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

    Neuronpedia

    INDEX
    Explanations

    utf unicode

    np_max-act · gemini-2.0-flash

    The neuron detects mentions of UTF-8 character encoding (and associated non-ASCII characters) in database or text handling contexts.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    ков
    -0.08
     //--------------------------------
    -0.07
    드를
    -0.07
    ortality
    -0.07
     Narrow
    -0.07
    nech
    -0.06
    тами
    -0.06
    _Project
    -0.06
    кою
    -0.06
    cstdio
    -0.06
    POSITIVE LOGITS
     generating
    0.07
    JT
    0.07
     activist
    0.07
     â
    0.06
    eguard
    0.06
    ред
    0.06
    (cb
    0.06
     Ads
    0.06
    )','
    0.06
    _create
    0.06
    Act Density 0.009%

    No Known Activations