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

    Neuronpedia

    INDEX
    Explanations

    Code and descriptions

    np_max-act · gemini-2.0-flash

    The neuron responds to the names of JavaScript built-in methods, properties, and descriptor attributes (e.g. “Object.defineProperty,” “Array.prototype.map,” “[[Configurable]],” etc.) as they appear in test descriptions.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     RF
    -0.06
    湯
    -0.06
    （平成
    -0.06
    Contains
    -0.06
    <h
    -0.06
     mCurrent
    -0.06
    ایل
    -0.06
     Населення
    -0.06
     Vacc
    -0.06
     dav
    -0.06
    POSITIVE LOGITS
    ráci
    0.07
     Lesb
    0.06
     relating
    0.06
    roken
    0.06
    _miss
    0.06
    _AVAILABLE
    0.06
     excited
    0.06
    hani
    0.06
    씀
    0.06
     banco
    0.06
    Act Density 0.013%

    No Known Activations