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

    Neuronpedia

    INDEX
    Explanations

    references to specific tragic events or disasters, particularly related to terrorist attacks or aviation incidents

    oai_token-act-pair · gpt-3.5-turbo

    references to terrorist attacks and related events

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    dit
    -0.81
    requires
    -0.76
    Dub
    -0.72
    utical
    -0.70
     Hawai
    -0.69
     Habit
    -0.69
    lishes
    -0.67
    olor
    -0.67
     Protoss
    -0.67
    Sense
    -0.66
    POSITIVE LOGITS
     massacre
    1.13
     bombing
    1.10
     bombings
    1.05
     victims
    1.04
     tragedy
    1.04
     aftermath
    1.02
     shootings
    1.01
     Massacre
    1.00
     shooter
    0.98
     killings
    0.96
    Act Density 0.220%

    No Known Activations