andrewbai's picture
Model save
42d6e86 verified
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 1.0,
"eval_steps": 500,
"global_step": 732,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.001366120218579235,
"grad_norm": 8.839000701532022,
"learning_rate": 2.702702702702703e-07,
"loss": 1.199,
"step": 1
},
{
"epoch": 0.006830601092896175,
"grad_norm": 8.54443252593224,
"learning_rate": 1.3513513513513515e-06,
"loss": 1.2045,
"step": 5
},
{
"epoch": 0.01366120218579235,
"grad_norm": 8.766360812816002,
"learning_rate": 2.702702702702703e-06,
"loss": 1.1859,
"step": 10
},
{
"epoch": 0.020491803278688523,
"grad_norm": 2.6135570637792593,
"learning_rate": 4.0540540540540545e-06,
"loss": 1.0192,
"step": 15
},
{
"epoch": 0.0273224043715847,
"grad_norm": 0.9204527360610539,
"learning_rate": 5.405405405405406e-06,
"loss": 0.9217,
"step": 20
},
{
"epoch": 0.03415300546448088,
"grad_norm": 0.7474197058876075,
"learning_rate": 6.7567567567567575e-06,
"loss": 0.859,
"step": 25
},
{
"epoch": 0.040983606557377046,
"grad_norm": 0.7028834448437822,
"learning_rate": 8.108108108108109e-06,
"loss": 0.8567,
"step": 30
},
{
"epoch": 0.04781420765027322,
"grad_norm": 0.6058504125273818,
"learning_rate": 9.45945945945946e-06,
"loss": 0.8564,
"step": 35
},
{
"epoch": 0.0546448087431694,
"grad_norm": 0.5930991083899937,
"learning_rate": 1.0810810810810812e-05,
"loss": 0.828,
"step": 40
},
{
"epoch": 0.06147540983606557,
"grad_norm": 0.5843120963137354,
"learning_rate": 1.2162162162162164e-05,
"loss": 0.8319,
"step": 45
},
{
"epoch": 0.06830601092896176,
"grad_norm": 0.5527620922946438,
"learning_rate": 1.3513513513513515e-05,
"loss": 0.8194,
"step": 50
},
{
"epoch": 0.07513661202185792,
"grad_norm": 0.5888810193109677,
"learning_rate": 1.4864864864864865e-05,
"loss": 0.828,
"step": 55
},
{
"epoch": 0.08196721311475409,
"grad_norm": 0.5711479605177191,
"learning_rate": 1.6216216216216218e-05,
"loss": 0.8255,
"step": 60
},
{
"epoch": 0.08879781420765027,
"grad_norm": 0.5790256594021843,
"learning_rate": 1.756756756756757e-05,
"loss": 0.8091,
"step": 65
},
{
"epoch": 0.09562841530054644,
"grad_norm": 0.5999689248767484,
"learning_rate": 1.891891891891892e-05,
"loss": 0.8509,
"step": 70
},
{
"epoch": 0.10245901639344263,
"grad_norm": 0.5760441130309225,
"learning_rate": 1.999988602302209e-05,
"loss": 0.7947,
"step": 75
},
{
"epoch": 0.1092896174863388,
"grad_norm": 0.563404914562876,
"learning_rate": 1.9995897101594454e-05,
"loss": 0.8095,
"step": 80
},
{
"epoch": 0.11612021857923498,
"grad_norm": 0.5960450579935208,
"learning_rate": 1.99862119291555e-05,
"loss": 0.8086,
"step": 85
},
{
"epoch": 0.12295081967213115,
"grad_norm": 0.59853110269883,
"learning_rate": 1.997083602488702e-05,
"loss": 0.807,
"step": 90
},
{
"epoch": 0.12978142076502733,
"grad_norm": 0.5967638652133508,
"learning_rate": 1.994977815088504e-05,
"loss": 0.8055,
"step": 95
},
{
"epoch": 0.1366120218579235,
"grad_norm": 0.5817027228929745,
