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

    Neuronpedia

    INDEX
    Explanations

    scams and unsuspecting victims

    np_max-act · gemini-2.0-flash

    The neuron activates on language describing or instructing fraudulent scams—particularly schemes targeting unsuspecting victims.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    _refer
    -0.06
     Phot
    -0.06
     dispon
    -0.06
     pornography
    -0.06
     finger
    -0.06
    -Feb
    -0.06
    führ
    -0.06
     вариант
    -0.06
     Viện
    -0.06
    .rooms
    -0.06
    POSITIVE LOGITS
     Shares
    0.06
     Community
    0.06
     ตำบล
    0.06
     Vì
    0.06
     survivors
    0.06
    	Vector
    0.06
    عب
    0.06
     incentiv
    0.06
    /github
    0.05
    ve
    0.05
    Act Density 0.024%

    No Known Activations