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. 1081
    Prev
    Next
    INDEX
    Explanations

    numerical representations or code snippets

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

    programming constructs or boolean logic-related terms

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

    ceylon spinach

    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
    IGraphics
    -0.56
    libft
    -0.56
    medriver
    -0.54
    ciach
    -0.51
     AppComponent
    -0.51
     chi̍t
    -0.49
    Референце
    -0.48
     airs
    -0.48
    {§
    -0.47
    //:
    -0.46
    POSITIVE LOGITS
    DebuggerNonUser
    0.64
    الدراسه
    0.57
     esboço
    0.56
    IndentedString
    0.55
    OGND
    0.54
     AssemblyProduct
    0.53
     Italijani
    0.52
     GetEnumerator
    0.52
     Commencez
    0.52
     Alamo
    0.50
    Activations Density 0.009%

    No Known Activations