[2405.17345] Exploring and steering the moral compass of Large Language Models

arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website. Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them. Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs. arXiv Operational Status Get status notifications via email or slack

[2405.17345] Exploring and steering the moral compass of Large Language Models Skip to main content arXiv is now an independent nonprofit! Learn more × Search arXiv Press Enter to search · Advanced search --> Computer Science > Artificial Intelligence arXiv:2405.17345 (cs) [Submitted on 27 May 2024 ( v1 ), last revised 6 Jun 2024 (this version, v2)] Title: Exploring and steering the moral compass of Large Language Models Authors: Alejandro Tlaie View a PDF of the paper titled Exploring and steering the moral compass of Large Language Models, by Alejandro Tlaie View PDF HTML (exper

related reading

• Discovering Language Model Behaviors with Model-Written Evaluations — LessWronglesswrong.com
• Refusal in LLMs is mediated by a single direction — LessWronglesswrong.com
• [2602.09416] Are Language Models Sensitive to Morally Irrelevant Distractors?arxiv.org
• [2203.02155] Training language models to follow instructions with human feedbackarxiv.org
• Productizing Large Language Modelsblog.replit.com
• Steering Might Stop Working Soon — LessWronglesswrong.com
• [2303.17548] Whose Opinions Do Language Models Reflect?arxiv.org
• Things we learned about LLMs in 2024simonwillison.net
• Stuff we figured out about AI in 2023simonwillison.net
• LMCA_dataset.pdfandrew.cmu.edu
• [2607.14345] Value Leakage: An LLM's Answers Are Silently Shaped by Its Own Valuesarxiv.org
• [2605.03907] Steer Like the LLM: Activation Steering that Mimics Promptingarxiv.org