606504512 +Step 36226 | loss:0.35669949650764465 lr:1.1734793444462266e-09 tokens_per_second_per_gpu:5071.562309959948 mfu:23.923424713837832 +Step 36227 | loss:0.31836068630218506 lr:1.1439591466377852e-09 tokens_per_second_per_gpu:3495.54912956465 mfu:16.489101646337 +Step 36228 | loss:0.3365161418914795 lr:1.1148149966000442e-09 tokens_per_second_per_gpu:5352.942472935449 mfu:25.250742950999545 +Step 36229 | loss:0.3969193398952484 lr:1.0860468945605995e-09 tokens_per_second_per_gpu:4180.059008588518 mfu:19.718051535120892 +Step 36230 | loss:0.27527302503585815 lr:1.0576548407303933e-09 tokens_per_second_per_gpu:5379.303106536279 mfu:25.375090557282526 +Step 36231 | loss:0.24228979647159576 lr:1.029638835325919e-09 tokens_per_second_per_gpu:3712.666919621046 mfu:17.513283306147873 +Step 36232 | loss:0.31466853618621826 lr:1.0019988785581192e-09 tokens_per_second_per_gpu:4520.5246131116455 mfu:21.32408588107861 +Step 36233 | loss:0.1857571005821228 lr:9.74734970637936e-10 tokens_per_second_per_gpu:3167.1128262220955 mfu:14.939811566458468 +Step 36234 | loss:0.310367226600647 lr:9.478471117596588e-10 tokens_per_second_per_gpu:4210.988043047082 mfu:19.86394906769937 +Step 36235 | loss:0.2829328775405884 lr:9.213353021397808e-10 tokens_per_second_per_gpu:3122.424259435696 mfu:14.729007972271297 +Step 36236 | loss:0.43717312812805176 lr:8.951995419614889e-10 tokens_per_second_per_gpu:3955.939916839689 mfu:18.66084354068157 +Step 36237 | loss:0.43648311495780945 lr:8.694398314357255e-10 tokens_per_second_per_gpu:4064.6239482030937 mfu:19.173524660030637 +Step 36238 | loss:0.24001772701740265 lr:8.440561707512285e-10 tokens_per_second_per_gpu:4177.657579344107 mfu:19.706723583648927 +Step 36239 | loss:0.3630330264568329 lr:8.190485601022869e-10 tokens_per_second_per_gpu:4743.5865224028075 mfu:22.376306965491192 +Step 36240 | loss:0.3468324542045593 lr:7.944169996665363e-10 tokens_per_second_per_gpu:5140.860601885749 mfu:24.25031658035964 +Step 36241 | loss:0.2773051857948303 lr:7.701614896382658e-10 tokens_per_second_per_gpu:5361.777203416168 mfu:25.292417882037547 +Step 36242 | loss:0.29587888717651367 lr:7.462820301951112e-10 tokens_per_second_per_gpu:4651.499280782042 mfu:21.94191573506248 +Step 36243 | loss:0.29312482476234436 lr:7.227786215258103e-10 tokens_per_second_per_gpu:4399.493225919861 mfu:20.75316017761094 +Step 36244 | loss:0.2128499150276184 lr:6.996512637913455e-10 tokens_per_second_per_gpu:5335.326287245135 mfu:25.16764439746712 +Step 36245 | loss:0.2203044891357422 lr:6.768999571804546e-10 tokens_per_second_per_gpu:4175.759683472423 mfu:19.697770885007767 +Step 36246 | loss:0.37478187680244446 lr:6.545247018596712e-10 tokens_per_second_per_gpu:4802.552993665548 mfu:22.65446187115496 +Step 36247 | loss:0.2882692217826843 lr:6.325254979955286e-10 tokens_per_second_per_gpu:4818.231311977356 mfu:22.728419173628705 +Step 36248 | loss:0.3043258488178253 lr:6.109023457490093e-10 tokens_per_second_per_gpu:4023.0262124272067 mfu:18.9773010430702 +Step 36249 | loss:0.31944146752357483 lr:5.896552452921977e-10 tokens_per_second_per_gpu:5377.403334856903 mfu:25.366129010136437 +Step 36250 | loss:0.36299675703048706 lr:5.687841967805252e-10 tokens_per_second_per_gpu:4224.8423737096755 mfu:19.92930230922738 +Step 36251 | loss:0.24773070216178894 lr:5.482892003694229e-10 tokens_per_second_per_gpu:4142.4208693708015 mfu:19.540506010702188 +Step 36252 | loss:0.3474345803260803 lr:5.281702562198732e-10 