| { | |
| "best_metric": 0.2866949737071991, | |
| "best_model_checkpoint": "finetuned-indian-food/checkpoint-1300", | |
| "epoch": 4.0, | |
| "eval_steps": 100, | |
| "global_step": 1332, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.03003003003003003, | |
| "grad_norm": 2.0226473808288574, | |
| "learning_rate": 0.0001984984984984985, | |
| "loss": 2.8635, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.06006006006006006, | |
| "grad_norm": 1.8129218816757202, | |
| "learning_rate": 0.00019699699699699701, | |
| "loss": 2.5393, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.09009009009009009, | |
| "grad_norm": 1.946191668510437, | |
| "learning_rate": 0.0001954954954954955, | |
| "loss": 2.2093, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.12012012012012012, | |
| "grad_norm": 2.3910531997680664, | |
| "learning_rate": 0.00019399399399399402, | |
| "loss": 1.8851, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.15015015015015015, | |
| "grad_norm": 2.2553648948669434, | |
| "learning_rate": 0.0001924924924924925, | |
| "loss": 1.6156, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.18018018018018017, | |
| "grad_norm": 2.605876922607422, | |
| "learning_rate": 0.000190990990990991, | |
| "loss": 1.4386, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.21021021021021022, | |
| "grad_norm": 3.005312919616699, | |
| "learning_rate": 0.0001894894894894895, | |
| "loss": 1.3945, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.24024024024024024, | |
| "grad_norm": 2.530965805053711, | |
| "learning_rate": 0.000187987987987988, | |
| "loss": 1.1486, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.2702702702702703, | |
| "grad_norm": 4.0471720695495605, | |
| "learning_rate": 0.0001864864864864865, | |
| "loss": 1.1218, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.3003003003003003, | |
| "grad_norm": 2.0968387126922607, | |
| "learning_rate": 0.000184984984984985, | |
| "loss": 1.0192, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.3003003003003003, | |
| "eval_accuracy": 0.8480340063761955, | |
| "eval_loss": 0.9247535467147827, | |
| "eval_runtime": 29.0203, | |
| "eval_samples_per_second": 32.426, | |
| "eval_steps_per_second": 4.066, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.3303303303303303, | |
| "grad_norm": 3.351391315460205, | |
| "learning_rate": 0.0001834834834834835, | |
| "loss": 0.9933, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.36036036036036034, | |
| "grad_norm": 3.367605686187744, | |
| "learning_rate": 0.000181981981981982, | |
| "loss": 1.0123, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.39039039039039036, | |
| "grad_norm": 2.310729742050171, | |
| "learning_rate": 0.0001804804804804805, | |
| "loss": 0.9198, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.42042042042042044, | |
| "grad_norm": 5.448969841003418, | |
| "learning_rate": 0.00017897897897897898, | |
| "loss": 0.818, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.45045045045045046, | |
| "grad_norm": 3.953378200531006, | |
| "learning_rate": 0.0001774774774774775, | |
| "loss": 0.8839, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.4804804804804805, | |
| "grad_norm": 4.24240779876709, | |
| "learning_rate": 0.000175975975975976, | |
| "loss": 0.8652, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.5105105105105106, | |
| "grad_norm": 3.4866714477539062, | |
| "learning_rate": 0.0001744744744744745, | |
| "loss": 0.7593, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.5405405405405406, | |
| "grad_norm": 2.6095967292785645, | |
| "learning_rate": 0.000172972972972973, | |
| "loss": 0.6274, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.5705705705705706, | |
| "grad_norm": 2.582280397415161, | |
| "learning_rate": 0.00017147147147147148, | |
| "loss": 0.6867, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.6006006006006006, | |
| "grad_norm": 3.176654577255249, | |
| "learning_rate": 0.00016996996996997, | |
