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. 77201
    Prev
    Next
    INDEX
    Explanations

    specific technical terms and numerical values related to measurements or constructs in engineering and science

    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
    complexContent
    -0.56
    帖最后由
    -0.54
    ьаж
    -0.52
     CreateTagHelper
    -0.52
    ItemBackground
    -0.52
    SequentialGroup
    -0.50
    AndroidJUnit
    -0.50
    GIVEREF
    -0.49
    openzeppelin
    -0.49
     <>",
    -0.48
    POSITIVE LOGITS
    CloseOperation
    0.59
    awtextra
    0.45
    parsedMessage
    0.44
    findpost
    0.43
    ⓧ
    0.36
     Walkover
    0.35
    [--
    0.32
     laſt
    0.32
    ArrowToggle
    0.31
    ContentValues
    0.30
    Activations Density 0.717%

    No Known Activations