© 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. 12505
    Prev
    Next
    INDEX
    Explanations

    names of cities and countries

    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
     Heere
    -1.03
     usados
    -0.94
     Somal
    -0.93
     Pria
    -0.92
     disponible
    -0.91
    RESUMEN
    -0.91
     Parlement
    -0.88
     alimenta
    -0.88
     レッド
    -0.88
     instead
    -0.88
    POSITIVE LOGITS
    ofed
    1.09
    ),),
    1.08
    ','$
    1.05
     хотите
    1.03
     Want
    1.02
    讲话
    1.01
    ENCIÓN
    1.01
     Danish
    0.99
     kawasan
    0.98
     cambian
    0.98
    Activations Density 0.056%

    No Known Activations