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. 793
    Prev
    Next
    INDEX
    Explanations

    HTML attributes and elements related to UI components

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

    HTML and CSS elements related to user interface components

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

    important relational

    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
    RegressionTest
    -0.64
    invokeLater
    -0.54
    Subview
    -0.54
    igshid
    -0.51
    といけない
    -0.48
    ruptedException
    -0.48
     EdgeInsets
    -0.47
    InteropServices
    -0.46
     først
    -0.46
    LEncoder
    -0.46
    POSITIVE LOGITS
    تقاوى
    0.70
    adaptiveStyles
    0.61
    Pflege
    0.57
    Cursors
    0.57
     Talis
    0.56
    trino
    0.56
     tabu
    0.55
    __':
    
    0.54
    iastes
    0.53
     Puch
    0.52
    Activations Density 0.024%

    No Known Activations