| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.0, | |
| "eval_steps": 30, | |
| "global_step": 450, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.022222222222222223, | |
| "grad_norm": 0.3012203276157379, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 0.3486, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.044444444444444446, | |
| "grad_norm": 0.19629184901714325, | |
| "learning_rate": 9.951340343707852e-05, | |
| "loss": 0.0694, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.06666666666666667, | |
| "grad_norm": 0.1222372055053711, | |
| "learning_rate": 9.890738003669029e-05, | |
| "loss": 0.0654, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.06666666666666667, | |
| "eval_loss": 0.06320454925298691, | |
| "eval_runtime": 136.4924, | |
| "eval_samples_per_second": 7.326, | |
| "eval_steps_per_second": 0.366, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.08888888888888889, | |
| "grad_norm": 0.12505225837230682, | |
| "learning_rate": 9.806308479691595e-05, | |
| "loss": 0.0637, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.1111111111111111, | |
| "grad_norm": 0.14867539703845978, | |
| "learning_rate": 9.698463103929542e-05, | |
| "loss": 0.0638, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.13333333333333333, | |
| "grad_norm": 0.10100489109754562, | |
| "learning_rate": 9.567727288213005e-05, | |
| "loss": 0.0595, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.13333333333333333, | |
| "eval_loss": 0.05767079070210457, | |
| "eval_runtime": 136.642, | |
| "eval_samples_per_second": 7.318, | |
| "eval_steps_per_second": 0.366, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.15555555555555556, | |
| "grad_norm": 0.17016883194446564, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 0.0569, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.17777777777777778, | |
| "grad_norm": 0.08618276566267014, | |
| "learning_rate": 9.24024048078213e-05, | |
| "loss": 0.0567, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 0.08416091650724411, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 0.0546, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "eval_loss": 0.053434912115335464, | |
| "eval_runtime": 136.6891, | |
| "eval_samples_per_second": 7.316, | |
| "eval_steps_per_second": 0.366, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.2222222222222222, | |
| "grad_norm": 0.09946251660585403, | |
| "learning_rate": 8.83022221559489e-05, | |
| "loss": 0.0558, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.24444444444444444, | |
| "grad_norm": 0.10791021585464478, | |
| "learning_rate": 8.596699001693255e-05, | |
| "loss": 0.0552, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.26666666666666666, | |
| "grad_norm": 0.10341854393482208, | |
| "learning_rate": 8.345653031794292e-05, | |
| "loss": 0.0555, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.26666666666666666, | |
| "eval_loss": 0.052995871752500534, | |
| "eval_runtime": 136.9161, | |
| "eval_samples_per_second": 7.304, | |
| "eval_steps_per_second": 0.365, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.28888888888888886, | |
| "grad_norm": 0.092866912484169, | |
| "learning_rate": 8.07830737662829e-05, | |
| "loss": 0.0504, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.3111111111111111, | |
| "grad_norm": 0.10099858045578003, | |
| "learning_rate": 7.795964517353735e-05, | |
| "loss": 0.0526, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.3333333333333333, | |
| "grad_norm": 0.17789088189601898, | |
| "learning_rate": 7.500000000000001e-05, | |
| "loss": 0.052, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.3333333333333333, | |
| "eval_loss": 0.052405789494514465, | |
| "eval_runtime": 136.4971, | |
| "eval_samples_per_second": 7.326, | |
| "eval_steps_per_second": 0.366, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.35555555555555557, | |
| "grad_norm": 0.13198032975196838, | |
| "learning_rate": 7.191855733945387e-05, | |
| "loss": 0.0525, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.37777777777777777, | |
| "grad_norm": 0.10188263654708862, | |
| "learning_rate": 6.873032967079561e-05, | |
| "loss": 0.0524, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 0.1550171822309494, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 0.0529, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "eval_loss": 0.05086047202348709, | |
| "eval_runtime": 136.9202, | |
| "eval_samples_per_second": 7.304, | |
| "eval_steps_per_second": 0.365, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.4222222222222222, | |
| "grad_norm": 0.08379051834344864, | |
| "learning_rate": 6.209609477998338e-05, | |
| "loss": 0.053, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.4444444444444444, | |
| "grad_norm": 0.1344887912273407, | |
| "learning_rate": 5.868240888334653e-05, | |
| "loss": 0.0511, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.4666666666666667, | |
| "grad_norm": 0.09979716688394547, | |
| "learning_rate": 5.522642316338268e-05, | |
| "loss": 0.0502, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.4666666666666667, | |
| "eval_loss": 0.0506291501224041, | |
| "eval_runtime": 136.7115, | |
| "eval_samples_per_second": 7.315, | |
| "eval_steps_per_second": 0.366, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.4888888888888889, | |
| "grad_norm": 0.07537568360567093, | |
| "learning_rate": 5.174497483512506e-05, | |
| "loss": 0.0497, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.5111111111111111, | |
| "grad_norm": 0.10576926916837692, | |
| "learning_rate": 4.825502516487497e-05, | |
| "loss": 0.0504, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.5333333333333333, | |
| "grad_norm": 0.10534786432981491, | |
| "learning_rate": 4.477357683661734e-05, | |
