amjadfqs's picture
End of training
563c0dd
raw
history blame contribute delete
209 Bytes
{
"epoch": 9.76,
"total_flos": 6.366510302768824e+18,
"train_loss": 0.0735103191435337,
"train_runtime": 27763.2532,
"train_samples_per_second": 2.934,
"train_steps_per_second": 0.007
}