// ⌖ Interpretability
Circuits & Induction Heads
Finds circuits implementing algorithms, like induction heads that copy patterns.
电路与归纳头 2022 Advanced core
IN PLAIN WORDS
Two attention heads form an induction head: after A B … A, predict B. They appear abruptly during training, coinciding with in-context learning.
Used in: Mechanistic interpretability research
TIMELINE · prerequisites → this → what builds on it
01Original paper
In-context Learning and Induction Heads
Catherine Olsson, Nelson Elhage, Neel Nanda, et al. · 2022
arXiv ↗ ✓ Verified · arXiv API
02Where it sits
Learn the left column first; the right column builds on this one.
Circuits & Induction Heads
Then you can learn
Related:Superposition
Left: prerequisites · right: builds on this · dashed: related. Full map →
03Learning resources
Official docs, explainers, courses and reference code. All links checked.
Papers
Explainers
Reference code
04Where the course teaches it
For AI / data
- Verification: verified, checked 2026-10-21.
- JSON: /api/learn/algorithms.json (id:
induction-heads)