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 consistently activates on numeric literal tokens in code.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    alysis
    -0.07
    >("
    -0.06
     jednání
    -0.06
     girlfriend
    -0.06
     Disease
    -0.06
    Between
    -0.06
     brief
    -0.06
     τους
    -0.06
     rend
    -0.06
    wait
    -0.06
    POSITIVE LOGITS
    uest
    0.07
    ledge
    0.07
     ApiController
    0.07
    asher
    0.06
     (@
    0.06
     مه
    0.06
     तम
    0.06
    식
    0.06
    <=$
    0.06
    认识
    0.06
    Act Density 0.017%

    No Known Activations