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

    Neuronpedia

    INDEX
    Explanations

    information related to crimes and legal proceedings

    oai_token-act-pair · gpt-3.5-turbo

    instances of the verb "to be" indicating the state of individuals involved in legal or violent contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    fy
    -0.71
     greatness
    -0.69
    heastern
    -0.66
    eor
    -0.65
    zsche
    -0.64
    cial
    -0.62
     Dynamics
    -0.61
    inav
    -0.61
    odes
    -0.61
    inventoryQuantity
    -0.60
    POSITIVE LOGITS
     suing
    1.24
     awaiting
    1.02
     recovering
    1.00
     facing
    1.00
     cooperating
    0.95
     believed
    0.93
     survived
    0.93
     accused
    0.90
     suspected
    0.90
     currently
    0.89
    Act Density 0.145%

    No Known Activations