Topics | Learn MI

Browse a structured mechanistic interpretability curriculum, from transformer foundations through causal methods, features, circuits, tools, and AI safety.

Topics Explore 84 articles across 17 blocks. Follow the suggested order or jump straight to the topic you need. 84 articles Each topic is a focused, standalone deep dive that still fits the larger sequence. 17 learning blocks Topics are grouped by theme so you can build intuition before moving to advanced tools. Suggested order Read top-to-bottom for a guided path, or jump into any block whenever you need. Curriculum Map Browse every block and topic in the recommended sequence. 1 Transformer Foundations 10 topics Prerequisites Transformer Architecture Intro Embeddings…

saved by

• Bhagyesh Kumar

related reading

• Neuronpedianeuronpedia.org
• Goodfire AIgoodfire.ai
• Transformer Circuits Threadtransformer-circuits.pub
• An Extremely Opinionated Annotated List of My Favourite Mechanistic Interpretability Papers v2 — AI Alignment Forumalignmentforum.org
• Transformer Explainer: LLM Transformer Model Visually Explainedpoloclub.github.io
• LLM Visualizationbbycroft.net
• Scaling Monosemanticity: Extracting Interpretable Features from Claude 3 Sonnettransformer-circuits.pub
• Parsed | Custom, interpretable AI systems that continuously learnparsed.com
• An Extremely Opinionated Annotated List of My Favourite Mechanistic Interpretability Papers v2 — AI Alignment Forumalignmentforum.org
• Natural Language Autoencoders Produce Unsupervised Explanations of LLM Activationstransformer-circuits.pub
• A Comprehensive Mechanistic Interpretability Explainer & Glossary — Neel Nandaneelnanda.io
• Datacurve | The data engine for frontier AIdatacurve.ai