distilgpt2_jje / train_results.json
AleBurzio's picture
distilgpt2-jje
84b3093
raw
history blame
197 Bytes
{
"epoch": 2.0,
"train_loss": 2.3804299571036704,
"train_runtime": 13242.0355,
"train_samples": 177922,
"train_samples_per_second": 26.872,
"train_steps_per_second": 2.687
}