INDEX
    Explanations
    New Auto-Interp
    Negative Logits
     roze
    -0.07
     counter
    -0.07
     Sparks
    -0.07
    Constant
    -0.07
    .counter
    -0.07
    -counter
    -0.07
    _constant
    -0.07
    602
    -0.07
    constant
    -0.07
     constant
    -0.07
    POSITIVE LOGITS
     مباشرة
    0.11
     attach
    0.11
    attach
    0.10
     attaches
    0.10
     مباشر
    0.09
     Attach
    0.09
    Attach
    0.09
     direkten
    0.09
     abonnement
    0.09
     accessing
    0.09
    Act Density 0.017%

    No Known Activations