| { |
| "best_global_step": null, |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 0.27292576419213976, |
| "eval_steps": 500, |
| "global_step": 500, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "entropy": 0.8506594449281693, |
| "epoch": 0.0054585152838427945, |
| "grad_norm": 6.625, |
| "learning_rate": 1.6363636363636365e-06, |
| "loss": 0.7328, |
| "mean_token_accuracy": 0.8114667266607285, |
| "num_tokens": 7873074.0, |
| "step": 10 |
| }, |
| { |
| "entropy": 0.8578470677137375, |
| "epoch": 0.010917030567685589, |
| "grad_norm": 3.234375, |
| "learning_rate": 3.454545454545455e-06, |
| "loss": 0.6724, |
| "mean_token_accuracy": 0.8153360992670059, |
| "num_tokens": 15735890.0, |
| "step": 20 |
| }, |
| { |
| "entropy": 0.868700721859932, |
| "epoch": 0.016375545851528384, |
| "grad_norm": 0.94140625, |
| "learning_rate": 5.272727272727273e-06, |
| "loss": 0.5629, |
| "mean_token_accuracy": 0.8309339731931686, |
| "num_tokens": 23601330.0, |
| "step": 30 |
| }, |
| { |
| "entropy": 0.8585465878248215, |
| "epoch": 0.021834061135371178, |
| "grad_norm": 0.58984375, |
| "learning_rate": 7.0909090909090916e-06, |
| "loss": 0.5071, |
| "mean_token_accuracy": 0.8447592586278916, |
| "num_tokens": 31458560.0, |
| "step": 40 |
| }, |
| { |
| "entropy": 0.8470574885606765, |
| "epoch": 0.027292576419213975, |
| "grad_norm": 0.5078125, |
| "learning_rate": 8.90909090909091e-06, |
| "loss": 0.4817, |
| "mean_token_accuracy": 0.8515126377344131, |
| "num_tokens": 39318330.0, |
| "step": 50 |
| }, |
| { |
| "entropy": 0.8366546362638474, |
| "epoch": 0.03275109170305677, |
| "grad_norm": 0.412109375, |
| "learning_rate": 9.999874978970385e-06, |
| "loss": 0.4536, |
| "mean_token_accuracy": 0.8550278633832932, |
| "num_tokens": 47187690.0, |
| "step": 60 |
| }, |
| { |
| "entropy": 0.8242797136306763, |
| "epoch": 0.03820960698689956, |
| "grad_norm": 0.5, |
| "learning_rate": 9.998468564187713e-06, |
| "loss": 0.4452, |
| "mean_token_accuracy": 0.8588810175657272, |
| "num_tokens": 55058356.0, |
| "step": 70 |
| }, |
| { |
| "entropy": 0.8176750153303146, |
| "epoch": 0.043668122270742356, |
| "grad_norm": 0.447265625, |
| "learning_rate": 9.995499899369634e-06, |
| "loss": 0.4375, |
| "mean_token_accuracy": 0.860920637845993, |
| "num_tokens": 62923390.0, |
| "step": 80 |
| }, |
| { |
| "entropy": 0.8128144711256027, |
| "epoch": 0.04912663755458515, |
| "grad_norm": 1.0078125, |
| "learning_rate": 9.990969912359675e-06, |
| "loss": 0.4483, |
| "mean_token_accuracy": 0.8596585601568222, |
| "num_tokens": 70785326.0, |
| "step": 90 |
| }, |
| { |
| "entropy": 0.8117900550365448, |
| "epoch": 0.05458515283842795, |
| "grad_norm": 0.486328125, |
| "learning_rate": 9.984880018985966e-06, |
| "loss": 0.4368, |
| "mean_token_accuracy": 0.8614727318286896, |
| "num_tokens": 78646604.0, |
| "step": 100 |
| }, |
| { |
| "entropy": 0.8099668920040131, |
| "epoch": 0.060043668122270744, |
| "grad_norm": 0.68359375, |
| "learning_rate": 9.977232122618704e-06, |
| "loss": 0.4429, |
| "mean_token_accuracy": 0.8630078643560409, |
| "num_tokens": 86509996.0, |
| "step": 110 |
| }, |
| { |
| "entropy": 0.8032392680644989, |
| "epoch": 0.06550218340611354, |
| "grad_norm": 0.6015625, |
| "learning_rate": 9.968028613575292e-06, |
| "loss": 0.4483, |
| "mean_token_accuracy": 0.8638610720634461, |
| "num_tokens": 94374220.0, |
| "step": 120 |
| }, |
| { |
| "entropy": 0.7966970652341843, |
| "epoch": 0.07096069868995633, |
