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

    Neuronpedia

    INDEX
    Explanations

    specific mentions of different court systems and legal procedures

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
    ec
    -0.75
    ku
    -0.72
    ubi
    -0.69
    dev
    -0.65
    edia
    -0.65
    eele
    -0.65
    aned
    -0.64
     ful
    -0.64
     OD
    -0.63
    ulic
    -0.62
    POSITIVE LOGITS
    rooms
    1.05
     Judge
    0.97
     judge
    0.95
    auld
    0.93
     magistrate
    0.91
     Clerk
    0.89
    room
    0.87
     proceedings
    0.81
     Settlement
    0.80
    iffs
    0.79
    Act Density 0.050%

    No Known Activations