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. 75298
    Prev
    Next
    INDEX
    Explanations

    historical names and entities related to legal and political contexts

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

    Zola Tongo or Wilkes

    np_acts-logits-general · gemini-2.5-flash-lite
    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
     straw
    -0.42
    ꜣ
    -0.40
     BoxFit
    -0.40
     Industry
    -0.39
     crown
    -0.38
     Victory
    -0.36
     Standards
    -0.36
    Qualifier
    -0.36
    mlı
    -0.36
     maquinaria
    -0.36
    POSITIVE LOGITS
    RRect
    0.57
    PreExecute
    0.54
    __*/
    0.50
    󠁢
    0.49
    Хьажоргаш
    0.45
    ViewFeatures
    0.43
     OrderedDict
    0.42
    middels
    0.42
     mCurrent
    0.42
     JAXBElement
    0.42
    Activations Density 0.077%

    No Known Activations