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

    Neuronpedia

    INDEX
    Explanations

    code and memory addresses

    np_max-act · gemini-2.0-flash

    The neuron fires on hexadecimal numeric fragments (e.g. parts of memory addresses or offsets) in stack dumps.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     ¿
    -0.07
    sitemap
    -0.07
    Understanding
    -0.07
    455
    -0.07
    .exam
    -0.07
    spm
    -0.06
    91
    -0.06
     uniformly
    -0.06
    /latest
    -0.06
     pohled
    -0.06
    POSITIVE LOGITS
    .Footer
    0.06
    findViewById
    0.06
     organic
    0.06
     Organic
    0.06
     Statue
    0.06
     역
    0.06
     InputDecoration
    0.06
    ghest
    0.06
     oxidative
    0.06
     Brewers
    0.06
    Act Density 0.002%

    No Known Activations