INDEX
    Explanations
    New Auto-Interp
    Negative Logits
     predicates
    0.73
     rhythmic
    0.73
     current
    0.71
     rhythm
    0.71
     subsum
    0.69
     functions
    0.69
     copious
    0.69
     familiar
    0.68
     constraint
    0.67
     contemplative
    0.66
    POSITIVE LOGITS
     연구
    1.17
     실험
    1.09
     исследования
    1.07
    试验
    1.07
     chercheurs
    1.07
     провели
    1.04
     الدراسة
    1.00
     estudo
    0.99
     исследование
    0.99
     조사
    0.98
    Act Density 0.276%

    No Known Activations