{ "best_metric": null, "best_model_checkpoint": null, "epoch": 2.996011396011396, "eval_steps": 500, "global_step": 1314, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.022792022792022793, "grad_norm": 14.146444752534505, "learning_rate": 3.787878787878788e-07, "loss": 0.8308, "step": 10 }, { "epoch": 0.045584045584045586, "grad_norm": 5.153758366693143, "learning_rate": 7.575757575757576e-07, "loss": 0.8105, "step": 20 }, { "epoch": 0.06837606837606838, "grad_norm": 1.5542878010616328, "learning_rate": 1.1363636363636364e-06, "loss": 0.7548, "step": 30 }, { "epoch": 0.09116809116809117, "grad_norm": 1.013899964163002, "learning_rate": 1.5151515151515152e-06, "loss": 0.7306, "step": 40 }, { "epoch": 0.11396011396011396, "grad_norm": 0.9161511187960373, "learning_rate": 1.8939393939393941e-06, "loss": 0.7157, "step": 50 }, { "epoch": 0.13675213675213677, "grad_norm": 0.8586592268438875, "learning_rate": 2.2727272727272728e-06, "loss": 0.6902, "step": 60 }, { "epoch": 0.15954415954415954, "grad_norm": 0.7309111880606206, "learning_rate": 2.6515151515151514e-06, "loss": 0.6845, "step": 70 }, { "epoch": 0.18233618233618235, "grad_norm": 0.8696315292996867, "learning_rate": 3.0303030303030305e-06, "loss": 0.6844, "step": 80 }, { "epoch": 0.20512820512820512, "grad_norm": 0.7998963491106827, "learning_rate": 3.409090909090909e-06, "loss": 0.6652, "step": 90 }, { "epoch": 0.22792022792022792, "grad_norm": 0.8523349303133146, "learning_rate": 3.7878787878787882e-06, "loss": 0.6617, "step": 100 }, { "epoch": 0.25071225071225073, "grad_norm": 0.6685447974331892, "learning_rate": 4.166666666666667e-06, "loss": 0.6479, "step": 110 }, { "epoch": 0.27350427350427353, "grad_norm": 0.5230920487997709, "learning_rate": 4.5454545454545455e-06, "loss": 0.6519, "step": 120 }, { "epoch": 0.2962962962962963, "grad_norm": 0.37258325894749, "learning_rate": 4.924242424242425e-06, "loss": 0.6501, "step": 130 }, { "epoch": 0.3190883190883191, "grad_norm": 0.3421753272014553, "learning_rate": 4.999434882941783e-06, "loss": 0.6507, "step": 140 }, { "epoch": 0.3418803418803419, "grad_norm": 0.32126906628196517, "learning_rate": 4.997139532754547e-06, "loss": 0.6424, "step": 150 }, { "epoch": 0.3646723646723647, "grad_norm": 0.3197148648781906, "learning_rate": 4.993080249767159e-06, "loss": 0.6469, "step": 160 }, { "epoch": 0.38746438746438744, "grad_norm": 0.36121332119458893, "learning_rate": 4.987259901381426e-06, "loss": 0.6383, "step": 170 }, { "epoch": 0.41025641025641024, "grad_norm": 0.314124834735473, "learning_rate": 4.979682598982912e-06, "loss": 0.6435, "step": 180 }, { "epoch": 0.43304843304843305, "grad_norm": 0.3036806041853383, "learning_rate": 4.970353695036733e-06, "loss": 0.6357, "step": 190 }, { "epoch": 0.45584045584045585, "grad_norm": 0.3559103875504945, "learning_rate": 4.959279779306672e-06, "loss": 0.6437, "step": 200 }, { "epoch": 0.47863247863247865, "grad_norm": 0.31838294132396994, "learning_rate": 4.9464686742003e-06, "loss": 0.6397, "step": 210 }, { "epoch": 0.5014245014245015, "grad_norm": 0.31931288370247984, "learning_rate": 4.931929429243368e-06, "loss": 0.6433, "step": 220 }, { "epoch": 0.5242165242165242, "grad_norm": 