OpenChat: Advancing open-source language models with imperfect data
1–10 of 27 posts
Re: OpenChat: Advancing open-source language models with imperfect data
#2Those numbers are quite impressive for a 7B model!
Re: OpenChat: Advancing open-source language models with imperfect data
#3“All you need is pretraining on the test set.”
Re: OpenChat: Advancing open-source language models with imperfect data
#4Wasn't there a thing about the mistake of using different tricks and techniques to beat benchmarks but in the end, the product would only be good for getting benchmark scores and nothing can surpass raw computation in general purposes?
Re: OpenChat: Advancing open-source language models with imperfect data
#5Is there a Gradio demo?
Re: OpenChat: Advancing open-source language models with imperfect data
#6Is there a Gradio demo?
Re: OpenChat: Advancing open-source language models with imperfect data
#7Surprised this is the first time I’ve heard of this, been mainly using Mistral 7B. Using their online demo, it’s pretty impressive so far.
Re: OpenChat: Advancing open-source language models with imperfect data
#8This is like back when we had image recognition. A new test set would come out and somehow everything new would be better than everything old but if you talked to anyone using, it would turn out that everything new sucked in general.
Goodhart came to take his slice.
Still I'm very excited about the open models. Lots of potential for true user tools because of what they can be.
Re: OpenChat: Advancing open-source language models with imperfect data
#9Time to change the benchmarks! Says openai.
Re: OpenChat: Advancing open-source language models with imperfect data
#10It can't be run locally,can it ?
I see that the training need 8xA100 80G and running need cuda but I doubt it need 8xA100 to run.