INDEX
    Explanations
    New Auto-Interp
    Negative Logits
     vœux
    -0.82
     bienfaits
    -0.81
     bénéfices
    -0.78
     récompenses
    -0.78
     lèvres
    -0.77
     épaules
    -0.77
     émissions
    -0.75
     pouvoirs
    -0.74
     âmes
    -0.73
     oreilles
    -0.73
    POSITIVE LOGITS
     forms
    0.77
     boards
    0.73
     modes
    0.70
     fleets
    0.69
     choirs
    0.69
     contexts
    0.68
     halls
    0.68
     groups
    0.67
     chains
    0.67
     environments
    0.67
    Act Density 0.050%

    No Known Activations