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

    Neuronpedia

    INDEX
    Explanations

    phrases related to conflict, warfare, or crisis

    oai_token-act-pair · gpt-3.5-turbo

    phrases related to stopping violence or conflict

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ledged
    -0.79
    ittal
    -0.75
    leground
    -0.72
    iences
    -0.69
     volunteered
    -0.68
     cautioned
    -0.68
    elson
    -0.67
    erved
    -0.66
     sacrificed
    -0.63
    soDeliveryDate
    -0.63
    POSITIVE LOGITS
     bleeding
    1.28
     flow
    1.24
     progress
    1.06
     progression
    1.04
     deterioration
    0.96
     tide
    0.95
     madness
    0.95
     onslaught
    0.95
     growth
    0.92
     runaway
    0.92
    Act Density 0.210%

    No Known Activations