© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 19810
    Prev
    Next
    INDEX
    Explanations

    European cities and countries

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

    This neuron detects proper names and named entities—e.g. people, teams, places, events, and organizations.

    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
    arrol
    -1.38
    銆
    -1.21
    ském
    -1.15
    Ꮑ
    -1.13
     spac
    -1.13
    없
    -1.12
     classics
    -1.07
     unveils
    -1.06
     stirs
    -1.06
    pullover
    -1.05
    POSITIVE LOGITS
     their
    1.52
     could
    1.21
     Though
    1.14
    ابر
    1.09
     Obwohl
    1.01
    ……」
    1.01
    ・・・」
    1.01
    asties
    0.98
     Notwithstanding
    0.98
    ˃
    0.98
    Activations Density 0.289%

    No Known Activations