| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 6.289308176100629, | |
| "eval_steps": 500, | |
| "global_step": 2000, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.6035483870967742, | |
| "eval_loss": 0.2065293788909912, | |
| "eval_runtime": 5.4973, | |
| "eval_samples_per_second": 563.918, | |
| "eval_steps_per_second": 11.824, | |
| "step": 318 | |
| }, | |
| { | |
| "epoch": 1.57, | |
| "learning_rate": 1.685534591194969e-05, | |
| "loss": 0.3298, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8422580645161291, | |
| "eval_loss": 0.09804486483335495, | |
| "eval_runtime": 5.6345, | |
| "eval_samples_per_second": 550.179, | |
| "eval_steps_per_second": 11.536, | |
| "step": 636 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8964516129032258, | |
| "eval_loss": 0.06418585777282715, | |
| "eval_runtime": 5.5011, | |
| "eval_samples_per_second": 563.528, | |
| "eval_steps_per_second": 11.816, | |
| "step": 954 | |
| }, | |
| { | |
| "epoch": 3.14, | |
| "learning_rate": 1.371069182389937e-05, | |
| "loss": 0.113, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.9045161290322581, | |
| "eval_loss": 0.04777143895626068, | |
| "eval_runtime": 5.6659, | |
| "eval_samples_per_second": 547.133, | |
| "eval_steps_per_second": 11.472, | |
| "step": 1272 | |
| }, | |
| { | |
| "epoch": 4.72, | |
| "learning_rate": 1.0566037735849058e-05, | |
| "loss": 0.0703, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.9151612903225806, | |
| "eval_loss": 0.03928793966770172, | |
| "eval_runtime": 5.7374, | |
| "eval_samples_per_second": 540.317, | |
| "eval_steps_per_second": 11.329, | |
| "step": 1590 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.9193548387096774, | |
| "eval_loss": 0.034144770354032516, | |
| "eval_runtime": 5.8666, | |
| "eval_samples_per_second": 528.418, | |
| "eval_steps_per_second": 11.08, | |
| "step": 1908 | |
| }, | |
| { | |
| "epoch": 6.29, | |
| "learning_rate": 7.421383647798742e-06, | |
| "loss": 0.0547, | |
| "step": 2000 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 3180, | |
| "num_train_epochs": 10, | |
| "save_steps": 500, | |
| "total_flos": 519927215063004.0, | |
| "trial_name": null, | |
| "trial_params": { | |
| "alpha": 0.10298372805208489, | |
| "num_train_epochs": 10, | |
| "temperature": 9 | |
| } | |
| } | |