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

    Neuronpedia

    APIAssistant AxisNEWCircuit 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. 5197
    Prev
    Next
    INDEX
    Explanations

    references to flags or flagella in various contexts

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

    "flag" or "flagellar"

    np_acts-logits-general · gemini-2.0-flash

    flagellar or flagstone

    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
    \"");
    -0.76
    الإنجليزية
    -0.66
     nahilalakip
    -0.64
     незавершена
    -0.63
    ChildScrollView
    -0.63
    Geplaatst
    -0.62
     Вікіпе
    -0.61
    Hozzáférés
    -0.61
     المعيارى
    -0.61
     ivelany
    -0.59
    POSITIVE LOGITS
     flags
    1.33
     Flags
    1.19
     flag
    1.18
     Flag
    1.15
     FLAG
    1.13
     FLAGS
    1.13
    flags
    1.08
    flag
    1.06
    Flags
    1.04
    FLAGS
    1.01
    Activations Density 0.087%

    No Known Activations