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-2-9B
    4. Residual Stream - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 4878
    Prev
    Next
    INDEX
    Explanations

    distinctions between similar terms or concepts

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

    The beginning characters of formatting or markup tags in technical text, particularly those used in HTML, markdown, or LaTeX typesetting contexts.

    eleuther_acts_top20 · claude-3-7-sonnet-20250219Triggered by @lu-christina
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-9b-pt-res/layer_20/width_131k/average_l0_114
    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
     omnia
    -0.33
    Citiți
    -0.32
     Nebenwirkungen
    -0.31
     gloria
    -0.30
    vertis
    -0.30
     quiet
    -0.29
     quæ
    -0.29
    Handled
    -0.29
     unpredictable
    -0.28
    fæ
    -0.28
    POSITIVE LOGITS
     kasarigan
    0.67
    omitempty
    0.66
    السكان
    0.63
     ModelExpression
    0.62
    ArgsConstructor
    0.60
     виправивши
    0.60
    HtmlAttribute
    0.59
     المعيارى
    0.57
     فريبيس
    0.57
    verwijspagina
    0.57
    Activations Density 0.013%

    No Known Activations