NDIF open question: do J-Lens-style interpretability lenses work on block-mixed residual streams?
SonglinYang4 · x · 2026-11-21
Interpretability team NDIF raises an open question: when layers write into 6-layer blocks instead of a single residual stream, with each layer reading its own learned mix of them, do J-Lens-style lenses still work the same way — and how should lenses be developed for such architectures?
More from Research
- Ramez Naam extrapolation: AI a decade away from completing 98% of hour-long human tasks — GaryMarcus · 2026-11-21
- Colgate paper: LLM roleplay hits 90% of human test-retest reliability in purchase-intent research — 2C_ornot2C · 2026-11-21
- Riemann Hypothesis proof gets full formal verification, experts call it 'very human math' — Kyrannio · 2026-11-21
- Claims surface: Navier-Stokes and Riemann proofs via LEAN, by OpenAI and Claude — BasketVolleyball11 · 2026-11-21
- Hassabis predicted 2 years for AI to crack a Millennium Prize problem; 6 months later, both Riemann and Navier–Stokes have proposed solutions — Hesamation · 2026-11-21
- GPT-2 'sleep' replay from a frozen snapshot keeps 58% of old facts vs 6% without — Junior_Reputation639 · 2026-11-21