INDEX
    Explanations
    New Auto-Interp
    Negative Logits
     ट्रिपल
    0.35
    assung
    0.35
    igat
    0.33
    ubhav
    0.32
    0.32
    ंतिक
    0.32
    ƙ
    0.32
    istungs
    0.31
     জেড
    0.31
    ܩ
    0.31
    POSITIVE LOGITS
     half
    4.66
    Half
    4.38
     Half
    4.34
    half
    4.16
    3.88
     HALF
    3.81
    HALF
    3.63
     نصف
    3.45
     setengah
    3.38
     nửa
    3.22
    Act Density 0.162%

    No Known Activations