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. 62900
    Prev
    Next
    INDEX
    Explanations

    references to web URLs and specific path components from a website

    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
    GEBURTSDATUM
    -0.77
    WebElementEntity
    -0.68
     Italijani
    -0.65
    afficheront
    -0.64
     ویکی‌پدی
    -0.61
     мәкал
    -0.60
     surla
    -0.57
    Panamoan
    -0.56
     queſta
    -0.56
     yyr
    -0.55
    POSITIVE LOGITS
     also
    0.46
    
    0.40
     announce
    0.40
     elsewhere
    0.40
     initComponents
    0.39
     juga
    0.38
     another
    0.37
     auch
    0.37
     announces
    0.36
    VersionUID
    0.35
    Activations Density 0.051%

    No Known Activations