© 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
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B
    4. Residual Stream - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 16358
    Prev
    Next
    INDEX
    Explanations

    patterns or groups of letters that may represent elements of a code or encoding structure

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

    document headings and title text in legal documents, scientific papers, and formal publications.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered 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
    rungsseite
    -0.54
     dimensión
    -0.49
     práctico
    -0.48
     almendras
    -0.47
    Sådan
    -0.47
     rozwiązania
    -0.47
     caneca
    -0.47
     mármol
    -0.46
     mogą
    -0.46
     lähe
    -0.46
    POSITIVE LOGITS
     umbi
    0.47
     EconPapers
    0.47
     SqlDbType
    0.44
    ftagPool
    0.44
    woordig
    0.44
    ly
    0.42
    chyma
    0.41
    ſt
    0.41
    Portail
    0.41
     DIC
    0.41
    Activations Density 0.084%

    No Known Activations