© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 14990
    Prev
    Next
    INDEX
    Explanations

    Class titles, election seats

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

    This neuron activates primarily on numeric tokens (e.g., years or standalone digits).

    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
     recently
    -1.05
    Already
    -1.00
     reciente
    -0.94
     aprender
    -0.92
     already
    -0.91
     recientes
    -0.89
     récente
    -0.89
    piecz
    -0.87
    CloseHandle
    -0.86
     sensational
    -0.85
    POSITIVE LOGITS
    变更
    0.94
     wichtigen
    0.93
    咔嚓
    0.92
    かは
    0.89
     Λ
    0.86
     himself
    0.85
     instead
    0.85
    Attacks
    0.84
    Ingestion
    0.83
    ët
    0.83
    Activations Density 0.009%

    No Known Activations