© 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. 22-GEMMASCOPE-RES-131K
    4. 13660
    Prev
    Next
    INDEX
    Explanations

    medical and technical terms

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

    hyphenated compound terms in medical and technical contexts.

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

    This neuron responds to numeric and hyphenated tokens (e.g. numbers, decimals, and dashes used in technical or compound terms).

    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
    odyne
    -1.28
     suom
    -1.27
    阵法
    -1.13
    ᜀ
    -1.13
     distanciation
    -1.10
     berço
    -1.09
     aima
    -1.08
     peur
    -1.08
     seigneur
    -1.07
     שלנו
    -1.07
    POSITIVE LOGITS
     and
    1.74
     mentre
    1.34
    了许多
    1.20
     weiterhin
    1.08
     comienzan
    1.07
    nesty
    1.05
    了不少
    1.05
     '-';
    1.02
    謠
    1.01
     easter
    0.98
    Activations Density 0.091%

    No Known Activations