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

    Neuronpedia

    INDEX
    Explanations

    drugs and harmful substances

    np_acts-logits-general · gemini-2.5-flash-lite

    statements that refuse or caution against illegal or harmful activities—especially drug manufacture/use—while delivering safety disclaimers, risk warnings, and help resources.

    oai_token-act-pair · gpt-5Triggered by @vetterc0
    New Auto-Interp
    Negative Logits
    Customer
    0.51
    Excel
    0.48
    舰
    0.48
    บริษัท
    0.47
    项目
    0.47
     ቅር
    0.47
     고객
    0.46
     Customer
    0.46
    创建
    0.46
    Owners
    0.46
    POSITIVE LOGITS
     drugs
    1.09
     heroin
    1.09
     drug
    1.08
     addiction
    1.07
     drogas
    1.05
     fentanyl
    1.05
     alcohol
    1.00
     нарко
    1.00
     cocaine
    0.97
     substances
    0.96
    Act Density 0.190%

    No Known Activations