© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 13070
    Prev
    Next
    INDEX
    Explanations

    html attribute access

    np_acts-logits-general · gemini-2.5-flash-lite

    The neuron is triggered by HTML (or DOM) attribute names (e.g. href, class, id, data-…), i.e. tokens that name element attributes.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     who
    -1.40
     the
    -1.15
     được
    -1.14
     up
    -1.10
     reached
    -1.09
     reaching
    -1.09
     duża
    -1.08
     Ortschaft
    -1.07
    ↵↵↵
    -1.06
    ro
    -1.05
    POSITIVE LOGITS
    Lá
    1.55
     pandemi
    1.52
    Hilsen
    1.49
    erobic
    1.39
    Kval
    1.38
    Ещё
    1.38
     diki
    1.37
    不可思议
    1.37
     maksi
    1.36
     schuhe
    1.35
    Activations Density 0.002%

    No Known Activations