INDEX
    Explanations
    New Auto-Interp
    Negative Logits
    вей
    -0.08
     vowel
    -0.08
     sinners
    -0.08
     голос
    -0.08
     venha
    -0.08
     parro
    -0.08
     conces
    -0.08
     vowels
    -0.08
     voters
    -0.08
    ąć
    -0.07
    POSITIVE LOGITS
     cookware
    0.11
     utens
    0.09
    abant
    0.08
     prized
    0.08
     stove
    0.08
    alatan
    0.08
    -balanced
    0.08
     Radeon
    0.08
    0.08
     Adapter
    0.08
    Act Density 0.020%

    No Known Activations