© 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. 2127
    Prev
    Next
    INDEX
    Explanations

    referenced questions and answers involving specific data points and metrics

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

    demographic categories and identifying information in technical or medical documents.

    oai_token-act-pair · claude-3-5-haiku-20241022Triggered by @neilrathi

    labels and headers in structured or tabular text, especially colon-delimited fields and column or section titles.

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

    formatting elements like colons, commas, and whitespace in structured documents such as tables, forms, and code.

    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
    AddTagHelper
    -0.58
    XmlAccessorType
    -0.51
    WriteTagHelper
    -0.50
     CreateTagHelper
    -0.50
    StoryboardSegue
    -0.47
     kaarangay
    -0.45
    whatever
    -0.44
    FormTagHelper
    -0.44
    tagHelperRunner
    -0.42
     Taktlose
    -0.42
    POSITIVE LOGITS
     ſtand
    0.53
    </thead>
    0.47
     immédi
    0.43
    oulli
    0.43
    conium
    0.42
    Referanser
    0.41
     cauſe
    0.41
    wiſe
    0.41
     ſtate
    0.41
     pleaſure
    0.40
    Activations Density 0.122%

    No Known Activations