© 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. 45340
    Prev
    Next
    INDEX
    Explanations

    patent citations and WO numbers

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

    The neuron detects numeric identifiers—especially patent and document numbers.

    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
    chtigkeit
    -0.93
     جديدة
    -0.93
    ORET
    -0.88
     corrom
    -0.88
     prostu
    -0.85
     leña
    -0.84
    caloosa
    -0.84
     hipote
    -0.84
     vás
    -0.82
    系統
    -0.82
    POSITIVE LOGITS
    OTHER
    1.00
     cited
    0.94
    FOREIGN
    0.92
     Other
    0.79
     OTHER
    0.79
    Responsable
    0.76
    日本では
    0.76
     схо
    0.75
     británico
    0.74
    我喜欢
    0.73
    Activations Density 0.022%

    No Known Activations