© 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. 34-GEMMASCOPE-RES-131K
    7. 102633
    Prev
    Next
    INDEX
    Explanations

    reserved to

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

    The neuron is tuned to detect numeric tokens—sequences of digits such as counts, IDs, version numbers, or other numeric values.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
    ฏิ
    -0.69
    ípio
    -0.68
    getGroup
    -0.68
     duży
    -0.65
     beaded
    -0.65
     گروه
    -0.64
     まで
    -0.63
     MAYO
    -0.63
     склады
    -0.63
    elde
    -0.63
    POSITIVE LOGITS
    NELL
    0.68
    grafic
    0.68
     nil
    0.64
     Weiter
    0.64
     Ш
    0.64
     Miel
    0.64
    хода
    0.63
    estre
    0.63
     Sre
    0.62
    sekten
    0.61
    Activations Density 0.068%

    No Known Activations