lixiqi's picture
End of training
b7576d7
{
"epoch": 3.0,
"total_flos": 6.668732964123095e+18,
"train_loss": 1.1616371373335521,
"train_runtime": 2357.3723,
"train_samples_per_second": 36.535,
"train_steps_per_second": 0.285
}