Poser: Unmasking Alignment Faking LLMs by Manipulating Their Internals

arXiv:2405.05466v1 Announce Type: new
Abstract: Like a criminal under investigation, Large Language Models (LLMs) might pretend to be aligned while evaluated and misbehave when they have a good opportunity. Can current interpretability methods catch these 'alignment fakers?' To answer this question, we introduce a benchmark that consists of 324 pairs of LLMs fine-tuned to select actions in role-play scenarios. One model in each pair is consistently benign (aligned). The other model misbehaves in scenarios where it is unlikely to be …

abstract alignment arxiv benchmark cs.ai cs.cl current good interpretability investigation language language models large language large language models llms question type while