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. 1022
    Prev
    Next
    INDEX
    Explanations

    patterns associated with image processing and data handling in programming

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

    terms related to image processing and photo data structures

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

    code closing delimiters

    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
     autorytatywna
    -0.55
    \{\\
    -0.49
    RDF
    -0.49
    涯
    -0.47
    <bos>
    -0.46
     JOY
    -0.45
    ctice
    -0.45
    afy
    -0.44
    JOY
    -0.44
    ંગ
    -0.44
    POSITIVE LOGITS
    setVerticalGroup
    0.82
    tagHelperRunner
    0.79
    featureID
    0.76
    ConstraintMaker
    0.73
    ########.
    0.72
     AspNetCore
    0.68
    ContentAsync
    0.68
    تفصیلات
    0.63
     vPvB
    0.63
    CodedInputStream
    0.62
    Activations Density 0.038%

    No Known Activations