| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.0107238605898123, | |
| "eval_steps": 500, | |
| "global_step": 1500, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.6702412868632708, | |
| "grad_norm": 9.020133018493652, | |
| "learning_rate": 1.4299394279479472e-06, | |
| "loss": 0.6244, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.7267851233482361, | |
| "eval_loss": 0.5142917633056641, | |
| "eval_runtime": 9.4466, | |
| "eval_samples_per_second": 315.773, | |
| "eval_steps_per_second": 19.795, | |
| "step": 746 | |
| }, | |
| { | |
| "epoch": 1.3404825737265416, | |
| "grad_norm": 10.139765739440918, | |
| "learning_rate": 1.3272138943309972e-06, | |
| "loss": 0.5438, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.7405296564102173, | |
| "eval_loss": 0.49428224563598633, | |
| "eval_runtime": 9.4794, | |
| "eval_samples_per_second": 314.682, | |
| "eval_steps_per_second": 19.727, | |
| "step": 1492 | |
| }, | |
| { | |
| "epoch": 2.0107238605898123, | |
| "grad_norm": 7.425673961639404, | |
| "learning_rate": 1.2244883607140467e-06, | |
| "loss": 0.5108, | |
| "step": 1500 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 7460, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 500, | |
| "total_flos": 3914742842009676.0, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": { | |
| "learning_rate": 1.5326649615648975e-06, | |
| "per_device_train_batch_size": 16 | |
| } | |
| } | |