Neuron atlas of Qwen finds 25 of 496 shared neurons carry a 'power' theme
jfonsecarivera · x · 2026-11-02
The author previously built an atlas of Qwen's neurons: turn up a single neuron's activation, then ask the model "who are you?" to observe how its self-description shifts.
This week's standout finding is a power theme: 25 of the 496 shared neurons carry power-related labels, roughly 2x what you'd expect from random neurons in the same layers. The thread shows what the untouched model says when these neurons are turned up.
Related event: Independently misaligned Qwen models light up the same power-themed neurons(4 posts)→
More from Research
- Paperclip builds biomedical datasets in one shot, cutting months-long curation to about a minute — james_y_zou · 2026-11-04
- Embodied Agent Arena: Closed-Weight VLMs Lead All Five Robot Task Domains, Open-Weight Models Lag — Haojian Huang · 2026-11-04
- FPT University Paper Shows RAG Improves Stock Market Information Retrieval Accuracy and Latency — Huan Phong Thai · 2026-11-04
- Alibaba's adaptive GUI agent training cuts environment interactions by 56-61% — Yuxiang Chai · 2026-11-04
- Large-Scale Study: AI Reviewers Often Ignore How Much You Improved Your Paper — umd-zhou-lab · 2026-11-04
- Kling Team's report: video generation bottleneck shifts from rendering to semantic control — KlingTeam · 2026-11-04