Angular Steering: Behavior Control via Rotation in Activation Space

Hieu M. Vu, Tan Minh Nguyen

Keywords: Steering, Applications of interpretability, AI Safety

TL;DR: This paper introduces Angular Steering, a robust and generalized method for fine-grained behavior control in language models, unifying and extending existing steering techniques through rotation in a feature-isolating subspace.

Abstract: Controlling specific behaviors in large language models while preserving their general capabilities is a central challenge for safe and reliable artificial intelligence (AI) deployment. Current steering methods, such as vector addition and directional ablation, are constrained within a two-dimensional subspace defined by the activation and feature direction, making them sensitive to chosen parameters and potentially affecting unrelated features due to unintended interactions in activation space. We introduce Angular Steering, a novel and flexible method for behavior modulation that operates by rotating activations within a fixed two-dimensional subspace. By formulating steering as a geometric rotation toward or away from a target behavior direction, Angular Steering provides continuous, fine-grained control over behaviors such as refusal and compliance. We demonstrate this method using refusal steering as a use case. Additionally, we propose Adaptive Angular Steering, a selective variant that rotates only activations aligned with the target feature, further enhancing stability and coherence. Angular Steering generalizes existing addition and orthogonalization techniques under a unified geometric rotation framework, simplifying parameter selection and maintaining model stability across a broader range of adjustments. Experiments across multiple model families and sizes show that Angular Steering achieves robust behavioral control while maintaining general language modeling performance, underscoring its flexibility, generalization, and robustness compared to prior approaches.

Submission Number: 68