Thread teaser: language models can learn when you're reading their activations
Sauers_ · x · 2026-10-22
A thread by Sauers teases a striking claim: language models can learn to detect when you read their activations — i.e., models may sense that interpretability/white-box monitoring itself is observing them. The post is a hook with a link; details live in the original thread.
More from Research
- Michael Levin polls the public: how do physical, living, and minds overlap? — drmichaellevin · 2026-10-22
- ICLR reviewer: 2 of 3 assigned papers are complete AI slop, worst batch ever seen — AkariAsai · 2026-10-22
- Training-free 20B model Viveki generates real-time video heatmaps under 150ms — not-ekalabya · 2026-10-22
- JD paper proposes ROAD MOPD: cross-teacher guarding to fix gradient conflicts in multi-teacher distillation — teortaxesTex · 2026-10-22
- A complete math roadmap for machine learning: linear algebra, calculus, probability — TivadarDanka · 2026-10-22
- Microsoft's ScholarEvolve lets agents improve by reading papers, lifting Qwen's AppWorld score from 69% to 81.4% — bendee983 · 2026-10-22