INDEX
    Explanations
    New Auto-Interp
    Negative Logits
    <eos>
    -0.47
     ...
    -0.45
     mer
    -0.44
     ra
    -0.43
    ...
    -0.42
    -0.41
     Mer
    -0.40
     a
    -0.40
    mer
    -0.39
     mo
    -0.39
    POSITIVE LOGITS
     nationaux
    0.75
    featureID
    0.73
     sélections
    0.70
     arşivlendi
    0.69
     attirer
    0.69
    -------------</
    0.68
     célè
    0.68
    出版年
    0.68
     Taktlose
    0.67
     propOrder
    0.67
    Act Density 0.970%

    No Known Activations