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

    Neuronpedia

    INDEX
    Explanations

    various fields

    np_max-act · gemini-2.0-flash

    The neuron strongly activates on list–item numerals and ordinal markers (e.g. “1.”, “2.”, “third-party”), effectively identifying numbered or ordered list entries in the text.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    .functional
    -0.06
     doll
    -0.06
     indoors
    -0.06
    >Title
    -0.06
     �
    -0.06
    INDOW
    -0.06
    ={
    -0.06
    doll
    -0.06
     blev
    -0.06
    -tree
    -0.06
    POSITIVE LOGITS
     Christoph
    0.07
    hões
    0.07
    0.06
     clan
    0.06
     gösterir
    0.06
    ़ें
    0.06
    γνω
    0.06
     @{↵
    0.06
    _YES
    0.06
     CRS
    0.06
    Act Density 0.799%

    No Known Activations