© 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. 9479
    Prev
    Next
    INDEX
    Explanations

    company suffixes

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

    The neuron fires on tokens that are uppercase strings or contain digits (e.g. parts of company names, model numbers, codes, and other capitalized or numeric identifiers).

    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
    informationen
    -1.06
    是怎么回事
    -1.05
    klassen
    -0.99
     ГА
    -0.99
    ತ್ತ
    -0.98
    prepareStatement
    -0.98
    🧃
    -0.98
    arım
    -0.97
    Créditos
    -0.97
    已經
    -0.95
    POSITIVE LOGITS
     such
    1.36
     chén
    1.06
     each
    1.03
     подобрать
    1.01
     its
    0.98
    ρεί
    0.98
    ellschaft
    0.95
     нач
    0.95
     trastorno
    0.95
     now
    0.93
    Activations Density 0.021%

    No Known Activations