ALM-AHME's picture
End of training
787bd70
raw
history blame contribute delete
210 Bytes
{
"epoch": 6.99,
"total_flos": 1.449450435763608e+19,
"train_loss": 0.10588535787143823,
"train_runtime": 11789.0344,
"train_samples_per_second": 5.344,
"train_steps_per_second": 0.167
}