[2301.05217] Progress measures for grokking via mechanistic interpretability

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

[2301.05217] Progress measures for grokking via mechanistic interpretability Skip to main content arXiv is now an independent nonprofit! Learn more × Search arXiv Press Enter to search · Advanced search --> Computer Science > Machine Learning arXiv:2301.05217 (cs) [Submitted on 12 Jan 2023 ( v1 ), last revised 19 Oct 2023 (this version, v3)] Title: Progress measures for grokking via mechanistic interpretability Authors: Neel Nanda , Lawrence Chan , Tom Lieberum , Jess Smith , Jacob Steinhardt View a PDF of the paper titled Progress measures for grokking via mechanistic interpretab

related reading

• [2301.05217] Progress measures for grokking via mechanistic interpretabilityarxiv.org
• [2201.02177] Grokking: Generalization Beyond Overfitting on Small Algorithmic Datasetsarxiv.org
• Do Machine Learning Models Memorize or Generalize?pair.withgoogle.com
• [2406.11717] Refusal in Language Models Is Mediated by a Single Directionarxiv.org
• [2512.19428] Attention Is Not What You Needarxiv.org
• Towards Automated Circuit Discovery for Mechanistic Interpretabilityarxiv.org
• [2309.01809] Are Emergent Abilities in Large Language Models just In-Context Learning?arxiv.org
• Mechanistic Interpretability of Binary and Ternary Transformersarxiv.org
• [2406.17583] Towards Compositional Interpretability for XAIarxiv.org
• [2406.02061] Alice in Wonderland: Simple Tasks Showing Complete Reasoning Breakdown in State-Of-the-Art Large Language Modelsarxiv.org
• [2305.18654] Faith and Fate: Limits of Transformers on Compositionalityarxiv.org
• A Mechanistic Interpretability Analysis of Grokking — LessWronglesswrong.com