© 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. 25722
    Prev
    Next
    INDEX
    Explanations

    table names or structure

    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
     even
    -1.23
     or
    -1.11
     Even
    -1.06
    ragende
    -1.01
     ἡ
    -1.01
     in
    -0.99
     tarda
    -0.99
     čet
    -0.99
     mesmo
    -0.99
    上学
    -0.98
    POSITIVE LOGITS
    ⽼
    1.16
    erobic
    1.15
     maïs
    1.08
    ;'>
    1.07
     Bahasa
    1.07
    ִּ
    1.03
     Muchas
    1.02
     fides
    1.02
    marginRight
    1.01
     Савез
    1.01
    Activations Density 0.005%

    No Known Activations