AWS Certified Cloud Practitioner

Explanation

The correct answer is C. max-tokens.

Why max-tokens is the correct parameter:

• max-tokens controls the maximum number of tokens (words/sub-words) that the model can generate in its response
• When summaries are too long, it directly increases:
  • Latency: Longer responses take more time to generate
  • Token cost: More tokens = higher API costs

How adjusting max-tokens helps:

• Setting a lower max-tokens limit forces the model to generate more concise summaries
• This reduces both generation time and cost
• The model learns to be more efficient within the token constraint

Why other options are incorrect:

• A. temperature: Controls randomness/creativity of responses, not length
• B. top-p: Controls vocabulary diversity through nucleus sampling, affects quality not length
• D. stop-sequences: Used to define stopping points for generation, but doesn't directly control overall response length

Best Practice:

For healthcare chatbots where concise, accurate information is crucial, setting an appropriate max-tokens limit ensures responses remain focused and cost-effective while maintaining quality.