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. 853
    Prev
    Next
    INDEX
    Explanations

    date references

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

    dates, specifically months and their respective days

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

    dates of birth

    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
    XmlAccessType
    -0.67
    }]{
    -0.64
     surla
    -0.57
    fony
    -0.56
    Trayectoria
    -0.56
    таратура
    -0.55
    RectangleBorder
    -0.54
    ﹏
    -0.53
    ducers
    -0.52
    bija
    -0.52
    POSITIVE LOGITS
     közel
    0.55
     ExecuteAsync
    0.52
    chọn
    0.49
    生まれ
    0.48
     myö
    0.48
    HomeAsUpEnabled
    0.47
    rolla
    0.46
    éraires
    0.46
     nonatomic
    0.45
     gyerme
    0.44
    Activations Density 0.029%

    No Known Activations