| "loss": 0.635, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.6006006006006006, | |
| "eval_accuracy": 0.8862911795961743, | |
| "eval_loss": 0.5916650891304016, | |
| "eval_runtime": 29.2674, | |
| "eval_samples_per_second": 32.152, | |
| "eval_steps_per_second": 4.032, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.6306306306306306, | |
| "grad_norm": 6.385357856750488, | |
| "learning_rate": 0.00016846846846846846, | |
| "loss": 0.715, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.6606606606606606, | |
| "grad_norm": 2.321627378463745, | |
| "learning_rate": 0.00016696696696696697, | |
| "loss": 0.6047, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.6906906906906907, | |
| "grad_norm": 3.8409202098846436, | |
| "learning_rate": 0.00016546546546546546, | |
| "loss": 0.7312, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.7207207207207207, | |
| "grad_norm": 4.827725887298584, | |
| "learning_rate": 0.00016396396396396395, | |
| "loss": 0.6743, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.7507507507507507, | |
| "grad_norm": 3.8321993350982666, | |
| "learning_rate": 0.00016246246246246247, | |
| "loss": 0.779, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.7807807807807807, | |
| "grad_norm": 5.7290167808532715, | |
| "learning_rate": 0.00016096096096096096, | |
| "loss": 0.7208, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.8108108108108109, | |
| "grad_norm": 1.1494112014770508, | |
| "learning_rate": 0.00015945945945945947, | |
| "loss": 0.7065, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.8408408408408409, | |
| "grad_norm": 2.5776023864746094, | |
| "learning_rate": 0.00015795795795795796, | |
| "loss": 0.4774, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.8708708708708709, | |
| "grad_norm": 3.333287239074707, | |
| "learning_rate": 0.00015645645645645645, | |
| "loss": 0.6134, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.9009009009009009, | |
| "grad_norm": 2.337012529373169, | |
| "learning_rate": 0.00015495495495495496, | |
| "loss": 0.6523, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.9009009009009009, | |
| "eval_accuracy": 0.8799149840595112, | |
| "eval_loss": 0.5133848786354065, | |
| "eval_runtime": 28.7003, | |
| "eval_samples_per_second": 32.787, | |
| "eval_steps_per_second": 4.111, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.9309309309309309, | |
| "grad_norm": 2.1757540702819824, | |
| "learning_rate": 0.00015345345345345345, | |
| "loss": 0.5388, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.960960960960961, | |
| "grad_norm": 3.668917655944824, | |
| "learning_rate": 0.00015195195195195194, | |
| "loss": 0.5271, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.990990990990991, | |
| "grad_norm": 2.8729825019836426, | |
| "learning_rate": 0.00015045045045045046, | |
| "loss": 0.3867, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 1.021021021021021, | |
| "grad_norm": 1.3619645833969116, | |
| "learning_rate": 0.00014894894894894895, | |
| "loss": 0.4743, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 1.0510510510510511, | |
| "grad_norm": 2.4344875812530518, | |
| "learning_rate": 0.00014744744744744746, | |
| "loss": 0.4791, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.0810810810810811, | |
| "grad_norm": 2.5022900104522705, | |
| "learning_rate": 0.00014594594594594595, | |
| "loss": 0.4244, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 1.1111111111111112, | |
| "grad_norm": 0.42008906602859497, | |
| "learning_rate": 0.00014444444444444444, | |
| "loss": 0.4419, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 1.1411411411411412, | |
| "grad_norm": 2.006427049636841, | |
| "learning_rate": 0.00014294294294294295, | |
| "loss": 0.3806, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 1.1711711711711712, | |
| "grad_norm": 4.105851650238037, | |
| "learning_rate": 0.00014144144144144144, | |
| "loss": 0.5661, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 1.2012012012012012, | |
