My impression is the general consensus is we don’t want huge corporations stealing data to train their AI models only to turn around and cram it down our throats anywhere they can with increasingly negative experiences.
That being said, while I would generally agree with that, I still find it interesting and especially if I can host it myself.
If forced to characterize the attitude of lemmy towards LLM/“AI,” I’d say people here are broadly interested in the tech but critical of the way it’s often used.
I dunno, with image models specifically it seems like they’re the devil because of the datasets they’re trained on, killing artists, and… that’s that. And LLMs to a lesser extent. There’s truth to all that, but there’s also a lot more.
I think most people don’t realize how much of an inflection point local running vs. corporate hosting could be, which is especially ironic on Lemmy.
Yeah, there’s a massive negative circlejerk going on, but mostly with parroted arguments. Being able to locally run a model with this kind of context is huge. Can’t wait for the finetunes that will result from this (*cough* NeverSleep’s *-maid models come to mind).
Oof - not on my 12gb 3060 it doesn’t :/ Even at 48k context and the Q4_K quantization, it’s ollama its doing a lot of offloading to the cpu.
What kind of hardware are you running it on?
But it should be fine on a 3060, with zero offloading.
Dump ollama for long context. Grab a 5-6bpw exl2 quantization and load it with Q4 or Q6 cache depending on how much context you want. I personally use EXUI, but text-gen-webui and tabbyapi (with some other frontend) will also load them.
I haven’t given it a very thorough testing, and I’m by no means an expert, but from the few prompts I’ve ran so far, I’d have to hand it to Nemo concerning quality.
Using openrouter.ai, I’ve also given llama3.1 405B a shot, and that seems to be at least on par with (if not better than) Claude 3.5 Sonnet, whilst being a bit cheaper as well.
128k token context is pretty sweet. Mistral nemo also just launched with a similar context. Good times.
How does the Nemo 12B compare to the Llama 3.1 8B?
At long context (close to the full 128K), Nemo is way better than llama 8B in my testing.
Turns out they are both very sensitive to quantization though.
TBH I didn’t know people here were running LLMs. Seems like most of Lemmy is very broadly anti AI?
My impression is the general consensus is we don’t want huge corporations stealing data to train their AI models only to turn around and cram it down our throats anywhere they can with increasingly negative experiences. That being said, while I would generally agree with that, I still find it interesting and especially if I can host it myself.
If forced to characterize the attitude of lemmy towards LLM/“AI,” I’d say people here are broadly interested in the tech but critical of the way it’s often used.
If by interested you mean willing to bullshit… Talking about AI here is like talking about evolution at bible camp in the deep south.
I dunno, with image models specifically it seems like they’re the devil because of the datasets they’re trained on, killing artists, and… that’s that. And LLMs to a lesser extent. There’s truth to all that, but there’s also a lot more.
I think most people don’t realize how much of an inflection point local running vs. corporate hosting could be, which is especially ironic on Lemmy.
The loud minority is really loud.
Yeah, there’s a massive negative circlejerk going on, but mostly with parroted arguments. Being able to locally run a model with this kind of context is huge. Can’t wait for the finetunes that will result from this (*cough* NeverSleep’s *-maid models come to mind).
I am looking into doing it on the 12B for myself, not so much for RP but novel style prose.
I am thinking literature + a fanfic dump as a dataset?
Ah, that’s a wonderful use case. One of my favourite models has a storytelling lora applied to it, maybe that would be useful to you too?
At any rate, if you’d end up publishing your model, I’d love to hear about it.
[Oh, my friend, you have to switch to this: https://huggingface.co/BeaverAI/mistral-doryV2-12b
It’s so much smarter than llama 13B. And it goes all the way out to 128K!
Oof - not on my 12gb 3060 it doesn’t :/ Even at 48k context and the Q4_K quantization, it’s ollama its doing a lot of offloading to the cpu. What kind of hardware are you running it on?
A 3090.
But it should be fine on a 3060, with zero offloading.
Dump ollama for long context. Grab a 5-6bpw exl2 quantization and load it with Q4 or Q6 cache depending on how much context you want. I personally use EXUI, but text-gen-webui and tabbyapi (with some other frontend) will also load them.
I haven’t given it a very thorough testing, and I’m by no means an expert, but from the few prompts I’ve ran so far, I’d have to hand it to Nemo concerning quality.
Using openrouter.ai, I’ve also given llama3.1 405B a shot, and that seems to be at least on par with (if not better than) Claude 3.5 Sonnet, whilst being a bit cheaper as well.
Llama 70B is probably where its at, if you go the API route. It’s distilled from 405B, and its benchmarks are pretty close.