I'm unsure how to include n on the POST to get multiple completions. The documentation makes it sound like setting 10 would produce to versions of the completion. But I can't seem to make it affect the output at all, 1 and 10 behave the same.
const baseCompletion = await openai.createCompletion({ model: 'davinci:ft-123456789123456789', prompt: ${basePromptPrefix}${req.body.userInput}, temperature: 0.5, n: 10, max_tokens: 1000, stop: 'END' });