INDEX
    Explanations

    phrases related to thinness or slimness

    New Auto-Interp
    Negative Logits
    ivals
    -0.83
    church
    -0.73
    masters
    -0.72
    amaz
    -0.69
    kamp
    -0.68
     Tours
    -0.67
    eering
    -0.65
    abad
    -0.63
    hurst
    -0.63
     celebrations
    -0.62
    POSITIVE LOGITS
     thin
    3.72
     thinner
    2.59
    thin
    2.57
     Thin
    2.28
     thick
    1.94
     thicker
    1.91
     slim
    1.77
     slender
    1.58
     thinly
    1.53
     thickness
    1.50
    Act Density 0.008%

    No Known Activations