INDEX
    Explanations
    New Auto-Interp
    Negative Logits
     گئی
    0.68
    uruh
    0.61
     Finished
    0.61
    と思った
    0.61
     вышел
    0.60
    Finished
    0.60
    えた
    0.59
     megh
    0.57
    0.56
     あの
    0.55
    POSITIVE LOGITS
     historically
    1.57
     always
    1.51
     traditionally
    1.41
     long
    1.31
     ALWAYS
    1.28
     consistently
    1.27
     longstanding
    1.27
     всегда
    1.26
    always
    1.25
     siempre
    1.18
    Act Density 0.108%

    No Known Activations