© 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-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 357
    Prev
    Next
    INDEX
    Explanations

    The neuron is detecting numeric tokens—especially numbers and measurement values—within technical or specification‐style text.

    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
     使い
    0.66
     ﺍﻟ
    0.63
     अल्को
    0.63
     时尚
    0.63
     那麼
    0.62
    ȩ
    0.62
     コットン
    0.62
     綿
    0.62
     biosolids
    0.61
    ṓ
    0.61
    POSITIVE LOGITS
    0.74
    <unused62>
    0.66
    <unused60>
    0.64
    <unused61>
    0.64
    ↵
    0.62
    ((
    0.61
    i
    0.60
      
    0.59
    get
    0.59
    String
    0.59
    Activations Density 0.197%

    No Known Activations