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. 117252
    Prev
    Next
    INDEX
    Explanations

    elements related to medical or biological concepts involving transformation, wounds, or healing processes

    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
     محفوظة
    -0.39
     térm
    -0.38
    通販
    -0.36
    transQ
    -0.36
    ulemon
    -0.36
     Pérez
    -0.35
    perty
    -0.35
     empty
    -0.34
     iVar
    -0.34
    empty
    -0.33
    POSITIVE LOGITS
    DockStyle
    0.57
    UnusedPrivate
    0.55
    AnimationsModule
    0.54
    новништво
    0.53
     AssemblyCulture
    0.53
    CloseOperation
    0.52
    findpost
    0.51
    parsedMessage
    0.49
     initComponents
    0.49
    ChildScrollView
    0.48
    Activations Density 0.243%

    No Known Activations