| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.1446540880503147, | |
| "eval_steps": 500, | |
| "global_step": 1000, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.6451612903225806, | |
| "eval_loss": 0.2875964641571045, | |
| "eval_runtime": 1.4249, | |
| "eval_samples_per_second": 2175.625, | |
| "eval_steps_per_second": 45.618, | |
| "step": 318 | |
| }, | |
| { | |
| "epoch": 1.57, | |
| "learning_rate": 1.685534591194969e-05, | |
| "loss": 0.4599, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8493548387096774, | |
| "eval_loss": 0.11622390896081924, | |
| "eval_runtime": 1.4485, | |
| "eval_samples_per_second": 2140.189, | |
| "eval_steps_per_second": 44.875, | |
| "step": 636 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8948387096774194, | |
| "eval_loss": 0.06665954738855362, | |
| "eval_runtime": 1.4577, | |
| "eval_samples_per_second": 2126.625, | |
| "eval_steps_per_second": 44.591, | |
| "step": 954 | |
| }, | |
| { | |
| "epoch": 3.14, | |
| "learning_rate": 1.371069182389937e-05, | |
| "loss": 0.1359, | |
| "step": 1000 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 3180, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 500, | |
| "total_flos": 260572968372528.0, | |
| "train_batch_size": 48, | |
| "trial_name": null, | |
| "trial_params": { | |
| "alpha": 0.6133894128091588, | |
| "num_train_epochs": 10, | |
| "temperature": 3 | |
| } | |
| } | |