INDEX
    Explanations

    emotional expressions or reactions

    New Auto-Interp
    Negative Logits
    agnostic
    -0.15
    ooth
    -0.14
    deaux
    -0.14
    azio
    -0.14
     Fay
    -0.13
     bail
    -0.13
     Charm
    -0.13
    aisy
    -0.13
    okia
    -0.13
     GOT
    -0.13
    POSITIVE LOGITS
    eca
    0.19
    женÑĮ
    0.17
    ippi
    0.15
    UGIN
    0.15
    eker
    0.14
    -abs
    0.14
    chet
    0.14
    ucks
    0.14
    unched
    0.14
     Abs
    0.14
    Act Density 0.049%

    No Known Activations