© 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. 22-GEMMASCOPE-RES-131K
    7. 75312
    Prev
    Next
    INDEX
    Explanations

    foreign place names

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

    The neuron responds to non-ASCII Latin characters—especially vowels marked with diacritics (e.g. macrons or tone marks) as used in pinyin.

    oai_token-act-pair · o4-miniTriggered 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
    Not in Any Lists

    No Comments

    Negative Logits
    camis
    -1.23
     Zuhause
    -1.12
    kimono
    -1.09
    dsquared
    -1.09
     of
    -1.03
     «
    -1.02
     žaid
    -1.02
    seiko
    -1.02
     vannak
    -1.00
    「
    -1.00
    POSITIVE LOGITS
    fordshire
    1.27
    tivamente
    1.15
    wać
    1.15
    zembro
    1.12
    iamo
    1.12
    Langkah
    1.11
    смартфон
    1.11
    خخ
    1.11
    مقدمه
    1.10
    .(*
    1.09
    Activations Density 0.038%

    No Known Activations