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. 69378
    Prev
    Next
    INDEX
    Explanations

    class declarations and their relationships in programming code

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    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
     חיצוניים
    -0.56
    期刊论文
    -0.56
     NSCoder
    -0.55
    IntoConstraints
    -0.52
     AppCompatTheme
    -0.52
    queryInterface
    -0.51
     disambiguazione
    -0.50
     JpaRepository
    -0.50
     Signalez
    -0.49
    原始内容存档于
    -0.49
    POSITIVE LOGITS
     Inner
    0.49
    typeDefinition
    0.47
    Inner
    0.44
     inner
    0.44
    localctx
    0.44
    inner
    0.44
     '\\;'
    0.42
    unmodifiable
    0.40
    内
    0.39
     Internal
    0.39
    Activations Density 0.018%

    No Known Activations