© 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. 39316
    Prev
    Next
    INDEX
    Explanations

    code definitions and commands

    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
     Dinas
    -1.00
    唐揚げ
    -0.98
    <%@
    -0.95
    的道
    -0.91
    persky
    -0.84
     burbuja
    -0.84
    årige
    -0.84
    tlement
    -0.84
    portun
    -0.84
    复
    -0.84
    POSITIVE LOGITS
     including
    1.25
     typical
    1.24
     include
    1.22
     their
    1.10
     includes
    1.08
    包括
    1.06
     for
    1.02
     typique
    1.01
    خخ
    1.00
    Ингредиенты
    0.99
    Activations Density 0.029%

    No Known Activations