INDEX
    Explanations

    punctuation marks, particularly commas and periods

    New Auto-Interp
    Negative Logits
    rosse
    -0.15
    ÙħØ©
    -0.14
    leton
    -0.14
     (__
    -0.14
    idis
    -0.14
     Icon
    -0.14
    vel
    -0.13
     Homepage
    -0.13
    Icon
    -0.13
     fora
    -0.13
    POSITIVE LOGITS
    ogi
    0.18
     <<-
    0.17
    Sharper
    0.17
    onto
    0.15
    /Typography
    0.15
    ollen
    0.15
    oki
    0.14
    rah
    0.13
    AAD
    0.13
     kolo
    0.13
    Act Density 0.007%

    No Known Activations