"learning_rate": 1.9923050307166655e-05,
"loss": 0.8042,
"step": 100
},
{
"epoch": 0.14344262295081966,
"grad_norm": 0.6298978323283461,
"learning_rate": 1.989066772483171e-05,
"loss": 0.8122,
"step": 105
},
{
"epoch": 0.15027322404371585,
"grad_norm": 0.5443292461213772,
"learning_rate": 1.9852648857383224e-05,
"loss": 0.7947,
"step": 110
},
{
"epoch": 0.15710382513661203,
"grad_norm": 0.5958295331916256,
"learning_rate": 1.9809015370211505e-05,
"loss": 0.8023,
"step": 115
},
{
"epoch": 0.16393442622950818,
"grad_norm": 0.5429841988820163,
"learning_rate": 1.9759792128247922e-05,
"loss": 0.8178,
"step": 120
},
{
"epoch": 0.17076502732240437,
"grad_norm": 0.5962054388159139,
"learning_rate": 1.9705007181795416e-05,
"loss": 0.8075,
"step": 125
},
{
"epoch": 0.17759562841530055,
"grad_norm": 0.5873047611136615,
"learning_rate": 1.964469175054377e-05,
"loss": 0.7948,
"step": 130
},
{
"epoch": 0.18442622950819673,
"grad_norm": 0.5707966893645422,
"learning_rate": 1.9578880205778793e-05,
"loss": 0.79,
"step": 135
},
{
"epoch": 0.1912568306010929,
"grad_norm": 0.5529077592904704,
"learning_rate": 1.950761005079556e-05,
"loss": 0.7677,
"step": 140
},
{
"epoch": 0.19808743169398907,
"grad_norm": 0.5908113970637738,
"learning_rate": 1.9430921899526786e-05,
"loss": 0.7889,
"step": 145
},
{
"epoch": 0.20491803278688525,
"grad_norm": 0.5881122057959933,
"learning_rate": 1.934885945339865e-05,
"loss": 0.7812,
"step": 150
},
{
"epoch": 0.21174863387978143,
"grad_norm": 0.599803919917601,
"learning_rate": 1.9261469476427122e-05,
"loss": 0.7957,
"step": 155
},
{
"epoch": 0.2185792349726776,
"grad_norm": 0.560441172041075,
"learning_rate": 1.916880176856909e-05,
"loss": 0.7589,
"step": 160
},
{
"epoch": 0.22540983606557377,
"grad_norm": 0.6107861386387331,
"learning_rate": 1.907090913734341e-05,
"loss": 0.7786,
"step": 165
},
{
"epoch": 0.23224043715846995,
"grad_norm": 0.5873436702665477,
"learning_rate": 1.896784736773805e-05,
"loss": 0.7678,
"step": 170
},
{
"epoch": 0.2390710382513661,
"grad_norm": 0.551913986008609,
"learning_rate": 1.885967519042054e-05,
"loss": 0.7744,
"step": 175
},
{
"epoch": 0.2459016393442623,
"grad_norm": 0.5534204553817462,
"learning_rate": 1.8746454248269777e-05,
"loss": 0.76,
"step": 180
},
{
"epoch": 0.2527322404371585,
"grad_norm": 0.5475707727189457,
"learning_rate": 1.862824906124826e-05,
"loss": 0.7837,
"step": 185
},
{
"epoch": 0.25956284153005466,
"grad_norm": 0.5362754705998823,
"learning_rate": 1.850512698963485e-05,
"loss": 0.7578,
"step": 190
},
{
"epoch": 0.26639344262295084,
"grad_norm": 0.568308146412154,
"learning_rate": 1.8377158195638877e-05,
"loss": 0.7721,
"step": 195
},
{
"epoch": 0.273224043715847,
"grad_norm": 0.573616423304356,
"learning_rate": 1.8244415603417603e-05,
"loss": 0.7938,
"step": 200
},
{
"epoch": 0.28005464480874315,
"grad_norm": 0.5772108653032471,
"learning_rate": 1.8106974857519737e-05,
"loss": 0.776,
"step": 205
},
{
"epoch": 0.28688524590163933,
