mlsanityData to 7 Oct 2026

The Geometry of Truth: Emergent Linear Structure in Large Language Model Representations of True/False Datasets

arXiv 2310.06824 ↗First posted 10 Oct 2023cs.AIcited by 673

Not on the map. Why?

Cited by 58 in LLM reasoning, but placed by its own abstract, which didn’t fit.

Abstract (arXiv)

Large Language Models (LLMs) have impressive capabilities, but are prone to outputting falsehoods. Recent work has developed techniques for inferring whether a LLM is telling the truth by training probes on the LLM's internal activations. However, this line of work is controversial, with some authors pointing out failures of these probes to generalize in basic ways, among other conceptual issues. In this work, we use high-quality datasets of simple true/false statements to study in detail the structure of LLM representations of truth, drawing on three lines of evidence: 1. Visualizations of LLM true/false statement representations, which reveal clear linear structure. 2. Transfer experiments in which probes trained on one dataset generalize to different datasets. 3. Causal evidence obtained by surgically intervening in a LLM's forward pass, causing it to treat false statements as true and vice versa. Overall, we present evidence that at sufficient scale, LLMs linearly represent the truth or falsehood of factual statements. We also show that simple difference-in-mean probes generalize as well as other probing techniques while identifying directions which are more causally implicated in model outputs.

Key facts. Cited by 673 papers in the map (data through 7 Oct 2026). Builds on 33 of its 52 references that are in the map. First posted on arXiv 10 Oct 2023.

Builds on

33 of 52 references.

References · 33

  1. 2023LLaMA: Open and Efficient Foundation Language Modelscited by 13,958
  2. 2021TruthfulQA: Measuring How Models Mimic Human Falsehoodscited by 2,919
  3. 2022Locating and Editing Factual Associations in GPTcited by 2,609
  4. 2016Concrete Problems in AI Safetycited by 1,582
  5. 2023Sparse Autoencoders Find Highly Interpretable Features in Language Modelscited by 1,359
  6. 2016Understanding intermediate layers using linear classifier probescited by 1,279
  7. 2023Inference-Time Intervention: Eliciting Truthful Answers from a Language Modelcited by 1,259
  8. 2022Toy Models of Superpositioncited by 928
  9. 2021Probing Classifiers: Promises, Shortcomings, and Advancescited by 831
  10. 2022Discovering Language Model Behaviors with Model-Written Evaluationscited by 805
  11. 2022Discovering Latent Knowledge in Language Models Without Supervisioncited by 733
  12. 2017The Implicit Bias of Gradient Descent on Separable Datacited by 649
21 more
  1. 2023The Internal State of an LLM Knows When It's Lyingcited by 644
  2. 2019Designing and Interpreting Probes with Control Taskscited by 608
  3. 2023Dissecting Recall of Factual Associations in Auto-Regressive Language Modelscited by 512
  4. 2022Emergent World Representations: Exploring a Sequence Model Trained on a Synthetic Taskcited by 417
  5. 2021Causal Abstractions of Neural Networkscited by 401
  6. 2023Finding Neurons in a Haystack: Case Studies with Sparse Probingcited by 328
  7. 2013Direct and Indirect Effectscited by 276
  8. 2020Understanding the Role of Individual Units in a Deep Neural Networkcited by 254
  9. 2023LEACE: Perfect linear concept erasure in closed formcited by 254
  10. 2023Linearity of Relation Decoding in Transformer Language Modelscited by 190
  11. 2023AI Deception: A Survey of Examples, Risks, and Potential Solutionscited by 185
  12. 2023Localizing Model Behavior with Path Patchingcited by 182
  13. 2021Implicit Representations of Meaning in Neural Language Modelscited by 179
  14. 2021Causal Analysis of Syntactic Agreement Mechanisms in Neural Language Modelscited by 127
  15. 2018What Is One Grain of Sand in the Desert? Analyzing Individual Neurons in Deep NLP Modelscited by 124
  16. 2021Can Language Models Encode Perceptual Structure Without Grounding? A Case Study in Colorcited by 118
  17. 2023Explore, Establish, Exploit: Red Teaming Language Models from Scratchcited by 111
  18. 2021CREAK: A Dataset for Commonsense Reasoning over Entity Knowledgecited by 88
  19. 2021Neuron-level Interpretation of Deep NLP Models: A Surveycited by 81
  20. 2023Still No Lie Detector for Language Models: Probing Empirical and Conceptual Roadblockscited by 77
  21. 2022Finding Skill Neurons in Pre-trained Transformer-based Language Modelscited by 65

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2023: 5; 2024: 101; 2025: 193; 2026: 374.

673 papers.

