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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B
    4. Residual Stream - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 9001
    Prev
    Next
    INDEX
    Explanations

    terms related to different types of bases or frameworks that structure various subjects or phenomena

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_20/width_16k/average_l0_68
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    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
     Evangelio
    -0.34
     posesión
    -0.34
    Selección
    -0.34
     Moscú
    -0.33
     juez
    -0.33
     TAMBIÉN
    -0.32
     keputusan
    -0.32
     juges
    -0.32
     cooperación
    -0.31
     kebiasaan
    -0.31
    POSITIVE LOGITS
    ніципалі
    0.73
    onaceous
    0.68
     lenker
    0.66
    IContainer
    0.64
     autorytatywna
    0.63
    iented
    0.62
    irited
    0.61
    webElementXpaths
    0.61
    tvguidetime
    0.61
    splitlines
    0.60
    Activations Density 0.400%

    No Known Activations