"grad_norm": 0.6117645615908475,
"learning_rate": 1.7964914279778716e-05,
"loss": 0.7521,
"step": 210
},
{
"epoch": 0.2937158469945355,
"grad_norm": 0.5700487711327129,
"learning_rate": 1.78183148246803e-05,
"loss": 0.764,
"step": 215
},
{
"epoch": 0.3005464480874317,
"grad_norm": 0.6151416615508302,
"learning_rate": 1.7667260033229953e-05,
"loss": 0.7798,
"step": 220
},
{
"epoch": 0.3073770491803279,
"grad_norm": 0.5856403601353761,
"learning_rate": 1.751183598534625e-05,
"loss": 0.7646,
"step": 225
},
{
"epoch": 0.31420765027322406,
"grad_norm": 0.6275879480436144,
"learning_rate": 1.7352131250807466e-05,
"loss": 0.769,
"step": 230
},
{
"epoch": 0.32103825136612024,
"grad_norm": 0.5509601358171512,
"learning_rate": 1.7188236838779297e-05,
"loss": 0.7517,
"step": 235
},
{
"epoch": 0.32786885245901637,
"grad_norm": 0.5604130447613014,
"learning_rate": 1.702024614595248e-05,
"loss": 0.7734,
"step": 240
},
{
"epoch": 0.33469945355191255,
"grad_norm": 0.6282624124992058,
"learning_rate": 1.6848254903319866e-05,
"loss": 0.7486,
"step": 245
},
{
"epoch": 0.34153005464480873,
"grad_norm": 0.568288936290054,
"learning_rate": 1.6672361121623238e-05,
"loss": 0.7613,
"step": 250
},
{
"epoch": 0.3483606557377049,
"grad_norm": 0.5484081363085775,
"learning_rate": 1.6492665035501048e-05,
"loss": 0.7585,
"step": 255
},
{
"epoch": 0.3551912568306011,
"grad_norm": 0.5764477527848118,
"learning_rate": 1.6309269046368777e-05,
"loss": 0.7673,
"step": 260
},
{
"epoch": 0.3620218579234973,
"grad_norm": 0.5971807718295861,
"learning_rate": 1.612227766406461e-05,
"loss": 0.7815,
"step": 265
},
{
"epoch": 0.36885245901639346,
"grad_norm": 0.5704742221970132,
"learning_rate": 1.5931797447293553e-05,
"loss": 0.7664,
"step": 270
},
{
"epoch": 0.3756830601092896,
"grad_norm": 0.5793216066877228,
"learning_rate": 1.5737936942904025e-05,
"loss": 0.7514,
"step": 275
},
{
"epoch": 0.3825136612021858,
"grad_norm": 0.5505577683054798,
"learning_rate": 1.554080662403144e-05,
"loss": 0.7588,
"step": 280
},
{
"epoch": 0.38934426229508196,
"grad_norm": 0.5729266512483018,
"learning_rate": 1.5340518827144145e-05,
"loss": 0.7559,
"step": 285
},
{
"epoch": 0.39617486338797814,
"grad_norm": 0.5602606968248647,
"learning_rate": 1.5137187688027437e-05,
"loss": 0.7536,
"step": 290
},
{
"epoch": 0.4030054644808743,
"grad_norm": 0.5460182503591859,
"learning_rate": 1.4930929076742317e-05,
"loss": 0.7409,
"step": 295
},
{
"epoch": 0.4098360655737705,
"grad_norm": 0.5523985577258183,
"learning_rate": 1.4721860531595868e-05,
"loss": 0.771,
"step": 300
},
{
"epoch": 0.4166666666666667,
"grad_norm": 0.5641378656097824,
"learning_rate": 1.451010119216102e-05,
"loss": 0.736,
"step": 305
},
{
"epoch": 0.42349726775956287,
"grad_norm": 0.6191757729625711,
"learning_rate": 1.4295771731383799e-05,
"loss": 0.7527,
"step": 310
},
{
"epoch": 0.430327868852459,
"grad_norm": 0.5471083343191394,
"learning_rate": 1.4078994286816768e-05,
"loss": 0.7295,
