© 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. 25381
    Prev
    Next
    INDEX
    Explanations

    describing conditions and capabilities

    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
     inond
    -1.78
     célè
    -1.76
     extré
    -1.71
     obiet
    -1.71
    킁
    -1.68
     éte
    -1.66
    えっ
    -1.65
     réinitial
    -1.65
     intermédiaire
    -1.63
    渽
    -1.63
    POSITIVE LOGITS
     for
    1.65
     T
    1.59
     L
    1.56
     d
    1.55
     all
    1.52
     in
    1.52
     is
    1.50
     use
    1.49
     has
    1.49
     N
    1.48
    Activations Density 0.015%

    No Known Activations