Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-2B
    4. Residual Stream - 65k
    5. 20-GEMMASCOPE-RES-65K
    6. 960
    Prev
    Next
    INDEX
    Explanations

    the beginning and end of documents or sections within documents

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

    the beginning of statements or sections in the document

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

    bb followed by ABOUT

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 20-gemmascope-res-65k
    Configuration
    google/gemma-scope-2b-pt-res/layer_20/width_65k/average_l0_114
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    65,536
    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
     AssemblyVersion
    -0.61
    +#+#
    -0.58
    rigkeit
    -0.54
    Хьажоргаш
    -0.53
     JSTOR
    -0.53
     minimum
    -0.52
     Gual
    -0.52
     ویکی‌پدیای
    -0.51
     közül
    -0.51
     <<<<<<<<<<<<<<
    -0.51
    POSITIVE LOGITS
     &___
    0.76
    </td>
    0.75
    </caption>
    0.70
    <h6>
    0.64
    ↩
    0.64
     betweenstory
    0.61
     Saltar
    0.60
    <table>
    0.58
    )];
    
    0.58
    ՚
    0.58
    Activations Density 0.003%

    No Known Activations