© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 66487
    Prev
    Next
    INDEX
    Explanations

    register access memory

    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
     konkret
    -1.69
     města
    -1.55
     אנחנו
    -1.48
     struk
    -1.47
    すごく
    -1.41
     nuestros
    -1.40
     玩具
    -1.39
     техника
    -1.38
     ekip
    -1.38
     kaupung
    -1.38
    POSITIVE LOGITS
     at
    1.88
     has
    1.63
     each
    1.60
     there
    1.55
    There
    1.52
    奬
    1.48
     during
    1.45
     only
    1.42
     on
    1.38
    Each
    1.34
    Activations Density 0.007%

    No Known Activations