overview for sergeant113

Train Smarter, Not Harder? - MiniSymposium 7b in c/localllama@poweruser.forum

[–] sergeant113@alien.top 1 points 10 months ago (2 children)

I think the idea has merits. How do you propose for others to contribute to the training dataset?

Yi-23B-Llama: Distil version of Yi-34B-Llama in c/localllama@poweruser.forum

[–] sergeant113@alien.top 1 points 10 months ago

Can’t wait!!!

What is the best 7B model for reading comprehension and instruction following at the moment? in c/localllama@poweruser.forum

[–] sergeant113@alien.top 1 points 10 months ago

I agree. My use case involves generating json outputs, and no other mistral finetune has even come close to matching the outputs of mistral-instruct. Instruct comes very close to the performance of gpt-3.5-turbo even.

API to a service that offers access to open source LLMs? in c/localllama@poweruser.forum

[–] sergeant113@alien.top 1 points 10 months ago

OpenRouter.ai It’s an openai proxy server that also offer many opensource models.

I scaled Mistral 7B to 200 GPUs in less than 5 minutes in c/localllama@poweruser.forum

[–] sergeant113@alien.top 1 points 10 months ago (1 children)

I’m in the middle of building my app on Modal. Guess I’ll adapt it to run on your service and see. Thanks for sharing!

Finally, a diffusion based LMM! in c/localllama@poweruser.forum

[–] sergeant113@alien.top 1 points 10 months ago

How does it make a lot of sense, please explain?