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. 9447
    Prev
    Next
    INDEX
    Explanations

    instances of comment and directive syntax in coding languages

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

    C preprocessor directives

    np_acts-logits-general · gemini-2.0-flash

    code directives and comments

    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
    
    -0.77
    PMailer
    -0.75
    @[+][
    -0.73
    Посилання
    -0.68
    queryInterface
    -0.68
     poffible
    -0.66
    ]")]
    -0.66
     zitate
    -0.65
     ThemeData
    -0.65
     Efq
    -0.65
    POSITIVE LOGITS
    endregion
    0.63
    region
    0.55
    line
    0.52
     Beyond
    0.48
    guli
    0.46
    gra
    0.45
     y
    0.43
    Region
    0.43
     Mer
    0.43
    opf
    0.42
    Activations Density 0.062%

    No Known Activations