| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 6.0, | |
| "eval_steps": 500, | |
| "global_step": 1908, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.9968553459119497, | |
| "grad_norm": 0.6420425772666931, | |
| "learning_rate": 1.6677148846960168e-05, | |
| "loss": 0.417, | |
| "step": 317 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.5764516129032258, | |
| "eval_loss": 0.21416573226451874, | |
| "eval_runtime": 1.3706, | |
| "eval_samples_per_second": 2261.824, | |
| "eval_steps_per_second": 47.425, | |
| "step": 318 | |
| }, | |
| { | |
| "epoch": 1.9937106918238994, | |
| "grad_norm": 0.5413391590118408, | |
| "learning_rate": 1.3354297693920338e-05, | |
| "loss": 0.1795, | |
| "step": 634 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8187096774193549, | |
| "eval_loss": 0.10729999095201492, | |
| "eval_runtime": 1.387, | |
| "eval_samples_per_second": 2235.111, | |
| "eval_steps_per_second": 46.865, | |
| "step": 636 | |
| }, | |
| { | |
| "epoch": 2.990566037735849, | |
| "grad_norm": 0.49032193422317505, | |
| "learning_rate": 1.0031446540880504e-05, | |
| "loss": 0.1149, | |
| "step": 951 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8693548387096774, | |
| "eval_loss": 0.0734153613448143, | |
| "eval_runtime": 1.5874, | |
| "eval_samples_per_second": 1952.882, | |
| "eval_steps_per_second": 40.948, | |
| "step": 954 | |
| }, | |
| { | |
| "epoch": 3.9874213836477987, | |
| "grad_norm": 0.45065024495124817, | |
| "learning_rate": 6.708595387840672e-06, | |
| "loss": 0.0894, | |
| "step": 1268 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8890322580645161, | |
| "eval_loss": 0.05864861235022545, | |
| "eval_runtime": 1.5959, | |
| "eval_samples_per_second": 1942.513, | |
| "eval_steps_per_second": 40.73, | |
| "step": 1272 | |
| }, | |
| { | |
| "epoch": 4.984276729559748, | |
| "grad_norm": 0.3604884445667267, | |
| "learning_rate": 3.385744234800839e-06, | |
| "loss": 0.0768, | |
| "step": 1585 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.8938709677419355, | |
| "eval_loss": 0.05153821408748627, | |
| "eval_runtime": 1.4183, | |
| "eval_samples_per_second": 2185.76, | |
| "eval_steps_per_second": 45.83, | |
| "step": 1590 | |
| }, | |
| { | |
| "epoch": 5.981132075471698, | |
| "grad_norm": 0.4372672736644745, | |
| "learning_rate": 6.289308176100629e-08, | |
| "loss": 0.0712, | |
| "step": 1902 | |
| } | |
| ], | |
| "logging_steps": 317, | |
| "max_steps": 1908, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 6, | |
| "save_steps": 1000000000.0, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 495439677408900.0, | |
| "train_batch_size": 48, | |
| "trial_name": null, | |
| "trial_params": { | |
| "alpha": 0.39186236224820015, | |
| "num_train_epochs": 6, | |
| "temperature": 8 | |
| } | |
| } | |