0.3387661303053822, "learning_rate": 4.915672314687389e-06, "loss": 0.6386, "step": 230 }, { "epoch": 0.5470085470085471, "grad_norm": 0.307483802697865, "learning_rate": 4.897708814254929e-06, "loss": 0.6344, "step": 240 }, { "epoch": 0.5698005698005698, "grad_norm": 0.32874061771123697, "learning_rate": 4.878051617027716e-06, "loss": 0.6392, "step": 250 }, { "epoch": 0.5925925925925926, "grad_norm": 0.30196467550126727, "learning_rate": 4.856714608483312e-06, "loss": 0.6372, "step": 260 }, { "epoch": 0.6153846153846154, "grad_norm": 0.31343992161469575, "learning_rate": 4.833712860686666e-06, "loss": 0.6342, "step": 270 }, { "epoch": 0.6381766381766382, "grad_norm": 0.31574574613486234, "learning_rate": 4.809062621643494e-06, "loss": 0.6449, "step": 280 }, { "epoch": 0.6609686609686609, "grad_norm": 0.32324064230462446, "learning_rate": 4.782781303822987e-06, "loss": 0.6335, "step": 290 }, { "epoch": 0.6837606837606838, "grad_norm": 0.287913175925187, "learning_rate": 4.75488747185797e-06, "loss": 0.6314, "step": 300 }, { "epoch": 0.7065527065527065, "grad_norm": 0.29874584820110867, "learning_rate": 4.7254008294311856e-06, "loss": 0.6312, "step": 310 }, { "epoch": 0.7293447293447294, "grad_norm": 0.3296980657627112, "learning_rate": 4.694342205356989e-06, "loss": 0.6317, "step": 320 }, { "epoch": 0.7521367521367521, "grad_norm": 0.29794479816918795, "learning_rate": 4.661733538868256e-06, "loss": 0.6283, "step": 330 }, { "epoch": 0.7749287749287749, "grad_norm": 0.3170073715761538, "learning_rate": 4.627597864118919e-06, "loss": 0.634, "step": 340 }, { "epoch": 0.7977207977207977, "grad_norm": 0.32128376305549455, "learning_rate": 4.591959293913071e-06, "loss": 0.6411, "step": 350 }, { "epoch": 0.8205128205128205, "grad_norm": 0.3204925483417397, "learning_rate": 4.554843002672129e-06, "loss": 0.6328, "step": 360 }, { "epoch": 0.8433048433048433, "grad_norm": 0.3329534096064245, "learning_rate": 4.516275208652094e-06, "loss": 0.6265, "step": 370 }, { "epoch": 0.8660968660968661, "grad_norm": 0.32160572276482535, "learning_rate": 4.4762831554234655e-06, "loss": 0.6387, "step": 380 }, { "epoch": 0.8888888888888888, "grad_norm": 0.30402339621917396, "learning_rate": 4.434895092626883e-06, "loss": 0.6372, "step": 390 }, { "epoch": 0.9116809116809117, "grad_norm": 0.30894589289734836, "learning_rate": 4.392140256018117e-06, "loss": 0.6404, "step": 400 }, { "epoch": 0.9344729344729344, "grad_norm": 0.2721461491779075, "learning_rate": 4.348048846816472e-06, "loss": 0.6366, "step": 410 }, { "epoch": 0.9572649572649573, "grad_norm": 0.2931868634895431, "learning_rate": 4.302652010371205e-06, "loss": 0.6234, "step": 420 }, { "epoch": 0.98005698005698, "grad_norm": 0.33224372295864196, "learning_rate": 4.255981814161038e-06, "loss": 0.6365, "step": 430 }, { "epoch": 0.9982905982905983, "eval_loss": 0.6270219683647156, "eval_runtime": 441.4607, "eval_samples_per_second": 26.782, "eval_steps_per_second": 0.419, "step": 438 }, { "epoch": 1.0034188034188034, "grad_norm": 0.30982191873103315, "learning_rate": 4.208071225142283e-06, "loss": 0.6498, "step": 440 }, { "epoch": 1.0262108262108263, "grad_norm": 