© 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. 22-GEMMASCOPE-RES-131K
    7. 84119
    Prev
    Next
    INDEX
    Explanations

    UI colors and code syntax

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

    The neuron fires on tokens naming or referencing colors (e.g. color properties, color resource keys, tint/background/brush terms) in code or markup.

    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
     Δια
    -1.27
    ようです
    -1.21
    んでしょう
    -1.19
    __':
    -1.16
    isling
    -1.15
     Με
    -1.13
    に見える
    -1.13
     介绍
    -1.10
     nguyễn
    -1.09
    jdź
    -1.09
    POSITIVE LOGITS
    がありますが
    1.27
    などを
    1.20
    などは
    1.13
    Напомним
    1.13
    与此同时
    1.08
     Aktivität
    1.08
     мероприятий
    1.08
    Якщо
    1.08
    より
    1.07
    Sejarah
    1.07
    Activations Density 0.003%

    No Known Activations