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
    4. Residual Stream - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 16442
    Prev
    Next
    INDEX
    Explanations

    opinions and suggestions regarding features and improvements

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-9b-pt-res/layer_20/width_131k/average_l0_114
    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
     snippetHide
    -0.59
     AssemblyProduct
    -0.57
     ویکی‌پدی
    -0.57
    хьтан
    -0.57
     समीक्षाओं
    -0.53
     للاسماء
    -0.51
    parsedMessage
    -0.51
    NameInMap
    -0.51
    Personensuche
    -0.50
    WindowConstants
    -0.50
    POSITIVE LOGITS
     gustaría
    0.35
     Privacidade
    0.35
    ุ
    0.34
     customizable
    0.34
    StackNavigator
    0.33
    Fug
    0.32
     funcionalidades
    0.32
    مصادر
    0.32
    fug
    0.32
    reszcie
    0.32
    Activations Density 0.050%

    No Known Activations