© 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. 34-GEMMASCOPE-RES-131K
    7. 90453
    Prev
    Next
    INDEX
    Explanations

    pronounced dead, after brain, nth-child, css, id

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

    This neuron is mainly firing on “identifier”–style tokens in code or structured markup (e.g. CSS selectors like “nth‐child,” XML attributes like “xml:id”), essentially spotting code/markup identifiers.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
    }_{\
    -0.76
    inium
    -0.70
    Major
    -0.69
    semitism
    -0.69
    fame
    -0.67
     parlare
    -0.63
     Bers
    -0.61
     يك
    -0.61
    visi
    -0.60
     emlé
    -0.60
    POSITIVE LOGITS
    brain
    0.90
     brain
    0.85
     Jacobi
    0.83
     река
    0.76
     eugen
    0.75
    jag
    0.75
     Brain
    0.72
    وردار
    0.71
    ഞ്
    0.69
     outfield
    0.69
    Activations Density 0.067%

    No Known Activations