INDEX
    Explanations
    New Auto-Interp
    Negative Logits
     conflicts
    -0.07
    164
    -0.07
    COVERY
    -0.07
    ètre
    -0.07
    isecond
    -0.07
     doivent
    -0.07
     floats
    -0.07
     Cors
    -0.06
     notions
    -0.06
     downward
    -0.06
    POSITIVE LOGITS
     emph
    0.15
     emphasize
    0.14
     emphas
    0.11
     emphasis
    0.10
     emphasized
    0.10
    emphasis
    0.09
     Memphis
    0.09
     emphasizing
    0.09
    	JPanel
    0.08
    phasis
    0.07
    Act Density 0.007%

    No Known Activations