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

    Neuronpedia

    INDEX
    Explanations

    lookahead

    np_max-act · gemini-2.0-flash

    The neuron fires on the non‐literal, metasyntactic parts of regular expressions—especially lookaround assertions and boundary markers (e.g. (?=…), (?<=…), (?!…), \b, etc.).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     Calculation
    -0.07
     дет
    -0.06
    .next
    -0.06
    Deserializer
    -0.06
     EFF
    -0.06
    .startswith
    -0.06
     Teeth
    -0.06
     compound
    -0.06
     STDMETHOD
    -0.06
    Semaphore
    -0.06
    POSITIVE LOGITS
    );\
    0.07
     ngại
    0.07
    acılık
    0.06
    Ω
    0.06
    vido
    0.06
    cki
    0.06
     omdat
    0.06
     обеспе
    0.06
     geo
    0.06
     Silent
    0.06
    Act Density 0.002%

    No Known Activations