© 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. 126268
    Prev
    Next
    INDEX
    Explanations

    references to baked goods or pastries

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

    muffins, rolls, and buns

    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
     kasarigan
    -0.47
    mern
    -0.43
    preduce
    -0.42
    iastes
    -0.42
    ViewImports
    -0.39
     groundwater
    -0.37
     matériel
    -0.37
     Sangu
    -0.36
     accès
    -0.36
     Kansas
    -0.36
    POSITIVE LOGITS
     pastries
    0.72
     pastry
    0.66
     Pastry
    0.63
     bagel
    0.63
     muffins
    0.62
     croissants
    0.61
     Biscuit
    0.61
     muffin
    0.61
    PreferredItem
    0.60
     bagels
    0.59
    Activations Density 0.005%

    No Known Activations