INDEX
    Explanations

    mentions of Germany or Germans

    New Auto-Interp
    Negative Logits
    ciating
    -0.75
    ctica
    -0.72
    ````
    -0.69
    heed
    -0.68
    clusively
    -0.67
    pole
    -0.67
    olulu
    -0.67
    Pokemon
    -0.67
    WHERE
    -0.64
    aban
    -0.63
    POSITIVE LOGITS
     Chancellor
    1.07
    geist
    1.05
     Munich
    1.02
     Bundesliga
    1.02
     Mü
    0.97
     Bundes
    0.91
    stadt
    0.90
    enstein
    0.89
     Reich
    0.88
    wings
    0.88
    Act Density 0.986%

    No Known Activations