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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    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. 91366
    Prev
    Next
    INDEX
    Explanations

    URL patterns and structural elements in code

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

    path prefixes

    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
     iyon
    -0.34
     étui
    -0.33
    hpp
    -0.32
     آش
    -0.29
     Absch
    -0.29
     regno
    -0.28
     vôtre
    -0.28
     juges
    -0.28
     ministerio
    -0.27
     мет
    -0.26
    POSITIVE LOGITS
    Jeografia
    0.78
    VersionUID
    0.78
    printStackTrace
    0.77
     betweenstory
    0.77
     ModelExpression
    0.77
    GEBURTSDATUM
    0.76
    OGND
    0.75
    ✨:
    0.73
     Paglinawan
    0.72
    setVerticalGroup
    0.72
    Activations Density 0.020%

    No Known Activations