INDEX
    Explanations

    expressions of gratitude and familial relationships

    New Auto-Interp
    Negative Logits
     μη
    -0.16
    ylland
    -0.15
    ohl
    -0.15
    ImageUrl
    -0.14
    INU
    -0.14
    ÙĪÙĦÙĬ
    -0.14
    ihan
    -0.14
    abor
    -0.14
    EMALE
    -0.14
    ORY
    -0.14
    POSITIVE LOGITS
     wonderful
    0.23
     dear
    0.21
     beautiful
    0.21
     little
    0.20
     boys
    0.20
     lovely
    0.19
     handsome
    0.18
     sweet
    0.18
     girls
    0.17
    little
    0.17
    Act Density 0.152%

    No Known Activations