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

    Neuronpedia

    INDEX
    Explanations

    initial letter followed by word start

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    as
    0.77
    RE
    0.72
    use
    0.70
    ة
    0.70
    LEN
    0.68
    ék
    0.67
    Ion
    0.66
    עים
    0.64
    rows
    0.64
    ים
    0.64
    POSITIVE LOGITS
    り
    0.91
    ки
    0.88
    ಯ
    0.85
    oretically
    0.77
    ৯
    0.74
    ுள்ளனர்
    0.73
    いた
    0.73
     insuff
    0.71
    stood
    0.70
    서
    0.70
    Act Density 0.400%

    No Known Activations