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

    Neuronpedia

    INDEX
    Explanations

    News articles

    np_max-act · gemini-2.0-flash

    This neuron detects long runs of the same character (strongly activated by long repeated "a" sequences).

    oai_token-act-pair · gpt-5-miniTriggered by @vetterc0
    New Auto-Interp
    Negative Logits
     watts
    -0.07
     hammer
    -0.07
    .booking
    -0.07
     symbolic
    -0.07
     shirt
    -0.07
    $('#
    -0.07
     Apartment
    -0.07
    itable
    -0.07
    -command
    -0.06
    cats
    -0.06
    POSITIVE LOGITS
     เพ
    0.07
     tiến
    0.06
    Wel
    0.06
    建设工程
    0.06
    嫦
    0.06
    פוט
    0.06
     potrze
    0.06
    �
    0.06
    .cz
    0.06
     verschied
    0.06
    Act Density 1.718%

    No Known Activations