© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 67164
    Prev
    Next
    INDEX
    Explanations

    thousand separator character grouping

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

    The neuron detects mentions of digit‐grouping or thousand‐separator instructions (e.g. adding spaces or separators every three digits).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    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
     maxHeight
    -1.00
    ficie
    -0.96
    dieran
    -0.94
     conducted
    -0.94
    yeli
    -0.94
    zieć
    -0.93
     compared
    -0.93
     dispoz
    -0.91
    rekli
    -0.91
     colorida
    -0.90
    POSITIVE LOGITS
    ãng
    0.93
    ̓
    0.91
    organised
    0.91
     пациентов
    0.90
    begin
    0.88
    raeli
    0.86
     emerg
    0.85
     Angew
    0.85
    ctuation
    0.84
     analyte
    0.83
    Activations Density 0.011%

    No Known Activations