| [ |
| { |
| "loss": 1.9427, |
| "grad_norm": 0.9304783344268799, |
| "learning_rate": 1.8e-05, |
| "entropy": 1.4844072341918946, |
| "num_tokens": 236335.0, |
| "mean_token_accuracy": 0.5745298147201539, |
| "epoch": 0.016, |
| "step": 10 |
| }, |
| { |
| "loss": 1.8013, |
| "grad_norm": 0.723517656326294, |
| "learning_rate": 3.8e-05, |
| "entropy": 1.7161851286888123, |
| "num_tokens": 471120.0, |
| "mean_token_accuracy": 0.5827198266983032, |
| "epoch": 0.032, |
| "step": 20 |
| }, |
| { |
| "loss": 1.4628, |
| "grad_norm": 1.4500237703323364, |
| "learning_rate": 5.8e-05, |
| "entropy": 1.5137762308120728, |
| "num_tokens": 713049.0, |
| "mean_token_accuracy": 0.6408967107534409, |
| "epoch": 0.048, |
| "step": 30 |
| }, |
| { |
| "loss": 1.0328, |
| "grad_norm": 0.41915225982666016, |
| "learning_rate": 7.800000000000001e-05, |
| "entropy": 1.0311345964670182, |
| "num_tokens": 950524.0, |
| "mean_token_accuracy": 0.7369566798210144, |
| "epoch": 0.064, |
| "step": 40 |
| }, |
| { |
| "loss": 0.9175, |
| "grad_norm": 0.3350801169872284, |
| "learning_rate": 9.8e-05, |
| "entropy": 0.9200641274452209, |
| "num_tokens": 1181734.0, |
| "mean_token_accuracy": 0.7648449391126633, |
| "epoch": 0.08, |
| "step": 50 |
| }, |
| { |
| "loss": 0.8685, |
| "grad_norm": 0.33790573477745056, |
| "learning_rate": 9.993956318446873e-05, |
| "entropy": 0.8751217514276505, |
| "num_tokens": 1417485.0, |
| "mean_token_accuracy": 0.7721945941448212, |
| "epoch": 0.096, |
| "step": 60 |
| }, |
| { |
| "loss": 0.8228, |
| "grad_norm": 0.3235560357570648, |
| "learning_rate": 9.973083336646172e-05, |
| "entropy": 0.8255001336336136, |
| "num_tokens": 1654923.0, |
| "mean_token_accuracy": 0.7804904520511627, |
| "epoch": 0.112, |
| "step": 70 |
| }, |
| { |
| "loss": 0.7851, |
| "grad_norm": 0.34785518050193787, |
| "learning_rate": 9.937368719576374e-05, |
| "entropy": 0.7923080742359161, |
| "num_tokens": 1893268.0, |
| "mean_token_accuracy": 0.7877459049224853, |
| "epoch": 0.128, |
| "step": 80 |
| }, |
| { |
| "loss": 0.784, |
| "grad_norm": 0.33706751465797424, |
| "learning_rate": 9.886919053691883e-05, |
| "entropy": 0.7805845379829407, |
| "num_tokens": 2131912.0, |
| "mean_token_accuracy": 0.7893630683422088, |
| "epoch": 0.144, |
| "step": 90 |
| }, |
| { |
| "loss": 0.7473, |
| "grad_norm": 0.3280414938926697, |
| "learning_rate": 9.82188490062415e-05, |
| "entropy": 0.7436369717121124, |
| "num_tokens": 2371393.0, |
| "mean_token_accuracy": 0.7966683268547058, |
| "epoch": 0.16, |
| "step": 100 |
| }, |
| { |
| "loss": 0.7439, |
| "grad_norm": 0.3483212888240814, |
| "learning_rate": 9.742460347846587e-05, |
| "entropy": 0.7480131626129151, |
| "num_tokens": 2608341.0, |
| "mean_token_accuracy": 0.7970409005880356, |
| "epoch": 0.176, |
| "step": 110 |
| }, |
| { |
| "loss": 0.741, |
| "grad_norm": 0.34795472025871277, |
| "learning_rate": 9.648882429441257e-05, |
| "entropy": 0.7411326467990875, |
| "num_tokens": 2845727.0, |
| "mean_token_accuracy": 0.7980288416147232, |
| "epoch": 0.192, |
| "step": 120 |
| }, |
| { |
| "loss": 0.7874, |
| "grad_norm": 0.4145350754261017, |
| "learning_rate": 9.541430418696018e-05, |
| "entropy": 0.7868094980716706, |
| "num_tokens": 3082504.0, |
| "mean_token_accuracy": 0.7893972307443619, |
| "epoch": 0.208, |
