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

    Neuronpedia

    INDEX
    Explanations

    Internal consistency, you need

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ozo
    1.61
    कुंठ
    1.59
    গাঁও
    1.57
    lL
    1.55
     compressibility
    1.54
    ی
    1.52
    س
    1.45
    ือบ
    1.43
    幾
    1.42
    adip
    1.42
    POSITIVE LOGITS
     hate
    1.60
     banda
    1.41
    ungu
    1.29
     ve
    1.29
     wort
    1.26
    forall
    1.25
    ра
    1.24
     ditt
    1.22
     ob
    1.22
     hated
    1.18
    Act Density 0.002%

    No Known Activations