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

    Neuronpedia

    INDEX
    Explanations

    Affirmation

    np_max-act · gemini-2.0-flash

    The neuron reliably lights up on grammar‐check commentary—tokens around evaluating or correcting sentence grammar (e.g. “sentence,” “grammatically,” “correct,” “improve,” “revision”).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    (digits
    -0.07
    _keeper
    -0.06
    经过
    -0.06
    ords
    -0.06
     oppos
    -0.06
    ITTE
    -0.06
    θούν
    -0.06
    Js
    -0.06
     spent
    -0.06
    위원
    -0.06
    POSITIVE LOGITS
     testName
    0.07
    �
    0.06
     Mourinho
    0.06
    대로
    0.06
     clearInterval
    0.06
    илось
    0.06
    _reply
    0.06
    >\<^
    0.06
     automáticamente
    0.06
     номер
    0.06
    Act Density 0.021%

    No Known Activations