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

    Neuronpedia

    INDEX
    Explanations

    CSS flexbox

    np_max-act · gemini-2.0-flash

    The neuron strongly activates on the literal “<” and “>” tokens that mark HTML/XML tags, i.e. it’s looking for HTML‐style tag delimiters.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     vit
    -0.07
     awake
    -0.07
     towns
    -0.07
     bước
    -0.07
     bring
    -0.07
    кими
    -0.06
     sein
    -0.06
    WebKit
    -0.06
    -submit
    -0.06
    _Resource
    -0.06
    POSITIVE LOGITS
    anded
    0.07
    .username
    0.07
    /js
    0.07
     корист
    0.07
    押
    0.07
     rozvoj
    0.06
    mandatory
    0.06
     зависим
    0.06
    eced
    0.06
    $(".
    0.06
    Act Density 0.009%

    No Known Activations