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

    Neuronpedia

    INDEX
    Explanations

    measuring things

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    し
    0.88
     موسي
    0.86
     ७
    0.86
    țile
    0.84
    لي
    0.84
     моём
    0.84
     கூடிய
    0.82
     Și
    0.82
    ților
    0.82
    ۹
    0.82
    POSITIVE LOGITS
    '
    1.23
    "
    1.03
    d
    0.99
    (
    0.99
    g
    0.93
    r
    0.93
    2
    0.90
    ก
    0.90
    ร
    0.89
    ת
    0.89
    Act Density 0.019%

    No Known Activations