© 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. 104907
    Prev
    Next
    INDEX
    Explanations

    states about numbers

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

    The neuron strongly activates on numeric tokens or numerical data (counts, measures, percentages) in the text.

    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
     first
    -1.41
    蛲
    -1.40
    ecé
    -1.31
     intermédiaire
    -1.28
    鹩
    -1.27
     posled
    -1.27
    
    -1.24
    ТОВ
    -1.23
    Réponse
    -1.23
    輅
    -1.23
    POSITIVE LOGITS
     that
    1.89
    -
    1.86
     is
    1.48
     would
    1.47
     but
    1.44
     all
    1.42
     всички
    1.33
      
    1.30
    ..
    1.30
     should
    1.28
    Activations Density 0.063%

    No Known Activations