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 serialization‐related code constructs (e.g. Data(Input|Output)(View|Stream), serialize/write/read methods, IOException), i.e. it detects serialization/deserialization routines.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    .setTimeout
    -0.08
    ющую
    -0.07
    ывает
    -0.07
     BrowserModule
    -0.07
     vanish
    -0.06
    thread
    -0.06
    一切
    -0.06
    -version
    -0.06
    STD
    -0.06
    Protect
    -0.06
    POSITIVE LOGITS
    erguson
    0.07
     Petra
    0.07
    brook
    0.06
     Ferguson
    0.06
     nominated
    0.06
     Ple
    0.06
    +',
    0.06
    овор
    0.06
     recruiter
    0.06
    \M
    0.06
    Act Density 0.011%

    No Known Activations