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
    4. Residual Stream - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 73616
    Prev
    Next
    INDEX
    Explanations

    HTML anchor tags or links in the text

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-9b-pt-res/layer_20/width_131k/average_l0_114
    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
    الحياه
    -0.59
    MemoryWarning
    -0.56
    tagHelperRunner
    -0.55
    principalColumn
    -0.54
    TypedDataSet
    -0.51
    sizeCache
    -0.51
    FormTagHelper
    -0.51
    bindingNavigator
    -0.50
     indígen
    -0.49
    лтемелер
    -0.49
    POSITIVE LOGITS
     uri
    0.42
    Архівовано
    0.40
    OGND
    0.39
     href
    0.39
    frac
    0.39
    nomin
    0.37
    sendRedirect
    0.37
    AnchorTagHelper
    0.36
    로그
    0.35
     stump
    0.35
    Activations Density 0.005%

    No Known Activations