INDEX
    Explanations
    New Auto-Interp
    Negative Logits
    ThroughAttribute
    -1.08
    GEBURTSDATUM
    -1.02
    expandindo
    -0.86
    تقاوى
    -0.86
    contentLoaded
    -0.82
    oredCriteria
    -0.81
    rungsseite
    -0.81
     Roskov
    -0.78
    WireFormatLite
    -0.76
     saites
    -0.74
    POSITIVE LOGITS
     better
    0.77
     how
    0.69
     whether
    0.64
     to
    0.59
    better
    0.54
     meglio
    0.51
     if
    0.51
     بهتر
    0.50
    Better
    0.49
     lepiej
    0.49
    Act Density 0.004%

    No Known Activations