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

    Neuronpedia

    INDEX
    Explanations

    phrases indicating criminal activity, particularly those that suggest concealment or punishment

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    abox
    -0.16
    isay
    -0.16
    .getInteger
    -0.14
     ì¶ķ
    -0.14
    .LayoutStyle
    -0.14
    imid
    -0.14
    agi
    -0.14
     جÙĦ
    -0.14
    quest
    -0.14
    odom
    -0.14
    POSITIVE LOGITS
     Stealth
    0.15
    cover
    0.15
     خد
    0.15
     b
    0.14
     Cover
    0.14
     cover
    0.14
    ouflage
    0.14
    женÑĮ
    0.14
    ìŀ¬
    0.14
    uth
    0.14
    Act Density 0.284%

    No Known Activations