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

    Neuronpedia

    INDEX
    Explanations

    grounded

    np_max-act · gemini-2.0-flash

    The neuron activates on proper‐name style suggestions (like “Zephyr,” “Lumina,” “Apex,” etc.), i.e. it spots proposed project or product names.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    위원회
    -0.06
     ancestors
    -0.06
    iations
    -0.06
     Winchester
    -0.06
     Cross
    -0.06
    거나
    -0.06
    (infile
    -0.06
    (import
    -0.06
    ーデ
    -0.06
     viet
    -0.06
    POSITIVE LOGITS
     Chris
    0.07
     poem
    0.07
    REN
    0.07
    divider
    0.06
     deja
    0.06
     sắc
    0.06
    ố
    0.06
     mRNA
    0.06
    losure
    0.06
     ray
    0.06
    Act Density 0.367%

    No Known Activations