| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.0, | |
| "eval_steps": 500, | |
| "global_step": 7812, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 1.975703324808184e-05, | |
| "loss": 10.4239, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 2.8920341394025606e-05, | |
| "loss": 7.9837, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 2.6701280227596018e-05, | |
| "loss": 7.7629, | |
| "step": 1560 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 2.4482219061166427e-05, | |
| "loss": 7.7335, | |
| "step": 2080 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 2.2263157894736843e-05, | |
| "loss": 7.723, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 2.0044096728307255e-05, | |
| "loss": 7.7143, | |
| "step": 3120 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "learning_rate": 1.7825035561877668e-05, | |
| "loss": 7.7118, | |
| "step": 3640 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 1.560597439544808e-05, | |
| "loss": 7.7059, | |
| "step": 4160 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "learning_rate": 1.3386913229018493e-05, | |
| "loss": 7.7045, | |
| "step": 4680 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 1.1167852062588905e-05, | |
| "loss": 7.7006, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 2.2, | |
| "learning_rate": 8.948790896159317e-06, | |
| "loss": 7.6989, | |
| "step": 5720 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "learning_rate": 6.729729729729731e-06, | |
| "loss": 7.7004, | |
| "step": 6240 | |
| }, | |
| { | |
| "epoch": 2.6, | |
| "learning_rate": 4.510668563300143e-06, | |
| "loss": 7.6977, | |
| "step": 6760 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "learning_rate": 2.2916073968705546e-06, | |
| "loss": 7.6962, | |
| "step": 7280 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 7.254623044096729e-08, | |
| "loss": 7.6984, | |
| "step": 7800 | |
| } | |
| ], | |
| "logging_steps": 520, | |
| "max_steps": 7812, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "total_flos": 1.9499709634707456e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |