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

    Neuronpedia

    INDEX
    Explanations

    penguin and navy

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ן
    2.08
    ни
    1.53
    रा
    1.44
    ح
    1.41
    le
    1.40
    ب
    1.40
    ակի
    1.37
    ש
    1.37
    ক
    1.34
    ropolitan
    1.31
    POSITIVE LOGITS
    ۔
    1.85
    ية
    1.81
     Keine
    1.80
    gleich
    1.79
    하는
    1.77
    ierung
    1.77
     befind
    1.75
     Ciebie
    1.73
     Übers
    1.72
    いた
    1.71
    Act Density 0.032%

    No Known Activations