© 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
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 25891
    Prev
    Next
    INDEX
    Explanations

    defining relative clauses

    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
    oczes
    -1.17
     and
    -1.14
    reté
    -1.08
    Rote
    -1.08
    Sitz
    -1.07
    Tó
    -1.06
    Reina
    -1.05
    Vener
    -1.04
    Trem
    -1.04
     комнате
    -1.02
    POSITIVE LOGITS
     that
    1.78
     I
    1.34
     we
    1.27
    </h1>
    1.15
     violência
    1.10
    ₦
    1.07
     you
    1.06
     cuyas
    1.05
     of
    1.04
     cuyos
    1.04
    Activations Density 0.024%

    No Known Activations