INDEX
    Explanations
    New Auto-Interp
    Negative Logits
    _geometry
    -0.07
    amphetamine
    -0.07
     powdered
    -0.07
    нями
    -0.07
     tem
    -0.06
     cellar
    -0.06
    GREEN
    -0.06
    	vm
    -0.06
     состоянии
    -0.06
    838
    -0.06
    POSITIVE LOGITS
    tığı
    0.07
    poke
    0.06
     různé
    0.06
     cartoon
    0.06
    :normal
    0.06
    _Row
    0.06
    _pred
    0.06
     website
    0.06
    -cond
    0.06
     demo
    0.06
    Act Density 0.144%

    No Known Activations