| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.0, | |
| "eval_steps": 500, | |
| "global_step": 4221, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.35536602700781805, | |
| "grad_norm": 3.349424362182617, | |
| "learning_rate": 4.4077232883203036e-05, | |
| "loss": 1.0722, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.7107320540156361, | |
| "grad_norm": 3.930595636367798, | |
| "learning_rate": 3.815446576640607e-05, | |
| "loss": 1.0092, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.5769, | |
| "eval_f1": 0.5719399737001815, | |
| "eval_loss": 0.9482702016830444, | |
| "eval_precision": 0.5727437080672995, | |
| "eval_recall": 0.5769, | |
| "eval_runtime": 95.581, | |
| "eval_samples_per_second": 104.623, | |
| "eval_steps_per_second": 1.643, | |
| "step": 1407 | |
| }, | |
| { | |
| "epoch": 1.0660980810234542, | |
| "grad_norm": 2.6278605461120605, | |
| "learning_rate": 3.2231698649609096e-05, | |
| "loss": 0.9693, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.4214641080312722, | |
| "grad_norm": 3.751405715942383, | |
| "learning_rate": 2.630893153281213e-05, | |
| "loss": 0.8914, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.7768301350390903, | |
| "grad_norm": 3.272271156311035, | |
| "learning_rate": 2.0386164416015163e-05, | |
| "loss": 0.8831, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.5826, | |
| "eval_f1": 0.581743457079248, | |
| "eval_loss": 0.9372864365577698, | |
| "eval_precision": 0.5848366372292547, | |
| "eval_recall": 0.5826, | |
| "eval_runtime": 95.5055, | |
| "eval_samples_per_second": 104.706, | |
| "eval_steps_per_second": 1.644, | |
| "step": 2814 | |
| }, | |
| { | |
| "epoch": 2.1321961620469083, | |
| "grad_norm": 4.172728061676025, | |
| "learning_rate": 1.4463397299218195e-05, | |
| "loss": 0.8467, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 2.487562189054726, | |
| "grad_norm": 3.4016871452331543, | |
| "learning_rate": 8.540630182421228e-06, | |
| "loss": 0.7759, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 2.8429282160625444, | |
| "grad_norm": 4.218382358551025, | |
| "learning_rate": 2.6178630656242596e-06, | |
| "loss": 0.7738, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.582, | |
| "eval_f1": 0.5809528539010538, | |
| "eval_loss": 0.9759224653244019, | |
| "eval_precision": 0.581423790890455, | |
| "eval_recall": 0.582, | |
| "eval_runtime": 95.4692, | |
| "eval_samples_per_second": 104.746, | |
| "eval_steps_per_second": 1.645, | |
| "step": 4221 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 4221, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "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": 3.57681111552e+16, | |
| "train_batch_size": 64, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |