I have my companies policy document. which contains all of the information about their policy, allowances (.gym, health, travel, dinner ). on some prompts the model give the output but most of the cases the model is giving me unexpected outputs. what I’ve done is, I gave chunks of my pdfs to the openai to parse it and convert it into a format in which openAI accepts for fine tunning. the training data is created fine. but after the fine tunning. it’s not giving me porper results. I’m stuck and need your help if anyone can.
Hi and welcome to the Developer Forum!
Fine-tuning is not intended for what you are trying to achieve. Fine-tuning is primarily meant to get the model respond in a certain style or format, or to get it to perform a task in a specific way. You should look into RAG-based solutions instead (see also here for an explanation for the difference).
One straightforward option if you‘d like to use the API is OpenAI‘s Assistant.
2 Likes