| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.0, | |
| "eval_steps": 500, | |
| "global_step": 14769, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 0.00013324306025727826, | |
| "loss": 2.3278, | |
| "step": 984 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 0.00019261209750225702, | |
| "loss": 2.1512, | |
| "step": 1968 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 0.00017780619921757448, | |
| "loss": 2.0221, | |
| "step": 2952 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 0.00016300030093289197, | |
| "loss": 1.979, | |
| "step": 3936 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 0.00014819440264820946, | |
| "loss": 2.0714, | |
| "step": 4920 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 0.00013338850436352695, | |
| "loss": 2.0495, | |
| "step": 5904 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "learning_rate": 0.00011858260607884442, | |
| "loss": 2.0241, | |
| "step": 6888 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 0.00010377670779416191, | |
| "loss": 2.0703, | |
| "step": 7872 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "learning_rate": 8.89708095094794e-05, | |
| "loss": 2.0445, | |
| "step": 8856 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 7.416491122479687e-05, | |
| "loss": 1.995, | |
| "step": 9840 | |
| }, | |
| { | |
| "epoch": 2.2, | |
| "learning_rate": 5.935901294011436e-05, | |
| "loss": 2.0327, | |
| "step": 10824 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "learning_rate": 4.455311465543184e-05, | |
| "loss": 2.1159, | |
| "step": 11808 | |
| }, | |
| { | |
| "epoch": 2.6, | |
| "learning_rate": 2.9747216370749325e-05, | |
| "loss": 2.1016, | |
| "step": 12792 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "learning_rate": 1.4941318086066807e-05, | |
| "loss": 2.0736, | |
| "step": 13776 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 1.354198013842913e-07, | |
| "loss": 2.0239, | |
| "step": 14760 | |
| } | |
| ], | |
| "logging_steps": 984, | |
| "max_steps": 14769, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "total_flos": 1.1759674170802176e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |