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

    Neuronpedia

    INDEX
    Explanations

    period

    np_max-act · gemini-2.0-flash

    This neuron fires on section‐opening or heading–style lines—that is, titles, abstracts, and the first sentence of major document sections.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    �
    -0.06
    -count
    -0.06
    ون
    -0.06
    �
    -0.06
    44
    -0.06
     are
    -0.06
    _CI
    -0.06
     Vanderbilt
    -0.06
    ностей
    -0.06
    voří
    -0.06
    POSITIVE LOGITS
     ген
    0.07
     国家
    0.07
    _byte
    0.07
     Pyramid
    0.06
     слож
    0.06
    interpre
    0.06
    .transitions
    0.06
    坐
    0.06
     appended
    0.06
    láda
    0.06
    Act Density 0.214%

    No Known Activations