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

    Neuronpedia

    INDEX
    Explanations

    phrases related to the act of creating or inventing something

    oai_token-act-pair · gpt-3.5-turbo

    phrases indicating the action of "building up" or "increasing" something

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    "]=>
    -0.60
     benefit
    -0.59
     bonded
    -0.59
    heses
    -0.58
     segregated
    -0.58
     segreg
    -0.57
     beneficiary
    -0.57
    Maker
    -0.56
    iw
    -0.55
    gee
    -0.55
    POSITIVE LOGITS
    river
    0.82
    dL
    0.70
    ̶
    0.66
    rights
    0.66
    dt
    0.65
    olicy
    0.63
    skirts
    0.62
    UFF
    0.61
    votes
    0.61
    raised
    0.61
    Act Density 0.063%

    No Known Activations