© 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
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B
    4. Residual Stream - 16k
    5. 36-GEMMASCOPE-RES-16K
    6. 5041
    Prev
    Next
    INDEX
    Explanations

    JavaScript import and export statements

    oai_token-act-pair · gpt-4o-mini
    Triggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 36-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_36/width_16k/average_l0_120
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.36.hook_resid_post
    Hook Layer
    36
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Negative Logits
    httphttps
    -0.69
     @"/
    -0.66
    DockStyle
    -0.61
    Hentet
    -0.61
    AddTagHelper
    -0.61
     ivelany
    -0.57
    FormTagHelper
    -0.57
     AssemblyTitle
    -0.56
    aarrggbb
    -0.55
     חיצוניים
    -0.54
    POSITIVE LOGITS
    const
    1.21
     const
    1.08
    let
    0.90
    function
    0.80
     let
    0.75
    export
    0.75
     function
    0.68
    var
    0.64
     Const
    0.56
     Let
    0.52
    Activations Density 0.089%

    No Known Activations