Hacker News: ant28https://news.ycombinator.com/user?id=ant28Hacker News RSShttps://hnrss.org/hnrss v2.1.1Fri, 29 May 2026 16:14:08 +0000<![CDATA[New comment by ant28 in "The Claude Code Source Leak: fake tools, frustration regexes, undercover mode"]]>How do you know this? I think of myself as being decent at spotting AI-generated text, so that I may have missed something is odd.

]]>Wed, 01 Apr 2026 14:18:07 +0000https://news.ycombinator.com/item?id=47601276ant28https://news.ycombinator.com/item?id=47601276https://news.ycombinator.com/item?id=47601276<![CDATA[New comment by ant28 in "Show HN: 1-Bit Bonsai, the First Commercially Viable 1-Bit LLMs"]]>What's up with - log error / model size? I'm not an LLM person, but a ratio of ~1 means a roughly 40% error rate for its size? I don't follow

(math: - log error / model size = 1 <-> error / model size = 1/e )

]]>Wed, 01 Apr 2026 14:11:22 +0000https://news.ycombinator.com/item?id=47601151ant28https://news.ycombinator.com/item?id=47601151https://news.ycombinator.com/item?id=47601151