surajjoshi's picture
End of training
1633a06
{
"epoch": 3.0,
"total_flos": 5.409667845997019e+18,
"train_loss": 0.1632997931153686,
"train_runtime": 53634.0894,
"train_samples_per_second": 1.287,
"train_steps_per_second": 0.01
}