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. 825
    Prev
    Next
    INDEX
    Explanations

    sections of code or commands, particularly involving package imports and function closures

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

    occurrences of semicolons

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

    Handbook of, symbols:, disclosed in

    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
     بيها
    -0.49
    ľov
    -0.48
     &
    -0.44
    </thead>
    -0.43
    spf
    -0.42
     تخت
    -0.42
    qtype
    -0.42
    usercontent
    -0.42
    UTC
    -0.42
    μως
    -0.41
    POSITIVE LOGITS
    ValueStyle
    0.73
    tagHelperRunner
    0.71
    __":
    
    0.69
     незавершена
    0.67
    dafx
    0.64
    __;
    0.63
     mouseY
    0.62
    ();
    
    
    0.61
    ceptre
    0.60
    __":
    0.60
    Activations Density 0.012%

    No Known Activations