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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B
    4. Residual Stream - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 485
    Prev
    Next
    INDEX
    Explanations

    HTML and database-related structures in code

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    HTML/XML code structures, particularly table rows, cells, and related markup elements.

    oai_token-act-pair · claude-3-5-haiku-20241022Triggered by @neilrathi
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_20/width_16k/average_l0_68
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.20.hook_resid_post
    Hook Layer
    20
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    enterOuterAlt
    -0.50
    Ligações
    -0.50
     FetchType
    -0.49
    RectangleBorder
    -0.47
    writeFieldEnd
    -0.47
    Personendaten
    -0.46
     resourceCulture
    -0.46
     AssemblyCompany
    -0.45
     ſtill
    -0.45
    BASEPATH
    -0.45
    POSITIVE LOGITS
    toarele
    0.42
    NM
    0.34
    期刊论文
    0.33
     Wicidata
    0.33
     "..\..\
    0.33
     indefinitely
    0.33
     hebdomada
    0.33
    hloro
    0.32
     episode
    0.32
     regularly
    0.32
    Activations Density 0.077%

    No Known Activations