Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit 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. 16-GEMMASCOPE-RES-16K
    6. 11942
    Prev
    Next
    INDEX
    Explanations

    keywords related to structured performance assessment factors and their statistical implications

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

    after specific word fragments

    np_acts-logits-general · gemini-2.0-flash

    citizenship law childbirth

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 16-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_16/width_16k/average_l0_78
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.16.hook_resid_post
    Hook Layer
    16
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     تعدى
    -0.68
    .*")]
    -0.68
    المراجع
    -0.63
    
    -0.60
    expandindo
    -0.59
    ﹍﹍﹍
    -0.59
    useNewUrlParser
    -0.59
    LikeLiked
    -0.58
     AssemblyCompany
    -0.55
     Перейти
    -0.55
    POSITIVE LOGITS
    󠁿
    1.37
    }{*}{}
    0.78
    μβρίου
    0.75
    かわらず
    0.74
    UVWXYZ
    0.73
     سكانية
    0.67
    __*/
    0.67
    Personensuche
    0.64
     contextLoads
    0.63
    StoryboardSegue
    0.61
    Activations Density 10.028%

    No Known Activations