A software developer and Linux nerd, living in Germany. I’m usually a chill dude but my online persona doesn’t always reflect my true personality. Take what I say with a grain of salt, I usually try to be nice and give good advice, though.

I’m into Free Software, selfhosting, microcontrollers and electronics, freedom, privacy and the usual stuff. And a few select other random things as well.

MRZ LCK 00

  • 2 posts
  • 5 comments
Joined 5 years ago
Cake day: August 21st, 2021
  • Linux@lemmy.ml•Is there a way to know the specifications of your cables?byhendrik@palaver.p3x.deEnglish
    14 hours

    Usually if a cable doesn’t come with specifications, you’d have to measure it yourself. Resistance / voltage drop tells us what kind of power can be transferred. It’s the same as with regular power cables. And I guess which contacts are connected can be measured with one of those (cheap) rigs. There’s small devices available for that job. And of course a computer can try to transfer data and see how that goes.

LocalLLaMA@sh.itjust.worksbyhendrik@palaver.p3x.de
16 days

How many proper open-source LLMs are there?

English

Sometimes I get into arguments about AI and people comment weird stuff here on Lemmy. Like we have ethical AI models (meaning they’re not made up of unethically sourced training material). Or we got plenty “libre” or “free and open-source” AI models.

Am I missing something? Or is this just a bunch of people trying to bullshit me? Is anyone here aware of such large language models and would be so kind as to educate me? I mean we do have a ton of open-weights models. But proper open-source or ethical??? I’ve been following the news for quite a while and I’m aware of a very select few I’d call LLMs with training datasets available. I’m aware of:

  • Pythia by EleutherAI (up to 12B, from 2023) and Dolly 2 by Databricks (12B, also from 2023)
  • BLOOM by BigScience (176B, from 2022)
  • GPT-2 by OpenAI (1.5B, from 2019)
  • The recreation of LLaMA by the RedPajama-Project
  • Apertus by the ETH Zürich

Edit: In the comments we got:

  • LLM360 with K2-65B, Crystal-7b, Amber-7B
  • the Soofi-Project
  • OLMoE

That list needs a lot of footnotes, though. Some of them come with restrictions, terms and conditions… And none of them can do modern tasks like AI-assisted coding. The only one getting close is the Swiss model. Sadly the earlier version I tried didn’t perform well. Seems they released an updated v1.5 version this summer, I wonder if that’s better and anybody uses it for real-world applications.

So… Back to my initial question: Are there models out there with datasets available? I mean except the exactly one somewhat usable one I already found?

0
    • Fediverse@lemmy.world•Quadruple moderation is killing lemmy. It’s a Gift to Redditbyhendrik@palaver.p3x.deEnglish
      21 days

      Seems you post a lot of duplicates. No wonder your posts get deleted if you keep spamming exactly the same post multiple times??? And then you violate community rules… After that you seem to be doing ban-evasion… I mean… Seems to me like Lemmy is working exactly as intended?! This isn’t a place to dump random spam posts? You gotta learn what this place is about.

      You can’t just throw some dogfood at people and demand they like it. You’re not the center of the universe. Maybe start your own community if that’s your thing. You’d be able to make up your own rules there. There’s a button in the top menu bar.

    About Lemmit@lemmit.onlinebyhendrik@palaver.p3x.de
    2 years

    Request: Remove amateur pornography subs

    English

    We’ve had a bit of a conversation, over in the big NoStupidQuestions community:
    https://lemmy.dbzer0.com/post/37540045

    While I have my own opinions on lemmit.online, I think it’s relatively uncontroversial, that copying content from amateur and indie creators is unethical.
    I’d like to request differentiating between the regular Reddit content, and amateur pornography plus OF creators and their original content. And deactivating the bridging for subreddits that contain a decent amount of the latter.

    My rationale is more or less that it’s not very Robin Hood to take things from people who aren’t well off in the first place. And that more or less regular people have the right to decide what happens with pictures of their naked bodies, and we can’t just spread them across the internet without their consent or ability to closely control their intimate stuff.

    0