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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 13519
    Prev
    Next
    INDEX
    Explanations

    specific terms related to organizational structure and culture

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

    content-heavy, domain-specific keywords—especially technical identifiers, proper names, and core nouns—rather than function words.

    oai_token-act-pair · gpt-5Triggered by @jyhe0408

    compound technical identifiers or names, particularly in code consisting of multiple concatenated words (like `DataReductionProxySettings`, `nodeToValue`, `OutletBufferReserve`).

    oai_token-act-pair · claude-4-5-sonnetTriggered by @jyhe0408
    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
     a
    -0.32
     Nationen
    -0.29
     lamang
    -0.29
     Faktoren
    -0.28
     uç
    -0.28
    break
    -0.26
     udě
    -0.26
     beira
    -0.25
     Gefähr
    -0.25
     Schritte
    -0.24
    POSITIVE LOGITS
     initComponents
    0.75
    ロウィン
    0.71
     MainAxisSize
    0.68
    évaluateur
    0.68
    出版年
    0.68
    IntoConstraints
    0.65
    TypedDataSet
    0.64
    
    0.63
     ―――――
    0.63
    久しぶ
    0.62
    Activations Density 0.855%

    No Known Activations