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

    Neuronpedia

    INDEX
    Explanations

    set add

    np_max-act · gemini-2.0-flash

    This neuron detects and highlights code blocks or fenced code snippets in the conversation.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     outer
    -0.07
    ("/")↵
    -0.07
     Minutes
    -0.06
     yas
    -0.06
     remained
    -0.06
     humming
    -0.06
     f
    -0.06
     spurred
    -0.06
    оступ
    -0.06
    OG
    -0.06
    POSITIVE LOGITS
    هنگ
    0.07
    -div
    0.07
    RGB
    0.06
    �
    0.06
     Schwartz
    0.06
     overwhel
    0.06
    <Transform
    0.06
     Deutschland
    0.06
     JsonRequestBehavior
    0.06
    �
    0.06
    Act Density 0.084%

    No Known Activations