© 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. 40443
    Prev
    Next
    INDEX
    Explanations

    programming commands and configuration settings

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

    current configuration or state

    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
     proposed
    -0.42
    ecycle
    -0.40
     duch
    -0.40
     تعدى
    -0.39
     nonUne
    -0.38
    Diweddarwch
    -0.38
     pushing
    -0.38
    /******/
    -0.36
    Revenir
    -0.36
    라고
    -0.35
    POSITIVE LOGITS
     resourceCulture
    0.57
    EndInit
    0.51
    twimg
    0.50
    IsMutable
    0.46
    rrggbb
    0.46
    InjectAttribute
    0.45
    WebServlet
    0.45
     estekak
    0.44
     smartest
    0.44
    tableFuture
    0.43
    Activations Density 0.005%

    No Known Activations