Interpretability's visible leap: same neurons, far better microscope than 2017
MajmudarAdam · x · 2026-10-16
Replying to skepticism that a recent interpretability post offers little empirical update, Majmudar Adam argues the progress is huge: compared to classic Distill.pub activation-atlas work, today's feature interpretations show a massive jump in intelligibility and in the share of neurons that are understandable. The accompanying image contrasts the same InceptionV1 neurons visualized 2009-style, in 2017, and in 2026 — evolving from noise to discrete, human-readable concepts. "Same neurons, better microscope."
More from Models
- Meta's Muse Spark Open scores 39 vs 48 for closed Spark 1.3 on AA Intelligence Index — EAccelerate_42 · 2026-10-16
- PhoneLLM Mini (30B) and Max (120B) target sub-100ms voice AI latency on Workers AI — ritakozlov · 2026-10-16
- Yacine on LLMs: "I'm sick of how stupid these language models are. Accelerate!" — yacineMTB · 2026-10-16
- Gemini 4 Argon rolls out in Antigravity for paid AI Pro users with a separate weekly quota — GeminiApp · 2026-10-16
- GPT-4o API dies January 14 as OpenAI refuses to release weights, pushing users to Qwen — mimi10v3 · 2026-10-16
- Opus 5.5 value may end the habit of using the top model for everything — adonis_singh · 2026-10-16