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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Gemma-2-27B
    3. 34-GEMMASCOPE-RES-131K
    4. 7812
    Prev
    Next
    INDEX
    Explanations

    code structure and non-English characters

    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
     even
    -1.51
     whose
    -1.11
     même
    -1.09
    Даже
    -1.07
    Mär
    -1.00
     thậm
    -1.00
     information
    -0.95
     same
    -0.94
     mesmo
    -0.91
     went
    -0.89
    POSITIVE LOGITS
    ).
    1.04
    '),
    1.00
    achio
    0.97
    ]=='
    0.95
     olika
    0.95
     очередь
    0.94
    ₦
    0.94
     nimmer
    0.93
     ſy
    0.93
     maïs
    0.92
    Activations Density 0.008%

    No Known Activations