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

    Neuronpedia

    APIAssistant AxisNEWCircuit 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 - 16k
    5. 21-GEMMASCOPE-RES-16K
    6. 15592
    Prev
    Next
    INDEX
    Explanations

    programming language constructs and syntax related to the definition and manipulation of data structures

    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
    rungsseite
    -0.69
     للاسماء
    -0.69
     otomatig
    -0.67
    LEncoder
    -0.64
     ویکی‌پدی
    -0.64
    
    -0.60
    cshtml
    -0.59
    parsedMessage
    -0.59
    最快更新
    -0.53
    TemporalType
    -0.52
    POSITIVE LOGITS
     solution
    0.41
     oplossing
    0.36
     solve
    0.36
     solutions
    0.35
     Solution
    0.34
     çöz
    0.33
     approach
    0.33
    אים
    0.32
     method
    0.32
     soluzione
    0.30
    Activations Density 0.168%

    No Known Activations