© 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-3-12B
    3. 31-GEMMASCOPE-2-RES-16K
    4. 16001
    Prev
    Next
    INDEX
    Explanations

    closing parenthesis followed by math/code symbols

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

    This neuron fires on numbers or numeric markers enclosed in parentheses—e.g., parenthesized citation indices, ratings, or percentage/statistical annotations.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-12b-pt/resid_post/layer_31_width_16k_l0_medium
    Prompts (Dashboard)
    392,802 prompts, 256 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    oyne
    0.90
    frid
    0.78
    .
    0.77
     underlines
    0.77
    FH
    0.76
    Vermont
    0.75
    rée
    0.74
    NESS
    0.73
    {//
    0.73
    RAF
    0.72
    POSITIVE LOGITS
     AppleWebKit
    0.85
     IBOutlet
    0.80
    א
    0.74
    形态
    0.66
    ار
    0.65
    崤
    0.65
    的大
    0.64
     dailySales
    0.63
     อ
    0.63
     До
    0.62
    Activations Density 0.192%

    No Known Activations