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 - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 4860
    Prev
    Next
    INDEX
    Explanations

    java event listener syntax in code snippets

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

    Opens and closes parentheses

    np_acts-logits-general · gemini-2.0-flash

    regulated genes, atsu drama

    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-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    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
     الدولى
    -0.81
     newBuilder
    -0.68
     ActivatedRoute
    -0.66
     jadx
    -0.63
    GeneratedValue
    -0.63
    ngOn
    -0.63
    Numerology
    -0.62
    خصة
    -0.61
     RouterModule
    -0.60
    LEncoder
    -0.58
    POSITIVE LOGITS
    this
    0.50
    (_,
    0.46
     ciudadana
    0.45
    aimerais
    0.45
     corporativa
    0.45
    lambda
    0.44
    function
    0.44
    ولايات
    0.43
     Handle
    0.41
    Handle
    0.41
    Activations Density 0.057%

    No Known Activations