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

    Neuronpedia

    INDEX
    Explanations

    conditions

    np_max-act · gemini-2.0-flash

    The neuron detects conditional/hypothetical triggers (e.g. “if,” “when”) that introduce “what happens” style questions.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    يا
    -0.06
     вел
    -0.06
    ">↵↵↵
    -0.06
    gnore
    -0.06
    冬
    -0.06
    296
    -0.06
     tuyệt
    -0.06
     궁
    -0.06
    ERSIST
    -0.06
    さら
    -0.06
    POSITIVE LOGITS
     Moses
    0.08
     pr
    0.07
    (acc
    0.06
     shortest
    0.06
     امروز
    0.06
     Received
    0.06
     Moder
    0.06
    [Z
    0.06
     Yesterday
    0.06
    "/>
    0.06
    Act Density 0.033%

    No Known Activations