Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B
    4. Residual Stream - 16k
    5. 21-GEMMASCOPE-RES-16K
    6. 15751
    Prev
    Next
    INDEX
    Explanations

    technical code-related terms and programming constructs

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.21.hook_resid_post
    Hook Layer
    21
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    -0.66
    mathrm
    -0.59
    ’
    -0.56
    '
    -0.51
    begin
    -0.51
     
    
    -0.42
     
    -0.41
    <tbody>
    -0.39
    _
    
    -0.35
     colspan
    -0.34
    POSITIVE LOGITS
     Савезне
    1.42
     ویکی‌پدی
    1.23
     Мексичка
    1.18
    GEBURTSDATUM
    1.17
    Personensuche
    1.11
    SourceChecksum
    1.10
    TagMode
    1.09
    تقاوى
    1.06
    tagHelperRunner
    1.05
    RegressionTest
    1.01
    Activations Density 0.024%

    No Known Activations