© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 127918
    Prev
    Next
    INDEX
    Explanations

    The neuron fires on the “date:” metadata line (and its surrounding punctuation) in a document header.

    oai_token-act-pair · o4-mini
    Triggered by @jyhe0408

    date and address fields

    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_22/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Negative Logits
    hermosa
    -1.05
    了一会儿
    -1.05
    tieth
    -0.99
    ismus
    -0.99
    笵
    -0.98
    
    -0.97
     mers
    -0.96
    ttes
    -0.96
     republiky
    -0.95
     العديد
    -0.94
    POSITIVE LOGITS
     entregó
    1.05
    &
    1.05
    ಝ
    1.01
     uživ
    0.99
    ,:);
    0.99
     кстати
    0.99
     škoda
    0.98
    )
    0.96
    师父
    0.95
    Sì
    0.94
    Activations Density 0.005%

    No Known Activations