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. 8697
    Prev
    Next
    INDEX
    Explanations

    patterns and structures in numerical or coded data

    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
    writeFieldEnd
    -0.44
     ویکی‌آمباردا
    -0.38
     ModelRenderer
    -0.37
     Fredrik
    -0.34
    ](#
    -0.34
    ToBounds
    -0.34
    endpush
    -0.34
    ParallelGroup
    -0.33
    /*++
    -0.33
     Mole
    -0.32
    POSITIVE LOGITS
     useState
    0.78
    useState
    0.73
    useEffect
    0.56
    ScopeManager
    0.54
    Chham
    0.54
     useEffect
    0.52
     useRouter
    0.52
     useContext
    0.49
     useStyles
    0.43
     OnInit
    0.42
    Activations Density 0.129%

    No Known Activations