© 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. 22-GEMMASCOPE-RES-131K
    7. 79930
    Prev
    Next
    INDEX
    Explanations

    numbers, dates, and medical terms

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

    numeric references and citations in academic or scientific papers.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    This neuron activates on numeric tokens (integers and decimals) such as years, measurements, and reference 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_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
     администрации
    -0.94
     quase
    -0.94
    只剩
    -0.93
    enet
    -0.90
    了一口
    -0.89
    čnih
    -0.88
    以及
    -0.87
    ʻ
    -0.87
     Fr
    -0.87
     и
    -0.86
    POSITIVE LOGITS
     inqui
    0.98
    Ikus
    0.93
     oeil
    0.89
     Fuente
    0.88
     Clase
    0.87
    🧮
    0.87
     honte
    0.86
     each
    0.86
     diser
    0.86
     lópez
    0.85
    Activations Density 0.009%

    No Known Activations