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. 24486
    Prev
    Next
    INDEX
    Explanations

    references to colors and their descriptions

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

    colors and appearance

    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
     <=",
    -0.53
     crossorigin
    -0.49
     <>",
    -0.49
    
    -0.43
     تضيفلها
    -0.43
     LIGHTS
    -0.43
    niająca
    -0.42
    didSet
    -0.40
    rungsseite
    -0.40
    postsleuth
    -0.40
    POSITIVE LOGITS
     בארץ
    0.40
    ynos
    0.39
    PreferredItem
    0.39
     agricoles
    0.38
     verze
    0.38
     marino
    0.38
    FormTagHelper
    0.37
     AssemblyCompany
    0.36
    loh
    0.36
     secundarios
    0.35
    Activations Density 0.019%

    No Known Activations