mansee's picture
End of training
d7a183b
raw
history blame
210 Bytes
{
"epoch": 4.95,
"total_flos": 1.1422818298339983e+18,
"train_loss": 0.08449313590923944,
"train_runtime": 710.4711,
"train_samples_per_second": 65.316,
"train_steps_per_second": 0.507
}