// ⌖ Interpretability
Locating & Editing Facts (ROME)
Locates where facts are stored and edits those weights directly.
知识定位与编辑 ROME 2022 Advanced extended
IN PLAIN WORDS
Causal tracing shows facts are recalled by mid-layer MLPs at the subject's last token; a rank-one update to that MLP rewrites a fact.
Used in: Knowledge editing research
TIMELINE · prerequisites → this → what builds on it
01Original paper
Locating and Editing Factual Associations in GPT
Kevin Meng, David Bau, Alex Andonian, et al. · 2022
arXiv ↗ ✓ Verified · arXiv API
02Where it sits
Learn the left column first; the right column builds on this one.
Then you can learn
- Nothing in the atlas builds on it yet
Related:Activation Steering
Left: prerequisites · right: builds on this · dashed: related. Full map →
03Learning resources
Official docs, explainers, courses and reference code. All links checked.
Official docs
04Where the course teaches it
For AI / data
- Verification: verified, checked 2026-10-21.
- JSON: /api/learn/algorithms.json (id:
rome)