Interp Researchers Clash: Is Superposition Just a Function of Parameter Count?
livgorton · x · 2026-10-16
An interpretability debate broke out on X: one participant argued a model's interpretability can go from 0% to 99% without changing model size, claiming nearly everyone simply had the wrong causal explanation. A researcher with field experience pushed back, arguing the amount of superposition and downstream polysemanticity is not purely a function of parameter count. The exchange highlights ongoing disagreement in the interp community over mechanistic explanations versus ontological differences.
More from AGI Musings
- Blogger predicts recursive self-improvement arrives around November 2027 — imjustnewatai · 2026-10-16
- 31 Tech Curves Visualized: Kurzweil's $1,000-Brain-Compute Call Missed by 400x — cpaik · 2026-10-16
- DeepMind Paper: First-Price Auctions in Agentic Markets Reach 1-1/e^2 Efficiency — dair_ai · 2026-10-16
- Garrison Lovely's new book 'Obsolete' argues AGI is neither impossible nor inevitable — GarrisonLovely · 2026-10-16
- Majmudar clarifies: controllability may be solved in 1-2 years, not full alignment — MajmudarAdam · 2026-10-16
- AI-generated fake Feynman lectures are flooding the internet, and they're bad — dreamwieber · 2026-10-16