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

    Neuronpedia

    INDEX
    Explanations

    phrases related to legal proceedings

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "arraign" in various forms

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    margin
    -0.67
    tz
    -0.65
    hub
    -0.64
    idge
    -0.63
    Roman
    -0.62
    IRO
    -0.62
    itarian
    -0.61
    ebook
    -0.60
    Jordan
    -0.60
     Heritage
    -0.59
    POSITIVE LOGITS
     arra
    1.24
    ignment
    0.92
    uations
    0.86
     plea
    0.81
    engers
    0.80
     hearings
    0.77
    terness
    0.77
    arnaev
    0.76
    urations
    0.76
    halla
    0.75
    Act Density 0.007%

    No Known Activations