© 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. 10-GEMMASCOPE-RES-131K
    7. 103595
    Prev
    Next
    INDEX
    Explanations

    primary structure

    np_acts-logits-general · gemini-2.5-flash-lite

    section headers or title elements in a document, particularly those marked as "primary" or that function as main navigation elements.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi
    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
     are
    -1.89
    Our
    -1.72
     It
    -1.66
     ‘
    -1.63
     types
    -1.56
    ":
    -1.55
    </strong>
    -1.54
     zealous
    -1.54
     `
    -1.51
     особы
    -1.48
    POSITIVE LOGITS
    primary
    1.81
     poud
    1.66
     PRIMARY
    1.66
    Primary
    1.63
     départements
    1.60
    庒
    1.59
    铄
    1.54
     spiele
    1.53
     schuhe
    1.52
    鮠
    1.52
    Activations Density 0.070%

    No Known Activations