© 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. 76060
    Prev
    Next
    INDEX
    Explanations

    number formatting with commas

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

    The neuron fires on mentions of formatting numbers with a thousands separator (e.g. “thousand separator,” adding commas/spaces to large numbers).

    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
     Ux
    -0.80
     Capsule
    -0.77
     exiles
    -0.76
     كري
    -0.73
    belongsTo
    -0.73
    Capsule
    -0.73
     酷
    -0.71
    Lic
    -0.71
     gala
    -0.70
     LIC
    -0.70
    POSITIVE LOGITS
     thousands
    1.75
     thousand
    1.52
    thousands
    1.41
    Thousands
    1.38
     Thousands
    1.38
     millions
    1.36
     million
    1.27
     commas
    1.23
    millions
    1.22
    thousand
    1.20
    Activations Density 0.021%

    No Known Activations