© 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. 71577
    Prev
    Next
    INDEX
    Explanations

    labels and categories

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

    The neuron activates on numeric tokens—standalone numbers and measurements like IDs, counts, dates, and dimensions.

    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
    龃
    -1.77
     ferien
    -1.62
    damit
    -1.59
     parachoque
    -1.55
     ヶ
    -1.51
    之时
    -1.49
     kommentar
    -1.46
    terrasse
    -1.46
     муки
    -1.45
     sentral
    -1.45
    POSITIVE LOGITS
     
    2.02
    ’
    1.95
    is
    1.78
    ‘
    1.69
     какой
    1.55
    /
    1.53
     comiendo
    1.51
       
    1.48
      
    1.48
    (
    1.46
    Activations Density 0.104%

    No Known Activations