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

    Neuronpedia

    INDEX
    Explanations

    Math/physics problems

    np_max-act · gemini-2.0-flash

    The neuron detects instructional and operation words in math problem statements and solutions (e.g. verbs like “find,” “calculate,” “need,” and related function words around totals, numbers, and steps).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     pancakes
    -0.07
     Lon
    -0.07
     remix
    -0.07
     Values
    -0.06
     ingredients
    -0.06
    -stock
    -0.06
    _char
    -0.06
    ationToken
    -0.06
    redits
    -0.06
     Indicates
    -0.06
    POSITIVE LOGITS
    \D
    0.07
     rooft
    0.06
    +='<
    0.06
    ボ
    0.06
     negot
    0.06
     다양
    0.06
    %%%%%%%%%%%%%%%%
    0.06
    砲
    0.06
     соблюд
    0.06
    协
    0.06
    Act Density 0.040%

    No Known Activations