chchen's picture
End of training
61a8c2a verified
raw
history blame contribute delete
223 Bytes
{
"epoch": 4.938271604938271,
"total_flos": 1.2628766887457587e+17,
"train_loss": 0.20852369618415834,
"train_runtime": 3506.5823,
"train_samples_per_second": 0.577,
"train_steps_per_second": 0.036
}