INDEX
    Explanations

    directions and boundaries

    New Auto-Interp
    Negative Logits
    Reason
    0.79
    Prima
    0.78
    İlk
    0.72
    车型
    0.71
    AssemblyName
    0.70
     yerden
    0.70
    Estado
    0.69
     utilisateurs
    0.69
    Usuario
    0.69
    Arvore
    0.69
    POSITIVE LOGITS
     portion
    1.13
     half
    1.01
     part
    0.97
     частина
    0.89
     bound
    0.87
     edge
    0.86
     portions
    0.86
    portion
    0.86
     Portion
    0.85
     boundary
    0.84
    Act Density 0.085%

    No Known Activations