AI People Wiki

Builders & open source

Chris Olah

Co-founder, Anthropic

Chris Olah works on reverse engineering artificial neural networks into human-understandable algorithms and is one of the co-founders of Anthropic. Olah previously led interpretability research at OpenAI, worked at Google Brain, and co-founded Distill, a scientific journal focused on clear communication. At Anthropic, Olah is the corresponding author of "A Mathematical Framework for Transformer Circuits" (2021), last-listed author of "Toy Models of Superposition" (2022), and a co-author of "On the Biology of a Large Language Model" (2025), which studies the internal mechanisms of Claude 3.5 Haiku with circuit tracing.1234

At a glance

Current role
Co-founder
Company
Anthropic
Sector
Company
Work listed
3
Sources cited
5
Profile reviewed

Career

Chris Olah's academic CV lists research internships at Google Brain in 2014, hosted by Jeff Dean, and in 2015, hosted by Greg Corrado, working on techniques for visualizing neural networks. Olah then joined Google Brain as a research associate in 2015 and became a research scientist in 2016.5

In October 2018 Olah joined OpenAI as a member of technical staff and led a new team, called Clarity, working on neural network interpretability. Olah is one of the co-founders of Anthropic, where Olah continues to work on reverse engineering neural networks.51

Distill and explanatory writing

Olah co-founded Distill, a scientific journal focused on clear communication of machine learning research. Olah's Distill articles include "Attention and Augmented Recurrent Neural Networks" (2016) with Shan Carter, "Research Debt" (2017), "Feature Visualization" (2017), "The Building Blocks of Interpretability" (2018) and "Activation Atlases" (2019).15

Interpretability research

At Anthropic, Olah's work on the Transformer Circuits Thread includes "A Mathematical Framework for Transformer Circuits" (2021), which reverse engineers small attention-only transformers, and "Toy Models of Superposition" (2022), which studies how models store additional sparse features in superposition, a source of polysemantic neurons.23

Olah is also a co-author of "On the Biology of a Large Language Model" (2025), which uses attribution graphs to trace internal mechanisms in Claude 3.5 Haiku. Earlier, Olah was a co-author of "Concrete Problems in AI Safety" (2016) and of the TensorFlow systems paper (2015).45

Work and contributions

  • A Mathematical Framework for Transformer Circuits Published research

    Corresponding author of the 2021 Anthropic article on the Transformer Circuits Thread, which develops a framework for reverse engineering small attention-only transformers.2

  • Toy Models of Superposition Published research

    Last-listed author of the 2022 paper, which gives a toy model in which polysemantic neurons arise from models storing additional sparse features in superposition.3

  • On the Biology of a Large Language Model Published research

    Co-author of the 2025 Anthropic study; its contributions statement credits Olah with work on the replacement model and attribution-graph computation, intervention analyses, and crosscoder ideas.4

Timeline

  1. Began a research internship at Google Brain hosted by Jeff Dean, creating the meta-SNE visualization algorithm.5
  2. Joined Google Brain as a research associate, becoming a research scientist in October 2016.5
  3. Published "Attention and Augmented Recurrent Neural Networks" with Shan Carter on Distill.5
  4. Published "Feature Visualization" on Distill with Alexander Mordvintsev and Ludwig Schubert.5
  5. Joined OpenAI as a member of technical staff and led the Clarity interpretability team.5
  6. Published "A Mathematical Framework for Transformer Circuits" as corresponding author.2
  7. Published "Toy Models of Superposition" as last-listed author.3
  8. Co-authored "On the Biology of a Large Language Model" on the Transformer Circuits Thread.4

Other documented roles

Roles the sources document that are not confirmed as current.

Where they work

Suggest a correction or removal

Profiles cover public professional work only, and each statement cites a source. If something is out of date, missing or wrong, tell us and link a source. People can also ask for their profile to be corrected or removed.

References

5 sources · profile reviewed Oct 21, 2026 · data updated Oct 21, 2026

  1. About Me — colah's blogChristopher Olah · Official page · accessed Oct 21, 2026
  2. A Mathematical Framework for Transformer Circuits (Transformer Circuits Thread, Anthropic)Anthropic · Paper · published Dec 22, 2021 · accessed Oct 21, 2026
  3. Toy Models of Superposition (arXiv 2209.10652)arXiv · Paper · published Sep 21, 2022 · accessed Oct 21, 2026
  4. On the Biology of a Large Language Model (Transformer Circuits Thread, Anthropic)Anthropic · Paper · published Mar 27, 2025 · accessed Oct 21, 2026
  5. Chris Olah — academic CV (PDF)Chris Olah · Official page · accessed Oct 21, 2026