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

    Neuronpedia

    INDEX
    Explanations

    200

    np_max-act · gemini-2.0-flash

    The neuron selectively activates on numerical tokens—years, docket and statute numbers, and similar legal citation identifiers.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     segue
    -0.07
    �
    -0.06
    思考
    -0.06
     nicotine
    -0.06
     Alle
    -0.06
    ﾟ
    -0.06
    asmine
    -0.06
    ايل
    -0.06
     Rol
    -0.06
    Recursive
    -0.06
    POSITIVE LOGITS
    xac
    0.07
    haus
    0.06
     alım
    0.06
     committing
    0.06
     conscious
    0.06
     adec
    0.06
     спас
    0.06
    ulen
    0.06
    mins
    0.06
     envis
    0.06
    Act Density 0.003%

    No Known Activations