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

    Neuronpedia

    INDEX
    Explanations

    Generic text

    np_max-act · gemini-2.0-flash

    This neuron activates on the formal definition and instruction language used to specify sexual‐content policy (e.g. words like “Content,” “meant,” “arouse,” “excitement,” “such,” “description,” “excluding”).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     Abyss
    -0.07
     vacant
    -0.07
     wreckage
    -0.06
     bliss
    -0.06
     expected
    -0.06
    ircuit
    -0.06
     gone
    -0.06
    obs
    -0.06
    _build
    -0.06
    Injection
    -0.06
    POSITIVE LOGITS
     ANT
    0.07
    %=
    0.06
     kuş
    0.06
    เค
    0.06
     زیبا
    0.06
    工
    0.06
     кора
    0.06
    ="")↵
    0.06
    United
    0.06
     başlayan
    0.06
    Act Density 0.012%

    No Known Activations