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-2-2B
    4. Residual Stream - 65k
    5. 20-GEMMASCOPE-RES-65K
    6. 921
    Prev
    Next
    INDEX
    Explanations

    repetitive questioning or confusion about identity or events

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

    the phrase "What" as a question or inquiry

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

    questions starting with what

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 20-gemmascope-res-65k
    Configuration
    google/gemma-scope-2b-pt-res/layer_20/width_65k/average_l0_114
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    65,536
    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
    DockStyle
    -1.02
     AssemblyTitle
    -0.98
    RegistryLite
    -0.92
     EconPapers
    -0.91
    MigrationBuilder
    -0.90
     ExecuteAsync
    -0.90
     <",
    -0.88
     <=",
    -0.87
    AnchorStyles
    -0.87
    ();)
    -0.86
    POSITIVE LOGITS
    —
    0.54
     do
    0.51
     about
    0.51
    !
    0.49
    ,
    0.48
    ..
    0.44
    urlpatterns
    0.43
     now
    0.43
     second
    0.43
    now
    0.43
    Activations Density 0.026%

    No Known Activations