Why do LLMs attend to the first token? | Hacker News

Hacker Newsnew | past | comments | ask | show | jobs | submit

		Why do LLMs attend to the first token? (arxiv.org)
		2 points by adhi01 on May 15, 2025 \| hide \| past \| favorite \| 1 comment

maytc on May 16, 2025 [–]

Curious if the authors had a chance to look at the Softpick paper? https://arxiv.org/abs/2504.20966

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact