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

    Neuronpedia

    INDEX
    Explanations

    code snippets or database schemas

    np_max-act · gemini-2.0-flash

    SQL statements involving data selection and aggregation.

    oai_token-act-pair · gpt-4o-miniTriggered by @xinyanhu8

    The neuron activates on numeric literals and numeric‐looking tokens (e.g. default values, NULL tolerances, and column sizes) in schema definitions.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    -edge
    -0.07
    วาม
    -0.07
     गई
    -0.07
    大
    -0.06
     dominated
    -0.06
    盛
    -0.06
    -face
    -0.06
     ند
    -0.06
     immediate
    -0.06
    圈
    -0.06
    POSITIVE LOGITS
    "${
    0.07
    Okay
    0.06
    )","
    0.06
     firing
    0.06
     Lear
    0.06
    “你
    0.06
    >I
    0.06
    CLICK
    0.06
    ünk
    0.06
    #c
    0.06
    Act Density 0.003%

    No Known Activations