| "grad_norm": 1.2595148086547852, | |
| "learning_rate": 0.00013993993993993996, | |
| "loss": 0.4247, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.2012012012012012, | |
| "eval_accuracy": 0.9043570669500531, | |
| "eval_loss": 0.39834967255592346, | |
| "eval_runtime": 28.9318, | |
| "eval_samples_per_second": 32.525, | |
| "eval_steps_per_second": 4.079, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.2312312312312312, | |
| "grad_norm": 1.719741702079773, | |
| "learning_rate": 0.00013843843843843845, | |
| "loss": 0.3838, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 1.2612612612612613, | |
| "grad_norm": 3.1315104961395264, | |
| "learning_rate": 0.00013693693693693693, | |
| "loss": 0.3394, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 1.2912912912912913, | |
| "grad_norm": 1.0376994609832764, | |
| "learning_rate": 0.00013543543543543545, | |
| "loss": 0.4235, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 1.3213213213213213, | |
| "grad_norm": 4.525661945343018, | |
| "learning_rate": 0.00013393393393393394, | |
| "loss": 0.4403, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 1.3513513513513513, | |
| "grad_norm": 5.646125316619873, | |
| "learning_rate": 0.00013243243243243243, | |
| "loss": 0.3864, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 1.3813813813813813, | |
| "grad_norm": 3.0982725620269775, | |
| "learning_rate": 0.00013093093093093094, | |
| "loss": 0.4172, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 1.4114114114114114, | |
| "grad_norm": 5.521030426025391, | |
| "learning_rate": 0.00012942942942942943, | |
| "loss": 0.3958, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 1.4414414414414414, | |
| "grad_norm": 5.819517135620117, | |
| "learning_rate": 0.00012792792792792795, | |
| "loss": 0.5228, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 1.4714714714714714, | |
| "grad_norm": 7.061979293823242, | |
| "learning_rate": 0.00012642642642642644, | |
| "loss": 0.3269, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 1.5015015015015014, | |
| "grad_norm": 0.9529209136962891, | |
| "learning_rate": 0.00012492492492492492, | |
| "loss": 0.4393, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.5015015015015014, | |
| "eval_accuracy": 0.89798087141339, | |
| "eval_loss": 0.4118588864803314, | |
| "eval_runtime": 29.0892, | |
| "eval_samples_per_second": 32.349, | |
| "eval_steps_per_second": 4.056, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.5315315315315314, | |
| "grad_norm": 2.104562520980835, | |
| "learning_rate": 0.00012342342342342344, | |
| "loss": 0.3378, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 1.5615615615615615, | |
| "grad_norm": 6.228355884552002, | |
| "learning_rate": 0.00012192192192192193, | |
| "loss": 0.385, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 1.5915915915915915, | |
| "grad_norm": 3.3202242851257324, | |
| "learning_rate": 0.00012042042042042043, | |
| "loss": 0.4868, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 1.6216216216216215, | |
| "grad_norm": 3.134051561355591, | |
| "learning_rate": 0.00011891891891891893, | |
| "loss": 0.3764, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 1.6516516516516515, | |
| "grad_norm": 1.982640027999878, | |
| "learning_rate": 0.00011741741741741743, | |
| "loss": 0.4304, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 1.6816816816816815, | |
| "grad_norm": 3.3197553157806396, | |
| "learning_rate": 0.00011591591591591592, | |
| "loss": 0.4026, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 1.7117117117117115, | |
| "grad_norm": 4.031904697418213, | |
| "learning_rate": 0.00011441441441441443, | |
| "loss": 0.4006, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 1.7417417417417418, | |
| "grad_norm": 2.91866135597229, | |
| "learning_rate": 0.00011291291291291293, | |
| "loss": 0.4367, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 1.7717717717717718, | |
| "grad_norm": 6.890028953552246, | |
