| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.9690721649484537, | |
| "eval_steps": 500, | |
| "global_step": 72, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.03, | |
| "grad_norm": 1.7972885656308806, | |
| "learning_rate": 2.203125e-05, | |
| "loss": 1.0025, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 2.06, | |
| "grad_norm": 1.887297196674392, | |
| "learning_rate": 1.03125e-05, | |
| "loss": 0.7677, | |
| "step": 50 | |
| } | |
| ], | |
| "logging_steps": 25, | |
| "max_steps": 72, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "total_flos": 30098325504000.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |