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

    Neuronpedia

    INDEX
    Explanations

    baby products

    np_max-act · gemini-2.0-flash

    The neuron activates on individual baby‐related product or list‐item nouns (e.g. “baby,” “clothes,” “wipes,” “sponge,” “swaddles,” etc.).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     device
    -0.07
    隐
    -0.07
    opacity
    -0.07
    _buf
    -0.07
     detainees
    -0.07
     Manager
    -0.07
     Siege
    -0.07
     Payload
    -0.07
     tub
    -0.07
     RK
    -0.06
    POSITIVE LOGITS
    ाए
    0.07
     человеч
    0.06
    .SaveChanges
    0.06
    Kent
    0.06
    ฝ
    0.06
    اهر
    0.06
    看看
    0.06
     آغاز
    0.06
    ินท
    0.06
    cks
    0.06
    Act Density 0.021%

    No Known Activations