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

    Neuronpedia

    INDEX
    Explanations

    customer service

    np_max-act · gemini-2.0-flash

    The neuron activates on short procedural or transactional statements that give customers instructions or details about how to contact, pay, ship, or access a service (e.g. availability times, payment methods, shipping costs, or contact channels).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    .pt
    -0.07
    .par
    -0.07
    !",
    -0.07
    :,
    -0.07
    "),
    -0.06
     міжнарод
    -0.06
    aval
    -0.06
     rahat
    -0.06
    -breaking
    -0.06
     Takes
    -0.06
    POSITIVE LOGITS
    	app
    0.07
    озя
    0.06
     craftsmanship
    0.06
     Kingston
    0.06
     plast
    0.06
    正在
    0.06
    (power
    0.06
    subnet
    0.06
     věci
    0.06
     neob
    0.06
    Act Density 0.076%

    No Known Activations