• 3 Posts
  • 113 Comments
Joined 1 year ago
cake
Cake day: August 8th, 2023

help-circle
  • Hmm. I just assumed 14B was distilled from 72B, because that’s what I thought llama was doing, and that would just make sense. On further research it’s not clear if llama did the traditional teacher method or just trained the smaller models on synthetic data generated from a large model. I suppose training smaller models on a larger amount of data generated by larger models is similar though. It does seem like Qwen was also trained on synthetic data, because it sometimes thinks it’s Claude, lol.

    Thanks for the tip on Medius. Just tried it out, and it does seem better than Qwen 14B.





  • I assume the “kill it” comment was a little tongue-in-cheek. On small SBCs, like a Pi, or old hardware, it could be a problem. I’ve seen people with flatpaks taking up 30GB of space, which is significant. I’m not sure how much RAM it wastes. I assume running 6 different applications that have loaded 6 different versions of Qt libraries would also use significantly more RAM than just loading the system’s shared Qt libraries once.











  • AI are people, my friend. /s

    But, really, I think people should be able to run algorithms on whatever data they want. It’s whether the output is sufficiently different or “transformative” that matters (and other laws like using people’s likeness). Otherwise, I think the laws will get complex and nonsensical once you start adding special cases for “AI.” And I’d bet if new laws are written, they’d be written by lobbiests to further erode the threat of competition (from free software, for instance).


  • There’s plenty of open source projects that distribute executables (i.e. all that use compiled languages). The projects just provide checksums, ensure their builds are reproducible, or provide some other method to verify.

    In practice, you’re going to wind up in dependency hell before pypi stops hosting the package. E.g. you need to use package A and package B, but package A depends on v1 of package C, and package B depends on v2 of package C.

    And you don’t need to use pypi or pip at all. You could just download the code and directly from tbe repo, import it into your project (possibly needing to build if it has binary components). However, if it was on pypi before, then the source repo likely had all the code pip needs to install it (i.e. contains setup.py and any related files).







  • The problem is that HP writes drivers and software for those things for Windows, but not for Linux, so Linux depends on random people to write software for those things for free (which often involves complex reverse-engineering). With Linux you need to make sure you use widely-used hardware that someone has already written support for (this is mostly applicable to laptops and peripherals, which often use custom non-standard hardware). There may be a way to fix your problems, but you’ll have to search forums or issue trackers for the solutions, and they’re probably pretty involved to get working correctly. The router crashing thing is probably just a coincidence though, or the laptop is using a feature that’s broken on your router.