© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-9B
    7. Residual Stream - 16k
    9. 20-GEMMASCOPE-RES-16K
    11. 5220
    Prev
    Next
    INDEX
    Explanations

    code-related terms that indicate data manipulation and querying processes

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    programming code or syntax, specifically related to C++, Nginx, or database operations.

    oai_token-act-pair · claude-3-5-haiku-20241022Triggered by @neilrathi
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_20/width_16k/average_l0_68
    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
     surla
    -0.45
     continúas
    -0.44
    ----</
    -0.42
     Autorisations
    -0.40
    Cyfeiriadau
    -0.40
     accountId
    -0.36
     ſche
    -0.36
     محفوظة
    -0.36
    匿名使用者
    -0.35
    localctx
    -0.34
    POSITIVE LOGITS
    aarrggbb
    0.56
     defaultstate
    0.52
     Paglinawan
    0.52
    AlterField
    0.49
    /**
    0.49
     oprot
    0.48
    ########.
    0.47
    XmlAccessorType
    0.46
     disambiguazione
    0.45
    endphp
    0.45
    Activations Density 0.150%

    No Known Activations