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

    Neuronpedia

    INDEX
    Explanations

    God

    np_max-act · gemini-2.0-flash

    This neuron activates on references to God’s classical attributes—especially phrases like “all-powerful,” “all-knowing,” and “benevolent” (and related divine descriptors).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     Mp
    -0.06
    stoi
    -0.06
    	fwrite
    -0.06
     cigar
    -0.06
    .cons
    -0.06
     chaud
    -0.06
    _tag
    -0.06
    ะ
    -0.06
     samen
    -0.05
    errar
    -0.05
    POSITIVE LOGITS
    AZ
    0.07
    Normalization
    0.07
     ú
    0.07
     warehouse
    0.06
     ارتفاع
    0.06
     tus
    0.06
     Military
    0.06
    _Enter
    0.06
     Thánh
    0.06
    بن
    0.06
    Act Density 0.014%

    No Known Activations