INDEX
    Explanations
    New Auto-Interp
    Negative Logits
     đau
    0.50
    0.41
     kullanıcı
    0.41
     troubleshoot
    0.38
    estimate
    0.38
     douleurs
    0.38
     rencana
    0.38
     perencanaan
    0.38
     đức
    0.37
     بتاعتنا
    0.37
    POSITIVE LOGITS
     irregular
    0.58
     interleaved
    0.55
     encoding
    0.55
     OE
    0.54
     Polynomial
    0.53
     polynomial
    0.53
     concatenation
    0.53
     subsequence
    0.52
     alternating
    0.51
     irregularity
    0.51
    Act Density 0.013%

    No Known Activations