INDEX
    Explanations
    New Auto-Interp
    Negative Logits
     fingert
    -0.09
     upgr
    -0.09
     oppon
    -0.08
     Chill
    -0.08
     vibrator
    -0.08
     माय
    -0.08
     miner
    -0.08
    Dam
    -0.08
     Česk
    -0.08
     rumo
    -0.08
    POSITIVE LOGITS
    but
    0.08
    ня
    0.08
    	n
    0.07
    voir
    0.07
    няка
    0.07
    	sort
    0.07
    язательно
    0.07
    ewa
    0.07
    haft
    0.07
     achten
    0.07
    Act Density 0.001%

    No Known Activations