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

    Neuronpedia

    INDEX
    Explanations

    algebra

    np_max-act · gemini-2.0-flash

    The neuron activates on names of mathematical subjects or topics (e.g. “calculus,” “linear algebra,” “discrete mathematics,” etc.).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    отов
    -0.07
    .magnitude
    -0.07
    ену
    -0.07
    цер
    -0.07
     neglect
    -0.07
    ันทร
    -0.07
     criticism
    -0.07
     pareja
    -0.06
    滿
    -0.06
    _SELECTION
    -0.06
    POSITIVE LOGITS
     */
↵
    0.07
    .linalg
    0.07
    ,,
    0.06
    	DECLARE
    0.06
    	glog
    0.06
     prime
    0.06
    _LED
    0.06
    (express
    0.06
     InvalidOperationException
    0.06
     Forty
    0.06
    Act Density 0.004%

    No Known Activations