Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2026
    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. 79029
    Prev
    Next
    INDEX
    Explanations

    patterns and structures within code or data related to programming

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

    long-anticipated musical

    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
    AsUp
    -0.44
    minecraft
    -0.35
    uxxxx
    -0.34
     ब्रेकडाउन
    -0.34
     Dete
    -0.34
     ciri
    -0.34
    履
    -0.33
     Jau
    -0.32
    rgba
    -0.32
    cstdlib
    -0.32
    POSITIVE LOGITS
     EconPapers
    0.64
     StatelessWidget
    0.58
     ModelExpression
    0.50
     pungkas
    0.47
     pungkasnya
    0.46
     psicológica
    0.46
    Према
    0.46
    Geplaatst
    0.45
    MigrationBuilder
    0.45
     jadx
    0.45
    Activations Density 0.057%

    No Known Activations