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

    Neuronpedia

    INDEX
    Explanations

    code/math context

    np_max-act · gemini-2.0-flash

    This neuron fires on LaTeX‐style mathematical notation—Greek symbols (e.g. ε), integrals, partial‐derivative operators, and other formula tokens.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    ume
    -0.07
     initiative
    -0.06
    770
    -0.06
    Để
    -0.06
    عار
    -0.06
    ighth
    -0.06
     Wizards
    -0.06
    uuid
    -0.06
     tiêu
    -0.06
    ayar
    -0.06
    POSITIVE LOGITS
    рещ
    0.06
    LAS
    0.06
     कन
    0.06
     edin
    0.06
    擦
    0.06
    anti
    0.06
    createElement
    0.06
    .median
    0.06
     nob
    0.05
    oved
    0.05
    Act Density 0.142%

    No Known Activations