© 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. 35841
    Prev
    Next
    INDEX
    Explanations

    JSON structures and metadata

    np_acts-logits-general · gemini-2.5-flash-lite
    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
    Zubereitung
    -1.36
    ubblica
    -1.20
    </tfoot>
    -1.17
    (!__
    -1.16
    Many
    -1.14
    bourhood
    -1.14
     podem
    -1.13
    Like
    -1.12
    vább
    -1.12
    既に
    -1.10
    POSITIVE LOGITS
     not
    1.32
    ↵
    1.27
     ולא
    1.23
     takich
    1.16
    '
    1.14
    Algérie
    1.14
    frase
    1.13
    などと
    1.11
     betreffende
    1.07
    ’
    1.07
    Activations Density 0.003%

    No Known Activations