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-2-9B-IT
    4. Residual Stream - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 5807
    Prev
    Next
    INDEX
    Explanations

    key terms related to health insurance and the healthcare system in America

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

    health insurance and healthcare

    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
    ỡng
    -0.42
    новништво
    -0.39
    odend
    -0.37
    twór
    -0.36
    못
    -0.35
    tură
    -0.35
    HomeAsUpEnabled
    -0.35
    SequentialGroup
    -0.35
    antd
    -0.34
     Frieden
    -0.34
    POSITIVE LOGITS
     insurance
    0.60
     houſe
    0.59
     myſelf
    0.57
     pleaſure
    0.56
     themſelves
    0.55
     AppCompatTheme
    0.54
    Personensuche
    0.54
     ſche
    0.53
     healthcare
    0.52
    Insurance
    0.52
    Activations Density 0.014%

    No Known Activations