© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    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. 101479
    Prev
    Next
    INDEX
    Explanations

    specific formatted references or identifiers related to documents and cases

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

    Numbers or terms in citations that appear in legal documents, academic papers, or technical reports, often between authors/sources and page numbers or dates. These include specific page numbers, section numbers, or reference identifiers.

    eleuther_acts_top20 · claude-3-7-sonnet-20250219Triggered by @lu-christina
    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
     &___
    -0.38
     alternativ
    -0.37
     <=",
    -0.34
     OFDb
    -0.33
    forName
    -0.33
    bewerken
    -0.32
    
    -0.31
     />\
    -0.31
     myſelf
    -0.31
    SharedDtor
    -0.31
    POSITIVE LOGITS
    XmlAccessorType
    0.56
    ismet
    0.56
     ProtoMessage
    0.55
    SharedCtor
    0.55
    Przypisy
    0.55
    setVerticalGroup
    0.52
    ArgumentParser
    0.50
     Administrativna
    0.49
    ContentLoaded
    0.48
     שוליים
    0.48
    Activations Density 0.218%

    No Known Activations