distilled-mt5-small-010099-full / train_results.json
Lvxue's picture
End of training
afb285f
{
"epoch": 10.0,
"train_loss": 1.7898551662303561,
"train_runtime": 329774.3211,
"train_samples": 610320,
"train_samples_per_second": 18.507,
"train_steps_per_second": 4.627
}