rwang5688's picture
End of training
761ad74
{
"epoch": 3.0,
"total_flos": 5.649571096410931e+18,
"train_loss": 0.25466423411118355,
"train_runtime": 3239.4244,
"train_samples_per_second": 22.504,
"train_steps_per_second": 0.176
}