© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    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. 7068
    Prev
    Next
    INDEX
    Explanations

    conditional statements in programming or logic contexts

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

    Immediately after periods, parentheses, or hyphens

    np_acts-logits-general · gemini-2.0-flash

    mathematical and programming contexts

    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
    ↵↵
    -0.50
    ↵
    -0.48
    рус
    -0.43
    tetten
    -0.42
    CodedInputStream
    -0.41
     proces
    -0.41
    <bos>
    -0.41
    tella
    -0.41
    brun
    -0.40
     des
    -0.39
    POSITIVE LOGITS
    InjectAttribute
    1.52
    IUrlHelper
    1.33
    ViewFeatures
    1.20
     ModelExpression
    0.96
     useAppContext
    0.95
     ویکی‌پدیا
    0.89
     فريبيس
    0.86
    UnsafeEnabled
    0.83
     '\\;'
    0.82
    SourceChecksum
    0.80
    Activations Density 0.022%

    No Known Activations