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. 18-GEMMASCOPE-RES-16K
    6. 578
    Prev
    Next
    INDEX
    Explanations

    code syntax elements and structural definitions in programming contexts

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

    C++ code or similar code snippets

    np_acts-logits-general · gemini-2.0-flash

    special characters and code structure

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.18.hook_resid_post
    Hook Layer
    18
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    TagMode
    -0.92
    MLLoader
    -0.90
     ModelExpression
    -0.79
    ScopeManager
    -0.78
     ExecuteAsync
    -0.76
     HttpServlet
    -0.75
    Portály
    -0.75
     joaat
    -0.74
    帖最后由
    -0.73
     BoxFit
    -0.72
    POSITIVE LOGITS
    ↵
    0.53
    <eos>
    0.51
    ёз
    0.49
    ↵↵
    0.49
     (
    0.48
     
    0.47
     capital
    0.46
    stddef
    0.42
     [
    0.42
    mainAxisSize
    0.41
    Activations Density 0.063%

    No Known Activations