INDEX
    Explanations

    initialization

    New Auto-Interp
    Negative Logits
     Ler
    -0.06
    =%
    -0.06
    _ll
    -0.06
    -0.06
    ег
    -0.06
    -Semitic
    -0.06
    ien
    -0.06
     Beit
    -0.06
    ́c
    -0.06
     grandchildren
    -0.06
    POSITIVE LOGITS
     strive
    0.07
     cites
    0.07
     consequ
    0.07
     piv
    0.06
     "]";↵
    0.06
     εμφ
    0.06
    .Transactional
    0.06
    0.06
    mam
    0.06
     Japanese
    0.06
    Act Density 0.021%

    No Known Activations