Terence Tao on O1

Re:Terence Tao on O1 by roenxi on Saturday September 14, 2024 @10:20PM (#41544713)

You don't need to call them at 4am, you can keep a git log of the prompts that were used to generate the code and some professional 4am debugger can sit there and use an LLM to fix it.

Probably not a practical option yet, but if we're looking at the long term that is where we are heading. Or, realistically, the even longer term where the LLM self-heals broken systems.

[ Parent | on HN ]

Re:Terence Tao on O1 by __loam on Sunday September 15, 2024 @02:24AM (#41545624)

Lol, yeah the prompt is definitely going to help clarify what the code actually does.

[ Parent | on HN ]

Re:Terence Tao on O1 by dvfjsdhgfv on Sunday September 15, 2024 @05:53AM (#41546458)

While a git log of prompts seems like a novel idea to me, I don't believe it would work - not because of temperature and LLMs being non-deterministic and the context window overflowing, but because at a certain level of complexity LLMs simply fail, even though they are excellent at fixing simple bugs.

[ Parent | on HN ]