News for hackers. Stuff that matters.

A comment on StackLlama: A hands-on guide to train LlaMa with RLHF.

Re:StackLlama: A hands-on guide to train LlaMa with RLHF by scottydog51834 on Friday April 07, 2023 @01:42PM (#35484507)
I'd love a tool where I can upload a private dataset and RLHF a model (even better if the tool provides the pre-trained model) without me having to worry about GPUs, memory, commercial access, or even writing any python code. I'd happy pay several hundred and maybe several thousands for access to this.
[ Parent | on HN ]
Re:StackLlama: A hands-on guide to train LlaMa with RLHF by Tepix on Friday April 07, 2023 @02:11PM (#35484855)
Even better would be a crowdfunding platform where you can collect some money from a bunch of people and then use it to fine-tune the bigger models (30B and 65B in the case of LLaMA).
[ Parent | on HN ]
Re:StackLlama: A hands-on guide to train LlaMa with RLHF by chankstein38 on Friday April 07, 2023 @02:51PM (#35485465)
OpenAI allows you to fine tune any of their LLM models. I'm not sure if your requirements were "without paying OpenAI something" but if you are ok with that, fine tuning with them is pretty straightforward. IANAL or anything but if I recall they even promise they won't use your data for their training.
[ Parent | on HN ]
Re:StackLlama: A hands-on guide to train LlaMa with RLHF by sebzim4500 on Friday April 07, 2023 @03:25PM (#35485910)
They let you do this with their GPT-3 derived models, but not the GPT-3.5 or GPT-4 ones.

That would be unsafe lol

[ Parent | on HN ]
Re:StackLlama: A hands-on guide to train LlaMa with RLHF by ttul on Friday April 07, 2023 @03:39PM (#35486059)
I think that tool would have broad appeal, but I wonder if the most likely case is that it would be buried inside other, higher-level systems, such as customer support automation SaaS.
[ Parent | on HN ]