has anyone come across a scenario where openAI completion API returns the partial response ? The context has the entire informtion and api some how skips over last few lines and summarizes 70% of the context
by the way ..is there a way to set this number dynamically , based on token consumption in the prompt ? I guess I can count that prompt token first and then set a number , what do you think ? I feel fixed number might be limiting
I know text-davinci-003 let you set max_tokens to minus one (which functions like you mention), but they removed the feature from gpt-3.5/4 last time I checked 🤔
