# Unsupervised Concept Vector Extraction for Bias Control in LLMs

arXiv: https://arxiv.org/abs/2502.19721 · HTML: https://mlsanity.com/paper/2502.19721/

## Key facts

- Cited by 7 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 21 of its 38 references that are in the map.
- First posted on arXiv 27 Feb 2025.

## Placed in

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

## Builds on (most cited first, 21 of 21)

- [Measuring Massive Multitask Language Understanding](https://mlsanity.com/paper/2009.03300/) (2020)
- [Representation Engineering: A Top-Down Approach to AI Transparency](https://mlsanity.com/paper/2310.01405/) (2023)
- [Steering Llama 2 via Contrastive Activation Addition](https://mlsanity.com/paper/2312.06681/) (2023)
- [Refusal in Language Models Is Mediated by a Single Direction](https://mlsanity.com/paper/2406.11717/) (2024)
- [The Geometry of Truth: Emergent Linear Structure in Large Language Model Representations of True/False Datasets](https://mlsanity.com/paper/2310.06824/) (2023)
- [Gender Bias in Coreference Resolution](https://mlsanity.com/paper/1804.09301/) (2018)
- [DExperts: Decoding-Time Controlled Text Generation with Experts and Anti-Experts](https://mlsanity.com/paper/2105.03023/) (2021)
- [Preventing Fairness Gerrymandering: Auditing and Learning for Subgroup Fairness](https://mlsanity.com/paper/1711.05144/) (2017)
- [Marked Personas: Using Natural Language Prompts to Measure Stereotypes in Language Models](https://mlsanity.com/paper/2305.18189/) (2023)
- [Programming Refusal with Conditional Activation Steering](https://mlsanity.com/paper/2409.05907/) (2024)
- [Linear Representations of Sentiment in Large Language Models](https://mlsanity.com/paper/2310.15154/) (2023)
- [NeuroLogic Decoding: (Un)supervised Neural Text Generation with Predicate Logic Constraints](https://mlsanity.com/paper/2010.12884/) (2020)
- [Toward Gender-Inclusive Coreference Resolution](https://mlsanity.com/paper/1910.13913/) (2019)
- [Theories of "Gender" in NLP Bias Research](https://mlsanity.com/paper/2205.02526/) (2022)
- [Linguistic Bias in ChatGPT: Language Models Reinforce Dialect Discrimination](https://mlsanity.com/paper/2406.08818/) (2024)
- [Investigating African-American Vernacular English in Transformer-Based Text Generation](https://mlsanity.com/paper/2010.02510/) (2020)
- [Evaluating Gender Bias in Large Language Models via Chain-of-Thought Prompting](https://mlsanity.com/paper/2401.15585/) (2024)
- [A Trip Towards Fairness: Bias and De-Biasing in Large Language Models](https://mlsanity.com/paper/2305.13862/) (2023)
- [Identifying and Adapting Transformer-Components Responsible for Gender Bias in an English Language Model](https://mlsanity.com/paper/2310.12611/) (2023)
- [BiasAlert: A Plug-and-play Tool for Social Bias Detection in LLMs](https://mlsanity.com/paper/2407.10241/) (2024)
- [Rejected Dialects: Biases Against African American Language in Reward Models](https://mlsanity.com/paper/2502.12858/) (2025)

## Built on by (most cited first, 7 of 7)

- [Steering the CensorShip: Uncovering Representation Vectors for LLM "Thought" Control](https://mlsanity.com/paper/2504.17130/) (2025)
- [Refusal Steering: Fine-grained Control over LLM Refusal Behaviour for Sensitive Topics](https://mlsanity.com/paper/2512.16602/) (2025)
- [Neuron-Level Interventions for Gendered and Gender-Neutral Generation in Language Models](https://mlsanity.com/paper/2605.30717/) (2026)
- [Understanding Formal Reasoning Failures in LLMs as Abstract Interpreters](https://mlsanity.com/paper/2503.12686/) (2025)
- [NeuroFilter: Activation-Based Guardrails for Privacy-Conscious LLM Agents](https://mlsanity.com/paper/2601.14660/) (2026)
- [White-Box Sensitivity Auditing with Steering Vectors](https://mlsanity.com/paper/2601.16398/) (2026)
- [When Do Model Internals Help? Exploring the Role of Representation Engineering in LLM Safety](https://mlsanity.com/paper/2609.34771/) (2026)

## Abstract (arXiv)

Large language models (LLMs) are known to perpetuate stereotypes and exhibit biases. Various strategies have been proposed to mitigate these biases, but most work studies biases as a black-box problem without considering how concepts are represented within the model. We adapt techniques from representation engineering to study how the concept of "gender" is represented within LLMs. We introduce a new method that extracts concept representations via probability weighting without labeled data and efficiently selects a steering vector for measuring and manipulating the model's representation. We develop a projection-based method that enables precise steering of model predictions and demonstrate its effectiveness in mitigating gender bias in LLMs and show that it also generalizes to racial bias. Our code is available at: https://github.com/hannahxchen/gender-bias-steering