| { | |
| "best_metric": 0.7896053362979433, | |
| "best_model_checkpoint": "tinybert-TG-HS-HX-parentpretrained\\run-6\\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": 1.0611700032566007e-05, | |
| "loss": 0.1259, | |
| "step": 197 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.7536131183991106, | |
| "eval_loss": 0.12161022424697876, | |
| "eval_runtime": 125.6824, | |
| "eval_samples_per_second": 57.255, | |
| "eval_steps_per_second": 0.454, | |
| "step": 197 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 9.095742885056578e-06, | |
| "loss": 0.1236, | |
| "step": 394 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.7745969983324069, | |
| "eval_loss": 0.12096704542636871, | |
| "eval_runtime": 125.4944, | |
| "eval_samples_per_second": 57.341, | |
| "eval_steps_per_second": 0.454, | |
| "step": 394 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 7.579785737547148e-06, | |
| "loss": 0.1229, | |
| "step": 591 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.7848804891606448, | |
| "eval_loss": 0.12066732347011566, | |
| "eval_runtime": 125.5947, | |
| "eval_samples_per_second": 57.295, | |
| "eval_steps_per_second": 0.454, | |
| "step": 591 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 6.063828590037718e-06, | |
| "loss": 0.1224, | |
| "step": 788 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.7804335742078933, | |
| "eval_loss": 0.12036041915416718, | |
| "eval_runtime": 125.5933, | |
| "eval_samples_per_second": 57.296, | |
| "eval_steps_per_second": 0.454, | |
| "step": 788 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 4.547871442528289e-06, | |
| "loss": 0.1221, | |
| "step": 985 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.7887715397443024, | |
| "eval_loss": 0.12023670971393585, | |
| "eval_runtime": 125.4555, | |
| "eval_samples_per_second": 57.359, | |
| "eval_steps_per_second": 0.454, | |
| "step": 985 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 3.031914295018859e-06, | |
| "loss": 0.1218, | |
| "step": 1182 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.7896053362979433, | |
| "eval_loss": 0.12014192342758179, | |
| "eval_runtime": 125.3237, | |
| "eval_samples_per_second": 57.419, | |
| "eval_steps_per_second": 0.455, | |
| "step": 1182 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 1576, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 8, | |
| "save_steps": 500, | |
| "total_flos": 265501203931764.0, | |
| "trial_name": null, | |
| "trial_params": { | |
| "alpha": 0.18717092492624732, | |
| "learning_rate": 1.2127657180075437e-05, | |
| "num_train_epochs": 8, | |
| "temperature": 5 | |
| } | |
| } | |