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-2-2B
    4. Residual Stream - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 13819
    Prev
    Next
    INDEX
    Explanations

    functions related to asynchronous programming in JavaScript

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

    Code-related terms followed by "state" or "options"

    np_acts-logits-general · gemini-2.0-flash

    code parameters and variables

    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
     dumne
    -0.46
     însă
    -0.44
    ¬
    -0.43
     musikk
    -0.42
     "");
    
    -0.42
     "\""
    -0.41
     blitt
    -0.41
    newswire
    -0.41
     contactez
    -0.40
     blevet
    -0.40
    POSITIVE LOGITS
    WEBPACK
    0.91
     eslint
    0.90
     disambiguazione
    0.86
     Sequelize
    0.75
    yarnpkg
    0.74
     NSCoder
    0.72
     TypeScript
    0.72
     propOrder
    0.71
     Promise
    0.71
    InjectAttribute
    0.70
    Activations Density 0.292%

    No Known Activations