Researcher critiques anti-superposition argument as circular descriptive refitting
jiaxinwen22 · x · 2026-10-16
Researcher jiaxinwen22 critiques the structure of recent arguments against superposition:
- Prior neuron description A can't fit N activation patterns, so people claim superposition
- Find description B that fits those N patterns, so "it's not superposition"
- If M more patterns appear that B can't fit, just ask Claude to generate description C fitting M+N — still "not superposition"
The thread highlights a methodological concern in interpretability research: whether endlessly refitting descriptions to absorb counterexamples makes the anti-superposition argument circular.
More from Research
- Free Textbook: Introduction to Machine Learning Covers the Math from Optimization to Generative Models — TheTuringPost · 2026-10-16
- Free ML math textbook: Laurent Younes' Introduction to Machine Learning covers Adam to generative models — TheTuringPost · 2026-10-16
- RFdiffusion3 published in Science, alongside Chaperone agent harness for bio-molecular design — anshulkundaje · 2026-10-16
- Microsoft paper: history-aware model routing lifts multi-turn agent task pass rate by 3.3 points — dair_ai · 2026-10-16
- Researcher: everyone should build a good eval at least once — bad evals are everywhere — gabriberton · 2026-10-16
- The square lattice connective constant remains unsolved — and might be ripe for AI — tak3sh8 · 2026-10-16