Chat GPT / Davinci API for long responses

I"m using the api to generate long responses and save them to a file. The prompt is great, it provides the exact response we need - but it’s limited by the number of tokens, so it sometimes stops mid sentence.

In playground or on the chat gpt interface, i can tell it to continue and it does.

But how can we do that with the API?

If the original prompt is long, on thing you can try is removing it and providing the incomplete output with a simple “Continue:” prompt.