longt5_xl_sfd_bp_40 / train_results.json
learn3r's picture
End of training
e2b0062
{
"epoch": 19.48,
"train_loss": 0.049460863295410364,
"train_runtime": 102546.6903,
"train_samples": 3673,
"train_samples_per_second": 0.716,
"train_steps_per_second": 0.003
}