INDEX
    Explanations

    graphic depictions of violence and brutality

    New Auto-Interp
    Negative Logits
    ipa
    -0.20
    rouch
    -0.17
     Dangerous
    -0.16
     exhaust
    -0.15
    ãĤĥ
    -0.14
     temper
    -0.13
    anes
    -0.13
    anik
    -0.13
    378
    -0.13
    ption
    -0.13
    POSITIVE LOGITS
     organs
    0.26
     dec
    0.24
     gore
    0.23
     severed
    0.23
     entr
    0.22
     limbs
    0.22
     dis
    0.21
     hacked
    0.21
     киÑĪ
    0.20
    /org
    0.20
    Act Density 0.163%

    No Known Activations