INDEX
    Explanations
    New Auto-Interp
    Negative Logits
    _por
    -0.08
     estil
    -0.08
     stylist
    -0.08
     evident
    -0.08
     beginner
    -0.08
     typography
    -0.08
     portrait
    -0.08
     Typography
    -0.07
     strtolower
    -0.07
    ЕК
    -0.07
    POSITIVE LOGITS
    共享
    0.19
     shared
    0.17
    Shared
    0.16
     공유
    0.16
     साझा
    0.15
    shared
    0.15
     mmap
    0.15
    (shared
    0.15
     partagé
    0.15
    _shared
    0.14
    Act Density 0.004%

    No Known Activations