distilled-mt5-small-b1 / train_results.json
Lvxue's picture
End of training
384c192
raw
history blame contribute delete
No virus
194 Bytes
{
"epoch": 5.0,
"train_loss": 4.035194541015625,
"train_runtime": 2658.5865,
"train_samples": 10000,
"train_samples_per_second": 18.807,
"train_steps_per_second": 4.702
}