| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 6.289308176100629, | |
| "eval_steps": 500, | |
| "global_step": 2000, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.6509677419354839, | |
| "eval_loss": 0.2873552143573761, | |
| "eval_runtime": 5.4953, | |
| "eval_samples_per_second": 564.118, | |
| "eval_steps_per_second": 11.828, | |
| "step": 318 | |
| }, | |
| { | |
| "epoch": 1.5723270440251573, | |
| "grad_norm": 0.6625721454620361, | |
| "learning_rate": 1.650593990216632e-05, | |
| "loss": 0.4591, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8429032258064516, | |
| "eval_loss": 0.11606641113758087, | |
| "eval_runtime": 5.7777, | |
| "eval_samples_per_second": 536.547, | |
| "eval_steps_per_second": 11.25, | |
| "step": 636 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8948387096774194, | |
| "eval_loss": 0.06758537143468857, | |
| "eval_runtime": 5.6402, | |
| "eval_samples_per_second": 549.628, | |
| "eval_steps_per_second": 11.524, | |
| "step": 954 | |
| }, | |
| { | |
| "epoch": 3.1446540880503147, | |
| "grad_norm": 0.5751062631607056, | |
| "learning_rate": 1.3011879804332637e-05, | |
| "loss": 0.1369, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.91, | |
| "eval_loss": 0.04885578155517578, | |
| "eval_runtime": 5.5152, | |
| "eval_samples_per_second": 562.081, | |
| "eval_steps_per_second": 11.786, | |
| "step": 1272 | |
| }, | |
| { | |
| "epoch": 4.716981132075472, | |
| "grad_norm": 0.373975932598114, | |
| "learning_rate": 9.517819706498952e-06, | |
| "loss": 0.0773, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.9180645161290323, | |
| "eval_loss": 0.03952872380614281, | |
| "eval_runtime": 5.702, | |
| "eval_samples_per_second": 543.671, | |
| "eval_steps_per_second": 11.4, | |
| "step": 1590 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.9261290322580645, | |
| "eval_loss": 0.03494011610746384, | |
| "eval_runtime": 5.7679, | |
| "eval_samples_per_second": 537.462, | |
| "eval_steps_per_second": 11.269, | |
| "step": 1908 | |
| }, | |
| { | |
| "epoch": 6.289308176100629, | |
| "grad_norm": 0.30554112792015076, | |
| "learning_rate": 6.02375960866527e-06, | |
| "loss": 0.0588, | |
| "step": 2000 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 2862, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 9, | |
| "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": 520991326672152.0, | |
| "train_batch_size": 48, | |
| "trial_name": null, | |
| "trial_params": { | |
| "alpha": 0.8846264724466298, | |
| "num_train_epochs": 9, | |
| "temperature": 3 | |
| } | |
| } | |