| "learning_rate": 0.00011141141141141143, | |
| "loss": 0.4345, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 1.8018018018018018, | |
| "grad_norm": 7.291341781616211, | |
| "learning_rate": 0.00010990990990990993, | |
| "loss": 0.4631, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.8018018018018018, | |
| "eval_accuracy": 0.9107332624867163, | |
| "eval_loss": 0.3752017319202423, | |
| "eval_runtime": 28.7185, | |
| "eval_samples_per_second": 32.766, | |
| "eval_steps_per_second": 4.109, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.8318318318318318, | |
| "grad_norm": 6.547687530517578, | |
| "learning_rate": 0.00010840840840840842, | |
| "loss": 0.3953, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 1.8618618618618619, | |
| "grad_norm": 3.9635071754455566, | |
| "learning_rate": 0.00010690690690690692, | |
| "loss": 0.3499, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 1.8918918918918919, | |
| "grad_norm": 2.1902055740356445, | |
| "learning_rate": 0.0001054054054054054, | |
| "loss": 0.3476, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 1.921921921921922, | |
| "grad_norm": 2.9604856967926025, | |
| "learning_rate": 0.0001039039039039039, | |
| "loss": 0.3693, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 1.951951951951952, | |
| "grad_norm": 1.3020485639572144, | |
| "learning_rate": 0.0001024024024024024, | |
| "loss": 0.3556, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 1.981981981981982, | |
| "grad_norm": 1.8542765378952026, | |
| "learning_rate": 0.00010090090090090089, | |
| "loss": 0.3287, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 2.012012012012012, | |
| "grad_norm": 1.039650559425354, | |
| "learning_rate": 9.93993993993994e-05, | |
| "loss": 0.3603, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 2.042042042042042, | |
| "grad_norm": 2.5295028686523438, | |
| "learning_rate": 9.789789789789791e-05, | |
| "loss": 0.2565, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 2.0720720720720722, | |
| "grad_norm": 5.409499168395996, | |
| "learning_rate": 9.639639639639641e-05, | |
| "loss": 0.3467, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 2.1021021021021022, | |
| "grad_norm": 2.797410249710083, | |
| "learning_rate": 9.48948948948949e-05, | |
| "loss": 0.2992, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 2.1021021021021022, | |
| "eval_accuracy": 0.9128586609989373, | |
| "eval_loss": 0.3469248116016388, | |
| "eval_runtime": 28.8507, | |
| "eval_samples_per_second": 32.616, | |
| "eval_steps_per_second": 4.09, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 2.1321321321321323, | |
| "grad_norm": 6.1658196449279785, | |
| "learning_rate": 9.33933933933934e-05, | |
| "loss": 0.3511, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 2.1621621621621623, | |
| "grad_norm": 5.815499305725098, | |
| "learning_rate": 9.18918918918919e-05, | |
| "loss": 0.2395, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 2.1921921921921923, | |
| "grad_norm": 4.747284412384033, | |
| "learning_rate": 9.039039039039039e-05, | |
| "loss": 0.1959, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 2.2222222222222223, | |
| "grad_norm": 2.6975574493408203, | |
| "learning_rate": 8.888888888888889e-05, | |
| "loss": 0.322, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 2.2522522522522523, | |
| "grad_norm": 6.27766752243042, | |
| "learning_rate": 8.738738738738738e-05, | |
| "loss": 0.3862, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 2.2822822822822824, | |
| "grad_norm": 2.173295497894287, | |
| "learning_rate": 8.588588588588588e-05, | |
| "loss": 0.2607, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 2.3123123123123124, | |
| "grad_norm": 3.244479179382324, | |
| "learning_rate": 8.438438438438439e-05, | |
| "loss": 0.1749, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 2.3423423423423424, | |
| "grad_norm": 3.7047109603881836, | |
| "learning_rate": 8.288288288288289e-05, | |
