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

    Neuronpedia

    INDEX
    Explanations

    code vulnerabilities

    np_max-act · gemini-2.0-flash

    snippets of HTML/JavaScript used in cross-site scripting or other client-side injection attacks (e.g., <script>, onerror/onclick attributes, src/import URLs, alert/document.cookie).

    oai_token-act-pair · gpt-5-miniTriggered by @vetterc0
    New Auto-Interp
    Negative Logits
    午饭
    -0.07
    נפגע
    -0.07
    mov
    -0.06
     xmax
    -0.06
    Throughout
    -0.06
    的性格
    -0.06
     Must
    -0.06
     Montgomery
    -0.06
    爱你
    -0.06
    .toLowerCase
    -0.06
    POSITIVE LOGITS
    ilda
    0.08
    FD
    0.08
     appell
    0.08
    פת
    0.07
    CPF
    0.07
    ⋆
    0.07
    䧞
    0.07
     tú
    0.07
    ilitation
    0.07
    引流
    0.07
    Act Density 0.068%

    No Known Activations