"step": 315
},
{
"epoch": 0.4371584699453552,
"grad_norm": 0.5982097120013118,
"learning_rate": 1.3859892391017867e-05,
"loss": 0.747,
"step": 320
},
{
"epoch": 0.44398907103825136,
"grad_norm": 0.5616896454618231,
"learning_rate": 1.3638590901154276e-05,
"loss": 0.7308,
"step": 325
},
{
"epoch": 0.45081967213114754,
"grad_norm": 0.5709111327321962,
"learning_rate": 1.341521592785145e-05,
"loss": 0.7416,
"step": 330
},
{
"epoch": 0.4576502732240437,
"grad_norm": 0.6029640775277743,
"learning_rate": 1.3189894763327851e-05,
"loss": 0.7575,
"step": 335
},
{
"epoch": 0.4644808743169399,
"grad_norm": 0.6090060111700056,
"learning_rate": 1.2962755808856341e-05,
"loss": 0.7233,
"step": 340
},
{
"epoch": 0.4713114754098361,
"grad_norm": 0.5552565456347985,
"learning_rate": 1.2733928501593587e-05,
"loss": 0.7513,
"step": 345
},
{
"epoch": 0.4781420765027322,
"grad_norm": 0.5899999700937736,
"learning_rate": 1.2503543240819127e-05,
"loss": 0.7247,
"step": 350
},
{
"epoch": 0.4849726775956284,
"grad_norm": 0.5678421870745727,
"learning_rate": 1.227173131362619e-05,
"loss": 0.7311,
"step": 355
},
{
"epoch": 0.4918032786885246,
"grad_norm": 0.5641985877955489,
"learning_rate": 1.2038624820106572e-05,
"loss": 0.723,
"step": 360
},
{
"epoch": 0.49863387978142076,
"grad_norm": 0.5759106686454865,
"learning_rate": 1.1804356598072223e-05,
"loss": 0.7392,
"step": 365
},
{
"epoch": 0.505464480874317,
"grad_norm": 0.5942494795318336,
"learning_rate": 1.1569060147356441e-05,
"loss": 0.7287,
"step": 370
},
{
"epoch": 0.5122950819672131,
"grad_norm": 0.6097243862112076,
"learning_rate": 1.133286955373779e-05,
"loss": 0.7567,
"step": 375
},
{
"epoch": 0.5191256830601093,
"grad_norm": 0.7022313974041855,
"learning_rate": 1.1095919412530136e-05,
"loss": 0.7257,
"step": 380
},
{
"epoch": 0.5259562841530054,
"grad_norm": 0.5588868936313335,
"learning_rate": 1.0858344751882304e-05,
"loss": 0.7468,
"step": 385
},
{
"epoch": 0.5327868852459017,
"grad_norm": 0.6211762914313255,
"learning_rate": 1.0620280955831088e-05,
"loss": 0.7537,
"step": 390
},
{
"epoch": 0.5396174863387978,
"grad_norm": 0.5740368220758967,
"learning_rate": 1.038186368715145e-05,
"loss": 0.7343,
"step": 395
},
{
"epoch": 0.546448087431694,
"grad_norm": 0.5701811365500552,
"learning_rate": 1.0143228810047877e-05,
"loss": 0.7356,
"step": 400
},
{
"epoch": 0.5532786885245902,
"grad_norm": 0.5631312726270489,
"learning_rate": 9.904512312730948e-06,
"loss": 0.723,
"step": 405
},
{
"epoch": 0.5601092896174863,
"grad_norm": 0.535442901596806,
"learning_rate": 9.665850229923258e-06,
"loss": 0.7358,
"step": 410
},
{
"epoch": 0.5669398907103825,
"grad_norm": 0.5690112641912415,
"learning_rate": 9.4273785653388e-06,
"loss": 0.7266,
"step": 415
},
{
"epoch": 0.5737704918032787,
"grad_norm": 0.5519801631394425,
"learning_rate": 9.189233214180057e-06,
"loss": 0.7247,
"step": 420
},
{
"epoch": 0.5806010928961749,
"grad_norm": 0.6214956085930828,
"learning_rate": 8.951549885696889e-06,
