| { | |
| "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": 0.6258064516129033, | |
| "eval_loss": 0.2538500130176544, | |
| "eval_runtime": 3.2111, | |
| "eval_samples_per_second": 965.397, | |
| "eval_steps_per_second": 20.242, | |
| "step": 318 | |
| }, | |
| { | |
| "epoch": 1.5723270440251573, | |
| "grad_norm": 0.6582645177841187, | |
| "learning_rate": 1.550763701707098e-05, | |
| "loss": 0.4023, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8419354838709677, | |
| "eval_loss": 0.1154777854681015, | |
| "eval_runtime": 3.3276, | |
| "eval_samples_per_second": 931.608, | |
| "eval_steps_per_second": 19.534, | |
| "step": 636 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8861290322580645, | |
| "eval_loss": 0.07336921989917755, | |
| "eval_runtime": 3.2154, | |
| "eval_samples_per_second": 964.125, | |
| "eval_steps_per_second": 20.216, | |
| "step": 954 | |
| }, | |
| { | |
| "epoch": 3.1446540880503147, | |
| "grad_norm": 0.6269882321357727, | |
| "learning_rate": 1.101527403414196e-05, | |
| "loss": 0.1345, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8996774193548387, | |
| "eval_loss": 0.05509715527296066, | |
| "eval_runtime": 3.2247, | |
| "eval_samples_per_second": 961.335, | |
| "eval_steps_per_second": 20.157, | |
| "step": 1272 | |
| }, | |
| { | |
| "epoch": 4.716981132075472, | |
| "grad_norm": 0.4124581515789032, | |
| "learning_rate": 6.522911051212939e-06, | |
| "loss": 0.0826, | |
| "step": 1500 | |
| } | |
| ], | |
| "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": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 391368939443328.0, | |
| "train_batch_size": 48, | |
| "trial_name": null, | |
| "trial_params": { | |
| "alpha": 0.35058021873169465, | |
| "num_train_epochs": 7, | |
| "temperature": 4 | |
| } | |
| } | |