Heretic removes restrictions from language models

(heretic-project.org)

61 points | by Bluestein 7 hours ago

4 comments

  • Almondsetat 1 hour ago
    I have a chinese IP camera. From superficial research I know it has some CVEs to take control of it. Unfortunately, I don't have the technical knowledge to perform an attack and run some software to extend the camera's functionalities. No model from a provider accepts my RE and hacking requests, so these abliterated ones have been vital to reclaim possession over my stuff
    • inexcf 46 minutes ago
      I did that exact thing with GLM-5.3 from Z.ai with a chinese IP Camera. And i did not have to trick it in any way.
    • dgellow 36 minutes ago
      What model did you try? Chinese models have no issues with that type of stuff
    • blurbleblurble 27 minutes ago
      Existential
  • Tepix 1 hour ago
    Keep a close eye on abliterated and "heretic" open weight models. They will be outlawed first.
    • api 0 minutes ago
      This is the test. If the speech that's easiest to dislike is legal, then we all have free speech.

      IMO math is free speech, and outlawing math is censorship.

    • luxpir 34 minutes ago
      Agree. I took a look at these last few months, did a write-up: https://languageops.com/blog/ai-safety-pdoom-local-vs-fronti... and I don't know if I agree or not on outlawing completely, but I think an age restriction *at least* like for alcohol, firearms and driving would be not unwise.
      • bilsbie 11 minutes ago
        Make sure they ban books with dangerous knowledge too.
      • mitxela 26 minutes ago
        The hardware requirements are already quite restrictive
        • redoxate 19 minutes ago
          Oh no, some run on iPhones
    • roenxi 1 hour ago
      It is not feasible. They never made much of an inroad against torrents and that is a much easier target than abliterated models. As the linked website shows; the process to abliterate a model can be as simple as

      pip install -U heretic-llm && heretic Qwen/Qwen3.5-4B

      let alone people just putting the weights up in a torrent. All assuming that someone even tried to ban abliterated models.

      • Sayrus 1 hour ago
        The torrents you are talking about are outlawed. Whether enforcement is working or not is another issue.
    • petra 14 minutes ago
      Like they've outlawed drugs? Illegal weapons? Hacking?
    • thih9 1 hour ago
      I'm not sure what is your point. It reads as defeatism to me but I'm not sure.

      Could you elaborate? Do you find it good or bad? What actions can be taken?

      • cyanydeez 1 hour ago
        Hes of the mind that american fascism will hold together long enough to be competent decesion makers
    • ben_w 1 hour ago
      Good.

      If you think closed source software/binaries only is bad, wait until you see how awful the state of the art is with a clear-as-mud bucket of matrix weights.

      We know it's possible to train an LLM to secretly respond to certain trigger phrases, and last I checked these could only be detected with the assistance of whoever chose those phrases.

      The trigger condition for such backdoors is not something anyone can do a systematic brute-force check for, for the same reason we had to invent LLMs in order to do natural language processing: combinatorial explosion.

      Passing around open weight models from known sources is already asking you to trust those sources; because of how difficult this is to do correctly even without deliberately inserting such things, we still don't know if China has already put such trigger conditions into their models despite headlines such as these: https://venturebeat.com/security/deepseek-injects-50-more-se...

      Regardless of if it was deliberate or not, we don't know if we caught all of these misbehaviours. We don't know how to.

      And note, I'm not saying "and therefore you should trust the Big Name Models". If open weight models score 2/100 in this context, closed ones score 1/100.

  • N_Lens 7 hours ago
    Looks like a well engineered, automated abliteration pipeline. The claims seem a bit overstated though, since the metrics mentioned are cherrypicking refusal count and KL divergence, both of which make the outcome seem the most dramatic.
    • tacomagick 3 hours ago
      I personally never saw much of a quality drop from models put through Heretic if that amounts to anything. They have been working quite well on small local models so far.
  • phoronixrly 1 hour ago
    Can the load-bearing gaps that are worth being flagged for pinning down be abliterated out of a model?