Skip Navigation

InitialsDiceBearhttps://github.com/dicebear/dicebearhttps://creativecommons.org/publicdomain/zero/1.0/„Initials” (https://github.com/dicebear/dicebear) by „DiceBear”, licensed under „CC0 1.0” (https://creativecommons.org/publicdomain/zero/1.0/)F

FrankLaskey

@ FrankLaskey @lemmy.ml

Posts
3
Comments
41
Joined
6 yr. ago

  • Do you participate in this hobby without a formal IT education or a career in a similar field?

    Jump

  • I have no formal training and work in real estate so I’ve taught myself GNU/Linux CLI commands, docker and basic networking stuff with the help of the Internet, YouTube and some LLM assistance. I now self-host around a dozen services via docker on an old Dell XPS laptop for myself. This did lead me to consider a career change and I applied for and got into a free study course for the CompTIA Network+ and Security+ certificates so we’ll see if that ends up just becoming extra help for my hobby or a means to get into IT more broadly.

  • Qwen3.8-2.4T-A95B Open Weights Released!

    Jump

  • Maybe if I quantize this down to 0.1-bit it will run on my 3090 and every so often spit out a sentence that isn’t gibberish or random characters..

  • Thoughts on Hermes?

    Jump

  • I have only very recently set it up and I’m running it with a cloud model (Gemini 3.5 Flash) but I do think its marquee feature (learning skills based on what you ask it to do and then being able to use those in the future) is pretty valuable. I haven’t used OpenClaw but my understanding is it has a massive library of skills (essentially repeatable functions or tasks) while Hermes has a somewhat smaller library of shared skills but crucially you can teach it your own skills by just asking it to do something and giving it notes on how to do it.

  • Gochujang Mushroom Ragu Pasta

    Jump

  • Looks great! Will have to try it.

  • I rescued a stray kitty today

    Jump

  • Gorgeous tuxedo cat. I have one and he’s great.

  • Qwen3.6-35B-A3B released

    Jump

  • Yes I did see that as well. That does seem to be the real Achilles heel here. Will have to try it myself to see how much it exacerbates context size limitations given I would be running it on a single 24 GB VRAM GPU. I wonder if adjusting reasoning effort parameters could make a difference without affecting quality too much?

  • Qwen3.6-35B-A3B released

    Jump

  • Was just looking at the benchmarks for it on artificialanalysis.ai and it looks great for the size. Probably best available for general use if you’re looking for something under 40b parameters I’d say. Even more impressive is the agentic capabilities and the fact that it is actually decent in terms of hallucinations (not amazing given it’s a small-medium size model but decent).

  • Deleted

    Permanently Deleted

    Jump

  • Hmmm.. if an ‘organism’ has no organs, is it still an organism? 🤔

  • It is now better than ever to find alternatives to fossil fuels, do you have any plans?

    Jump

  • Absolutely. They do tend to be quite a bit more expensive per watt though. From my research you can get cheap panels and just plug them into a LiPo4 battery or even lead acid batteries as well. It’s just that batteries (even though they’ve gotten a lot cheaper) are still expensive enough to nearly double the cost of the system in many cases so the plug-in option with 400 watts of panels etc gives you the best bang for the buck for sure. But it’s not legal or possible in many US states (though that is changing fast)

  • It is now better than ever to find alternatives to fossil fuels, do you have any plans?

    Jump

  • Biking (e-bikes are great if you can afford one and greatly increase range and decrease effort in many environments)

    Installing solar (I’m currently renting but my state is debating a new law which seems likely to pass modeled on the Utah law - and others which started in Europe - which allows small solar systems to be plugged directly into a home outlet to supplement energy needs with minimal cost)

  • How Much Do LLMs Hallucinate in Document Q&A Scenarios? A 172-Billion-Token Study Across Temperatures, Context Lengths, and Hardware Platforms [TLDR: 25%]

    Jump

  • My biggest takeaway here is that choosing the context length and (to a lesser extent) the temperature carefully is important for reducing hallucinations. I expected model families to vary widely between themselves but not for context length to have such a massive impact tbh.

