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

    Neuronpedia

    INDEX
    Explanations

    punctuation

    np_max-act · gemini-2.0-flash

    The neuron fires on tokens that serve as structural labels or markers—things like list bullets (“–”), ordered‐list numbers or letters with periods (“1.”, “a.”, “I.”), and heading/subtitle labels (“Title,” “Subtitle”)—i.e. outline or section identifiers.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     وص
    -0.06
     derived
    -0.06
    .Time
    -0.06
     професій
    -0.06
    Frozen
    -0.06
     onCreate
    -0.06
     Installed
    -0.06
     регуляр
    -0.06
    為
    -0.06
     Litecoin
    -0.06
    POSITIVE LOGITS
    STACK
    0.07
     entrusted
    0.06
    emin
    0.06
    条
    0.06
     herd
    0.06
     filesystem
    0.06
    <Integer
    0.06
    starttime
    0.06
     glEnd
    0.06
     assortment
    0.06
    Act Density 0.028%

    No Known Activations