longt5_xl_gov_memsum_bp_5 / train_results.json
learn3r's picture
End of training
f6c33a3
raw
history blame
196 Bytes
{
"epoch": 4.99,
"train_loss": 0.8627048472709515,
"train_runtime": 92306.6703,
"train_samples": 17457,
"train_samples_per_second": 0.946,
"train_steps_per_second": 0.015
}