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

    Neuronpedia

    INDEX
    Explanations

    Amazon URLs

    np_max-act · gemini-2.0-flash

    This neuron detects URL query components (e.g. “?”, “=”, parameter names like “ref” or “utm_medium”) within hyperlinks.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     calming
    -0.06
    .(
    -0.06
     lun
    -0.06
    .Json
    -0.06
    下载
    -0.06
     en
    -0.06
     educ
    -0.06
     scene
    -0.06
     Voor
    -0.06
    دث
    -0.06
    POSITIVE LOGITS
    -May
    0.06
    -test
    0.06
    >(&
    0.06
     withholding
    0.06
     getResult
    0.06
     للإ
    0.06
     Tại
    0.06
    itě
    0.06
     GE
    0.06
     performan
    0.06
    Act Density 0.005%

    No Known Activations