INDEX
    Explanations

    mentions of fruit in the text

    New Auto-Interp
    Negative Logits
    aution
    -0.80
    DOS
    -0.71
    orld
    -0.71
    kson
    -0.64
    entric
    -0.64
     Schwarzenegger
    -0.62
     Worldwide
    -0.62
    govern
    -0.62
    awar
    -0.61
     Govern
    -0.61
    POSITIVE LOGITS
    cake
    1.22
    fruit
    1.15
     juice
    1.06
    cakes
    1.04
     juices
    0.94
    fulness
    0.93
     fruit
    0.90
    nect
    0.85
    less
    0.83
    ruit
    0.82
    Act Density 0.021%

    No Known Activations