| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 4.716981132075472, | |
| "eval_steps": 500, | |
| "global_step": 1500, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.6103225806451613, | |
| "eval_loss": 0.2610521614551544, | |
| "eval_runtime": 5.9505, | |
| "eval_samples_per_second": 520.964, | |
| "eval_steps_per_second": 10.923, | |
| "step": 318 | |
| }, | |
| { | |
| "epoch": 1.57, | |
| "learning_rate": 1.371069182389937e-05, | |
| "loss": 0.4067, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8264516129032258, | |
| "eval_loss": 0.12506552040576935, | |
| "eval_runtime": 6.095, | |
| "eval_samples_per_second": 508.612, | |
| "eval_steps_per_second": 10.664, | |
| "step": 636 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8706451612903225, | |
| "eval_loss": 0.08441388607025146, | |
| "eval_runtime": 5.9795, | |
| "eval_samples_per_second": 518.442, | |
| "eval_steps_per_second": 10.871, | |
| "step": 954 | |
| }, | |
| { | |
| "epoch": 3.14, | |
| "learning_rate": 7.421383647798742e-06, | |
| "loss": 0.1449, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8887096774193548, | |
| "eval_loss": 0.06862777471542358, | |
| "eval_runtime": 6.1044, | |
| "eval_samples_per_second": 507.834, | |
| "eval_steps_per_second": 10.648, | |
| "step": 1272 | |
| }, | |
| { | |
| "epoch": 4.72, | |
| "learning_rate": 1.1320754716981133e-06, | |
| "loss": 0.0985, | |
| "step": 1500 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 1590, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "save_steps": 500, | |
| "total_flos": 391111083046572.0, | |
| "train_batch_size": 48, | |
| "trial_name": null, | |
| "trial_params": { | |
| "alpha": 0.21279867704385869, | |
| "num_train_epochs": 5, | |
| "temperature": 4 | |
| } | |
| } | |