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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 130917
    Prev
    Next
    INDEX
    Explanations

    patent document

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

    This neuron detects structural metadata labels and numeric fields in Japanese-style event schedules (e.g. bracketed date, venue, time, ticket‐price 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_10/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
     or
    -2.27
     nahezu
    -2.00
    年初
    -1.96
     Ironically
    -1.94
     epitom
    -1.85
     With
    -1.81
     strives
    -1.76
    just
    -1.75
    檔案
    -1.74
     .....
    -1.70
    POSITIVE LOGITS
     terrific
    2.27
    鋱
    2.22
    錼
    2.20
    genodigd
    2.17
    䊐
    2.17
    ݮ
    2.08
    暖かい
    2.05
    Hmmm
    2.05
    癀
    2.02
     gruesome
    2.00
    Activations Density 0.003%

    No Known Activations