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. 15505
    Prev
    Next
    INDEX
    Explanations

    structured data formats or code-like syntax

    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
     betweenstory
    -0.71
    sizeCache
    -0.69
    abestanden
    -0.67
     onCreateView
    -0.66
    XmlAccessorType
    -0.66
     ویکی‌پدی
    -0.65
    Бахар
    -0.65
    expandindo
    -0.64
    oredCriteria
    -0.64
    protoimpl
    -0.63
    POSITIVE LOGITS
    )
    0.34
     and
    0.34
     possibles
    0.33
     ninguno
    0.33
    ,
    0.33
     Gesicht
    0.32
    day
    0.32
     days
    0.32
     daarvan
    0.32
    days
    0.31
    Activations Density 0.485%

    No Known Activations