INDEX
    Explanations
    New Auto-Interp
    Negative Logits
     siding
    -0.07
    υν
    -0.07
    _sibling
    -0.06
     COLOR
    -0.06
    PAIR
    -0.06
    -e
    -0.06
     Technique
    -0.06
    TOTYPE
    -0.06
    isté
    -0.06
     indicate
    -0.06
    POSITIVE LOGITS
     pense
    0.07
    	bl
    0.06
    Closing
    0.06
     perd
    0.06
    	os
    0.06
    enemy
    0.06
    0.06
     cout
    0.06
    .build
    0.06
    284
    0.06
    Act Density 0.019%

    No Known Activations