I want to create a chatbot for domain specific application. For the same I want to use GPT 3.5 Turbo as base model, fine tuning on custom dataset and OpenAI embeddings. Kindly guide me what is exact procedure and pricing for the same
Want to use GPT 3.5 Turbo as base model, fine tuning on custom dataset and embeddings for a chatbot develpment
Hi there - you should take at the following guidance from OpenAI, which breaks down the process: https://platform.openai.com/docs/guides/fine-tuning
Note that since early November there is now also a dedicated user interface for finetuning, which makes the process for submission of files for training very easy: https://platform.openai.com/finetune
Pricing will really depend on the volume of training examples. Under the first link, the OpenAI team has provided an example for pricing.
I hope this helps as a starting point. Let me know if you have any specific question and I might be able to help further.