LLMs can be specialized to a task without any fine-tuning where the prompt acts as a form of in-context learning approach, how does that work and what seems to happen internally in the network?
In: 0
It’s fascinating! When LLMs are prompted without fine-tuning, they can still learn and specialize in a task by using the prompt as a way of understanding the context
It’s like they have their own way of learning and adapting internally within the network.
My understanding of the technology, and one of the “scary” things, is that those who created them cannot tell you how they do what they do. I don’t think there is currently an answer to this question.
Latest Answers