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

    Neuronpedia

    INDEX
    Explanations

    Crucially, bribery, primary, thrive, brain

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ों
    2.59
    러
    2.56
    w
    2.45
    ת
    2.45
    ais
    2.42
    v
    2.42
    اته
    2.33
    ات
    2.31
    ური
    2.27
    im
    2.23
    POSITIVE LOGITS
    ции
    2.11
    HAEL
    1.88
    Պ
    1.84
     При
    1.76
    BERS
    1.71
    綦
    1.69
    רים
    1.63
    хождения
    1.62
     altura
    1.60
    IT
    1.59
    Act Density 2.340%

    No Known Activations