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-2-2B
    4. Residual Stream - 16k
    5. 5-GEMMASCOPE-RES-16K
    6. 4721
    Prev
    Next
    INDEX
    Explanations

    instances of the word "But" and its variations, indicating contrast or exception in ideas

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.5.hook_resid_post
    Hook Layer
    5
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    )");
    
    -1.22
    .";
    
    -1.13
    `;
    
    -1.11
    )";
    
    -1.10
    %");
    -1.06
    >");
    
    -1.02
    )++;
    -1.01
    ]";
    -1.01
    ніципалі
    -1.00
    "],
    
    -0.99
    POSITIVE LOGITS
    But
    1.42
     But
    1.31
    And
    1.10
     And
    0.98
    So
    0.84
    In
    0.76
    With
    0.75
     Pero
    0.72
    Pero
    0.71
     So
    0.71
    Activations Density 0.156%

    No Known Activations