ALM-AHME's picture
End of training
f7e13ef
{
"epoch": 15.0,
"total_flos": 2.615871779941884e+19,
"train_loss": 0.02590897159305381,
"train_runtime": 7346.925,
"train_samples_per_second": 12.993,
"train_steps_per_second": 0.406
}