| { | |
| "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": 0.5812903225806452, | |
| "eval_loss": 0.19345349073410034, | |
| "eval_runtime": 5.6622, | |
| "eval_samples_per_second": 547.491, | |
| "eval_steps_per_second": 11.48, | |
| "step": 318 | |
| }, | |
| { | |
| "epoch": 1.5723270440251573, | |
| "grad_norm": 0.5203747153282166, | |
| "learning_rate": 1.650593990216632e-05, | |
| "loss": 0.3117, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8167741935483871, | |
| "eval_loss": 0.09445264935493469, | |
| "eval_runtime": 6.0175, | |
| "eval_samples_per_second": 515.163, | |
| "eval_steps_per_second": 10.802, | |
| "step": 636 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8764516129032258, | |
| "eval_loss": 0.06364042311906815, | |
| "eval_runtime": 5.673, | |
| "eval_samples_per_second": 546.447, | |
| "eval_steps_per_second": 11.458, | |
| "step": 954 | |
| }, | |
| { | |
| "epoch": 3.1446540880503147, | |
| "grad_norm": 0.43803247809410095, | |
| "learning_rate": 1.3011879804332637e-05, | |
| "loss": 0.1095, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.9006451612903226, | |
| "eval_loss": 0.049462586641311646, | |
| "eval_runtime": 5.8278, | |
| "eval_samples_per_second": 531.931, | |
| "eval_steps_per_second": 11.153, | |
| "step": 1272 | |
| }, | |
| { | |
| "epoch": 4.716981132075472, | |
| "grad_norm": 0.2977636158466339, | |
| "learning_rate": 9.517819706498952e-06, | |
| "loss": 0.0705, | |
| "step": 1500 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 2862, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 9, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 391368939443328.0, | |
| "train_batch_size": 48, | |
| "trial_name": null, | |
| "trial_params": { | |
| "alpha": 0.7805733466574912, | |
| "num_train_epochs": 9, | |
| "temperature": 15 | |
| } | |
| } | |