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

    Neuronpedia

    INDEX
    Explanations

    stealth

    np_max-act · gemini-2.0-flash

    The neuron responds to descriptions of covert combat or infiltration roles—i.e. text about clandestine, stealthy professions or actions.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     magnet
    -0.06
     antiqu
    -0.06
    (userID
    -0.06
    cete
    -0.06
    姉
    -0.06
     klas
    -0.06
    serir
    -0.06
    Income
    -0.06
     induced
    -0.06
     окра
    -0.06
    POSITIVE LOGITS
     stealth
    0.12
     Stealth
    0.11
     Cam
    0.07
    .low
    0.07
    ////
    0.07
     Cool
    0.07
    itsu
    0.07
    uest
    0.06
    ATTLE
    0.06
     surprise
    0.06
    Act Density 0.004%

    No Known Activations