© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 70634
    Prev
    Next
    INDEX
    Explanations

    X prefixes

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

    The neuron fires on instances of the uppercase “X” used as a prefix in technical codes or labels (e.g. chromosome arms like “Xq28,” gene‐locus shorthand like “XR,” pathogen or product names beginning with “X,” or document/format identifiers starting with “X”).

    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
     criar
    -2.38
     spettacolo
    -2.34
     limpar
    -2.27
     avoids
    -2.25
    );
    -2.25
    ます
    -2.20
     advises
    -2.20
    quelize
    -2.19
    </
    -2.19
     undermines
    -2.17
    POSITIVE LOGITS
    </i>
    2.75
     byproduct
    2.34
    ist
    2.27
    ARIUS
    2.25
     peores
    2.23
     '
    2.22
     penchant
    2.20
     انگلیسی
    2.08
    te
    2.06
     "
    2.06
    Activations Density 0.057%

    No Known Activations