© 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. 22-GEMMASCOPE-RES-131K
    4. 130911
    Prev
    Next
    INDEX
    Explanations

    The neuron fires on numeric tokens and mathematical formula components (e.g. digits, subscripts, and math symbols).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    distance or difference calculation

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
    tillinger
    -0.78
    ալ
    -0.77
    PointCloud
    -0.75
     المؤلف
    -0.74
    жении
    -0.71
    gamb
    -0.70
    pág
    -0.70
    がり
    -0.69
    nergie
    -0.69
     Dean
    -0.69
    POSITIVE LOGITS
     difference
    1.60
    Difference
    1.48
     Difference
    1.46
    difference
    1.37
     differences
    1.34
     diferença
    1.30
     DIFFERENCE
    1.26
     mismatch
    1.23
     deviation
    1.20
    Differences
    1.20
    Activations Density 0.023%

    No Known Activations