0.3223209055964483, "learning_rate": 4.158954086461603e-06, "loss": 0.6077, "step": 450 }, { "epoch": 1.049002849002849, "grad_norm": 0.3279808548257942, "learning_rate": 4.108665093549844e-06, "loss": 0.6014, "step": 460 }, { "epoch": 1.0717948717948718, "grad_norm": 0.2800454775432525, "learning_rate": 4.057239769613827e-06, "loss": 0.6023, "step": 470 }, { "epoch": 1.0945868945868946, "grad_norm": 0.29883377707281294, "learning_rate": 4.004714440543418e-06, "loss": 0.6098, "step": 480 }, { "epoch": 1.1173789173789175, "grad_norm": 0.3092428877092738, "learning_rate": 3.951126209251599e-06, "loss": 0.6148, "step": 490 }, { "epoch": 1.1401709401709401, "grad_norm": 0.27196645349799226, "learning_rate": 3.8965129294656586e-06, "loss": 0.6046, "step": 500 }, { "epoch": 1.162962962962963, "grad_norm": 0.31199904774590587, "learning_rate": 3.84091317898803e-06, "loss": 0.6023, "step": 510 }, { "epoch": 1.1857549857549858, "grad_norm": 0.3074229624216695, "learning_rate": 3.7843662324456465e-06, "loss": 0.6079, "step": 520 }, { "epoch": 1.2085470085470085, "grad_norm": 0.28879942705874306, "learning_rate": 3.726912033547076e-06, "loss": 0.6009, "step": 530 }, { "epoch": 1.2313390313390313, "grad_norm": 0.29573529015381267, "learning_rate": 3.668591166867035e-06, "loss": 0.6111, "step": 540 }, { "epoch": 1.2541310541310542, "grad_norm": 0.2943249221910953, "learning_rate": 3.6094448291781992e-06, "loss": 0.6075, "step": 550 }, { "epoch": 1.2769230769230768, "grad_norm": 0.2913348604970632, "learning_rate": 3.549514800350572e-06, "loss": 0.6091, "step": 560 }, { "epoch": 1.2997150997150997, "grad_norm": 0.30264400504017175, "learning_rate": 3.488843413838963e-06, "loss": 0.6029, "step": 570 }, { "epoch": 1.3225071225071225, "grad_norm": 0.2662727717857523, "learning_rate": 3.4274735267794246e-06, "loss": 0.6013, "step": 580 }, { "epoch": 1.3452991452991454, "grad_norm": 0.2681498436832575, "learning_rate": 3.3654484897157717e-06, "loss": 0.6011, "step": 590 }, { "epoch": 1.368091168091168, "grad_norm": 0.2984325688299942, "learning_rate": 3.3028121159775658e-06, "loss": 0.6066, "step": 600 }, { "epoch": 1.390883190883191, "grad_norm": 0.2976332928085903, "learning_rate": 3.239608650731193e-06, "loss": 0.6066, "step": 610 }, { "epoch": 1.4136752136752135, "grad_norm": 0.2944058336808568, "learning_rate": 3.1758827397259078e-06, "loss": 0.6093, "step": 620 }, { "epoch": 1.4364672364672364, "grad_norm": 0.28155079188959703, "learning_rate": 3.111679397756906e-06, "loss": 0.607, "step": 630 }, { "epoch": 1.4592592592592593, "grad_norm": 0.29389202117647023, "learning_rate": 3.0470439768677116e-06, "loss": 0.6018, "step": 640 }, { "epoch": 1.4820512820512821, "grad_norm": 0.29252690351217875, "learning_rate": 2.982022134314344e-06, "loss": 0.5996, "step": 650 }, { "epoch": 1.504843304843305, "grad_norm": 0.2750049021563567, "learning_rate": 2.9166598003138768e-06, "loss": 0.5995, "step": 660 }, { "epoch": 1.5276353276353276, "grad_norm": 0.294567165891501, "learning_rate": 2.8510031456001946e-06, "loss": 0.607, "step": 670 }, { "epoch": 1.5504273504273505, "grad_norm": 0.2653990825203286, "learning_rate": 2.785098548809844e-06, "loss": 0.5956, "step": 