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

    Neuronpedia

    INDEX
    Explanations

    python `if __name__ == "__main__":`

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     مجال
    1.04
    )$$
    1.02
     uomo
    0.97
    )#
    0.95
    Тем
    0.95
     혹은
    0.94
    >∕
    0.94
    räume
    0.94
    <unused1769>
    0.93
    )$.
    0.92
    POSITIVE LOGITS
    ל
    0.92
     guided
    0.92
     each
    0.89
    hip
    0.88
     several
    0.88
    lar
    0.87
     fact
    0.85
     behalf
    0.84
     competing
    0.84
     the
    0.81
    Act Density 0.007%

    No Known Activations