| { | |
| "best_global_step": null, | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.0, | |
| "eval_steps": 500, | |
| "global_step": 830, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.060350030175015085, | |
| "grad_norm": 0.17640693485736847, | |
| "learning_rate": 9.599999999999999e-05, | |
| "loss": 1.7813, | |
| "mean_token_accuracy": 0.6300852990150452, | |
| "num_tokens": 156656.0, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.12070006035003017, | |
| "grad_norm": 0.24193964898586273, | |
| "learning_rate": 0.00019599999999999997, | |
| "loss": 0.9197, | |
| "mean_token_accuracy": 0.768382026553154, | |
| "num_tokens": 282982.0, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.18105009052504525, | |
| "grad_norm": 0.16499435901641846, | |
| "learning_rate": 0.000296, | |
| "loss": 0.5906, | |
| "mean_token_accuracy": 0.8341364151239395, | |
| "num_tokens": 441181.0, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.24140012070006034, | |
| "grad_norm": 0.23563049733638763, | |
| "learning_rate": 0.0002999269005776963, | |
| "loss": 0.4832, | |
| "mean_token_accuracy": 0.8592967188358307, | |
| "num_tokens": 567644.0, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.30175015087507545, | |
| "grad_norm": 0.22556591033935547, | |
| "learning_rate": 0.0002996953705789175, | |
| "loss": 0.3612, | |
| "mean_token_accuracy": 0.8925437909364701, | |
| "num_tokens": 725987.0, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.3621001810500905, | |
| "grad_norm": 0.33429527282714844, | |
| "learning_rate": 0.00029930552794275785, | |
| "loss": 0.3126, | |
| "mean_token_accuracy": 0.9086851555109025, | |
| "num_tokens": 853185.0, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.4224502112251056, | |
| "grad_norm": 0.27340370416641235, | |
| "learning_rate": 0.0002987577849532824, | |
| "loss": 0.2343, | |
| "mean_token_accuracy": 0.9301495373249054, | |
| "num_tokens": 1011232.0, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.4828002414001207, | |
| "grad_norm": 0.2711191475391388, | |
| "learning_rate": 0.00029805272088449905, | |
| "loss": 0.2021, | |
| "mean_token_accuracy": 0.9406860828399658, | |
| "num_tokens": 1138074.0, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.5431502715751357, | |
| "grad_norm": 0.19240038096904755, | |
| "learning_rate": 0.00029719108138773827, | |
| "loss": 0.1508, | |
| "mean_token_accuracy": 0.9550948125123978, | |
| "num_tokens": 1293601.0, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 0.6035003017501509, | |
| "grad_norm": 0.27221739292144775, | |
| "learning_rate": 0.00029617377770307837, | |
| "loss": 0.1563, | |
| "mean_token_accuracy": 0.9542003554105759, | |
| "num_tokens": 1418074.0, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.663850331925166, | |
| "grad_norm": 0.25977134704589844, | |
| "learning_rate": 0.0002950018856956494, | |
| "loss": 0.1228, | |
| "mean_token_accuracy": 0.9640595990419388, | |
| "num_tokens": 1577856.0, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 0.724200362100181, | |
| "grad_norm": 0.2311161458492279, | |
| "learning_rate": 0.0002936766447178356, | |
| "loss": 0.1229, | |
| "mean_token_accuracy": 0.9646531140804291, | |
| "num_tokens": 1704393.0, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.7845503922751962, | |
| "grad_norm": 0.1375264674425125, | |
| "learning_rate": 0.0002921994562985788, | |
| "loss": 0.0972, | |
| "mean_token_accuracy": 0.9722524845600128, | |
| "num_tokens": 1860935.0, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 0.8449004224502112, | |
| "grad_norm": 0.2860631048679352, | |
| "learning_rate": 0.0002905718826611708, | |
| "loss": 0.0853, | |
| "mean_token_accuracy": 0.9756521546840667, | |
| "num_tokens": 1988266.0, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.9052504526252263, | |
| "grad_norm": 0.11123040318489075, | |
| "learning_rate": 0.00028879564507109946, | |
| "loss": 0.0814, | |
| "mean_token_accuracy": 0.9769123244285584, | |
| "num_tokens": 2146122.0, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 0.9656004828002414, | |
| "grad_norm": 0.239139586687088, | |
| "learning_rate": 0.0002868726220156981, | |
| "loss": 0.0696, | |
| "mean_token_accuracy": 0.9802538651227951, | |
| "num_tokens": 2273996.0, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.07211296260356903, | |
| "eval_mean_token_accuracy": 0.979879263285044, | |
| "eval_num_tokens": 2354180.0, | |
| "eval_runtime": 29.4819, | |
| "eval_samples_per_second": 12.516, | |
| "eval_steps_per_second": 6.275, | |
| "step": 415 | |
| }, | |
| { | |
| "epoch": 1.024140012070006, | |
| "grad_norm": 0.13852086663246155, | |
| "learning_rate": 0.0002848048472175225, | |
| "loss": 0.0764, | |
| "mean_token_accuracy": 0.9782544571099822, | |
