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

    Neuronpedia

    INDEX
    Explanations

    code

    np_max-act · gemini-2.0-flash

    The neuron consistently activates on numeric literals (both integers and floating‐point constants) in the code.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     Just
    -0.07
     LEGO
    -0.07
     Although
    -0.06
     bullying
    -0.06
     Gates
    -0.06
    ']>;↵
    -0.06
    Dll
    -0.06
    Parsing
    -0.06
     perform
    -0.06
     Woo
    -0.06
    POSITIVE LOGITS
    .constraints
    0.07
    )`
    0.06
     свое
    0.06
     migliori
    0.06
     unanimous
    0.06
    CREEN
    0.06
    (raw
    0.06
    สภ
    0.06
    /an
    0.06
    ificial
    0.06
    Act Density 0.067%

    No Known Activations