mansee's picture
End of training
173e935
raw
history blame
208 Bytes
{
"epoch": 9.88,
"total_flos": 6.325810471729889e+17,
"train_loss": 0.1487744191288948,
"train_runtime": 421.2195,
"train_samples_per_second": 61.132,
"train_steps_per_second": 0.475
}