2 ms·

I guess you'd want to fine tune for content that wasn't already parsed before

• underlines 4y ago
  to my understanding there are 4 levels to add information: 1. train a model 2. fine tune a model 3. create embeddings for a model 4. use few shot prompt examples at inference time These have decreasing resource need, but also decreasing quality. For example, the GPT-3 API (not yet the GPT-4 API) has a functionality to send it your own embeddings, for example of your own source code documentation. Then you can query GPT-3 and it "knows" your source code doc and answers specifically with that in mind.
  • curl-up 4y ago
    Where in the API docs is this described?