chchen's picture
End of training
1fbe65a verified
{
"epoch": 9.977728285077951,
"total_flos": 5.23282185018409e+16,
"train_loss": 0.4115582968507494,
"train_runtime": 2017.6604,
"train_samples_per_second": 4.451,
"train_steps_per_second": 0.278
}