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. 16-GEMMASCOPE-RES-16K
    6. 15813
    Prev
    Next
    INDEX
    Explanations

    phrases indicating intent or accusations related to individuals or entities

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

    Follows punctuation (, . ?) or "saying"

    np_acts-logits-general · gemini-2.0-flash

    exclamations and questions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 16-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_16/width_16k/average_l0_78
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.16.hook_resid_post
    Hook Layer
    16
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    `,
    
    -0.94
     itſelf
    -0.92
    =?";
    -0.91
    .";
    
    -0.91
    "]);
    
    -0.90
     myſelf
    -0.90
     />';
    -0.88
     ""));
    -0.87
    `;
    
    -0.87
    });*/
    -0.86
    POSITIVE LOGITS
     fucking
    0.55
    .
    0.50
    ,
    0.49
     [
    0.49
     I
    0.47
     you
    0.45
    fucking
    0.45
     No
    0.44
     because
    0.43
     obligé
    0.42
    Activations Density 0.201%

    No Known Activations