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-2-27B
    3. 34-GEMMASCOPE-RES-131K
    4. 62
    Prev
    Next
    INDEX
    Explanations

    Crystal Palace

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    alne
    -0.80
    ちゃう
    -0.79
     JButton
    -0.79
    зонта
    -0.77
    katan
    -0.75
     Viva
    -0.72
     Polen
    -0.70
     NOK
    -0.70
    kend
    -0.69
     Fij
    -0.69
    POSITIVE LOGITS
    Crystal
    0.90
    crystal
    0.89
     Crystal
    0.89
     Palace
    0.88
    Palace
    0.78
     crystal
    0.76
    سور
    0.74
     crystallized
    0.71
     Croydon
    0.70
    palace
    0.70
    Activations Density 0.039%

    No Known Activations