[kewp]

are these LLMs just answering the question "if you found this text on the internet (the prompt) what would most likely follow" ?

[Enginerrrd]

In essence, yes I think, but... isn't that essentially not much different than what I'm doing in making this comment?

[sebzim4500]

That's how they are trained initially, but the resulting model isn't all that useful (was SOTA two years ago but this field moves fast).

A lot of the utility comes from the later finetuning. You can see this using the examples from the article, every mistake they identify with GPT-3 (which is the unfinetuned version) is answered correctly by chatGPT, which has gone through an extensive finetuning process called RLHF.

[colechristensen]

Yes, they are being trained, to simplify, to complete sentences. You can then use the resulting model to do lots of things.

How you train a model and the inference jobs it can do don't necessarily have to be the same.

[astrange]

That's how the text decoder works, but the model gets to define "most likely" and an RLHF model uses this to make the text decoder produce useful answers instead.