News for hackers. Stuff that matters.
Re:Mistral 7B Paper on ArXiv by ramesh31 on Wednesday October 11, 2023 @10:27AM (#37844817)
Is it better than llama 2?
Re:Mistral 7B Paper on ArXiv by tarruda on Wednesday October 11, 2023 @10:29AM (#37844847)
It is better than llama 2 7b and 13b. I tried the OpenOrca fine tune and it is very good, even when 4-bit quantized
Re:Mistral 7B Paper on ArXiv by faizshah on Wednesday October 11, 2023 @12:01PM (#37846206)
What does OpenOrca do? It’s just instruction tuning it?
Re:Mistral 7B Paper on ArXiv by tarruda on Wednesday October 11, 2023 @12:38PM (#37846710)
Yes, it is a instruction tune dataset: https://huggingface.co/datasets/Open-Orca/OpenOrca

It felt different from the official Mistral7B-Instruct. One of the highlights with the OpenOrca version is that you can steer the model with a system prompt (eg "You are a 5 year old")

Re:Mistral 7B Paper on ArXiv by TheRoque on Wednesday October 11, 2023 @10:27AM (#37844824)
Yes
Re:Mistral 7B Paper on ArXiv by sebzim4500 on Wednesday October 11, 2023 @10:47AM (#37845113)
For its size, yes. In absolute terms it is obviously less capable than llama-2-70B
Re:Mistral 7B Paper on ArXiv by espadrine on Wednesday October 11, 2023 @11:34AM (#37845814)
For now. Huggingface[0] mentioned a DPO-fine-tuned version, Zephyr 7B, which it claims is competitive with Llama2-70B[1].

[0]: https://huggingface.co/spaces/HuggingFaceH4/zephyr-chat

[1]: https://twitter.com/huggingface/status/1711780979574976661

Re:Mistral 7B Paper on ArXiv by brucethemoose2 on Wednesday October 11, 2023 @03:58PM (#37849499)
Oh, they uploaded the weights. I missed this one, cheers!
Re:Mistral 7B Paper on ArXiv by andai on Wednesday October 11, 2023 @11:58AM (#37846157)
I found llama-2-70B to be a bit worse than GPT-4. (So, pretty good!) But I did not compare with GPT-3.

How do llama-2-70B and Mistral 7B compare with GPT-3?