© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 131005
    Prev
    Next
    INDEX
    Explanations

    names starting with Tr, Gar, Z, Noble, Ign

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

    The neuron consistently responds to subword tokens that begin with an uppercase letter—i.e. the initial pieces of proper names or tagged elements.

    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
     these
    -1.80
     食物
    -1.80
    garmin
    -1.73
    ~~~~
    -1.69
    inspira
    -1.68
     に関する
    -1.67
    Each
    -1.55
    该如何
    -1.55
    barcelona
    -1.53
    afro
    -1.52
    POSITIVE LOGITS
    kande
    1.67
    Уже
    1.51
    ]));
    1.51
    自从
    1.50
    数
    1.48
    >
    
    1.46
    尽管
    1.38
     LIABLE
    1.38
    ();
    1.38
    '){
    
    1.38
    Activations Density 0.003%

    No Known Activations