© 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. 4483
    Prev
    Next
    INDEX
    Explanations

    passages about human mating systems, social structures, and genetics, especially related to patriarchy/matriarchy and polygyny.

    oai_token-act-pair · gemini-2.0-flash

    terms related to social structures

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

    social/kinship studies

    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
    idosis
    -0.48
    LookAnd
    -0.45
     bibliote
    -0.44
     phủ
    -0.42
    MenuView
    -0.41
     zove
    -0.41
     idiot
    -0.40
    liv
    -0.39
     libraries
    -0.39
    epiece
    -0.39
    POSITIVE LOGITS
    ScopeManager
    0.75
    xase
    0.68
    CloseOperation
    0.65
    /******/
    0.60
     typelib
    0.60
     social
    0.59
    mbre
    0.59
     group
    0.59
     socially
    0.59
    Rohy
    0.57
    Activations Density 3.219%

    No Known Activations