INDEX
    Explanations

    terms related to civilian casualties and warfare

    New Auto-Interp
    Negative Logits
    rafted
    -0.15
    oram
    -0.14
    igan
    -0.14
    Attempt
    -0.14
    usted
    -0.14
    تÙģ
    -0.13
    mith
    -0.13
     pers
    -0.13
     пеÑģ
    -0.13
    Canon
    -0.13
    POSITIVE LOGITS
    olini
    0.15
     Sole
    0.14
    UILayout
    0.14
    ÄĽr
    0.14
     Geneva
    0.14
    sole
    0.14
    odyn
    0.14
    리ìĹIJ
    0.14
     Leaf
    0.14
    Leaf
    0.14
    Act Density 0.032%

    No Known Activations