    It seems from this like reducing context length in applications where it isn’t essential for the model to hold very large amounts of context simultaneously would be best practice no?

  • Are you doing something to fuck over "the powerful"? If so, what?

    Jump

  • In a similar boat in regards to cutting out invasive, rapidly-enshittifying corporate tech providers but I agree that YouTube still has some good content worth it. Good news is that with the use of Invidious servers on desktop and SmartTube / NewPipe on Android TV or mobile you can pretty much have your cake and eat it too.

    P.S. if you’re looking for a less shitty / ad bloated way of watching Yt videos on Nvidia Shield Pro or similar Android TV I recently switched from NewPipe to SmartTube and it’s much better since it integrates Sponsorblock and the UI is actually designed for a TV.

  • Garlic, potatoes and crispy pork bits !

    Jump

  • Awesome thanks! Will give it a shot. I also learned the term lardon which I never knew before today..

  • Garlic, potatoes and crispy pork bits !

    Jump

  • Looks great! Would love a recipe.

  • 30B-A3B GLM-4.7-Flash Released

    Jump

  • Will be interesting to see how it stacks up to Nemotron 3 nano. I’m hoping to have somewhat more reliable models at this size that I can run entirely locally on my 3090. Really hoping that with MCP servers and tools, etc. they can be functional enough for use cases which require more security and local operation. For most things, though, I have switched to using larger models via open router.

  • Loops publishes illustration of their recommender algorithm

    Jump

  • Technically, yes the info would be available but not everyone is able to read code so this is a welcome additional bit of info and transparency. I also see it as an example of how future open source social media platforms can promote and demonstrate the way their algorithms work as opposed to the black boxes of proprietary web apps like Instagram etc (which, while not algorithmically transparent or open, have been pretty well established to have algorithms that prioritize maximizing time on platform and engagement over all else with some serious negative repercussions for that).

  • Deleted

    Permanently Deleted

    Jump

  • This article in the Guardian is definitely worth a read if you’re not intimately familiar with just how it got this way.. It’s 8 years old so it won’t cover recent history but does give you an idea of how it started.

    And yes Robert Maxwell (father of Ghislaine) is mostly to blame.

  • Deciding To Win – Toward A Common Sense Renewal of the Democratic Party

    Jump

  • Much of this reads like what the Democrats’ strategy has already been for the last 10 years or so with some additional calls towards a “moderate” “centrism” that has not proven to be as popular as left populist policies with most Americans. And a lot of it seems confused or just wrong. For example, Medicare for All is not an unpopular policy. Last I checked it it was polling at around a 60% approval.

    Even if your politics are more centrist, I don’t see how this represents a substantive shift in any way. It’s minor tinkering around the edges or slightly altering messaging. That’s what the moment calls for? That’s the key to success and the way to fight back against an increasingly overt and ascendant strain of fascism in the country? If that’s what the Democratic leadership thinks then one must honestly wonder if the party is institutionally incapable of the change that would need to happen to actually win elections and improve their godawful approval ratings. And based on their donor base, this shouldn’t actually be surprising. The leadership is still utterly captured by and beholden to wealthy interests that would rather jump off a bridge than acknowledge that what Americans want is a party that will fight for meaningful material benefits in their standard of living as this implies meaningfully raising taxes and imposing costs or regulations on their businesses to ensure benefits for regular working class people who are struggling mightily.

    .

  • Anyone using z.ai?

    Jump

  • Yes. GLM 4.5 is excellent. I mostly use that along with Qwen3 and DeepSeek 3.1 Terminus (Thinking) these days.

  • LocalLLaMA @sh.itjust.works
    FrankLaskey @lemmy.ml

    Best LLMs for PC/Tech Troubleshooting?

  • Technology @lemmy.world
    FrankLaskey @lemmy.ml
    Gaussian Splatting

    The Tech to Build the Holodeck

    www.theverge.com /2025/1/19/24345491/gaussian-splats-3d-scanning-scaniverse-niantic
  • Technology @lemmy.world
    FrankLaskey @lemmy.ml

    How do Graphics Cards Work? Exploring GPU Architecture - YouTube by Branch Education (28:29 minutes) from Oct 19, 2024