© 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
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B-IT
    4. Residual Stream - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 10760
    Prev
    Next
    INDEX
    Explanations

    occurrences of the word "palette" and related terms in a technical context

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

    TeX math commands

    np_acts-logits-general · gemini-2.0-flash

    \protect\mathpalette

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91
    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
    ########.
    -0.49
     declare
    -0.37
     boy
    -0.37
    baomidou
    -0.37
    KommentareTeilen
    -0.36
     declares
    -0.35
     Jau
    -0.35
     Ris
    -0.35
     BufferedImage
    -0.34
     boto
    -0.34
    POSITIVE LOGITS
    spath
    1.22
    webElementXpaths
    0.85
    GEBURTSDATUM
    0.82
    AccessorTable
    0.81
     AssemblyVersion
    0.81
     AssemblyCulture
    0.71
     ProtoMessage
    0.69
    IndentedString
    0.69
    ArrowToggle
    0.68
     ویکی‌پدی
    0.67
    Activations Density 0.007%

    No Known Activations