chchen's picture
End of training
71b259d verified
{
"epoch": 4.938271604938271,
"total_flos": 1.2628766887457587e+17,
"train_loss": 0.19458558702468873,
"train_runtime": 3405.889,
"train_samples_per_second": 0.595,
"train_steps_per_second": 0.037
}