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

    Neuronpedia

    INDEX
    Explanations

    label

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    crate
    -0.06
    顔
    -0.06
    owanie
    -0.06
    bject
    -0.06
    beiten
    -0.06
     Affordable
    -0.06
     Qué
    -0.06
     ноября
    -0.06
     brisk
    -0.06
    謝
    -0.06
    POSITIVE LOGITS
     grounded
    0.07
     DataType
    0.06
    icycle
    0.06
     thơm
    0.06
     tavern
    0.06
     decl
    0.06
     Danh
    0.06
    Displayed
    0.06
    imbus
    0.06
     atención
    0.06
    Act Density 0.006%

    No Known Activations