INDEX
    Explanations
    New Auto-Interp
    Negative Logits
     opts
    -0.07
     Athlete
    -0.07
     विव
    -0.07
    -0.07
     reh
    -0.07
    -0.07
     athlete
    -0.07
     Thickness
    -0.07
     weich
    -0.07
     thicker
    -0.07
    POSITIVE LOGITS
    _ng
    0.08
     وجه
    0.08
     gjerne
    0.08
     visage
    0.07
     مل
    0.07
    -kind
    0.07
    .ng
    0.07
     grote
    0.07
    ->[
    0.07
    147
    0.07
    Act Density 0.001%

    No Known Activations