beit-base / all_results.json
ChasingMercer's picture
f
031ddf6
{
"epoch": 3.0,
"total_flos": 4.3513294045912105e+18,
"train_loss": 0.08035437971183718,
"train_runtime": 5091.7575,
"train_samples_per_second": 11.034,
"train_steps_per_second": 0.345
}