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

    Neuronpedia

    INDEX
    Explanations

    phrases related to rankings and superlatives, especially in the context of different categories such as sports, economic indicators, and geographical features

    oai_token-act-pair · gpt-3.5-turbo

    phrases that reference hierarchical rankings or status

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    bsp
    -0.71
    unctions
    -0.68
    abel
    -0.68
    okane
    -0.67
    edu
    -0.67
    Query
    -0.67
    =#
    -0.66
    isson
    -0.66
    arten
    -0.66
    leased
    -0.65
    POSITIVE LOGITS
     shortest
    1.06
     world
    0.92
     weakest
    0.92
     smallest
    0.90
     pursuit
    0.88
     country
    0.86
     largest
    0.85
     toughest
    0.83
     nation
    0.82
     slightest
    0.80
    Act Density 0.202%

    No Known Activations