INDEX
    Explanations
    New Auto-Interp
    Negative Logits
    eering
    -0.97
     Amend
    -0.68
    Files
    -0.64
    doms
    -0.63
    ures
    -0.63
     tremend
    -0.62
    Reloaded
    -0.61
    Sax
    -0.61
    */(
    -0.60
    -+-+-+-+
    -0.60
    POSITIVE LOGITS
    melon
    1.46
    colour
    1.06
    Bott
    1.03
     bottles
    1.03
    bott
    1.02
    tight
    0.98
     bottle
    0.97
     Bottle
    0.93
    loo
    0.92
     fluor
    0.90
    Act Density 1.992%

    No Known Activations