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 - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 8587
    Prev
    Next
    INDEX
    Explanations

    code structure and component configurations in programming contexts

    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-16k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    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
    UserScript
    -0.43
    Inform
    -0.40
    jsonwebtoken
    -0.40
    ulemon
    -0.39
    Cyfarwyddwr
    -0.37
     uyar
    -0.36
     Inform
    -0.35
    autogui
    -0.35
    readLine
    -0.33
     Hitchcock
    -0.32
    POSITIVE LOGITS
    NgModule
    0.79
     initComponents
    0.79
    declarations
    0.76
    出版年
    0.73
    Personendaten
    0.68
    Hentet
    0.66
    MigrationBuilder
    0.66
    IntoConstraints
    0.64
    ViewFeatures
    0.63
    Autoritní
    0.60
    Activations Density 0.022%

    No Known Activations