© 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. 12636
    Prev
    Next
    INDEX
    Explanations

    features and capabilities

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

    The neuron selectively fires on capitalized names of software products, branded features, or project titles.

    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
     deoarece
    -1.23
    endő
    -1.15
     Именно
    -1.14
    並沒有
    -1.13
     ≈
    -1.12
    Diseño
    -1.11
    誒
    -1.11
    ঈ
    -1.10
    étaires
    -1.08
     latach
    -1.08
    POSITIVE LOGITS
     capability
    1.73
     option
    1.72
     function
    1.70
     feature
    1.66
     you
    1.59
     capabilities
    1.50
     functionality
    1.47
     ability
    1.39
     in
    1.36
     program
    1.34
    Activations Density 0.076%

    No Known Activations