680 }, { "epoch": 1.573219373219373, "grad_norm": 0.32424963851879257, "learning_rate": 2.7189925637210325e-06, "loss": 0.5994, "step": 690 }, { "epoch": 1.596011396011396, "grad_norm": 0.28116188877931586, "learning_rate": 2.652731886368906e-06, "loss": 0.6152, "step": 700 }, { "epoch": 1.6188034188034188, "grad_norm": 0.289124746715315, "learning_rate": 2.586363322060339e-06, "loss": 0.6022, "step": 710 }, { "epoch": 1.6415954415954417, "grad_norm": 0.2896827266034778, "learning_rate": 2.519933752311542e-06, "loss": 0.6066, "step": 720 }, { "epoch": 1.6643874643874645, "grad_norm": 0.292122351078682, "learning_rate": 2.4534901017318296e-06, "loss": 0.5995, "step": 730 }, { "epoch": 1.6871794871794872, "grad_norm": 0.28733382706761834, "learning_rate": 2.387079304876954e-06, "loss": 0.6006, "step": 740 }, { "epoch": 1.7099715099715098, "grad_norm": 0.28600009597025633, "learning_rate": 2.3207482730954066e-06, "loss": 0.5978, "step": 750 }, { "epoch": 1.7327635327635327, "grad_norm": 0.28872990570576484, "learning_rate": 2.2545438613911212e-06, "loss": 0.602, "step": 760 }, { "epoch": 1.7555555555555555, "grad_norm": 0.2898958466462769, "learning_rate": 2.1885128353259687e-06, "loss": 0.5985, "step": 770 }, { "epoch": 1.7783475783475784, "grad_norm": 0.28131336352135394, "learning_rate": 2.1227018379854384e-06, "loss": 0.6056, "step": 780 }, { "epoch": 1.8011396011396013, "grad_norm": 0.2712379996070474, "learning_rate": 2.0571573570308296e-06, "loss": 0.603, "step": 790 }, { "epoch": 1.823931623931624, "grad_norm": 0.2790752087160646, "learning_rate": 1.9919256918612407e-06, "loss": 0.5898, "step": 800 }, { "epoch": 1.8467236467236468, "grad_norm": 0.29154871136330335, "learning_rate": 1.927052920908528e-06, "loss": 0.5977, "step": 810 }, { "epoch": 1.8695156695156694, "grad_norm": 0.277334855491975, "learning_rate": 1.8625848690883686e-06, "loss": 0.5954, "step": 820 }, { "epoch": 1.8923076923076922, "grad_norm": 0.27268605650616284, "learning_rate": 1.7985670754303907e-06, "loss": 0.5992, "step": 830 }, { "epoch": 1.915099715099715, "grad_norm": 0.2844929476101612, "learning_rate": 1.7350447609102513e-06, "loss": 0.6038, "step": 840 }, { "epoch": 1.937891737891738, "grad_norm": 0.28770081260298885, "learning_rate": 1.6720627965063876e-06, "loss": 0.6018, "step": 850 }, { "epoch": 1.9606837606837608, "grad_norm": 0.27708084645145203, "learning_rate": 1.6096656715039868e-06, "loss": 0.603, "step": 860 }, { "epoch": 1.9834757834757835, "grad_norm": 0.2975526235682639, "learning_rate": 1.5478974620685922e-06, "loss": 0.6006, "step": 870 }, { "epoch": 1.9994301994301993, "eval_loss": 0.6204226016998291, "eval_runtime": 442.6386, "eval_samples_per_second": 26.71, "eval_steps_per_second": 0.418, "step": 877 }, { "epoch": 2.006837606837607, "grad_norm": 0.2660113640695103, "learning_rate": 1.4868018001115166e-06, "loss": 0.6233, "step": 880 }, { "epoch": 2.0296296296296297, "grad_norm": 0.28026483024484405, "learning_rate": 1.4264218424690759e-06, "loss": 0.5696, "step": 890 }, { "epoch": 2.0524216524216525, "grad_norm": 0.28181694489301495, "learning_rate": 1.3668002404174047e-06, "loss": 0.5855, "step": 900 }, { "epoch": 