INDEX
    Explanations

    "ain't" and "if" in various languages

    New Auto-Interp
    Negative Logits
     dimers
    0.36
     lancéolées
    0.35
     oscillations
    0.34
     seeded
    0.34
     interstices
    0.33
     ansatz
    0.33
     prismatic
    0.33
     sinusoidal
    0.32
     gradients
    0.32
     dimer
    0.32
    POSITIVE LOGITS
    Hvis
    0.41
    ần
    0.39
    Jeśli
    0.39
    Pokud
    0.36
    Не
    0.36
    у
    0.36
    Га
    0.35
    і
    0.35
    ны
    0.34
    На
    0.34
    Act Density 0.358%

    No Known Activations