INDEX
    Explanations
    New Auto-Interp
    Negative Logits
     get
    -1.06
    get
    -0.93
     propres
    -0.78
     rése
    -0.69
     rayures
    -0.69
    Get
    -0.67
     vectoriales
    -0.66
     aikana
    -0.66
     preuves
    -0.66
     sèche
    -0.65
    POSITIVE LOGITS
     a
    0.94
     the
    0.88
     an
    0.82
     something
    0.66
     anything
    0.64
     lots
    0.62
     some
    0.59
     several
    0.59
     another
    0.59
     caught
    0.59
    Act Density 0.072%

    No Known Activations