Tokenizers and context budget (step 7/7) · cnns, transformers, and useful llm internals

Checkpoint

One last thing before we move on. Same surface as a write step — but the lesson doesn't complete until this passes.

Checkpoint. Size a prompt against the budget before sending. Write budget_check(messages, budget) that totals the tokens (word-split stand-in) and returns:

7/5 tokens: over budget

Format {used}/{budget} tokens: {verdict} where verdict is fits if used <= budget, else over budget.

Checkpoint

One last thing before we move on. Same surface as a write step — but the lesson doesn't complete until this passes.

Checkpoint. Size a prompt against the budget before sending. Write budget_check(messages, budget) that totals the tokens (word-split stand-in) and returns:

7/5 tokens: over budget

Format {used}/{budget} tokens: {verdict} where verdict is fits if used <= budget, else over budget.

full-screen editor opens — close anytime to keep reading.

Tokenizers and context budget — step 7 of 7

Tokenizers and context budget — step 7 of 7