© 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
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 84095
    Prev
    Next
    INDEX
    Explanations

    words that follow

    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
     bijna
    -1.02
     velmi
    -0.99
     zowel
    -0.98
     bardzo
    -0.96
     naast
    -0.96
     daardoor
    -0.92
     наиболее
    -0.92
    ganggu
    -0.91
    非常に
    -0.90
     behaup
    -0.88
    POSITIVE LOGITS
     digno
    0.92
    нились
    0.89
     夫
    0.88
     Werken
    0.86
    bbl
    0.86
     عشر
    0.86
    anderen
    0.86
     Januari
    0.85
     Qaeda
    0.85
    🛡
    0.85
    Activations Density 0.159%

    No Known Activations