| "step": 130 |
| }, |
| { |
| "loss": 0.7625, |
| "grad_norm": 1.9233369827270508, |
| "learning_rate": 9.420424994643266e-05, |
| "entropy": 0.7673351168632507, |
| "num_tokens": 3318727.0, |
| "mean_token_accuracy": 0.7942031443119049, |
| "epoch": 0.224, |
| "step": 140 |
| }, |
| { |
| "loss": 0.7391, |
| "grad_norm": 0.43922004103660583, |
| "learning_rate": 9.28622728502766e-05, |
| "entropy": 0.7414912819862366, |
| "num_tokens": 3551849.0, |
| "mean_token_accuracy": 0.7978828340768814, |
| "epoch": 0.24, |
| "step": 150 |
| }, |
| { |
| "loss": 0.702, |
| "grad_norm": 0.3687349259853363, |
| "learning_rate": 9.139237788559e-05, |
| "entropy": 0.7002590656280517, |
| "num_tokens": 3784063.0, |
| "mean_token_accuracy": 0.8059363931417465, |
| "epoch": 0.256, |
| "step": 160 |
| }, |
| { |
| "loss": 0.7214, |
| "grad_norm": 0.3384023904800415, |
| "learning_rate": 8.979895179666673e-05, |
| "entropy": 0.7226290732622147, |
| "num_tokens": 4030658.0, |
| "mean_token_accuracy": 0.80358906686306, |
| "epoch": 0.272, |
| "step": 170 |
| }, |
| { |
| "loss": 0.7132, |
| "grad_norm": 0.39796605706214905, |
| "learning_rate": 8.808674999322728e-05, |
| "entropy": 0.7169670343399048, |
| "num_tokens": 4264808.0, |
| "mean_token_accuracy": 0.8048111826181412, |
| "epoch": 0.288, |
| "step": 180 |
| }, |
| { |
| "loss": 0.721, |
| "grad_norm": 0.4140866696834564, |
| "learning_rate": 8.626088235840726e-05, |
| "entropy": 0.7286807566881179, |
| "num_tokens": 4500615.0, |
| "mean_token_accuracy": 0.8025989532470703, |
| "epoch": 0.304, |
| "step": 190 |
| }, |
| { |
| "loss": 0.693, |
| "grad_norm": 0.35368794202804565, |
| "learning_rate": 8.43267979988576e-05, |
| "entropy": 0.6942046850919723, |
| "num_tokens": 4727422.0, |
| "mean_token_accuracy": 0.8079935729503631, |
| "epoch": 0.32, |
| "step": 200 |
| }, |
| { |
| "loss": 0.6868, |
| "grad_norm": 0.36108162999153137, |
| "learning_rate": 8.229026898246885e-05, |
| "entropy": 0.6922369688749314, |
| "num_tokens": 4970679.0, |
| "mean_token_accuracy": 0.8092731922864914, |
| "epoch": 0.336, |
| "step": 210 |
| }, |
| { |
| "loss": 0.6775, |
| "grad_norm": 0.41774165630340576, |
| "learning_rate": 8.015737311225172e-05, |
| "entropy": 0.6763984620571136, |
| "num_tokens": 5205857.0, |
| "mean_token_accuracy": 0.8115724682807922, |
| "epoch": 0.352, |
| "step": 220 |
| }, |
| { |
| "loss": 0.6551, |
| "grad_norm": 0.3897970914840698, |
| "learning_rate": 7.793447578778427e-05, |
| "entropy": 0.6601776152849197, |
| "num_tokens": 5440774.0, |
| "mean_token_accuracy": 0.8163723379373551, |
| "epoch": 0.368, |
| "step": 230 |
| }, |
| { |
| "loss": 0.694, |
| "grad_norm": 0.4008243978023529, |
| "learning_rate": 7.56282110083577e-05, |
| "entropy": 0.7006691724061966, |
| "num_tokens": 5681945.0, |
| "mean_token_accuracy": 0.8072743356227875, |
| "epoch": 0.384, |
| "step": 240 |
| }, |
| { |
| "loss": 0.6635, |
| "grad_norm": 0.38448140025138855, |
| "learning_rate": 7.324546157451508e-05, |
| "entropy": 0.6688638418912888, |
| "num_tokens": 5911653.0, |
| "mean_token_accuracy": 0.8145491361618042, |
| "epoch": 0.4, |
| "step": 250 |
| }, |
| { |
| "loss": 0.6778, |
| "grad_norm": 0.3691696226596832, |
| "learning_rate": 7.079333854706938e-05, |
| "entropy": 0.6748764783143997, |
