Pacific Design/ artificial intelligence

section 02 · status: live · 6 entries · updated 2026-08-27

Large Language Models

The technology that pulled AI into everyday life. How transformers actually work, what tokens are, how a raw internet-scale model becomes an assistant, why models hallucinate, and what scale does and doesn't buy.

live · attention weighing the context, then sampling the next token

  1. Transformers & attentionevery word looks at every other word
  2. Tokens & tokenizationthe atoms of model perception
  3. How LLMs are trainedfrom internet prediction to instruction following
  4. Context windows & memorythe model's whole world, measured in tokens
  5. Hallucinationfluent, confident, and wrong by design
  6. Scaling laws & small modelspredictable gains, surprising arrivals, shrinking costs