INDEX
    Explanations
    New Auto-Interp
    Negative Logits
     forecasting
    0.52
     हारा
    0.46
     Forecasting
    0.43
     unilateral
    0.42
     hostile
    0.41
     multilateral
    0.41
     predictions
    0.40
     geopolitical
    0.40
    0.40
     ousted
    0.39
    POSITIVE LOGITS
     더욱
    0.50
     그런
    0.42
     supaya
    0.42
    變得
    0.41
    还是很
    0.41
     Still
    0.41
     개선
    0.38
    Still
    0.38
    そんな
    0.38
     Verbesser
    0.38
    Act Density 0.002%

    No Known Activations