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

    Neuronpedia

    INDEX
    Explanations

    math equations

    np_max-act · gemini-2.0-flash

    The neuron responds to sentence‐ or turn‐boundary markers—e.g. periods and end-of-text/header tokens—i.e. it’s detecting punctuation or special delimiters that end segments.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     صف
    -0.07
     homicides
    -0.07
    инки
    -0.06
    他の
    -0.06
    映画
    -0.06
    Examples
    -0.06
     Users
    -0.06
     آباد
    -0.06
    Connected
    -0.06
    üfus
    -0.06
    POSITIVE LOGITS
     vel
    0.07
     deeper
    0.07
    setFlash
    0.07
     impunity
    0.06
     stalled
    0.06
     klar
    0.06
     larger
    0.06
     كتاب
    0.06
    .bpm
    0.06
     στη
    0.06
    Act Density 0.039%

    No Known Activations