| { | |
| "best_metric": 0.8245412844036697, | |
| "best_model_checkpoint": "tiny-bert-sst2-distilled/run-2/checkpoint-527", | |
| "epoch": 4.0, | |
| "eval_steps": 500, | |
| "global_step": 2108, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 18.61258888244629, | |
| "learning_rate": 0.00029723759979073097, | |
| "loss": 1.527, | |
| "step": 527 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.8245412844036697, | |
| "eval_loss": 1.6911760568618774, | |
| "eval_runtime": 2.6073, | |
| "eval_samples_per_second": 334.44, | |
| "eval_steps_per_second": 2.685, | |
| "step": 527 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 6.284183979034424, | |
| "learning_rate": 0.00019815839986048731, | |
| "loss": 0.67, | |
| "step": 1054 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8245412844036697, | |
| "eval_loss": 1.8715265989303589, | |
| "eval_runtime": 2.6081, | |
| "eval_samples_per_second": 334.343, | |
| "eval_steps_per_second": 2.684, | |
| "step": 1054 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 15.639009475708008, | |
| "learning_rate": 9.907919993024366e-05, | |
| "loss": 0.4531, | |
| "step": 1581 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8176605504587156, | |
| "eval_loss": 1.803161859512329, | |
| "eval_runtime": 2.5962, | |
| "eval_samples_per_second": 335.871, | |
| "eval_steps_per_second": 2.696, | |
| "step": 1581 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 16.141427993774414, | |
| "learning_rate": 0.0, | |
| "loss": 0.3446, | |
| "step": 2108 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8153669724770642, | |
| "eval_loss": 1.8865516185760498, | |
| "eval_runtime": 2.6209, | |
| "eval_samples_per_second": 332.708, | |
| "eval_steps_per_second": 2.671, | |
| "step": 2108 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 2108, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 4, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 32337743850420.0, | |
| "train_batch_size": 128, | |
| "trial_name": null, | |
| "trial_params": { | |
| "alpha": 0.22408062690203823, | |
| "learning_rate": 0.00039631679972097463, | |
| "num_train_epochs": 4, | |
| "temperature": 30 | |
| } | |
| } | |