| { | |
| "best_metric": 0.8337155963302753, | |
| "best_model_checkpoint": "tiny-bert-sst2-distilled/run-18/checkpoint-2108", | |
| "epoch": 5.0, | |
| "eval_steps": 500, | |
| "global_step": 2635, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 0.00014893639687145338, | |
| "loss": 0.8667, | |
| "step": 527 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.8165137614678899, | |
| "eval_loss": 0.8296506404876709, | |
| "eval_runtime": 6.9326, | |
| "eval_samples_per_second": 125.782, | |
| "eval_steps_per_second": 1.01, | |
| "step": 527 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 0.00013031934726252174, | |
| "loss": 0.4473, | |
| "step": 1054 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8256880733944955, | |
| "eval_loss": 0.9700397849082947, | |
| "eval_runtime": 7.0917, | |
| "eval_samples_per_second": 122.961, | |
| "eval_steps_per_second": 0.987, | |
| "step": 1054 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 0.00011170229765359004, | |
| "loss": 0.3367, | |
| "step": 1581 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8302752293577982, | |
| "eval_loss": 0.971959114074707, | |
| "eval_runtime": 7.4513, | |
| "eval_samples_per_second": 117.027, | |
| "eval_steps_per_second": 0.939, | |
| "step": 1581 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 9.308524804465839e-05, | |
| "loss": 0.2772, | |
| "step": 2108 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8337155963302753, | |
| "eval_loss": 0.9656078219413757, | |
| "eval_runtime": 7.1382, | |
| "eval_samples_per_second": 122.16, | |
| "eval_steps_per_second": 0.981, | |
| "step": 2108 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 7.446819843572669e-05, | |
| "loss": 0.2345, | |
| "step": 2635 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.823394495412844, | |
| "eval_loss": 1.0300672054290771, | |
| "eval_runtime": 7.485, | |
| "eval_samples_per_second": 116.5, | |
| "eval_steps_per_second": 0.935, | |
| "step": 2635 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 4743, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 9, | |
| "save_steps": 500, | |
| "total_flos": 40436714110320.0, | |
| "train_batch_size": 128, | |
| "trial_name": null, | |
| "trial_params": { | |
| "alpha": 0.7585638468345914, | |
| "learning_rate": 0.00016755344648038508, | |
| "num_train_epochs": 9, | |
| "temperature": 8 | |
| } | |
| } | |