| { | |
| "best_global_step": null, | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.9489751417357173, | |
| "eval_steps": 500, | |
| "global_step": 284, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.013955516790231139, | |
| "grad_norm": 4.9564723930657655, | |
| "learning_rate": 0.0001, | |
| "loss": 0.3584, | |
| "mean_token_accuracy": 0.916381947696209, | |
| "num_tokens": 794149.0, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.13955516790231137, | |
| "grad_norm": 0.2761209164773756, | |
| "learning_rate": 0.0001, | |
| "loss": 0.2481, | |
| "mean_token_accuracy": 0.9236801053500838, | |
| "num_tokens": 7995437.0, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.27911033580462274, | |
| "grad_norm": 0.09969812837236781, | |
| "learning_rate": 0.0001, | |
| "loss": 0.1429, | |
| "mean_token_accuracy": 0.9472733726724982, | |
| "num_tokens": 15966760.0, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.4186655037069342, | |
| "grad_norm": 0.06494669674949567, | |
| "learning_rate": 0.0001, | |
| "loss": 0.1206, | |
| "mean_token_accuracy": 0.9548009386286139, | |
| "num_tokens": 23944501.0, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.5582206716092455, | |
| "grad_norm": 0.06828323186623608, | |
| "learning_rate": 0.0001, | |
| "loss": 0.1069, | |
| "mean_token_accuracy": 0.959625343978405, | |
| "num_tokens": 31937884.0, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.6977758395115569, | |
| "grad_norm": 0.06165242715272168, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0979, | |
| "mean_token_accuracy": 0.9630069149658084, | |
| "num_tokens": 39961024.0, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.8373310074138683, | |
| "grad_norm": 0.11090642053667181, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0915, | |
| "mean_token_accuracy": 0.965560769662261, | |
| "num_tokens": 47949799.0, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.9768861753161797, | |
| "grad_norm": 0.06459506102599961, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0873, | |
| "mean_token_accuracy": 0.9674661625176668, | |
| "num_tokens": 55962901.0, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.111644134321849, | |
| "grad_norm": 0.10339694562330957, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0795, | |
| "mean_token_accuracy": 0.9700713963956124, | |
| "num_tokens": 63670548.0, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.2511993022241605, | |
| "grad_norm": 0.0950394988703407, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0758, | |
| "mean_token_accuracy": 0.9714846231043339, | |
| "num_tokens": 71651053.0, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.3907544701264718, | |
| "grad_norm": 0.07782860090437102, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0739, | |
| "mean_token_accuracy": 0.9722085015848279, | |
| "num_tokens": 79640732.0, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.5303096380287833, | |
| "grad_norm": 0.07920993179360719, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0736, | |
| "mean_token_accuracy": 0.9723761970177293, | |
| "num_tokens": 87641211.0, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 1.6698648059310948, | |
| "grad_norm": 0.07411448710660809, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0722, | |
| "mean_token_accuracy": 0.9729392532259226, | |
| "num_tokens": 95648655.0, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.809419973833406, | |
| "grad_norm": 0.05976275661821771, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0721, | |
| "mean_token_accuracy": 0.9728728512302041, | |
| "num_tokens": 103620940.0, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.9489751417357173, | |
| "grad_norm": 0.06809511858490988, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0718, | |
| "mean_token_accuracy": 0.9729661472141743, | |
| "num_tokens": 111618326.0, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 2.083733100741387, | |
| "grad_norm": 0.07797576138180032, | |
| "learning_rate": 0.0001, | |
| "loss": 0.063, | |
| "mean_token_accuracy": 0.9762524333586585, | |
| "num_tokens": 119354987.0, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 2.223288268643698, | |
| "grad_norm": 0.06579714178222891, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0579, | |
| "mean_token_accuracy": 0.9778176054358483, | |
| "num_tokens": 127354165.0, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 2.36284343654601, | |
| "grad_norm": 0.06657618820877312, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0585, | |
| "mean_token_accuracy": 0.9775434693321585, | |
| "num_tokens": 135368633.0, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 2.502398604448321, | |
| "grad_norm": 0.0804632921365247, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0601, | |
| "mean_token_accuracy": 0.9770329371094704, | |
| "num_tokens": 143337210.0, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 2.6419537723506323, | |
| "grad_norm": 0.06313698724150443, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0602, | |
| "mean_token_accuracy": 0.9769743764773011, | |
| "num_tokens": 151337995.0, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 2.7815089402529436, | |
| "grad_norm": 0.06376621347327172, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0611, | |
| "mean_token_accuracy": 0.9767070030793548, | |
| "num_tokens": 159338042.0, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.921064108155255, | |
| "grad_norm": 0.06559121682492687, | |
| "learning_rate": 0.0001, | |
| "loss": 0.062, | |
| "mean_token_accuracy": 0.9763033416122198, | |
| "num_tokens": 167307219.0, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 3.0558220671609244, | |
| "grad_norm": 0.10041231718814958, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0559, | |
| "mean_token_accuracy": 0.9786613514122454, | |
| "num_tokens": 175030018.0, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 3.195377235063236, | |
| "grad_norm": 0.08091195651854872, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0473, | |
| "mean_token_accuracy": 0.9815034126862884, | |
| "num_tokens": 183021967.0, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 3.3349324029655474, | |
| "grad_norm": 0.07575741175199704, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0489, | |
| "mean_token_accuracy": 0.9808894643560052, | |
| "num_tokens": 191014779.0, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 3.4744875708678586, | |
| "grad_norm": 0.07829702223737112, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0498, | |
| "mean_token_accuracy": 0.9806048579514026, | |
| "num_tokens": 199025571.0, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 3.61404273877017, | |
| "grad_norm": 0.06942538891461214, | |
| "learning_rate": 0.0001, | |
| "loss": 0.051, | |
| "mean_token_accuracy": 0.9801934180781245, | |
| "num_tokens": 207011679.0, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 3.7535979066724816, | |
| "grad_norm": 0.07675336174060962, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0515, | |
| "mean_token_accuracy": 0.9799842674285173, | |
| "num_tokens": 215009066.0, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 3.893153074574793, | |
| "grad_norm": 0.07733704641759509, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0527, | |
| "mean_token_accuracy": 0.9795725893229246, | |
| "num_tokens": 223003497.0, | |
| "step": 280 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 284, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 4, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 507232006963200.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |