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

    Neuronpedia

    INDEX
    Explanations

    Code

    np_max-act · gemini-2.0-flash

    The neuron is triggered by the “…” ellipsis token (i.e. sequences of three dots) that mark omitted or continued text.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     IQueryable
    -0.07
     jsonData
    -0.06
     ettik
    -0.06
    _reviews
    -0.06
    พวก
    -0.06
    محمد
    -0.06
    рус
    -0.06
     Roger
    -0.06
    ायल
    -0.06
     descriptive
    -0.06
    POSITIVE LOGITS
     boa
    0.08
    -working
    0.07
     Bill
    0.07
     START
    0.06
     latch
    0.06
     guy
    0.06
     sala
    0.06
    -gr
    0.06
    _bill
    0.06
     backyard
    0.06
    Act Density 0.001%

    No Known Activations