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

    Neuronpedia

    INDEX
    Explanations

    code

    np_max-act · gemini-2.0-flash

    The neuron activates on HTML element markup—essentially it’s spotting opening tags (the “<…>” constructs) in HTML.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    bac
    -0.07
     udrž
    -0.06
     Café
    -0.06
     типа
    -0.06
     سو
    -0.06
     nécessaire
    -0.06
    пеки
    -0.06
    _equ
    -0.06
     покол
    -0.06
     ощущ
    -0.06
    POSITIVE LOGITS
    _examples
    0.07
     getInfo
    0.07
     prizes
    0.07
    Colors
    0.06
     verilm
    0.06
    ากร
    0.06
    ?↵
    0.06
     citing
    0.06
    uluğ
    0.06
    `}↵
    0.06
    Act Density 0.040%

    No Known Activations