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

    Neuronpedia

    INDEX
    Explanations

    age descriptions in terms of years, which are often related to allegations or incidents

    oai_token-act-pair · gpt-3.5-turbo

    references to the ages of individuals involved in serious incidents, particularly minors

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     lobb
    -0.62
     akin
    -0.60
    Dialogue
    -0.59
     Parables
    -0.56
    hops
    -0.56
    morrow
    -0.56
    omsky
    -0.55
     balcon
    -0.55
    essler
    -0.55
    peak
    -0.55
    POSITIVE LOGITS
     olds
    1.03
    old
    0.82
    OLD
    0.80
     veteran
    0.78
     old
    0.76
    olds
    0.73
     ago
    0.71
     pregnant
    0.66
    iversary
    0.66
    ieving
    0.65
    Act Density 0.060%

    No Known Activations