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-2-9B-IT
    4. Residual Stream - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 45768
    Prev
    Next
    INDEX
    Explanations

    references to developed or industrialized countries

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

    developed countries

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    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
     Barbier
    -0.53
    Alexandre
    -0.50
     affectionately
    -0.48
    #
    -0.46
     downstairs
    -0.46
     resourceCulture
    -0.45
    Nickname
    -0.45
     Améli
    -0.44
     fondly
    -0.44
     himself
    -0.44
    POSITIVE LOGITS
    bootstrapcdn
    0.61
     countries
    0.56
    发达
    0.52
    OECD
    0.51
     OECD
    0.49
     Länder
    0.49
     Bundesrepublik
    0.47
     Ländern
    0.45
     Countries
    0.44
     AssemblyCulture
    0.44
    Activations Density 0.013%

    No Known Activations