INDEX
    Explanations
    New Auto-Interp
    Negative Logits
     Bir
    -0.06
     insol
    -0.06
    -cn
    -0.06
    output
    -0.06
     nut
    -0.06
     cables
    -0.06
     apps
    -0.06
     maze
    -0.06
     digits
    -0.06
     círk
    -0.06
    POSITIVE LOGITS
     tomato
    0.12
     Tomato
    0.10
     tomatoes
    0.09
     strawberry
    0.08
    .FONT
    0.07
    τομα
    0.07
     FactoryGirl
    0.07
    .uml
    0.07
     submissive
    0.07
    	EIF
    0.06
    Act Density 0.003%

    No Known Activations