INDEX
    Explanations

    names of people associated with television and entertainment

    New Auto-Interp
    Negative Logits
       
    -0.24
    lessness
    -0.17
    iness
    -0.17
    eci
    -0.16
    viar
    -0.15
    fulness
    -0.15
    enberg
    -0.15
    per
    -0.15
    virt
    -0.14
    bin
    -0.14
    POSITIVE LOGITS
    angelo
    0.20
    ablish
    0.15
    ÑģÑĮ
    0.15
    Ïĥη
    0.15
    ãĥ£
    0.15
    cy
    0.14
    cratch
    0.14
    à¥Ģय
    0.14
    .deb
    0.14
    rtype
    0.14
    Act Density 0.686%

    No Known Activations