So I was just reading this thread about deepseek refusing to answer questions about Tianenmen square.

It seems obvious from screenshots of people trying to jailbreak the webapp that there’s some middleware that just drops the connection when the incident is mentioned. However I’ve already asked the self hosted model multiple controversial China questions and it’s answered them all.

The poster of the thread was also running the model locally, the 14b model to be specific, so what’s happening? I decide to check for myself and lo and behold, I get the same “I am sorry, I cannot answer that question. I am an AI assistant designed to provide helpful and harmless responses.”

Is it just that specific model being censored? Is it because it’s the qwen model it’s distilled from that’s censored? But isn’t the 7b model also distilled from qwen?

So I check the 7b model again, and this time round that’s also censored. I panic for a few seconds. Have the Chinese somehow broken into my local model to cover it up after I downloaded it.

I check the screenshot I have of it answering the first time I asked and ask the exact same question again, and not only does it work, it acknowledges the previous question.

So wtf is going on? It seems that “Tianenmen square” will clumsily shut down any kind of response, but Tiananmen square is completely fine to discuss.

So the local model actually is censored, but the filter is so shit, you might not even notice it.

It’ll be interesting to see what happens with the next release. Will the censorship be less thorough, stay the same, or will china again piss away a massive amount of soft power and goodwill over something that everybody knows about anyway?

  • @Albbi@lemmy.ca
    link
    fedilink
    English
    -13
    edit-2
    27 days ago

    downvotes are me

    Is using multiple accounts for voting against any of Lemmy’s rules?

    • @selfA
      link
      English
      1627 days ago

      if I wanted to cheat the downvote count I’d just modify our instance’s database. our view of votes is different and neither our posters nor our instance really give a fuck which posts random federated weirdos like or don’t like

      feel free to report me to me though

      • @swlabr
        link
        English
        1427 days ago

        reported because I want my internet points and I want them NOW (jk)

        • @froztbyte
          link
          English
          1127 days ago

          no internet points today, only internet cookies (good kind)

      • @froztbyte
        link
        English
        927 days ago

        neither our posters nor our instance really give a fuck which posts random federated weirdos like or don’t like

        although every now and then some of those come by and get a fedifuckton of updoots, but when you look at it it’s all very 🤨 and

        the gif of malcolm reynolds with a raised finger, struggling to respond to something absurd

    • flere-imsaho
      link
      English
      1326 days ago

      i always thought that some canadians do get sarcasm.

      • @Albbi@lemmy.ca
        link
        fedilink
        English
        -226 days ago

        I was joking about using a plural (downvotes) with a singular (are me), and then I was just curious if Lemmy has any sort of rules about this. I have no idea.

    • David GerardMA
      link
      English
      1227 days ago

      you should definitely report @self lots