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. 15619
    Prev
    Next
    INDEX
    Explanations

    phrases related to cleaning products and their effectiveness

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

    Names of detergents or cleaning agents

    np_acts-logits-general · gemini-2.0-flash

    detergents and soap

    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
     Gat
    -0.52
    Gat
    -0.46
     lạc
    -0.45
    akana
    -0.45
    ArrowToggle
    -0.43
     leader
    -0.42
     vite
    -0.41
     Stray
    -0.41
    culas
    -0.41
    ort
    -0.41
    POSITIVE LOGITS
     soap
    1.42
     soaps
    1.30
     detergent
    1.29
     Soap
    1.24
     detergents
    1.22
     shampoo
    1.19
    soap
    1.19
     sabun
    1.14
    Soap
    1.11
     savon
    1.03
    Activations Density 0.174%

    No Known Activations