| "grad_norm": 0.546875, |
| "learning_rate": 9.957272368373235e-06, |
| "loss": 0.4306, |
| "mean_token_accuracy": 0.8641012936830521, |
| "num_tokens": 102245070.0, |
| "step": 130 |
| }, |
| { |
| "entropy": 0.7891126424074173, |
| "epoch": 0.07641921397379912, |
| "grad_norm": 0.65234375, |
| "learning_rate": 9.944966748831104e-06, |
| "loss": 0.4376, |
| "mean_token_accuracy": 0.8643449753522873, |
| "num_tokens": 110114714.0, |
| "step": 140 |
| }, |
| { |
| "entropy": 0.7812703251838684, |
| "epoch": 0.08187772925764192, |
| "grad_norm": 0.494140625, |
| "learning_rate": 9.931115601017816e-06, |
| "loss": 0.4412, |
| "mean_token_accuracy": 0.8659342557191849, |
| "num_tokens": 117983612.0, |
| "step": 150 |
| }, |
| { |
| "entropy": 0.7815882682800293, |
| "epoch": 0.08733624454148471, |
| "grad_norm": 0.60546875, |
| "learning_rate": 9.915723254050551e-06, |
| "loss": 0.4264, |
| "mean_token_accuracy": 0.865845575928688, |
| "num_tokens": 125857638.0, |
| "step": 160 |
| }, |
| { |
| "entropy": 0.7799684792757035, |
| "epoch": 0.0927947598253275, |
| "grad_norm": 0.71875, |
| "learning_rate": 9.898794518741725e-06, |
| "loss": 0.4217, |
| "mean_token_accuracy": 0.8657624512910843, |
| "num_tokens": 133718934.0, |
| "step": 170 |
| }, |
| { |
| "entropy": 0.7746502578258514, |
| "epoch": 0.0982532751091703, |
| "grad_norm": 0.50390625, |
| "learning_rate": 9.880334686095365e-06, |
| "loss": 0.4375, |
| "mean_token_accuracy": 0.865676325559616, |
| "num_tokens": 141587252.0, |
| "step": 180 |
| }, |
| { |
| "entropy": 0.7698926359415055, |
| "epoch": 0.1037117903930131, |
| "grad_norm": 0.87890625, |
| "learning_rate": 9.860349525653455e-06, |
| "loss": 0.4142, |
| "mean_token_accuracy": 0.8676840603351593, |
| "num_tokens": 149454258.0, |
| "step": 190 |
| }, |
| { |
| "entropy": 0.7676965594291687, |
| "epoch": 0.1091703056768559, |
| "grad_norm": 0.75390625, |
| "learning_rate": 9.838845283692665e-06, |
| "loss": 0.4132, |
| "mean_token_accuracy": 0.8689810872077942, |
| "num_tokens": 157314466.0, |
| "step": 200 |
| }, |
| { |
| "entropy": 0.7642738163471222, |
| "epoch": 0.1146288209606987, |
| "grad_norm": 0.7265625, |
| "learning_rate": 9.815828681272127e-06, |
| "loss": 0.4098, |
| "mean_token_accuracy": 0.8706588715314865, |
| "num_tokens": 165182774.0, |
| "step": 210 |
| }, |
| { |
| "entropy": 0.7638502717018127, |
| "epoch": 0.12008733624454149, |
| "grad_norm": 1.484375, |
| "learning_rate": 9.791306912132776e-06, |
| "loss": 0.4081, |
| "mean_token_accuracy": 0.8702643483877182, |
| "num_tokens": 173042008.0, |
| "step": 220 |
| }, |
| { |
| "entropy": 0.7644800752401352, |
| "epoch": 0.12554585152838427, |
| "grad_norm": 1.0390625, |
| "learning_rate": 9.765287640448996e-06, |
| "loss": 0.4065, |
| "mean_token_accuracy": 0.8699748694896698, |
| "num_tokens": 180901588.0, |
| "step": 230 |
| }, |
| { |
| "entropy": 0.7578901439905167, |
| "epoch": 0.13100436681222707, |
| "grad_norm": 0.9453125, |
| "learning_rate": 9.737778998433198e-06, |
| "loss": 0.3974, |
| "mean_token_accuracy": 0.8707681357860565, |
| "num_tokens": 188769044.0, |
| "step": 240 |
| }, |
| { |
| "entropy": 0.7537958860397339, |
| "epoch": 0.13646288209606988, |
| "grad_norm": 0.77734375, |
| "learning_rate": 9.708789583794144e-06, |
| "loss": 0.4011, |
| "mean_token_accuracy": 0.8707371085882187, |
| "num_tokens": 196634434.0, |
| "step": 250 |
| }, |
| { |
| "entropy": 0.7602073311805725, |
| "epoch": 0.14192139737991266, |
