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

    Neuronpedia

    INDEX
    Explanations

    or

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    分别为
    0.95
     And
    0.74
    And
    0.74
     everywhere
    0.71
     four
    0.70
     Everywhere
    0.69
    一共
    0.69
    4
    0.69
    and
    0.69
    ıları
    0.68
    POSITIVE LOGITS
     или
    3.16
     or
    3.01
     hoặc
    2.98
    或
    2.97
     oder
    2.90
     atau
    2.86
     nebo
    2.85
     أو
    2.84
     veya
    2.83
     یا
    2.80
    Act Density 2.818%

    No Known Activations