© 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
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B
    4. Residual Stream - 16k
    5. 31-GEMMASCOPE-RES-16K
    6. 15514
    Prev
    Next
    INDEX
    Explanations

    metadata elements in structured documents

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

    metadata fields and structured document headers.

    oai_token-act-pair · claude-4-5-haikuTriggered by @emiglarou

    metadata fields in structured documents, particularly labels like "title:", "author:", "version:", and "date:" in academic papers and technical files.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @jyhe0408

    structured document metadata and front-matter headers (e.g., bibliographic, YAML/config, or code/file descriptors) indicating fields like authorship, titles, versions, dates, and similar labels.

    oai_token-act-pair · gpt-5Triggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 31-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_31/width_16k/average_l0_114
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.31.hook_resid_post
    Hook Layer
    31
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    《
    -0.41
    WireFormatLite
    -0.35
    nonatomic
    -0.35
     
    -0.35
    (@"%@",
    -0.34
    👈
    -0.34
    upiter
    -0.34
    зулта
    -0.33
     кос
    -0.33
    ==
    -0.32
    POSITIVE LOGITS
     AssemblyCompany
    0.75
    
    0.72
     kasarigan
    0.71
     resourceCulture
    0.68
     autorytatywna
    0.61
    MigrationBuilder
    0.57
    expandindo
    0.57
     Paglinawan
    0.55
    ロウィン
    0.54
     насељу
    0.54
    Activations Density 0.391%

    No Known Activations