INDEX
    Explanations

    references and citations in academic texts

    New Auto-Interp
    Negative Logits
    acco
    -0.14
     Також
    -0.14
    cznie
    -0.14
    aland
    -0.14
    ube
    -0.14
    ноÑģÑı
    -0.13
    лик
    -0.13
     Záp
    -0.13
    å·»
    -0.13
    иж
    -0.13
    POSITIVE LOGITS
     et
    0.24
     ed
    0.23
     eds
    0.21
     compiler
    0.17
    ed
    0.17
    (ed
    0.17
    etal
    0.16
     editor
    0.15
    _editor
    0.15
    _ed
    0.15
    Act Density 0.180%

    No Known Activations