Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & 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. 11814
    Prev
    Next
    INDEX
    Explanations

    functional programming constructs and data structure manipulations.

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

    specific code patterns and structures in a programming context.

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot

    programming language syntax elements and structures, particularly in C++, Java, and similar languages. It activates on things like function parameters, template declarations, class methods, and code block structures.

    oai_token-act-pair · claude-3-5-sonnet-20240620Triggered by @bot

    Code snippets with parentheses and semicolons

    np_acts-logits-general · gemini-2.0-flash

    programming method calls

    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
    <eos>
    -0.53
     pun
    -0.47
     har
    -0.47
    webElementXpaths
    -0.46
     om
    -0.44
    ">//
    -0.44
    WebServlet
    -0.43
     dos
    -0.43
    ազմ
    -0.43
     구
    -0.41
    POSITIVE LOGITS
    ↵
    0.99
     Efq
    0.85
     فريبيس
    0.84
     myſelf
    0.82
    .”)
    0.80
     Theſe
    0.74
     Jefus
    0.73
    %");
    0.73
     Monfieur
    0.72
     becauſe
    0.72
    Activations Density 0.238%

    No Known Activations