tsware's picture
End of training
5a46ef2 verified
{
"epoch": 2.98,
"total_flos": 1.2014751945240576e+18,
"train_loss": 0.17176837376540616,
"train_runtime": 8690.1401,
"train_samples_per_second": 5.591,
"train_steps_per_second": 0.043
}