© 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. 10-GEMMASCOPE-RES-131K
    7. 12755
    Prev
    Next
    INDEX
    Explanations

    XML file processing

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

    The neuron activates on occurrences of “xml” (and other markup‐language tokens like HTML or DOM) in code or document text.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
     colorear
    -2.25
     It
    -1.93
     almacenar
    -1.86
     liberar
    -1.66
     fijar
    -1.62
     fabricar
    -1.61
     efectuar
    -1.59
    妄
    -1.57
    mband
    -1.55
     seleccionar
    -1.55
    POSITIVE LOGITS
     chwili
    2.14
     takim
    1.92
     び
    1.89
    </u>
    1.88
     אחרים
    1.88
     enemigo
    1.82
    \
    1.82
    덫
    1.81
     guerrero
    1.79
    蛧
    1.79
    Activations Density 0.058%

    No Known Activations