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

    Neuronpedia

    INDEX
    Explanations

    :

    np_max-act · gemini-2.0-flash

    The neuron detects administrative and legal boilerplate—things like registration numbers, VAT identifiers, “Company Reg No,” “Registered office,” and similar regulatory or disclaimer metadata.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     screwed
    -0.07
    이비
    -0.06
    lew
    -0.06
    ปร
    -0.06
     Năm
    -0.06
    wj
    -0.06
    _PLACE
    -0.06
     حکم
    -0.06
     Atari
    -0.06
    fast
    -0.06
    POSITIVE LOGITS
    AndHashCode
    0.07
     latex
    0.07
     Lag
    0.06
    >'.
    0.06
    _lstm
    0.06
    _End
    0.06
    _INTERFACE
    0.06
     trump
    0.06
    (loss
    0.06
    jam
    0.06
    Act Density 0.055%

    No Known Activations