© 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
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B-IT
    4. Residual Stream - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 122472
    Prev
    Next
    INDEX
    Explanations

    instances of quotation marks

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

    C++ header includes

    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
    bosky
    -0.54
    exitRule
    -0.50
    ➯
    -0.48
    SOT
    -0.48
     Italijani
    -0.46
    IsContent
    -0.46
    GEBURTS
    -0.46
    estacks
    -0.46
    gonic
    -0.46
     Administrativna
    -0.46
    POSITIVE LOGITS
    antMatchers
    0.55
    import
    0.41
     the
    0.37
    ://
    0.37
     المعيارى
    0.37
    TestingModule
    0.36
    endcsname
    0.36
     introduces
    0.36
     craindre
    0.35
    таратура
    0.35
    Activations Density 0.003%

    No Known Activations