longt5_xl_sfd_4096_e10 / train_results.json
learn3r's picture
End of training
114d833 verified
raw
history blame contribute delete
195 Bytes
{
"epoch": 9.74,
"train_loss": 1.5745136559009552,
"train_runtime": 36489.2213,
"train_samples": 3673,
"train_samples_per_second": 1.007,
"train_steps_per_second": 0.004
}