INDEX
    Explanations
    New Auto-Interp
    Negative Logits
    doms
    -1.11
    uthor
    -0.88
    _-
    -0.87
    olicy
    -0.82
    idem
    -0.80
    rolet
    -0.73
    ernel
    -0.73
    uckland
    -0.70
     Cheong
    -0.69
    ancock
    -0.68
    POSITIVE LOGITS
     palette
    1.14
    blind
    1.14
     tint
    0.92
     coded
    0.92
    anguage
    0.91
    ="#
    0.90
     hue
    0.86
     dye
    0.86
     color
    0.85
     Spray
    0.85
    Act Density 0.039%

    No Known Activations