{ "best_metric": null, "best_model_checkpoint": null, "epoch": 0.001537082105802485, "eval_steps": 3, "global_step": 3, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.0005123607019341616, "grad_norm": 0.3183790445327759, "learning_rate": 2e-05, "loss": 0.7016, "step": 1 }, { "epoch": 0.0005123607019341616, "eval_loss": 1.1740705966949463, "eval_runtime": 91.1727, "eval_samples_per_second": 4.508, "eval_steps_per_second": 4.508, "step": 1 }, { "epoch": 0.0010247214038683233, "grad_norm": 0.9751269221305847, "learning_rate": 4e-05, "loss": 0.6813, "step": 2 }, { "epoch": 0.001537082105802485, "grad_norm": 0.8691899180412292, "learning_rate": 6e-05, "loss": 0.6713, "step": 3 }, { "epoch": 0.001537082105802485, "eval_loss": 1.1680519580841064, "eval_runtime": 89.7416, "eval_samples_per_second": 4.58, "eval_steps_per_second": 4.58, "step": 3 } ], "logging_steps": 1, "max_steps": 10, "num_input_tokens_seen": 0, "num_train_epochs": 1, "save_steps": 3, "stateful_callbacks": { "TrainerControl": { "args": { "should_epoch_stop": false, "should_evaluate": false, "should_log": false, "should_save": true, "should_training_stop": false }, "attributes": {} } }, "total_flos": 2097116079980544.0, "train_batch_size": 1, "trial_name": null, "trial_params": null }