LLMs & Tokens Explained

What an LLM actually is, what a token is, why context windows matter, and why model choice changes the answer.

TourM1 -- Wed Apr 29, ~20 min

What we cover

LLMs in plain English -- next-word prediction trained on a lot of text
What a token is, why it matters for cost and quality
Context windows: 200k vs 1M, when each matters
Why model choice (Opus / Sonnet / Haiku) changes the answer you get back

Why it matters

You'll spend the rest of the program writing prompts. Knowing what the system is actually doing under the hood makes your prompts five times sharper without any extra effort.

Hands-on moment

Open Claude.ai. Ask the same question two ways -- once with a 50-word context, once with a 500-word context. Notice the difference in the answer. Discuss with a partner what changed.

Related artifacts

pre-readM1 Pre-read../../public-site/github-pages/course-pages/module-1-pre-read/index.html
deckM1 Kickoff Deck (slides 9-16)../../00-current-workshop/m1-kickoff-master.pptx

Source files live alongside this site under clients/greenridge-growth/; paths above are relative to the syllabus-site root.