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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-9B-IT
    7. Residual Stream - 16k
    9. 20-GEMMASCOPE-RES-16K
    11. 2999
    Prev
    Next
    INDEX
    Explanations

    indicative phrases and their context in reports or summaries

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

    technical or scientific terms, names, and quantities, especially those related to measurements, numerical values, biological terms, and specialized scientific vocabulary in academic or technical contexts.

    oai_token-act-pair · gpt-4.1-2025-04-14Triggered by @ywhp2c8tjm

    followed by units of measurement

    np_acts-logits-general · gemini-2.0-flash

    physical dimensions and structures

    np_acts-logits-general · gemini-2.5-flash-lite

    technical, academic-style expository text featuring domain-specific jargon, measurements and units, acronyms, and formal structural or formatting markers characteristic of scientific or encyclopedic writing.

    oai_token-act-pair · gpt-5Triggered by @vetterc0
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91
    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
    transQ
    -0.56
    tagHelperRunner
    -0.55
     Italijani
    -0.54
    Ӕ
    -0.50
     brancas
    -0.47
     @"/
    -0.46
     AssemblyCompany
    -0.45
     betweenstory
    -0.44
    bcryptjs
    -0.43
    OGND
    -0.43
    POSITIVE LOGITS
    IntoConstraints
    0.97
     defaultstate
    0.55
    ]';
    0.47
    setViewportView
    0.46
    хьтан
    0.46
    astéro
    0.45
     فريبيس
    0.43
    AnimationsModule
    0.40
     ujednoznacz
    0.40
     linkovi
    0.40
    Activations Density 32.519%

    No Known Activations