| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.769230769230769, | |
| "eval_steps": 500, | |
| "global_step": 18, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.15, | |
| "grad_norm": 8.216391327503926, | |
| "learning_rate": 1e-05, | |
| "loss": 1.1406, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "grad_norm": 15.166642977674293, | |
| "learning_rate": 1.8314696123025456e-05, | |
| "loss": 1.424, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "eval_loss": 0.996313214302063, | |
| "eval_runtime": 60.6333, | |
| "eval_samples_per_second": 11.99, | |
| "eval_steps_per_second": 0.759, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 1.54, | |
| "grad_norm": 4.105427910884771, | |
| "learning_rate": 1e-05, | |
| "loss": 1.2201, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.9208275675773621, | |
| "eval_runtime": 60.4501, | |
| "eval_samples_per_second": 12.026, | |
| "eval_steps_per_second": 0.761, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 2.31, | |
| "grad_norm": 3.0392951701447033, | |
| "learning_rate": 1.6853038769745466e-06, | |
| "loss": 1.0574, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 2.77, | |
| "eval_loss": 0.9125745892524719, | |
| "eval_runtime": 60.4665, | |
| "eval_samples_per_second": 12.023, | |
| "eval_steps_per_second": 0.761, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 2.77, | |
| "step": 18, | |
| "total_flos": 3716488888320.0, | |
| "train_loss": 1.177431093321906, | |
| "train_runtime": 489.2099, | |
| "train_samples_per_second": 2.471, | |
| "train_steps_per_second": 0.037 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 18, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "total_flos": 3716488888320.0, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |