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

    Neuronpedia

    INDEX
    Explanations

    code

    np_max-act · gemini-2.0-flash

    function definitions and method signatures in programming code.

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

    This neuron activates on programming-style syntax—tokens like semicolons, braces, method calls, and other code punctuation—so it’s detecting code fragments.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     исслед
    -0.07
    (keys
    -0.06
     penc
    -0.06
     věku
    -0.06
    .sharedInstance
    -0.06
     дв
    -0.06
     نه
    -0.06
     основе
    -0.06
     Χα
    -0.06
     serotonin
    -0.06
    POSITIVE LOGITS
     maneuver
    0.07
    onus
    0.07
    �
    0.07
     transactions
    0.07
    aza
    0.07
     ＼↵
    0.07
     Suarez
    0.06
    iso
    0.06
    AZ
    0.06
    ）。↵↵
    0.06
    Act Density 0.043%

    No Known Activations