Despite the outstanding performance of Large language Models (LLMs) in diverse tasks, they are vulnerable to jailbreak attacks, wherein adversarial prompts are crafted to bypass their security mechanisms and elicit unexpected responses. Although jailbreak attacks are prevalent, the understanding of their underlying mechanisms remains limited. Recent studies have explained typical jailbreaking behavior (e.g., the degree to which the model refuses to respond) of LLMs by analyzing representation shifts in their latent space caused by jailbreak prompts or identifying key neurons that contribute to the success of jailbreak attacks. However, these studies neither explore diverse jailbreak patterns nor provide a fine-grained explanation from the failure of circuit to the changes of representational, leaving significant gaps in uncovering the jailbreak mechanism. In this paper, we propose JailbreakLens, an interpretation framework that analyzes jailbreak mechanisms from both representation (which reveals how jailbreaks alter the model's harmfulness perception) and circuit perspectives~(which uncovers the causes of these deceptions by identifying key circuits contributing to the vulnerability), tracking their evolution throughout the entire response generation process. We then conduct an in-depth evaluation of jailbreak behavior on five mainstream LLMs under seven jailbreak strategies. Our evaluation reveals that jailbreak prompts amplify components that reinforce affirmative responses while suppressing those that produce refusal. This manipulation shifts model representations toward safe clusters to deceive the LLM, leading it to provide detailed responses instead of refusals. Notably, we find a strong and consistent correlation between representation deception and activation shift of key circuits across diverse jailbreak methods and multiple LLMs.
View on arXiv@article{he2025_2411.11114,
title={ JailbreakLens: Interpreting Jailbreak Mechanism in the Lens of Representation and Circuit },
author={ Zeqing He and Zhibo Wang and Zhixuan Chu and Huiyu Xu and Wenhui Zhang and Qinglong Wang and Rui Zheng },
journal={arXiv preprint arXiv:2411.11114},
year={ 2025 }
}