gcperk20's picture
End of training
77ff61f
raw
history blame
210 Bytes
{
"epoch": 19.51,
"total_flos": 6.358977448272691e+16,
"train_loss": 0.40896380335092547,
"train_runtime": 131.9389,
"train_samples_per_second": 98.985,
"train_steps_per_second": 3.032
}