© 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
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B
    4. Residual Stream - 16k
    5. 21-GEMMASCOPE-RES-16K
    6. 10907
    Prev
    Next
    INDEX
    Explanations

    references to Canada in the context of various topics, especially related to media and geography

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.21.hook_resid_post
    Hook Layer
    21
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     riun
    -0.45
    vanju
    -0.45
    bottomRight
    -0.44
     Marlow
    -0.44
     Etat
    -0.43
    tisgarh
    -0.43
     cohé
    -0.42
    fazione
    -0.42
     Popov
    -0.41
    matite
    -0.41
    POSITIVE LOGITS
     Canada
    1.35
     Canadian
    1.32
    Canada
    1.27
     Canadians
    1.25
    Canadian
    1.20
     CANADA
    1.15
    カナダ
    1.11
     Canadá
    1.10
     Kanada
    1.09
    CANADA
    1.09
    Activations Density 0.366%

    No Known Activations