© 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. 22-GEMMASCOPE-RES-131K
    4. 74924
    Prev
    Next
    INDEX
    Explanations

    food toppings and ingredients

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

    food and culinary terms, particularly focusing on ingredients and toppings.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    The neuron fires strongly on words naming foods, ingredients, and food-related items.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
     through
    -1.48
     this
    -1.30
    āja
    -1.25
    bergh
    -1.25
    нент
    -1.15
    тенок
    -1.14
    Результа
    -1.13
    frica
    -1.13
     in
    -1.13
     data
    -1.11
    POSITIVE LOGITS
    ()]
    
    1.25
     الشاشة
    1.16
    ycznie
    1.13
     choisi
    1.12
     repuesto
    1.09
    daun
    1.09
     paillettes
    1.09
     Relaciones
    1.08
     aggiungere
    1.06
    icestershire
    1.06
    Activations Density 0.022%

    No Known Activations