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

    Neuronpedia

    INDEX
    Explanations

    color

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     overwrite
    0.68
     鳥
    0.63
    രേ
    0.62
    Ợ
    0.61
     Kay
    0.61
     زی
    0.60
     පු
    0.58
     createContext
    0.58
    munderover
    0.58
     Lago
    0.57
    POSITIVE LOGITS
    #
    3.92
     #
    3.61
    .#
    2.92
     (#
    2.90
    #,
    2.86
     \#
    2.86
     $\#
    2.82
    ＃
    2.80
     "#
    2.79
    ,#
    2.75
    Act Density 0.023%

    No Known Activations