| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.0, | |
| "global_step": 8301, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 2.8644741597397903e-05, | |
| "loss": 2.0959, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 2.728948319479581e-05, | |
| "loss": 1.3446, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 2.5934224792193712e-05, | |
| "loss": 1.2113, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 2.4578966389591615e-05, | |
| "loss": 1.1509, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 2.3223707986989517e-05, | |
| "loss": 1.0866, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_exact_match": 79.10122989593188, | |
| "eval_f1": 86.91234473539434, | |
| "step": 2767 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "learning_rate": 2.1868449584387424e-05, | |
| "loss": 0.9377, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.26, | |
| "learning_rate": 2.051319118178533e-05, | |
| "loss": 0.7975, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "learning_rate": 1.9157932779183232e-05, | |
| "loss": 0.814, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.63, | |
| "learning_rate": 1.7802674376581135e-05, | |
| "loss": 0.7997, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 1.81, | |
| "learning_rate": 1.644741597397904e-05, | |
| "loss": 0.8012, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.99, | |
| "learning_rate": 1.5092157571376944e-05, | |
| "loss": 0.791, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_exact_match": 79.78240302743615, | |
| "eval_f1": 87.66969355486262, | |
| "step": 5534 | |
| }, | |
| { | |
| "epoch": 2.17, | |
| "learning_rate": 1.3736899168774846e-05, | |
| "loss": 0.5813, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 2.35, | |
| "learning_rate": 1.238164076617275e-05, | |
| "loss": 0.571, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 2.53, | |
| "learning_rate": 1.1026382363570653e-05, | |
| "loss": 0.5749, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 2.71, | |
| "learning_rate": 9.671123960968558e-06, | |
| "loss": 0.5654, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 2.89, | |
| "learning_rate": 8.315865558366462e-06, | |
| "loss": 0.568, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_exact_match": 80.9271523178808, | |
| "eval_f1": 88.19682357031269, | |
| "step": 8301 | |
| } | |
| ], | |
| "max_steps": 11068, | |
| "num_train_epochs": 4, | |
| "total_flos": 5.204717742535475e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |