tsware's picture
End of training
4494137 verified
raw
history blame contribute delete
No virus
211 Bytes
{
"epoch": 6.96,
"total_flos": 2.8025139821285376e+18,
"train_loss": 0.13459565003810117,
"train_runtime": 2248.9328,
"train_samples_per_second": 50.411,
"train_steps_per_second": 0.392
}