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. 4616
    Prev
    Next
    INDEX
    Explanations

    specific references to websites and their attributes

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

    Wikipedia and wiki-related content

    np_acts-logits-general · gemini-2.0-flash

    Wikipedia links

    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
    indd
    -0.65
    UpInside
    -0.61
    PointerException
    -0.58
     سكانية
    -0.56
    stateParams
    -0.55
    matchCondition
    -0.55
    awtextra
    -0.55
    IBOutlet
    -0.54
    ThemeOverlay
    -0.53
     ModelExpression
    -0.52
    POSITIVE LOGITS
     Wikipedia
    1.45
     wiki
    1.39
     Wiki
    1.39
    wiki
    1.30
    Wiki
    1.22
    Wikipedia
    1.17
     wikipedia
    1.16
    wik
    1.05
    wikipedia
    0.98
     wik
    0.96
    Activations Density 0.412%

    No Known Activations