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

    Neuronpedia

    INDEX
    Explanations

    people's names, specifically those containing the letter 'A'

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
     anime
    -0.66
     undergrad
    -0.64
     Eclipse
    -0.59
     enterprise
    -0.58
     adults
    -0.57
     aggregate
    -0.57
     intended
    -0.57
     millions
    -0.57
    Army
    -0.57
     enclosed
    -0.56
    POSITIVE LOGITS
    hern
    1.12
    veyard
    1.12
    oki
    1.07
    cknowled
    1.06
    vil
    1.03
    ileen
    1.01
    verett
    1.01
    uld
    1.00
    plin
    1.00
    lder
    1.00
    Act Density 0.023%

    No Known Activations