© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-9B
    7. Residual Stream - 131k
    9. 20-GEMMASCOPE-RES-131K
    11. 105374
    Prev
    Next
    INDEX
    Explanations

    punctuation marks and structural elements within the text

    oai_token-act-pair · gpt-4o-mini
    Triggered by @bot

    Sentence-ending or clause-separating punctuation marks (primarily periods and commas), often followed by closing quotation marks or parentheses in technical, academic, or online content.

    eleuther_acts_top20 · claude-3-7-sonnet-20250219
    Triggered by @lu-christina
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-9b-pt-res/layer_20/width_131k/average_l0_114
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    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
    Negative Logits
     autorytatywna
    -0.41
    хьтан
    -0.36
    XtraReports
    -0.35
    AccessorTable
    -0.34
    livion
    -0.33
     <>",
    -0.33
     hẹn
    -0.31
     disponibilités
    -0.31
     mxArray
    -0.31
     konz
    -0.31
    POSITIVE LOGITS
     Савез
    0.50
    /**
    0.47
    +#+
    0.46
    enderror
    0.46
    DockStyle
    0.46
    Leggi
    0.44
    verwijs
    0.44
    stick
    0.44
    Tikang
    0.44
    Stick
    0.44
    Activations Density 0.179%

    No Known Activations