© 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. 110871
    Prev
    Next
    INDEX
    Explanations

    place names and routes

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

    The neuron identifies and strongly activates on proper noun phrases—especially place names, institutional names, and other named entities.

    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
    𝔼
    -1.11
    филь
    -1.09
     andern
    -1.09
     attente
    -1.09
    ῾
    -1.06
    𝙽
    -1.06
     extremists
    -1.05
    鴴
    -1.04
     खरीद
    -1.04
     primero
    -1.02
    POSITIVE LOGITS
     because
    1.30
     people
    1.27
     due
    1.14
    ”。
    1.09
     good
    1.09
     well
    1.07
     will
    1.06
    )=\
    1.04
     children
    1.03
     two
    1.01
    Activations Density 0.266%

    No Known Activations