INDEX
    Explanations

    punctuation marks and their associated emotional expressions

    New Auto-Interp
    Negative Logits
    ereum
    -0.17
    rome
    -0.16
    cery
    -0.15
    burgh
    -0.14
    dong
    -0.14
    ptune
    -0.14
    SHIFT
    -0.13
    ä»ĺãģį
    -0.13
    bish
    -0.13
    lass
    -0.13
    POSITIVE LOGITS
    .fac
    0.16
    ense
    0.16
    kö
    0.16
    _closure
    0.16
    flix
    0.15
    iese
    0.14
    ằng
    0.14
    atter
    0.14
    oint
    0.14
     Ze
    0.14
    Act Density 0.005%

    No Known Activations