© 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. 70821
    Prev
    Next
    INDEX
    Explanations

    initials or abbreviations related to names

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

    This neuron fires on alphanumeric identifier tokens that combine uppercase letters with digits (e.g. “C 183,” “GG 53.75,” “CE 20.375”).

    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
     leur
    -1.52
     algunas
    -1.33
     quelques
    -1.25
     In
    -1.25
     on
    -1.24
    ↵
    -1.24
     re
    -1.22
     двумя
    -1.22
     otras
    -1.21
     in
    -1.20
    POSITIVE LOGITS
    閖
    1.77
     erhi
    1.66
    hmigung
    1.66
     margiela
    1.53
     torino
    1.51
     CALIFORNI
    1.50
     iveco
    1.50
    ーズン
    1.48
     verona
    1.48
     niedersachsen
    1.46
    Activations Density 0.042%

    No Known Activations