© 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
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 38692
    Prev
    Next
    INDEX
    Explanations

    code structure elements

    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
     and
    -2.92
     an
    -2.47
     the
    -2.41
    為に
    -2.38
     as
    -2.28
     The
    -2.28
     In
    -2.08
     Of
    -2.08
     creates
    -1.88
     There
    -1.85
    POSITIVE LOGITS
     diki
    2.50
    焢
    2.47
     vokal
    2.44
     kulak
    2.39
    これは
    2.39
     makam
    2.38
     唯美
    2.36
     seksi
    2.33
     eſſ
    2.30
    킁
    2.30
    Activations Density 0.174%

    No Known Activations