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

    Neuronpedia

    INDEX
    Explanations

    math symbols

    np_max-act · gemini-2.0-flash

    This neuron fires on numeric tokens—i.e. it detects digits and numbers in the text.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     Threshold
    -0.06
    IMPLEMENT
    -0.06
    appropri
    -0.06
    (cnt
    -0.06
    	Register
    -0.06
    Reviewed
    -0.06
    errMsg
    -0.06
     YEAR
    -0.06
     BLUE
    -0.06
    parer
    -0.06
    POSITIVE LOGITS
    ляються
    0.07
    alarda
    0.07
    _mob
    0.07
    иной
    0.06
    버지
    0.06
    σμα
    0.06
    тим
    0.06
     Bağ
    0.06
    undaki
    0.06
     снова
    0.06
    Act Density 0.002%

    No Known Activations