venetis's picture
End of training
aad7306
{
"epoch": 15.0,
"total_flos": 8.179980758810437e+19,
"train_loss": 1.7412154973476806,
"train_runtime": 20817.4419,
"train_samples_per_second": 154.783,
"train_steps_per_second": 0.605
}