| "loss": 0.0479, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.5333333333333333, | |
| "eval_loss": 0.048654135316610336, | |
| "eval_runtime": 136.5401, | |
| "eval_samples_per_second": 7.324, | |
| "eval_steps_per_second": 0.366, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.5555555555555556, | |
| "grad_norm": 0.10896391421556473, | |
| "learning_rate": 4.131759111665349e-05, | |
| "loss": 0.0506, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.5777777777777777, | |
| "grad_norm": 0.10746324062347412, | |
| "learning_rate": 3.790390522001662e-05, | |
| "loss": 0.0477, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "grad_norm": 0.16898374259471893, | |
| "learning_rate": 3.4549150281252636e-05, | |
| "loss": 0.0479, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "eval_loss": 0.04828697443008423, | |
| "eval_runtime": 136.3988, | |
| "eval_samples_per_second": 7.331, | |
| "eval_steps_per_second": 0.367, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.6222222222222222, | |
| "grad_norm": 0.23446787893772125, | |
| "learning_rate": 3.12696703292044e-05, | |
| "loss": 0.0487, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.6444444444444445, | |
| "grad_norm": 0.18817700445652008, | |
| "learning_rate": 2.8081442660546125e-05, | |
| "loss": 0.0483, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.6666666666666666, | |
| "grad_norm": 0.1385546326637268, | |
| "learning_rate": 2.500000000000001e-05, | |
| "loss": 0.047, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.6666666666666666, | |
| "eval_loss": 0.04631952568888664, | |
| "eval_runtime": 136.5399, | |
| "eval_samples_per_second": 7.324, | |
| "eval_steps_per_second": 0.366, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.6888888888888889, | |
| "grad_norm": 0.1836208701133728, | |
| "learning_rate": 2.2040354826462668e-05, | |
| "loss": 0.0477, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.7111111111111111, | |
| "grad_norm": 0.17699767649173737, | |
| "learning_rate": 1.9216926233717085e-05, | |
| "loss": 0.0455, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.7333333333333333, | |
| "grad_norm": 0.17500163614749908, | |
| "learning_rate": 1.6543469682057106e-05, | |
| "loss": 0.0452, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.7333333333333333, | |
| "eval_loss": 0.0455147922039032, | |
| "eval_runtime": 136.6527, | |
| "eval_samples_per_second": 7.318, | |
| "eval_steps_per_second": 0.366, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.7555555555555555, | |
| "grad_norm": 0.4250011146068573, | |
| "learning_rate": 1.4033009983067452e-05, | |
| "loss": 0.045, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.7777777777777778, | |
| "grad_norm": 0.22023910284042358, | |
| "learning_rate": 1.1697777844051105e-05, | |
| "loss": 0.0436, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 0.1621396541595459, | |
| "learning_rate": 9.549150281252633e-06, | |
| "loss": 0.0437, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "eval_loss": 0.04442232847213745, | |
| "eval_runtime": 136.5105, | |
| "eval_samples_per_second": 7.325, | |
| "eval_steps_per_second": 0.366, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.8222222222222222, | |
| "grad_norm": 0.1345316618680954, | |
| "learning_rate": 7.597595192178702e-06, | |
| "loss": 0.043, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.8444444444444444, | |
| "grad_norm": 0.1296633929014206, | |
| "learning_rate": 5.852620357053651e-06, | |
| "loss": 0.0451, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.8666666666666667, | |
| "grad_norm": 0.1679375320672989, | |
| "learning_rate": 4.322727117869951e-06, | |
| "loss": 0.0442, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.8666666666666667, | |
| "eval_loss": 0.04389314725995064, | |
| "eval_runtime": 136.6458, | |
| "eval_samples_per_second": 7.318, | |
| "eval_steps_per_second": 0.366, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.8888888888888888, | |
| "grad_norm": 0.1358930468559265, | |
| "learning_rate": 3.0153689607045845e-06, | |
| "loss": 0.042, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.9111111111111111, | |
| "grad_norm": 0.12815573811531067, | |
| "learning_rate": 1.9369152030840556e-06, | |
| "loss": 0.0435, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 0.9333333333333333, | |
| "grad_norm": 0.16702179610729218, | |
| "learning_rate": 1.0926199633097157e-06, | |
| "loss": 0.0439, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.9333333333333333, | |
| "eval_loss": 0.04338783398270607, | |
| "eval_runtime": 136.4534, | |
| "eval_samples_per_second": 7.329, | |
| "eval_steps_per_second": 0.366, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.9555555555555556, | |
| "grad_norm": 0.2307462841272354, | |
| "learning_rate": 4.865965629214819e-07, | |
| "loss": 0.0406, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 0.9777777777777777, | |
| "grad_norm": 0.1788487583398819, | |
| "learning_rate": 1.2179748700879012e-07, | |
| "loss": 0.0434, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 0.22977402806282043, | |
| "learning_rate": 0.0, | |
| "loss": 0.0444, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.04335245117545128, | |
| "eval_runtime": 136.5818, | |
| "eval_samples_per_second": 7.322, | |
| "eval_steps_per_second": 0.366, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "step": 450, | |
| "total_flos": 8.965157752615731e+17, | |
| "train_loss": 0.0571367746591568, | |
| "train_runtime": 6071.8085, | |
| "train_samples_per_second": 1.482, | |
| "train_steps_per_second": 0.074 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 450, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 2000, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 8.965157752615731e+17, | |
| "train_batch_size": 10, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |