| { |
| "best_metric": 84.05797101449275, |
| "best_model_checkpoint": "outputs/bitfit/t5-base/rte/checkpoint-300", |
| "epoch": 20.0, |
| "global_step": 1560, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 1.28, |
| "eval_accuracy": 76.81159420289855, |
| "eval_average_metrics": 76.81159420289855, |
| "eval_loss": 0.21982769668102264, |
| "eval_runtime": 1.0149, |
| "eval_samples_per_second": 135.972, |
| "step": 100 |
| }, |
| { |
| "epoch": 2.56, |
| "eval_accuracy": 78.26086956521739, |
| "eval_average_metrics": 78.26086956521739, |
| "eval_loss": 0.18932242691516876, |
| "eval_runtime": 0.6715, |
| "eval_samples_per_second": 205.52, |
| "step": 200 |
| }, |
| { |
| "epoch": 3.85, |
| "eval_accuracy": 84.05797101449275, |
| "eval_average_metrics": 84.05797101449275, |
| "eval_loss": 0.16963252425193787, |
| "eval_runtime": 0.8653, |
| "eval_samples_per_second": 159.484, |
| "step": 300 |
| }, |
| { |
| "epoch": 5.13, |
| "eval_accuracy": 80.43478260869566, |
| "eval_average_metrics": 80.43478260869566, |
| "eval_loss": 0.18479692935943604, |
| "eval_runtime": 0.7171, |
| "eval_samples_per_second": 192.431, |
| "step": 400 |
| }, |
| { |
| "epoch": 6.41, |
| "learning_rate": 0.00020384615384615385, |
| "loss": 0.2327, |
| "step": 500 |
| }, |
| { |
| "epoch": 6.41, |
| "eval_accuracy": 81.88405797101449, |
| "eval_average_metrics": 81.88405797101449, |
| "eval_loss": 0.23633617162704468, |
| "eval_runtime": 0.6787, |
| "eval_samples_per_second": 203.343, |
| "step": 500 |
| }, |
| { |
| "epoch": 7.69, |
| "eval_accuracy": 79.71014492753623, |
| "eval_average_metrics": 79.71014492753623, |
| "eval_loss": 0.23751917481422424, |
| "eval_runtime": 0.8803, |
| "eval_samples_per_second": 156.767, |
| "step": 600 |
| }, |
| { |
| "epoch": 8.97, |
| "eval_accuracy": 84.05797101449275, |
| "eval_average_metrics": 84.05797101449275, |
| "eval_loss": 0.23690848052501678, |
| "eval_runtime": 0.5629, |
| "eval_samples_per_second": 245.142, |
| "step": 700 |
| }, |
| { |
| "epoch": 10.26, |
| "eval_accuracy": 83.33333333333334, |
| "eval_average_metrics": 83.33333333333334, |
| "eval_loss": 0.2846378684043884, |
| "eval_runtime": 0.9472, |
| "eval_samples_per_second": 145.689, |
| "step": 800 |
| }, |
| { |
| "epoch": 11.54, |
| "eval_accuracy": 82.6086956521739, |
| "eval_average_metrics": 82.6086956521739, |
| "eval_loss": 0.2770601212978363, |
| "eval_runtime": 0.8799, |
| "eval_samples_per_second": 156.839, |
| "step": 900 |
| }, |
| { |
| "epoch": 12.82, |
| "learning_rate": 0.00010769230769230768, |
| "loss": 0.0774, |
| "step": 1000 |
| }, |
| { |
| "epoch": 12.82, |
| "eval_accuracy": 82.6086956521739, |
| "eval_average_metrics": 82.6086956521739, |
| "eval_loss": 0.3579100966453552, |
| "eval_runtime": 0.9343, |
| "eval_samples_per_second": 147.71, |
| "step": 1000 |
| }, |
| { |
| "epoch": 14.1, |
| "eval_accuracy": 81.15942028985508, |
| "eval_average_metrics": 81.15942028985508, |
| "eval_loss": 0.4031548798084259, |
| "eval_runtime": 0.6763, |
| "eval_samples_per_second": 204.046, |
| "step": 1100 |
| }, |
| { |
| "epoch": 15.38, |
| "eval_accuracy": 81.15942028985508, |
| "eval_average_metrics": 81.15942028985508, |
| "eval_loss": 0.4019993841648102, |
| "eval_runtime": 0.9288, |
| "eval_samples_per_second": 148.583, |
| "step": 1200 |
| }, |
| { |
| "epoch": 16.67, |
| "eval_accuracy": 78.98550724637681, |
| "eval_average_metrics": 78.98550724637681, |
| "eval_loss": 0.47033587098121643, |
| "eval_runtime": 0.6015, |
| "eval_samples_per_second": 229.417, |
| "step": 1300 |
| }, |
| { |
| "epoch": 17.95, |
| "eval_accuracy": 81.88405797101449, |
| "eval_average_metrics": 81.88405797101449, |
| "eval_loss": 0.4693973958492279, |
| "eval_runtime": 0.8377, |
| "eval_samples_per_second": 164.743, |
| "step": 1400 |
| }, |
| { |
| "epoch": 19.23, |
| "learning_rate": 1.1538461538461538e-05, |
| "loss": 0.0312, |
| "step": 1500 |
| }, |
| { |
| "epoch": 19.23, |
| "eval_accuracy": 79.71014492753623, |
| "eval_average_metrics": 79.71014492753623, |
| "eval_loss": 0.5053138136863708, |
| "eval_runtime": 1.053, |
| "eval_samples_per_second": 131.051, |
| "step": 1500 |
| }, |
| { |
| "epoch": 20.0, |
| "step": 1560, |
| "total_flos": 7649465032433664.0, |
| "train_loss": 0.1103338890350782, |
| "train_runtime": 529.9082, |
| "train_samples_per_second": 93.979, |
| "train_steps_per_second": 2.944 |
| } |
| ], |
| "max_steps": 1560, |
| "num_train_epochs": 20, |
| "total_flos": 7649465032433664.0, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
