| { | |
| "best_metric": 0.8246247915508615, | |
| "best_model_checkpoint": "tinybert-TG-HS-HX-parentpretrained\\run-3\\checkpoint-985", | |
| "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": 4.53412212955664e-05, | |
| "loss": 0.2002, | |
| "step": 197 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.796692607003891, | |
| "eval_loss": 0.19448761641979218, | |
| "eval_runtime": 125.5379, | |
| "eval_samples_per_second": 57.321, | |
| "eval_steps_per_second": 0.454, | |
| "step": 197 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 3.778435107963867e-05, | |
| "loss": 0.1946, | |
| "step": 394 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8086436909394108, | |
| "eval_loss": 0.19145084917545319, | |
| "eval_runtime": 125.3349, | |
| "eval_samples_per_second": 57.414, | |
| "eval_steps_per_second": 0.455, | |
| "step": 394 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 3.0227480863710935e-05, | |
| "loss": 0.1912, | |
| "step": 591 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8153140633685381, | |
| "eval_loss": 0.1914054900407791, | |
| "eval_runtime": 125.3098, | |
| "eval_samples_per_second": 57.426, | |
| "eval_steps_per_second": 0.455, | |
| "step": 591 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 2.26706106477832e-05, | |
| "loss": 0.189, | |
| "step": 788 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8187882156753752, | |
| "eval_loss": 0.1904531717300415, | |
| "eval_runtime": 125.5411, | |
| "eval_samples_per_second": 57.32, | |
| "eval_steps_per_second": 0.454, | |
| "step": 788 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 1.5113740431855467e-05, | |
| "loss": 0.1871, | |
| "step": 985 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.8246247915508615, | |
| "eval_loss": 0.19075776636600494, | |
| "eval_runtime": 125.5068, | |
| "eval_samples_per_second": 57.336, | |
| "eval_steps_per_second": 0.454, | |
| "step": 985 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 7.556870215927734e-06, | |
| "loss": 0.1853, | |
| "step": 1182 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.8246247915508615, | |
| "eval_loss": 0.1908973753452301, | |
| "eval_runtime": 125.4292, | |
| "eval_samples_per_second": 57.371, | |
| "eval_steps_per_second": 0.454, | |
| "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": 0.3140746400350408, | |
| "learning_rate": 5.2898091511494136e-05, | |
| "num_train_epochs": 7, | |
| "temperature": 24 | |
| } | |
| } | |