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

    Neuronpedia

    INDEX
    Explanations

    XML-like code

    np_max-act · gemini-2.0-flash

    This neuron detects XML‐style markup tags and their component tokens (e.g. folder, filename, source, annotation, etc.).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    _tw
    -0.07
     рабо
    -0.06
    .getRoot
    -0.06
     नगर
    -0.06
    .returnValue
    -0.06
     nad
    -0.06
    [Index
    -0.06
     oğlu
    -0.06
     crews
    -0.06
    break
    -0.06
    POSITIVE LOGITS
    -L
    0.07
    alıdır
    0.07
    ΥΝ
    0.07
    ?;↵↵
    0.07
     enticing
    0.07
    ottenham
    0.06
     gerç
    0.06
     फ
    0.06
    "){
    0.06
     strange
    0.06
    Act Density 0.003%

    No Known Activations