© 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
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 65012
    Prev
    Next
    INDEX
    Explanations

    words ending in multiple languages

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

    This neuron detects tokens containing non-ASCII letters (i.e. accented or Cyrillic characters), effectively flagging foreign-language or diacritic-bearing word fragments.

    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
     a
    -1.55
    ить
    -1.44
     configura
    -1.37
    bitkan
    -1.36
     opérateur
    -1.36
     from
    -1.32
    大人気
    -1.30
     zcela
    -1.27
     inaugu
    -1.27
     viciss
    -1.26
    POSITIVE LOGITS
     quando
    1.65
    iamo
    1.55
     when
    1.38
     esperan
    1.34
    f
    1.33
     cuando
    1.32
    ↵↵
    1.27
    in
    1.27
    r
    1.26
    auce
    1.25
    Activations Density 0.043%

    No Known Activations