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

    Neuronpedia

    INDEX
    Explanations

    words related to acts of terrorism

    oai_token-act-pair · gpt-3.5-turbo

    references to terrorism and related threats

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    bye
    -0.75
     Quartz
    -0.70
    lease
    -0.69
    resso
    -0.68
     Scotch
    -0.66
    galitarian
    -0.65
     Clerk
    -0.65
    flush
    -0.64
     Decker
    -0.63
    Ģ
    -0.63
    POSITIVE LOGITS
    izing
    1.11
    istic
    1.01
    ising
    1.01
     attacks
    0.95
    ization
    0.94
    ised
    0.91
    icious
    0.90
    ized
    0.90
    isation
    0.89
     bombings
    0.86
    Act Density 0.021%

    No Known Activations