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

    Neuronpedia

    INDEX
    Explanations

    family relationship terms

    oai_token-act-pair · gpt-3.5-turbo

    references to familial and legal relationships

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    spir
    -0.92
    uve
    -0.83
    ankind
    -0.80
    interstitial
    -0.79
    STEM
    -0.67
     Carbuncle
    -0.66
    issions
    -0.65
    ahime
    -0.65
    kefeller
    -0.65
    tnc
    -0.65
    POSITIVE LOGITS
     solicitor
    0.86
     uncle
    0.81
     attorney
    0.77
     sister
    0.76
     lawyer
    0.75
     cried
    0.75
     plead
    0.74
     pleaded
    0.74
     aunt
    0.73
     objected
    0.72
    Act Density 0.107%

    No Known Activations