llm_NLP / run-0 /checkpoint-1070 /trainer_state.json
LarryTW's picture
Training in progress, epoch 2
fe6fdc0
{
"best_metric": 0.0,
"best_model_checkpoint": "distilbert-base-uncased-finetuned-cola/run-0/checkpoint-535",
"epoch": 2.0,
"eval_steps": 500,
"global_step": 1070,
"is_hyper_param_search": true,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.93,
"learning_rate": 6.837169275748106e-07,
"loss": 0.6115,
"step": 500
},
{
"epoch": 1.0,
"eval_loss": 0.5967965126037598,
"eval_matthews_correlation": 0.0,
"eval_runtime": 0.7905,
"eval_samples_per_second": 1319.501,
"eval_steps_per_second": 83.497,
"step": 535
},
{
"epoch": 1.87,
"learning_rate": 8.396523671971359e-08,
"loss": 0.5803,
"step": 1000
},
{
"epoch": 2.0,
"eval_loss": 0.5847072005271912,
"eval_matthews_correlation": 0.0,
"eval_runtime": 0.8007,
"eval_samples_per_second": 1302.61,
"eval_steps_per_second": 82.428,
"step": 1070
}
],
"logging_steps": 500,
"max_steps": 1070,
"num_input_tokens_seen": 0,
"num_train_epochs": 2,
"save_steps": 500,
"total_flos": 85092192859320.0,
"train_batch_size": 16,
"trial_name": null,
"trial_params": {
"learning_rate": 1.283468618429908e-06,
"num_train_epochs": 2,
"per_device_train_batch_size": 16,
"seed": 29
}
}