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

    Neuronpedia

    INDEX
    Explanations

    code parsing and structure

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    적으로
    0.85
     وعلى
    0.79
    膛
    0.78
     deswegen
    0.75
    ে
    0.74
    ात्
    0.68
    ی
    0.68
    遢
    0.67
    هما
    0.67
    ally
    0.65
    POSITIVE LOGITS
    t
    0.96
    িন
    0.92
    ad
    0.86
    ir
    0.80
    c
    0.79
    éristique
    0.77
    ون
    0.75
    お子
    0.75
     déclar
    0.74
    ariyam
    0.73
    Act Density 0.273%

    No Known Activations