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

    Neuronpedia

    INDEX
    Explanations

    rest Response, window Builder, request Parameter

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ():
    0.64
     bothered
    0.63
    ()){
    0.62
     ТО
    0.59
    ળી
    0.55
     минима
    0.55
    بيوتر
    0.55
     )))
    0.55
    шки
    0.54
    ()
    0.54
    POSITIVE LOGITS
    nič
    0.64
     decarbon
    0.63
    dé
    0.59
    ৬৮
    0.59
    ;
    0.59
     unanim
    0.58
     topl
    0.58
     संबंधों
    0.58
     adulter
    0.58
    arian
    0.57
    Act Density 0.008%

    No Known Activations