shuvayanti's picture
Training in progress, epoch 4
746fca6
raw
history blame
1.35 kB
{
"best_metric": 0.447681158536829,
"best_model_checkpoint": "distilbert-base-uncased-finetuned-cola/run-5/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": 1.6879780082730344e-05,
"loss": 0.5883,
"step": 500
},
{
"epoch": 0.47,
"learning_rate": 1.583160695290357e-05,
"loss": 0.5765,
"step": 1000
},
{
"epoch": 0.7,
"learning_rate": 1.47834338230768e-05,
"loss": 0.5618,
"step": 1500
},
{
"epoch": 0.94,
"learning_rate": 1.3735260693250027e-05,
"loss": 0.5987,
"step": 2000
},
{
"epoch": 1.0,
"eval_loss": 0.5104408264160156,
"eval_matthews_correlation": 0.447681158536829,
"eval_runtime": 0.6946,
"eval_samples_per_second": 1501.589,
"eval_steps_per_second": 95.019,
"step": 2138
}
],
"logging_steps": 500,
"max_steps": 8552,
"num_train_epochs": 4,
"save_steps": 500,
"total_flos": 32623201623696.0,
"trial_name": null,
"trial_params": {
"learning_rate": 1.7927953212557118e-05,
"num_train_epochs": 4,
"per_device_train_batch_size": 4,
"seed": 35
}
}