INDEX
    Explanations

    mentions of email and login-related terms

    New Auto-Interp
    Negative Logits
    alach
    -0.17
     segment
    -0.16
    ниÑĤ
    -0.16
    avez
    -0.16
    yn
    -0.16
    ange
    -0.15
    ili
    -0.15
    iversit
    -0.15
    anky
    -0.15
    hausen
    -0.15
    POSITIVE LOGITS
     protected
    0.23
    protected
    0.18
    ekil
    0.17
    Protected
    0.16
    rescia
    0.15
     Protected
    0.15
    ZY
    0.14
    mailto
    0.14
     red
    0.14
    /vendors
    0.14
    Act Density 0.003%

    No Known Activations