tokens_per_second_per_gpu:4639.269727776441 mfu:21.884226846958157 +Step 36253 | loss:0.34462401270866394 lr:5.084273644706539e-10 tokens_per_second_per_gpu:5319.329709902431 mfu:25.092185812844058 +Step 36254 | loss:0.29808610677719116 lr:4.890605252827474e-10 tokens_per_second_per_gpu:4796.284218953806 mfu:22.62489099127642 +Step 36255 | loss:0.31782153248786926 lr:4.700697387949315e-10 tokens_per_second_per_gpu:4280.477238993895 mfu:20.191741461059873 +Step 36256 | loss:0.33391183614730835 lr:4.514550051515354e-10 tokens_per_second_per_gpu:5129.978336225641 mfu:24.198983076534933 +Step 36257 | loss:0.3558836877346039 lr:4.3321632449133676e-10 tokens_per_second_per_gpu:3358.9138047569127 mfu:15.844569506828595 +Step 36258 | loss:0.2632720470428467 lr:4.1535369695311357e-10 tokens_per_second_per_gpu:5213.657687233359 mfu:24.593712852407833 +Step 36259 | loss:0.24505648016929626 lr:3.978671226700925e-10 tokens_per_second_per_gpu:4926.753931823551 mfu:23.240338887311445 +Step 36260 | loss:0.24152283370494843 lr:3.807566017755004e-10 tokens_per_second_per_gpu:5348.994355597555 mfu:25.23211900789842 +Step 36261 | loss:0.28758564591407776 lr:3.6402213440256404e-10 tokens_per_second_per_gpu:3502.717412037795 mfu:16.52291565779793 +Step 36262 | loss:0.3358950912952423 lr:3.476637206678568e-10 tokens_per_second_per_gpu:4662.841843557532 mfu:21.99542053891523 +Step 36263 | loss:0.28394755721092224 lr:3.316813606935032e-10 tokens_per_second_per_gpu:4057.5913620203096 mfu:19.14035075112351 +Step 36264 | loss:0.3141486644744873 lr:3.1607505461273e-10 tokens_per_second_per_gpu:5369.254049801352 mfu:25.327687442118002 +Step 36265 | loss:0.3534218668937683 lr:3.008448025310084e-10 tokens_per_second_per_gpu:3907.3838852964127 mfu:18.43179645537868 +Step 36266 | loss:0.29558464884757996 lr:2.859906045649119e-10 tokens_per_second_per_gpu:3329.298400252235 mfu:15.704868590870984 +Step 36267 | loss:0.46285000443458557 lr:2.715124608310138e-10 tokens_per_second_per_gpu:5411.687289430793 mfu:25.5278522733073 +Step 36268 | loss:0.3646628260612488 lr:2.574103714347853e-10 tokens_per_second_per_gpu:4191.9433428239745 mfu:19.77411196738605 +Step 36269 | loss:0.3479219675064087 lr:2.436843364816976e-10 tokens_per_second_per_gpu:4914.440733799791 mfu:23.18225543138515 +Step 36270 | loss:0.3486219048500061 lr:2.3033435607167086e-10 tokens_per_second_per_gpu:4485.226439909098 mfu:21.157578375592824 +Step 36271 | loss:0.2864948809146881 lr:2.1736043031572728e-10 tokens_per_second_per_gpu:5279.822386478239 mfu:24.905823027607696 +Step 36272 | loss:0.2750260531902313 lr:2.0476255930268473e-10 tokens_per_second_per_gpu:4261.764576202307 mfu:20.10347063796246 +Step 36273 | loss:0.29126912355422974 lr:1.9254074312691216e-10 tokens_per_second_per_gpu:3597.1615068208207 mfu:16.96842456671364 +Step 36274 | loss:0.3386561870574951 lr:1.8069498188832968e-10 tokens_per_second_per_gpu:5357.009873588686 mfu:25.269929573850973 +Step 36275 | loss:0.22872182726860046 lr:1.6922527567020398e-10 tokens_per_second_per_gpu:4735.291917027196 mfu:22.337179896728824 +Step 36276 | loss:0.3325962424278259 lr:1.5813162455580176e-10 tokens_per_second_per_gpu:5196.547788232071 mfu:24.513002539568966 +Step 36277 | loss:0.26904916763305664 lr:1.4741402863394094e-10 tokens_per_second_per_gpu:5372.638463609399 mfu:25.343652299491097 +Step 36278 | loss:0.2651248276233673 lr:1.3707248798233707e-10 tokens_per_second_per_gpu:3536.5451988713244 mfu:16.682487099907274 +Step 36279 | loss:0.32268020510673523 