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 - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 1024
    Prev
    Next
    INDEX
    Explanations

    specific roles or titles related to legal and healthcare professions

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

    Types of professional practices

    np_acts-logits-general · gemini-2.0-flash

    law, medicine, therapy, veterinary

    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-16k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91
    Prompts (Dashboard)
    24,576 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
     lenker
    -0.64
    lusconi
    -0.60
    createTextNode
    -0.56
    adori
    -0.56
    quiao
    -0.55
    chymal
    -0.54
    bihan
    -0.54
     handleMessage
    -0.54
    AnchorStyles
    -0.54
    featureID
    -0.53
    POSITIVE LOGITS
     parson
    0.37
    ViewImports
    0.31
     practice
    0.30
     practicing
    0.30
    BarStyle
    0.29
     Constitución
    0.28
    Jeografia
    0.27
    UserScript
    0.27
     technician
    0.26
     ligiloj
    0.26
    Activations Density 0.316%

    No Known Activations