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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B
    4. Residual Stream - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 70680
    Prev
    Next
    INDEX
    Explanations

    curly braces and extensions in code structures

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-9b-pt-res/layer_20/width_131k/average_l0_114
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    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
     intptr
    -0.59
    ẖ
    -0.50
    MemoryWarning
    -0.49
    UserScript
    -0.48
    İstinadlar
    -0.48
     ControllerBase
    -0.48
     utafitiHapana
    -0.47
    baric
    -0.47
     HttpNotFound
    -0.46
     ModelExpression
    -0.46
    POSITIVE LOGITS
    .*")]
    0.47
    Rhestr
    0.45
     Umum
    0.41
     Geraadpleegd
    0.41
     <<<<<<<<<<<<<<
    0.40
    AutoresizingMask
    0.40
    ėmis
    0.39
     GENERATED
    0.39
     kaldı
    0.38
    </thead>
    0.38
    Activations Density 0.008%

    No Known Activations