INDEX
    Explanations
    New Auto-Interp
    Negative Logits
     simplicity
    0.47
     becomes
    0.46
     simplification
    0.45
     menjadi
    0.45
     astonishment
    0.45
     become
    0.44
     diventa
    0.44
     aure
    0.43
     elit
    0.42
     يصير
    0.42
    POSITIVE LOGITS
     Paar
    0.43
     чтобы
    0.38
     Ramp
    0.38
     Deutschlands
    0.37
    بیداری
    0.37
     щоб
    0.37
     تاکہ
    0.37
    ຕົວ
    0.37
    Roasted
    0.36
     wèi
    0.36
    Act Density 0.014%

    No Known Activations