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 - 65k
    5. 20-GEMMASCOPE-RES-65K
    6. 1167
    Prev
    Next
    INDEX
    Explanations

    programming language syntax related to function definitions and import statements

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

    programming constructs related to function definitions and imports in code

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

    case; src="/; The S

    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-65k
    Configuration
    google/gemma-scope-2b-pt-res/layer_20/width_65k/average_l0_114
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    65,536
    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
    OGND
    -0.96
    TagMode
    -0.70
    __":
    
    -0.69
    AndEndTag
    -0.69
    HtmlAttribute
    -0.69
    RectangleBorder
    -0.66
    BeginContext
    -0.66
    __':
    
    -0.63
    ^(@)
    -0.63
     }}"></
    -0.63
    POSITIVE LOGITS
     change
    0.52
    change
    0.45
     actions
    0.45
    actions
    0.44
     decrease
    0.42
    hy
    0.41
     cambio
    0.41
     set
    0.41
     increase
    0.40
     CHANGE
    0.40
    Activations Density 0.011%

    No Known Activations