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. 1004
    Prev
    Next
    INDEX
    Explanations

    HTML link elements referencing stylesheet files

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

    hyperlinks related to CSS stylesheets

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

    CSS links and stylesheets

    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
    endregion
    -0.56
     Political
    -0.56
    
    -0.56
     snippetHide
    -0.56
    पया
    -0.55
     Residency
    -0.53
    الحياه
    -0.52
    fromCharCode
    -0.51
     morire
    -0.51
     political
    -0.51
    POSITIVE LOGITS
     style
    1.14
     styles
    1.09
     styling
    1.08
     Styles
    1.00
     Style
    0.99
    Styles
    0.99
    stylesheet
    0.96
    style
    0.96
    Style
    0.95
    styles
    0.93
    Activations Density 0.027%

    No Known Activations