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 - 16k
    5. 11-GEMMASCOPE-RES-16K
    6. 6078
    Prev
    Next
    INDEX
    Explanations

    references to Canada and its cities

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 11-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_11/width_16k/average_l0_80
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.11.hook_resid_post
    Hook Layer
    11
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     strerror
    -0.67
     couche
    -0.67
    ⎨
    -0.65
    y
    -0.63
     Steine
    -0.61
     Harian
    -0.60
     Grub
    -0.57
    エコ
    -0.57
     lei
    -0.57
    oughby
    -0.56
    POSITIVE LOGITS
     Canadians
    1.50
     Canada
    1.45
     Canadá
    1.38
     canad
    1.34
     Canadian
    1.32
    Canada
    1.31
     CANADA
    1.24
     Canad
    1.22
    Canadian
    1.20
    Cana
    1.20
    Activations Density 0.123%

    No Known Activations