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

    Neuronpedia

    INDEX
    Explanations

    function calls

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    s
    1.01
    ات
    0.74
     the
    0.70
    št
    0.70
    ের
    0.67
     וא
    0.65
    ség
    0.65
    も含
    0.65
    ambuk
    0.64
     ル
    0.63
    POSITIVE LOGITS
     extraordinaire
    0.65
     étudier
    0.59
    珊
    0.57
    bedaan
    0.57
     inutil
    0.56
     adalah
    0.55
     ቀላል
    0.55
     envió
    0.55
     craindre
    0.55
     occupe
    0.55
    Act Density 0.610%

    No Known Activations