Keyboard shortcuts

Press or to navigate between chapters

Press S or / to search in the book

Press ? to show this help

Press Esc to hide this help

Exercises

Count the Tokens
Write 150 words about your favorite character. Run it through a tokenizer. Compare words vs. tokens.

Token Rent Audit
Paste your Personality + Scenario into a counter. Is your “rent” under 1,800 tokens? Trim if not.

Memory Curve Test
Paste 2,000 words into a chatbot. Ask about the start vs. end. Observe recency vs. anchor bias.

Shrink to Improve
Take a 300-word backstory. Rewrite into 100 words of functional reminders. Test both.

Send Order Drill
Write one Personality line, one Scenario line, one Advanced Prompt, and one user message. Put them in order. Which dominates?

✅ End of Chapter 1
In this chapter, you learned how the model sees the world: tokens, context windows, pretraining, and prompt order. In Chapter 2, we will dig into token efficiency — how to write prompts that stay lean, clear, and powerful without wasting space.