| { | |
| "best_metric": 0.044754352420568466, | |
| "best_model_checkpoint": "/tmp/model/checkpoint-1026", | |
| "epoch": 3.0, | |
| "eval_steps": 500, | |
| "global_step": 1026, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 3.155339805825243e-05, | |
| "loss": 1.2096, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 4.837486457204768e-05, | |
| "loss": 0.2349, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 4.469122426868906e-05, | |
| "loss": 0.3095, | |
| "step": 204 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 4.1007583965330447e-05, | |
| "loss": 0.2824, | |
| "step": 272 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "learning_rate": 3.7323943661971835e-05, | |
| "loss": 0.2215, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.9751098096632503, | |
| "eval_auc": 0.9967504189228713, | |
| "eval_f1": 0.9735202492211837, | |
| "eval_loss": 0.07951382547616959, | |
| "eval_precision": 0.962278675904542, | |
| "eval_recall": 0.9850275807722616, | |
| "eval_runtime": 50.4793, | |
| "eval_samples_per_second": 54.121, | |
| "eval_steps_per_second": 3.388, | |
| "step": 342 | |
| }, | |
| { | |
| "epoch": 1.19, | |
| "learning_rate": 3.3640303358613216e-05, | |
| "loss": 0.1574, | |
| "step": 408 | |
| }, | |
| { | |
| "epoch": 1.39, | |
| "learning_rate": 2.9956663055254608e-05, | |
| "loss": 0.1602, | |
| "step": 476 | |
| }, | |
| { | |
| "epoch": 1.59, | |
| "learning_rate": 2.6273022751895993e-05, | |
| "loss": 0.1773, | |
| "step": 544 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "learning_rate": 2.258938244853738e-05, | |
| "loss": 0.1794, | |
| "step": 612 | |
| }, | |
| { | |
| "epoch": 1.99, | |
| "learning_rate": 1.8905742145178766e-05, | |
| "loss": 0.1738, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.9838945827232797, | |
| "eval_auc": 0.9987853795244613, | |
| "eval_f1": 0.9824281150159744, | |
| "eval_loss": 0.04984632134437561, | |
| "eval_precision": 0.9959514170040485, | |
| "eval_recall": 0.9692671394799054, | |
| "eval_runtime": 52.4203, | |
| "eval_samples_per_second": 52.117, | |
| "eval_steps_per_second": 3.262, | |
| "step": 684 | |
| }, | |
| { | |
| "epoch": 2.19, | |
| "learning_rate": 1.5222101841820153e-05, | |
| "loss": 0.0957, | |
| "step": 748 | |
| }, | |
| { | |
| "epoch": 2.39, | |
| "learning_rate": 1.153846153846154e-05, | |
| "loss": 0.1417, | |
| "step": 816 | |
| }, | |
| { | |
| "epoch": 2.58, | |
| "learning_rate": 7.854821235102926e-06, | |
| "loss": 0.0833, | |
| "step": 884 | |
| }, | |
| { | |
| "epoch": 2.78, | |
| "learning_rate": 4.171180931744313e-06, | |
| "loss": 0.0681, | |
| "step": 952 | |
| }, | |
| { | |
| "epoch": 2.98, | |
| "learning_rate": 4.875406283856988e-07, | |
| "loss": 0.103, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.9864568081991215, | |
| "eval_auc": 0.9994751008188858, | |
| "eval_f1": 0.9852413242919825, | |
| "eval_loss": 0.044754352420568466, | |
| "eval_precision": 0.9975767366720517, | |
| "eval_recall": 0.9732072498029944, | |
| "eval_runtime": 52.4079, | |
| "eval_samples_per_second": 52.13, | |
| "eval_steps_per_second": 3.263, | |
| "step": 1026 | |
| } | |
| ], | |
| "logging_steps": 68, | |
| "max_steps": 1026, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "total_flos": 6.35124346881196e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |