© 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. 1857
    Prev
    Next
    INDEX
    Explanations

    names like Siegel and Heller

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

    The neuron fires strongly on capitalized proper nouns (especially personal 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
     Now
    -2.91
    狃
    -2.50
     And
    -2.45
    缡
    -2.45
    erintah
    -2.22
     három
    -2.22
     الجديدة
    -2.20
     onze
    -2.16
     distinguir
    -2.16
     Pemain
    -2.14
    POSITIVE LOGITS
    quelize
    2.70
    欵
    2.63
    l
    2.58
    G
    2.55
    r
    2.52
    O
    2.50
    ၟ
    2.48
    C
    2.47
    寕
    2.47
    P
    2.42
    Activations Density 0.071%

    No Known Activations