INDEX
    Explanations
    New Auto-Interp
    Negative Logits
    Prov
    -0.08
     ni
    -0.08
    hum
    -0.07
     provoc
    -0.07
    -0.07
     murdering
    -0.07
                                      
    -0.07
    ිරි
    -0.07
     बनी
    -0.07
    _check
    -0.07
    POSITIVE LOGITS
     sapphire
    0.09
     hinweg
    0.08
    thin
    0.07
     घोषणा
    0.07
     Dao
    0.07
     reporters
    0.07
     centimeter
    0.07
    по
    0.07
     trough
    0.07
     Gamer
    0.07
    Act Density 0.012%

    No Known Activations