INDEX
    Explanations
    New Auto-Interp
    Negative Logits
    (Id
    -0.08
    ullan
    -0.08
     недвижимости
    -0.08
     недвиж
    -0.08
    -0.07
    -0.07
    vy
    -0.07
     Lan
    -0.07
     Skyl
    -0.07
     Schwartz
    -0.07
    POSITIVE LOGITS
    0.10
     picturesque
    0.09
     rustic
    0.09
     creamy
    0.09
    Punch
    0.08
    hout
    0.08
     distortion
    0.08
     distort
    0.08
     distorted
    0.08
    整体
    0.08
    Act Density 0.017%

    No Known Activations