Get The Stanford LLM Watch Guide

The companion to “Skip the 2-Hour Show: These 2 Free Stanford Lectures Explain How ChatGPT Works”. PAGE 1 — why these two lectures, the links to both (Lecture 1 Transformers, 1:44:28; Lecture 2 Large Language Models, 1:43:15), who teaches them (twin brothers Afshine and Shervine Amidi) and a two-evening plan. PAGES 2–3 — every Lecture 1 chapter with a clickable timestamp, a WATCH / SKIM / SKIP verdict and plain-English notes. PAGES 4–5 — the same for all 19 Lecture 2 chapters, from BERT and GPT to mixture of experts, temperature and context rot. PAGE 6 — attention worked by hand with small numbers, plus the real GPT-4o tokenizer counts (English 7 tokens, Hindi 12, strawberry = 1 token). PAGE 7 — the five questions from the video with full answers, and a plain-words glossary. Facts checked on October 18, 2026. Independent guide from Hyperautomation Labs — not affiliated with Stanford, Stanford Online or the instructors.

By submitting, you'll be subscribed to the Hyperautomation AI Report newsletter — three practical AI emails a week. Unsubscribe anytime.

Free. No spam. Unsubscribe anytime.