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

    Neuronpedia

    INDEX
    Explanations

    #

    np_max-act · gemini-2.0-flash

    This neuron detects numerical citation markers and reference‐style numbers (e.g. bracketed or parenthesized citation indices and DOI/numeric identifiers).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     Taiwanese
    -0.07
    怎么
    -0.07
     Tour
    -0.07
     Thailand
    -0.06
     Leon
    -0.06
     dans
    -0.06
    Scott
    -0.06
     Я
    -0.06
    xBC
    -0.06
    双
    -0.06
    POSITIVE LOGITS
    imizer
    0.06
     [↵
    0.06
     sne
    0.06
    .NoError
    0.06
    _Ref
    0.06
     diffic
    0.06
    ajan
    0.06
    ordinal
    0.05
    iei
    0.05
    ीवन
    0.05
    Act Density 0.005%

    No Known Activations