© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 130401
    Prev
    Next
    INDEX
    Explanations

    This neuron detects Chinese numeric expressions and quantifiers (e.g. words like “第一,” “一种,” “一个,” “多个”) that signal quantity or order.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    multiple kinds/examples/versions

    np_acts-logits-general · gemini-2.5-flash-lite
    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
     alcuni
    -1.76
     抠
    -1.75
    
    -1.73
    が多く
    -1.68
    ㋛
    -1.66
     belangrijk
    -1.66
     quantità
    -1.65
    1
    -1.62
    皷
    -1.56
     verrà
    -1.55
    POSITIVE LOGITS
     ‘
    2.05
    
    1.91
     profundas
    1.83
     There
    1.79
     ‘‘
    1.76
     (‘
    1.75
     machen
    1.73
     ”
    1.72
     variadas
    1.70
     consideradas
    1.69
    Activations Density 0.005%

    No Known Activations