How to fine-tuning with a no-code app?

Hello everyone :smiling_face: !

Can you help me better understand how to fine-tune chatgpt using a no-code app (in particular bubble. io) ?

Thanks !

Welcome to the community @_Alina

What are you trying to fine-tune for?

Thanks Sukhman :smiling_face:

I’m trying to fine-tune my chatgpt model on bubble .io.
So I had to connect the chatgpt api to bubble to be able to connect it with the app I made.
But the problem is that I’d like to fine-tune this model in curl and json but I can’t manage to do it and in particular to host my json file online so that it can be retrieved from the curl and json commands.
Do you understand what I’m trying to do?

Thanks for the reply. That part was clear in your first post.


What I was curious about was your use-case for the fine-tuned model.

Here’s how the fine-tuning process after the creation of JSONL file looks like:

  1. The file is uploaded with the purpose ‘fine-tune’ to the OpenAI’s cloud.
  2. The id for the file is passed along with hyperparams to the fine-tune endpoint to request a fine-tune job.
  3. Once the job is created successfully, it is enqueued and results in a successful or failure status. The job can also be cancelled by the user before fine-tune process starts.

Okay thanks
But how do I download the file so that it’s in the cloud?

Use the /Files endpoint.

Ok so can I retrieve the url of the upload file ?

Can we have a little bit more discussion on this topic? How exactly are you uploading the files to OpenAI? I’m currently using Embeddings in order to interact with CHatGPT when I have a large PDF file, but I don’t think that is the right way of doing it. Please show me how I can use this method you just explained, I want to fine-tune the model for a question-answering chatbot that I’m currently developing. The question-answering chatbot should be fine-tuned on a large PDF file.

Also, where can I find the fine-tune endpoint?

I am curious about this topic too. Is there more to learn how to do this?