3. How do LLMs work?
Week 3 (February 9)A conceptual and hands-on overview of how large language models work.
Agenda
- Present initial conjectures.
- Guest speaker: Sara Vogel
Readings
- Norman, D. A. (2013). Design thinking. In The design of everyday things. Basic Books. (Link)
- Norman, D. A. (2005). Human-centered design considered harmful. Interactions, 12(4), 14–19. (Link)
- DRAFT chapter from forthcoming book: Learning Centered Technology Design (Link)
Want to go deeper?
Neural networks: Zero to hero, the best introduction to machine learning, which claims to require only "basic knowledge of Python and a vague recollection of calculus from high school." This goes deeper than we will have time for in this course, but I'd be happy to support you if you want to dive in.
This course is taught by Andrej Karpathy, one of the most important people in creating our current AI moment. He co-founded OpenAI (the company behind ChatGPT) and led AI research at Tesla. What I particularly admire about him is that he has now shifted his focus to education, helping people learn about machine learning and AI.