© 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. 65322
    Prev
    Next
    INDEX
    Explanations

    research and development verbs

    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
     really
    -1.39
     or
    -1.37
     our
    -1.28
     just
    -1.21
     tenuta
    -1.19
     actually
    -1.18
     macetas
    -1.16
     different
    -1.11
     this
    -1.10
     sırasında
    -1.10
    POSITIVE LOGITS
     to
    2.31
     langage
    1.28
     for
    1.21
    €)
    1.20
     minuter
    1.17
    ssssssss
    1.17
     NIGERIA
    1.15
     chev
    1.14
    '';
    1.13
    УД
    1.13
    Activations Density 0.203%

    No Known Activations