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-2-2B
    4. Residual Stream - 65k
    5. 13-GEMMASCOPE-RES-65K
    6. 39278
    Prev
    Next
    INDEX
    Explanations

    occurrences of the word "NBC" in different contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 13-gemmascope-res-65k
    Configuration
    google/gemma-scope-2b-pt-res/layer_13/width_65k/average_l0_75
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    65,536
    Data Type
    float32
    Hook Name
    blocks.13.hook_resid_post
    Hook Layer
    13
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    <!--[
    -0.68
     '
    
    -0.64
    '],
    
    -0.61
    `,
    
    -0.60
    ]:
    
    -0.60
     resources
    -0.59
     everything
    -0.58
     itſelf
    -0.57
     duy
    -0.56
    `;
    
    -0.56
    POSITIVE LOGITS
     illetve
    0.70
     valamint
    0.58
     közül
    0.57
    <bos>
    0.57
     그리고
    0.57
     élevées
    0.51
    FormBuilder
    0.50
    หน่อย
    0.50
    IUrlHelper
    0.50
     etc
    0.49
    Activations Density 0.095%

    No Known Activations