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

    Neuronpedia

    INDEX
    Explanations

    code punctuations

    np_max-act · gemini-2.0-flash

    The neuron is detecting numeric literals—especially version numbers and decimal values—in code or config files.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     Ung
    -0.07
    JC
    -0.07
    .average
    -0.07
     wall
    -0.07
    _DE
    -0.06
     mostr
    -0.06
    anooga
    -0.06
     codes
    -0.06
     whitelist
    -0.06
     Because
    -0.06
    POSITIVE LOGITS
    общ
    0.07
    formData
    0.07
    className
    0.07
     emotionally
    0.07
    $conn
    0.06
    $id
    0.06
     dost
    0.06
    997
    0.06
    	TEST
    0.06
    []=$
    0.06
    Act Density 0.034%

    No Known Activations