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

    Neuronpedia

    INDEX
    Explanations

    Omega

    np_max-act · gemini-2.0-flash

    The neuron activates on chunks of mathematical notation and LaTeX‐style formulas (e.g. equations, symbols, variable declarations, domain definitions).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     pencils
    -0.06
    (batch
    -0.06
     Utt
    -0.06
    uctor
    -0.06
    言葉
    -0.06
     Penis
    -0.06
    الك
    -0.06
    ynth
    -0.06
    _DELAY
    -0.06
    wo
    -0.06
    POSITIVE LOGITS
     �
    0.07
     많이
    0.06
    International
    0.06
     ROUND
    0.06
     bringing
    0.06
    JOIN
    0.06
     роки
    0.06
    幸福
    0.06
    _DEFAULT
    0.06
    ная
    0.06
    Act Density 0.011%

    No Known Activations