longt5_xl_gov_bp_5 / train_results.json
learn3r's picture
End of training
b61d936
Raw
History Blame
196 Bytes
{
"epoch": 4.99,
"train_loss": 1.064733072940041,
"train_runtime": 117328.7303,
"train_samples": 17457,
"train_samples_per_second": 0.744,
"train_steps_per_second": 0.006
}