{ | |
"best_metric": 0.6198716753557617, | |
"best_model_checkpoint": "test-klue/ynat/run-3/checkpoint-691", | |
"epoch": 1.0, | |
"eval_steps": 500, | |
"global_step": 691, | |
"is_hyper_param_search": true, | |
"is_local_process_zero": true, | |
"is_world_process_zero": true, | |
"log_history": [ | |
{ | |
"epoch": 0.72, | |
"learning_rate": 1.7061206107369956e-05, | |
"loss": 0.8931, | |
"step": 500 | |
}, | |
{ | |
"epoch": 1.0, | |
"eval_f1": 0.6198716753557617, | |
"eval_loss": 0.7939329743385315, | |
"eval_runtime": 9.5969, | |
"eval_samples_per_second": 246.85, | |
"eval_steps_per_second": 1.98, | |
"step": 691 | |
} | |
], | |
"logging_steps": 500, | |
"max_steps": 2073, | |
"num_input_tokens_seen": 0, | |
"num_train_epochs": 3, | |
"save_steps": 500, | |
"total_flos": 168043139838000.0, | |
"train_batch_size": 8, | |
"trial_name": null, | |
"trial_params": { | |
"learning_rate": 2.237588569580688e-05, | |
"num_train_epochs": 3, | |
"per_device_train_batch_size": 8, | |
"seed": 39 | |
} | |
} | |