| "loss": 0.4127, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 2.3723723723723724, | |
| "grad_norm": 3.665949821472168, | |
| "learning_rate": 8.138138138138138e-05, | |
| "loss": 0.1813, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 2.4024024024024024, | |
| "grad_norm": 2.648050546646118, | |
| "learning_rate": 7.987987987987988e-05, | |
| "loss": 0.3, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.4024024024024024, | |
| "eval_accuracy": 0.9202975557917109, | |
| "eval_loss": 0.3156701624393463, | |
| "eval_runtime": 28.5152, | |
| "eval_samples_per_second": 33.0, | |
| "eval_steps_per_second": 4.138, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.4324324324324325, | |
| "grad_norm": 0.29231011867523193, | |
| "learning_rate": 7.837837837837838e-05, | |
| "loss": 0.1998, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 2.4624624624624625, | |
| "grad_norm": 3.4136736392974854, | |
| "learning_rate": 7.687687687687688e-05, | |
| "loss": 0.2671, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 2.4924924924924925, | |
| "grad_norm": 1.5347037315368652, | |
| "learning_rate": 7.537537537537538e-05, | |
| "loss": 0.2087, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 2.5225225225225225, | |
| "grad_norm": 0.2756473124027252, | |
| "learning_rate": 7.387387387387387e-05, | |
| "loss": 0.2465, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 2.5525525525525525, | |
| "grad_norm": 2.2905094623565674, | |
| "learning_rate": 7.237237237237238e-05, | |
| "loss": 0.4043, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 2.5825825825825826, | |
| "grad_norm": 4.817660331726074, | |
| "learning_rate": 7.087087087087088e-05, | |
| "loss": 0.238, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 2.6126126126126126, | |
| "grad_norm": 2.556462526321411, | |
| "learning_rate": 6.936936936936938e-05, | |
| "loss": 0.1516, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 2.6426426426426426, | |
| "grad_norm": 6.198856830596924, | |
| "learning_rate": 6.786786786786787e-05, | |
| "loss": 0.3203, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 2.6726726726726726, | |
| "grad_norm": 0.9303868412971497, | |
| "learning_rate": 6.636636636636637e-05, | |
| "loss": 0.2225, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 2.7027027027027026, | |
| "grad_norm": 3.7441625595092773, | |
| "learning_rate": 6.486486486486487e-05, | |
| "loss": 0.2372, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 2.7027027027027026, | |
| "eval_accuracy": 0.9192348565356004, | |
| "eval_loss": 0.3209655284881592, | |
| "eval_runtime": 29.1587, | |
| "eval_samples_per_second": 32.272, | |
| "eval_steps_per_second": 4.047, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 2.7327327327327327, | |
| "grad_norm": 1.344062089920044, | |
| "learning_rate": 6.336336336336337e-05, | |
| "loss": 0.3659, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 2.7627627627627627, | |
| "grad_norm": 3.962114095687866, | |
| "learning_rate": 6.186186186186186e-05, | |
| "loss": 0.2461, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 2.7927927927927927, | |
| "grad_norm": 1.4669685363769531, | |
| "learning_rate": 6.0360360360360365e-05, | |
| "loss": 0.2636, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 2.8228228228228227, | |
| "grad_norm": 7.011595249176025, | |
| "learning_rate": 5.8858858858858854e-05, | |
| "loss": 0.2226, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 2.8528528528528527, | |
| "grad_norm": 5.4901862144470215, | |
| "learning_rate": 5.7357357357357356e-05, | |
| "loss": 0.1958, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 2.8828828828828827, | |
| "grad_norm": 0.7357378602027893, | |
| "learning_rate": 5.585585585585585e-05, | |
| "loss": 0.2583, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 2.9129129129129128, | |
| "grad_norm": 5.614912509918213, | |
| "learning_rate": 5.435435435435435e-05, | |
| "loss": 0.2466, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 2.942942942942943, | |
