© 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. 31-GEMMASCOPE-RES-16K
    11. 15436
    Prev
    Next
    INDEX
    Explanations

    patterns and elements related to punctuation and formatting in text

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

    words and punctuation related to technical or formal writing about systems, codes, and specifications.

    oai_token-act-pair · claude-4-5-haikuTriggered by @emiglarou
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 31-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_31/width_16k/average_l0_114
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.31.hook_resid_post
    Hook Layer
    31
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    featureID
    -1.07
    writeFieldEnd
    -0.88
    RTLI
    -0.83
    parsedMessage
    -0.82
    AddTagHelper
    -0.75
    InjectAttribute
    -0.74
    principalTable
    -0.74
    EndGlobalSection
    -0.71
    UserScript
    -0.70
     ***!
    -0.69
    POSITIVE LOGITS
     symbols
    0.69
     formatting
    0.68
     format
    0.67
     abbreviations
    0.61
     readability
    0.60
     text
    0.57
     syntax
    0.57
     punctuation
    0.56
     coding
    0.56
     graphical
    0.56
    Activations Density 1.271%

    No Known Activations