distilgpt2-wikitext2 / train_results.json
Mingxiao's picture
fine-tuned model
674af44
raw
history blame contribute delete
No virus
192 Bytes
{
"epoch": 3.0,
"train_loss": 3.429908208737428,
"train_runtime": 1005.8166,
"train_samples": 2318,
"train_samples_per_second": 6.914,
"train_steps_per_second": 0.865
}