chchen's picture
End of training
c2796a2 verified
raw
history blame contribute delete
210 Bytes
{
"epoch": 5.0,
"total_flos": 8.718478050646426e+17,
"train_loss": 0.049027663960821866,
"train_runtime": 36714.2196,
"train_samples_per_second": 1.405,
"train_steps_per_second": 0.088
}