Tell
Andrej Karpathy: "99% Use AI & LLMs Wrong" (Here’s How He Does It)
Andrej Karpathy's lecture provides a mental model for navigating the current landscape of Large Language Models (LLMs).
Karpathy uses the "Zip File" analogy to explain the probabilistic nature of LLM memory: the model "read the internet... and I only remember vaguely"
Link: https://youtu.be/7YEHs2SMaVQ