On 13 July 2021 OpenAI enabled fine-tuning for all users who have API access. This feature is currently in beta, so some parameters most probably will changed.
The idea from OpenAI is that fine-tuning of this nature afford users the opportunity to train a model, which will should yield answers in keeping with the training data.
All tests were performed using the OpenAI CLI (Command Line Interface). In some instances cURL, the Playground or Python code can be used. However, the OpenAI CLI lends the best structure to the training process.
Once a model has been fine-tuned, you won’t need to provide examples in the prompt anymore.
For example training a general purpose chatbot could require minimal training data.
Perhaps 20 examples per intent. However, when creating a data set for training it is advised that you use a few hundred or more training examples.
For classification at least 100 training examples are required per class, for some training examples more than 500 records of training data is demanded.
This is not in keeping with other environments like Rasa, IBM Watson Assistant, Microsoft LUIS etc., where astounding results can be achieved with relative few training examples.
At a high level, fine-tuning involves the following steps:
- Prepare and upload training data
- Train a new fine-tuned model
- Use your fine-tuned model
The potential for using the powerful black box of GGPT-3 for a highly customized personal assistant now seems within reach. I can imagine training a version to respond like me to kblog comments or a chat bot to answer common Angular development questions. The possibilities seem endless!
Learn more at openAi here