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

    Neuronpedia

    INDEX
    Explanations

    include

    np_max-act · gemini-2.0-flash

    This neuron detects legal text specifying racial eligibility or classification (e.g. references to “white persons” in statutes or rulings).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    WebDriver
    -0.07
     Family
    -0.07
    füg
    -0.06
    унд
    -0.06
     investor
    -0.06
    	add
    -0.06
    ieten
    -0.06
    Seattle
    -0.06
    Honda
    -0.06
    参照
    -0.06
    POSITIVE LOGITS
     proving
    0.06
    vlan
    0.06
    /String
    0.06
    _##
    0.06
     vara
    0.06
    ธ
    0.06
     preserve
    0.06
     lifts
    0.06
     discrete
    0.06
     FF
    0.06
    Act Density 0.029%

    No Known Activations