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

    Neuronpedia

    INDEX
    Explanations

    death and killing

    np_max-act · gemini-2.0-flash

    The neuron activates on words describing killing or lethal methods (e.g. kill, killing, killed, death, painless, humane).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     NA
    -0.07
    앙
    -0.06
     желуд
    -0.06
    ி
    -0.06
     Psych
    -0.06
    _GAP
    -0.06
    Matchers
    -0.06
    UIT
    -0.06
     sympathetic
    -0.06
    ESİ
    -0.06
    POSITIVE LOGITS
     travelling
    0.06
     maritime
    0.06
     Updating
    0.06
    OLUM
    0.06
     sources
    0.06
    nbr
    0.06
     ≈
    0.06
     Catholics
    0.06
    大家
    0.06
     installed
    0.06
    Act Density 0.021%

    No Known Activations