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

    Neuronpedia

    INDEX
    Explanations

    words associated with intense negative events or situations

    oai_token-act-pair · gpt-3.5-turbo

    references to tragic events or situations

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    aton
    -0.84
    ipher
    -0.77
    ebus
    -0.76
    ership
    -0.74
    orney
    -0.71
    igree
    -0.70
    resp
    -0.70
    secut
    -0.70
    elled
    -0.70
    collar
    -0.69
    POSITIVE LOGITS
     tragedy
    0.90
     tragedies
    0.84
     Royale
    0.80
     traged
    0.79
     involving
    0.77
     spree
    0.76
     plag
    0.76
     perpetrated
    0.75
     unfold
    0.75
     Ambro
    0.72
    Act Density 0.011%

    No Known Activations