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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-9B
    7. Residual Stream - 16k
    9. 20-GEMMASCOPE-RES-16K
    11. 11975
    Prev
    Next
    INDEX
    Explanations

    mathematical notation and symbols related to set theory and functions

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

    technical syntax from programming and LaTeX-style mathematical notation, especially structured code tokens, operators, indices, brackets, and numeric/formula markup.

    oai_token-act-pair · gpt-5Triggered by @jyhe0408

    structural or formatting tokens in technical code and mathematical notation, particularly brackets, braces, and delimiters.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @jyhe0408
    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
    름
    -0.26
     r
    -0.24
    انتهای
    -0.24
    株式会社
    -0.24
     endTime
    -0.23
    也会
    -0.23
     squ
    -0.23
     arra
    -0.23
     Gug
    -0.23
     disp
    -0.23
    POSITIVE LOGITS
     '\\;'
    0.95
    RenderAtEndOf
    0.77
     AssemblyCulture
    0.75
    WriteTagHelper
    0.74
    parsedMessage
    0.73
    principalTable
    0.72
    AndEndTag
    0.72
    enderror
    0.70
     utafitiHapana
    0.64
    DockStyle
    0.64
    Activations Density 0.808%

    No Known Activations