# Localizing Lying in Llama: Understanding Instructed Dishonesty on True-False Questions Through Prompting, Probing, and Patching

arXiv: https://arxiv.org/abs/2311.15131 · HTML: https://mlsanity.com/paper/2311.15131/

## Key facts

- Cited by 24 papers in the map (data through 14 Oct 2026).
- Placed in Probing deception, bias and beliefs, one of 13 directions in Mechanistic interpretability.
- Builds on 6 of its 13 references that are in the map.
- First posted on arXiv 25 Nov 2023.

## Placed in

- Mechanistic interpretability / [Probing deception, bias and beliefs](https://mlsanity.com/mechinterp/probing-deception-bias-and-beliefs/)

## Builds on (most cited first, 6 of 6)

- [Inference-Time Intervention: Eliciting Truthful Answers from a Language Model](https://mlsanity.com/paper/2306.03341/) (2023)
- [The Internal State of an LLM Knows When It's Lying](https://mlsanity.com/paper/2304.13734/) (2023)
- [An Overview of Catastrophic AI Risks](https://mlsanity.com/paper/2306.12001/) (2023)
- [Overthinking the Truth: Understanding how Language Models Process False Demonstrations](https://mlsanity.com/paper/2307.09476/) (2023)
- [How to Catch an AI Liar: Lie Detection in Black-Box LLMs by Asking Unrelated Questions](https://mlsanity.com/paper/2309.15840/) (2023)
- [Still No Lie Detector for Language Models: Probing Empirical and Conceptual Roadblocks](https://mlsanity.com/paper/2307.00175/) (2023)

## Built on by (most cited first, 24 of 24)

- [Assessing the Brittleness of Safety Alignment via Pruning and Low-Rank Modifications](https://mlsanity.com/paper/2402.05162/) (2024)
- [The MASK Benchmark: Disentangling Honesty From Accuracy in AI Systems](https://mlsanity.com/paper/2503.03750/) (2025)
- [On the Role of Attention Heads in Large Language Model Safety](https://mlsanity.com/paper/2410.13708/) (2024)
- [Usable XAI: 10 Strategies Towards Exploiting Explainability in the LLM Era](https://mlsanity.com/paper/2403.08946/) (2024)
- [Model Editing as a Robust and Denoised variant of DPO: A Case Study on Toxicity](https://mlsanity.com/paper/2405.13967/) (2024)
- [Standards for Belief Representations in LLMs](https://mlsanity.com/paper/2405.21030/) (2024)
- [AI Risk Management Should Incorporate Both Safety and Security](https://mlsanity.com/paper/2405.19524/) (2024)
- [Difficulties with Evaluating a Deception Detector for AIs](https://mlsanity.com/paper/2511.22662/) (2025)
- [LLMs Faithfully and Iteratively Compute Answers During CoT: A Systematic Analysis With Multi-step Arithmetics](https://mlsanity.com/paper/2412.01113/) (2024)
- [LLMs Know They're Wrong and Agree Anyway: The Shared Sycophancy-Lying Circuit](https://mlsanity.com/paper/2604.19117/) (2026)
- [Predicting the Performance of Black-box LLMs through Follow-up Queries](https://mlsanity.com/paper/2501.01558/) (2025)
- [Improving Preference Extraction In LLMs By Identifying Latent Knowledge Through Classifying Probes](https://mlsanity.com/paper/2503.17755/) (2025)
- [Rep2Text: Decoding Full Text from a Single LLM Token Representation](https://mlsanity.com/paper/2511.06571/) (2025)
- [Exploring and Exploiting the Inherent Efficiency within Large Reasoning Models for Self-Guided Efficiency Enhancement](https://mlsanity.com/paper/2506.15647/) (2025)
- [Can LLMs Lie? Investigation beyond Hallucination](https://mlsanity.com/paper/2509.03518/) (2025)
- [Large Language Models as Misleading Assistants in Conversation](https://mlsanity.com/paper/2407.11789/) (2024)
- [Lens: Rethinking Multilingual Enhancement for Large Language Models](https://mlsanity.com/paper/2410.04407/) (2024)
- [Cluster-norm for Unsupervised Probing of Knowledge](https://mlsanity.com/paper/2407.18712/) (2024)
- [Logistic Regression makes small LLMs strong and explainable "tens-of-shot" classifiers](https://mlsanity.com/paper/2408.03414/) (2024)
- [This Is Your Doge, If It Please You: Exploring Deception and Robustness in Mixture of LLMs](https://mlsanity.com/paper/2503.05856/) (2025)
- [Trick or Neat: Adversarial Ambiguity and Language Model Evaluation](https://mlsanity.com/paper/2506.01205/) (2025)
- [Detecting Contextual Hallucinations in LLMs with Frequency-Aware Attention](https://mlsanity.com/paper/2602.18145/) (2026)
- [Nodes Are Early, Edges Are Late: Probing Diagram Representations in Large Vision-Language Models](https://mlsanity.com/paper/2603.02865/) (2026)
- [A Lie Detector Test for Language Models: Reading Knowledge a Model Won't Reveal](https://mlsanity.com/paper/2609.21996/) (2026)

## Abstract (arXiv)

Large language models (LLMs) demonstrate significant knowledge through their outputs, though it is often unclear whether false outputs are due to a lack of knowledge or dishonesty. In this paper, we investigate instructed dishonesty, wherein we explicitly prompt LLaMA-2-70b-chat to lie. We perform prompt engineering to find which prompts best induce lying behavior, and then use mechanistic interpretability approaches to localize where in the network this behavior occurs. Using linear probing and activation patching, we localize five layers that appear especially important for lying. We then find just 46 attention heads within these layers that enable us to causally intervene such that the lying model instead answers honestly. We show that these interventions work robustly across many prompts and dataset splits. Overall, our work contributes a greater understanding of dishonesty in LLMs so that we may hope to prevent it.