INDEX
    Explanations
    New Auto-Interp
    Negative Logits
     group
    0.63
     incriminating
    0.61
     Company
    0.60
     친구
    0.60
    集团
    0.59
     identific
    0.59
     apothecary
    0.59
     secretario
    0.58
     botan
    0.58
     Methodist
    0.57
    POSITIVE LOGITS
     largeur
    1.15
     width
    1.09
     largura
    1.08
     dimensioni
    1.04
     resizing
    1.04
     resize
    1.02
    width
    1.01
    尺寸
    1.01
     Resize
    0.98
     tamaños
    0.98
    Act Density 0.479%

    No Known Activations