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

    Neuronpedia

    INDEX
    Explanations

    Code

    np_max-act · gemini-2.0-flash

    The neuron fires on HTML/XML markup tokens—especially “<div>” elements (and their accompanying class/type attributes).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    Hur
    -0.07
    !\
    -0.07
    ۸
    -0.07
     Hur
    -0.07
    ("
    -0.07
    [C
    -0.06
     Esper
    -0.06
    AV
    -0.06
    ernen
    -0.06
    ­
    -0.06
    POSITIVE LOGITS
     mı
    0.07
     css
    0.07
     mask
    0.07
     labeling
    0.07
     empower
    0.07
    cls
    0.07
    ]';↵
    0.07
    (class
    0.06
    "])↵
    0.06
    ')))↵
    0.06
    Act Density 0.018%

    No Known Activations