| { | |
| "best_metric": 52.55846445665645, | |
| "best_model_checkpoint": "outputs/bitfit/t5-base/cola/checkpoint-800", | |
| "epoch": 20.0, | |
| "global_step": 1720, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 2.33, | |
| "eval_average_metrics": 21.053609210794303, | |
| "eval_loss": 0.2652588486671448, | |
| "eval_matthews_correlation": 21.053609210794303, | |
| "eval_runtime": 0.8287, | |
| "eval_samples_per_second": 628.717, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 4.65, | |
| "eval_average_metrics": 44.5285953164197, | |
| "eval_loss": 0.23253390192985535, | |
| "eval_matthews_correlation": 44.5285953164197, | |
| "eval_runtime": 0.8258, | |
| "eval_samples_per_second": 630.924, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 5.81, | |
| "learning_rate": 0.0003, | |
| "loss": 1.3817, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 6.98, | |
| "eval_average_metrics": 47.511676688521085, | |
| "eval_loss": 0.25159838795661926, | |
| "eval_matthews_correlation": 47.511676688521085, | |
| "eval_runtime": 0.8388, | |
| "eval_samples_per_second": 621.09, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 9.3, | |
| "eval_average_metrics": 52.55846445665645, | |
| "eval_loss": 0.22525694966316223, | |
| "eval_matthews_correlation": 52.55846445665645, | |
| "eval_runtime": 1.5205, | |
| "eval_samples_per_second": 342.658, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 11.63, | |
| "learning_rate": 0.00017704918032786883, | |
| "loss": 0.1859, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 11.63, | |
| "eval_average_metrics": 50.296653045050874, | |
| "eval_loss": 0.23001490533351898, | |
| "eval_matthews_correlation": 50.296653045050874, | |
| "eval_runtime": 1.2989, | |
| "eval_samples_per_second": 401.107, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 13.95, | |
| "eval_average_metrics": 51.11915736464516, | |
| "eval_loss": 0.22642329335212708, | |
| "eval_matthews_correlation": 51.11915736464516, | |
| "eval_runtime": 1.2231, | |
| "eval_samples_per_second": 425.968, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 16.28, | |
| "eval_average_metrics": 50.87277150375961, | |
| "eval_loss": 0.22626039385795593, | |
| "eval_matthews_correlation": 50.87277150375961, | |
| "eval_runtime": 1.1916, | |
| "eval_samples_per_second": 437.245, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 17.44, | |
| "learning_rate": 5.40983606557377e-05, | |
| "loss": 0.1748, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 18.6, | |
| "eval_average_metrics": 52.035569173764515, | |
| "eval_loss": 0.22766831517219543, | |
| "eval_matthews_correlation": 52.035569173764515, | |
| "eval_runtime": 1.2214, | |
| "eval_samples_per_second": 426.575, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "step": 1720, | |
| "total_flos": 6646762904679216.0, | |
| "train_loss": 0.5282898814179177, | |
| "train_runtime": 422.8779, | |
| "train_samples_per_second": 404.419, | |
| "train_steps_per_second": 4.067 | |
| } | |
| ], | |
| "max_steps": 1720, | |
| "num_train_epochs": 20, | |
| "total_flos": 6646762904679216.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |