Supervised Fine Tuning

Another technique used in pre-training AI is called Supervised Fine Tuning, or SFT for short.

With this technique, a contributor, in other words, a human element, gets to review, improve, or re-engage the conversation in a single session with the model, but generally is just reviewing what an actual conversation looks like. The conversation having taken place prior between another human and the model.

Leave a Reply

Your email address will not be published. Required fields are marked *