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

    Neuronpedia

    INDEX
    Explanations

    references to physical restraint or suffocation actions

    oai_token-act-pair · gpt-3.5-turbo

    terms related to instances of choking and physical restraint in law enforcement contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    estate
    -0.79
    issance
    -0.76
    iances
    -0.73
    endar
    -0.71
    estine
    -0.71
    roth
    -0.69
    atism
    -0.68
    Ô
    -0.67
    iday
    -0.67
    utenberg
    -0.66
    POSITIVE LOGITS
    hold
    1.13
     choke
    0.86
     ropes
    0.86
     straps
    0.83
     rope
    0.77
     restrain
    0.77
     wrists
    0.76
     hold
    0.76
     leash
    0.75
     strap
    0.75
    Act Density 0.164%

    No Known Activations