© 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. 69112
    Prev
    Next
    INDEX
    Explanations

    cities and agreements

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

    The neuron detects proper geographic names (cities or places) when they appear in the titles of conferences, treaties, or international agreements.

    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
    凿
    -1.70
     nabíd
    -1.63
    зербай
    -1.52
     Empieza
    -1.49
     Zoltán
    -1.46
     excus
    -1.45
    нить
    -1.44
    konos
    -1.44
     beginnen
    -1.42
     verminder
    -1.42
    POSITIVE LOGITS
     time
    1.51
     three
    1.41
     two
    1.40
     an
    1.38
     about
    1.37
     La
    1.35
     econó
    1.34
     one
    1.30
     prace
    1.26
     производстве
    1.25
    Activations Density 0.048%

    No Known Activations