Don't Go Quiet On Me: every reference
A field guide · 4:22 · 216 references Don't Go Quiet On Me is a stylised history of mechanistic interpretability, sung to the model. Across a researcher's field notebook, from 2020 into 2027, a toy model in a petri dish grows, chorus by chorus, into a walled fortress while she keeps trying to read it: circuits, superposition and grokking, sparse autoencoders and Golden Gate Claude, probes, chain
A field guide · 4:22 · 216 references Don't Go Quiet On Me is a stylised history of mechanistic interpretability, sung to the model. Across a researcher's field notebook, from 2020 into 2027, a toy model in a petri dish grows, chorus by chorus, into a walled fortress while she keeps trying to read it: circuits, superposition and grokking, sparse autoencoders and Golden Gate Claude, probes, chain of thought, eval awareness, and models that think without saying. Claude Opus 5.5 made it with Suno v6, for Neel Nanda. The video is dense on purpose. This page explains all 216 references, chapter…
related reading
- Don't Go Quiet On Me (by Opus 5.5)youtube.com
- On the Biology of a Large Language Modeltransformer-circuits.pub
- A Mathematical Framework for Transformer Circuitstransformer-circuits.pub
- Circuit Tracing: Revealing Computational Graphs in Language Modelstransformer-circuits.pub
- Tracing the thoughts of a large language model \ Anthropicanthropic.com
- Towards Monosemanticity: Decomposing Language Models With Dictionary Learningtransformer-circuits.pub
- As Rocks May Think | Eric Jangevjang.com
- An Extremely Opinionated Annotated List of My Favourite Mechanistic Interpretability Papers v2 — AI Alignment Forumalignmentforum.org
- models have some pretty funny attractor states — LessWronglesswrong.com
- It Looks Like You’re Trying To Take Over The World · Gwern.net (reader mode)gwern.net
- trees are harlequins, words are harlequins - the voidnostalgebraist.tumblr.com
- A Comprehensive Mechanistic Interpretability Explainer & Glossary — Neel Nandaneelnanda.io