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 - 16k
    5. 21-GEMMASCOPE-RES-16K
    6. 15328
    Prev
    Next
    INDEX
    Explanations

    programming constructs related to object-oriented classes and methods

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

    programming/code snippets, especially identifiers and syntax from class definitions, namespaces, annotations, and build/config directives.

    oai_token-act-pair · gpt-5Triggered by @arielgoldsteinlab
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.21.hook_resid_post
    Hook Layer
    21
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    MigrationBuilder
    -0.82
    IsMutable
    -0.74
     protoimpl
    -0.72
    UseVisualStyle
    -0.66
    IUrlHelper
    -0.62
     invokingState
    -0.61
    CppMethod
    -0.60
     PeEnEo
    -0.59
    Архівовано
    -0.59
     isInitialized
    -0.59
    POSITIVE LOGITS
    <eos>
    0.82
    bewerken
    0.53
     podjela
    0.50
    crí
    0.38
     particuliers
    0.38
    transQ
    0.37
    ecuted
    0.36
     Obras
    0.36
     ویکی‌آمباردا
    0.36
     ADVICE
    0.36
    Activations Density 0.359%

    No Known Activations