INDEX
    Explanations

    numerical values and mathematical expressions

    New Auto-Interp
    Negative Logits
    amarin
    -0.16
    steller
    -0.15
    bject
    -0.15
    enstein
    -0.14
    amburg
    -0.14
    stands
    -0.14
    gars
    -0.13
     marked
    -0.13
    marked
    -0.13
    ocl
    -0.13
    POSITIVE LOGITS
    orado
    0.16
    éry
    0.16
    CurrentValue
    0.15
    лев
    0.15
    eniable
    0.15
    ãĤ¾
    0.15
    endon
    0.15
    idth
    0.14
    åIJIJ
    0.14
    çļ
    0.14
    Act Density 0.008%

    No Known Activations