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

    Neuronpedia

    INDEX
    Explanations

    code

    np_max-act · gemini-2.0-flash

    This neuron responds to code comparison expressions, i.e. the “if (… == …)” pattern in source code.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     April
    -0.07
     recursion
    -0.07
    April
    -0.07
     VIDEO
    -0.07
     trailed
    -0.07
    Reached
    -0.07
    _socket
    -0.06
    プ
    -0.06
    questions
    -0.06
    Game
    -0.06
    POSITIVE LOGITS
    _status
    0.06
     cortex
    0.06
    Jets
    0.06
     حم
    0.06
     효
    0.06
    บรร
    0.06
     впол
    0.06
    -operator
    0.06
    _subtype
    0.06
    人人
    0.06
    Act Density 0.192%

    No Known Activations