longt5_xl_summ_screen_bp_only_30 / train_results.json
learn3r's picture
End of training
ddf30ed
raw
history blame
197 Bytes
{
"epoch": 13.18,
"train_loss": 0.16495737830797833,
"train_runtime": 12728.0808,
"train_samples": 3673,
"train_samples_per_second": 4.329,
"train_steps_per_second": 0.004
}