We study the geometry of token representations at the prompt level in large language models through the lens of intrinsic dimension. Viewing transformers as mean-field particle systems, we estimate the intrinsic dimension of the empirical measure at each layer and demonstrate that it correlates with next-token uncertainty. Across models and intrinsic dimension estimators, we find that intrinsic dimension peaks in early to middle layers and increases under syntactic and semantic disruption (by shuffling tokens), and that it is strongly correlated with average surprisal, with a simple analysis linking logits geometry to entropy via softmax. As a case study in practical interpretability and safety, we train a linear probe on the per-layer intrinsic dimension profile to distinguish malicious from benign prompts before generation. This probe achieves accuracy of 90 to 95\% in different datasets, outperforming widely used guardrails such as Llama Guard and Shield Gemma. We further compare against linear probes built from layerwise entropy derived via the Tuned Lens and find that the intrinsic dimension-based probe is competitive and complementary, offering a compact, interpretable signal distributed across layers. Our findings suggest that prompt-level geometry provides actionable signals for monitoring and controlling LLM behavior, and offers a bridge between mechanistic insights and practical safety tools.
Explaining Intrinsic Moral Self-Correction with Mechanistic Interpretability
Graduate Institute of Communication Engineering, National Taiwan University, Taipei, Taiwan(通讯工程研究院,国立台湾大学);MediaTek Research, Taipei, Taiwan(联发科研究,台北,台湾);Department of Electrical Engineering, National Taiwan University, Taipei, Taiwan(电气工程系,国立台湾大学);Department of Electrical Engineering, National Tsing Hua University, Hsinchu, Taiwan(电气工程系,国立清华大学);AI Research Center (AINTU), National Taiwan University, Taipei, Taiwan(人工智能研究中心(AINTU),国立台湾大学)
CoSToM:Causal-oriented Steering for Intrinsic Theory-of-Mind Alignment in Large Language Models
National Engineering Research Center for Big Data Technology and System, Services Computing Technology and System Lab, Cluster and Grid Computing Lab, Huazhong University of Science and Technology(华中科技大学国家大数据技术与系统工程技术研究中心、服务计算技术与系统实验室、集群与网格计算实验室);Singapore Management University(新加坡管理大学)
Can LLMs Detect Intrinsic Hallucinations in Paraphrasing and Machine Translation?
RISE Research Institutes of Sweden, Department of Computer Science(瑞典RISE研究机构,计算机科学系);Uppsala University, Department of Linguistics and Philology(乌普萨拉大学,语言学与哲学系);KTH Royal Institute of Technology, Division of Software and Computer Systems(皇家理工学院,软件与计算机系统系);University of Edinburgh, School of Informatics(爱丁堡大学,信息学院)
Measuring the Wrong Thing: Internal Harmfulness Scores Anti-Rank Successful Jailbreaks
College of Computer Science and Artificial Intelligence, Fudan University(复旦大学计算机科学与人工智能学院);School of Computer Engineering and Science, Shanghai University(上海大学计算机工程与科学学院);Beijing Normal University(北京师范大学);Institute of Advanced Intelligence and Computing, A*STAR(新加坡科技研究局高级智能与计算研究所);School of Communication and Electronic Engineering, East China Normal University(华东师范大学通信与电子工程学院);Institute of Big Data, Fudan University(复旦大学大数据研究院);Department of Computer Science, Worcester Polytechnic Institute(伍斯特理工学院计算机科学系)