SOURCE-LINKED INTELLIGENCE

LLM Forensics: Where Do Backdoors Hide? Localizing and Controlling Trigger Mechanisms with Sparse Autoencoders

arXiv · AI, language, vision and robotics · article · Sep 7, 2026 · UTC

Even though backdoors in LLMs have been a growing concern, their inner workings are still under heavy scrutiny. Trigger-based backdoors are easy to define behaviorally, a rare input that makes the model switch to a chosen response pattern, but the mechanism between triggers and their responses is less clear. We study this mechanism in a controlled, harmless language-switching setting, where fixed trigger sequences make 1B and 8B language models continue English prompts in French or German. For this, we train sparse autoencoders (SAEs) across layers and transformer components, then compare trig

Read original source ↗ Open in workspace

recordType

paper

region

Global