© 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. 20-GEMMASCOPE-RES-16K
    6. 15565
    Prev
    Next
    INDEX
    Explanations

    references to political campaigns and elections, particularly concerning reelections and candidacies

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_20/width_16k/average_l0_68
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    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
     typelib
    -0.46
    ớm
    -0.43
     Réponses
    -0.41
    endcsname
    -0.40
     breadcrumbs
    -0.39
     Obrador
    -0.39
     الحره
    -0.39
     verminder
    -0.39
    Interactive
    -0.38
     Interactive
    -0.37
    POSITIVE LOGITS
     rematch
    0.50
     propOrder
    0.46
     renewal
    0.46
    SharedCtor
    0.45
    again
    0.43
     CanadaChoose
    0.42
     retorno
    0.42
     renewing
    0.42
    原始内容存档于
    0.41
     erneut
    0.41
    Activations Density 0.175%

    No Known Activations