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

    Neuronpedia

    INDEX
    Explanations

    proper names related to legal matters

    oai_token-act-pair · gpt-3.5-turbo

    mentions of female names

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    nir
    -1.13
    lasses
    -0.99
    rance
    -0.86
    ugu
    -0.83
    eling
    -0.81
    doms
    -0.76
    indal
    -0.75
    eled
    -0.75
    dfx
    -0.74
    inary
    -0.74
    POSITIVE LOGITS
     Nicole
    1.15
     Knox
    0.96
     herself
    0.93
     Marie
    0.91
    Marie
    0.86
     Mae
    0.85
     Louise
    0.83
     Judd
    0.82
     Amanda
    0.82
    Madison
    0.81
    Act Density 0.043%

    No Known Activations