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

    Neuronpedia

    APISteerSAE EvalsBlog/PodcastNEWSlackPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlog/PodcastGitHubSlackTwitterContact
    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. 33415
    Prev
    Next
    INDEX
    Explanations

    terms related to chemical reactions and properties

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    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
    !*\
    -0.37
     Integrated
    -0.35
    Integrated
    -0.35
    rophoresis
    -0.34
     nhật
    -0.33
     running
    -0.33
    HEP
    -0.33
     pegged
    -0.33
     magically
    -0.33
     meeting
    -0.33
    POSITIVE LOGITS
     OFDb
    0.63
     ModelExpression
    0.58
     InputDecoration
    0.55
     >=",
    0.52
    Jîn
    0.50
    ThroughAttribute
    0.47
    
    0.46
     ujednoznacz
    0.46
    MigrationBuilder
    0.45
     henvisninger
    0.43
    Activations Density 0.144%

    No Known Activations