mcQuantum's picture
End of training
13ea50b
raw
history blame contribute delete
210 Bytes
{
"epoch": 3.0,
"total_flos": 1.8124066505760768e+18,
"train_loss": 0.33898268457044634,
"train_runtime": 1181.8644,
"train_samples_per_second": 61.682,
"train_steps_per_second": 0.482
}