{ "best_metric": 0.5723373889923096, "best_model_checkpoint": "distilbert_lda_20_v1_book_mrpc/checkpoint-30", "epoch": 7.0, "eval_steps": 500, "global_step": 105, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 1.0, "grad_norm": 1.5297983884811401, "learning_rate": 4.9e-05, "loss": 0.6229, "step": 15 }, { "epoch": 1.0, "eval_accuracy": 0.6887254901960784, "eval_combined_score": 0.744988723188806, "eval_f1": 0.8012519561815337, "eval_loss": 0.5868853330612183, "eval_runtime": 0.2038, "eval_samples_per_second": 2001.902, "eval_steps_per_second": 9.813, "step": 15 }, { "epoch": 2.0, "grad_norm": 2.2882039546966553, "learning_rate": 4.8e-05, "loss": 0.5861, "step": 30 }, { "epoch": 2.0, "eval_accuracy": 0.6985294117647058, "eval_combined_score": 0.7524143121815655, "eval_f1": 0.8062992125984252, "eval_loss": 0.5723373889923096, "eval_runtime": 0.1927, "eval_samples_per_second": 2117.072, "eval_steps_per_second": 10.378, "step": 30 }, { "epoch": 3.0, "grad_norm": 7.32234525680542, "learning_rate": 4.7e-05, "loss": 0.5298, "step": 45 }, { "epoch": 3.0, "eval_accuracy": 0.696078431372549, "eval_combined_score": 0.7404003267973855, "eval_f1": 0.7847222222222222, "eval_loss": 0.5741574168205261, "eval_runtime": 0.2263, "eval_samples_per_second": 1802.594, "eval_steps_per_second": 8.836, "step": 45 }, { "epoch": 4.0, "grad_norm": 4.486771583557129, "learning_rate": 4.600000000000001e-05, "loss": 0.4213, "step": 60 }, { "epoch": 4.0, "eval_accuracy": 0.7279411764705882, "eval_combined_score": 0.7677834131919665, "eval_f1": 0.8076256499133448, "eval_loss": 0.6085153222084045, "eval_runtime": 0.1966, "eval_samples_per_second": 2075.315, "eval_steps_per_second": 10.173, "step": 60 }, { "epoch": 5.0, "grad_norm": 22.681074142456055, "learning_rate": 4.5e-05, "loss": 0.3198, "step": 75 }, { "epoch": 5.0, "eval_accuracy": 0.7426470588235294, "eval_combined_score": 0.7862343884069025, "eval_f1": 0.8298217179902755, "eval_loss": 0.7630409002304077, "eval_runtime": 0.2359, "eval_samples_per_second": 1729.612, "eval_steps_per_second": 8.478, "step": 75 }, { "epoch": 6.0, "grad_norm": 12.322725296020508, "learning_rate": 4.4000000000000006e-05, "loss": 0.3685, "step": 90 }, { "epoch": 6.0, "eval_accuracy": 0.7303921568627451, "eval_combined_score": 0.7683650925158796, "eval_f1": 0.8063380281690141, "eval_loss": 0.673819899559021, "eval_runtime": 0.1889, "eval_samples_per_second": 2159.851, "eval_steps_per_second": 10.588, "step": 90 }, { "epoch": 7.0, "grad_norm": 4.493818759918213, "learning_rate": 4.3e-05, "loss": 0.1999, "step": 105 }, { "epoch": 7.0, "eval_accuracy": 0.7083333333333334, "eval_combined_score": 0.7657565626547795, "eval_f1": 0.8231797919762258, "eval_loss": 1.124142050743103, "eval_runtime": 0.1919, "eval_samples_per_second": 2125.851, "eval_steps_per_second": 10.421, "step": 105 }, { "epoch": 7.0, "step": 105, "total_flos": 1700616463945728.0, "train_loss": 0.4354898702530634, "train_runtime": 38.232, "train_samples_per_second": 4797.032, "train_steps_per_second": 19.617 } ], "logging_steps": 1, "max_steps": 750, "num_input_tokens_seen": 0, "num_train_epochs": 50, "save_steps": 500, "stateful_callbacks": { "EarlyStoppingCallback": { "args": { "early_stopping_patience": 5, "early_stopping_threshold": 0.0 }, "attributes": { "early_stopping_patience_counter": 5 } }, "TrainerControl": { "args": { "should_epoch_stop": false, "should_evaluate": false, "should_log": false, "should_save": true, "should_training_stop": true }, "attributes": {} } }, "total_flos": 1700616463945728.0, "train_batch_size": 256, "trial_name": null, "trial_params": null }