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

    Neuronpedia

    INDEX
    Explanations

    digits 9

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Potts
    0.79
    જા
    0.78
     Lockhart
    0.77
    0.76
     kira
    0.74
     ESO
    0.74
     Жа
    0.74
     Aziz
    0.74
    ril
    0.74
     Linton
    0.73
    POSITIVE LOGITS
    9
    2.37
    九
    1.72
     九
    1.67
     Nine
    1.62
    ９
    1.61
    ۹
    1.59
     ninth
    1.58
     Ninth
    1.54
    ९
    1.52
     nine
    1.52
    Act Density 0.608%

    No Known Activations