INDEX
    Explanations

    phrases that emphasize inclusivity and universality

    New Auto-Interp
    Negative Logits
    mens
    -0.14
    dney
    -0.14
    out
    -0.14
    è¼Ķ
    -0.13
    addtogroup
    -0.13
    onomy
    -0.13
    _unix
    -0.13
    ounty
    -0.13
    iffer
    -0.13
    jong
    -0.13
    POSITIVE LOGITS
    amat
    0.15
    æ¯ķ
    0.15
    uv
    0.15
    eryl
    0.15
    Gate
    0.15
    ÑĨиÑĤ
    0.15
    AGING
    0.14
     else
    0.14
     ########.
    0.14
     Tet
    0.14
    Act Density 0.020%

    No Known Activations