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. 30474
    Prev
    Next
    INDEX
    Explanations

    references to names and acronyms associated with film and production-related terms

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

    soldiers stationed

    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
    webElementGuid
    -0.54
    ruptedException
    -0.54
    السكان
    -0.51
     čierna
    -0.47
    apeake
    -0.44
    ništ
    -0.43
    الدراسه
    -0.43
    ulemon
    -0.42
     estadounidenses
    -0.42
     dew
    -0.41
    POSITIVE LOGITS
    ########.
    0.50
    scriptcase
    0.43
    ̈́
    0.39
     övers
    0.38
     German
    0.37
     ویکی‌پدی
    0.37
     translated
    0.37
    ্ত
    0.37
     CanadaChoose
    0.37
     European
    0.36
    Activations Density 0.131%

    No Known Activations