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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-2B
    4. Residual Stream - 16k
    5. 4-GEMMASCOPE-RES-16K
    6. 3541
    Prev
    Next
    INDEX
    Explanations

    specific identifiers or labels that may relate to sequences or coding systems in scientific contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 4-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_4/width_16k/average_l0_124
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.4.hook_resid_post
    Hook Layer
    4
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     linkovi
    -0.72
     Wikimedijinoj
    -0.66
    Portale
    -0.63
    extAlignment
    -0.60
    DockStyle
    -0.59
    TestTools
    -0.58
     Wikispecies
    -0.54
    NameInMap
    -0.54
    addContainerGap
    -0.52
    localctx
    -0.51
    POSITIVE LOGITS
    HideFlags
    0.63
    клопе
    0.57
    PerformLayout
    0.50
    onViewCreated
    0.50
    /#{
    0.50
    ulho
    0.48
    ContextHolder
    0.48
    ⋙
    0.47
    mejores
    0.47
     doa
    0.47
    Activations Density 3.434%

    No Known Activations