chchen's picture
End of training
c4eefbf verified
raw
history blame contribute delete
223 Bytes
{
"epoch": 4.938271604938271,
"total_flos": 1.9835335038939955e+17,
"train_loss": 0.024442945823073388,
"train_runtime": 6275.5304,
"train_samples_per_second": 0.323,
"train_steps_per_second": 0.02
}