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. 85641
    Prev
    Next
    INDEX
    Explanations

    HTML and JavaScript tags, particularly related to script elements

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

    closing script tags

    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
     ujednoznacz
    -0.69
    tvguidetime
    -0.60
    ConstraintMaker
    -0.59
    titleMargin
    -0.57
    URBANA
    -0.57
    PhysRevD
    -0.57
    ValueGenerated
    -0.57
    principalColumn
    -0.56
     ſei
    -0.55
    anatom
    -0.53
    POSITIVE LOGITS
    endphp
    0.57
    UserScript
    0.40
    script
    0.38
    ViewFeatures
    0.36
     Lä
    0.33
    UseVisualStyle
    0.31
     script
    0.31
    noscript
    0.31
    portál
    0.30
    AccessorTable
    0.30
    Activations Density 0.004%

    No Known Activations