INDEX
    Explanations

    references to Tibetan culture and related terms

    New Auto-Interp
    Negative Logits
    clave
    -0.18
    abin
    -0.15
    ulan
    -0.14
    arrison
    -0.14
    rig
    -0.14
    umpy
    -0.14
    forge
    -0.14
    habi
    -0.14
    bye
    -0.14
    owns
    -0.14
    POSITIVE LOGITS
    leck
    0.14
    beb
    0.14
    ück
    0.14
     cuckold
    0.14
    ationToken
    0.14
    оÑİ
    0.13
     ull
    0.13
    urm
    0.13
    agy
    0.13
    ocom
    0.13
    Act Density 0.004%

    No Known Activations