© 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. 5846
    Prev
    Next
    INDEX
    Explanations

    references to popular culture, specifically movies, characters, and various media franchises

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

    popular fictional franchises and entertainment media properties.

    oai_token-act-pair · claude-3-5-haiku-20241022Triggered by @neilrathi
    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
     resourceCulture
    -0.62
    endpush
    -0.61
    transQ
    -0.57
     zdrav
    -0.54
    /**
    -0.54
    rrggbb
    -0.50
    IndentedString
    -0.48
    WithIOException
    -0.48
    greja
    -0.47
     indipendente
    -0.46
    POSITIVE LOGITS
    Fel
    0.37
     characters
    0.37
     Franchise
    0.37
     cele
    0.36
     themed
    0.36
    titleMargin
    0.35
     celebr
    0.35
     Characters
    0.35
    ╗
    0.35
     replica
    0.34
    Activations Density 0.128%

    No Known Activations