© 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
    3. Gemma-2-2B
    5. 21-GEMMASCOPE-TRANSCODER-16K
    7. 12109
    Prev
    Next
    INDEX
    Explanations

    words associated with kingdoms, leadership, and/or royalty

    oai_token-act-pair · gemini-2.0-flash

    titles of nobility

    np_token-act-pair-logits · gpt-4o-mini

    royal titles

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_21/width_16k/average_l0_13
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.21.ln2.hook_normalized
    Architecture
    jumprelu_transcoder
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     homens
    -0.54
     umana
    -0.53
     mensen
    -0.52
     människor
    -0.52
     Humanos
    -0.51
     mense
    -0.51
    CodeAttribute
    -0.50
     umane
    -0.49
    omenta
    -0.49
    NewUrlParser
    -0.49
    POSITIVE LOGITS
     duke
    0.68
    WebElementEntity
    0.68
     king
    0.63
     Duke
    0.60
     governor
    0.59
     Viceroy
    0.59
     Jefus
    0.59
    كونات
    0.59
    duke
    0.58
     chancellor
    0.58
    Activations Density 1.925%

    No Known Activations