Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-2B
    4. Residual Stream - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 6664
    Prev
    Next
    INDEX
    Explanations

    references to specific entities, names, or identifiers related to research and diagnostics

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

    Japanese text or abbreviations

    np_acts-logits-general · gemini-2.0-flash

    RS, Mason, Japanese questions, errors

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71
    Prompts (Dashboard)
    36,864 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
    ation
    -0.81
     estekak
    -0.64
    reibt
    -0.63
    й
    -0.63
    posedge
    -0.62
    vernight
    -0.61
     initComponents
    -0.61
    SaveChangesAsync
    -0.60
     requestCode
    -0.60
     referenties
    -0.60
    POSITIVE LOGITS
    bergh
    0.62
    úsqueda
    0.58
    wali
    0.58
    '}>
    0.57
    Lur
    0.56
    rows
    0.55
    +</
    0.55
     SPDX
    0.55
     Osh
    0.54
    Ы
    0.54
    Activations Density 0.514%

    No Known Activations