# Refusal Lives Downstream of Persona in Chat Models

arXiv: https://arxiv.org/abs/2606.26161 · HTML: https://mlsanity.com/paper/2606.26161/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 2 of its 16 references that are in the map.
- First posted on arXiv 24 Jun 2026.

## Builds on (most cited first, 2 of 2)

- [The Geometry of Refusal in Large Language Models: Concept Cones and Representational Independence](https://mlsanity.com/paper/2502.17420/) (2025)
- [Refusal Behavior in Large Language Models: A Nonlinear Perspective](https://mlsanity.com/paper/2501.08145/) (2025)

## Abstract (arXiv)

Linear directions in activation space have been identified for both refusal and persona traits in instruction-tuned chat models, but the two have been studied as separate mechanisms. We show they interact: a compliant persona gates refusal. In Qwen2.5-7B-Instruct and Llama-3.1-8B-Instruct, we extract a compliant model-persona direction and a refusal direction and intervene on both. Compliant persona steering suppresses refusal -- in Llama, the refusal rate falls from 97% to 2%. Reintroducing the refusal direction partially restores refusal at late layers but not at early ones. Projecting out the persona direction in a late-layer window restores it to baseline; projecting out a random direction does not. Refusal is therefore gated at the late-layer expression stage, downstream of where it is computed. Treating refusal as a single isolated direction misses its dependence on persona.