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

    Neuronpedia

    INDEX
    Explanations

    phrases related to destruction or violence

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "destroy."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ETF
    -0.83
    annis
    -0.74
    DragonMagazine
    -0.68
    çīĪ
    -0.67
    liner
    -0.66
    matic
    -0.66
    culation
    -0.64
    brow
    -0.63
    BuyableInstoreAndOnline
    -0.63
    Chart
    -0.63
    POSITIVE LOGITS
     havoc
    1.08
    roying
    0.95
     wre
    0.74
     arte
    0.71
     spree
    0.70
     destruction
    0.69
    ishing
    0.67
     warheads
    0.67
    ãĥĩãĤ£
    0.66
    ãĥł
    0.66
    Act Density 0.032%

    No Known Activations