INDEX
    Explanations
    New Auto-Interp
    Negative Logits
     wouldn
    0.93
     wouldnt
    0.91
     kunne
    0.89
     могли
    0.88
     not
    0.88
    not
    0.86
     pudieran
    0.84
     Could
    0.83
     kemungkinan
    0.83
     couldn
    0.82
    POSITIVE LOGITS
     sometimes
    1.75
     often
    1.63
     parfois
    1.54
     oftentimes
    1.49
    sometimes
    1.46
     souvent
    1.45
     Sometimes
    1.42
    Sometimes
    1.41
     soms
    1.41
     spesso
    1.39
    Act Density 0.143%

    No Known Activations