Fetching the paper…

LLaMA-NAS: Efficient Neural Architecture Search for Large Language Models · Around