INDEX
    Explanations
    New Auto-Interp
    Negative Logits
     obese
    -0.10
    ubi
    -0.10
     swath
    -0.10
     Brilliant
    -0.10
    lid
    -0.09
     Fat
    -0.09
     wizard
    -0.09
    stag
    -0.09
     Prest
    -0.09
     handsome
    -0.09
    POSITIVE LOGITS
     fit
    0.11
    -fit
    0.11
     beach
    0.11
     iconic
    0.11
    örper
    0.10
     babel
    0.10
    InputDialog
    0.10
    arger
    0.10
    icon
    0.09
     abs
    0.09
    Act Density 0.055%

    No Known Activations