Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

How to Finetune GPT-Like Large Language Models on a Custom Dataset

lightning.ai

Post reply on HN
Print view View original

21–30 of 126 posts

Re: How to Finetune GPT-Like Large Language Models on a Custom Dataset

#21

Post by ramesh1994 »

fallingmeat wrote: post #6
artembugara wrote: post #2

Have a question to the Generative AI experts here. So, I can use smthg like GPT-4 to label data and then use that as a train set for my own LLM, right? EDIT: adding this from OpenAI Restriction TOS: "(iii) use output from the Services to develop models that compete with OpenAI;"

That is against their ToS though if you use your new LLM commercially.

It prohibits anything that competes with OpenAI services i.e as long as you're not literally providing an LLM API commercially you should be fine

Re: How to Finetune GPT-Like Large Language Models on a Custom Dataset

#22

Post by artembugara »

foobarbecue wrote: post #17
artembugara wrote: post #2

Have a question to the Generative AI experts here. So, I can use smthg like GPT-4 to label data and then use that as a train set for my own LLM, right? EDIT: adding this from OpenAI Restriction TOS: "(iii) use output from the Services to develop models that compete with OpenAI;"

Is "ca" "can" or "can't"?

can

Re: How to Finetune GPT-Like Large Language Models on a Custom Dataset

#23

Post by postsantum »

pmoriarty wrote: post #7
fallingmeat wrote: post #6

Earlier quoted context omitted.

That is against their ToS though if you use your new LLM commercially.

So what are they going to do about it?

MS lawyers have a good track record at sending out those scary cease&desist letters

Re: How to Finetune GPT-Like Large Language Models on a Custom Dataset

#25

Post by mgfist »

slenocchio wrote: post #24

Can someone explain why I'd want to use fine-tuning instead of a vector database (or some other way of storing data/context)?

First reason that comes to mind is you can make much smaller models, which helps with latency, cost and may enable you to run the model locally.

Re: How to Finetune GPT-Like Large Language Models on a Custom Dataset

#26

Post by pid-1 »

slenocchio wrote: post #24

Can someone explain why I'd want to use fine-tuning instead of a vector database (or some other way of storing data/context)?

I've been playing with using documents as OpenAI embeddings for the past weeks and, at least for my use case, the results are meh. It seems sometimes just using context is not enough.

My next step is to play with fine tunning, but I have no results to report yet.

Re: How to Finetune GPT-Like Large Language Models on a Custom Dataset

#27

Post by notpublic »

artembugara wrote: post #2

Have a question to the Generative AI experts here. So, I can use smthg like GPT-4 to label data and then use that as a train set for my own LLM, right? EDIT: adding this from OpenAI Restriction TOS: "(iii) use output from the Services to develop models that compete with OpenAI;"

not an AI expert but from a talk I recently heard... if there is a mismatch in training data between the "teacher" LLM and "student" LLM, you risk teaching the student to hallucinate or to ignore information

Re: How to Finetune GPT-Like Large Language Models on a Custom Dataset

#28

Post by mountainriver »

slenocchio wrote: post #24

Can someone explain why I'd want to use fine-tuning instead of a vector database (or some other way of storing data/context)?

I think it probably works a lot better, but I would love to see some research validating this

Re: How to Finetune GPT-Like Large Language Models on a Custom Dataset

#29

Post by moffkalast »

artembugara wrote: post #2

Have a question to the Generative AI experts here. So, I can use smthg like GPT-4 to label data and then use that as a train set for my own LLM, right? EDIT: adding this from OpenAI Restriction TOS: "(iii) use output from the Services to develop models that compete with OpenAI;"

> I can use smthg like GPT-4 to label data and then use that as a train set for my own LLM, right?

Yes, almost all improved LLama models are tuned exactly that way (trained on examples of questions and answers from say GPT 4). If OpenAI stole copyrighted works to train their models it is morally fair game to do the same to them regardless of their TOS. It's not like they can prove it anyway.

Plus there's the other point where they also say that everything generated by their models is public domain, so which one is it eh?

Re: How to Finetune GPT-Like Large Language Models on a Custom Dataset

#30

Post by sirsinsalot »

moffkalast wrote: post #29
artembugara wrote: post #2

Have a question to the Generative AI experts here. So, I can use smthg like GPT-4 to label data and then use that as a train set for my own LLM, right? EDIT: adding this from OpenAI Restriction TOS: "(iii) use output from the Services to develop models that compete with OpenAI;"

> I can use smthg like GPT-4 to label data and then use that as a train set for my own LLM, right? Yes, almost all improved LLama models are tuned exactly that way (trained on examples of questions and answers from say GPT 4). If OpenAI stole copyrighted works to train their models it is morally fair game to do the same to them regardless of their TOS. It's not like they can prove it anyway. Plus there's the other po…

This ... but we all know business is corrupt.

The current attempts to spur on regulation by OpenAI is moat building

Post reply on HN