Miscellaneous Commentary:.
- GPT3 by including data not used in the study as reference material in the prompts.
- Up-to-date information that did not exist at the time of the study
- Data that cannot be made accessible on the Internet for copyright reasons, such as e-books
- Items that cannot be disclosed outside the company, such as customer contact history data in groupware
- 〜Mechanism that allows GPT3 to use ~, etc.
- Typical usage:.
- Create embedded vector from a document at hand with OpenAI API
- For a query sentence, use vector search to find related documents.
- Embed that document in the prompt to GPT3
repo: GitHub - jerryjliu/gpt_index: LlamaIndex (GPT Index) is a project that provides a central interface to connect your LLM’s with external data. docs: Welcome to LlamaIndex 🦙 (GPT Index)! — LlamaIndex documentation 3rd party blog: LlamaIndex Quick Start Guide｜npaka｜note
old name: GPT-Index
This page is auto-translated from /nishio/LlamaIndex using DeepL. If you looks something interesting but the auto-translated English is not good enough to understand it, feel free to let me know at @nishio_en. I’m very happy to spread my thought to non-Japanese readers.