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

    Neuronpedia

    INDEX
    Explanations

    references to individuals who have been convicted of crimes

    oai_token-act-pair · gpt-3.5-turbo

    references to individuals who have been convicted of crimes

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    mentation
    -0.86
    mented
    -0.80
    earable
    -0.75
    aven
    -0.71
    eah
    -0.71
    yip
    -0.69
    erved
    -0.69
    idge
    -0.67
    hare
    -0.66
    agnetic
    -0.66
    POSITIVE LOGITS
     felon
    1.03
    icts
    0.79
     convict
    0.73
    rupt
    0.73
     guilty
    0.72
     fel
    0.72
    llor
    0.72
     rehabilit
    0.69
    iary
    0.68
     perjury
    0.68
    Act Density 0.047%

    No Known Activations