| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.0, | |
| "eval_steps": 500, | |
| "global_step": 6948, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.2158894645941278, | |
| "grad_norm": 1.0430593490600586, | |
| "learning_rate": 1.856649395509499e-05, | |
| "loss": 0.8043, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.4317789291882556, | |
| "grad_norm": 0.8806478977203369, | |
| "learning_rate": 1.7127230857800806e-05, | |
| "loss": 0.4945, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.6476683937823834, | |
| "grad_norm": 0.49951252341270447, | |
| "learning_rate": 1.568796776050662e-05, | |
| "loss": 0.4456, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.8635578583765112, | |
| "grad_norm": 0.8921486139297485, | |
| "learning_rate": 1.4248704663212436e-05, | |
| "loss": 0.4194, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.33670297265052795, | |
| "eval_runtime": 28.3351, | |
| "eval_samples_per_second": 16.658, | |
| "eval_steps_per_second": 4.164, | |
| "step": 2316 | |
| }, | |
| { | |
| "epoch": 1.079447322970639, | |
| "grad_norm": 3.1095924377441406, | |
| "learning_rate": 1.280944156591825e-05, | |
| "loss": 0.4036, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.2953367875647668, | |
| "grad_norm": 1.0692085027694702, | |
| "learning_rate": 1.1370178468624064e-05, | |
| "loss": 0.3879, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.5112262521588946, | |
| "grad_norm": 0.9439179301261902, | |
| "learning_rate": 9.93091537132988e-06, | |
| "loss": 0.3728, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.7271157167530224, | |
| "grad_norm": 0.9635007381439209, | |
| "learning_rate": 8.491652274035695e-06, | |
| "loss": 0.3675, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.9430051813471503, | |
| "grad_norm": 1.1892434358596802, | |
| "learning_rate": 7.052389176741509e-06, | |
| "loss": 0.362, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.3095574975013733, | |
| "eval_runtime": 28.3225, | |
| "eval_samples_per_second": 16.665, | |
| "eval_steps_per_second": 4.166, | |
| "step": 4632 | |
| }, | |
| { | |
| "epoch": 2.158894645941278, | |
| "grad_norm": 0.6397312879562378, | |
| "learning_rate": 5.613126079447323e-06, | |
| "loss": 0.3467, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 2.3747841105354057, | |
| "grad_norm": 0.6125323176383972, | |
| "learning_rate": 4.173862982153138e-06, | |
| "loss": 0.3513, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 2.5906735751295336, | |
| "grad_norm": 0.5765931606292725, | |
| "learning_rate": 2.734599884858952e-06, | |
| "loss": 0.3428, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 2.8065630397236614, | |
| "grad_norm": 0.8941567540168762, | |
| "learning_rate": 1.2982153137593554e-06, | |
| "loss": 0.353, | |
| "step": 6500 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 6948, | |
| "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": 1.692233082077184e+16, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |