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. 1013
    Prev
    Next
    INDEX
    Explanations

    variables and parameters in code or programming contexts

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

    links or references to other resources

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

    mathematical expressions and code

    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
    DebuggerNonUser
    -0.69
    RenderAtEndOf
    -0.67
    
    -0.67
    TagMode
    -0.63
    PerformLayout
    -0.63
     JpaRepository
    -0.58
    WaitGroup
    -0.57
     smtplib
    -0.56
     subtree
    -0.54
    ItemBackground
    -0.54
    POSITIVE LOGITS
    awtextra
    0.62
    OGND
    0.59
    ednesdays
    0.55
    rawDesc
    0.55
     مرئيه
    0.53
    stalline
    0.53
    ourdes
    0.52
     olev
    0.52
    NUMX
    0.51
    renar
    0.51
    Activations Density 0.004%

    No Known Activations