© 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. 15919
    Prev
    Next
    INDEX
    Explanations

    technical terms related to digital communication and networking

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

    content related to specific technical or explicit topics across various domains (software, education, adult content, trails, etc.).

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    content-heavy subject nouns and phrases that signal specific topics or categories (technical systems, health/medical, places/routes, education, etc.).

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

    technical or specialized discourse, particularly in software, educational, or explicit adult content 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
    ViewFeatures
    -0.47
    Eksteraj
    -0.45
    Distribuzione
    -0.39
     rebajas
    -0.39
     hochzeit
    -0.39
    StructEnd
    -0.37
     łą
    -0.36
     lunak
    -0.36
     resulta
    -0.35
     escol
    -0.35
    POSITIVE LOGITS
     viewDidLoad
    0.52
    ########.
    0.51
     TextInputType
    0.50
     snippetHide
    0.49
     насељу
    0.49
    WebElementEntity
    0.48
    PRNewswire
    0.48
    ContentLoaded
    0.48
    RegressionTest
    0.48
    ″]
    0.47
    Activations Density 0.386%

    No Known Activations