2.0752136752136754, "grad_norm": 0.28298520415629186, "learning_rate": 1.307979109544388e-06, "loss": 0.5834, "step": 910 }, { "epoch": 2.098005698005698, "grad_norm": 0.2704142615239853, "learning_rate": 1.2500000000000007e-06, "loss": 0.5865, "step": 920 }, { "epoch": 2.1207977207977207, "grad_norm": 0.2711904141610201, "learning_rate": 1.1929038671460486e-06, "loss": 0.584, "step": 930 }, { "epoch": 2.1435897435897435, "grad_norm": 0.28015029336959785, "learning_rate": 1.136731042626073e-06, "loss": 0.5787, "step": 940 }, { "epoch": 2.1663817663817664, "grad_norm": 0.2705809516221485, "learning_rate": 1.0815212058758218e-06, "loss": 0.5796, "step": 950 }, { "epoch": 2.1891737891737892, "grad_norm": 0.27273162517988947, "learning_rate": 1.0273133560944432e-06, "loss": 0.5845, "step": 960 }, { "epoch": 2.211965811965812, "grad_norm": 0.2830067338695591, "learning_rate": 9.741457846961721e-07, "loss": 0.5833, "step": 970 }, { "epoch": 2.234757834757835, "grad_norm": 0.28075024037195573, "learning_rate": 9.220560482619956e-07, "loss": 0.573, "step": 980 }, { "epoch": 2.2575498575498574, "grad_norm": 0.27521631417439996, "learning_rate": 8.710809420103788e-07, "loss": 0.5819, "step": 990 }, { "epoch": 2.2803418803418802, "grad_norm": 0.2641332093050321, "learning_rate": 8.21256473805811e-07, "loss": 0.5821, "step": 1000 }, { "epoch": 2.303133903133903, "grad_norm": 0.28140142311128546, "learning_rate": 7.726178387235234e-07, "loss": 0.5796, "step": 1010 }, { "epoch": 2.325925925925926, "grad_norm": 0.2813678146041498, "learning_rate": 7.251993941883428e-07, "loss": 0.5825, "step": 1020 }, { "epoch": 2.348717948717949, "grad_norm": 0.26733817473037325, "learning_rate": 6.790346357052443e-07, "loss": 0.584, "step": 1030 }, { "epoch": 2.3715099715099717, "grad_norm": 0.2604883848198127, "learning_rate": 6.341561731987572e-07, "loss": 0.5816, "step": 1040 }, { "epoch": 2.394301994301994, "grad_norm": 0.289012080255221, "learning_rate": 5.905957079779187e-07, "loss": 0.5808, "step": 1050 }, { "epoch": 2.417094017094017, "grad_norm": 0.2695833722813263, "learning_rate": 5.483840103430599e-07, "loss": 0.589, "step": 1060 }, { "epoch": 2.43988603988604, "grad_norm": 0.26691946928176913, "learning_rate": 5.0755089785024e-07, "loss": 0.5922, "step": 1070 }, { "epoch": 2.4626780626780627, "grad_norm": 0.26813063058675607, "learning_rate": 4.6812521424868416e-07, "loss": 0.5874, "step": 1080 }, { "epoch": 2.4854700854700855, "grad_norm": 0.2638423364986993, "learning_rate": 4.301348091060906e-07, "loss": 0.5847, "step": 1090 }, { "epoch": 2.5082621082621084, "grad_norm": 0.28075479357187255, "learning_rate": 3.936065181362211e-07, "loss": 0.5914, "step": 1100 }, { "epoch": 2.5310541310541312, "grad_norm": 0.26465289459340213, "learning_rate": 3.585661442426494e-07, "loss": 0.5844, "step": 1110 }, { "epoch": 2.5538461538461537, "grad_norm": 0.260920376837073, "learning_rate": 3.250384392920741e-07, "loss": 0.5826, "step": 1120 }, { "epoch": 2.5766381766381765, "grad_norm": 0.26750993710726595, "learning_rate": 2.930470866300583e-07, "loss": 0.5819, "step": 1130 }, { "epoch": 2.5994301994301994, "grad_norm": 0.2622362163032793, "learning_rate": 