Inference Parameters Quiz
5 questions Pass: 70% +25 pts (50 with Pro)
Quiz covering Foundation Model Design
What you'll be asked
- Which parameter controls the randomness of the model's output by flattening or sharpening the probability distribution of the next token?
- You notice that your model is frequently repeating the same phrases in long-form generation. Which parameter should you adjust to discourage this behavior?
- When using Top-P (Nucleus) sampling with a value of 0.9, how does the model select the next token?
- A user wants a model to be strictly factual and avoid creative hallucinations. Which combination of settings is most appropriate?
- If you set 'Top-K' to 50 and 'Top-P' to 0.9 simultaneously, how does the inference engine typically apply these constraints?
Answers and explanations come after you submit.
Course Content