| { |
| "best_global_step": 4266, |
| "best_metric": 0.41152405738830566, |
| "best_model_checkpoint": "tiny_bert_km_100_v2_qqp/checkpoint-4266", |
| "epoch": 8.0, |
| "eval_steps": 500, |
| "global_step": 11376, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 1.0, |
| "grad_norm": 1.5886011123657227, |
| "learning_rate": 4.9e-05, |
| "loss": 0.5003, |
| "step": 1422 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_accuracy": 0.7805342567400445, |
| "eval_combined_score": 0.7278828818197496, |
| "eval_f1": 0.6752315068994547, |
| "eval_loss": 0.4545815587043762, |
| "eval_runtime": 11.6701, |
| "eval_samples_per_second": 3464.394, |
| "eval_steps_per_second": 13.539, |
| "step": 1422 |
| }, |
| { |
| "epoch": 2.0, |
| "grad_norm": 2.2501890659332275, |
| "learning_rate": 4.8e-05, |
| "loss": 0.413, |
| "step": 2844 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_accuracy": 0.8009151620084096, |
| "eval_combined_score": 0.7596275428616426, |
| "eval_f1": 0.7183399237148756, |
| "eval_loss": 0.4230986535549164, |
| "eval_runtime": 11.4578, |
| "eval_samples_per_second": 3528.59, |
| "eval_steps_per_second": 13.79, |
| "step": 2844 |
| }, |
| { |
| "epoch": 3.0, |
| "grad_norm": 3.183741807937622, |
| "learning_rate": 4.7e-05, |
| "loss": 0.3574, |
| "step": 4266 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_accuracy": 0.8119218402176601, |
| "eval_combined_score": 0.7730424471679826, |
| "eval_f1": 0.7341630541183052, |
| "eval_loss": 0.41152405738830566, |
| "eval_runtime": 11.4586, |
| "eval_samples_per_second": 3528.351, |
| "eval_steps_per_second": 13.789, |
| "step": 4266 |
| }, |
| { |
| "epoch": 4.0, |
| "grad_norm": 3.2916994094848633, |
| "learning_rate": 4.600000000000001e-05, |
| "loss": 0.3115, |
| "step": 5688 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_accuracy": 0.815730892901311, |
| "eval_combined_score": 0.7747822024349356, |
| "eval_f1": 0.7338335119685602, |
| "eval_loss": 0.43493619561195374, |
| "eval_runtime": 11.4616, |
| "eval_samples_per_second": 3527.418, |
| "eval_steps_per_second": 13.785, |
| "step": 5688 |
| }, |
| { |
| "epoch": 5.0, |
| "grad_norm": 2.382847785949707, |
| "learning_rate": 4.5e-05, |
| "loss": 0.2734, |
| "step": 7110 |
| }, |
| { |
| "epoch": 5.0, |
| "eval_accuracy": 0.8201582982933465, |
| "eval_combined_score": 0.7860133538356129, |
| "eval_f1": 0.7518684093778794, |
| "eval_loss": 0.4389333128929138, |
| "eval_runtime": 11.436, |
| "eval_samples_per_second": 3535.321, |
| "eval_steps_per_second": 13.816, |
| "step": 7110 |
| }, |
| { |
| "epoch": 6.0, |
| "grad_norm": 2.9211905002593994, |
| "learning_rate": 4.4000000000000006e-05, |
| "loss": 0.2411, |
| "step": 8532 |
| }, |
| { |
| "epoch": 6.0, |
| "eval_accuracy": 0.8221864951768488, |
| "eval_combined_score": 0.7906220205883909, |
| "eval_f1": 0.7590575459999329, |
| "eval_loss": 0.4614395797252655, |
| "eval_runtime": 11.4342, |
| "eval_samples_per_second": 3535.872, |
| "eval_steps_per_second": 13.818, |
| "step": 8532 |
| }, |
| { |
| "epoch": 7.0, |
| "grad_norm": 3.270599126815796, |
| "learning_rate": 4.3e-05, |
| "loss": 0.2126, |
| "step": 9954 |
| }, |
| { |
| "epoch": 7.0, |
| "eval_accuracy": 0.8209992579767499, |
| "eval_combined_score": 0.7892971503415821, |
| "eval_f1": 0.7575950427064143, |
| "eval_loss": 0.4831831157207489, |
| "eval_runtime": 11.4554, |
| "eval_samples_per_second": 3529.348, |
| "eval_steps_per_second": 13.793, |
| "step": 9954 |
| }, |
| { |
| "epoch": 8.0, |
| "grad_norm": 2.8055524826049805, |
| "learning_rate": 4.2e-05, |
| "loss": 0.1876, |
| "step": 11376 |
| }, |
| { |
| "epoch": 8.0, |
| "eval_accuracy": 0.8208508533267376, |
| "eval_combined_score": 0.7897690881659508, |
| "eval_f1": 0.7586873230051641, |
| "eval_loss": 0.5212997198104858, |
| "eval_runtime": 11.4485, |
| "eval_samples_per_second": 3531.467, |
| "eval_steps_per_second": 13.801, |
| "step": 11376 |
| }, |
| { |
| "epoch": 8.0, |
| "step": 11376, |
| "total_flos": 7.633075201391002e+16, |
| "train_loss": 0.312113081855613, |
| "train_runtime": 1604.4109, |
| "train_samples_per_second": 11338.929, |
| "train_steps_per_second": 44.315 |
| } |
| ], |
| "logging_steps": 1, |
| "max_steps": 71100, |
| "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": 7.633075201391002e+16, |
| "train_batch_size": 256, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
