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

    Neuronpedia

    INDEX
    Explanations

    multiplication and boxed answers

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ద్ధ
    0.59
    ChooserView
    0.54
    quasi
    0.53
    <unused344>
    0.53
    埜
    0.51
    eya
    0.51
    ˜</
    0.50
    <unused1757>
    0.50
    <unused294>
    0.50
    <unused238>
    0.50
    POSITIVE LOGITS
     color
    0.67
     require
    0.63
     Цвет
    0.63
     цвет
    0.58
    require
    0.58
     ×
    0.55
    boxed
    0.55
     Color
    0.55
    color
    0.55
     requires
    0.54
    Act Density 0.013%

    No Known Activations