© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 38868
    Prev
    Next
    INDEX
    Explanations

    performance outcomes and comparisons

    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
     السياس
    -1.29
     sogenannten
    -1.27
    عرض
    -1.18
    יך
    -1.15
     kathol
    -1.13
     शरीर
    -1.11
     sebagai
    -1.08
     påvir
    -1.08
     الحكومة
    -1.06
    Jó
    -1.06
    POSITIVE LOGITS
     results
    1.35
     الشاشة
    1.30
     especially
    1.26
    \
    1.22
     Dalio
    1.20
     Approx
    1.17
     cérami
    1.15
     problem
    1.14
     résultats
    1.13
     performance
    1.13
    Activations Density 0.031%

    No Known Activations