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

    Neuronpedia

    INDEX
    Explanations

    so [descriptor]

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ó
    1.07
    此
    0.93
    nél
    0.91
    imizi
    0.77
    ére
    0.76
    ल
    0.75
    and
    0.75
    ává
    0.73
     I
    0.73
    nici
    0.73
    POSITIVE LOGITS
    an
    1.15
    ان
    1.14
    ва
    1.05
     an
    1.01
    ма
    0.93
     он
    0.86
     isn
    0.85
    か
    0.82
    া
    0.82
    that
    0.79
    Act Density 0.027%

    No Known Activations