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 fires on code tokens that are names of types or classes—i.e. capitalized identifiers (like Pet, Module, Config, Impl) in source code.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    Songs
    -0.07
    '?
    -0.07
     adultery
    -0.07
    	fmt
    -0.06
     '$
    -0.06
    ()
    -0.06
     layered
    -0.06
     Eph
    -0.06
     Value
    -0.06
     Kan
    -0.06
    POSITIVE LOGITS
     буде
    0.08
    しており
    0.07
     tháng
    0.07
     없음
    0.07
     erh
    0.06
     را
    0.06
    ẵn
    0.06
     опера
    0.06
     تغییر
    0.06
     개인
    0.06
    Act Density 0.327%

    No Known Activations