© 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. 84428
    Prev
    Next
    INDEX
    Explanations

    reading and writing XML-like structures

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

    The neuron fires on code‐style identifiers and tag names (e.g. XML or programming keywords) rather than ordinary prose.

    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
    を
    -1.18
    は
    -1.16
    aneity
    -1.09
    则是
    -1.05
     haup
    -1.04
    が
    -1.02
    の
    -0.99
    である
    -0.98
     will
    -0.91
    应该是
    -0.91
    POSITIVE LOGITS
    ROGEN
    1.14
     Dě
    1.03
    читать
    1.02
    clearTimeout
    1.01
     ]);
    1.00
    xtick
    1.00
    "]));
    0.99
    INUM
    0.97
    RTRS
    0.97
    mbda
    0.96
    Activations Density 0.043%

    No Known Activations