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. 1158
    Prev
    Next
    INDEX
    Explanations

    code or technical terms related to streaming data and API functionalities

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

    references to stream processing and data manipulation in code

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

    file.write, path, cb

    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
    SBATCH
    -0.81
    GraphicsUnit
    -0.67
    DotNetBar
    -0.62
     forehead
    -0.62
    //
    -0.60
     defStyleAttr
    -0.59
    Pyx
    -0.58
     kasarigan
    -0.57
    исленность
    -0.56
    addCriterion
    -0.55
    POSITIVE LOGITS
     End
    0.52
    Escrit
    0.49
    ếp
    0.48
     Buffer
    0.47
     write
    0.47
     Escrit
    0.46
    $_['
    0.46
    end
    0.46
     saada
    0.43
     kaarangay
    0.43
    Activations Density 0.210%

    No Known Activations