© 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. 2-GEMMASCOPE-RES-16K
    6. 4251
    Prev
    Next
    INDEX
    Explanations

    the presence of the beginning of ordered sequences in the text

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 2-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_2/width_16k/average_l0_67
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.2.hook_resid_post
    Hook Layer
    2
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    -0.48
    '
    -0.34
     […]
    -0.32
    <unused62>
    -0.31
    ’
    -0.28
    mathrm
    -0.28
    <unused61>
    -0.26
    <tbody>
    -0.26
    [toxicity=0]
    -0.23
    begin
    -0.22
    POSITIVE LOGITS
     Савезне
    1.28
    Personendaten
    1.21
    SharedCtor
    1.16
     ویکی‌پدی
    1.14
    TagMode
    1.10
     Normdatei
    1.09
    RegressionTest
    1.05
    SourceChecksum
    1.05
     мәкал
    1.05
    setVerticalGroup
    1.03
    Activations Density 0.006%

    No Known Activations