© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 127777
    Prev
    Next
    INDEX
    Explanations

    code strings

    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
    Negative Logits
     suscit
    -1.43
     incess
    -1.38
    鳚
    -1.37
    柔ら
    -1.36
     intervi
    -1.34
     aident
    -1.34
    incredible
    -1.34
     différences
    -1.32
     rése
    -1.32
     magnific
    -1.32
    POSITIVE LOGITS
     to
    1.97
     without
    1.87
     which
    1.75
     that
    1.63
     every
    1.59
     as
    1.59
     not
    1.52
     trying
    1.51
     each
    1.51
     only
    1.49
    Activations Density 0.053%

    No Known Activations