INDEX
    Explanations
    New Auto-Interp
    Negative Logits
     itſelf
    -1.05
     parson
    -1.04
     carriages
    -1.02
     donkeys
    -1.02
     chapels
    -1.01
     handcuffs
    -1.00
     violins
    -1.00
     barbarians
    -1.00
     photolibrary
    -0.99
     Octave
    -0.99
    POSITIVE LOGITS
     in
    0.64
    0.61
     or
    0.61
     en
    0.60
    ful
    0.60
     of
    0.59
     for
    0.58
     —
    0.57
     type
    0.56
     the
    0.55
    Act Density 0.100%

    No Known Activations