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. 14468
    Prev
    Next
    INDEX
    Explanations

    specific functions related to webpage interaction or DOM manipulation

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    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
    routeProvider
    -0.59
    usermodel
    -0.55
    exitRule
    -0.55
    stateProvider
    -0.54
    jooq
    -0.51
    ThroughAttribute
    -0.50
    fillType
    -0.50
    RegistryLite
    -0.50
    BorderFactory
    -0.49
    SharedCtor
    -0.49
    POSITIVE LOGITS
     līdz
    0.36
     Hayward
    0.34
     بيها
    0.32
    iora
    0.32
    自
    0.32
     Dow
    0.31
     Laura
    0.31
    PARATUS
    0.31
     hand
    0.31
    forChild
    0.31
    Activations Density 0.215%

    No Known Activations