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

    Neuronpedia

    INDEX
    Explanations

    constant or constants

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    م
    1.91
    𝒂
    1.88
    を集
    1.87
    де
    1.80
    ες
    1.74
    &$\
    1.73
    $("
    1.72
    是由
    1.70
     совет
    1.66
    从而
    1.66
    POSITIVE LOGITS
    inued
    2.28
    ق
    2.06
     itulah
    1.91
    alities
    1.77
    ellations
    1.77
    ূতন
    1.75
    ra
    1.72
    س
    1.72
    lagt
    1.70
     declinations
    1.66
    Act Density 0.076%

    No Known Activations