INDEX
    Explanations
    New Auto-Interp
    Negative Logits
    _office
    -0.07
     Feature
    -0.07
     
    -0.07
    -0.06
    итив
    -0.06
    ition
    -0.06
    -eslint
    -0.06
    Spo
    -0.06
    ANCED
    -0.06
    OTHER
    -0.06
    POSITIVE LOGITS
     margins
    0.07
    <ID
    0.07
    oad
    0.06
    ellipse
    0.06
    kyt
    0.06
    Brightness
    0.06
     elasticity
    0.06
     ECM
    0.06
    .cf
    0.06
     Bağ
    0.06
    Act Density 0.003%

    No Known Activations