INDEX
    Explanations
    New Auto-Interp
    Negative Logits
     trin
    -0.11
    Th
    -0.08
     Hv
    -0.08
    Ic
    -0.07
     Lar
    -0.07
    lis
    -0.07
    trin
    -0.07
     lar
    -0.07
    KT
    -0.07
    wi
    -0.07
    POSITIVE LOGITS
     tun
    0.10
     пристав
    0.08
     occurring
    0.08
     conjunct
    0.08
     denial
    0.07
     Brad
    0.07
     Mechanics
    0.07
    _MAG
    0.07
     addressing
    0.07
     firewall
    0.07
    Act Density 0.006%

    No Known Activations