| "grad_norm": 0.70703125, |
| "learning_rate": 9.678328457049766e-06, |
| "loss": 0.3991, |
| "mean_token_accuracy": 0.8711989611387253, |
| "num_tokens": 204486562.0, |
| "step": 260 |
| }, |
| { |
| "entropy": 0.7597428560256958, |
| "epoch": 0.14737991266375547, |
| "grad_norm": 0.95703125, |
| "learning_rate": 9.646405138695331e-06, |
| "loss": 0.3997, |
| "mean_token_accuracy": 0.8710573464632034, |
| "num_tokens": 212351956.0, |
| "step": 270 |
| }, |
| { |
| "entropy": 0.7577679485082627, |
| "epoch": 0.15283842794759825, |
| "grad_norm": 2.1875, |
| "learning_rate": 9.613029606227867e-06, |
| "loss": 0.399, |
| "mean_token_accuracy": 0.8701757520437241, |
| "num_tokens": 220205590.0, |
| "step": 280 |
| }, |
| { |
| "entropy": 0.7598035603761673, |
| "epoch": 0.15829694323144106, |
| "grad_norm": 2.78125, |
| "learning_rate": 9.57821229102773e-06, |
| "loss": 0.4217, |
| "mean_token_accuracy": 0.8680002480745316, |
| "num_tokens": 228076474.0, |
| "step": 290 |
| }, |
| { |
| "entropy": 0.7590495645999908, |
| "epoch": 0.16375545851528384, |
| "grad_norm": 2.578125, |
| "learning_rate": 9.541964075098313e-06, |
| "loss": 0.4285, |
| "mean_token_accuracy": 0.8676898956298829, |
| "num_tokens": 235936872.0, |
| "step": 300 |
| }, |
| { |
| "entropy": 0.757094269990921, |
| "epoch": 0.16921397379912664, |
| "grad_norm": 2.359375, |
| "learning_rate": 9.50429628766494e-06, |
| "loss": 0.4185, |
| "mean_token_accuracy": 0.8680381834506988, |
| "num_tokens": 243803792.0, |
| "step": 310 |
| }, |
| { |
| "entropy": 0.7542549550533295, |
| "epoch": 0.17467248908296942, |
| "grad_norm": 3.875, |
| "learning_rate": 9.465220701633953e-06, |
| "loss": 0.4055, |
| "mean_token_accuracy": 0.8681020706892013, |
| "num_tokens": 251674582.0, |
| "step": 320 |
| }, |
| { |
| "entropy": 0.7625361144542694, |
| "epoch": 0.18013100436681223, |
| "grad_norm": 1.5, |
| "learning_rate": 9.424749529913138e-06, |
| "loss": 0.4199, |
| "mean_token_accuracy": 0.8669591844081879, |
| "num_tokens": 259538060.0, |
| "step": 330 |
| }, |
| { |
| "entropy": 0.7554975986480713, |
| "epoch": 0.185589519650655, |
| "grad_norm": 0.68359375, |
| "learning_rate": 9.38289542159464e-06, |
| "loss": 0.4121, |
| "mean_token_accuracy": 0.8684647858142853, |
| "num_tokens": 267404072.0, |
| "step": 340 |
| }, |
| { |
| "entropy": 0.7535371214151383, |
| "epoch": 0.19104803493449782, |
| "grad_norm": 0.8828125, |
| "learning_rate": 9.339671458001543e-06, |
| "loss": 0.3962, |
| "mean_token_accuracy": 0.8698400855064392, |
| "num_tokens": 275270380.0, |
| "step": 350 |
| }, |
| { |
| "entropy": 0.754734218120575, |
| "epoch": 0.1965065502183406, |
| "grad_norm": 0.63671875, |
| "learning_rate": 9.295091148599353e-06, |
| "loss": 0.4112, |
| "mean_token_accuracy": 0.8690466344356537, |
| "num_tokens": 283145890.0, |
| "step": 360 |
| }, |
| { |
| "entropy": 0.7563666641712189, |
| "epoch": 0.2019650655021834, |
| "grad_norm": 0.95703125, |
| "learning_rate": 9.249168426773675e-06, |
| "loss": 0.4091, |
| "mean_token_accuracy": 0.8709859728813172, |
| "num_tokens": 291005988.0, |
| "step": 370 |
| }, |
| { |
| "entropy": 0.7565184265375138, |
| "epoch": 0.2074235807860262, |
| "grad_norm": 0.74609375, |
| "learning_rate": 9.201917645475393e-06, |
| "loss": 0.3981, |
| "mean_token_accuracy": 0.8720056504011154, |
| "num_tokens": 298864526.0, |
| "step": 380 |
| }, |
| { |
| "entropy": 0.760869923233986, |
| "epoch": 0.212882096069869, |
