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-2-9B-IT
    4. Residual Stream - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 2474
    Prev
    Next
    INDEX
    Explanations

    concepts related to health and wellness, particularly nutrition and drug awareness

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

    healthy food and nutrition

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    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
    ChildScrollView
    -0.82
     BrowserModule
    -0.68
     erſt
    -0.67
     OFDb
    -0.67
    Hochspringen
    -0.65
    iſchen
    -0.65
    ">//
    -0.63
     vooz
    -0.61
    UnsafeEnabled
    -0.61
     dieſe
    -0.60
    POSITIVE LOGITS
     awareness
    0.48
     Awareness
    0.40
     encourage
    0.36
     message
    0.36
     cyber
    0.35
     education
    0.35
     attitudes
    0.34
     counter
    0.33
    awareness
    0.33
     promoting
    0.33
    Activations Density 0.017%

    No Known Activations