# Probing Ethical Framework Representations in Large Language Models: Structure, Entanglement, and Methodological Challenges

arXiv: https://arxiv.org/abs/2603.23659 · HTML: https://mlsanity.com/paper/2603.23659/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 17 of its 40 references that are in the map.
- First posted on arXiv 24 Mar 2026.

## Builds on (most cited first, 17 of 17)

- [Inference-Time Intervention: Eliciting Truthful Answers from a Language Model](https://mlsanity.com/paper/2306.03341/) (2023)
- [BERT Rediscovers the Classical NLP Pipeline](https://mlsanity.com/paper/1905.05950/) (2019)
- [Probing Classifiers: Promises, Shortcomings, and Advances](https://mlsanity.com/paper/2102.12452/) (2021)
- [Discovering Latent Knowledge in Language Models Without Supervision](https://mlsanity.com/paper/2212.03827/) (2022)
- [Aligning AI With Shared Human Values](https://mlsanity.com/paper/2008.02275/) (2020)
- [Transformer Feed-Forward Layers Build Predictions by Promoting Concepts in the Vocabulary Space](https://mlsanity.com/paper/2203.14680/) (2022)
- [Entity-Based Knowledge Conflicts in Question Answering](https://mlsanity.com/paper/2109.05052/) (2021)
- [Evaluating the Moral Beliefs Encoded in LLMs](https://mlsanity.com/paper/2307.14324/) (2023)
- [What do Neural Machine Translation Models Learn about Morphology?](https://mlsanity.com/paper/1704.03471/) (2017)
- [DisentQA: Disentangling Parametric and Contextual Knowledge with Counterfactual Question Answering](https://mlsanity.com/paper/2211.05655/) (2022)
- [The Moral Integrity Corpus: A Benchmark for Ethical Dialogue Systems](https://mlsanity.com/paper/2204.03021/) (2022)
- [Moral Mimicry: Large Language Models Produce Moral Rationalizations Tailored to Political Identity](https://mlsanity.com/paper/2209.12106/) (2022)
- [Probing the Moral Development of Large Language Models through Defining Issues Test](https://mlsanity.com/paper/2309.13356/) (2023)
- [Exploring and steering the moral compass of Large Language Models](https://mlsanity.com/paper/2405.17345/) (2024)
- [Unsupervised Contrast-Consistent Ranking with Language Models](https://mlsanity.com/paper/2309.06991/) (2023)
- [The Greatest Good Benchmark: Measuring LLMs' Alignment with Utilitarian Moral Dilemmas](https://mlsanity.com/paper/2503.19598/) (2025)
- [Right vs. Right: Can LLMs Make Tough Choices?](https://mlsanity.com/paper/2412.19926/) (2024)

## Abstract (arXiv)

When large language models make ethical judgments, do their internal representations distinguish between normative frameworks, or collapse ethics into a single acceptability dimension? We probe hidden representations across five ethical frameworks (deontology, utilitarianism, virtue, justice, commonsense) in six LLMs spanning 4B--72B parameters. Our analysis reveals differentiated ethical subspaces with asymmetric transfer patterns -- e.g., deontology probes partially generalize to virtue scenarios while commonsense probes fail catastrophically on justice. Disagreement between deontological and utilitarian probes correlates with higher behavioral entropy across architectures, though this relationship may partly reflect shared sensitivity to scenario difficulty. Post-hoc validation reveals that probes partially depend on surface features of benchmark templates, motivating cautious interpretation. We discuss both the structural insights these methods provide and their epistemological limitations.