LLM_T1 / run-6 /checkpoint-2138 /trainer_state.json
CCChenRyan's picture
Training in progress, epoch 1
a401230
{
"best_metric": 0.46718843812774224,
"best_model_checkpoint": "distilbert-base-uncased-finetuned-cola/run-6/checkpoint-2138",
"epoch": 1.0,
"eval_steps": 500,
"global_step": 2138,
"is_hyper_param_search": true,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.23,
"learning_rate": 1.1719121295456048e-05,
"loss": 0.5754,
"step": 500
},
{
"epoch": 0.47,
"learning_rate": 8.141855942752735e-06,
"loss": 0.5394,
"step": 1000
},
{
"epoch": 0.7,
"learning_rate": 4.564590590049425e-06,
"loss": 0.5396,
"step": 1500
},
{
"epoch": 0.94,
"learning_rate": 9.87325237346114e-07,
"loss": 0.5291,
"step": 2000
},
{
"epoch": 1.0,
"eval_loss": 0.5313145518302917,
"eval_matthews_correlation": 0.46718843812774224,
"eval_runtime": 0.7498,
"eval_samples_per_second": 1391.059,
"eval_steps_per_second": 88.025,
"step": 2138
}
],
"logging_steps": 500,
"max_steps": 2138,
"num_input_tokens_seen": 0,
"num_train_epochs": 1,
"save_steps": 500,
"total_flos": 32685295716816.0,
"train_batch_size": 4,
"trial_name": null,
"trial_params": {
"learning_rate": 1.529638664815936e-05,
"num_train_epochs": 1,
"per_device_train_batch_size": 4,
"seed": 13
}
}