| "num_tokens": 6153386.0, |
| "mean_token_accuracy": 0.8123239934444427, |
| "epoch": 0.416, |
| "step": 260 |
| }, |
| { |
| "loss": 0.6931, |
| "grad_norm": 0.34841006994247437, |
| "learning_rate": 6.827916002490304e-05, |
| "entropy": 0.698054638504982, |
| "num_tokens": 6392270.0, |
| "mean_token_accuracy": 0.8090349048376083, |
| "epoch": 0.432, |
| "step": 270 |
| }, |
| { |
| "loss": 0.6833, |
| "grad_norm": 0.36822986602783203, |
| "learning_rate": 6.571042930488474e-05, |
| "entropy": 0.6903767049312591, |
| "num_tokens": 6628878.0, |
| "mean_token_accuracy": 0.8101357609033585, |
| "epoch": 0.448, |
| "step": 280 |
| }, |
| { |
| "loss": 0.6587, |
| "grad_norm": 0.3246494233608246, |
| "learning_rate": 6.309481248908235e-05, |
| "entropy": 0.6623686283826828, |
| "num_tokens": 6866329.0, |
| "mean_token_accuracy": 0.8156650364398956, |
| "epoch": 0.464, |
| "step": 290 |
| }, |
| { |
| "loss": 0.6773, |
| "grad_norm": 0.3834746778011322, |
| "learning_rate": 6.044011560610153e-05, |
| "entropy": 0.6803743422031403, |
| "num_tokens": 7096086.0, |
| "mean_token_accuracy": 0.8116874188184738, |
| "epoch": 0.48, |
| "step": 300 |
| }, |
| { |
| "loss": 0.6648, |
| "grad_norm": 0.33701953291893005, |
| "learning_rate": 5.775426131482811e-05, |
| "entropy": 0.6692022681236267, |
| "num_tokens": 7332753.0, |
| "mean_token_accuracy": 0.8145207434892654, |
| "epoch": 0.496, |
| "step": 310 |
| }, |
| { |
| "loss": 0.6334, |
| "grad_norm": 0.3483853340148926, |
| "learning_rate": 5.5045265260100234e-05, |
| "entropy": 0.6391948401927948, |
| "num_tokens": 7562764.0, |
| "mean_token_accuracy": 0.8210806369781494, |
| "epoch": 0.512, |
| "step": 320 |
| }, |
| { |
| "loss": 0.6424, |
| "grad_norm": 0.2986826002597809, |
| "learning_rate": 5.232121215087369e-05, |
| "entropy": 0.6503159761428833, |
| "num_tokens": 7810030.0, |
| "mean_token_accuracy": 0.8197412937879562, |
| "epoch": 0.528, |
| "step": 330 |
| }, |
| { |
| "loss": 0.6784, |
| "grad_norm": 0.3261159360408783, |
| "learning_rate": 4.959023163227284e-05, |
| "entropy": 0.6812007486820221, |
| "num_tokens": 8046816.0, |
| "mean_token_accuracy": 0.8112091690301895, |
| "epoch": 0.544, |
| "step": 340 |
| }, |
| { |
| "loss": 0.6711, |
| "grad_norm": 0.37481242418289185, |
| "learning_rate": 4.6860474023534335e-05, |
| "entropy": 0.6745136559009552, |
| "num_tokens": 8279743.0, |
| "mean_token_accuracy": 0.8127087384462357, |
| "epoch": 0.56, |
| "step": 350 |
| }, |
| { |
| "loss": 0.6573, |
| "grad_norm": 0.38329869508743286, |
| "learning_rate": 4.4140085994251136e-05, |
| "entropy": 0.657204994559288, |
| "num_tokens": 8513323.0, |
| "mean_token_accuracy": 0.8160663604736328, |
| "epoch": 0.576, |
| "step": 360 |
| }, |
| { |
| "loss": 0.6356, |
| "grad_norm": 0.4051035940647125, |
| "learning_rate": 4.143718625150854e-05, |
| "entropy": 0.6373627185821533, |
| "num_tokens": 8744995.0, |
| "mean_token_accuracy": 0.8212027668952941, |
| "epoch": 0.592, |
| "step": 370 |
| }, |
| { |
| "loss": 0.6721, |
| "grad_norm": 0.34723058342933655, |
| "learning_rate": 3.875984131047135e-05, |
| "entropy": 0.6792070597410202, |
| "num_tokens": 8976168.0, |
| "mean_token_accuracy": 0.8130046099424362, |
| "epoch": 0.608, |
| "step": 380 |
| }, |
| { |
| "loss": 0.6579, |
| "grad_norm": 0.3466307520866394, |