lr:1.2710700267870578e-10 tokens_per_second_per_gpu:3625.4025361371455 mfu:17.101642320415205 +Step 36280 | loss:0.2787432074546814 lr:1.175175728007627e-10 tokens_per_second_per_gpu:5383.521369414053 mfu:25.394988822986846 +Step 36281 | loss:0.3623773455619812 lr:1.083041984206723e-10 tokens_per_second_per_gpu:4514.895801393352 mfu:21.29753381582911 +Step 36282 | loss:0.2544284164905548 lr:9.946687960504797e-11 tokens_per_second_per_gpu:4061.0915320809368 mfu:19.156861650490082 +Step 36283 | loss:0.2582242488861084 lr:9.100561642050309e-11 tokens_per_second_per_gpu:4295.789456273912 mfu:20.263971802503377 +Step 36284 | loss:0.2567087411880493 lr:8.292040893365105e-11 tokens_per_second_per_gpu:4022.758658746228 mfu:18.97603894670766 +Step 36285 | loss:0.28284284472465515 lr:7.521125720000299e-11 tokens_per_second_per_gpu:3795.542246883569 mfu:17.904220364834096 +Step 36286 | loss:0.40104636549949646 lr:6.787816128062118e-11 tokens_per_second_per_gpu:3972.572555458842 mfu:18.73930253486889 +Step 36287 | loss:0.2882140576839447 lr:6.092112123101679e-11 tokens_per_second_per_gpu:5244.787690711225 mfu:24.740558390139206 +Step 36288 | loss:0.23594950139522552 lr:5.434013710670094e-11 tokens_per_second_per_gpu:3887.8115789576977 mfu:18.339470546999156 +Step 36289 | loss:0.31583139300346375 lr:4.8135208952082566e-11 tokens_per_second_per_gpu:5366.92365318865 mfu:25.31669456368922 +Step 36290 | loss:0.2902151942253113 lr:4.230633681712171e-11 tokens_per_second_per_gpu:5345.699394830686 mfu:25.21657611578282 +Step 36291 | loss:0.2937644422054291 lr:3.685352074067616e-11 tokens_per_second_per_gpu:5343.4021652992205 mfu:25.205739692135122 +Step 36292 | loss:0.2934243679046631 lr:3.177676076715486e-11 tokens_per_second_per_gpu:5367.118274236941 mfu:25.317612624379784 +Step 36293 | loss:0.27855992317199707 lr:2.7076056935415596e-11 tokens_per_second_per_gpu:5350.576041332183 mfu:25.239580089372073 +Step 36294 | loss:0.30331647396087646 lr:2.2751409278765067e-11 tokens_per_second_per_gpu:4445.011217949237 mfu:20.967876312183765 +Step 36295 | loss:0.31621235609054565 lr:1.880281783606108e-11 tokens_per_second_per_gpu:4099.136968972519 mfu:19.336328467528563 +Step 36296 | loss:0.2965101897716522 lr:1.5230282629508098e-11 tokens_per_second_per_gpu:4383.632798982104 mfu:20.67834383767766 +Step 36297 | loss:0.37591296434402466 lr:1.203380368686169e-11 tokens_per_second_per_gpu:5363.592423386662 mfu:25.30098058434683 +Step 36298 | loss:0.3476930260658264 lr:9.213381035877434e-12 tokens_per_second_per_gpu:4937.761127936141 mfu:23.292261709395845 +Step 36299 | loss:0.2785347104072571 lr:6.7690146932086755e-12 tokens_per_second_per_gpu:5053.6758994331285 mfu:23.839051463646502 +Step 36300 | loss:0.3373561501502991 lr:4.700704681059875e-12 tokens_per_second_per_gpu:3982.9635591069123 mfu:18.788318671964873 +Step 36301 | loss:0.3569736182689667 lr:3.008451010533264e-12 tokens_per_second_per_gpu:5149.5847281367805 mfu:24.291469772375184 +Step 36302 | loss:0.2967053949832916 lr:1.6922537038333019e-12 tokens_per_second_per_gpu:5353.868129901658 mfu:25.255109432842517 +Step 36303 | loss:0.3732341527938843 lr:7.521127609599887e-13 tokens_per_second_per_gpu:3518.1420880933774 mfu:16.595676486473224 +Step 36304 | loss:0.2695417106151581 lr:1.880281874644396e-13 tokens_per_second_per_gpu:5278.378327741036 mfu:24.89901115616248 +Step 36305 | loss:0.21345533430576324 lr:0.0 tokens_per_second_per_gpu:4413.674564864958 mfu:20.820055973000883