INDEX
    Explanations

    say "capitalized word"

    New Auto-Interp
    Negative Logits
     PATH
    -0.08
     муль
    -0.08
    DUCT
    -0.08
     विल
    -0.08
     фин
    -0.07
    /sdk
    -0.07
     scratch
    -0.07
     पहली
    -0.07
    	Path
    -0.07
    	IL
    -0.07
    POSITIVE LOGITS
     Neb
    0.10
     Ob
    0.10
    Ob
    0.10
     mwezi
    0.09
     grape
    0.09
    Tan
    0.09
    olloin
    0.08
    Cherry
    0.08
    ilight
    0.08
     Albuquerque
    0.08
    Act Density 0.003%

    No Known Activations