© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 78284
    Prev
    Next
    INDEX
    Explanations

    historical texts and scholars

    np_acts-logits-general · gemini-2.5-flash-lite

    The neuron is tuned to detect numeric tokens (e.g. years, page counts, volume or issue numbers) typically found in reference or bibliography sections.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    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
    üdis
    -0.74
    记录
    -0.73
    的神
    -0.69
     colds
    -0.69
     cherish
    -0.69
     Geh
    -0.69
     report
    -0.68
    記録
    -0.67
     collections
    -0.66
     Maggio
    -0.66
    POSITIVE LOGITS
     Medieval
    1.07
     medieval
    0.94
     美術
    0.88
    Medieval
    0.85
    いため
    0.75
    Clause
    0.75
     Clause
    0.74
    ઃ
    0.73
     Dyn
    0.73
     Ibn
    0.72
    Activations Density 0.019%

    No Known Activations