Ask HN: How do I train a custom LLM/ChatGPT on my own documents?

Re:Ask HN: How do I train a custom LLM/ChatGPT on my own doc... by npsomaratna on Sunday July 23, 2023 @01:13PM (#36837385)

Re: step 3, I think you mean text-embedding-ada-002—OpenAI's current embeddings model, which replaces all 15 (or is it 16?) first generation embeddings models.

With respect to open source embeddings models, instructor-xl is the state of the art currently—as effective as text-embedding-ada-002.

That said, instructor-xl has a context length of 512 tokens, while text-embedding-ada-002 has a context length of 8192 tokens, which is markedly more convenient.

Last but not least, parent's comment re: langchain is spot on. It's simple and straightforward to write these few lines of code yourself.

[ Parent | on HN ]

Re:Ask HN: How do I train a custom LLM/ChatGPT on my own doc... by KRAKRISMOTT on Sunday July 23, 2023 @02:47PM (#36838373)

You are right, my mistake, it should be text-embedding-ada. Thanks for the info about instructor, will check it out. What do you usually use LLMs for?

[ Parent | on HN ]

Re:Ask HN: How do I train a custom LLM/ChatGPT on my own doc... by npsomaratna on Sunday July 23, 2023 @11:00PM (#36842641)

Mainly for data processing, and almost exclusively "open-source" LLMs now (essential given our focus on industries with tons of confidential data). Let me plug my startup's website, which describes what we do: https://www.amw.ai/

[ Parent | on HN ]

Re:Ask HN: How do I train a custom LLM/ChatGPT on my own doc... by ianpurton on Monday July 24, 2023 @02:25AM (#36843762)

From you home page, I couldn't figure out exactly what you do.

Can you parse PDF into embeddings for a vector database?

[ Parent | on HN ]

Re:Ask HN: How do I train a custom LLM/ChatGPT on my own doc... by kirdiekirdie on Thursday July 27, 2023 @11:19AM (#36894906)

localGPT can parse PDF into embeddings, see <https://github.com/PromtEngineer/localGPT>.

[ Parent | on HN ]