longt5_xl_sfd_memsum_30 / train_results.json
learn3r's picture
End of training
68d715b verified
raw
history blame contribute delete
No virus
196 Bytes
{
"epoch": 29.22,
"train_loss": 0.5600467140298514,
"train_runtime": 18789.4517,
"train_samples": 3673,
"train_samples_per_second": 5.864,
"train_steps_per_second": 0.022
}