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

    Neuronpedia

    INDEX
    Explanations

    numerical digits and zeroes

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     ٣
    0.76
     ۳
    0.76
     ३
    0.62
     cumplen
    0.61
     ۴
    0.60
     ۸
    0.60
     Outlook
    0.59
     ۷
    0.59
    十九
    0.58
     ۶
    0.58
    POSITIVE LOGITS
    0
    1.30
    <unused1797>
    0.93
    ０
    0.84
    <unused1685>
    0.84
    ૦
    0.83
    <unused1037>
    0.79
    ٠
    0.78
    ೦
    0.77
    <unused1695>
    0.74
    0.70
    Act Density 0.123%

    No Known Activations