Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    INDEX
    Explanations

    references to violence and casualties in political or conflict-related contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     damaging
    -0.15
     thief
    -0.15
    linkplain
    -0.15
    heimer
    -0.14
    _pci
    -0.14
    urum
    -0.14
    dam
    -0.14
    ÄĻż
    -0.14
    亡
    -0.14
     æIJ
    -0.14
    POSITIVE LOGITS
     Mass
    0.45
     mass
    0.44
    mass
    0.41
     massacre
    0.41
     slaughter
    0.39
    Mass
    0.38
     sla
    0.38
    _mass
    0.34
     butcher
    0.34
     massa
    0.34
    Act Density 0.029%

    No Known Activations