Syntax checking and comments in Fine-Tune JSONL

API
1

Hi Community,
Two (hopefully easy) questions regarding fine tuning JSONL dataset files.

What is the generally accepted best practice method for:

  1. Adding comments into JSONL files given we don’t want bad data to be imported into our fine-tuned models?
  2. Syntax checking the file before sending it off using the fine_tunes.create?

Sorry if these seem like silly questions, I’m still very new to all this.

2

JSON does not allow comments.

If you need to do this, you might consider storing your JSON in a JavaScript file as it will allow you to put comments wherever.

You will still have to export the data in the JS file to the proper JSONL format before using it for the fine-tuning, though.