100rab25's picture
End of training
a81ec82
raw
history blame
208 Bytes
{
"epoch": 10.0,
"total_flos": 2.6335377898183066e+18,
"train_loss": 0.2811273262443313,
"train_runtime": 1705.57,
"train_samples_per_second": 62.114,
"train_steps_per_second": 0.487
}