| "learning_rate": 3.6116041420732103e-05, |
| "entropy": 0.6591042011976243, |
| "num_tokens": 9217079.0, |
| "mean_token_accuracy": 0.8165153503417969, |
| "epoch": 0.624, |
| "step": 390 |
| }, |
| { |
| "loss": 0.655, |
| "grad_norm": 0.3664017915725708, |
| "learning_rate": 3.3513676720265934e-05, |
| "entropy": 0.6642171382904053, |
| "num_tokens": 9448775.0, |
| "mean_token_accuracy": 0.8166080981492996, |
| "epoch": 0.64, |
| "step": 400 |
| }, |
| { |
| "loss": 0.6423, |
| "grad_norm": 0.35895591974258423, |
| "learning_rate": 3.0960513688157754e-05, |
| "entropy": 0.6416015654802323, |
| "num_tokens": 9685315.0, |
| "mean_token_accuracy": 0.8195697069168091, |
| "epoch": 0.656, |
| "step": 410 |
| }, |
| { |
| "loss": 0.6279, |
| "grad_norm": 0.3837035894393921, |
| "learning_rate": 2.846417196637613e-05, |
| "entropy": 0.6338138192892074, |
| "num_tokens": 9922050.0, |
| "mean_token_accuracy": 0.8214808911085129, |
| "epoch": 0.672, |
| "step": 420 |
| }, |
| { |
| "loss": 0.6638, |
| "grad_norm": 0.3423479199409485, |
| "learning_rate": 2.603210161976687e-05, |
| "entropy": 0.6659493237733841, |
| "num_tokens": 10152825.0, |
| "mean_token_accuracy": 0.8158191829919815, |
| "epoch": 0.688, |
| "step": 430 |
| }, |
| { |
| "loss": 0.6551, |
| "grad_norm": 0.3698706328868866, |
| "learning_rate": 2.3671560902131447e-05, |
| "entropy": 0.654210501909256, |
| "num_tokens": 10389209.0, |
| "mean_token_accuracy": 0.8156277418136597, |
| "epoch": 0.704, |
| "step": 440 |
| }, |
| { |
| "loss": 0.6283, |
| "grad_norm": 0.38697871565818787, |
| "learning_rate": 2.1389594594744864e-05, |
| "entropy": 0.6335553735494613, |
| "num_tokens": 10627038.0, |
| "mean_token_accuracy": 0.8218480378389359, |
| "epoch": 0.72, |
| "step": 450 |
| }, |
| { |
| "loss": 0.6455, |
| "grad_norm": 0.33989399671554565, |
| "learning_rate": 1.9193012981959475e-05, |
| "entropy": 0.6531029403209686, |
| "num_tokens": 10865908.0, |
| "mean_token_accuracy": 0.8178681671619416, |
| "epoch": 0.736, |
| "step": 460 |
| }, |
| { |
| "loss": 0.6205, |
| "grad_norm": 0.37868356704711914, |
| "learning_rate": 1.7088371526639853e-05, |
| "entropy": 0.6222113043069839, |
| "num_tokens": 11105057.0, |
| "mean_token_accuracy": 0.8235622078180314, |
| "epoch": 0.752, |
| "step": 470 |
| }, |
| { |
| "loss": 0.6771, |
| "grad_norm": 0.35013335943222046, |
| "learning_rate": 1.508195130608504e-05, |
| "entropy": 0.6883874475955963, |
| "num_tokens": 11347042.0, |
| "mean_token_accuracy": 0.8107833832502365, |
| "epoch": 0.768, |
| "step": 480 |
| }, |
| { |
| "loss": 0.6578, |
| "grad_norm": 0.3570718467235565, |
| "learning_rate": 1.3179740266825253e-05, |
| "entropy": 0.6517588019371032, |
| "num_tokens": 11585916.0, |
| "mean_token_accuracy": 0.8149342775344849, |
| "epoch": 0.784, |
| "step": 490 |
| }, |
| { |
| "loss": 0.6338, |
| "grad_norm": 0.3437284529209137, |
| "learning_rate": 1.1387415354235887e-05, |
| "entropy": 0.6454322338104248, |
| "num_tokens": 11824489.0, |
| "mean_token_accuracy": 0.8217582374811172, |
| "epoch": 0.8, |
| "step": 500 |
| }, |
| { |
| "loss": 0.6403, |
| "grad_norm": 0.3438985347747803, |
| "learning_rate": 9.710325570301348e-06, |
| "entropy": 0.6410525500774383, |
| "num_tokens": 12062163.0, |
| "mean_token_accuracy": 0.8210669010877609, |
