3 ms·

Question. How does one feed content to ChatGPT and get such output?

• swyx 4y ago
  not an expert, but i think usually one of two ways - either use openai’s finetune api (the right way) or embed all the docs, retrieve them using embedding of the questions, and add just enough of them to the prompt to generate a response (gpt-index or langchain help automate some of this). downside of former is lack of sources, downside of latter is context constraint
  • firasd 4y ago
    I'm not sure fine tuning helps with this kinda thing. Because the fine tuning mechanism involves sending the API a bunch of data like {"prompt": "<prompt text>", "completion": "<ideal generated text>"} So I'm not sure how you would feed it PG's essays in this mechanism
• sargstuff 4y ago
  chatgp isnt limited to spoonfuls, so guessing internet setup / tty typing speed. would tying infinate google search feeds to chatgp crash the internet or just reduce down to 42?