lixiqi's picture
End of training
3a04ddc
{
"epoch": 9.97,
"total_flos": 1.427863337526786e+18,
"train_loss": 1.1358447994504657,
"train_runtime": 1053.1568,
"train_samples_per_second": 17.557,
"train_steps_per_second": 0.133
}