| "epoch": 0.816, |
| "step": 510 |
| }, |
| { |
| "loss": 0.6843, |
| "grad_norm": 0.3453381359577179, |
| "learning_rate": 8.153476010090788e-06, |
| "entropy": 0.6865164041519165, |
| "num_tokens": 12302722.0, |
| "mean_token_accuracy": 0.8096443176269531, |
| "epoch": 0.832, |
| "step": 520 |
| }, |
| { |
| "loss": 0.6299, |
| "grad_norm": 0.3145497143268585, |
| "learning_rate": 6.721512924587242e-06, |
| "entropy": 0.6282441258430481, |
| "num_tokens": 12542828.0, |
| "mean_token_accuracy": 0.8224331319332123, |
| "epoch": 0.848, |
| "step": 530 |
| }, |
| { |
| "loss": 0.6521, |
| "grad_norm": 0.7085707783699036, |
| "learning_rate": 5.418709854448489e-06, |
| "entropy": 0.658638134598732, |
| "num_tokens": 12778750.0, |
| "mean_token_accuracy": 0.8189221560955048, |
| "epoch": 0.864, |
| "step": 540 |
| }, |
| { |
| "loss": 0.6516, |
| "grad_norm": 0.3605988025665283, |
| "learning_rate": 4.248954876082195e-06, |
| "entropy": 0.6549220532178879, |
| "num_tokens": 13017993.0, |
| "mean_token_accuracy": 0.8162762761116028, |
| "epoch": 0.88, |
| "step": 550 |
| }, |
| { |
| "loss": 0.6227, |
| "grad_norm": 0.3576490581035614, |
| "learning_rate": 3.2157389980976783e-06, |
| "entropy": 0.6235672175884247, |
| "num_tokens": 13248686.0, |
| "mean_token_accuracy": 0.8248269021511078, |
| "epoch": 0.896, |
| "step": 560 |
| }, |
| { |
| "loss": 0.6533, |
| "grad_norm": 0.35294073820114136, |
| "learning_rate": 2.322145742764181e-06, |
| "entropy": 0.6572938799858093, |
| "num_tokens": 13479101.0, |
| "mean_token_accuracy": 0.8174732059240342, |
| "epoch": 0.912, |
| "step": 570 |
| }, |
| { |
| "loss": 0.6294, |
| "grad_norm": 0.3723297417163849, |
| "learning_rate": 1.5708419435684462e-06, |
| "entropy": 0.6358791559934616, |
| "num_tokens": 13711722.0, |
| "mean_token_accuracy": 0.8224119067192077, |
| "epoch": 0.928, |
| "step": 580 |
| }, |
| { |
| "loss": 0.6359, |
| "grad_norm": 0.3650200068950653, |
| "learning_rate": 9.640697863354698e-07, |
| "entropy": 0.6374465972185135, |
| "num_tokens": 13947211.0, |
| "mean_token_accuracy": 0.821335780620575, |
| "epoch": 0.944, |
| "step": 590 |
| }, |
| { |
| "loss": 0.6595, |
| "grad_norm": 0.3581603169441223, |
| "learning_rate": 5.036401176647332e-07, |
| "entropy": 0.6631550431251526, |
| "num_tokens": 14186581.0, |
| "mean_token_accuracy": 0.8151085019111634, |
| "epoch": 0.96, |
| "step": 600 |
| }, |
| { |
| "loss": 0.6501, |
| "grad_norm": 0.36077970266342163, |
| "learning_rate": 1.9092704065222788e-07, |
| "entropy": 0.6615756928920746, |
| "num_tokens": 14416922.0, |
| "mean_token_accuracy": 0.8183146119117737, |
| "epoch": 0.976, |
| "step": 610 |
| }, |
| { |
| "loss": 0.6589, |
| "grad_norm": 0.3554404377937317, |
| "learning_rate": 2.6863814026734102e-08, |
| "entropy": 0.6655615210533142, |
| "num_tokens": 14652110.0, |
| "mean_token_accuracy": 0.8152088820934296, |
| "epoch": 0.992, |
| "step": 620 |
| }, |
| { |
| "train_runtime": 6311.5248, |
| "train_samples_per_second": 3.167, |
| "train_steps_per_second": 0.099, |
| "total_flos": 6.939581601096253e+18, |
| "train_loss": 0.7419871192932129, |
| "entropy": 0.6707779943943024, |
| "num_tokens": 14764501.0, |
| "mean_token_accuracy": 0.8151742398738862, |
| "epoch": 1.0, |
| "step": 625 |
| } |
| ] |
