@article{scalinglawsforactivationsteeringwithllam, title = {Scaling laws for activation steering with Llama 2 models and refusal mechanisms}, author = {Sheikh Abdur Raheem Ali and Justin Xu and Ivory Yang and Jasmine Xinze Li and Ayse Arslan and Clark Benham}, year = {2025}, eprint = {2507.11771}, archivePrefix = {arXiv}, url = {https://arxiv.org/abs/2507.11771}, }