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

    Neuronpedia

    INDEX
    Explanations

    phrases or descriptions related to violent actions or gruesome incidents

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Negative Logits
     Volunteers
    -0.62
     Anthem
    -0.61
     Observ
    -0.61
    orea
    -0.58
     anten
    -0.58
     deliberations
    -0.57
     hospitality
    -0.56
    pard
    -0.56
     outgoing
    -0.56
     Observer
    -0.56
    POSITIVE LOGITS
     perfection
    0.86
     shred
    0.85
     pieces
    0.82
     pulp
    0.82
    wered
    0.78
     bits
    0.76
     Pieces
    0.75
    ãĤ©
    0.75
     hell
    0.73
    ãĥİ
    0.73
    Act Density 0.320%

    No Known Activations