{ "best_metric": 0.8222477064220184, "best_model_checkpoint": "tiny-bert-sst2-distilled/run-1/checkpoint-1581", "epoch": 3.0, "eval_steps": 500, "global_step": 1581, "is_hyper_param_search": true, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 1.0, "grad_norm": 17.658132553100586, "learning_rate": 7.63128750557968e-05, "loss": 1.3361, "step": 527 }, { "epoch": 1.0, "eval_accuracy": 0.8107798165137615, "eval_loss": 1.132369041442871, "eval_runtime": 2.6024, "eval_samples_per_second": 335.073, "eval_steps_per_second": 2.69, "step": 527 }, { "epoch": 2.0, "grad_norm": 17.63370704650879, "learning_rate": 3.81564375278984e-05, "loss": 0.7175, "step": 1054 }, { "epoch": 2.0, "eval_accuracy": 0.819954128440367, "eval_loss": 1.094281554222107, "eval_runtime": 2.5939, "eval_samples_per_second": 336.174, "eval_steps_per_second": 2.699, "step": 1054 }, { "epoch": 3.0, "grad_norm": Infinity, "learning_rate": 7.240310726356433e-08, "loss": 0.5796, "step": 1581 }, { "epoch": 3.0, "eval_accuracy": 0.8222477064220184, "eval_loss": 1.1151666641235352, "eval_runtime": 2.56, "eval_samples_per_second": 340.627, "eval_steps_per_second": 2.734, "step": 1581 } ], "logging_steps": 500, "max_steps": 1581, "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": 24228458327580.0, "train_batch_size": 128, "trial_name": null, "trial_params": { "alpha": 0.5918699224275719, "learning_rate": 0.00011446931258369519, "num_train_epochs": 3, "temperature": 9 } }