© 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-IT
    4. Residual Stream - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 13577
    Prev
    Next
    INDEX
    Explanations

    code-related formatting commands and documentation tags

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

    LaTeX commands

    np_acts-logits-general · gemini-2.0-flash

    LaTeX `\notag` and `\\`

    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
    GraphicsUnit
    -0.57
     propOrder
    -0.56
     queſta
    -0.56
    новништво
    -0.54
    tanleria
    -0.51
    pgsql
    -0.51
    afone
    -0.51
    setopt
    -0.51
     ſind
    -0.51
    ORAGE
    -0.51
    POSITIVE LOGITS
    maketitle
    1.84
    inheritDoc
    0.51
     title
    0.47
    ładka
    0.45
    inheritdoc
    0.44
    Espèce
    0.43
    title
    0.42
     Monteiro
    0.42
     Titel
    0.42
     Scherer
    0.42
    Activations Density 0.010%

    No Known Activations