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

    Neuronpedia

    INDEX
    Explanations

    variable types

    np_max-act · gemini-2.0-flash

    The neuron is chiefly activated by numeric literals (integer or floating-point constants) in the code.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     questionnaire
    -0.07
    meniz
    -0.07
     weaving
    -0.06
     Pa
    -0.06
     acet
    -0.06
    buy
    -0.06
    ीटर
    -0.06
    .Owner
    -0.06
     Millennium
    -0.06
     exon
    -0.06
    POSITIVE LOGITS
     committed
    0.07
    .getInt
    0.07
     rank
    0.07
    �
    0.07
    Scoped
    0.07
    テ
    0.06
    мів
    0.06
    �
    0.06
    _planes
    0.06
    缓
    0.06
    Act Density 0.002%

    No Known Activations