distilled-mt5-small-0.07-0.25 / train_results.json
Lvxue's picture
End of training
595dacf
{
"epoch": 5.0,
"train_loss": 5.0029668359375,
"train_runtime": 2664.5208,
"train_samples": 10000,
"train_samples_per_second": 18.765,
"train_steps_per_second": 4.691
}