News for hackers. Stuff that matters.

ant28

ant28 on Hacker News

Re:The Claude Code Source Leak: fake tools, frustration regexes, under... on Wednesday April 01, 2026 @10:18AM (#47601276)
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.
Re:Show HN: 1-Bit Bonsai, the First Commercially Viable 1-Bit LLMs on Wednesday April 01, 2026 @10:11AM (#47601151)
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 )