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

    Neuronpedia

    INDEX
    Explanations

    code formatting, file operations, listening, status updates

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    s
    1.46
    ס
    1.27
     to
    1.23
     from
    1.12
     with
    1.08
    خ
    1.08
    ش
    1.07
    س
    1.05
     at
    1.04
    ス
    1.03
    POSITIVE LOGITS
    ag
    0.96
    età
    0.86
    ↵
    0.82
    ong
    0.79
    I
    0.79
    Pedido
    0.78
    ਈ
    0.75
     Fundación
    0.75
    ä
    0.75
    ură
    0.71
    Act Density 0.866%

    No Known Activations