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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    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. 75234
    Prev
    Next
    INDEX
    Explanations

    elements related to website navigation and URL manipulation

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

    web page navigation and redirects

    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
    Tembelea
    -0.88
     ModelExpression
    -0.84
     GenerationType
    -0.84
    +:+
    -0.82
     otomatig
    -0.78
    Personensuche
    -0.78
    RTEX
    -0.77
     autorytatywna
    -0.75
     незавершена
    -0.75
    <pad>
    -0.74
    POSITIVE LOGITS
    SceneManagement
    0.52
    sendRedirect
    0.38
     s
    0.33
    s
    0.33
     espont
    0.33
    D
    0.31
    czki
    0.31
    H
    0.30
    ↵
    0.30
    WebServlet
    0.30
    Activations Density 0.003%

    No Known Activations