Sort:

By area:

  1. 2024Refusal in Language Models Is Mediated by a Single Directioncited by 960
  2. 2024Do Llamas Work in English? On the Latent Language of Multilingual Transformerscited by 275
  3. 2024Semantic Entropy Probes: Robust and Cheap Hallucination Detection in LLMscited by 246
  4. 2024LLMs Know More Than They Show: On the Intrinsic Representation of LLM Hallucinationscited by 230
  5. 2024Not All Language Model Features Are One-Dimensionally Linearcited by 201
  6. 2025AxBench: Steering LLMs? Even Simple Baselines Outperform Sparse Autoencoderscited by 199
  7. 2024The Geometry of Categorical and Hierarchical Concepts in Large Language Modelscited by 151
  8. 2024Improving Instruction-Following in Language Models through Activation Steeringcited by 149
  9. 2024The Remarkable Robustness of LLMs: Stages of Inference?cited by 138
  10. 2024Analyzing the Generalization and Reliability of Steering Vectorscited by 130
  11. 2024Black-Box Access is Insufficient for Rigorous AI Auditscited by 123
  12. 2024Do I Know This Entity? Knowledge Awareness and Hallucinations in Language Modelscited by 114
38 more
  1. 2024A Practical Review of Mechanistic Interpretability for Transformer-Based Language Modelscited by 113
  2. 2025Are Sparse Autoencoders Useful? A Case Study in Sparse Probingcited by 109
  3. 2024TruthX: Alleviating Hallucinations by Editing Large Language Models in Truthful Spacecited by 90
  4. 2024A Primer on the Inner Workings of Transformer-based Language Modelscited by 83
  5. 2024Truth is Universal: Robust Detection of Lies in LLMscited by 78
  6. 2024RAVEL: Evaluating Interpretability Methods on Disentangling Language Model Representationscited by 77
  7. 2025LLMs Encode Harmfulness and Refusal Separatelycited by 75
  8. 2023Cognitive Dissonance: Why Do Language Model Outputs Disagree with Internal Representations of Truthfulness?cited by 62
  9. 2024Robust LLM safeguarding via refusal feature adversarial trainingcited by 60
  10. 2024Confidence Regulation Neurons in Language Modelscited by 57
  11. 2025Large Language Models Share Representations of Latent Grammatical Concepts Across Typologically Diverse Languagescited by 57
  12. 2023Eliciting Latent Knowledge from Quirky Language Modelscited by 56
  13. 2024The Geometry of Concepts: Sparse Autoencoder Feature Structurecited by 52
  14. 2024A Mechanistic Analysis of a Transformer Trained on a Symbolic Multi-Step Reasoning Taskcited by 49
  15. 2024ICLR: In-Context Learning of Representationscited by 49
  16. 2025Steer LLM Latents for Hallucination Detectioncited by 49
  17. 2025MIB: A Mechanistic Interpretability Benchmarkcited by 49
  18. 2025Calibrating Verbal Uncertainty as a Linear Feature to Reduce Hallucinationscited by 48
  19. 2025Attention Sinks and Compression Valleys in LLMs are Two Sides of the Same Coincited by 47
  20. 2024CausalGym: Benchmarking causal interpretability methods on linguistic taskscited by 45
  21. 2024Characterizing Truthfulness in Large Language Model Generations with Local Intrinsic Dimensioncited by 45
  22. 2024Usable XAI: 10 Strategies Towards Exploiting Explainability in the LLM Eracited by 45
  23. 2025From Flat to Hierarchical: Extracting Sparse Representations with Matching Pursuitcited by 45
  24. 2025Detecting High-Stakes Interactions with Activation Probescited by 45
  25. 2025Angular Steering: Behavior Control via Rotation in Activation Spacecited by 45
  26. 2025Massive Values in Self-Attention Modules are the Key to Contextual Knowledge Understandingcited by 44
  27. 2024A Language Model's Guide Through Latent Spacecited by 43
  28. 2024Designing a Dashboard for Transparency and Control of Conversational AIcited by 43
  29. 2026When Models Manipulate Manifolds: The Geometry of a Counting Taskcited by 43
  30. 2025Hallucination Detection in LLMs Using Spectral Features of Attention Mapscited by 42
  31. 2025Activation Oracles: Training and Evaluating LLMs as General-Purpose Activation Explainerscited by 41
  32. 2024The Unreasonable Effectiveness of Easy Training Data for Hard Taskscited by 40
  33. 2025A Unified Understanding and Evaluation of Steering Methodscited by 40
  34. 2024Estimating Knowledge in Large Language Models Without Generating a Single Tokencited by 39
  35. 2024Monitoring Latent World States in Language Models with Propositional Probescited by 39
  36. 2025Understanding (Un)Reliability of Steering Vectors in Language Modelscited by 39
  37. 2024Language Ranker: A Metric for Quantifying LLM Performance Across High and Low-Resource Languagescited by 38
  38. 2025LLM Unlearning via Neural Activation Redirectioncited by 38

50 most cited of 673 All 673 (JSON)