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. 2458
    Prev
    Next
    INDEX
    Explanations

    references to different types of flooring and associated maintenance tasks

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

    floor and flooring material

    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
    selig
    -0.47
    量
    -0.44
     Letters
    -0.43
    prepareStatement
    -0.43
    Stateless
    -0.43
     Complex
    -0.42
     ChromeDriver
    -0.42
    JsonObject
    -0.42
    XmlRootElement
    -0.42
     '+':
    -0.40
    POSITIVE LOGITS
     flooring
    0.93
     floors
    0.90
     floor
    0.86
     vloer
    0.69
     Floors
    0.69
    floor
    0.68
     FLOOR
    0.67
    floors
    0.67
    Floor
    0.66
     Flooring
    0.65
    Activations Density 0.008%

    No Known Activations