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

    Neuronpedia

    INDEX
    Explanations

    phrases related to family and social relationships

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
     bridges
    -0.75
    Gate
    -0.67
     issu
    -0.65
    worthiness
    -0.63
    uncture
    -0.63
    resy
    -0.63
     repetition
    -0.62
     surpassed
    -0.61
    vell
    -0.61
    ivably
    -0.60
    POSITIVE LOGITS
    mates
    0.85
     mates
    0.82
     downstairs
    0.77
     belongings
    0.76
    forts
    0.70
     cousins
    0.70
    selves
    0.70
     sleeping
    0.70
     goodbye
    0.69
    emate
    0.66
    Act Density 0.248%

    No Known Activations