INDEX
    Explanations

    expressions conveying strong emotions or opinions

    phrases that express strong opinions or hyperbolic statements

    New Auto-Interp
    Negative Logits
    elight
    -0.65
     ¶
    -0.64
    =~=~
    -0.61
    eworks
    -0.61
    =~
    -0.60
    office
    -0.60
    ESPN
    -0.60
    é¾įå¥ij士
    -0.60
    onduct
    -0.60
     Oswald
    -0.59
    POSITIVE LOGITS
    ils
    0.92
     lot
    0.88
     coincidence
    0.88
     heck
    0.81
     waste
    0.77
     bunch
    0.77
     lovely
    0.76
     wonderful
    0.75
     fuss
    0.74
     hypocr
    0.72
    Act Density 0.045%

    No Known Activations