| "grad_norm": 2.3125, |
| "learning_rate": 9.153353572734716e-06, |
| "loss": 0.4139, |
| "mean_token_accuracy": 0.869957122206688, |
| "num_tokens": 306735082.0, |
| "step": 390 |
| }, |
| { |
| "entropy": 0.7582542330026627, |
| "epoch": 0.2183406113537118, |
| "grad_norm": 3.609375, |
| "learning_rate": 9.103491387045495e-06, |
| "loss": 0.4144, |
| "mean_token_accuracy": 0.8715152114629745, |
| "num_tokens": 314604158.0, |
| "step": 400 |
| }, |
| { |
| "entropy": 0.7573513597249985, |
| "epoch": 0.22379912663755458, |
| "grad_norm": 1.6484375, |
| "learning_rate": 9.052346672621246e-06, |
| "loss": 0.4132, |
| "mean_token_accuracy": 0.8716101557016372, |
| "num_tokens": 322465290.0, |
| "step": 410 |
| }, |
| { |
| "entropy": 0.7590894103050232, |
| "epoch": 0.2292576419213974, |
| "grad_norm": 1.90625, |
| "learning_rate": 8.999935414524373e-06, |
| "loss": 0.417, |
| "mean_token_accuracy": 0.8695549905300141, |
| "num_tokens": 330336414.0, |
| "step": 420 |
| }, |
| { |
| "entropy": 0.7527199715375901, |
| "epoch": 0.23471615720524017, |
| "grad_norm": 0.8359375, |
| "learning_rate": 8.9462739936701e-06, |
| "loss": 0.4077, |
| "mean_token_accuracy": 0.8720698267221451, |
| "num_tokens": 338189854.0, |
| "step": 430 |
| }, |
| { |
| "entropy": 0.7517860412597657, |
| "epoch": 0.24017467248908297, |
| "grad_norm": 1.0859375, |
| "learning_rate": 8.891379181706697e-06, |
| "loss": 0.3975, |
| "mean_token_accuracy": 0.8719705253839493, |
| "num_tokens": 346051832.0, |
| "step": 440 |
| }, |
| { |
| "entropy": 0.7506425231695175, |
| "epoch": 0.24563318777292575, |
| "grad_norm": 0.90625, |
| "learning_rate": 8.835268135773561e-06, |
| "loss": 0.4144, |
| "mean_token_accuracy": 0.8714013040065766, |
| "num_tokens": 353923892.0, |
| "step": 450 |
| }, |
| { |
| "entropy": 0.7511818319559097, |
| "epoch": 0.25109170305676853, |
| "grad_norm": 0.671875, |
| "learning_rate": 8.777958393138833e-06, |
| "loss": 0.3958, |
| "mean_token_accuracy": 0.8729292124509811, |
| "num_tokens": 361787270.0, |
| "step": 460 |
| }, |
| { |
| "entropy": 0.7484469264745712, |
| "epoch": 0.25655021834061137, |
| "grad_norm": 1.234375, |
| "learning_rate": 8.719467865718192e-06, |
| "loss": 0.4055, |
| "mean_token_accuracy": 0.8726571470499038, |
| "num_tokens": 369659224.0, |
| "step": 470 |
| }, |
| { |
| "entropy": 0.7491981714963913, |
| "epoch": 0.26200873362445415, |
| "grad_norm": 0.9375, |
| "learning_rate": 8.659814834476571e-06, |
| "loss": 0.3889, |
| "mean_token_accuracy": 0.8721846073865891, |
| "num_tokens": 377517650.0, |
| "step": 480 |
| }, |
| { |
| "entropy": 0.745842233300209, |
| "epoch": 0.26746724890829693, |
| "grad_norm": 1.0, |
| "learning_rate": 8.599017943714514e-06, |
| "loss": 0.3995, |
| "mean_token_accuracy": 0.8731576293706894, |
| "num_tokens": 385374510.0, |
| "step": 490 |
| }, |
| { |
| "entropy": 0.7489482790231705, |
| "epoch": 0.27292576419213976, |
| "grad_norm": 0.68359375, |
| "learning_rate": 8.537096195240982e-06, |
| "loss": 0.4032, |
| "mean_token_accuracy": 0.8717446327209473, |
| "num_tokens": 393229928.0, |
| "step": 500 |
| } |
| ], |
| "logging_steps": 10, |
| "max_steps": 1832, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 9223372036854775807, |
| "save_steps": 500, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": false |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 4.896978740027851e+18, |
| "train_batch_size": 16, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
