© 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. 100386
    Prev
    Next
    INDEX
    Explanations

    Co- prefix or occipital regions

    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
    which
    -3.20
    lossians
    -2.80
    ."
    -2.77
    ixante
    -2.53
    g
    -2.38
    ABLE
    -2.31
    most
    -2.30
    
    -2.30
    ;
    -2.25
    ggish
    -2.25
    POSITIVE LOGITS
     difficulty
    2.69
     thể
    2.39
     upsetting
    2.28
     但是
    2.25
    
    2.20
    }
    2.17
     concentrado
    2.16
     appeared
    2.11
     become
    2.11
     democrat
    2.05
    Activations Density 0.001%

    No Known Activations