ALM-AHME's picture
End of training
62900df
raw
history blame contribute delete
209 Bytes
{
"epoch": 14.93,
"total_flos": 9.675577376037974e+18,
"train_loss": 0.8123349746416884,
"train_runtime": 8647.4379,
"train_samples_per_second": 6.333,
"train_steps_per_second": 0.198
}