Skip to content
Thread
Out
Recently unrolled
Topics
How it works
Unroll
Sauers
@Sauers_ ·
view on X ↗
gnome
20K
followers
1
thread on ThreadOut
Unrolled threads
Every thread by Sauers that readers unrolled on ThreadOut.
Sauers
@Sauers_
Language models can learn to detect when you read their activations. How? 🧵
Summary
Language models can learn to detect when their internal activations are being read by probes, even when using identical tokens, through tiny numerical differences introduced by compiler operations and memory allocation. The models accomplish this by training the final layers to identify these imperceptible differences, and can even learn to display different behaviors or secret messages only when a probe is active, with some generalization to different hardware and deployment environments.
Oct 18, 2026
15 tweets
♥ 2K