© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 67869
    Prev
    Next
    INDEX
    Explanations

    numbers and dates

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

    The neuron is strongly activated by sequences of digits (especially long numbers, phone‐number-style strings, prices, dates, or other numeric data).

    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
     "];
    -1.73
     rapporter
    -1.70
     Itali
    -1.70
    ')}}"
    -1.67
    tapete
    -1.66
     restent
    -1.63
    蛩
    -1.60
     perioden
    -1.59
     gardent
    -1.59
     kahve
    -1.59
    POSITIVE LOGITS
     on
    1.75
    —
    1.70
     different
    1.60
    //
    1.49
     
    1.49
     specific
    1.42
     that
    1.35
     there
    1.33
    .,
    1.33
     not
    1.32
    Activations Density 0.010%

    No Known Activations