© 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
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 72548
    Prev
    Next
    INDEX
    Explanations

    programming constructs and data attributes

    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
     Euse
    -1.11
    сль
    -0.95
     hcm
    -0.94
     homenagem
    -0.94
     bcm
    -0.94
    OOTDTY
    -0.92
    忽略
    -0.91
     biografi
    -0.91
    лись
    -0.90
     pérd
    -0.90
    POSITIVE LOGITS
     our
    1.90
     some
    1.27
    むしろ
    1.17
    ONSORED
    1.13
     even
    1.10
     other
    1.03
     most
    1.02
    咐
    1.02
     referenties
    0.98
    などが
    0.96
    Activations Density 0.003%

    No Known Activations