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

    Neuronpedia

    INDEX
    Explanations

    real estate

    np_max-act · gemini-2.0-flash

    The neuron is keyed to the markers and first words of numbered “step‐by‐step” instructions (e.g. list numbers like “1.”, “2.” and the words that immediately follow).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     Woody
    -0.06
    当
    -0.06
     대통령
    -0.06
     Od
    -0.06
    oom
    -0.06
    従
    -0.06
     Ka
    -0.06
    Agents
    -0.06
     cunning
    -0.06
     Nate
    -0.06
    POSITIVE LOGITS
     пример
    0.07
     payer
    0.07
     Jahren
    0.06
     procent
    0.06
    /native
    0.06
    "),"
    0.06
    -prev
    0.06
     annotations
    0.06
    ünden
    0.06
     Sonuç
    0.06
    Act Density 0.055%

    No Known Activations