© 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. 22-GEMMASCOPE-RES-131K
    7. 39908
    Prev
    Next
    INDEX
    Explanations

    locations and historical sites

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

    The neuron specializes in picking out mentions of “site” as in a named or designated location (e.g., historical or geographic site).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
     hierdie
    -1.11
     участия
    -1.06
     дет
    -1.05
     Dinas
    -1.05
    ษัท
    -1.04
    -1.03
     lange
    -1.02
     Yojana
    -1.02
    әм
    -1.02
    pex
    -1.01
    POSITIVE LOGITS
     where
    1.77
     σημαν
    1.16
     ιστο
    1.09
    ЦИЯ
    1.05
     location
    1.05
     महत्वपूर्ण
    1.03
    ruzione
    1.01
    实际
    1.01
    Where
    0.99
     פֿ
    0.98
    Activations Density 0.058%

    No Known Activations