| { | |
| "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": 0.5967741935483871, | |
| "eval_loss": 0.19606828689575195, | |
| "eval_runtime": 5.7638, | |
| "eval_samples_per_second": 537.836, | |
| "eval_steps_per_second": 11.277, | |
| "step": 318 | |
| }, | |
| { | |
| "epoch": 1.57, | |
| "learning_rate": 1.4758909853249476e-05, | |
| "loss": 0.312, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8087096774193548, | |
| "eval_loss": 0.10075896233320236, | |
| "eval_runtime": 5.972, | |
| "eval_samples_per_second": 519.086, | |
| "eval_steps_per_second": 10.884, | |
| "step": 636 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8683870967741936, | |
| "eval_loss": 0.06981873512268066, | |
| "eval_runtime": 5.7111, | |
| "eval_samples_per_second": 542.803, | |
| "eval_steps_per_second": 11.381, | |
| "step": 954 | |
| }, | |
| { | |
| "epoch": 3.14, | |
| "learning_rate": 9.517819706498952e-06, | |
| "loss": 0.1152, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8925806451612903, | |
| "eval_loss": 0.05667635425925255, | |
| "eval_runtime": 5.7563, | |
| "eval_samples_per_second": 538.542, | |
| "eval_steps_per_second": 11.292, | |
| "step": 1272 | |
| }, | |
| { | |
| "epoch": 4.72, | |
| "learning_rate": 4.276729559748428e-06, | |
| "loss": 0.0791, | |
| "step": 1500 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 1908, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 6, | |
| "save_steps": 500, | |
| "total_flos": 390310534917408.0, | |
| "train_batch_size": 48, | |
| "trial_name": null, | |
| "trial_params": { | |
| "alpha": 0.7650015549981369, | |
| "num_train_epochs": 6, | |
| "temperature": 19 | |
| } | |
| } | |