INDEX
    Explanations
    New Auto-Interp
    Negative Logits
     cup
    -0.83
     све
    -0.79
    buttonText
    -0.76
    üz
    -0.76
     chute
    -0.74
     napkins
    -0.74
    สือ
    -0.74
    cups
    -0.74
     Cup
    -0.73
    -0.70
    POSITIVE LOGITS
     drums
    1.77
     drum
    1.63
     barrels
    1.41
     Drum
    1.35
    Drum
    1.30
     Drums
    1.30
     barrel
    1.29
    ドラム
    1.27
    drum
    1.26
    drums
    1.18
    Act Density 0.029%

    No Known Activations