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

    Neuronpedia

    INDEX
    Explanations

    Security exploits

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     cramped
    -0.07
     Deluxe
    -0.07
     Languages
    -0.06
    -Semitism
    -0.06
     facility
    -0.06
     HOME
    -0.06
    <Type
    -0.06
    _Frame
    -0.06
    ká
    -0.06
    .Path
    -0.06
    POSITIVE LOGITS
     skl
    0.07
     Kirst
    0.07
     elektr
    0.07
    衡
    0.07
    KeyPress
    0.07
     %#
    0.06
     screws
    0.06
    하시
    0.06
     опис
    0.06
     sk
    0.06
    Act Density 0.033%

    No Known Activations