INDEX
    Explanations

    mentions of Balkan-related context and terminology

    New Auto-Interp
    Negative Logits
    esty
    -0.16
     Erf
    -0.14
    landa
    -0.14
     ERC
    -0.14
    787
    -0.14
    osoph
    -0.14
    igail
    -0.14
    aepernick
    -0.13
    atte
    -0.13
    ilin
    -0.13
    POSITIVE LOGITS
    ancing
    0.18
    azaar
    0.17
    iffs
    0.16
    aji
    0.16
    wick
    0.15
    azar
    0.15
    anced
    0.15
    лив
    0.14
    enticated
    0.14
    boa
    0.14
    Act Density 0.025%

    No Known Activations