| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.0, | |
| "eval_steps": 10000, | |
| "global_step": 7276, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 9.359487165438073e-05, | |
| "loss": 1.3028, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 6.618156843109719e-05, | |
| "loss": 0.9004, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 5.403702434442518e-05, | |
| "loss": 0.7984, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 1.0141562223434448, | |
| "eval_runtime": 3.363, | |
| "eval_samples_per_second": 264.642, | |
| "eval_steps_per_second": 33.303, | |
| "step": 3638 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "learning_rate": 4.679743582719036e-05, | |
| "loss": 0.6662, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.37, | |
| "learning_rate": 4.1856899072912704e-05, | |
| "loss": 0.4999, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.65, | |
| "learning_rate": 3.820994634908561e-05, | |
| "loss": 0.4938, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "learning_rate": 3.537553634121427e-05, | |
| "loss": 0.4857, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 1.0224509239196777, | |
| "eval_runtime": 3.3432, | |
| "eval_samples_per_second": 266.209, | |
| "eval_steps_per_second": 33.5, | |
| "step": 7276 | |
| } | |
| ], | |
| "logging_steps": 1000, | |
| "max_steps": 7276, | |
| "num_train_epochs": 2, | |
| "save_steps": 500, | |
| "total_flos": 2.5033888640414515e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |