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. 40780
    Prev
    Next
    INDEX
    Explanations

    code snippets

    np_max-act · gemini-2.0-flash

    It activates for short, high-frequency function words—articles, conjunctions, prepositions, pronouns, etc.—that appear preceded by a space.

    oai_token-act-pair · o3Triggered by @giuseppebirardi
    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
     vectoriales
    -0.63
     vérit
    -0.56
     giustizia
    -0.52
     rayures
    -0.52
     nélk
    -0.51
     cadeaux
    -0.51
     adicionais
    -0.51
     tromper
    -0.51
     larmes
    -0.50
     jouets
    -0.50
    POSITIVE LOGITS
    <eos>
    0.85
    ↵↵
    0.76
    >');
    0.76
    '},
    
    0.72
    )]);
    0.71
    IUrlHelper
    0.70
    '};
    0.69
    ()]);
    0.69
    ]))
    
    0.66
    ...'
    0.66
    Activations Density 47.664%

    No Known Activations