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. 53832
    Prev
    Next
    INDEX
    Explanations

    code structure and control flow elements, such as function definitions and return statements

    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
    WriteTagHelper
    -0.61
     <>",
    -0.59
    DllImport
    -0.58
    󠁮
    -0.55
     للاسماء
    -0.54
     wikipagina
    -0.53
    gdx
    -0.53
    
    -0.53
    الحياه
    -0.52
    WebVitals
    -0.52
    POSITIVE LOGITS
     postre
    0.35
     dàng
    0.35
     Haft
    0.33
    regler
    0.32
    Term
    0.32
     Guten
    0.32
    queryInterface
    0.31
     étrangère
    0.31
     métiers
    0.30
     pratiques
    0.30
    Activations Density 0.008%

    No Known Activations