Interpretability Researcher at Google DeepMind. ex-director of the ARENA program.
ARENA_3.0. Jupyter Notebook
1.2ksae_vis. Create feature-centric and prompt-centric visualizations for sparse autoencoders (like those from Anthropic's published research).
262ARENA_2.0. Resources for skilling up in AI alignment research engineering. Covers basics of deep learning, mechanistic interpretability, and RL.
246computational-thread-art. Code that renders an image as a series of lines connecting pins around a circular frame (for more detail, see my Medium page).
140sae_visualizer. HTML
31sae-exercises-mats. Python
26TransformerLens-intro. HTML
20path_patching. Implementation of path patching & activation patching (will eventually add to TransformerLens).
15computational-thread-art-v2. Updated version of my computational thread art algorithm, which also includes code for generating full-color images.
10SERI-MATS-2023-Streamlit-pages. Repo for hosting Streamlit pages for my 2023 SERI MATS project with Arthur Conmy (mentored by Neel Nanda).
9CircuitsVis. Mechanistic Interpretability Visualizations using React
8jupyter-to-anki. Python functions to convert Jupyter Notebooks into Anki cards.
7eindex. My interpretation of what einops indexing would look like (created to work on during my SERI MATS project).
6arena-heroku. HTML
6anki_templates. Contains templates for Anki cards, and code to auto-generate Anki cards (coming soon)
5computational-thread-art-v3. Jupyter Notebook
5arena-pragmatic-interp. Jupyter Notebook
5SERIMATS_app. Application for Neel's SERI MATS stream, 2023.
3arena-v1. HTML
3plotly-widgets. Jupyter Notebook
2SAE_Bench_Template. Python
2Fundamentals. HTML
1arena-v1-ldn-exercises-new. HTML
1copy-suppression. HTML
1