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 mainly lights up on numeric literals (digits and numbers) appearing in code snippets.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     pension
    -0.07
     Bu
    -0.07
    BED
    -0.07
    Bu
    -0.06
     immersion
    -0.06
     copyright
    -0.06
    ію
    -0.06
     jeep
    -0.06
     governance
    -0.06
    рест
    -0.06
    POSITIVE LOGITS
    /slick
    0.09
    .componentInstance
    0.08
     aussi
    0.07
     Sms
    0.06
    .tencent
    0.06
    ैं.↵
    0.06
    ينات
    0.06
    .offsetTop
    0.06
     ")");↵
    0.06
    кат
    0.06
    Act Density 0.114%

    No Known Activations