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

    Neuronpedia

    INDEX
    Explanations

    simple descriptions of tasks

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    的多
    1.03
     to
    1.01
    는
    1.00
    ↵↵
    0.93
    та
    0.92
     dàng
    0.90
    것
    0.86
    จะ
    0.86
    的外
    0.86
    是
    0.86
    POSITIVE LOGITS
    ী
    0.99
    ம்
    0.94
     einfacher
    0.86
     basit
    0.86
    THAN
    0.80
    SIMPLE
    0.80
    Simple
    0.79
     Simple
    0.77
    。
    0.77
     einfachen
    0.76
    Act Density 0.057%

    No Known Activations