| { | |
| "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": 0.5783870967741935, | |
| "eval_loss": 0.19767619669437408, | |
| "eval_runtime": 6.8983, | |
| "eval_samples_per_second": 449.384, | |
| "eval_steps_per_second": 9.423, | |
| "step": 318 | |
| }, | |
| { | |
| "epoch": 1.5723270440251573, | |
| "grad_norm": 0.5253991484642029, | |
| "learning_rate": 1.550763701707098e-05, | |
| "loss": 0.3154, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8151612903225807, | |
| "eval_loss": 0.09857291728258133, | |
| "eval_runtime": 6.9285, | |
| "eval_samples_per_second": 447.424, | |
| "eval_steps_per_second": 9.381, | |
| "step": 636 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8722580645161291, | |
| "eval_loss": 0.06783536076545715, | |
| "eval_runtime": 5.8131, | |
| "eval_samples_per_second": 533.278, | |
| "eval_steps_per_second": 11.182, | |
| "step": 954 | |
| }, | |
| { | |
| "epoch": 3.1446540880503147, | |
| "grad_norm": 0.4484029710292816, | |
| "learning_rate": 1.101527403414196e-05, | |
| "loss": 0.1139, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8938709677419355, | |
| "eval_loss": 0.05406318977475166, | |
| "eval_runtime": 5.9462, | |
| "eval_samples_per_second": 521.342, | |
| "eval_steps_per_second": 10.931, | |
| "step": 1272 | |
| }, | |
| { | |
| "epoch": 4.716981132075472, | |
| "grad_norm": 0.31517869234085083, | |
| "learning_rate": 6.522911051212939e-06, | |
| "loss": 0.0759, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.9012903225806451, | |
| "eval_loss": 0.04607924818992615, | |
| "eval_runtime": 6.3687, | |
| "eval_samples_per_second": 486.755, | |
| "eval_steps_per_second": 10.206, | |
| "step": 1590 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.9038709677419355, | |
| "eval_loss": 0.04200601950287819, | |
| "eval_runtime": 5.4702, | |
| "eval_samples_per_second": 566.709, | |
| "eval_steps_per_second": 11.883, | |
| "step": 1908 | |
| }, | |
| { | |
| "epoch": 6.289308176100629, | |
| "grad_norm": 0.29168257117271423, | |
| "learning_rate": 2.0305480682839176e-06, | |
| "loss": 0.0626, | |
| "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": 0.6369202392102897, | |
| "num_train_epochs": 7, | |
| "temperature": 14 | |
| } | |
| } | |