| { | |
| "best_metric": 0.7864091161756531, | |
| "best_model_checkpoint": "tinybert-TG-HS-HX-parentpretrained\\run-5\\checkpoint-1182", | |
| "epoch": 6.0, | |
| "eval_steps": 500, | |
| "global_step": 1182, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 4.941433573951871e-06, | |
| "loss": 0.3472, | |
| "step": 197 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.7540300166759311, | |
| "eval_loss": 0.3264233469963074, | |
| "eval_runtime": 125.6102, | |
| "eval_samples_per_second": 57.288, | |
| "eval_steps_per_second": 0.454, | |
| "step": 197 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 4.117861311626559e-06, | |
| "loss": 0.3304, | |
| "step": 394 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.7727904391328516, | |
| "eval_loss": 0.3193582594394684, | |
| "eval_runtime": 125.5179, | |
| "eval_samples_per_second": 57.33, | |
| "eval_steps_per_second": 0.454, | |
| "step": 394 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 3.2942890493012468e-06, | |
| "loss": 0.3247, | |
| "step": 591 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.7782101167315175, | |
| "eval_loss": 0.31648218631744385, | |
| "eval_runtime": 125.7064, | |
| "eval_samples_per_second": 57.244, | |
| "eval_steps_per_second": 0.453, | |
| "step": 591 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 2.4707167869759353e-06, | |
| "loss": 0.3224, | |
| "step": 788 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.7812673707615342, | |
| "eval_loss": 0.31448855996131897, | |
| "eval_runtime": 125.4689, | |
| "eval_samples_per_second": 57.353, | |
| "eval_steps_per_second": 0.454, | |
| "step": 788 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 1.6471445246506234e-06, | |
| "loss": 0.3204, | |
| "step": 985 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.7847415230683713, | |
| "eval_loss": 0.31334513425827026, | |
| "eval_runtime": 125.4012, | |
| "eval_samples_per_second": 57.384, | |
| "eval_steps_per_second": 0.455, | |
| "step": 985 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 8.235722623253117e-07, | |
| "loss": 0.3185, | |
| "step": 1182 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.7864091161756531, | |
| "eval_loss": 0.3128955364227295, | |
| "eval_runtime": 125.3008, | |
| "eval_samples_per_second": 57.43, | |
| "eval_steps_per_second": 0.455, | |
| "step": 1182 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 1379, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 7, | |
| "save_steps": 500, | |
| "total_flos": 265501203931764.0, | |
| "trial_name": null, | |
| "trial_params": { | |
| "alpha": 0.5501861057464915, | |
| "learning_rate": 5.7650058362771825e-06, | |
| "num_train_epochs": 7, | |
| "temperature": 15 | |
| } | |
| } | |