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

    Neuronpedia

    INDEX
    Explanations

    Legal/Formal Documents

    np_max-act · gemini-2.0-flash

    This neuron activates on formatting‐requirement keywords in the system instructions—terms like “with,” “proper,” and “and” that specify how the answer should be structured.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     math
    -0.07
    _positions
    -0.06
    madan
    -0.06
    nesia
    -0.06
    Titles
    -0.06
     turmoil
    -0.06
    san
    -0.06
    |:
    -0.06
    notifications
    -0.06
    Euro
    -0.06
    POSITIVE LOGITS
     hete
    0.07
     użytk
    0.06
    azar
    0.06
    lıyor
    0.06
     اس
    0.06
    ViewChild
    0.06
     INSTALL
    0.06
    rij
    0.06
     Lifecycle
    0.06
     NSStringFromClass
    0.06
    Act Density 0.003%

    No Known Activations