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

    Neuronpedia

    INDEX
    Explanations

    create

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     break
    0.68
     flax
    0.66
    royd
    0.64
    ంబే
    0.63
    婿
    0.62
     bells
    0.61
     rooftop
    0.60
    ല്ലോ
    0.60
     rejection
    0.60
     failure
    0.59
    POSITIVE LOGITS
     създа
    0.89
     creato
    0.87
     crean
    0.85
     crear
    0.85
     cré
    0.81
     creare
    0.79
     생성
    0.78
    创建一个
    0.78
     crea
    0.77
     созда
    0.77
    Act Density 0.246%

    No Known Activations