Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B
    4. Residual Stream - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 73680
    Prev
    Next
    INDEX
    Explanations

    coding-related terminology and syntax

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

    Variable names, function parameters, and technical terminology in code, scientific classifications, references, and API calls - often showing relationships between related technical elements like parent/child components or taxonomic hierarchies.

    eleuther_acts_top20 · claude-3-7-sonnet-20250219Triggered 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
    Not in Any Lists

    No Comments

    Negative Logits
    AnchorTagHelper
    -0.55
    MethodManager
    -0.50
    EndContext
    -0.47
    (!__
    -0.46
     ***!
    -0.43
    WriteBarrier
    -0.42
    IsMutable
    -0.40
    WithIdentifier
    -0.39
    havior
    -0.39
     oprot
    -0.37
    POSITIVE LOGITS
     Administrativna
    0.80
    :✨
    0.65
    MigrationBuilder
    0.63
    Bibliograf
    0.59
     Мексичка
    0.58
     Wikimedijinoj
    0.56
     Савезне
    0.56
     transférez
    0.55
     Roskov
    0.53
     otomatig
    0.51
    Activations Density 40.793%

    No Known Activations