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

    Neuronpedia

    INDEX
    Explanations

    topic subject

    np_max-act · gemini-2.0-flash

    The neuron selectively activates on instances of the word “topics,” particularly when the assistant lists or discusses the range of topics it can handle.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    ‌تواند
    -0.07
     komb
    -0.07
    spb
    -0.06
    .getBlock
    -0.06
     courtyard
    -0.06
    kode
    -0.06
     Champions
    -0.06
     scape
    -0.06
    :first
    -0.06
    fight
    -0.06
    POSITIVE LOGITS
     karşı
    0.07
    219
    0.07
     bitter
    0.06
    252
    0.06
    Projected
    0.06
    loss
    0.06
    [out
    0.06
     kapit
    0.06
     zx
    0.06
    uku
    0.06
    Act Density 0.019%

    No Known Activations