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

    Neuronpedia

    INDEX
    Explanations

    chat / chatbot interaction

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ן
    0.95
    ils
    0.94
     (
    0.92
    ின்
    0.78
    ych
    0.77
    är
    0.77
    än
    0.77
    ather
    0.77
    hammad
    0.77
    air
    0.75
    POSITIVE LOGITS
    ли
    1.28
    у
    1.16
    мо
    1.14
    ૫
    1.10
    ви
    1.05
     lumea
    1.04
    l
    1.04
    d
    1.02
    w
    1.02
    по
    1.02
    Act Density 0.040%

    No Known Activations