Hi, I’m Daphne.

I’m currently learning machine learning, with a focus on AI safety and interpretability. This site is where I document what I’m learning — the concepts that click, the ones that don’t, and everything in between.

I believe in learning in public: sharing knowledge as it’s acquired, not after it’s perfected.

What I’m working on

• Working through the ARENA curriculum (AI safety research training)
• Building fluency with NumPy and PyTorch
• Studying transformer interpretability and mechanistic interpretability

Elsewhere

• GitHub

This site is built with Hugo and the Risotto theme.