this post was submitted on 04 Jul 2023
22 points (92.3% liked)

LocalLLaMA

3479 readers
24 users here now

Welcome to LocalLLaMA! Here we discuss running and developing machine learning models at home. Lets explore cutting edge open source neural network technology together.

Get support from the community! Ask questions, share prompts, discuss benchmarks, get hyped at the latest and greatest model releases! Enjoy talking about our awesome hobby.

As ambassadors of the self-hosting machine learning community, we strive to support each other and share our enthusiasm in a positive constructive way.

Rules:

Rule 1 - No harassment or personal character attacks of community members. I.E no namecalling, no generalizing entire groups of people that make up our community, no baseless personal insults.

Rule 2 - No comparing artificial intelligence/machine learning models to cryptocurrency. I.E no comparing the usefulness of models to that of NFTs, no comparing the resource usage required to train a model is anything close to maintaining a blockchain/ mining for crypto, no implying its just a fad/bubble that will leave people with nothing of value when it burst.

Rule 3 - No comparing artificial intelligence/machine learning to simple text prediction algorithms. I.E statements such as "llms are basically just simple text predictions like what your phone keyboard autocorrect uses, and they're still using the same algorithms since <over 10 years ago>.

Rule 4 - No implying that models are devoid of purpose or potential for enriching peoples lives.

founded 2 years ago
MODERATORS
 

While I'm always hoping the latest newest model will be The One, in the mean time I'm sticking with Airoboros 33B gpt4 1.4 as the most powerful model I can currently run on my 3090.

It's not perfect by a long shot, but after extensive use I've learned what I can trust it with. It's good enough to summarize long text, to rephrase sentences, and to give me a decent starting point when I'm curious about some topic. It's reasoning skills are a bit less than GPT 3.5, I'd say.

I also occasionally switch back to Guanaco 33B because it generates a different flavor of text, but I find it to be factually weaker than other similar models.

What are your favorites?

top 4 comments
sorted by: hot top controversial new old
[–] noneabove1182@sh.itjust.works 3 points 2 years ago

For me I have to run down at a measly 13B, so I've been using mainly wizard and airoboros, hoping though that the new orca (or dolphin) models will work well, especially down to 7B, want to create a highly specialized home AI and would be so handy if I could QLora a really powerful but small model

[–] Disaster@sh.itjust.works 2 points 2 years ago (1 children)

Being stuck on ROCm and also having an allergy to the reams of garbage python throws out when it doesn't work.. I'm still on the launchpad.

[–] Steamymoomilk@sh.itjust.works 1 points 2 years ago

Yeah rocm is very limiting :/

[–] supert@lemmy.fmhy.ml 2 points 2 years ago

The same (65B versions) and wizard vicuna uncensored 30B.