© 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. 12182
    Prev
    Next
    INDEX
    Explanations

    terminology related to therapy, coaching, and professional qualifications.

    oai_token-act-pair · gemini-2.0-flash

    terms related to certification and training

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

    mathematics

    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
     AssemblyCompany
    -0.52
     محفوظة
    -0.45
    ξύ
    -0.44
    ępu
    -0.44
     quadruple
    -0.43
     Triple
    -0.43
    dibles
    -0.43
     Inſ
    -0.42
    Triple
    -0.42
     conserved
    -0.42
    POSITIVE LOGITS
    Carriera
    0.66
    XPATH
    0.63
     &___
    0.62
    WebElementEntity
    0.62
    verwijspagina
    0.59
    jsxFileName
    0.57
     Automata
    0.57
     Drapeau
    0.56
    mazione
    0.55
    PEAT
    0.55
    Activations Density 3.259%

    No Known Activations