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

    Neuronpedia

    INDEX
    Explanations

    words related to intelligence and strategic thinking

    oai_token-act-pair · gpt-3.5-turbo

    terms associated with cleverness and cunning strategies

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    asta
    -0.72
    acid
    -0.70
    thood
    -0.69
    astery
    -0.64
    thur
    -0.63
    aru
    -0.63
    imester
    -0.62
    olina
    -0.62
    astered
    -0.61
    artifacts
    -0.60
    POSITIVE LOGITS
    ly
    1.04
    sonian
    0.87
     tricks
    0.83
     twist
    0.81
     disgu
    0.80
     enough
    0.80
     clever
    0.80
     ploy
    0.80
     dece
    0.77
    nesses
    0.75
    Act Density 0.073%

    No Known Activations