© 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. 9950
    Prev
    Next
    INDEX
    Explanations

    race names like Dale Stakes

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

    The neuron activates on multi‐word capitalized proper names of Thoroughbred racehorses (e.g. “Sea the Stars,” “Stradivarius,” “Galileo,” “Teofilo”).

    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
     of
    -1.74
     comportements
    -1.48
    雠
    -1.41
     dizaines
    -1.40
     centaines
    -1.39
     camiset
    -1.38
     Antropo
    -1.34
     Bretaña
    -1.31
     milliers
    -1.31
     dritten
    -1.31
    POSITIVE LOGITS
     to
    2.00
     You
    1.38
     overras
    1.38
     ongeveer
    1.33
    عر
    1.29
    Then
    1.28
    
    1.27
     leh
    1.23
     then
    1.21
    ospatial
    1.20
    Activations Density 0.006%

    No Known Activations