kanishka's picture
End of training
3f74250 verified
raw
history blame
238 Bytes
{
"epoch": 24.0,
"total_flos": 1.550560221462528e+18,
"train_loss": 3.0875192876465034,
"train_runtime": 49059.1729,
"train_samples": 494517,
"train_samples_per_second": 403.2,
"train_steps_per_second": 1.574
}