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

    Neuronpedia

    INDEX
    Explanations

    </

    np_max-act · gemini-2.0-flash

    The neuron strongly activates on HTML markup tokens—particularly the angle‐bracket and slash sequences (e.g. “</”) and the tag‐name pieces—i.e. parts of HTML tags.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    SA
    -0.07
    /tree
    -0.07
     sewing
    -0.07
    Rating
    -0.07
    Per
    -0.07
    Lines
    -0.07
     ByteArray
    -0.07
    GraphQL
    -0.06
    SAMPLE
    -0.06
    Fake
    -0.06
    POSITIVE LOGITS
     Tai
    0.06
     малыш
    0.06
     Due
    0.06
     kvinn
    0.05
    �
    0.05
     bakış
    0.05
    ovaného
    0.05
     nominations
    0.05
    ："+
    0.05
     candidacy
    0.05
    Act Density 0.002%

    No Known Activations