Augusto777's picture
End of training
b6011ca
{
"epoch": 36.13,
"total_flos": 8.783603110659686e+17,
"train_loss": 0.42505432409899574,
"train_runtime": 631.588,
"train_samples_per_second": 61.939,
"train_steps_per_second": 0.443
}