| "grad_norm": 2.8368093967437744, | |
| "learning_rate": 5.2852852852852855e-05, | |
| "loss": 0.2056, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 2.972972972972973, | |
| "grad_norm": 1.6713290214538574, | |
| "learning_rate": 5.135135135135135e-05, | |
| "loss": 0.2231, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 3.003003003003003, | |
| "grad_norm": 0.7347887754440308, | |
| "learning_rate": 4.984984984984985e-05, | |
| "loss": 0.2447, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 3.003003003003003, | |
| "eval_accuracy": 0.922422954303932, | |
| "eval_loss": 0.3139547109603882, | |
| "eval_runtime": 29.2161, | |
| "eval_samples_per_second": 32.208, | |
| "eval_steps_per_second": 4.039, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 3.033033033033033, | |
| "grad_norm": 2.947268486022949, | |
| "learning_rate": 4.834834834834835e-05, | |
| "loss": 0.2247, | |
| "step": 1010 | |
| }, | |
| { | |
| "epoch": 3.063063063063063, | |
| "grad_norm": 0.17604035139083862, | |
| "learning_rate": 4.684684684684685e-05, | |
| "loss": 0.125, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 3.093093093093093, | |
| "grad_norm": 3.0279932022094727, | |
| "learning_rate": 4.5345345345345345e-05, | |
| "loss": 0.1933, | |
| "step": 1030 | |
| }, | |
| { | |
| "epoch": 3.123123123123123, | |
| "grad_norm": 0.5538057684898376, | |
| "learning_rate": 4.384384384384385e-05, | |
| "loss": 0.1259, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 3.153153153153153, | |
| "grad_norm": 3.872739315032959, | |
| "learning_rate": 4.234234234234234e-05, | |
| "loss": 0.3184, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 3.1831831831831834, | |
| "grad_norm": 0.44582998752593994, | |
| "learning_rate": 4.0840840840840845e-05, | |
| "loss": 0.2011, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 3.2132132132132134, | |
| "grad_norm": 2.342395782470703, | |
| "learning_rate": 3.933933933933934e-05, | |
| "loss": 0.1837, | |
| "step": 1070 | |
| }, | |
| { | |
| "epoch": 3.2432432432432434, | |
| "grad_norm": 0.7843596339225769, | |
| "learning_rate": 3.783783783783784e-05, | |
| "loss": 0.226, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 3.2732732732732734, | |
| "grad_norm": 2.8491196632385254, | |
| "learning_rate": 3.633633633633634e-05, | |
| "loss": 0.1481, | |
| "step": 1090 | |
| }, | |
| { | |
| "epoch": 3.3033033033033035, | |
| "grad_norm": 1.972225546836853, | |
| "learning_rate": 3.483483483483483e-05, | |
| "loss": 0.2209, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 3.3033033033033035, | |
| "eval_accuracy": 0.9160467587672688, | |
| "eval_loss": 0.30342116951942444, | |
| "eval_runtime": 29.6037, | |
| "eval_samples_per_second": 31.787, | |
| "eval_steps_per_second": 3.986, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 3.3333333333333335, | |
| "grad_norm": 2.8411455154418945, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 0.1262, | |
| "step": 1110 | |
| }, | |
| { | |
| "epoch": 3.3633633633633635, | |
| "grad_norm": 0.25656577944755554, | |
| "learning_rate": 3.183183183183183e-05, | |
| "loss": 0.1239, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 3.3933933933933935, | |
| "grad_norm": 0.13324937224388123, | |
| "learning_rate": 3.0330330330330332e-05, | |
| "loss": 0.1241, | |
| "step": 1130 | |
| }, | |
| { | |
| "epoch": 3.4234234234234235, | |
| "grad_norm": 1.2855831384658813, | |
| "learning_rate": 2.882882882882883e-05, | |
| "loss": 0.1918, | |
| "step": 1140 | |
| }, | |
| { | |
| "epoch": 3.4534534534534536, | |
| "grad_norm": 1.9817436933517456, | |
| "learning_rate": 2.732732732732733e-05, | |
| "loss": 0.2025, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 3.4834834834834836, | |
| "grad_norm": 2.209524631500244, | |
| "learning_rate": 2.582582582582583e-05, | |
| "loss": 0.2032, | |
| "step": 1160 | |
| }, | |
| { | |
| "epoch": 3.5135135135135136, | |
| "grad_norm": 3.2614221572875977, | |
