INDEX
    Explanations

    expressions of strong personal preferences or feelings

    New Auto-Interp
    Negative Logits
    agi
    -0.19
    éis
    -0.15
     Woche
    -0.14
    obao
    -0.14
    ilig
    -0.14
    avis
    -0.14
    ContentType
    -0.14
    exual
    -0.13
    amas
    -0.13
     Bless
    -0.13
    POSITIVE LOGITS
    μη
    0.15
    HEMA
    0.14
     Sierra
    0.14
     Güven
    0.14
    anje
    0.14
     Stap
    0.14
     Buen
    0.13
     ÃĶ
    0.13
    /locale
    0.13
    ¼
    0.13
    Act Density 0.000%

    No Known Activations