| { | |
| "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": 0.603225806451613, | |
| "eval_loss": 0.23406817018985748, | |
| "eval_runtime": 5.1784, | |
| "eval_samples_per_second": 598.637, | |
| "eval_steps_per_second": 12.552, | |
| "step": 318 | |
| }, | |
| { | |
| "epoch": 1.5723270440251573, | |
| "grad_norm": 0.57925945520401, | |
| "learning_rate": 1.550763701707098e-05, | |
| "loss": 0.3716, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8309677419354838, | |
| "eval_loss": 0.1094982922077179, | |
| "eval_runtime": 5.4255, | |
| "eval_samples_per_second": 571.371, | |
| "eval_steps_per_second": 11.98, | |
| "step": 636 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8819354838709678, | |
| "eval_loss": 0.07135984301567078, | |
| "eval_runtime": 5.3165, | |
| "eval_samples_per_second": 583.086, | |
| "eval_steps_per_second": 12.226, | |
| "step": 954 | |
| }, | |
| { | |
| "epoch": 3.1446540880503147, | |
| "grad_norm": 0.5091490149497986, | |
| "learning_rate": 1.101527403414196e-05, | |
| "loss": 0.1274, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8980645161290323, | |
| "eval_loss": 0.055049341171979904, | |
| "eval_runtime": 5.431, | |
| "eval_samples_per_second": 570.802, | |
| "eval_steps_per_second": 11.968, | |
| "step": 1272 | |
| }, | |
| { | |
| "epoch": 4.716981132075472, | |
| "grad_norm": 0.3538859486579895, | |
| "learning_rate": 6.522911051212939e-06, | |
| "loss": 0.0809, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.9061290322580645, | |
| "eval_loss": 0.04594043269753456, | |
| "eval_runtime": 5.5266, | |
| "eval_samples_per_second": 560.928, | |
| "eval_steps_per_second": 11.761, | |
| "step": 1590 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.9090322580645162, | |
| "eval_loss": 0.04156717658042908, | |
| "eval_runtime": 5.3097, | |
| "eval_samples_per_second": 583.842, | |
| "eval_steps_per_second": 12.242, | |
| "step": 1908 | |
| }, | |
| { | |
| "epoch": 6.289308176100629, | |
| "grad_norm": 0.31923583149909973, | |
| "learning_rate": 2.0305480682839176e-06, | |
| "loss": 0.0654, | |
| "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": 0.6968668210901579, | |
| "num_train_epochs": 7, | |
| "temperature": 5 | |
| } | |
| } | |