Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Gemma-2-2B
    3. 0-CLT-HP
    4. 40771
    Prev
    Next
    INDEX
    Explanations

    punctuation

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ;
    -1.05
    ;
    -0.83
     étoient
    -0.79
    ؛
    -0.78
    ;
    
    -0.77
    ';
    -0.71
     réfrig
    -0.70
     enfans
    -0.69
    ;\
    -0.69
    ;;
    -0.69
    POSITIVE LOGITS
     but
    0.72
    存于互联网档案馆
    0.72
     it
    0.71
     they
    0.70
     дописавши
    0.70
     we
    0.68
    msgTypes
    0.68
     she
    0.67
     however
    0.67
    انيف
    0.65
    Activations Density 0.262%

    No Known Activations