# SteeringSafety: Benchmarking Representation Steering in LLMs Across Safety Perspectives

arXiv: https://arxiv.org/abs/2509.13450 · HTML: https://mlsanity.com/paper/2509.13450/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Placed in Steering and probing for LLM safety, one of 13 directions in Mechanistic interpretability.
- First posted on arXiv 16 Sep 2025.

## Placed in

- Mechanistic interpretability / [Steering and probing for LLM safety](https://mlsanity.com/mechinterp/steering-and-probing-for-llm-safety/)

## Built on by (most cited first, 2 of 2)

- [Dynamic Latent Routing](https://mlsanity.com/paper/2605.14323/) (2026)
- [Steering Interference Reflects the Model's Defaults, Not the Behavior Directions](https://mlsanity.com/paper/2609.06951/) (2026)

## Abstract (arXiv)

We introduce SteeringSafety, a benchmark for evaluating representation steering methods across nine safety perspectives spanning 18 datasets. While prior work highlights the general capabilities of representation steering, we focus on safety perspectives including refusal, bias, hallucination, social behaviors, reasoning, epistemic integrity, and normative judgment. SteeringSafety provides modularized building blocks for state-of-the-art steering methods, enabling unified implementation of DIM, ACE, CAA, PCA, and LAT with recent enhancements such as conditional steering. Results on Gemma-2-2B, Llama-3.1-8B, and Qwen-2.5-7B show that strong steering performance depends on the pairing of method, model, and specific perspective. For instance, DIM is consistently effective, yet all methods exhibit substantial entanglement, where improving effectiveness on one safety perspective often significantly changes performance on others. Social behaviors are most vulnerable (degradation up to 76%), refusal steering (jailbreaking) frequently compromises normative judgment such as commonsense morality (up to 26%), and hallucination steering shifts political views unpredictably across models, ranging from a 25% shift to the right to a 28% shift to the left. These findings show the need to understand steering methods through multiple safety angles rather than a single target behavior.