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. 15718
    Prev
    Next
    INDEX
    Explanations

    specific terms and phrases related to interactions and characteristics of sports, biology, and medical testing

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    specialized, domain-specific nouns and jargon terms, particularly in technical or set-phrase collocations.

    oai_token-act-pair · gpt-5Triggered by @arielgoldsteinlab
    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
     the
    -0.23
     a
    -0.21
     colspan
    -0.21
     zdan
    -0.21
     I
    -0.21
     off
    -0.20
     on
    -0.19
    ↵
    -0.19
     it
    -0.18
     with
    -0.18
    POSITIVE LOGITS
    StructEnd
    1.02
    Diwedd
    1.01
    
    0.98
    featureID
    0.96
    +#+#
    0.94
    pushFollow
    0.94
     ویکی‌پدی
    0.93
    MessageOf
    0.93
     مرئيه
    0.92
     lenker
    0.88
    Activations Density 3.284%

    No Known Activations