Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    INDEX
    Explanations

    Asking questions

    np_max-act · gemini-2.0-flash

    phrasing that introduces a user instruction or request (e.g., "my first question is", "I want you to...", question/colon markers).

    oai_token-act-pair · gpt-5-miniTriggered by @vetterc0
    New Auto-Interp
    Negative Logits
     Unix
    -0.07
     bất
    -0.07
    бор
    -0.07
    ереж
    -0.06
     converge
    -0.06
     sucht
    -0.06
     Bayern
    -0.06
    	version
    -0.06
     正
    -0.06
     الشي
    -0.06
    POSITIVE LOGITS
    ="<?=
    0.07
     😀
    0.07
    fer
    0.07
     cancelButtonTitle
    0.06
     doctr
    0.06
     Vậy
    0.06
     <=>
    0.06
    ató
    0.06
     Tes
    0.06
    /{$
    0.06
    Act Density 0.045%

    No Known Activations