INDEX
    Explanations

    proper names of individuals, likely focusing on names related to articles or topics discussed

    New Auto-Interp
    Negative Logits
     accord
    -0.15
    etr
    -0.14
    edis
    -0.14
    ãĥĨãĥ«
    -0.14
    gres
    -0.14
     ÙĨÙħ
    -0.14
     JIT
    -0.13
    ondo
    -0.13
    jas
    -0.13
    iesta
    -0.13
    POSITIVE LOGITS
    qml
    0.18
     Lid
    0.15
    sik
    0.15
    elsing
    0.15
    WithTitle
    0.14
    αιν
    0.13
    pis
    0.13
    eck
    0.13
    zy
    0.13
    swers
    0.13
    Act Density 0.009%

    No Known Activations