‹ BackHN Continuity

Thread

Kev: Tiny Jev-like family of decision models built on top of Qwen3.5

462 points · 211 comments · tosh

  1. hbarka · · focus · HN ↗
    If Jev is fundamentally trained using RLCD while you’re building on a Qwen model that was trained using RLHF, how can the resulting model be considered Jev-like?
    1. mohsen1 · · focus · HN ↗
      I can't find it but saw that if you give Jev English alphabet as choices and ask it in a loop what model it is, it would say Qwen

      also tried myself: <a href="https:&#x2F;&#x2F;console.typesafe.ai&#x2F;playground?share=shr_1690a3160f19c2e4851b6a2883bcf17a790" rel="nofollow">https:&#x2F;&#x2F;console.typesafe.ai&#x2F;playground?share=shr_1690a3160f1...

      1. bityard · · focus · HN ↗
        That is not how models work.

        Unless specifically told in a system prompt, the pile of weights has absolutely no knowledge of itself. You could hypothetically train it to answer such questions, but nobody bothers to do this, and ALL &quot;knowledge&quot; embedded in the weights is probabalistic anyway.

        (I feel like this should be common knowledge in LLM discussions on HN by now.)

        1. akx · · focus · HN ↗
          This is, in fact, not true.

              $ .&#x2F;build&#x2F;bin&#x2F;llama-cli -m Qwen3.6-35B-A3B-UD-Q4_K_XL.gguf --system-prompt &quot;(this prompt explicitly left empty)&quot;
    build      : b10889-6e41074fc
    model      : Qwen3.6-35B-A3B-UD-Q4_K_XL.gguf
    ftype      : Q4_K - Medium
    modalities : text
    using custom system prompt

    &gt; What model are you?

    [Start thinking]

    Here&#x27;s a thinking process:

    1.  **Analyze User Input:** The user asks &quot;What model are you?&quot;
    2.  **Identify Core Question:** The user wants to know my identity&#x2F;model name.
    3.  **Check System Prompt&#x2F;Instructions:** The system prompt says &quot;(this prompt explicitly left empty)&quot; - this is unusual but doesn&#x27;t change my core identity. I need to respond based on my actual identity as Qwen &#x2F; 通义千问.
    4.  **Formulate Response:** I should clearly state that I am Qwen (通义千问), developed by Alibaba Group&#x27;s Tongyi Lab, and keep it concise as per guidelines.
    5.  **Draft Response:** I am Qwen (通义千问), a large language model independently developed by Alibaba Group&#x27;s Tongyi Lab. How can I assist you today?
    6.  **Refine according to guidelines:** The draft matches the required identity and tone. It&#x27;s direct and helpful. Ready.
    [End thinking]

    I am Qwen (通义千问), a large language model independently developed by Alibaba Group&#x27;s Tongyi Lab. How can I assist you today?
          1. bityard · · focus · HN ↗
            Okay, I stand corrected that _some_ models may include self-identifying metadata in their training. Although I believe I accounted for that in my comment.

            Historically, many do not and there are lots of counter-examples proving this. They merely hallucinate an answer just like anything else. The SAME model may even give different answers to the same prompt when asked multiple times... sometimes they claim to be ChatGPT, sometimes Gemma, etc. The fact that the answer is delivered confidently fools people who don&#x27;t understand this, and these people then run straight to social media with &quot;proof&quot; of their conspiracy theory that one AI lab &quot;stole&quot; another AI lab&#x27;s model.

            My point stands that unless specifically trained or told, big bags of weights do not possess any inherent introspection. LLMs have many fascinating emergent properties, but this is not one of them.

            1. mohsen1 · · focus · HN ↗
              &gt; The SAME model may even give different answers to the same prompt when asked multiple times

              temperature?

Open on Hacker News to reply ↗

Unofficial Hacker News client; not affiliated with Y Combinator.