@article{featureguidedsaesteeringforrefusalrateco, title = {Feature-Guided SAE Steering for Refusal-Rate Control using Contrasting Prompts}, author = {Samaksh Bhargav and Zining Zhu}, year = {2025}, eprint = {2511.00029}, archivePrefix = {arXiv}, url = {https://arxiv.org/abs/2511.00029}, }