Notable Links
Work
Tweets
- “LLMs work best by simulating engaged people”
- You won’t measure what you don’t understand
- on attempts to commodify the act of prompting a language model
- on agency, in response to Michael Nielsen
- thoughts on the Waluigi Effect, Feb. 2023
- janus’s favorite book
- a chat about a FUSE filesystem which lazily renders nested contents with LLMs
Reposts
- on being the first person to use “the right tool for a task”