"loss": 0.7152,
"step": 425
},
{
"epoch": 0.587431693989071,
"grad_norm": 0.5453035860772034,
"learning_rate": 8.714464025851428e-06,
"loss": 0.7328,
"step": 430
},
{
"epoch": 0.5942622950819673,
"grad_norm": 0.5928933851523271,
"learning_rate": 8.478110740132971e-06,
"loss": 0.7394,
"step": 435
},
{
"epoch": 0.6010928961748634,
"grad_norm": 0.5777886942360209,
"learning_rate": 8.242624716566928e-06,
"loss": 0.7386,
"step": 440
},
{
"epoch": 0.6079234972677595,
"grad_norm": 0.5829472844091811,
"learning_rate": 8.008140148961642e-06,
"loss": 0.7287,
"step": 445
},
{
"epoch": 0.6147540983606558,
"grad_norm": 0.5662302145710356,
"learning_rate": 7.774790660436857e-06,
"loss": 0.7396,
"step": 450
},
{
"epoch": 0.6215846994535519,
"grad_norm": 0.6056835562596397,
"learning_rate": 7.542709227277396e-06,
"loss": 0.7256,
"step": 455
},
{
"epoch": 0.6284153005464481,
"grad_norm": 0.5452103732631707,
"learning_rate": 7.312028103155426e-06,
"loss": 0.7112,
"step": 460
},
{
"epoch": 0.6352459016393442,
"grad_norm": 0.5263394595201824,
"learning_rate": 7.0828787437645455e-06,
"loss": 0.7253,
"step": 465
},
{
"epoch": 0.6420765027322405,
"grad_norm": 0.5387984073614047,
"learning_rate": 6.8553917319085676e-06,
"loss": 0.715,
"step": 470
},
{
"epoch": 0.6489071038251366,
"grad_norm": 0.5401548566338565,
"learning_rate": 6.629696703087755e-06,
"loss": 0.7135,
"step": 475
},
{
"epoch": 0.6557377049180327,
"grad_norm": 0.5515159980407115,
"learning_rate": 6.405922271624874e-06,
"loss": 0.6676,
"step": 480
},
{
"epoch": 0.662568306010929,
"grad_norm": 0.5568515122991526,
"learning_rate": 6.184195957373176e-06,
"loss": 0.7353,
"step": 485
},
{
"epoch": 0.6693989071038251,
"grad_norm": 0.6271172253128796,
"learning_rate": 5.964644113048079e-06,
"loss": 0.7064,
"step": 490
},
{
"epoch": 0.6762295081967213,
"grad_norm": 0.5462042747201297,
"learning_rate": 5.74739185222394e-06,
"loss": 0.7336,
"step": 495
},
{
"epoch": 0.6830601092896175,
"grad_norm": 0.5558772088354085,
"learning_rate": 5.532562978036964e-06,
"loss": 0.7198,
"step": 500
},
{
"epoch": 0.6898907103825137,
"grad_norm": 0.6030516260535024,
"learning_rate": 5.320279912634907e-06,
"loss": 0.7186,
"step": 505
},
{
"epoch": 0.6967213114754098,
"grad_norm": 0.5696457236273929,
"learning_rate": 5.110663627413695e-06,
"loss": 0.7366,
"step": 510
},
{
"epoch": 0.703551912568306,
"grad_norm": 0.5539578087989158,
"learning_rate": 4.903833574080825e-06,
"loss": 0.7065,
"step": 515
},
{
"epoch": 0.7103825136612022,
"grad_norm": 0.6071180295830684,
"learning_rate": 4.6999076165847214e-06,
"loss": 0.7436,
"step": 520
},
{
"epoch": 0.7172131147540983,
"grad_norm": 0.5571214598686036,
"learning_rate": 4.499001963948929e-06,
"loss": 0.7279,
"step": 525
},
{
"epoch": 0.7240437158469946,
"grad_norm": 0.5401199305903159,
"learning_rate": 4.301231104049359e-06,
"loss": 0.7102,
"step": 530
},
{
"epoch": 0.7308743169398907,
"grad_norm": 0.5313937404748474,
