| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.657142857142857, | |
| "eval_steps": 500, | |
| "global_step": 16, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.45714285714285713, | |
| "grad_norm": 0.9680992960929871, | |
| "learning_rate": 0.000175, | |
| "loss": 9.5072, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.9142857142857143, | |
| "grad_norm": 1.5162838697433472, | |
| "learning_rate": 0.00015000000000000001, | |
| "loss": 9.2188, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 1.3714285714285714, | |
| "grad_norm": 1.7571940422058105, | |
| "learning_rate": 0.000125, | |
| "loss": 8.7662, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 1.8285714285714287, | |
| "grad_norm": 1.9175825119018555, | |
| "learning_rate": 0.0001, | |
| "loss": 8.3673, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 2.2857142857142856, | |
| "grad_norm": 1.9747986793518066, | |
| "learning_rate": 7.500000000000001e-05, | |
| "loss": 7.9358, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 2.742857142857143, | |
| "grad_norm": 1.8220804929733276, | |
| "learning_rate": 5e-05, | |
| "loss": 7.7996, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "grad_norm": NaN, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 9.0907, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 3.657142857142857, | |
| "grad_norm": 1.4926611185073853, | |
| "learning_rate": 1.25e-05, | |
| "loss": 7.5739, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 3.657142857142857, | |
| "step": 16, | |
| "total_flos": 73613559088512.0, | |
| "train_loss": 8.532437920570374, | |
| "train_runtime": 488.0905, | |
| "train_samples_per_second": 0.574, | |
| "train_steps_per_second": 0.033 | |
| } | |
| ], | |
| "logging_steps": 2, | |
| "max_steps": 16, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 4, | |
| "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": 73613559088512.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |