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. 85398
    Prev
    Next
    INDEX
    Explanations

    references to embassies and diplomatic interactions

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

    diplomatic missions and representatives

    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
     يتيمه
    -0.50
     <>",
    -0.45
     nahilalakip
    -0.43
    MessageTagHelper
    -0.40
    Erreferentziak
    -0.39
    Искәрмәләр
    -0.38
    LayoutConstraint
    -0.37
     Jogos
    -0.37
    ครั้ง
    -0.36
    号線
    -0.36
    POSITIVE LOGITS
     Embassy
    0.63
    Ambassador
    0.62
     Ambassador
    0.57
    ambassade
    0.57
     Diplomatic
    0.56
     diplomatic
    0.56
    Diplom
    0.55
     ambassador
    0.55
     consular
    0.54
     Ambassadors
    0.53
    Activations Density 0.014%

    No Known Activations