{ | |
"best_metric": 0.5270758122743683, | |
"best_model_checkpoint": "microsoft/deberta-v3-smallsed_rte-finetuned-rte/run-3/checkpoint-32", | |
"epoch": 1.0, | |
"eval_steps": 500, | |
"global_step": 32, | |
"is_hyper_param_search": true, | |
"is_local_process_zero": true, | |
"is_world_process_zero": true, | |
"log_history": [ | |
{ | |
"epoch": 1.0, | |
"eval_accuracy": 0.5270758122743683, | |
"eval_loss": 0.693800151348114, | |
"eval_runtime": 1.7818, | |
"eval_samples_per_second": 155.458, | |
"eval_steps_per_second": 10.102, | |
"step": 32 | |
} | |
], | |
"logging_steps": 500, | |
"max_steps": 32, | |
"num_input_tokens_seen": 0, | |
"num_train_epochs": 1, | |
"save_steps": 500, | |
"total_flos": 0, | |
"train_batch_size": 8, | |
"trial_name": null, | |
"trial_params": { | |
"learning_rate": 4.078702802904116e-05, | |
"num_train_epochs": 1, | |
"per_device_train_batch_size": 8, | |
"seed": 2 | |
} | |
} | |