sedrickkeh's picture
End of training
bb66fed verified
{
"epoch": 3.0,
"total_flos": 406391461183488.0,
"train_loss": 0.4046504932661026,
"train_runtime": 10117.0351,
"train_samples_per_second": 2.964,
"train_steps_per_second": 0.093
}