© 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. 10-GEMMASCOPE-RES-131K
    7. 59155
    Prev
    Next
    INDEX
    Explanations

    contractions like wasn't, doesn't, isn't

    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_10/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
     after
    -2.23
     had
    -1.98
     “
    -1.95
     not
    -1.93
     
    -1.87
     because
    -1.80
     using
    -1.79
     those
    -1.77
     during
    -1.70
     t
    -1.70
    POSITIVE LOGITS
    毟
    2.44
    
    2.13
     mawar
    2.08
    擽
    2.08
     célè
    2.02
    塒
    2.02
     sepatu
    1.99
     poë
    1.98
     serre
    1.92
    这一幕
    1.90
    Activations Density 0.576%

    No Known Activations