polejowska's picture
End of training
28a3eaf
{
"epoch": 4.98,
"total_flos": 1.6665072271223685e+18,
"train_loss": 0.47145734555793534,
"train_runtime": 824.3744,
"train_samples_per_second": 26.202,
"train_steps_per_second": 0.2
}