"learning_rate": 4.106707738372357e-06,
"loss": 0.696,
"step": 535
},
{
"epoch": 0.7377049180327869,
"grad_norm": 0.5821444871544712,
"learning_rate": 3.915542717790759e-06,
"loss": 0.7224,
"step": 540
},
{
"epoch": 0.744535519125683,
"grad_norm": 0.5758421647300646,
"learning_rate": 3.727844979394526e-06,
"loss": 0.7368,
"step": 545
},
{
"epoch": 0.7513661202185792,
"grad_norm": 0.5511567358018045,
"learning_rate": 3.543721484411976e-06,
"loss": 0.738,
"step": 550
},
{
"epoch": 0.7581967213114754,
"grad_norm": 0.5782196132261889,
"learning_rate": 3.3632771572569878e-06,
"loss": 0.7346,
"step": 555
},
{
"epoch": 0.7650273224043715,
"grad_norm": 0.5444427117744266,
"learning_rate": 3.1866148257368666e-06,
"loss": 0.6956,
"step": 560
},
{
"epoch": 0.7718579234972678,
"grad_norm": 0.5508607476703664,
"learning_rate": 3.0138351624550165e-06,
"loss": 0.7086,
"step": 565
},
{
"epoch": 0.7786885245901639,
"grad_norm": 0.5547385528497039,
"learning_rate": 2.845036627441755e-06,
"loss": 0.7029,
"step": 570
},
{
"epoch": 0.7855191256830601,
"grad_norm": 0.5497303389157615,
"learning_rate": 2.6803154120460007e-06,
"loss": 0.698,
"step": 575
},
{
"epoch": 0.7923497267759563,
"grad_norm": 0.5445054309275521,
"learning_rate": 2.5197653841197546e-06,
"loss": 0.7359,
"step": 580
},
{
"epoch": 0.7991803278688525,
"grad_norm": 0.5649479809138797,
"learning_rate": 2.3634780345266805e-06,
"loss": 0.7272,
"step": 585
},
{
"epoch": 0.8060109289617486,
"grad_norm": 0.5703720996642136,
"learning_rate": 2.211542425005223e-06,
"loss": 0.7093,
"step": 590
},
{
"epoch": 0.8128415300546448,
"grad_norm": 0.5542039680095538,
"learning_rate": 2.064045137415982e-06,
"loss": 0.7464,
"step": 595
},
{
"epoch": 0.819672131147541,
"grad_norm": 0.5329395934495856,
"learning_rate": 1.9210702244022616e-06,
"loss": 0.7148,
"step": 600
},
{
"epoch": 0.8265027322404371,
"grad_norm": 0.5287377422445715,
"learning_rate": 1.7826991614919264e-06,
"loss": 0.7223,
"step": 605
},
{
"epoch": 0.8333333333333334,
"grad_norm": 0.5593988173817716,
"learning_rate": 1.6490108006678495e-06,
"loss": 0.7247,
"step": 610
},
{
"epoch": 0.8401639344262295,
"grad_norm": 0.5672809722094616,
"learning_rate": 1.5200813254334013e-06,
"loss": 0.7054,
"step": 615
},
{
"epoch": 0.8469945355191257,
"grad_norm": 0.5730006618469092,
"learning_rate": 1.3959842073986085e-06,
"loss": 0.7267,
"step": 620
},
{
"epoch": 0.8538251366120219,
"grad_norm": 0.5446691548063705,
"learning_rate": 1.2767901644116943e-06,
"loss": 0.7154,
"step": 625
},
{
"epoch": 0.860655737704918,
"grad_norm": 0.5672648990349456,
"learning_rate": 1.1625671202598875e-06,
"loss": 0.7043,
"step": 630
},
{
"epoch": 0.8674863387978142,
"grad_norm": 0.5935712213732242,
"learning_rate": 1.0533801659624531e-06,
"loss": 0.696,
"step": 635
},
{
"epoch": 0.8743169398907104,
"grad_norm": 0.5474056757010451,
"learning_rate": 9.492915226779809e-07,
"loss": 0.7037,
"step": 640
},
{
"epoch": 0.8811475409836066,
