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

    Neuronpedia

    INDEX
    Explanations

    Abbreviations/Non-English

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    m
    -0.82
    t
    -0.75
    k
    -0.73
    d
    -0.69
    に
    -0.67
    mente
    -0.64
    p
    -0.63
    c
    -0.63
    了
    -0.60
    n
    -0.59
    POSITIVE LOGITS
     automatiques
    0.72
     fédé
    0.71
     démocr
    0.69
     africains
    0.68
     umana
    0.67
     parlent
    0.67
     dieux
    0.65
     blessés
    0.64
     américains
    0.63
     chré
    0.62
    Act Density 0.088%

    No Known Activations