INDEX
    Explanations

    special characters and diacritics in the text

    New Auto-Interp
    Negative Logits
    ander
    -0.17
    eward
    -0.17
    ql
    -0.15
    ocity
    -0.15
     Flux
    -0.14
    lut
    -0.14
    opia
    -0.14
    ĩa
    -0.14
    oming
    -0.14
    .ms
    -0.14
    POSITIVE LOGITS
    istically
    0.17
    .Sdk
    0.17
    Ÿ
    0.17
    alist
    0.16
    emens
    0.15
    ²
    0.15
    zac
    0.14
    gid
    0.14
    ¼
    0.14
    keit
    0.14
    Act Density 0.006%

    No Known Activations