© 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
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B-IT
    4. Residual Stream - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 76772
    Prev
    Next
    INDEX
    Explanations

    properties and characteristics of graphical elements or layout styles in a document

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

    ${\mkern2mu\underline{\mkern-2mu\smash{\phi }\mkern-2mu}\mkern2mu}$

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    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
    OGND
    -0.85
    Ӕ
    -0.69
    migrationBuilder
    -0.63
    tagHelperRunner
    -0.61
     препратки
    -0.60
    HtmlAttribute
    -0.59
     виправивши
    -0.59
    adaptiveStyles
    -0.58
     дописавши
    -0.58
    Diweddarwch
    -0.58
    POSITIVE LOGITS
    Personendaten
    0.36
    ,
    0.35
     kap
    0.34
    AndEndTag
    0.33
     charme
    0.32
     formy
    0.32
     norm
    0.32
     AssemblyProduct
    0.31
    了嗎
    0.31
    ください
    0.31
    Activations Density 0.110%

    No Known Activations