| "num_tokens": 2422077.0, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 1.0844900422450212, | |
| "grad_norm": 0.17753440141677856, | |
| "learning_rate": 0.00028259450748355637, | |
| "loss": 0.0527, | |
| "mean_token_accuracy": 0.9847306323051452, | |
| "num_tokens": 2564180.0, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 1.1448400724200363, | |
| "grad_norm": 0.13473442196846008, | |
| "learning_rate": 0.00028024394039252005, | |
| "loss": 0.0697, | |
| "mean_token_accuracy": 0.9803690612316132, | |
| "num_tokens": 2705737.0, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 1.2051901025950513, | |
| "grad_norm": 0.04998031258583069, | |
| "learning_rate": 0.0002777556318227281, | |
| "loss": 0.0452, | |
| "mean_token_accuracy": 0.987179564833641, | |
| "num_tokens": 2848782.0, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.2655401327700664, | |
| "grad_norm": 0.09412259608507156, | |
| "learning_rate": 0.00027513221332311073, | |
| "loss": 0.0615, | |
| "mean_token_accuracy": 0.9825534737110138, | |
| "num_tokens": 2991024.0, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 1.3258901629450814, | |
| "grad_norm": 0.12740211188793182, | |
| "learning_rate": 0.0002723764593301788, | |
| "loss": 0.0452, | |
| "mean_token_accuracy": 0.9870287185907364, | |
| "num_tokens": 3132622.0, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 1.3862401931200965, | |
| "grad_norm": 0.11777028441429138, | |
| "learning_rate": 0.0002694912842338756, | |
| "loss": 0.0568, | |
| "mean_token_accuracy": 0.9838440799713135, | |
| "num_tokens": 3273143.0, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 1.4465902232951118, | |
| "grad_norm": 0.09758122265338898, | |
| "learning_rate": 0.0002664797392954194, | |
| "loss": 0.0444, | |
| "mean_token_accuracy": 0.986908946633339, | |
| "num_tokens": 3416381.0, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.5069402534701268, | |
| "grad_norm": 0.0658789575099945, | |
| "learning_rate": 0.0002633450094203953, | |
| "loss": 0.0535, | |
| "mean_token_accuracy": 0.9848115313053131, | |
| "num_tokens": 3558249.0, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 1.567290283645142, | |
| "grad_norm": 0.07210762798786163, | |
| "learning_rate": 0.000260090409790509, | |
| "loss": 0.0409, | |
| "mean_token_accuracy": 0.9878959685564042, | |
| "num_tokens": 3700350.0, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 1.627640313820157, | |
| "grad_norm": 0.09006072580814362, | |
| "learning_rate": 0.000256719382357566, | |
| "loss": 0.052, | |
| "mean_token_accuracy": 0.9852559435367584, | |
| "num_tokens": 3842983.0, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 1.687990343995172, | |
| "grad_norm": 0.07121206820011139, | |
| "learning_rate": 0.0002532354922033823, | |
| "loss": 0.0401, | |
| "mean_token_accuracy": 0.9882262688875199, | |
| "num_tokens": 3985789.0, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.748340374170187, | |
| "grad_norm": 0.049150578677654266, | |
| "learning_rate": 0.00024964242376947747, | |
| "loss": 0.0514, | |
| "mean_token_accuracy": 0.9852595126628876, | |
| "num_tokens": 4128158.0, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 1.8086904043452021, | |
| "grad_norm": 0.08217272907495499, | |
| "learning_rate": 0.000245943976960537, | |
| "loss": 0.0373, | |
| "mean_token_accuracy": 0.9887230151891708, | |
| "num_tokens": 4270705.0, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 1.8690404345202172, | |
| "grad_norm": 0.060381677001714706, | |
| "learning_rate": 0.00024214406312576472, | |
| "loss": 0.051, | |
| "mean_token_accuracy": 0.9850554609298706, | |
| "num_tokens": 4412064.0, | |
| "step": 775 | |
| }, | |
| { | |
| "epoch": 1.9293904646952322, | |
| "grad_norm": 0.07599000632762909, | |
| "learning_rate": 0.00023824670092237557, | |
| "loss": 0.0385, | |
| "mean_token_accuracy": 0.9883499753475189, | |
| "num_tokens": 4554646.0, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.9897404948702473, | |
| "grad_norm": 0.06878010928630829, | |
| "learning_rate": 0.00023425601206560257, | |
| "loss": 0.0432, | |
| "mean_token_accuracy": 0.9873087042570114, | |
| "num_tokens": 4688134.0, | |
| "step": 825 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.04542902857065201, | |
| "eval_mean_token_accuracy": 0.9870176924241556, | |
| "eval_num_tokens": 4708360.0, | |
| "eval_runtime": 29.4629, | |
| "eval_samples_per_second": 12.524, | |
| "eval_steps_per_second": 6.279, | |
| "step": 830 | |
| } | |
| ], | |
| "logging_steps": 25, | |
| "max_steps": 2490, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 6, | |
| "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": 3.996397729951826e+17, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |