INDEX
    Explanations
    New Auto-Interp
    Negative Logits
     turtle
    -0.07
     reconc
    -0.07
     recrut
    -0.07
    _until
    -0.07
     android
    -0.07
     AND
    -0.07
    	AND
    -0.07
     Zato
    -0.07
    един
    -0.07
     Prison
    -0.07
    POSITIVE LOGITS
    тоб
    0.08
    309
    0.08
    xfe
    0.08
    Sve
    0.08
    ligne
    0.08
    Jon
    0.08
    BBox
    0.08
     entschieden
    0.08
     reckon
    0.08
    VBox
    0.08
    Act Density 0.002%

    No Known Activations