| { |
| "best_global_step": 102, |
| "best_metric": 0.6190451383590698, |
| "best_model_checkpoint": "tiny_bert_km_20_v2_cola/checkpoint-102", |
| "epoch": 8.0, |
| "eval_steps": 500, |
| "global_step": 272, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 1.0, |
| "grad_norm": 0.6262004971504211, |
| "learning_rate": 4.9e-05, |
| "loss": 0.6209, |
| "step": 34 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_accuracy": 0.6912751793861389, |
| "eval_loss": 0.6208353042602539, |
| "eval_matthews_correlation": 0.0, |
| "eval_runtime": 0.339, |
| "eval_samples_per_second": 3076.545, |
| "eval_steps_per_second": 14.749, |
| "step": 34 |
| }, |
| { |
| "epoch": 2.0, |
| "grad_norm": 0.9489756226539612, |
| "learning_rate": 4.8e-05, |
| "loss": 0.6054, |
| "step": 68 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_accuracy": 0.6912751793861389, |
| "eval_loss": 0.6204327344894409, |
| "eval_matthews_correlation": 0.0, |
| "eval_runtime": 0.325, |
| "eval_samples_per_second": 3209.635, |
| "eval_steps_per_second": 15.387, |
| "step": 68 |
| }, |
| { |
| "epoch": 3.0, |
| "grad_norm": 1.023315668106079, |
| "learning_rate": 4.7e-05, |
| "loss": 0.5937, |
| "step": 102 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_accuracy": 0.6912751793861389, |
| "eval_loss": 0.6190451383590698, |
| "eval_matthews_correlation": 0.0, |
| "eval_runtime": 0.36, |
| "eval_samples_per_second": 2897.503, |
| "eval_steps_per_second": 13.89, |
| "step": 102 |
| }, |
| { |
| "epoch": 4.0, |
| "grad_norm": 1.2687016725540161, |
| "learning_rate": 4.600000000000001e-05, |
| "loss": 0.5732, |
| "step": 136 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_accuracy": 0.6423777341842651, |
| "eval_loss": 0.6476646065711975, |
| "eval_matthews_correlation": 0.0283653964710585, |
| "eval_runtime": 0.3402, |
| "eval_samples_per_second": 3065.737, |
| "eval_steps_per_second": 14.697, |
| "step": 136 |
| }, |
| { |
| "epoch": 5.0, |
| "grad_norm": 1.3729408979415894, |
| "learning_rate": 4.5e-05, |
| "loss": 0.5393, |
| "step": 170 |
| }, |
| { |
| "epoch": 5.0, |
| "eval_accuracy": 0.6730585098266602, |
| "eval_loss": 0.6414644122123718, |
| "eval_matthews_correlation": 0.06036235184714196, |
| "eval_runtime": 0.3303, |
| "eval_samples_per_second": 3157.942, |
| "eval_steps_per_second": 15.139, |
| "step": 170 |
| }, |
| { |
| "epoch": 6.0, |
| "grad_norm": 1.6535099744796753, |
| "learning_rate": 4.4000000000000006e-05, |
| "loss": 0.4907, |
| "step": 204 |
| }, |
| { |
| "epoch": 6.0, |
| "eval_accuracy": 0.6644295454025269, |
| "eval_loss": 0.6778461337089539, |
| "eval_matthews_correlation": 0.07404272693058336, |
| "eval_runtime": 0.3186, |
| "eval_samples_per_second": 3273.218, |
| "eval_steps_per_second": 15.691, |
| "step": 204 |
| }, |
| { |
| "epoch": 7.0, |
| "grad_norm": 1.7401143312454224, |
| "learning_rate": 4.3e-05, |
| "loss": 0.4491, |
| "step": 238 |
| }, |
| { |
| "epoch": 7.0, |
| "eval_accuracy": 0.6490891575813293, |
| "eval_loss": 0.7494480609893799, |
| "eval_matthews_correlation": 0.06645824521377343, |
| "eval_runtime": 0.3473, |
| "eval_samples_per_second": 3003.076, |
| "eval_steps_per_second": 14.396, |
| "step": 238 |
| }, |
| { |
| "epoch": 8.0, |
| "grad_norm": 2.450899362564087, |
| "learning_rate": 4.2e-05, |
| "loss": 0.4125, |
| "step": 272 |
| }, |
| { |
| "epoch": 8.0, |
| "eval_accuracy": 0.6088207364082336, |
| "eval_loss": 0.7905942797660828, |
| "eval_matthews_correlation": 0.09910073524633717, |
| "eval_runtime": 0.3109, |
| "eval_samples_per_second": 3354.746, |
| "eval_steps_per_second": 16.082, |
| "step": 272 |
| }, |
| { |
| "epoch": 8.0, |
| "step": 272, |
| "total_flos": 1793902531485696.0, |
| "train_loss": 0.5355994981877944, |
| "train_runtime": 45.1604, |
| "train_samples_per_second": 9467.364, |
| "train_steps_per_second": 37.644 |
| } |
| ], |
| "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": 1793902531485696.0, |
| "train_batch_size": 256, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
