INDEX
    Explanations

    options, depending, ranging

    New Auto-Interp
    Negative Logits
     щоб
    0.41
     quello
    0.39
     digamos
    0.39
     essi
    0.38
     sucedido
    0.38
     substitution
    0.38
     yada
    0.38
    hard
    0.37
     grandma
    0.37
     Floyd
    0.37
    POSITIVE LOGITS
     Depends
    0.48
     вариантов
    0.45
     다양한
    0.43
     различных
    0.43
     различными
    0.42
    ImageBeforeText
    0.42
     তান
    0.41
    गा
    0.40
    ('*
    0.40
    Depending
    0.39
    Act Density 0.001%

    No Known Activations