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

    Neuronpedia

    INDEX
    Explanations

    abstract

    np_max-act · gemini-2.0-flash

    This neuron detects the “abstract” section heading at the start of scholarly documents.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     dokonce
    -0.07
     yeşil
    -0.06
     Rows
    -0.06
     المح
    -0.06
    航
    -0.06
     mañana
    -0.06
    δή
    -0.06
    予
    -0.06
    Uuid
    -0.06
     والع
    -0.06
    POSITIVE LOGITS
     situations
    0.07
    (adapter
    0.06
    .clicked
    0.06
     seriously
    0.06
     mechanisms
    0.06
    aş
    0.06
     cars
    0.06
    nan
    0.06
     Overnight
    0.06
     troublesome
    0.06
    Act Density 0.002%

    No Known Activations