{ "best_metric": null, "best_model_checkpoint": null, "epoch": 3.0, "eval_steps": 500, "global_step": 68223, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 1.0, "grad_norm": 5.0593485832214355, "learning_rate": 1.3333333333333333e-05, "loss": 0.2942, "step": 22741 }, { "epoch": 1.0, "eval_accuracy": 0.9008656937917388, "eval_combined_score": 0.8844814218975149, "eval_f1": 0.868097150003291, "eval_loss": 0.2476482093334198, "eval_runtime": 125.5141, "eval_samples_per_second": 322.115, "eval_steps_per_second": 40.266, "step": 22741 }, { "epoch": 2.0, "grad_norm": 10.616366386413574, "learning_rate": 6.666666666666667e-06, "loss": 0.1919, "step": 45482 }, { "epoch": 2.0, "eval_accuracy": 0.9079643828839971, "eval_combined_score": 0.892029542512559, "eval_f1": 0.8760947021411208, "eval_loss": 0.2705819010734558, "eval_runtime": 126.0816, "eval_samples_per_second": 320.665, "eval_steps_per_second": 40.085, "step": 45482 }, { "epoch": 3.0, "grad_norm": 0.2998017370700836, "learning_rate": 0.0, "loss": 0.1342, "step": 68223 }, { "epoch": 3.0, "eval_accuracy": 0.9108830076675736, "eval_combined_score": 0.8956724185346925, "eval_f1": 0.8804618294018115, "eval_loss": 0.3567805886268616, "eval_runtime": 126.0333, "eval_samples_per_second": 320.788, "eval_steps_per_second": 40.101, "step": 68223 }, { "epoch": 3.0, "step": 68223, "total_flos": 2.8719571514554368e+17, "train_loss": 0.0, "train_runtime": 0.0017, "train_samples_per_second": 658929504.829, "train_steps_per_second": 41184225.934 } ], "logging_steps": 500, "max_steps": 68223, "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": 2.8719571514554368e+17, "train_batch_size": 16, "trial_name": null, "trial_params": null }