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

    Neuronpedia

    INDEX
    Explanations

    greetings

    np_max-act · gemini-2.0-flash

    The neuron strongly activates on the words “are” and “you,” i.e. it detects instances of the phrase “are you.”

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     Рос
    -0.07
    Trou
    -0.07
    ">(
    -0.06
    íte
    -0.06
     session
    -0.06
    INUE
    -0.06
    .getSession
    -0.06
     borderSide
    -0.06
    783
    -0.06
     cables
    -0.06
    POSITIVE LOGITS
    斗
    0.07
     зем
    0.07
     Ribbon
    0.06
     διά
    0.06
    Hostname
    0.06
     спост
    0.06
    .Board
    0.06
    стоя
    0.06
    onestly
    0.06
     неб
    0.06
    Act Density 0.032%

    No Known Activations