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

    Neuronpedia

    INDEX
    Explanations

    technical command lines

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     repos
    0.45
    也不能
    0.42
     Pel
    0.41
     dior
    0.40
     religieux
    0.40
    Pel
    0.40
     pretended
    0.40
     almo
    0.38
     نمود
    0.38
     moderate
    0.38
    POSITIVE LOGITS
    Setter
    0.39
    を採用
    0.38
     девушка
    0.38
    ㅋㅋ
    0.37
    breadcrumbs
    0.37
     impactful
    0.37
    外国人
    0.37
    Measurement
    0.36
    愁
    0.36
     ചേർ
    0.35
    Act Density 0.002%

    No Known Activations