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

    Neuronpedia

    INDEX
    Explanations

    reads the

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     berikutnya
    0.87
    t
    0.86
    ته
    0.85
    tained
    0.84
    say
    0.77
    ted
    0.75
    น
    0.74
    tional
    0.74
    td
    0.74
     antaranya
    0.73
    POSITIVE LOGITS
    urface
    0.95
    Với
    0.88
    <unused622>
    0.86
    在
    0.84
    인더
    0.83
    <unused2162>
    0.83
    ٹ
    0.83
    Я
    0.82
    čiai
    0.81
     ಮತ್ತು
    0.80
    Act Density 0.791%

    No Known Activations