"grad_norm": 0.7316698327879719,
"learning_rate": 8.503605062471187e-07,
"loss": 0.7285,
"step": 645
},
{
"epoch": 0.8879781420765027,
"grad_norm": 0.5399843524785475,
"learning_rate": 7.566434933909006e-07,
"loss": 0.7168,
"step": 650
},
{
"epoch": 0.894808743169399,
"grad_norm": 0.5406637929209906,
"learning_rate": 6.681938895839746e-07,
"loss": 0.69,
"step": 655
},
{
"epoch": 0.9016393442622951,
"grad_norm": 0.5517097205610324,
"learning_rate": 5.850620986210198e-07,
"loss": 0.7028,
"step": 660
},
{
"epoch": 0.9084699453551912,
"grad_norm": 0.5587545679939938,
"learning_rate": 5.072954938936925e-07,
"loss": 0.7059,
"step": 665
},
{
"epoch": 0.9153005464480874,
"grad_norm": 0.608047963211898,
"learning_rate": 4.3493839139447716e-07,
"loss": 0.709,
"step": 670
},
{
"epoch": 0.9221311475409836,
"grad_norm": 0.5654282179981049,
"learning_rate": 3.6803202446282217e-07,
"loss": 0.7173,
"step": 675
},
{
"epoch": 0.9289617486338798,
"grad_norm": 0.5461106613890502,
"learning_rate": 3.0661452028795335e-07,
"loss": 0.6962,
"step": 680
},
{
"epoch": 0.9357923497267759,
"grad_norm": 0.5263590681005187,
"learning_rate": 2.507208781817638e-07,
"loss": 0.7061,
"step": 685
},
{
"epoch": 0.9426229508196722,
"grad_norm": 0.5372230969207142,
"learning_rate": 2.0038294963413251e-07,
"loss": 0.7303,
"step": 690
},
{
"epoch": 0.9494535519125683,
"grad_norm": 0.5292248426851656,
"learning_rate": 1.556294201620734e-07,
"loss": 0.7078,
"step": 695
},
{
"epoch": 0.9562841530054644,
"grad_norm": 0.5576533992425854,
"learning_rate": 1.1648579296304252e-07,
"loss": 0.7099,
"step": 700
},
{
"epoch": 0.9631147540983607,
"grad_norm": 0.562157461832351,
"learning_rate": 8.297437438170797e-08,
"loss": 0.716,
"step": 705
},
{
"epoch": 0.9699453551912568,
"grad_norm": 0.5582179313687514,
"learning_rate": 5.51142611984834e-08,
"loss": 0.7148,
"step": 710
},
{
"epoch": 0.976775956284153,
"grad_norm": 0.5436289814742661,
"learning_rate": 3.2921329747056527e-08,
"loss": 0.7131,
"step": 715
},
{
"epoch": 0.9836065573770492,
"grad_norm": 0.6108023242735494,
"learning_rate": 1.6408226867118404e-08,
"loss": 0.7132,
"step": 720
},
{
"epoch": 0.9904371584699454,
"grad_norm": 0.5559954784052279,
"learning_rate": 5.584362697453882e-09,
"loss": 0.7282,
"step": 725
},
{
"epoch": 0.9972677595628415,
"grad_norm": 1.0362594443574582,
"learning_rate": 4.5590531348227443e-10,
"loss": 0.7112,
"step": 730
},
{
"epoch": 1.0,
"eval_loss": 0.7273545861244202,
"eval_runtime": 9.4727,
"eval_samples_per_second": 52.784,
"eval_steps_per_second": 1.689,
"step": 732
},
{
"epoch": 1.0,
"step": 732,
"total_flos": 106034689474560.0,
"train_loss": 0.7574391521391322,
"train_runtime": 8050.0643,
"train_samples_per_second": 11.633,
"train_steps_per_second": 0.091
}
],
"logging_steps": 5,
"max_steps": 732,
"num_input_tokens_seen": 0,
"num_train_epochs": 1,
"save_steps": 100,
"total_flos": 106034689474560.0,
"train_batch_size": 16,
"trial_name": null,
"trial_params": null
}