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 punctuation and single‐letter markers used in structured legal citations (e.g., the parentheses and subdivision letters in statutory references).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     Roll
    -0.07
     Acceler
    -0.07
    Market
    -0.07
     Mix
    -0.06
    Easy
    -0.06
    Clr
    -0.06
    Tank
    -0.06
     Ад
    -0.06
     Colour
    -0.06
    几
    -0.06
    POSITIVE LOGITS
    comment
    0.06
    _cur
    0.06
    acle
    0.06
    .loader
    0.06
    apse
    0.06
     afs
    0.06
    imiters
    0.06
    .finish
    0.06
     chce
    0.06
    steller
    0.06
    Act Density 0.010%

    No Known Activations