A report on an experiment I conducted for Boaz Barak's "CS 2881r: AI Alignment and Safety" at Harvard. I evaluate the effects of different system prompt types (minimal vs. principles vs. rules) and their combinations on over-refusal, toxic-refusal, and capability benchmarks.
An essay covering one of my favorite units in Prof. Mara Bollard's Phil 302 - Moral Psychology. Hopefully you can learn a few new frameworks to think about understanding and evaluating progress.
This is an essay I wrote for Phil 345: Language and Mind taught by Professor James Joyce. I loved this class and liked this essay.
I am extremely interested in the connections between physics and machine learning. This is a work in progress combining the ideas and connections I've been thinking about and filling in the (many) gaps in my knowledge.