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

    Neuronpedia

    INDEX
    Explanations

    initialize state or variables

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    τητα
    0.76
     geeft
    0.75
     pedidos
    0.73
    停留
    0.72
     :/
    0.70
    ቀር
    0.70
    ncia
    0.70
     =>
    0.68
     مجھے
    0.67
    nja
    0.67
    POSITIVE LOGITS
     américaine
    0.72
     Dunes
    0.72
    筮
    0.69
     tine
    0.68
     Swiss
    0.68
    ক
    0.66
     Dye
    0.64
     slopes
    0.63
     ধরনের
    0.62
     Clay
    0.62
    Act Density 0.001%

    No Known Activations