| { | |
| "best_metric": 0.7435897435897436, | |
| "best_model_checkpoint": "Output_llama2_70-15-15/checkpoint-460", | |
| "epoch": 20.0, | |
| "eval_steps": 500, | |
| "global_step": 920, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.6538461538461539, | |
| "eval_balanced_accuracy": 0.654945054945055, | |
| "eval_loss": 0.6548082232475281, | |
| "eval_runtime": 142.4927, | |
| "eval_samples_per_second": 1.095, | |
| "eval_steps_per_second": 0.07, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.6153846153846154, | |
| "eval_balanced_accuracy": 0.6180613090306546, | |
| "eval_loss": 0.659588634967804, | |
| "eval_runtime": 174.4587, | |
| "eval_samples_per_second": 0.894, | |
| "eval_steps_per_second": 0.057, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_balanced_accuracy": 0.7773972602739726, | |
| "eval_loss": 0.5924286842346191, | |
| "eval_runtime": 145.3834, | |
| "eval_samples_per_second": 1.073, | |
| "eval_steps_per_second": 0.069, | |
| "step": 138 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.6474358974358975, | |
| "eval_balanced_accuracy": 0.6468253968253967, | |
| "eval_loss": 0.621242105960846, | |
| "eval_runtime": 157.0532, | |
| "eval_samples_per_second": 0.993, | |
| "eval_steps_per_second": 0.064, | |
| "step": 184 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.6666666666666666, | |
| "eval_balanced_accuracy": 0.6741095162147794, | |
| "eval_loss": 0.5860592722892761, | |
| "eval_runtime": 152.4923, | |
| "eval_samples_per_second": 1.023, | |
| "eval_steps_per_second": 0.066, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.6538461538461539, | |
| "eval_balanced_accuracy": 0.6597350327016602, | |
| "eval_loss": 0.6232044696807861, | |
| "eval_runtime": 143.706, | |
| "eval_samples_per_second": 1.086, | |
| "eval_steps_per_second": 0.07, | |
| "step": 276 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.6346153846153846, | |
| "eval_balanced_accuracy": 0.6890120967741935, | |
| "eval_loss": 0.5387519001960754, | |
| "eval_runtime": 153.5233, | |
| "eval_samples_per_second": 1.016, | |
| "eval_steps_per_second": 0.065, | |
| "step": 322 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.6923076923076923, | |
| "eval_balanced_accuracy": 0.7074555779446785, | |
| "eval_loss": 0.5306457281112671, | |
| "eval_runtime": 158.9271, | |
| "eval_samples_per_second": 0.982, | |
| "eval_steps_per_second": 0.063, | |
| "step": 368 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.6923076923076923, | |
| "eval_balanced_accuracy": 0.6994583260527695, | |
| "eval_loss": 0.5140182375907898, | |
| "eval_runtime": 145.2382, | |
| "eval_samples_per_second": 1.074, | |
| "eval_steps_per_second": 0.069, | |
| "step": 414 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.7435897435897436, | |
| "eval_balanced_accuracy": 0.7438276719138359, | |
| "eval_loss": 0.513781726360321, | |
| "eval_runtime": 144.459, | |
| "eval_samples_per_second": 1.08, | |
| "eval_steps_per_second": 0.069, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 10.869565217391305, | |
| "grad_norm": 22.628210067749023, | |
| "learning_rate": 4.565217391304348e-06, | |
| "loss": 0.5585, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.7371794871794872, | |
| "eval_balanced_accuracy": 0.7398989898989898, | |
| "eval_loss": 0.49725937843322754, | |
| "eval_runtime": 145.8734, | |
| "eval_samples_per_second": 1.069, | |
| "eval_steps_per_second": 0.069, | |
| "step": 506 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.6794871794871795, | |
| "eval_balanced_accuracy": 0.6999809269502193, | |
| "eval_loss": 0.49190282821655273, | |
| "eval_runtime": 148.1722, | |
| "eval_samples_per_second": 1.053, | |
| "eval_steps_per_second": 0.067, | |
| "step": 552 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.717948717948718, | |
| "eval_balanced_accuracy": 0.7293992557150452, | |
| "eval_loss": 0.4932830035686493, | |
| "eval_runtime": 150.2623, | |
| "eval_samples_per_second": 1.038, | |
| "eval_steps_per_second": 0.067, | |
| "step": 598 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.7051282051282052, | |
| "eval_balanced_accuracy": 0.7297348846080488, | |
| "eval_loss": 0.49414268136024475, | |
| "eval_runtime": 152.7381, | |
| "eval_samples_per_second": 1.021, | |
| "eval_steps_per_second": 0.065, | |
| "step": 644 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.7115384615384616, | |
| "eval_balanced_accuracy": 0.7239285714285715, | |
| "eval_loss": 0.4975796341896057, | |
| "eval_runtime": 148.3264, | |
| "eval_samples_per_second": 1.052, | |
| "eval_steps_per_second": 0.067, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.7371794871794872, | |
| "eval_balanced_accuracy": 0.7386363636363636, | |
| "eval_loss": 0.5156741738319397, | |
| "eval_runtime": 151.5711, | |
| "eval_samples_per_second": 1.029, | |
| "eval_steps_per_second": 0.066, | |
| "step": 736 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.717948717948718, | |
| "eval_balanced_accuracy": 0.720879120879121, | |
| "eval_loss": 0.5169395208358765, | |
| "eval_runtime": 152.1754, | |
| "eval_samples_per_second": 1.025, | |
| "eval_steps_per_second": 0.066, | |
| "step": 782 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.7243589743589743, | |
| "eval_balanced_accuracy": 0.7282608695652174, | |
| "eval_loss": 0.5141614079475403, | |
| "eval_runtime": 146.7922, | |
| "eval_samples_per_second": 1.063, | |
| "eval_steps_per_second": 0.068, | |
| "step": 828 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.7115384615384616, | |
| "eval_balanced_accuracy": 0.7167124227865478, | |
| "eval_loss": 0.5163176655769348, | |
| "eval_runtime": 112.8422, | |
| "eval_samples_per_second": 1.382, | |
| "eval_steps_per_second": 0.089, | |
| "step": 874 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.7115384615384616, | |
| "eval_balanced_accuracy": 0.7113095238095238, | |
| "eval_loss": 0.5302315950393677, | |
| "eval_runtime": 112.7072, | |
| "eval_samples_per_second": 1.384, | |
| "eval_steps_per_second": 0.089, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "step": 920, | |
| "total_flos": 4.976764349571072e+17, | |
| "train_loss": 0.4557167053222656, | |
| "train_runtime": 45521.9577, | |
| "train_samples_per_second": 0.321, | |
| "train_steps_per_second": 0.02 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 920, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 20, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "EarlyStoppingCallback": { | |
| "args": { | |
| "early_stopping_patience": 10, | |
| "early_stopping_threshold": 0.0 | |
| }, | |
| "attributes": { | |
| "early_stopping_patience_counter": 0 | |
| } | |
| }, | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 4.976764349571072e+17, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |