distilled-mt5-small-0.0-0.5 / train_results.json
Lvxue's picture
End of training
ed1e7c1
raw history blame
No virus
194 Bytes
{
"epoch": 5.0,
"train_loss": 3.157569091796875,
"train_runtime": 2599.1072,
"train_samples": 10000,
"train_samples_per_second": 19.237,
"train_steps_per_second": 4.809
}