distilled-mt5-small-b5 / train_results.json
Lvxue's picture
End of training
bb3294b
raw
history blame
193 Bytes
{
"epoch": 5.0,
"train_loss": 4.99171919921875,
"train_runtime": 2742.9735,
"train_samples": 10000,
"train_samples_per_second": 18.228,
"train_steps_per_second": 4.557
}