© 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. 104624
    Prev
    Next
    INDEX
    Explanations

    Simmons

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

    The neuron fires most strongly on tokens written in a non-English script or coming from a foreign language (e.g. Japanese kana/kanji, German “Ähnlich,” etc.).

    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
    釙
    -3.06
     =
    -2.67
    猡
    -2.52
    骓
    -2.50
    壜
    -2.50
    栳
    -2.44
    跸
    -2.41
    梘
    -2.41
    𖥧
    -2.38
    鈈
    -2.38
    POSITIVE LOGITS
     You
    2.36
     Some
    2.17
    You
    2.17
    3
    2.17
    ?
    2.13
    th
    2.05
     traditionellen
    2.05
    6
    2.02
    Some
    1.95
     Einzelnen
    1.89
    Activations Density 0.002%

    No Known Activations