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

    Neuronpedia

    INDEX
    Explanations

    requires complexity

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    emptyDict
    0.36
    とおり
    0.35
    改良
    0.35
    能够在
    0.35
    สี
    0.34
     ጥላ
    0.34
    💖
    0.34
    普遍
    0.34
    inales
    0.33
    不必
    0.33
    POSITIVE LOGITS
     requires
    2.13
     Requires
    1.97
    requires
    1.96
     require
    1.95
    Requires
    1.91
     requiere
    1.87
     требует
    1.83
     wymaga
    1.73
    require
    1.69
     requiring
    1.68
    Act Density 0.109%

    No Known Activations