© 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. 31-GEMMASCOPE-RES-131K
    6. 14203
    Prev
    Next
    INDEX
    Explanations

    instances of code structure and class-related keywords in programming context

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

    programming keywords

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 31-gemmascope-res-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_31/width_131k/average_l0_109
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    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
     AssemblyTitle
    -0.73
    StoryboardSegue
    -0.70
    principalColumn
    -0.65
    OGND
    -0.59
     Савезне
    -0.59
     AssemblyCulture
    -0.57
     newState
    -0.55
    enderror
    -0.55
    KommentareTeilen
    -0.55
     MenuView
    -0.54
    POSITIVE LOGITS
     private
    0.86
    private
    0.80
    public
    0.65
    constructor
    0.65
     public
    0.65
     Private
    0.59
    Private
    0.52
     privat
    0.52
    readonly
    0.51
     PRIVATE
    0.48
    Activations Density 0.002%

    No Known Activations