# [Mei Lin](https://longwake.org/people/mei-lin) (she/her)

Mechanistic interpretability of planning in language models

- **Affiliations:** Research lead, Lattice Interpretability Lab
- **Location:** Berlin
- **Topics:** Mechanistic interpretability, Scheming and deception
- **Open to:** Reviewing drafts, Mentoring, Hiring
- **Contact:** Open to messages
- **Signal:** mei.42
- **Book a call:** https://example.org/book/mei
- **Links:** [alignment forum](https://example.org/af/mei), [github](https://example.org/gh/mei)
- **Selected work:**
  - [Finding goal representations in a 7B model](https://example.org/posts/goal-reps) (2026)

I look for internal representations of goals and plans. Happy to review interpretability drafts.

_Sample profile for development; not a real person._