longt5_xl_sfd_memsum_40 / train_results.json
learn3r's picture
End of training
a2b5bb6 verified
raw
history blame contribute delete
No virus
196 Bytes
{
"epoch": 38.96,
"train_loss": 0.32971865670822026,
"train_runtime": 25027.7089,
"train_samples": 3673,
"train_samples_per_second": 5.87,
"train_steps_per_second": 0.045
}