| { |
| "best_global_step": 820, |
| "best_metric": 0.6279211044311523, |
| "best_model_checkpoint": "bert_base_km_10_v1_qnli/checkpoint-820", |
| "epoch": 7.0, |
| "eval_steps": 500, |
| "global_step": 2870, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 1.0, |
| "grad_norm": 2.2646937370300293, |
| "learning_rate": 4.9e-05, |
| "loss": 0.6642, |
| "step": 410 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_accuracy": 0.6395753249130515, |
| "eval_loss": 0.6345698833465576, |
| "eval_runtime": 3.4904, |
| "eval_samples_per_second": 1565.133, |
| "eval_steps_per_second": 6.303, |
| "step": 410 |
| }, |
| { |
| "epoch": 2.0, |
| "grad_norm": 2.06400728225708, |
| "learning_rate": 4.8e-05, |
| "loss": 0.6203, |
| "step": 820 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_accuracy": 0.6498261028738789, |
| "eval_loss": 0.6279211044311523, |
| "eval_runtime": 3.4233, |
| "eval_samples_per_second": 1595.845, |
| "eval_steps_per_second": 6.427, |
| "step": 820 |
| }, |
| { |
| "epoch": 3.0, |
| "grad_norm": 3.729912281036377, |
| "learning_rate": 4.7e-05, |
| "loss": 0.5422, |
| "step": 1230 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_accuracy": 0.6304228445908842, |
| "eval_loss": 0.6750292181968689, |
| "eval_runtime": 3.4493, |
| "eval_samples_per_second": 1583.801, |
| "eval_steps_per_second": 6.378, |
| "step": 1230 |
| }, |
| { |
| "epoch": 4.0, |
| "grad_norm": 6.5136637687683105, |
| "learning_rate": 4.600000000000001e-05, |
| "loss": 0.422, |
| "step": 1640 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_accuracy": 0.6467142595643419, |
| "eval_loss": 0.7332958579063416, |
| "eval_runtime": 3.4349, |
| "eval_samples_per_second": 1590.425, |
| "eval_steps_per_second": 6.405, |
| "step": 1640 |
| }, |
| { |
| "epoch": 5.0, |
| "grad_norm": 6.036999702453613, |
| "learning_rate": 4.5e-05, |
| "loss": 0.2952, |
| "step": 2050 |
| }, |
| { |
| "epoch": 5.0, |
| "eval_accuracy": 0.6551345414607359, |
| "eval_loss": 0.8971741199493408, |
| "eval_runtime": 3.4347, |
| "eval_samples_per_second": 1590.509, |
| "eval_steps_per_second": 6.405, |
| "step": 2050 |
| }, |
| { |
| "epoch": 6.0, |
| "grad_norm": 8.021710395812988, |
| "learning_rate": 4.4000000000000006e-05, |
| "loss": 0.1998, |
| "step": 2460 |
| }, |
| { |
| "epoch": 6.0, |
| "eval_accuracy": 0.6445176642870217, |
| "eval_loss": 1.0509867668151855, |
| "eval_runtime": 3.4393, |
| "eval_samples_per_second": 1588.41, |
| "eval_steps_per_second": 6.397, |
| "step": 2460 |
| }, |
| { |
| "epoch": 7.0, |
| "grad_norm": 11.778338432312012, |
| "learning_rate": 4.3e-05, |
| "loss": 0.1454, |
| "step": 2870 |
| }, |
| { |
| "epoch": 7.0, |
| "eval_accuracy": 0.6468973091707853, |
| "eval_loss": 1.2517988681793213, |
| "eval_runtime": 3.434, |
| "eval_samples_per_second": 1590.876, |
| "eval_steps_per_second": 6.407, |
| "step": 2870 |
| }, |
| { |
| "epoch": 7.0, |
| "step": 2870, |
| "total_flos": 9.645664445050368e+16, |
| "train_loss": 0.4127295610381336, |
| "train_runtime": 1216.6029, |
| "train_samples_per_second": 4304.732, |
| "train_steps_per_second": 16.85 |
| } |
| ], |
| "logging_steps": 1, |
| "max_steps": 20500, |
| "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": 9.645664445050368e+16, |
| "train_batch_size": 256, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
