| { |
| "best_global_step": 30, |
| "best_metric": 0.6078192591667175, |
| "best_model_checkpoint": "bert_base_km_20_v2_mrpc/checkpoint-30", |
| "epoch": 7.0, |
| "eval_steps": 500, |
| "global_step": 105, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 1.0, |
| "grad_norm": 1.2124764919281006, |
| "learning_rate": 4.9e-05, |
| "loss": 0.6259, |
| "step": 15 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_accuracy": 0.6838235294117647, |
| "eval_combined_score": 0.740016030108726, |
| "eval_f1": 0.7962085308056872, |
| "eval_loss": 0.616814136505127, |
| "eval_runtime": 0.2744, |
| "eval_samples_per_second": 1486.649, |
| "eval_steps_per_second": 7.287, |
| "step": 15 |
| }, |
| { |
| "epoch": 2.0, |
| "grad_norm": 1.5705448389053345, |
| "learning_rate": 4.8e-05, |
| "loss": 0.5681, |
| "step": 30 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_accuracy": 0.7034313725490197, |
| "eval_combined_score": 0.7593492740607694, |
| "eval_f1": 0.815267175572519, |
| "eval_loss": 0.6078192591667175, |
| "eval_runtime": 0.2706, |
| "eval_samples_per_second": 1507.857, |
| "eval_steps_per_second": 7.391, |
| "step": 30 |
| }, |
| { |
| "epoch": 3.0, |
| "grad_norm": 1.5428181886672974, |
| "learning_rate": 4.7e-05, |
| "loss": 0.4951, |
| "step": 45 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_accuracy": 0.7009803921568627, |
| "eval_combined_score": 0.7530461066215624, |
| "eval_f1": 0.805111821086262, |
| "eval_loss": 0.6499011516571045, |
| "eval_runtime": 0.2726, |
| "eval_samples_per_second": 1496.591, |
| "eval_steps_per_second": 7.336, |
| "step": 45 |
| }, |
| { |
| "epoch": 4.0, |
| "grad_norm": 2.37738037109375, |
| "learning_rate": 4.600000000000001e-05, |
| "loss": 0.3946, |
| "step": 60 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_accuracy": 0.6936274509803921, |
| "eval_combined_score": 0.7396096088521171, |
| "eval_f1": 0.7855917667238422, |
| "eval_loss": 0.6939632892608643, |
| "eval_runtime": 0.2946, |
| "eval_samples_per_second": 1384.696, |
| "eval_steps_per_second": 6.788, |
| "step": 60 |
| }, |
| { |
| "epoch": 5.0, |
| "grad_norm": 4.476428031921387, |
| "learning_rate": 4.5e-05, |
| "loss": 0.2527, |
| "step": 75 |
| }, |
| { |
| "epoch": 5.0, |
| "eval_accuracy": 0.6397058823529411, |
| "eval_combined_score": 0.6798529411764705, |
| "eval_f1": 0.72, |
| "eval_loss": 0.876370906829834, |
| "eval_runtime": 0.2767, |
| "eval_samples_per_second": 1474.695, |
| "eval_steps_per_second": 7.229, |
| "step": 75 |
| }, |
| { |
| "epoch": 6.0, |
| "grad_norm": 2.77229380607605, |
| "learning_rate": 4.4000000000000006e-05, |
| "loss": 0.1245, |
| "step": 90 |
| }, |
| { |
| "epoch": 6.0, |
| "eval_accuracy": 0.6666666666666666, |
| "eval_combined_score": 0.7140350877192982, |
| "eval_f1": 0.7614035087719299, |
| "eval_loss": 1.1046664714813232, |
| "eval_runtime": 0.2943, |
| "eval_samples_per_second": 1386.409, |
| "eval_steps_per_second": 6.796, |
| "step": 90 |
| }, |
| { |
| "epoch": 7.0, |
| "grad_norm": 3.0406038761138916, |
| "learning_rate": 4.3e-05, |
| "loss": 0.06, |
| "step": 105 |
| }, |
| { |
| "epoch": 7.0, |
| "eval_accuracy": 0.6862745098039216, |
| "eval_combined_score": 0.7327924273157539, |
| "eval_f1": 0.7793103448275862, |
| "eval_loss": 1.2587286233901978, |
| "eval_runtime": 0.2836, |
| "eval_samples_per_second": 1438.785, |
| "eval_steps_per_second": 7.053, |
| "step": 105 |
| }, |
| { |
| "epoch": 7.0, |
| "step": 105, |
| "total_flos": 3377819728711680.0, |
| "train_loss": 0.36015133233297436, |
| "train_runtime": 59.2376, |
| "train_samples_per_second": 3096.005, |
| "train_steps_per_second": 12.661 |
| } |
| ], |
| "logging_steps": 1, |
| "max_steps": 750, |
| "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": 3377819728711680.0, |
| "train_batch_size": 256, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
