© 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. 10-GEMMASCOPE-RES-131K
    7. 19645
    Prev
    Next
    INDEX
    Explanations

    numbers followed by units or punctuation

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

    The neuron detects numeric measurement tokens—especially numbers (often with decimals or percentages) that report values, dosages, or metric readings.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
     penser
    -1.62
     豚
    -1.59
     henvisninger
    -1.57
    cising
    -1.52
    existent
    -1.51
     μαζί
    -1.50
     organisé
    -1.45
     estekak
    -1.41
     Confira
    -1.41
    嘞
    -1.41
    POSITIVE LOGITS
    ↵↵
    1.91
     because
    1.77
     although
    1.64
     
    1.52
    </b>
    1.48
     despite
    1.45
     if
    1.45
     budaya
    1.42
    ↵↵↵
    1.34
    ↵↵↵↵
    1.30
    Activations Density 0.095%

    No Known Activations