© 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. 10-GEMMASCOPE-RES-131K
    4. 61439
    Prev
    Next
    INDEX
    Explanations

    numbers following "No."

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

    The neuron detects numbered reference markers—tokens like “No.”/“Nos.” and adjoining digits that indicate document, grant, patent, or case identification numbers.

    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
    These
    -2.09
    Those
    -2.00
    One
    -1.85
     presumably
    -1.83
     utterly
    -1.83
    Just
    -1.82
     and
    -1.81
     primarily
    -1.79
    Two
    -1.78
    Different
    -1.73
    POSITIVE LOGITS
    翆
    1.88
    玙
    1.84
    ⎔
    1.77
    狴
    1.73
    虓
    1.71
    
    1.69
    翃
    1.67
    遊んで
    1.66
     améli
    1.66
     USING
    1.66
    Activations Density 0.074%

    No Known Activations