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

    Neuronpedia

    INDEX
    Explanations

    topped with

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    AC
    0.77
    Sync
    0.72
    AY
    0.71
    SC
    0.70
    RC
    0.68
    History
    0.66
     >,
    0.66
    RS
    0.65
    Bat
    0.65
    BC
    0.64
    POSITIVE LOGITS
     toppings
    0.93
     topping
    0.77
    topping
    0.73
     topped
    0.72
    堆
    0.72
    夫
    0.70
    颜
    0.70
     topper
    0.68
    鞍
    0.68
     essentielle
    0.67
    Act Density 0.017%

    No Known Activations