INDEX
    Explanations

    terms indicating close friendships or relationships

    New Auto-Interp
    Negative Logits
    égor
    -0.16
    ihn
    -0.15
    ardon
    -0.15
    tober
    -0.15
    agos
    -0.15
    vor
    -0.15
    ÏĦÏģÎŃ
    -0.14
    _endian
    -0.14
    hora
    -0.14
    eyer
    -0.14
    POSITIVE LOGITS
    etics
    0.17
     Quest
    0.15
    anken
    0.15
    pects
    0.14
    beiter
    0.14
     ØŃÙħ
    0.14
    ictim
    0.14
    nano
    0.14
    inx
    0.13
    inqu
    0.13
    Act Density 0.007%

    No Known Activations