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

    Neuronpedia

    INDEX
    Explanations

    -

    np_max-act · gemini-2.0-flash

    This neuron responds to structural or formatting punctuation in parse‐tree and document markup—e.g. parentheses, hyphens, and asterisks.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    ains
    -0.07
    AINS
    -0.06
     mind
    -0.06
     START
    -0.06
     Mac
    -0.06
     private
    -0.06
     OFF
    -0.06
    -val
    -0.06
    Servlet
    -0.06
    _MSB
    -0.06
    POSITIVE LOGITS
     Ryzen
    0.07
     anomalies
    0.07
    φέ
    0.07
     Carnegie
    0.07
     وغير
    0.07
     anomaly
    0.07
     التر
    0.07
    _verbose
    0.06
     disrupting
    0.06
    >G
    0.06
    Act Density 0.006%

    No Known Activations