© 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
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 13028
    Prev
    Next
    INDEX
    Explanations

    names derived from words

    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
    டுக
    -1.11
     totul
    -1.06
    ̶
    -1.01
    товая
    -1.01
     support
    -1.01
     you
    -0.99
    alties
    -0.95
     toată
    -0.95
     everything
    -0.94
    Під
    -0.94
    POSITIVE LOGITS
     of
    1.48
     from
    1.25
     Suomessa
    1.09
     with
    1.07
     German
    1.02
     viena
    1.00
    地说
    0.98
     but
    0.98
     two
    0.97
     Austrian
    0.97
    Activations Density 0.024%

    No Known Activations