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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-9B-IT
    7. Residual Stream - 131k
    9. 20-GEMMASCOPE-RES-131K
    11. 122044
    Prev
    Next
    INDEX
    Explanations

    relevance to quantity, specifically numerical values or factors

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

    the number 5

    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
     Bertram
    -0.60
    traveling
    -0.59
    Bibliograf
    -0.57
    逅
    -0.56
    cour
    -0.55
    chre
    -0.55
    urator
    -0.54
    ruitment
    -0.54
    etting
    -0.54
    odule
    -0.54
    POSITIVE LOGITS
    5
    1.42
    ５
    0.87
     cinquième
    0.87
    five
    0.82
     cinco
    0.81
     Five
    0.80
     five
    0.79
     cinq
    0.78
    Five
    0.77
     vijf
    0.75
    Activations Density 0.746%

    No Known Activations