© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-2B
    7. Residual Stream - 65k
    9. 20-GEMMASCOPE-RES-65K
    11. 572
    Prev
    Next
    INDEX
    Explanations

    references to customer interactions and purchasing processes

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

    phrases related to customer assistance and service offerings

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

    competition provides

    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
     ,
    -0.85
     )
    -0.69
    <eos>
    -0.65
     ‘
    -0.61
     ),
    -0.60
     .
    -0.60
    rawDesc
    -0.57
     :
    -0.56
     [
    -0.56
     del
    -0.55
    POSITIVE LOGITS
    aarrggbb
    0.75
    Personensuche
    0.71
    expandindo
    0.69
    webElement
    0.66
     ModelExpression
    0.65
    ⓧ
    0.64
    esez
    0.63
     للاسماء
    0.62
    intios
    0.62
    __":
    0.61
    Activations Density 0.001%

    No Known Activations