longt5_xl_gov_memsum_bp_15 / train_results.json
learn3r's picture
End of training
92a6484
{
"epoch": 4.99,
"train_loss": 0.10190736966974595,
"train_runtime": 144636.3155,
"train_samples": 17457,
"train_samples_per_second": 0.603,
"train_steps_per_second": 0.009
}