{ "best_global_step": null, "best_metric": null, "best_model_checkpoint": null, "epoch": 3.0, "eval_steps": 500, "global_step": 3048, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.4921259842519685, "grad_norm": 0.0023708906956017017, "learning_rate": 4.181430446194226e-05, "loss": 0.0182, "step": 500 }, { "epoch": 0.984251968503937, "grad_norm": 0.005748015828430653, "learning_rate": 3.361220472440945e-05, "loss": 0.0125, "step": 1000 }, { "epoch": 1.4763779527559056, "grad_norm": 0.020883610472083092, "learning_rate": 2.5410104986876644e-05, "loss": 0.0055, "step": 1500 }, { "epoch": 1.968503937007874, "grad_norm": 0.0016900873742997646, "learning_rate": 1.7208005249343835e-05, "loss": 0.0073, "step": 2000 }, { "epoch": 2.4606299212598426, "grad_norm": 0.0035135194193571806, "learning_rate": 9.005905511811025e-06, "loss": 0.0021, "step": 2500 }, { "epoch": 2.952755905511811, "grad_norm": 0.0004347392823547125, "learning_rate": 8.038057742782153e-07, "loss": 0.0024, "step": 3000 }, { "epoch": 3.0, "step": 3048, "total_flos": 3.7763396378982973e+18, "train_loss": 0.007946457738286554, "train_runtime": 2551.8222, "train_samples_per_second": 19.097, "train_steps_per_second": 1.194 } ], "logging_steps": 500, "max_steps": 3048, "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": 3.7763396378982973e+18, "train_batch_size": 16, "trial_name": null, "trial_params": null }