© 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. 25588
    Prev
    Next
    INDEX
    Explanations

    more important

    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
     faciliter
    -1.55
     sowieso
    -1.45
    愜
    -1.37
     víc
    -1.36
     tiveram
    -1.36
     vůbec
    -1.34
    Beantwoorden
    -1.33
     Tienen
    -1.31
     绘画
    -1.31
    ullaby
    -1.31
    POSITIVE LOGITS
     or
    2.23
     usually
    1.62
     typically
    1.54
    往往
    1.53
     often
    1.53
     your
    1.52
     for
    1.52
     السياس
    1.51
     substantial
    1.48
     an
    1.48
    Activations Density 0.019%

    No Known Activations