© 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. 128498
    Prev
    Next
    INDEX
    Explanations

    code comments and phrases

    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
     you
    -2.75
     does
    -2.64
     should
    -2.52
     nearing
    -2.52
     are
    -2.47
     amidst
    -2.47
     before
    -2.44
     where
    -2.42
     swiftly
    -2.36
     in
    -2.33
    POSITIVE LOGITS
    だいぶ
    2.56
    ’
    2.38
    ときに
    2.30
    〼
    2.27
     美甲
    2.25
     melding
    2.25
    
    2.23
     delightfully
    2.13
    菒
    2.13
    久し
    2.06
    Activations Density 0.000%

    No Known Activations