gcperk20's picture
End of training
6582364
raw
history blame
No virus
208 Bytes
{
"epoch": 19.51,
"total_flos": 2.493693637268521e+17,
"train_loss": 0.3432186938822269,
"train_runtime": 208.3396,
"train_samples_per_second": 62.686,
"train_steps_per_second": 1.92
}