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

    Neuronpedia

    INDEX
    Explanations

    _instance

    np_max-act · gemini-2.0-flash

    This neuron activates on code comments or annotations that specify formal verification conditions (e.g. “invariant”, “requires”, “ensures”, and paired “open”/“close” proof or predicate blocks).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     投稿日
    -0.06
     Dispose
    -0.06
     beverage
    -0.06
    withErrors
    -0.06
    /buttons
    -0.06
     "/",
    -0.06
     资
    -0.06
     Dead
    -0.06
    лика
    -0.06
    alborg
    -0.06
    POSITIVE LOGITS
    }`)↵
    0.07
     huge
    0.07
    .randrange
    0.06
    >');↵
    0.06
     někter
    0.06
    '),↵
    0.06
    '})↵
    0.06
    PROGRAM
    0.06
    dots
    0.06
    attachment
    0.06
    Act Density 0.046%

    No Known Activations