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 activates on occurrences of the word “numbers,” especially in code contexts (e.g., variable or function names referencing “numbers”).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     jade
    -0.06
    .One
    -0.06
    Magic
    -0.06
    .Black
    -0.06
    全
    -0.06
    	    ↵
    -0.06
    policy
    -0.06
     ////
    -0.06
     prefect
    -0.06
    /n
    -0.06
    POSITIVE LOGITS
     εισ
    0.07
    igi
    0.07
    Feb
    0.06
     hdc
    0.06
     vaping
    0.06
    тр
    0.06
     predecessor
    0.06
     Fraction
    0.06
    >'.
    0.06
     Pri
    0.06
    Act Density 0.016%

    No Known Activations