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-2-2B
    4. Residual Stream - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 14559
    Prev
    Next
    INDEX
    Explanations

    references to enterprise systems or tools

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

    "Enterprise" or "ring"

    np_acts-logits-general · gemini-2.0-flash

    ring, enterprise, although

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71
    Prompts (Dashboard)
    36,864 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
     للمعارف
    -0.81
    樣
    -0.65
    httphttps
    -0.63
    maxn
    -0.61
     tár
    -0.61
    IUrlHelper
    -0.60
    ֑
    -0.60
    خصة
    -0.60
     pobres
    -0.59
    الإنجليزية
    -0.59
    POSITIVE LOGITS
    Controllo
    0.69
    WebServlet
    0.68
    submitted
    0.65
    lipop
    0.63
    Besos
    0.62
    ollectionView
    0.60
    takings
    0.60
    jména
    0.59
     henvisninger
    0.59
     Finlay
    0.59
    Activations Density 0.847%

    No Known Activations