© 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. 24933
    Prev
    Next
    INDEX
    Explanations

    CJK languages and unicode support

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

    The neuron fires on identifiers of programming or typesetting libraries/packages (e.g. TeX package names like “soul,” “calc,” “filecontents,” “cjk,” or code imports like “xrange”).

    oai_token-act-pair · o4-mini
    Triggered by @jyhe0408
    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
     allemande
    -0.96
     måneder
    -0.94
    ござい
    -0.93
     thermique
    -0.93
     care
    -0.93
     ratings
    -0.91
    ソング
    -0.91
     to
    -0.90
     suisse
    -0.90
     cubana
    -0.88
    POSITIVE LOGITS
    )$\\
    1.16
     子ども
    1.09
    uddled
    0.98
     необхід
    0.97
    الج
    0.94
    홍
    0.93
    ሕ
    0.90
    تل
    0.90
    umpang
    0.89
     competición
    0.88
    Activations Density 0.000%

    No Known Activations