byt5_en_tsn_news / train_results.json
Davlan's picture
add MT model
9185455
{
"epoch": 10.0,
"train_loss": 1.0783850606282552,
"train_runtime": 1643.1228,
"train_samples": 2100,
"train_samples_per_second": 12.781,
"train_steps_per_second": 1.278
}