© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 14442
    Prev
    Next
    INDEX
    Explanations

    cities and geographic context

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

    The neuron fires on named geographic reference points in distance statements—that is, place names cited as endpoints when giving how far something lies from them.

    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
     especially
    -1.46
     szczególnie
    -1.31
     zawsze
    -1.24
     negociar
    -1.21
     these
    -1.20
    Posté
    -1.20
     particularly
    -1.18
     można
    -1.18
     convencer
    -1.18
     özellikle
    -1.17
    POSITIVE LOGITS
     your
    1.38
     cabaña
    1.29
     to
    1.29
    phòng
    1.28
    了一口
    1.25
     tý
    1.21
    +{
    1.19
    U
    1.17
     Wants
    1.17
    uste
    1.17
    Activations Density 0.092%

    No Known Activations