© 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
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-9B
    7. Residual Stream - 16k
    9. 30-GEMMASCOPE-RES-16K
    11. 10777
    Prev
    Next
    INDEX
    Explanations

    references to documents and license files

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

    Tokens found inside parentheses

    np_acts-logits-general · gemini-2.0-flash

    parentheses and code"},"file extensions, symbols, and code snippets"},"code, spreadsheets, and data formats"},"code, file types, and data"},"code, file extensions, and symbols"},"numeric identifiers and symbols"},"code snippets and symbols"},"code, extensions, and symbols"},"code, file types, and punctuation"},"code, spreadsheets, and punctuation"},"code, numeric data, and symbols"},"code, file names, and punctuation"},"code, file extensions, and punctuation"},"code, data entries, and symbols"},"code, version numbers, and symbols"},"code, numbers, and punctuation"},"code","numbers and symbols"},"code, punctuation, and data labels"},"code and data"},"code","code","code"},"code"},"code"},"code"},"code"},"code","code"},"code"},"code"},"code"},"code"},"code"},"code"},"code, spreadsheets, and data"},"code, file formats, and data"},"code, file names, and data"},"code, numbers, and file extensions"},"code, numbers, and file types"},"code, financial data, and symbols"},"code, data formats, and symbols"},"code, numbers, and spreadsheets"},"code, numbers, and file extensions"},"code, numbers, and file types"},"code, spreadsheets, and symbols"},"code, data, and punctuation"},"code, numbers, and punctuation"},"code, file extensions, and punctuation"},},"code, file extensions, and symbols"}

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 30-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_30/width_16k/average_l0_120
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.30.hook_resid_post
    Hook Layer
    30
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    rungsseite
    -0.77
    */].
    -0.51
     Audiodateien
    -0.50
    UserDTO
    -0.49
    ↀ
    -0.47
    Evid
    -0.47
    OGRAPH
    -0.45
     kohdetta
    -0.45
     pagină
    -0.44
     ]]
    -0.44
    POSITIVE LOGITS
    Personendaten
    0.59
    ValueStyle
    0.52
    )
    0.43
    ?)
    0.42
    oneofs
    0.42
    )?
    0.40
     ...)
    0.39
    UnusedPrivate
    0.39
    ):
    0.38
    +)
    0.38
    Activations Density 2.313%

    No Known Activations