© 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. 11441
    Prev
    Next
    INDEX
    Explanations

    topics related to health and safety

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

    topic-defining domain terms—specialized nouns and field/industry/context keywords that signal the passage’s subject area.

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

    specialized or technical terminology related to specific professional fields, industries, or institutional contexts.

    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
     للاسماء
    -0.85
    цездатний
    -0.82
     виправивши
    -0.72
     Italijanski
    -0.70
     kasarigan
    -0.68
    <unused47>
    -0.67
    <unused41>
    -0.67
    <unused43>
    -0.67
    [@BOS@]
    -0.67
    <unused3>
    -0.67
    POSITIVE LOGITS
     popular
    0.38
     Popular
    0.35
     POPULAR
    0.31
     Gemeinsame
    0.31
     Montag
    0.30
     Or
    0.30
     bar
    0.30
     rips
    0.30
    soort
    0.30
     therefore
    0.30
    Activations Density 0.549%

    No Known Activations