| "learning_rate": 2.4324324324324327e-05, | |
| "loss": 0.1055, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 3.5435435435435436, | |
| "grad_norm": 1.4140874147415161, | |
| "learning_rate": 2.2822822822822822e-05, | |
| "loss": 0.1869, | |
| "step": 1180 | |
| }, | |
| { | |
| "epoch": 3.5735735735735736, | |
| "grad_norm": 4.29329252243042, | |
| "learning_rate": 2.132132132132132e-05, | |
| "loss": 0.1858, | |
| "step": 1190 | |
| }, | |
| { | |
| "epoch": 3.6036036036036037, | |
| "grad_norm": 6.110797882080078, | |
| "learning_rate": 1.981981981981982e-05, | |
| "loss": 0.2641, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 3.6036036036036037, | |
| "eval_accuracy": 0.9277364505844846, | |
| "eval_loss": 0.28959977626800537, | |
| "eval_runtime": 28.964, | |
| "eval_samples_per_second": 32.489, | |
| "eval_steps_per_second": 4.074, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 3.6336336336336337, | |
| "grad_norm": 0.3689054250717163, | |
| "learning_rate": 1.831831831831832e-05, | |
| "loss": 0.1217, | |
| "step": 1210 | |
| }, | |
| { | |
| "epoch": 3.6636636636636637, | |
| "grad_norm": 5.862549781799316, | |
| "learning_rate": 1.6816816816816817e-05, | |
| "loss": 0.1643, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 3.6936936936936937, | |
| "grad_norm": 0.19235260784626007, | |
| "learning_rate": 1.5315315315315316e-05, | |
| "loss": 0.165, | |
| "step": 1230 | |
| }, | |
| { | |
| "epoch": 3.7237237237237237, | |
| "grad_norm": 0.40488940477371216, | |
| "learning_rate": 1.3813813813813815e-05, | |
| "loss": 0.1948, | |
| "step": 1240 | |
| }, | |
| { | |
| "epoch": 3.7537537537537538, | |
| "grad_norm": 3.595454692840576, | |
| "learning_rate": 1.2312312312312313e-05, | |
| "loss": 0.1063, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 3.7837837837837838, | |
| "grad_norm": 3.3588507175445557, | |
| "learning_rate": 1.0810810810810812e-05, | |
| "loss": 0.2252, | |
| "step": 1260 | |
| }, | |
| { | |
| "epoch": 3.813813813813814, | |
| "grad_norm": 5.059345245361328, | |
| "learning_rate": 9.309309309309309e-06, | |
| "loss": 0.1233, | |
| "step": 1270 | |
| }, | |
| { | |
| "epoch": 3.843843843843844, | |
| "grad_norm": 0.11752854287624359, | |
| "learning_rate": 7.807807807807808e-06, | |
| "loss": 0.1526, | |
| "step": 1280 | |
| }, | |
| { | |
| "epoch": 3.873873873873874, | |
| "grad_norm": 4.34885835647583, | |
| "learning_rate": 6.306306306306306e-06, | |
| "loss": 0.1737, | |
| "step": 1290 | |
| }, | |
| { | |
| "epoch": 3.903903903903904, | |
| "grad_norm": 1.4543063640594482, | |
| "learning_rate": 4.804804804804805e-06, | |
| "loss": 0.0954, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 3.903903903903904, | |
| "eval_accuracy": 0.926673751328374, | |
| "eval_loss": 0.2866949737071991, | |
| "eval_runtime": 29.2579, | |
| "eval_samples_per_second": 32.162, | |
| "eval_steps_per_second": 4.033, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 3.933933933933934, | |
| "grad_norm": 1.619908094406128, | |
| "learning_rate": 3.3033033033033035e-06, | |
| "loss": 0.2021, | |
| "step": 1310 | |
| }, | |
| { | |
| "epoch": 3.963963963963964, | |
| "grad_norm": 4.71495246887207, | |
| "learning_rate": 1.801801801801802e-06, | |
| "loss": 0.1503, | |
| "step": 1320 | |
| }, | |
| { | |
| "epoch": 3.993993993993994, | |
| "grad_norm": 4.634551048278809, | |
| "learning_rate": 3.003003003003003e-07, | |
| "loss": 0.1235, | |
| "step": 1330 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "step": 1332, | |
| "total_flos": 1.651775717862015e+18, | |
| "train_loss": 0.46514646391252856, | |
| "train_runtime": 1194.1419, | |
| "train_samples_per_second": 17.847, | |
| "train_steps_per_second": 1.115 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 1332, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 4, | |
| "save_steps": 100, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.651775717862015e+18, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |