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

    Neuronpedia

    INDEX
    Explanations

    specification

    np_max-act · gemini-2.0-flash

    The neuron activates on boilerplate references to the “specification” (e.g. “this specification,” “present specification”) in patent-style text.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     Muscle
    -0.07
    1
    -0.06
     ihre
    -0.06
    usr
    -0.06
    =train
    -0.06
    _H
    -0.06
     here
    -0.06
     alır
    -0.06
    okens
    -0.06
     giữa
    -0.06
    POSITIVE LOGITS
    (startTime
    0.07
    _ACTIV
    0.06
    학년도
    0.06
    ATUS
    0.06
     설정
    0.06
    кат
    0.06
    .cod
    0.06
    Backdrop
    0.06
     تماس
    0.06
     intervening
    0.06
    Act Density 0.002%

    No Known Activations