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

    Neuronpedia

    INDEX
    Explanations

    words related to legal terms, specifically the concept of guilt

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "guilty" in the context of legal judgments or accusations

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    andel
    -0.79
    ILA
    -0.75
    pid
    -0.73
    flies
    -0.73
    edia
    -0.72
    yip
    -0.72
    abwe
    -0.69
     Gork
    -0.68
    pora
    -0.67
    lav
    -0.67
    POSITIVE LOGITS
     plea
    0.86
     verdict
    0.85
     Guilty
    0.84
     pleas
    0.83
     guilty
    0.80
    isance
    0.78
     unfocusedRange
    0.76
    alty
    0.76
    thouse
    0.74
    alties
    0.73
    Act Density 0.015%

    No Known Activations