| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 7.0, | |
| "eval_steps": 500, | |
| "global_step": 2226, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.6445161290322581, | |
| "eval_loss": 0.29249975085258484, | |
| "eval_runtime": 5.1547, | |
| "eval_samples_per_second": 601.395, | |
| "eval_steps_per_second": 12.61, | |
| "step": 318 | |
| }, | |
| { | |
| "epoch": 1.5723270440251573, | |
| "grad_norm": 0.6633772850036621, | |
| "learning_rate": 1.550763701707098e-05, | |
| "loss": 0.4626, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8380645161290322, | |
| "eval_loss": 0.12248263508081436, | |
| "eval_runtime": 5.377, | |
| "eval_samples_per_second": 576.528, | |
| "eval_steps_per_second": 12.088, | |
| "step": 636 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8887096774193548, | |
| "eval_loss": 0.07343784719705582, | |
| "eval_runtime": 5.2352, | |
| "eval_samples_per_second": 592.149, | |
| "eval_steps_per_second": 12.416, | |
| "step": 954 | |
| }, | |
| { | |
| "epoch": 3.1446540880503147, | |
| "grad_norm": 0.59869384765625, | |
| "learning_rate": 1.101527403414196e-05, | |
| "loss": 0.1438, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.9045161290322581, | |
| "eval_loss": 0.05444410815834999, | |
| "eval_runtime": 5.4434, | |
| "eval_samples_per_second": 569.5, | |
| "eval_steps_per_second": 11.941, | |
| "step": 1272 | |
| }, | |
| { | |
| "epoch": 4.716981132075472, | |
| "grad_norm": 0.4075869023799896, | |
| "learning_rate": 6.522911051212939e-06, | |
| "loss": 0.0845, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.9141935483870968, | |
| "eval_loss": 0.045045968145132065, | |
| "eval_runtime": 5.238, | |
| "eval_samples_per_second": 591.824, | |
| "eval_steps_per_second": 12.409, | |
| "step": 1590 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.9190322580645162, | |
| "eval_loss": 0.04080257937312126, | |
| "eval_runtime": 5.1414, | |
| "eval_samples_per_second": 602.95, | |
| "eval_steps_per_second": 12.642, | |
| "step": 1908 | |
| }, | |
| { | |
| "epoch": 6.289308176100629, | |
| "grad_norm": 0.352282851934433, | |
| "learning_rate": 2.0305480682839176e-06, | |
| "loss": 0.0669, | |
| "step": 2000 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 2226, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 7, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 520991326672152.0, | |
| "train_batch_size": 48, | |
| "trial_name": null, | |
| "trial_params": { | |
| "alpha": 0.3308965359344769, | |
| "num_train_epochs": 7, | |
| "temperature": 3 | |
| } | |
| } | |