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

    Neuronpedia

    INDEX
    Explanations

    the Original

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    he
    0.40
    al
    0.35
    te
    0.35
    ل
    0.34
    en
    0.33
    as
    0.33
    il
    0.32
    et
    0.32
    an
    0.30
    ad
    0.30
    POSITIVE LOGITS
     to
    0.45
     of
    0.36
     be
    0.34
    。
    0.29
     éxitos
    0.29
    rystals
    0.28
     hijos
    0.28
     ojos
    0.27
     was
    0.27
    する
    0.27
    Act Density 5.281%

    No Known Activations