chchen's picture
End of training
2ed1f62 verified
{
"epoch": 4.938271604938271,
"total_flos": 1.3285578941359718e+17,
"train_loss": 0.09023197002708912,
"train_runtime": 3707.8482,
"train_samples_per_second": 0.546,
"train_steps_per_second": 0.034
}