INDEX
    Explanations
    New Auto-Interp
    Negative Logits
     vowel
    2.77
     covariant
    2.74
     bothered
    2.63
     urge
    2.61
     विराम
    2.57
    Brightness
    2.56
    Yeni
    2.54
     volver
    2.51
    2.47
     βά
    2.47
    POSITIVE LOGITS
    GINIA
    2.95
    ैस
    2.55
     Heel
    2.45
    èbre
    2.43
    ität
    2.42
    دون
    2.41
    вання
    2.35
    ني
    2.35
    স্পতি
    2.30
    णिज्य
    2.27
    Act Density 0.042%

    No Known Activations