INDEX
    Explanations
    New Auto-Interp
    Negative Logits
     обоих
    0.63
    どちら
    0.58
     entrambi
    0.57
    いずれ
    0.56
     ambos
    0.55
     obu
    0.55
     entrambe
    0.54
     Ambos
    0.52
     semuanya
    0.51
    而不是
    0.49
    POSITIVE LOGITS
     figuratively
    0.55
     figurative
    0.45
     psychological
    0.43
     sexes
    0.43
     теорети
    0.42
     genders
    0.40
     subtle
    0.40
     an
    0.39
     theoretical
    0.39
    мою
    0.38
    Act Density 0.017%

    No Known Activations