How LLMs Workbeginner11 min · 9 steps
Tokenization Deep Dive
Models read subword tokens, not letters or words — and that one fact drives cost, context limits, multilingual fairness and some famous failures.
datatoken
ProblemStep 1 of 9
The model never sees the letters
Ask a model “How many r’s are in strawberry?” and it often gets it wrong. Not because it can’t count — because it never saw the letters.
A tokenizer turns the word into subword pieces like str · aw · berry. Counting letters means reasoning about the inside of tokens the model only knows as whole units.
TipWorkaround from the guide: ask the model to spell the word out character by character first, then count. That forces one token per letter.
Lesson 2 of 6 in How LLMs Work