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

    Neuronpedia

    INDEX
    Explanations

    sources

    np_max-act · gemini-2.0-flash

    This neuron detects mentions of “sources” (i.e. references to information sources).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     Taliban
    -0.08
    Sale
    -0.07
    (tab
    -0.07
     Pamela
    -0.07
    uye
    -0.07
     polygons
    -0.06
    ७
    -0.06
    ruba
    -0.06
     Parkway
    -0.06
     설
    -0.06
    POSITIVE LOGITS
     sources
    0.10
     spies
    0.08
     source
    0.08
    Sources
    0.06
     oleh
    0.06
     outlets
    0.06
     officials
    0.06
     PLATFORM
    0.06
    ears
    0.06
     resources
    0.06
    Act Density 0.013%

    No Known Activations