© 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
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B-IT
    4. Residual Stream - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 62529
    Prev
    Next
    INDEX
    Explanations

    punctuation marks and special characters

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

    code and parentheses

    np_acts-logits-general · gemini-2.5-flash-lite

    code-related syntax and developer text, especially programming punctuation, preprocessor directives, and git diff/patch blocks.

    oai_token-act-pair · gpt-5Triggered by @vetterc0
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81
    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
    Not in Any Lists

    No Comments

    Negative Logits
    <unused41>
    -1.28
    <unused14>
    -1.27
    <unused52>
    -1.27
    <unused79>
    -1.27
    <unused23>
    -1.27
    <unused8>
    -1.27
    <unused68>
    -1.27
    <unused74>
    -1.27
    [@BOS@]
    -1.27
    <unused3>
    -1.27
    POSITIVE LOGITS
    The
    0.45
    1
    0.41
    For
    0.39
    (
    0.34
    2
    0.34
    My
    0.33
        
    0.32
    
    0.31
    To
    0.31
    This
    0.30
    Activations Density 0.001%

    No Known Activations