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

    Neuronpedia

    INDEX
    Explanations

    family relationships

    oai_token-act-pair · gpt-3.5-turbo

    mentions of parental figures and family relationships

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     hikers
    -0.69
    "]=>
    -0.67
    Increases
    -0.65
    cerning
    -0.63
    umbn
    -0.63
    owler
    -0.63
    =>
    -0.62
    awar
    -0.62
    estyles
    -0.61
    intensity
    -0.61
    POSITIVE LOGITS
     died
    1.00
     divorced
    0.95
     uncle
    0.80
    ma
    0.80
     perished
    0.77
    hood
    0.76
     deceased
    0.73
     disappro
    0.73
     ancestors
    0.73
     ancest
    0.72
    Act Density 0.066%

    No Known Activations