2.626146843515598e-07, "loss": 0.5816, "step": 1140 }, { "epoch": 2.6222222222222222, "grad_norm": 0.2707711955951654, "learning_rate": 2.3376272933805482e-07, "loss": 0.5817, "step": 1150 }, { "epoch": 2.645014245014245, "grad_norm": 0.27125876578009134, "learning_rate": 2.065116020725433e-07, "loss": 0.5824, "step": 1160 }, { "epoch": 2.667806267806268, "grad_norm": 0.2697526018968707, "learning_rate": 1.80880552243157e-07, "loss": 0.5858, "step": 1170 }, { "epoch": 2.690598290598291, "grad_norm": 0.2617655076275081, "learning_rate": 1.5688768514553587e-07, "loss": 0.5808, "step": 1180 }, { "epoch": 2.7133903133903132, "grad_norm": 0.25665192252847907, "learning_rate": 1.345499488935875e-07, "loss": 0.5817, "step": 1190 }, { "epoch": 2.736182336182336, "grad_norm": 0.264271353582254, "learning_rate": 1.1388312244765332e-07, "loss": 0.5849, "step": 1200 }, { "epoch": 2.758974358974359, "grad_norm": 0.25635457369854714, "learning_rate": 9.490180446854791e-08, "loss": 0.5881, "step": 1210 }, { "epoch": 2.781766381766382, "grad_norm": 0.2608245013747383, "learning_rate": 7.7619403005334e-08, "loss": 0.5778, "step": 1220 }, { "epoch": 2.8045584045584047, "grad_norm": 0.2520807093959864, "learning_rate": 6.204812602412901e-08, "loss": 0.5763, "step": 1230 }, { "epoch": 2.827350427350427, "grad_norm": 0.2587943339148964, "learning_rate": 4.8198972784625205e-08, "loss": 0.5913, "step": 1240 }, { "epoch": 2.8501424501424504, "grad_norm": 0.26075750237607453, "learning_rate": 3.608172607041632e-08, "loss": 0.5882, "step": 1250 }, { "epoch": 2.872934472934473, "grad_norm": 0.26447256486892573, "learning_rate": 2.5704945278623438e-08, "loss": 0.5852, "step": 1260 }, { "epoch": 2.8957264957264957, "grad_norm": 0.2675763366142142, "learning_rate": 1.7075960373695022e-08, "loss": 0.5861, "step": 1270 }, { "epoch": 2.9185185185185185, "grad_norm": 0.2625544683764617, "learning_rate": 1.0200866709657863e-08, "loss": 0.5763, "step": 1280 }, { "epoch": 2.9413105413105414, "grad_norm": 0.2636204922137769, "learning_rate": 5.08452072447152e-09, "loss": 0.5762, "step": 1290 }, { "epoch": 2.9641025641025642, "grad_norm": 0.2680005327249821, "learning_rate": 1.7305365095324211e-09, "loss": 0.5775, "step": 1300 }, { "epoch": 2.9868945868945866, "grad_norm": 0.269400024292336, "learning_rate": 1.412832567460054e-10, "loss": 0.5877, "step": 1310 }, { "epoch": 2.996011396011396, "eval_loss": 0.6208292245864868, "eval_runtime": 441.8042, "eval_samples_per_second": 26.761, "eval_steps_per_second": 0.419, "step": 1314 }, { "epoch": 2.996011396011396, "step": 1314, "total_flos": 2755219238682624.0, "train_loss": 0.6149487337565314, "train_runtime": 70718.2917, "train_samples_per_second": 9.529, "train_steps_per_second": 0.019 } ], "logging_steps": 10, "max_steps": 1314, "num_input_tokens_seen": 0, "num_train_epochs": 3, "save_steps": 500, "stateful_callbacks": { "TrainerControl": { "args": { "should_epoch_stop": false, "should_evaluate": false, "should_log": false, "should_save": true, "should_training_stop": true }, "attributes": {} } }, "total_flos": 2755219238682624.0, "train_batch_size": 8, "trial_name": null, "trial_params": null }