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 floating‐point numeric literals (decimal constants) in the code.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    versation
    -0.07
     coach
    -0.07
    बर
    -0.07
    بود
    -0.07
    ARE
    -0.07
     object
    -0.07
    Fixed
    -0.07
     baskets
    -0.07
     FG
    -0.07
     heel
    -0.07
    POSITIVE LOGITS
     переб
    0.06
     بحث
    0.06
     Natasha
    0.06
     stricter
    0.06
     потреб
    0.06
    styl
    0.06
     nichts
    0.06
     merkezi
    0.06
     gehen
    0.06
     Sergei
    0.06
    Act Density 0.012%

    No Known Activations