INDEX
    Explanations

    references to facial features, particularly related to cheeks

    New Auto-Interp
    Negative Logits
    пеÑĩ
    -0.14
     âĦķ
    -0.14
    bens
    -0.14
    tin
    -0.14
     THR
    -0.14
    æ¼
    -0.14
    ollo
    -0.14
    rowing
    -0.14
     Ear
    -0.14
    ÑĹ
    -0.13
    POSITIVE LOGITS
    termin
    0.16
    endar
    0.15
    ums
    0.14
    vertise
    0.14
    .named
    0.14
    Ymd
    0.14
    seys
    0.13
    OCR
    0.13
     migrationBuilder
    0.13
     Cunning
    0.13
    Act Density 0.012%

    No Known Activations