{ "best_metric": 0.5057714581489563, "best_model_checkpoint": "distilbert_lda_5_v1_book_cola/checkpoint-68", "epoch": 7.0, "eval_steps": 500, "global_step": 238, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 1.0, "grad_norm": 2.265052556991577, "learning_rate": 4.9e-05, "loss": 0.5951, "step": 34 }, { "epoch": 1.0, "eval_accuracy": 0.7277085185050964, "eval_loss": 0.5445801019668579, "eval_matthews_correlation": 0.27743272040414674, "eval_runtime": 0.3251, "eval_samples_per_second": 3207.861, "eval_steps_per_second": 15.378, "step": 34 }, { "epoch": 2.0, "grad_norm": 5.59794807434082, "learning_rate": 4.8e-05, "loss": 0.4788, "step": 68 }, { "epoch": 2.0, "eval_accuracy": 0.7660594582557678, "eval_loss": 0.5057714581489563, "eval_matthews_correlation": 0.4068166329179125, "eval_runtime": 0.3222, "eval_samples_per_second": 3237.203, "eval_steps_per_second": 15.519, "step": 68 }, { "epoch": 3.0, "grad_norm": 4.7205119132995605, "learning_rate": 4.7e-05, "loss": 0.3438, "step": 102 }, { "epoch": 3.0, "eval_accuracy": 0.772770881652832, "eval_loss": 0.5465193390846252, "eval_matthews_correlation": 0.4231731541032522, "eval_runtime": 0.3125, "eval_samples_per_second": 3337.587, "eval_steps_per_second": 16.0, "step": 102 }, { "epoch": 4.0, "grad_norm": 5.3492655754089355, "learning_rate": 4.600000000000001e-05, "loss": 0.2469, "step": 136 }, { "epoch": 4.0, "eval_accuracy": 0.7775647044181824, "eval_loss": 0.5831495523452759, "eval_matthews_correlation": 0.4439591175312219, "eval_runtime": 0.325, "eval_samples_per_second": 3208.748, "eval_steps_per_second": 15.382, "step": 136 }, { "epoch": 5.0, "grad_norm": 5.582437515258789, "learning_rate": 4.5e-05, "loss": 0.171, "step": 170 }, { "epoch": 5.0, "eval_accuracy": 0.7794822454452515, "eval_loss": 0.674407422542572, "eval_matthews_correlation": 0.4526212019715672, "eval_runtime": 0.3187, "eval_samples_per_second": 3272.554, "eval_steps_per_second": 15.688, "step": 170 }, { "epoch": 6.0, "grad_norm": 7.174558639526367, "learning_rate": 4.4000000000000006e-05, "loss": 0.1344, "step": 204 }, { "epoch": 6.0, "eval_accuracy": 0.7861936688423157, "eval_loss": 0.7193422913551331, "eval_matthews_correlation": 0.4712167809814141, "eval_runtime": 0.3292, "eval_samples_per_second": 3167.871, "eval_steps_per_second": 15.186, "step": 204 }, { "epoch": 7.0, "grad_norm": 4.941391468048096, "learning_rate": 4.3e-05, "loss": 0.1026, "step": 238 }, { "epoch": 7.0, "eval_accuracy": 0.7900287508964539, "eval_loss": 0.8283830881118774, "eval_matthews_correlation": 0.4791261224300467, "eval_runtime": 0.346, "eval_samples_per_second": 3014.071, "eval_steps_per_second": 14.449, "step": 238 }, { "epoch": 7.0, "step": 238, "total_flos": 3964550540676096.0, "train_loss": 0.296079120716127, "train_runtime": 50.1772, "train_samples_per_second": 8520.806, "train_steps_per_second": 33.88 } ], "logging_steps": 1, "max_steps": 1700, "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": 3964550540676096.0, "train_batch_size": 256, "trial_name": null, "trial_params": null }