@article{depthwiseactivationsteeringforhonestlang, title = {Depth-Wise Activation Steering for Honest Language Models}, author = {Gracjan Góral and Marysia Winkels and Steven Basart}, year = {2025}, eprint = {2512.07667}, archivePrefix = {arXiv}, url = {https://arxiv.org/abs/2512.07667}, }