distilled-mt5-small-0.2-0.25 / train_results.json
Lvxue's picture
End of training
809983d
raw
history blame
No virus
191 Bytes
{
"epoch": 5.0,
"train_loss": 7.9021027734375,
"train_runtime": 2649.083,
"train_samples": 10000,
"train_samples_per_second": 18.874,
"train_steps_per_second": 4.719
}