© 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. 10-GEMMASCOPE-RES-131K
    7. 15355
    Prev
    Next
    INDEX
    Explanations

    command: keyword

    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_10/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
     może
    -1.47
     الذي
    -1.44
    Should
    -1.38
    llavero
    -1.34
    mvh
    -1.34
     наших
    -1.33
    Приложение
    -1.32
    Lịch
    -1.31
     ambayo
    -1.30
     oleju
    -1.30
    POSITIVE LOGITS
     it
    1.98
     not
    1.82
     as
    1.78
     its
    1.64
     about
    1.48
     and
    1.38
     before
    1.29
     on
    1.27
     their
    1.21
     like
    1.21
    Activations Density 0.189%

    No Known Activations