Home/CommunityMechanistic interpretability researchers applying causality theory to LLMsadunk·Hacker News·Community·July 12, 2026https://arxiv.org/abs/2301.04709Read full article →Related ArticlesClaude Code sends 33k tokens before reading the prompt; OpenCode sends 7ksystima · Hacker News · 6h agoIrish datacenters now guzzle 23% of the country's electricityBender · Hacker News · 4h agoSince Chromium 148, Math.tanh is now fingerprintable to link underlying OSjoahnn_s · Hacker News · 3h agoOld and new apps, via modern coding agentssubset · Hacker News · 13h agoAI boosts research careers but narrow the span of ideas explored: studyzaikunzhang · Hacker News · 11h ago