Look Before You Steer: Geometry Predicts SAE Feature Steerability

arXiv:2609.22782 · cs.LG · Submitted 2026-09-19 · Read on arXiv

cs.LG

Submitted: 2026-09-19

Updated: 2026-09-19

Comments: 12 pages, 3 figures, 5 tables. Accepted at the Mechanistic Interpretability Workshop at ICML 2026 and AIW 2026 at COLM

License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Terminology

• Steering — 4× in this paper · explained in 2 episodes
• AUROC — 1× in this paper · explained in 2 episodes
• Cosine Similarity — 1× in this paper · explained in 3 episodes

Related papers

• Polynomial-Augmented Neural Networks (PANNs) with Weak Orthogonality Constraints for Enhanced Function and PDE Approximation
• AIRL-S: Unifying Reinforcement Learning and Search-Based Test-Time Scaling via Adversarial Inverse Reinforcement Learning
• Transformers as Bayesian In-Context Experimenters: Smoothness-Adaptive Efficient ATE Estimation
• Convergence issues in Relational Concept Analysis based on AOC-posets
• Beliefs Beyond Posteriors: Local-Consistency Optimisation for Bayesian Neural Networks
• Understanding Diffusion Models via Ratio-Based Function Approximation with SignReLU Networks