© 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. 71588
    Prev
    Next
    INDEX
    Explanations

    caused by, characterized by, down to

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

    the presence of numerical or statistical measurements (e.g. percentages, counts, or other numeric data) in the text.

    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
     our
    -1.48
    に於
    -1.47
     we
    -1.39
     kesel
    -1.34
    🤟
    -1.29
    👊
    -1.22
     kuliah
    -1.21
    至此
    -1.21
     verbeteren
    -1.20
     kuah
    -1.20
    POSITIVE LOGITS
     ‌
    1.62
     вас
    1.41
     вам
    1.39
     nyingi
    1.33
    <strong>
    1.30
     research
    1.30
    你
    1.28
     разных
    1.27
     kwenye
    1.23
    ‌
    1.23
    Activations Density 0.003%

    No Known Activations