© 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. 21646
    Prev
    Next
    INDEX
    Explanations

    historical figures and characters

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

    The neuron detects named locations—especially multiword proper‐noun place names like museums, historic sites, parks, and similar venues.

    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
     reciente
    -1.33
     всичко
    -1.30
     Вода
    -1.29
    ᯾
    -1.27
     subray
    -1.19
     vandaag
    -1.17
    来不及
    -1.17
    ᛬
    -1.14
     botão
    -1.13
     EVER
    -1.12
    POSITIVE LOGITS
     about
    1.38
     always
    1.11
     often
    1.09
     equilibr
    1.09
    not
    1.02
     ない
    1.02
     rapida
    1.02
    我是
    1.02
    lijn
    1.02
     experien
    1.01
    Activations Density 0.052%

    No Known Activations