Mechanistic Transparency for Machine Learning
[EDIT (added Jan 2023): it’s come to my attention that this post was likely influenced by conversations I had with Chris Olah related to the distinction between standard interpretability and the type called “mechanistic”, as well as early experiments he had which became the ‘circuits’ sequence of papers - my sincere apologies for not making this clearer earlier.]
Mechanistic Transparency for Machine Learning Mechanistic Transparency for Machine Learning Jul 10, 2018 Cross-posted to the AI Alignment Forum . [EDIT (added Jan 2023): it’s come to my attention that this post was likely influenced by conversations I had with Chris Olah related to the distinction between standard interpretability and the type called “mechanistic”, as well as early experiments he had which became the ‘circuits’ sequence of papers - my sincere apologies for not making this clearer earlier.] Lately I’ve been trying to come up with a thread of AI alignment research that (a) I can
related reading
- How To Become A Mechanistic Interpretability Researcher — AI Alignment Forumalignmentforum.org
- An Extremely Opinionated Annotated List of My Favourite Mechanistic Interpretability Papers v2 — AI Alignment Forumalignmentforum.org
- An Ambitious Vision for Interpretability — AI Alignment Forumalignmentforum.org
- A Pragmatic Vision for Interpretability — AI Alignment Forumalignmentforum.org
- Alignment remains a hard, unsolved problem — LessWronglesswrong.com
- An Extremely Opinionated Annotated List of My Favourite Mechanistic Interpretability Papers v2 — AI Alignment Forumalignmentforum.org
- On Optimism for Interpretabilitygoodfire.ai
- What is the purpose of interpretability?ericjmichaud.com
- God Help Us, Let’s Try To Learn About Mechanistic Interpretability Techniquesastralcodexten.com
- Neurotechnology is Critical for AI Alignmentmilan.cvitkovic.net
- The Building Blocks of Interpretabilitydistill.pub
- A gentle introduction to mechanistic anomaly detection — LessWronglesswrong.com