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

    Neuronpedia

    INDEX
    Explanations

    tags and categories

    np_max-act · gemini-2.0-flash

    This neuron detects metadata labels or section headers (like “Tag:”, “Category:”, “Articles tagged”) in the text.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    .answers
    -0.07
    ipes
    -0.07
    _POINTS
    -0.06
    -Д
    -0.06
    бол
    -0.06
     Sof
    -0.06
    iping
    -0.06
     Kai
    -0.06
    롯
    -0.06
     sparse
    -0.06
    POSITIVE LOGITS
    cle
    0.07
    ?',↵
    0.07
    **
↵
    0.06
    *
↵
    0.06
    (editor
    0.06
     RuntimeMethod
    0.06
     допом
    0.06
    -member
    0.06
    とう
    0.06
    __.'/
    0.06
    Act Density 0.038%

    No Known Activations